* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86, pvclock: Remove leftover scale_delta() function
x86, apic: Remove double #include
x86: Adjust section annotations in AMD Fam10 MMCONF enabling code
x86, UV: Update node controller MMRs
x86: Remove unnecessary casts of void ptr returning alloc function return values
x86: Address gcc4.6 "set but not used" warnings in apic.h
x86, mm: Fix section mismatch in tlb.c
--- /dev/null
+What: /proc/<pid>/oom_adj
+When: August 2012
+Why: /proc/<pid>/oom_adj allows userspace to influence the oom killer's
+ badness heuristic used to determine which task to kill when the kernel
+ is out of memory.
+
+ The badness heuristic has since been rewritten since the introduction of
+ this tunable such that its meaning is deprecated. The value was
+ implemented as a bitshift on a score generated by the badness()
+ function that did not have any precise units of measure. With the
+ rewrite, the score is given as a proportion of available memory to the
+ task allocating pages, so using a bitshift which grows the score
+ exponentially is, thus, impossible to tune with fine granularity.
+
+ A much more powerful interface, /proc/<pid>/oom_score_adj, was
+ introduced with the oom killer rewrite that allows users to increase or
+ decrease the badness() score linearly. This interface will replace
+ /proc/<pid>/oom_adj.
+
+ A warning will be emitted to the kernel log if an application uses this
+ deprecated interface. After it is printed once, future warnings will be
+ suppressed until the kernel is rebooted.
prototypes:
int (*fl_compare_owner)(struct file_lock *, struct file_lock *);
void (*fl_notify)(struct file_lock *); /* unblock callback */
- void (*fl_copy_lock)(struct file_lock *, struct file_lock *);
void (*fl_release_private)(struct file_lock *);
void (*fl_break)(struct file_lock *); /* break_lease callback */
BKL may block
fl_compare_owner: yes no
fl_notify: yes no
-fl_copy_lock: yes no
fl_release_private: yes yes
fl_break: yes no
Roadmap:
-2.6.37 Remove experimental tag from mount option
- => should be roughly 6 months after initial merge
- => enough time to:
- => gain confidence and fix problems reported by early
- adopters (a.k.a. guinea pigs)
- => address worst performance regressions and undesired
- behaviours
- => start tuning/optimising code for parallelism
- => start tuning/optimising algorithms consuming
- excessive CPU time
-
2.6.39 Switch default mount option to use delayed logging
=> should be roughly 12 months after initial merge
=> enough time to shake out remaining problems before next round of
* Intel 82801I (ICH9)
* Intel EP80579 (Tolapai)
* Intel 82801JI (ICH10)
- * Intel 3400/5 Series (PCH)
+ * Intel 5/3400 Series (PCH)
* Intel Cougar Point (PCH)
+ * Intel Patsburg (PCH)
Datasheets: Publicly available at the Intel website
+On Intel Patsburg and later chipsets, both the normal host SMBus controller
+and the additional 'Integrated Device Function' controllers are supported.
+
Authors:
Mark Studebaker <mdsxyz123@yahoo.com>
Jean Delvare <khali@linux-fr.org>
Some LEDs can be programmed to blink without any CPU interaction. To
support this feature, a LED driver can optionally implement the
-blink_set() function (see <linux/leds.h>). If implemented, triggers can
-attempt to use it before falling back to software timers. The blink_set()
-function should return 0 if the blink setting is supported, or -EINVAL
-otherwise, which means that LED blinking will be handled by software.
-
-The blink_set() function should choose a user friendly blinking
-value if it is called with *delay_on==0 && *delay_off==0 parameters. In
-this case the driver should give back the chosen value through delay_on
-and delay_off parameters to the leds subsystem.
+blink_set() function (see <linux/leds.h>). To set an LED to blinking,
+however, it is better to use use the API function led_blink_set(),
+as it will check and implement software fallback if necessary.
+
+To turn off blinking again, use the API function led_brightness_set()
+as that will not just set the LED brightness but also stop any software
+timers that may have been required for blinking.
+
+The blink_set() function should choose a user friendly blinking value
+if it is called with *delay_on==0 && *delay_off==0 parameters. In this
+case the driver should give back the chosen value through delay_on and
+delay_off parameters to the leds subsystem.
Setting the brightness to zero with brightness_set() callback function
should completely turn off the LED and cancel the previously programmed
--- /dev/null
+Kernel driver for lp5521
+========================
+
+* National Semiconductor LP5521 led driver chip
+* Datasheet: http://www.national.com/pf/LP/LP5521.html
+
+Authors: Mathias Nyman, Yuri Zaporozhets, Samu Onkalo
+Contact: Samu Onkalo (samu.p.onkalo-at-nokia.com)
+
+Description
+-----------
+
+LP5521 can drive up to 3 channels. Leds can be controlled directly via
+the led class control interface. Channels have generic names:
+lp5521:channelx, where x is 0 .. 2
+
+All three channels can be also controlled using the engine micro programs.
+More details of the instructions can be found from the public data sheet.
+
+Control interface for the engines:
+x is 1 .. 3
+enginex_mode : disabled, load, run
+enginex_load : store program (visible only in engine load mode)
+
+Example (start to blink the channel 2 led):
+cd /sys/class/leds/lp5521:channel2/device
+echo "load" > engine3_mode
+echo "037f4d0003ff6000" > engine3_load
+echo "run" > engine3_mode
+
+stop the engine:
+echo "disabled" > engine3_mode
+
+sysfs contains a selftest entry.
+The test communicates with the chip and checks that
+the clock mode is automatically set to the requested one.
+
+Each channel has its own led current settings.
+/sys/class/leds/lp5521:channel0/led_current - RW
+/sys/class/leds/lp5521:channel0/max_current - RO
+Format: 10x mA i.e 10 means 1.0 mA
+
+example platform data:
+
+Note: chan_nr can have values between 0 and 2.
+
+static struct lp5521_led_config lp5521_led_config[] = {
+ {
+ .chan_nr = 0,
+ .led_current = 50,
+ .max_current = 130,
+ }, {
+ .chan_nr = 1,
+ .led_current = 0,
+ .max_current = 130,
+ }, {
+ .chan_nr = 2,
+ .led_current = 0,
+ .max_current = 130,
+ }
+};
+
+static int lp5521_setup(void)
+{
+ /* setup HW resources */
+}
+
+static void lp5521_release(void)
+{
+ /* Release HW resources */
+}
+
+static void lp5521_enable(bool state)
+{
+ /* Control of chip enable signal */
+}
+
+static struct lp5521_platform_data lp5521_platform_data = {
+ .led_config = lp5521_led_config,
+ .num_channels = ARRAY_SIZE(lp5521_led_config),
+ .clock_mode = LP5521_CLOCK_EXT,
+ .setup_resources = lp5521_setup,
+ .release_resources = lp5521_release,
+ .enable = lp5521_enable,
+};
+
+If the current is set to 0 in the platform data, that channel is
+disabled and it is not visible in the sysfs.
--- /dev/null
+Kernel driver for lp5523
+========================
+
+* National Semiconductor LP5523 led driver chip
+* Datasheet: http://www.national.com/pf/LP/LP5523.html
+
+Authors: Mathias Nyman, Yuri Zaporozhets, Samu Onkalo
+Contact: Samu Onkalo (samu.p.onkalo-at-nokia.com)
+
+Description
+-----------
+LP5523 can drive up to 9 channels. Leds can be controlled directly via
+the led class control interface. Channels have generic names:
+lp5523:channelx where x is 0...8
+
+The chip provides 3 engines. Each engine can control channels without
+interaction from the main CPU. Details of the micro engine code can be found
+from the public data sheet. Leds can be muxed to different channels.
+
+Control interface for the engines:
+x is 1 .. 3
+enginex_mode : disabled, load, run
+enginex_load : microcode load (visible only in load mode)
+enginex_leds : led mux control (visible only in load mode)
+
+cd /sys/class/leds/lp5523:channel2/device
+echo "load" > engine3_mode
+echo "9d80400004ff05ff437f0000" > engine3_load
+echo "111111111" > engine3_leds
+echo "run" > engine3_mode
+
+sysfs contains a selftest entry. It measures each channel
+voltage level and checks if it looks reasonable. If the level is too high,
+the led is missing; if the level is too low, there is a short circuit.
+
+Selftest uses always the current from the platform data.
+
+Each channel contains led current settings.
+/sys/class/leds/lp5523:channel2/led_current - RW
+/sys/class/leds/lp5523:channel2/max_current - RO
+Format: 10x mA i.e 10 means 1.0 mA
+
+Example platform data:
+
+Note - chan_nr can have values between 0 and 8.
+
+static struct lp5523_led_config lp5523_led_config[] = {
+ {
+ .chan_nr = 0,
+ .led_current = 50,
+ .max_current = 130,
+ },
+...
+ }, {
+ .chan_nr = 8,
+ .led_current = 50,
+ .max_current = 130,
+ }
+};
+
+static int lp5523_setup(void)
+{
+ /* Setup HW resources */
+}
+
+static void lp5523_release(void)
+{
+ /* Release HW resources */
+}
+
+static void lp5523_enable(bool state)
+{
+ /* Control chip enable signal */
+}
+
+static struct lp5523_platform_data lp5523_platform_data = {
+ .led_config = lp5523_led_config,
+ .num_channels = ARRAY_SIZE(lp5523_led_config),
+ .clock_mode = LP5523_CLOCK_EXT,
+ .setup_resources = lp5523_setup,
+ .release_resources = lp5523_release,
+ .enable = lp5523_enable,
+};
+1 Release Date : Thur. May 03, 2010 09:12:45 PST 2009 -
+ (emaild-id:megaraidlinux@lsi.com)
+ Bo Yang
+
+2 Current Version : 00.00.04.31-rc1
+3 Older Version : 00.00.04.17.1-rc1
+
+1. Add the Online Controller Reset (OCR) to the Driver.
+ OCR is the new feature for megaraid_sas driver which
+ will allow the fw to do the chip reset which will not
+ affact the OS behavious.
+
+ To add the OCR support, driver need to do:
+ a). reset the controller chips -- Xscale and Gen2 which
+ will change the function calls and add the reset function
+ related to this two chips.
+
+ b). during the reset, driver will store the pending cmds
+ which not returned by FW to driver's pending queue. Driver
+ will re-issue those pending cmds again to FW after the OCR
+ finished.
+
+ c). In driver's timeout routine, driver will report to
+ OS as reset. Also driver's queue routine will block the
+ cmds until the OCR finished.
+
+ d). in Driver's ISR routine, if driver get the FW state as
+ state change, FW in Failure status and FW support online controller
+ reset (OCR), driver will start to do the controller reset.
+
+ e). In driver's IOCTL routine, the application cmds will wait for the
+ OCR to finish, then issue the cmds to FW.
+
+ f). Before driver kill adapter, driver will do last chance of
+ OCR to see if driver can bring back the FW.
+
+2. Add the support update flag to the driver to tell LSI megaraid_sas
+ application which driver will support the device update. So application
+ will not need to do the device update after application add/del the device
+ from the system.
+3. In driver's timeout routine, driver will do three time reset if fw is in
+ failed state. Driver will kill adapter if can't bring back FW after the
+ this three times reset.
+4. Add the input parameter max_sectors to 1MB support to our GEN2 controller.
+ customer can use the input paramenter max_sectors to add 1MB support to GEN2
+ controller.
+
1 Release Date : Thur. Oct 29, 2009 09:12:45 PST 2009 -
(emaild-id:megaraidlinux@lsi.com)
Bo Yang
- core_uses_pid
- ctrl-alt-del
- dentry-state
+- dmesg_restrict
- domainname
- hostname
- hotplug
==============================================================
+dmesg_restrict:
+
+This toggle indicates whether unprivileged users are prevented from using
+dmesg(8) to view messages from the kernel's log buffer. When
+dmesg_restrict is set to (0) there are no restrictions. When
+dmesg_restrict is set set to (1), users must have CAP_SYS_ADMIN to use
+dmesg(8).
+
+The kernel config option CONFIG_SECURITY_DMESG_RESTRICT sets the default
+value of dmesg_restrict.
+
+==============================================================
+
domainname & hostname:
These files can be used to set the NIS/YP domainname and the
W: http://developer.axis.com
S: Maintained
F: arch/cris/
+F: drivers/serial/crisv10.*
CRYPTO API
M: Herbert Xu <herbert@gondor.apana.org.au>
VERSION = 2
PATCHLEVEL = 6
-SUBLEVEL = 36
-EXTRAVERSION =
+SUBLEVEL = 37
+EXTRAVERSION = -rc1
NAME = Flesh-Eating Bats with Fangs
# *DOCUMENTATION*
config SH_CLK_CPG
bool
+source "drivers/sh/Kconfig"
+
endif
/* FSI */
#define IRQ_FSI evt2irq(0x1840)
+
+static int fsi_set_rate(int is_porta, int rate)
+{
+ struct clk *fsib_clk;
+ struct clk *fdiv_clk = &sh7372_fsidivb_clk;
+ int ret;
+
+ /* set_rate is not needed if port A */
+ if (is_porta)
+ return 0;
+
+ fsib_clk = clk_get(NULL, "fsib_clk");
+ if (IS_ERR(fsib_clk))
+ return -EINVAL;
+
+ switch (rate) {
+ case 48000:
+ clk_set_rate(fsib_clk, clk_round_rate(fsib_clk, 85428000));
+ clk_set_rate(fdiv_clk, clk_round_rate(fdiv_clk, 12204000));
+ ret = SH_FSI_ACKMD_256 | SH_FSI_BPFMD_64;
+ break;
+ default:
+ pr_err("unsupported rate in FSI2 port B\n");
+ ret = -EINVAL;
+ break;
+ }
+
+ clk_put(fsib_clk);
+
+ return ret;
+}
+
static struct sh_fsi_platform_info fsi_info = {
.porta_flags = SH_FSI_BRS_INV |
SH_FSI_OUT_SLAVE_MODE |
SH_FSI_IN_SLAVE_MODE |
SH_FSI_OFMT(PCM) |
SH_FSI_IFMT(PCM),
+
+ .portb_flags = SH_FSI_BRS_INV |
+ SH_FSI_BRM_INV |
+ SH_FSI_LRS_INV |
+ SH_FSI_OFMT(SPDIF),
+ .set_rate = fsi_set_rate,
};
static struct resource fsi_resources[] = {
static struct sh_mobile_hdmi_info hdmi_info = {
.lcd_chan = &sh_mobile_lcdc1_info.ch[0],
.lcd_dev = &lcdc1_device.dev,
+ .flags = HDMI_SND_SRC_SPDIF,
};
static struct resource hdmi_resources[] = {
#define GPIO_PORT9CR 0xE6051009
#define GPIO_PORT10CR 0xE605100A
+#define USCCR1 0xE6058144
static void __init ap4evb_init(void)
{
u32 srcr4;
/* setup USB phy */
__raw_writew(0x8a0a, 0xE6058130); /* USBCR2 */
- /* enable FSI2 */
+ /* enable FSI2 port A (ak4643) */
gpio_request(GPIO_FN_FSIAIBT, NULL);
gpio_request(GPIO_FN_FSIAILR, NULL);
gpio_request(GPIO_FN_FSIAISLD, NULL);
gpio_request(GPIO_PORT41, NULL);
gpio_direction_input(GPIO_PORT41);
+ /* setup FSI2 port B (HDMI) */
+ gpio_request(GPIO_FN_FSIBCK, NULL);
+ __raw_writew(__raw_readw(USCCR1) & ~(1 << 6), USCCR1); /* use SPDIF */
+
/* set SPU2 clock to 119.6 MHz */
clk = clk_get(NULL, "spu_clk");
if (!IS_ERR(clk)) {
#define SMSTPCR3 0xe615013c
#define SMSTPCR4 0xe6150140
+#define FSIDIVA 0xFE1F8000
+#define FSIDIVB 0xFE1F8008
+
/* Platforms must set frequency on their DV_CLKI pin */
struct clk sh7372_dv_clki_clk = {
};
.ops = &pllc2_clk_ops,
.parent = &extal1_div2_clk,
.freq_table = pllc2_freq_table,
+ .nr_freqs = ARRAY_SIZE(pllc2_freq_table) - 1,
.parent_table = pllc2_parent,
.parent_num = ARRAY_SIZE(pllc2_parent),
};
fsibckcr_parent, ARRAY_SIZE(fsibckcr_parent), 6, 2),
};
+/* FSI DIV */
+static unsigned long fsidiv_recalc(struct clk *clk)
+{
+ unsigned long value;
+
+ value = __raw_readl(clk->mapping->base);
+
+ if ((value & 0x3) != 0x3)
+ return 0;
+
+ value >>= 16;
+ if (value < 2)
+ return 0;
+
+ return clk->parent->rate / value;
+}
+
+static long fsidiv_round_rate(struct clk *clk, unsigned long rate)
+{
+ return clk_rate_div_range_round(clk, 2, 0xffff, rate);
+}
+
+static void fsidiv_disable(struct clk *clk)
+{
+ __raw_writel(0, clk->mapping->base);
+}
+
+static int fsidiv_enable(struct clk *clk)
+{
+ unsigned long value;
+
+ value = __raw_readl(clk->mapping->base) >> 16;
+ if (value < 2) {
+ fsidiv_disable(clk);
+ return -ENOENT;
+ }
+
+ __raw_writel((value << 16) | 0x3, clk->mapping->base);
+
+ return 0;
+}
+
+static int fsidiv_set_rate(struct clk *clk,
+ unsigned long rate, int algo_id)
+{
+ int idx;
+
+ if (clk->parent->rate == rate) {
+ fsidiv_disable(clk);
+ return 0;
+ }
+
+ idx = (clk->parent->rate / rate) & 0xffff;
+ if (idx < 2)
+ return -ENOENT;
+
+ __raw_writel(idx << 16, clk->mapping->base);
+ return fsidiv_enable(clk);
+}
+
+static struct clk_ops fsidiv_clk_ops = {
+ .recalc = fsidiv_recalc,
+ .round_rate = fsidiv_round_rate,
+ .set_rate = fsidiv_set_rate,
+ .enable = fsidiv_enable,
+ .disable = fsidiv_disable,
+};
+
+static struct clk_mapping sh7372_fsidiva_clk_mapping = {
+ .phys = FSIDIVA,
+ .len = 8,
+};
+
+struct clk sh7372_fsidiva_clk = {
+ .ops = &fsidiv_clk_ops,
+ .parent = &div6_reparent_clks[DIV6_FSIA], /* late install */
+ .mapping = &sh7372_fsidiva_clk_mapping,
+};
+
+static struct clk_mapping sh7372_fsidivb_clk_mapping = {
+ .phys = FSIDIVB,
+ .len = 8,
+};
+
+struct clk sh7372_fsidivb_clk = {
+ .ops = &fsidiv_clk_ops,
+ .parent = &div6_reparent_clks[DIV6_FSIB], /* late install */
+ .mapping = &sh7372_fsidivb_clk_mapping,
+};
+
+static struct clk *late_main_clks[] = {
+ &sh7372_fsidiva_clk,
+ &sh7372_fsidivb_clk,
+};
+
enum { MSTP001,
MSTP131, MSTP130,
MSTP129, MSTP128, MSTP127, MSTP126, MSTP125,
if (!ret)
ret = sh_clk_mstp32_register(mstp_clks, MSTP_NR);
+ for (k = 0; !ret && (k < ARRAY_SIZE(late_main_clks)); k++)
+ ret = clk_register(late_main_clks[k]);
+
clkdev_add_table(lookups, ARRAY_SIZE(lookups));
if (!ret)
static inline int gpio_to_irq(unsigned gpio)
{
- return -ENOSYS;
+ return __gpio_to_irq(gpio);
}
static inline int irq_to_gpio(unsigned int irq)
{
- return -EINVAL;
+ return -ENOSYS;
}
#endif /* CONFIG_GPIOLIB */
extern struct clk sh7372_pllc2_clk;
extern struct clk sh7372_fsiack_clk;
extern struct clk sh7372_fsibck_clk;
+extern struct clk sh7372_fsidiva_clk;
+extern struct clk sh7372_fsidivb_clk;
#endif /* __ASM_SH7372_H__ */
.bus_num = 0, /* Only one bus on this chip */
.chip_select = 0,
/* Means SPI_CS_HIGH, change if e.g low CS */
- .mode = SPI_MODE_1 | SPI_LSB_FIRST | SPI_LOOP,
+ .mode = SPI_MODE_1 | SPI_LOOP,
},
#endif
};
#define _M68K_IRQFLAGS_H
#include <linux/types.h>
+#ifdef CONFIG_MMU
#include <linux/hardirq.h>
+#endif
#include <linux/preempt.h>
#include <asm/thread_info.h>
#include <asm/entry.h>
extern irqreturn_t arch_timer_interrupt(int irq, void *dummy);
extern void config_BSP(char *command, int len);
+extern void do_IRQ(int irq, struct pt_regs *fp);
#endif /* _M68K_MACHDEP_H */
static void kvm_patch_ins_b(u32 *inst, int addr)
{
-#ifdef CONFIG_RELOCATABLE
+#if defined(CONFIG_RELOCATABLE) && defined(CONFIG_PPC_BOOK3S)
/* On relocatable kernels interrupts handlers and our code
can be in different regions, so we don't patch them */
lwz r3, VCPU_PC(r4)
mtsrr0 r3
lwz r3, VCPU_SHARED(r4)
- lwz r3, VCPU_SHARED_MSR(r3)
+ lwz r3, (VCPU_SHARED_MSR + 4)(r3)
oris r3, r3, KVMPPC_MSR_MASK@h
ori r3, r3, KVMPPC_MSR_MASK@l
mtsrr1 r3
struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
free_page((unsigned long)vcpu->arch.shared);
- kvmppc_e500_tlb_uninit(vcpu_e500);
kvm_vcpu_uninit(vcpu);
+ kvmppc_e500_tlb_uninit(vcpu_e500);
kmem_cache_free(kvm_vcpu_cache, vcpu_e500);
}
switch (ioctl) {
case KVM_PPC_GET_PVINFO: {
struct kvm_ppc_pvinfo pvinfo;
+ memset(&pvinfo, 0, sizeof(pvinfo));
r = kvm_vm_ioctl_get_pvinfo(&pvinfo);
if (copy_to_user(argp, &pvinfo, sizeof(pvinfo))) {
r = -EFAULT;
int i;
/* pause guest execution to avoid concurrent updates */
- local_irq_disable();
mutex_lock(&vcpu->mutex);
vcpu->arch.last_exit_type = 0xDEAD;
vcpu->arch.timing_last_enter.tv64 = 0;
mutex_unlock(&vcpu->mutex);
- local_irq_enable();
}
static void add_exit_timing(struct kvm_vcpu *vcpu, u64 duration, int type)
config CPU_SH2A
bool
select CPU_SH2
+ select UNCACHED_MAPPING
config CPU_SH3
bool
machdir-$(CONFIG_SH_HP6XX) += mach-hp6xx
machdir-$(CONFIG_SH_DREAMCAST) += mach-dreamcast
machdir-$(CONFIG_SH_SH03) += mach-sh03
-machdir-$(CONFIG_SH_SECUREEDGE5410) += mach-snapgear
machdir-$(CONFIG_SH_RTS7751R2D) += mach-r2d
-machdir-$(CONFIG_SH_7751_SYSTEMH) += mach-systemh
-machdir-$(CONFIG_SH_EDOSK7705) += mach-edosk7705
machdir-$(CONFIG_SH_HIGHLANDER) += mach-highlander
machdir-$(CONFIG_SH_MIGOR) += mach-migor
machdir-$(CONFIG_SH_AP325RXA) += mach-ap325rxa
Select 7343 SolutionEngine if configuring for a Hitachi
SH7343 (SH-Mobile 3AS) evaluation board.
-config SH_7751_SYSTEMH
- bool "SystemH7751R"
- depends on CPU_SUBTYPE_SH7751R
- help
- Select SystemH if you are configuring for a Renesas SystemH
- 7751R evaluation board.
-
config SH_HP6XX
bool "HP6XX"
select SYS_SUPPORTS_APM_EMULATION
# Specific board support, not covered by a mach group.
#
obj-$(CONFIG_SH_MAGIC_PANEL_R2) += board-magicpanelr2.o
+obj-$(CONFIG_SH_SECUREEDGE5410) += board-secureedge5410.o
obj-$(CONFIG_SH_SH2007) += board-sh2007.o
obj-$(CONFIG_SH_SH7785LCR) += board-sh7785lcr.o
obj-$(CONFIG_SH_URQUELL) += board-urquell.o
obj-$(CONFIG_SH_SHMIN) += board-shmin.o
+obj-$(CONFIG_SH_EDOSK7705) += board-edosk7705.o
obj-$(CONFIG_SH_EDOSK7760) += board-edosk7760.o
obj-$(CONFIG_SH_ESPT) += board-espt.o
obj-$(CONFIG_SH_POLARIS) += board-polaris.o
--- /dev/null
+/*
+ * arch/sh/boards/renesas/edosk7705/setup.c
+ *
+ * Copyright (C) 2000 Kazumoto Kojima
+ *
+ * Hitachi SolutionEngine Support.
+ *
+ * Modified for edosk7705 development
+ * board by S. Dunn, 2003.
+ */
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/platform_device.h>
+#include <linux/interrupt.h>
+#include <linux/smc91x.h>
+#include <asm/machvec.h>
+#include <asm/sizes.h>
+
+#define SMC_IOBASE 0xA2000000
+#define SMC_IO_OFFSET 0x300
+#define SMC_IOADDR (SMC_IOBASE + SMC_IO_OFFSET)
+
+#define ETHERNET_IRQ 0x09
+
+static void __init sh_edosk7705_init_irq(void)
+{
+ make_imask_irq(ETHERNET_IRQ);
+}
+
+/* eth initialization functions */
+static struct smc91x_platdata smc91x_info = {
+ .flags = SMC91X_USE_16BIT | SMC91X_IO_SHIFT_1 | IORESOURCE_IRQ_LOWLEVEL,
+};
+
+static struct resource smc91x_res[] = {
+ [0] = {
+ .start = SMC_IOADDR,
+ .end = SMC_IOADDR + SZ_32 - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [1] = {
+ .start = ETHERNET_IRQ,
+ .end = ETHERNET_IRQ,
+ .flags = IORESOURCE_IRQ ,
+ }
+};
+
+static struct platform_device smc91x_dev = {
+ .name = "smc91x",
+ .id = -1,
+ .num_resources = ARRAY_SIZE(smc91x_res),
+ .resource = smc91x_res,
+
+ .dev = {
+ .platform_data = &smc91x_info,
+ },
+};
+
+/* platform init code */
+static struct platform_device *edosk7705_devices[] __initdata = {
+ &smc91x_dev,
+};
+
+static int __init init_edosk7705_devices(void)
+{
+ return platform_add_devices(edosk7705_devices,
+ ARRAY_SIZE(edosk7705_devices));
+}
+__initcall(init_edosk7705_devices);
+
+/*
+ * The Machine Vector
+ */
+static struct sh_machine_vector mv_edosk7705 __initmv = {
+ .mv_name = "EDOSK7705",
+ .mv_nr_irqs = 80,
+ .mv_init_irq = sh_edosk7705_init_irq,
+};
/*
- * linux/arch/sh/boards/snapgear/setup.c
- *
* Copyright (C) 2002 David McCullough <davidm@snapgear.com>
* Copyright (C) 2003 Paul Mundt <lethal@linux-sh.org>
*
#include <linux/module.h>
#include <linux/sched.h>
#include <asm/machvec.h>
-#include <mach/snapgear.h>
+#include <mach/secureedge5410.h>
#include <asm/irq.h>
#include <asm/io.h>
#include <cpu/timer.h>
+unsigned short secureedge5410_ioport;
+
/*
* EraseConfig handling functions
*/
-
static irqreturn_t eraseconfig_interrupt(int irq, void *dev_id)
{
- (void)__raw_readb(0xb8000000); /* dummy read */
+ ctrl_delay(); /* dummy read */
printk("SnapGear: erase switch interrupt!\n");
static int __init eraseconfig_init(void)
{
+ unsigned int irq = evt2irq(0x240);
+
printk("SnapGear: EraseConfig init\n");
+
/* Setup "EraseConfig" switch on external IRQ 0 */
- if (request_irq(IRL0_IRQ, eraseconfig_interrupt, IRQF_DISABLED,
+ if (request_irq(irq, eraseconfig_interrupt, IRQF_DISABLED,
"Erase Config", NULL))
printk("SnapGear: failed to register IRQ%d for Reset witch\n",
- IRL0_IRQ);
+ irq);
else
printk("SnapGear: registered EraseConfig switch on IRQ%d\n",
- IRL0_IRQ);
- return(0);
+ irq);
+ return 0;
}
-
module_init(eraseconfig_init);
-/****************************************************************************/
/*
* Initialize IRQ setting
*
* IRL2 = eth1
* IRL3 = crypto
*/
-
static void __init init_snapgear_IRQ(void)
{
printk("Setup SnapGear IRQ/IPR ...\n");
static struct sh_machine_vector mv_snapgear __initmv = {
.mv_name = "SnapGear SecureEdge5410",
.mv_nr_irqs = 72,
-
- .mv_inb = snapgear_inb,
- .mv_inw = snapgear_inw,
- .mv_inl = snapgear_inl,
- .mv_outb = snapgear_outb,
- .mv_outw = snapgear_outw,
- .mv_outl = snapgear_outl,
-
- .mv_inb_p = snapgear_inb_p,
- .mv_inw_p = snapgear_inw,
- .mv_inl_p = snapgear_inl,
- .mv_outb_p = snapgear_outb_p,
- .mv_outw_p = snapgear_outw,
- .mv_outl_p = snapgear_outl,
-
.mv_init_irq = init_snapgear_IRQ,
};
+++ /dev/null
-#
-# Makefile for the EDOSK7705 specific parts of the kernel
-#
-
-obj-y := setup.o io.o
+++ /dev/null
-/*
- * arch/sh/boards/renesas/edosk7705/io.c
- *
- * Copyright (C) 2001 Ian da Silva, Jeremy Siegel
- * Based largely on io_se.c.
- *
- * I/O routines for Hitachi EDOSK7705 board.
- *
- */
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/io.h>
-#include <mach/edosk7705.h>
-#include <asm/addrspace.h>
-
-#define SMC_IOADDR 0xA2000000
-
-/* Map the Ethernet addresses as if it is at 0x300 - 0x320 */
-static unsigned long sh_edosk7705_isa_port2addr(unsigned long port)
-{
- /*
- * SMC91C96 registers are 4 byte aligned rather than the
- * usual 2 byte!
- */
- if (port >= 0x300 && port < 0x320)
- return SMC_IOADDR + ((port - 0x300) * 2);
-
- maybebadio(port);
- return port;
-}
-
-/* Trying to read / write bytes on odd-byte boundaries to the Ethernet
- * registers causes problems. So we bit-shift the value and read / write
- * in 2 byte chunks. Setting the low byte to 0 does not cause problems
- * now as odd byte writes are only made on the bit mask / interrupt
- * register. This may not be the case in future Mar-2003 SJD
- */
-unsigned char sh_edosk7705_inb(unsigned long port)
-{
- if (port >= 0x300 && port < 0x320 && port & 0x01)
- return __raw_readw(port - 1) >> 8;
-
- return __raw_readb(sh_edosk7705_isa_port2addr(port));
-}
-
-void sh_edosk7705_outb(unsigned char value, unsigned long port)
-{
- if (port >= 0x300 && port < 0x320 && port & 0x01) {
- __raw_writew(((unsigned short)value << 8), port - 1);
- return;
- }
-
- __raw_writeb(value, sh_edosk7705_isa_port2addr(port));
-}
-
-void sh_edosk7705_insb(unsigned long port, void *addr, unsigned long count)
-{
- unsigned char *p = addr;
-
- while (count--)
- *p++ = sh_edosk7705_inb(port);
-}
-
-void sh_edosk7705_outsb(unsigned long port, const void *addr, unsigned long count)
-{
- unsigned char *p = (unsigned char *)addr;
-
- while (count--)
- sh_edosk7705_outb(*p++, port);
-}
+++ /dev/null
-/*
- * arch/sh/boards/renesas/edosk7705/setup.c
- *
- * Copyright (C) 2000 Kazumoto Kojima
- *
- * Hitachi SolutionEngine Support.
- *
- * Modified for edosk7705 development
- * board by S. Dunn, 2003.
- */
-#include <linux/init.h>
-#include <linux/irq.h>
-#include <asm/machvec.h>
-#include <mach/edosk7705.h>
-
-static void __init sh_edosk7705_init_irq(void)
-{
- /* This is the Ethernet interrupt */
- make_imask_irq(0x09);
-}
-
-/*
- * The Machine Vector
- */
-static struct sh_machine_vector mv_edosk7705 __initmv = {
- .mv_name = "EDOSK7705",
- .mv_nr_irqs = 80,
-
- .mv_inb = sh_edosk7705_inb,
- .mv_outb = sh_edosk7705_outb,
-
- .mv_insb = sh_edosk7705_insb,
- .mv_outsb = sh_edosk7705_outsb,
-
- .mv_init_irq = sh_edosk7705_init_irq,
-};
/*
* map I/O ports to memory-mapped addresses
*/
-static unsigned long microdev_isa_port2addr(unsigned long offset)
+void __iomem *microdev_ioport_map(unsigned long offset, unsigned int len)
{
unsigned long result;
* Configuration Registers
*/
result = IO_SUPERIO_PHYS + (offset << 1);
-#if 0
- } else if (offset == KBD_DATA_REG || offset == KBD_CNTL_REG ||
- offset == KBD_STATUS_REG) {
- /*
- * SMSC FDC37C93xAPM SuperIO chip
- *
- * PS/2 Keyboard + Mouse (ports 0x60 and 0x64).
- */
- result = IO_SUPERIO_PHYS + (offset << 1);
-#endif
} else if (((offset >= IO_IDE1_BASE) &&
(offset < IO_IDE1_BASE + IO_IDE_EXTENT)) ||
(offset == IO_IDE1_MISC)) {
result = PVR;
}
- return result;
-}
-
-#define PORT2ADDR(x) (microdev_isa_port2addr(x))
-
-static inline void delay(void)
-{
-#if defined(CONFIG_PCI)
- /* System board present, just make a dummy SRAM access. (CS0 will be
- mapped to PCI memory, probably good to avoid it.) */
- __raw_readw(0xa6800000);
-#else
- /* CS0 will be mapped to flash, ROM etc so safe to access it. */
- __raw_readw(0xa0000000);
-#endif
-}
-
-unsigned char microdev_inb(unsigned long port)
-{
-#ifdef CONFIG_PCI
- if (port >= PCIBIOS_MIN_IO)
- return microdev_pci_inb(port);
-#endif
- return *(volatile unsigned char*)PORT2ADDR(port);
-}
-
-unsigned short microdev_inw(unsigned long port)
-{
-#ifdef CONFIG_PCI
- if (port >= PCIBIOS_MIN_IO)
- return microdev_pci_inw(port);
-#endif
- return *(volatile unsigned short*)PORT2ADDR(port);
-}
-
-unsigned int microdev_inl(unsigned long port)
-{
-#ifdef CONFIG_PCI
- if (port >= PCIBIOS_MIN_IO)
- return microdev_pci_inl(port);
-#endif
- return *(volatile unsigned int*)PORT2ADDR(port);
-}
-
-void microdev_outw(unsigned short b, unsigned long port)
-{
-#ifdef CONFIG_PCI
- if (port >= PCIBIOS_MIN_IO) {
- microdev_pci_outw(b, port);
- return;
- }
-#endif
- *(volatile unsigned short*)PORT2ADDR(port) = b;
-}
-
-void microdev_outb(unsigned char b, unsigned long port)
-{
-#ifdef CONFIG_PCI
- if (port >= PCIBIOS_MIN_IO) {
- microdev_pci_outb(b, port);
- return;
- }
-#endif
-
- /*
- * There is a board feature with the current SH4-202 MicroDev in
- * that the 2 byte enables (nBE0 and nBE1) are tied together (and
- * to the Chip Select Line (Ethernet_CS)). Due to this connectivity,
- * it is not possible to safely perform 8-bit writes to the
- * Ethernet registers, as 16-bits will be consumed from the Data
- * lines (corrupting the other byte). Hence, this function is
- * written to implement 16-bit read/modify/write for all byte-wide
- * accesses.
- *
- * Note: there is no problem with byte READS (even or odd).
- *
- * Sean McGoogan - 16th June 2003.
- */
- if ((port >= IO_LAN91C111_BASE) &&
- (port < IO_LAN91C111_BASE + IO_LAN91C111_EXTENT)) {
- /*
- * Then are trying to perform a byte-write to the
- * LAN91C111. This needs special care.
- */
- if (port % 2 == 1) { /* is the port odd ? */
- /* unset bit-0, i.e. make even */
- const unsigned long evenPort = port-1;
- unsigned short word;
-
- /*
- * do a 16-bit read/write to write to 'port',
- * preserving even byte.
- *
- * Even addresses are bits 0-7
- * Odd addresses are bits 8-15
- */
- word = microdev_inw(evenPort);
- word = (word & 0xffu) | (b << 8);
- microdev_outw(word, evenPort);
- } else {
- /* else, we are trying to do an even byte write */
- unsigned short word;
-
- /*
- * do a 16-bit read/write to write to 'port',
- * preserving odd byte.
- *
- * Even addresses are bits 0-7
- * Odd addresses are bits 8-15
- */
- word = microdev_inw(port);
- word = (word & 0xff00u) | (b);
- microdev_outw(word, port);
- }
- } else {
- *(volatile unsigned char*)PORT2ADDR(port) = b;
- }
-}
-
-void microdev_outl(unsigned int b, unsigned long port)
-{
-#ifdef CONFIG_PCI
- if (port >= PCIBIOS_MIN_IO) {
- microdev_pci_outl(b, port);
- return;
- }
-#endif
- *(volatile unsigned int*)PORT2ADDR(port) = b;
-}
-
-unsigned char microdev_inb_p(unsigned long port)
-{
- unsigned char v = microdev_inb(port);
- delay();
- return v;
-}
-
-unsigned short microdev_inw_p(unsigned long port)
-{
- unsigned short v = microdev_inw(port);
- delay();
- return v;
-}
-
-unsigned int microdev_inl_p(unsigned long port)
-{
- unsigned int v = microdev_inl(port);
- delay();
- return v;
-}
-
-void microdev_outb_p(unsigned char b, unsigned long port)
-{
- microdev_outb(b, port);
- delay();
-}
-
-void microdev_outw_p(unsigned short b, unsigned long port)
-{
- microdev_outw(b, port);
- delay();
-}
-
-void microdev_outl_p(unsigned int b, unsigned long port)
-{
- microdev_outl(b, port);
- delay();
-}
-
-void microdev_insb(unsigned long port, void *buffer, unsigned long count)
-{
- volatile unsigned char *port_addr;
- unsigned char *buf = buffer;
-
- port_addr = (volatile unsigned char *)PORT2ADDR(port);
-
- while (count--)
- *buf++ = *port_addr;
-}
-
-void microdev_insw(unsigned long port, void *buffer, unsigned long count)
-{
- volatile unsigned short *port_addr;
- unsigned short *buf = buffer;
-
- port_addr = (volatile unsigned short *)PORT2ADDR(port);
-
- while (count--)
- *buf++ = *port_addr;
-}
-
-void microdev_insl(unsigned long port, void *buffer, unsigned long count)
-{
- volatile unsigned long *port_addr;
- unsigned int *buf = buffer;
-
- port_addr = (volatile unsigned long *)PORT2ADDR(port);
-
- while (count--)
- *buf++ = *port_addr;
-}
-
-void microdev_outsb(unsigned long port, const void *buffer, unsigned long count)
-{
- volatile unsigned char *port_addr;
- const unsigned char *buf = buffer;
-
- port_addr = (volatile unsigned char *)PORT2ADDR(port);
-
- while (count--)
- *port_addr = *buf++;
-}
-
-void microdev_outsw(unsigned long port, const void *buffer, unsigned long count)
-{
- volatile unsigned short *port_addr;
- const unsigned short *buf = buffer;
-
- port_addr = (volatile unsigned short *)PORT2ADDR(port);
-
- while (count--)
- *port_addr = *buf++;
-}
-
-void microdev_outsl(unsigned long port, const void *buffer, unsigned long count)
-{
- volatile unsigned long *port_addr;
- const unsigned int *buf = buffer;
-
- port_addr = (volatile unsigned long *)PORT2ADDR(port);
-
- while (count--)
- *port_addr = *buf++;
+ return (void __iomem *)result;
}
static struct sh_machine_vector mv_sh4202_microdev __initmv = {
.mv_name = "SH4-202 MicroDev",
.mv_nr_irqs = 72,
-
- .mv_inb = microdev_inb,
- .mv_inw = microdev_inw,
- .mv_inl = microdev_inl,
- .mv_outb = microdev_outb,
- .mv_outw = microdev_outw,
- .mv_outl = microdev_outl,
-
- .mv_inb_p = microdev_inb_p,
- .mv_inw_p = microdev_inw_p,
- .mv_inl_p = microdev_inl_p,
- .mv_outb_p = microdev_outb_p,
- .mv_outw_p = microdev_outw_p,
- .mv_outl_p = microdev_outl_p,
-
- .mv_insb = microdev_insb,
- .mv_insw = microdev_insw,
- .mv_insl = microdev_insl,
- .mv_outsb = microdev_outsb,
- .mv_outsw = microdev_outsw,
- .mv_outsl = microdev_outsl,
-
+ .mv_ioport_map = microdev_ioport_map,
.mv_init_irq = init_microdev_irq,
};
# Makefile for the 7206 SolutionEngine specific parts of the kernel
#
-obj-y := setup.o io.o irq.o
+obj-y := setup.o irq.o
+++ /dev/null
-/* $Id: io.c,v 1.5 2004/02/22 23:08:43 kkojima Exp $
- *
- * linux/arch/sh/boards/se/7206/io.c
- *
- * Copyright (C) 2006 Yoshinori Sato
- *
- * I/O routine for Hitachi 7206 SolutionEngine.
- *
- */
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <asm/io.h>
-#include <mach-se/mach/se7206.h>
-
-
-static inline void delay(void)
-{
- __raw_readw(0x20000000); /* P2 ROM Area */
-}
-
-/* MS7750 requires special versions of in*, out* routines, since
- PC-like io ports are located at upper half byte of 16-bit word which
- can be accessed only with 16-bit wide. */
-
-static inline volatile __u16 *
-port2adr(unsigned int port)
-{
- if (port >= 0x2000 && port < 0x2020)
- return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
- else if (port >= 0x300 && port < 0x310)
- return (volatile __u16 *) (PA_SMSC + (port - 0x300));
-
- return (volatile __u16 *)port;
-}
-
-unsigned char se7206_inb(unsigned long port)
-{
- return (*port2adr(port)) & 0xff;
-}
-
-unsigned char se7206_inb_p(unsigned long port)
-{
- unsigned long v;
-
- v = (*port2adr(port)) & 0xff;
- delay();
- return v;
-}
-
-unsigned short se7206_inw(unsigned long port)
-{
- return *port2adr(port);
-}
-
-void se7206_outb(unsigned char value, unsigned long port)
-{
- *(port2adr(port)) = value;
-}
-
-void se7206_outb_p(unsigned char value, unsigned long port)
-{
- *(port2adr(port)) = value;
- delay();
-}
-
-void se7206_outw(unsigned short value, unsigned long port)
-{
- *port2adr(port) = value;
-}
-
-void se7206_insb(unsigned long port, void *addr, unsigned long count)
-{
- volatile __u16 *p = port2adr(port);
- __u8 *ap = addr;
-
- while (count--)
- *ap++ = *p;
-}
-
-void se7206_insw(unsigned long port, void *addr, unsigned long count)
-{
- volatile __u16 *p = port2adr(port);
- __u16 *ap = addr;
- while (count--)
- *ap++ = *p;
-}
-
-void se7206_outsb(unsigned long port, const void *addr, unsigned long count)
-{
- volatile __u16 *p = port2adr(port);
- const __u8 *ap = addr;
-
- while (count--)
- *p = *ap++;
-}
-
-void se7206_outsw(unsigned long port, const void *addr, unsigned long count)
-{
- volatile __u16 *p = port2adr(port);
- const __u16 *ap = addr;
- while (count--)
- *p = *ap++;
-}
make_se7206_irq(IRQ0_IRQ); /* SMC91C111 */
make_se7206_irq(IRQ1_IRQ); /* ATA */
make_se7206_irq(IRQ3_IRQ); /* SLOT / PCM */
- __raw_writew(inw(INTC_ICR1) | 0x000b ,INTC_ICR1 ) ; /* ICR1 */
+
+ __raw_writew(__raw_readw(INTC_ICR1) | 0x000b, INTC_ICR); /* ICR1 */
/* FPGA System register setup*/
__raw_writew(0x0000,INTSTS0); /* Clear INTSTS0 */
__raw_writew(0x0000,INTSTS1); /* Clear INTSTS1 */
+
/* IRQ0=LAN, IRQ1=ATA, IRQ3=SLT,PCM */
__raw_writew(0x0001,INTSEL);
}
static struct sh_machine_vector mv_se __initmv = {
.mv_name = "SolutionEngine",
.mv_nr_irqs = 256,
- .mv_inb = se7206_inb,
- .mv_inw = se7206_inw,
- .mv_outb = se7206_outb,
- .mv_outw = se7206_outw,
-
- .mv_inb_p = se7206_inb_p,
- .mv_inw_p = se7206_inw,
- .mv_outb_p = se7206_outb_p,
- .mv_outw_p = se7206_outw,
-
- .mv_insb = se7206_insb,
- .mv_insw = se7206_insw,
- .mv_outsb = se7206_outsb,
- .mv_outsw = se7206_outsw,
-
.mv_init_irq = init_se7206_IRQ,
};
# Makefile for the 770x SolutionEngine specific parts of the kernel
#
-obj-y := setup.o io.o irq.o
+obj-y := setup.o irq.o
+++ /dev/null
-/*
- * Copyright (C) 2000 Kazumoto Kojima
- *
- * I/O routine for Hitachi SolutionEngine.
- */
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <asm/io.h>
-#include <mach-se/mach/se.h>
-
-/* MS7750 requires special versions of in*, out* routines, since
- PC-like io ports are located at upper half byte of 16-bit word which
- can be accessed only with 16-bit wide. */
-
-static inline volatile __u16 *
-port2adr(unsigned int port)
-{
- if (port & 0xff000000)
- return ( volatile __u16 *) port;
- if (port >= 0x2000)
- return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
- else if (port >= 0x1000)
- return (volatile __u16 *) (PA_83902 + (port << 1));
- else
- return (volatile __u16 *) (PA_SUPERIO + (port << 1));
-}
-
-static inline int
-shifted_port(unsigned long port)
-{
- /* For IDE registers, value is not shifted */
- if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
- return 0;
- else
- return 1;
-}
-
-unsigned char se_inb(unsigned long port)
-{
- if (shifted_port(port))
- return (*port2adr(port) >> 8);
- else
- return (*port2adr(port))&0xff;
-}
-
-unsigned char se_inb_p(unsigned long port)
-{
- unsigned long v;
-
- if (shifted_port(port))
- v = (*port2adr(port) >> 8);
- else
- v = (*port2adr(port))&0xff;
- ctrl_delay();
- return v;
-}
-
-unsigned short se_inw(unsigned long port)
-{
- if (port >= 0x2000)
- return *port2adr(port);
- else
- maybebadio(port);
- return 0;
-}
-
-unsigned int se_inl(unsigned long port)
-{
- maybebadio(port);
- return 0;
-}
-
-void se_outb(unsigned char value, unsigned long port)
-{
- if (shifted_port(port))
- *(port2adr(port)) = value << 8;
- else
- *(port2adr(port)) = value;
-}
-
-void se_outb_p(unsigned char value, unsigned long port)
-{
- if (shifted_port(port))
- *(port2adr(port)) = value << 8;
- else
- *(port2adr(port)) = value;
- ctrl_delay();
-}
-
-void se_outw(unsigned short value, unsigned long port)
-{
- if (port >= 0x2000)
- *port2adr(port) = value;
- else
- maybebadio(port);
-}
-
-void se_outl(unsigned int value, unsigned long port)
-{
- maybebadio(port);
-}
-
-void se_insb(unsigned long port, void *addr, unsigned long count)
-{
- volatile __u16 *p = port2adr(port);
- __u8 *ap = addr;
-
- if (shifted_port(port)) {
- while (count--)
- *ap++ = *p >> 8;
- } else {
- while (count--)
- *ap++ = *p;
- }
-}
-
-void se_insw(unsigned long port, void *addr, unsigned long count)
-{
- volatile __u16 *p = port2adr(port);
- __u16 *ap = addr;
- while (count--)
- *ap++ = *p;
-}
-
-void se_insl(unsigned long port, void *addr, unsigned long count)
-{
- maybebadio(port);
-}
-
-void se_outsb(unsigned long port, const void *addr, unsigned long count)
-{
- volatile __u16 *p = port2adr(port);
- const __u8 *ap = addr;
-
- if (shifted_port(port)) {
- while (count--)
- *p = *ap++ << 8;
- } else {
- while (count--)
- *p = *ap++;
- }
-}
-
-void se_outsw(unsigned long port, const void *addr, unsigned long count)
-{
- volatile __u16 *p = port2adr(port);
- const __u16 *ap = addr;
-
- while (count--)
- *p = *ap++;
-}
-
-void se_outsl(unsigned long port, const void *addr, unsigned long count)
-{
- maybebadio(port);
-}
#elif defined(CONFIG_CPU_SUBTYPE_SH7710) || defined(CONFIG_CPU_SUBTYPE_SH7712)
.mv_nr_irqs = 104,
#endif
-
- .mv_inb = se_inb,
- .mv_inw = se_inw,
- .mv_inl = se_inl,
- .mv_outb = se_outb,
- .mv_outw = se_outw,
- .mv_outl = se_outl,
-
- .mv_inb_p = se_inb_p,
- .mv_inw_p = se_inw,
- .mv_inl_p = se_inl,
- .mv_outb_p = se_outb_p,
- .mv_outw_p = se_outw,
- .mv_outl_p = se_outl,
-
- .mv_insb = se_insb,
- .mv_insw = se_insw,
- .mv_insl = se_insl,
- .mv_outsb = se_outsb,
- .mv_outsw = se_outsw,
- .mv_outsl = se_outsl,
-
.mv_init_irq = init_se_IRQ,
};
# Makefile for the 7751 SolutionEngine specific parts of the kernel
#
-obj-y := setup.o io.o irq.o
+obj-y := setup.o irq.o
+++ /dev/null
-/*
- * Copyright (C) 2001 Ian da Silva, Jeremy Siegel
- * Based largely on io_se.c.
- *
- * I/O routine for Hitachi 7751 SolutionEngine.
- *
- * Initial version only to support LAN access; some
- * placeholder code from io_se.c left in with the
- * expectation of later SuperIO and PCMCIA access.
- */
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/pci.h>
-#include <asm/io.h>
-#include <mach-se/mach/se7751.h>
-#include <asm/addrspace.h>
-
-static inline volatile u16 *port2adr(unsigned int port)
-{
- if (port >= 0x2000)
- return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
- maybebadio((unsigned long)port);
- return (volatile __u16*)port;
-}
-
-/*
- * General outline: remap really low stuff [eventually] to SuperIO,
- * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
- * is mapped through the PCI IO window. Stuff with high bits (PXSEG)
- * should be way beyond the window, and is used w/o translation for
- * compatibility.
- */
-unsigned char sh7751se_inb(unsigned long port)
-{
- if (PXSEG(port))
- return *(volatile unsigned char *)port;
- else
- return (*port2adr(port)) & 0xff;
-}
-
-unsigned char sh7751se_inb_p(unsigned long port)
-{
- unsigned char v;
-
- if (PXSEG(port))
- v = *(volatile unsigned char *)port;
- else
- v = (*port2adr(port)) & 0xff;
- ctrl_delay();
- return v;
-}
-
-unsigned short sh7751se_inw(unsigned long port)
-{
- if (PXSEG(port))
- return *(volatile unsigned short *)port;
- else if (port >= 0x2000)
- return *port2adr(port);
- else
- maybebadio(port);
- return 0;
-}
-
-unsigned int sh7751se_inl(unsigned long port)
-{
- if (PXSEG(port))
- return *(volatile unsigned long *)port;
- else if (port >= 0x2000)
- return *port2adr(port);
- else
- maybebadio(port);
- return 0;
-}
-
-void sh7751se_outb(unsigned char value, unsigned long port)
-{
-
- if (PXSEG(port))
- *(volatile unsigned char *)port = value;
- else
- *(port2adr(port)) = value;
-}
-
-void sh7751se_outb_p(unsigned char value, unsigned long port)
-{
- if (PXSEG(port))
- *(volatile unsigned char *)port = value;
- else
- *(port2adr(port)) = value;
- ctrl_delay();
-}
-
-void sh7751se_outw(unsigned short value, unsigned long port)
-{
- if (PXSEG(port))
- *(volatile unsigned short *)port = value;
- else if (port >= 0x2000)
- *port2adr(port) = value;
- else
- maybebadio(port);
-}
-
-void sh7751se_outl(unsigned int value, unsigned long port)
-{
- if (PXSEG(port))
- *(volatile unsigned long *)port = value;
- else
- maybebadio(port);
-}
-
-void sh7751se_insl(unsigned long port, void *addr, unsigned long count)
-{
- maybebadio(port);
-}
-
-void sh7751se_outsl(unsigned long port, const void *addr, unsigned long count)
-{
- maybebadio(port);
-}
static struct sh_machine_vector mv_7751se __initmv = {
.mv_name = "7751 SolutionEngine",
.mv_nr_irqs = 72,
-
- .mv_inb = sh7751se_inb,
- .mv_inw = sh7751se_inw,
- .mv_inl = sh7751se_inl,
- .mv_outb = sh7751se_outb,
- .mv_outw = sh7751se_outw,
- .mv_outl = sh7751se_outl,
-
- .mv_inb_p = sh7751se_inb_p,
- .mv_inw_p = sh7751se_inw,
- .mv_inl_p = sh7751se_inl,
- .mv_outb_p = sh7751se_outb_p,
- .mv_outw_p = sh7751se_outw,
- .mv_outl_p = sh7751se_outl,
-
- .mv_insl = sh7751se_insl,
- .mv_outsl = sh7751se_outsl,
-
.mv_init_irq = init_7751se_IRQ,
};
+++ /dev/null
-#
-# Makefile for the SnapGear specific parts of the kernel
-#
-
-obj-y := setup.o io.o
+++ /dev/null
-/*
- * Copyright (C) 2002 David McCullough <davidm@snapgear.com>
- * Copyright (C) 2001 Ian da Silva, Jeremy Siegel
- * Based largely on io_se.c.
- *
- * I/O routine for Hitachi 7751 SolutionEngine.
- *
- * Initial version only to support LAN access; some
- * placeholder code from io_se.c left in with the
- * expectation of later SuperIO and PCMCIA access.
- */
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/pci.h>
-#include <asm/io.h>
-#include <asm/addrspace.h>
-
-#ifdef CONFIG_SH_SECUREEDGE5410
-unsigned short secureedge5410_ioport;
-#endif
-
-static inline volatile __u16 *port2adr(unsigned int port)
-{
- maybebadio((unsigned long)port);
- return (volatile __u16*)port;
-}
-
-/*
- * General outline: remap really low stuff [eventually] to SuperIO,
- * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
- * is mapped through the PCI IO window. Stuff with high bits (PXSEG)
- * should be way beyond the window, and is used w/o translation for
- * compatibility.
- */
-unsigned char snapgear_inb(unsigned long port)
-{
- if (PXSEG(port))
- return *(volatile unsigned char *)port;
- else
- return (*port2adr(port)) & 0xff;
-}
-
-unsigned char snapgear_inb_p(unsigned long port)
-{
- unsigned char v;
-
- if (PXSEG(port))
- v = *(volatile unsigned char *)port;
- else
- v = (*port2adr(port))&0xff;
- ctrl_delay();
- return v;
-}
-
-unsigned short snapgear_inw(unsigned long port)
-{
- if (PXSEG(port))
- return *(volatile unsigned short *)port;
- else if (port >= 0x2000)
- return *port2adr(port);
- else
- maybebadio(port);
- return 0;
-}
-
-unsigned int snapgear_inl(unsigned long port)
-{
- if (PXSEG(port))
- return *(volatile unsigned long *)port;
- else if (port >= 0x2000)
- return *port2adr(port);
- else
- maybebadio(port);
- return 0;
-}
-
-void snapgear_outb(unsigned char value, unsigned long port)
-{
-
- if (PXSEG(port))
- *(volatile unsigned char *)port = value;
- else
- *(port2adr(port)) = value;
-}
-
-void snapgear_outb_p(unsigned char value, unsigned long port)
-{
- if (PXSEG(port))
- *(volatile unsigned char *)port = value;
- else
- *(port2adr(port)) = value;
- ctrl_delay();
-}
-
-void snapgear_outw(unsigned short value, unsigned long port)
-{
- if (PXSEG(port))
- *(volatile unsigned short *)port = value;
- else if (port >= 0x2000)
- *port2adr(port) = value;
- else
- maybebadio(port);
-}
-
-void snapgear_outl(unsigned int value, unsigned long port)
-{
- if (PXSEG(port))
- *(volatile unsigned long *)port = value;
- else
- maybebadio(port);
-}
-
-void snapgear_insl(unsigned long port, void *addr, unsigned long count)
-{
- maybebadio(port);
-}
-
-void snapgear_outsl(unsigned long port, const void *addr, unsigned long count)
-{
- maybebadio(port);
-}
+++ /dev/null
-#
-# Makefile for the SystemH specific parts of the kernel
-#
-
-obj-y := setup.o irq.o io.o
-
-# XXX: This wants to be consolidated in arch/sh/drivers/pci, and more
-# importantly, with the generic sh7751_pcic_init() code. For now, we'll
-# just abuse the hell out of kbuild, because we can..
-
-obj-$(CONFIG_PCI) += pci.o
-pci-y := ../../se/7751/pci.o
-
+++ /dev/null
-/*
- * linux/arch/sh/boards/renesas/systemh/io.c
- *
- * Copyright (C) 2001 Ian da Silva, Jeremy Siegel
- * Based largely on io_se.c.
- *
- * I/O routine for Hitachi 7751 Systemh.
- */
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/pci.h>
-#include <mach/systemh7751.h>
-#include <asm/addrspace.h>
-#include <asm/io.h>
-
-#define ETHER_IOMAP(adr) (0xB3000000 + (adr)) /*map to 16bits access area
- of smc lan chip*/
-static inline volatile __u16 *
-port2adr(unsigned int port)
-{
- if (port >= 0x2000)
- return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
- maybebadio((unsigned long)port);
- return (volatile __u16*)port;
-}
-
-/*
- * General outline: remap really low stuff [eventually] to SuperIO,
- * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
- * is mapped through the PCI IO window. Stuff with high bits (PXSEG)
- * should be way beyond the window, and is used w/o translation for
- * compatibility.
- */
-unsigned char sh7751systemh_inb(unsigned long port)
-{
- if (PXSEG(port))
- return *(volatile unsigned char *)port;
- else if (port <= 0x3F1)
- return *(volatile unsigned char *)ETHER_IOMAP(port);
- else
- return (*port2adr(port))&0xff;
-}
-
-unsigned char sh7751systemh_inb_p(unsigned long port)
-{
- unsigned char v;
-
- if (PXSEG(port))
- v = *(volatile unsigned char *)port;
- else if (port <= 0x3F1)
- v = *(volatile unsigned char *)ETHER_IOMAP(port);
- else
- v = (*port2adr(port))&0xff;
- ctrl_delay();
- return v;
-}
-
-unsigned short sh7751systemh_inw(unsigned long port)
-{
- if (PXSEG(port))
- return *(volatile unsigned short *)port;
- else if (port >= 0x2000)
- return *port2adr(port);
- else if (port <= 0x3F1)
- return *(volatile unsigned int *)ETHER_IOMAP(port);
- else
- maybebadio(port);
- return 0;
-}
-
-unsigned int sh7751systemh_inl(unsigned long port)
-{
- if (PXSEG(port))
- return *(volatile unsigned long *)port;
- else if (port >= 0x2000)
- return *port2adr(port);
- else if (port <= 0x3F1)
- return *(volatile unsigned int *)ETHER_IOMAP(port);
- else
- maybebadio(port);
- return 0;
-}
-
-void sh7751systemh_outb(unsigned char value, unsigned long port)
-{
-
- if (PXSEG(port))
- *(volatile unsigned char *)port = value;
- else if (port <= 0x3F1)
- *(volatile unsigned char *)ETHER_IOMAP(port) = value;
- else
- *(port2adr(port)) = value;
-}
-
-void sh7751systemh_outb_p(unsigned char value, unsigned long port)
-{
- if (PXSEG(port))
- *(volatile unsigned char *)port = value;
- else if (port <= 0x3F1)
- *(volatile unsigned char *)ETHER_IOMAP(port) = value;
- else
- *(port2adr(port)) = value;
- ctrl_delay();
-}
-
-void sh7751systemh_outw(unsigned short value, unsigned long port)
-{
- if (PXSEG(port))
- *(volatile unsigned short *)port = value;
- else if (port >= 0x2000)
- *port2adr(port) = value;
- else if (port <= 0x3F1)
- *(volatile unsigned short *)ETHER_IOMAP(port) = value;
- else
- maybebadio(port);
-}
-
-void sh7751systemh_outl(unsigned int value, unsigned long port)
-{
- if (PXSEG(port))
- *(volatile unsigned long *)port = value;
- else
- maybebadio(port);
-}
-
-void sh7751systemh_insb(unsigned long port, void *addr, unsigned long count)
-{
- unsigned char *p = addr;
- while (count--) *p++ = sh7751systemh_inb(port);
-}
-
-void sh7751systemh_insw(unsigned long port, void *addr, unsigned long count)
-{
- unsigned short *p = addr;
- while (count--) *p++ = sh7751systemh_inw(port);
-}
-
-void sh7751systemh_insl(unsigned long port, void *addr, unsigned long count)
-{
- maybebadio(port);
-}
-
-void sh7751systemh_outsb(unsigned long port, const void *addr, unsigned long count)
-{
- unsigned char *p = (unsigned char*)addr;
- while (count--) sh7751systemh_outb(*p++, port);
-}
-
-void sh7751systemh_outsw(unsigned long port, const void *addr, unsigned long count)
-{
- unsigned short *p = (unsigned short*)addr;
- while (count--) sh7751systemh_outw(*p++, port);
-}
-
-void sh7751systemh_outsl(unsigned long port, const void *addr, unsigned long count)
-{
- maybebadio(port);
-}
+++ /dev/null
-/*
- * linux/arch/sh/boards/renesas/systemh/irq.c
- *
- * Copyright (C) 2000 Kazumoto Kojima
- *
- * Hitachi SystemH Support.
- *
- * Modified for 7751 SystemH by
- * Jonathan Short.
- */
-
-#include <linux/init.h>
-#include <linux/irq.h>
-#include <linux/interrupt.h>
-#include <linux/io.h>
-
-#include <mach/systemh7751.h>
-#include <asm/smc37c93x.h>
-
-/* address of external interrupt mask register
- * address must be set prior to use these (maybe in init_XXX_irq())
- * XXX : is it better to use .config than specifying it in code? */
-static unsigned long *systemh_irq_mask_register = (unsigned long *)0xB3F10004;
-static unsigned long *systemh_irq_request_register = (unsigned long *)0xB3F10000;
-
-static void disable_systemh_irq(struct irq_data *data)
-{
- unsigned long val, mask = 0x01 << 1;
-
- /* Clear the "irq"th bit in the mask and set it in the request */
- val = __raw_readl((unsigned long)systemh_irq_mask_register);
- val &= ~mask;
- __raw_writel(val, (unsigned long)systemh_irq_mask_register);
-
- val = __raw_readl((unsigned long)systemh_irq_request_register);
- val |= mask;
- __raw_writel(val, (unsigned long)systemh_irq_request_register);
-}
-
-static void enable_systemh_irq(struct irq_data *data)
-{
- unsigned long val, mask = 0x01 << 1;
-
- /* Set "irq"th bit in the mask register */
- val = __raw_readl((unsigned long)systemh_irq_mask_register);
- val |= mask;
- __raw_writel(val, (unsigned long)systemh_irq_mask_register);
-}
-
-static struct irq_chip systemh_irq_type = {
- .name = "SystemH Register",
- .irq_unmask = enable_systemh_irq,
- .irq_mask = disable_systemh_irq,
-};
-
-void make_systemh_irq(unsigned int irq)
-{
- disable_irq_nosync(irq);
- set_irq_chip_and_handler(irq, &systemh_irq_type, handle_level_irq);
- disable_systemh_irq(irq_get_irq_data(irq));
-}
+++ /dev/null
-/*
- * linux/arch/sh/boards/renesas/systemh/setup.c
- *
- * Copyright (C) 2000 Kazumoto Kojima
- * Copyright (C) 2003 Paul Mundt
- *
- * Hitachi SystemH Support.
- *
- * Modified for 7751 SystemH by Jonathan Short.
- *
- * Rewritten for 2.6 by Paul Mundt.
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License. See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#include <linux/init.h>
-#include <asm/machvec.h>
-#include <mach/systemh7751.h>
-
-extern void make_systemh_irq(unsigned int irq);
-
-/*
- * Initialize IRQ setting
- */
-static void __init sh7751systemh_init_irq(void)
-{
- make_systemh_irq(0xb); /* Ethernet interrupt */
-}
-
-static struct sh_machine_vector mv_7751systemh __initmv = {
- .mv_name = "7751 SystemH",
- .mv_nr_irqs = 72,
-
- .mv_inb = sh7751systemh_inb,
- .mv_inw = sh7751systemh_inw,
- .mv_inl = sh7751systemh_inl,
- .mv_outb = sh7751systemh_outb,
- .mv_outw = sh7751systemh_outw,
- .mv_outl = sh7751systemh_outl,
-
- .mv_inb_p = sh7751systemh_inb_p,
- .mv_inw_p = sh7751systemh_inw,
- .mv_inl_p = sh7751systemh_inl,
- .mv_outb_p = sh7751systemh_outb_p,
- .mv_outw_p = sh7751systemh_outw,
- .mv_outl_p = sh7751systemh_outl,
-
- .mv_insb = sh7751systemh_insb,
- .mv_insw = sh7751systemh_insw,
- .mv_insl = sh7751systemh_insl,
- .mv_outsb = sh7751systemh_outsb,
- .mv_outsw = sh7751systemh_outsw,
- .mv_outsl = sh7751systemh_outsl,
-
- .mv_init_irq = sh7751systemh_init_irq,
-};
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_BLK_DEV_INITRD=y
-# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
-# CONFIG_SYSCTL_SYSCALL is not set
-# CONFIG_HOTPLUG is not set
-CONFIG_SLAB=y
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_BLK_DEV_BSG is not set
-CONFIG_CPU_SUBTYPE_SH7751R=y
-CONFIG_MEMORY_START=0x0c000000
-CONFIG_MEMORY_SIZE=0x00400000
-CONFIG_FLATMEM_MANUAL=y
-CONFIG_SH_7751_SYSTEMH=y
-CONFIG_PREEMPT=y
-# CONFIG_STANDALONE is not set
-CONFIG_BLK_DEV_RAM=y
-CONFIG_BLK_DEV_RAM_SIZE=1024
-# CONFIG_INPUT is not set
-# CONFIG_SERIO_SERPORT is not set
-# CONFIG_VT is not set
-CONFIG_HW_RANDOM=y
-CONFIG_PROC_KCORE=y
-CONFIG_TMPFS=y
-CONFIG_CRAMFS=y
-CONFIG_ROMFS_FS=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
/*
* These will never work in 32-bit, don't even bother.
*/
-#define P1SEGADDR(a) __futile_remapping_attempt
-#define P2SEGADDR(a) __futile_remapping_attempt
-#define P3SEGADDR(a) __futile_remapping_attempt
-#define P4SEGADDR(a) __futile_remapping_attempt
+#define P1SEGADDR(a) ({ (void)(a); BUG(); NULL; })
+#define P2SEGADDR(a) ({ (void)(a); BUG(); NULL; })
+#define P3SEGADDR(a) ({ (void)(a); BUG(); NULL; })
+#define P4SEGADDR(a) ({ (void)(a); BUG(); NULL; })
#endif
#endif /* P1SEG */
#define PHYS_ADDR_MASK29 0x1fffffff
#define PHYS_ADDR_MASK32 0xffffffff
-#ifdef CONFIG_PMB
static inline unsigned long phys_addr_mask(void)
{
/* Is the MMU in 29bit mode? */
return PHYS_ADDR_MASK32;
}
-#elif defined(CONFIG_32BIT)
-static inline unsigned long phys_addr_mask(void)
-{
- return PHYS_ADDR_MASK32;
-}
-#else
-static inline unsigned long phys_addr_mask(void)
-{
- return PHYS_ADDR_MASK29;
-}
-#endif
#define PTE_PHYS_MASK (phys_addr_mask() & PAGE_MASK)
#define PTE_FLAGS_MASK (~(PTE_PHYS_MASK) << PAGE_SHIFT)
#include <linux/compiler.h>
#include <linux/linkage.h>
#include <asm/types.h>
+#include <asm/uncached.h>
#define AT_VECTOR_SIZE_ARCH 5 /* entries in ARCH_DLINFO */
#define instruction_size(insn) (4)
#endif
-extern unsigned long cached_to_uncached;
-extern unsigned long uncached_size;
-
void per_cpu_trap_init(void);
void default_idle(void);
void cpu_idle_wait(void);
__restore_dsp(prev); \
} while (0)
-/*
- * Jump to uncached area.
- * When handling TLB or caches, we need to do it from an uncached area.
- */
-#define jump_to_uncached() \
-do { \
- unsigned long __dummy; \
- \
- __asm__ __volatile__( \
- "mova 1f, %0\n\t" \
- "add %1, %0\n\t" \
- "jmp @%0\n\t" \
- " nop\n\t" \
- ".balign 4\n" \
- "1:" \
- : "=&z" (__dummy) \
- : "r" (cached_to_uncached)); \
-} while (0)
-
-/*
- * Back to cached area.
- */
-#define back_to_cached() \
-do { \
- unsigned long __dummy; \
- ctrl_barrier(); \
- __asm__ __volatile__( \
- "mov.l 1f, %0\n\t" \
- "jmp @%0\n\t" \
- " nop\n\t" \
- ".balign 4\n" \
- "1: .long 2f\n" \
- "2:" \
- : "=&r" (__dummy)); \
-} while (0)
-
#ifdef CONFIG_CPU_HAS_SR_RB
#define lookup_exception_vector() \
({ \
&next->thread); \
} while (0)
-#define jump_to_uncached() do { } while (0)
-#define back_to_cached() do { } while (0)
-
#define __icbi(addr) __asm__ __volatile__ ( "icbi %0, 0\n\t" : : "r" (addr))
#define __ocbp(addr) __asm__ __volatile__ ( "ocbp %0, 0\n\t" : : "r" (addr))
#define __ocbi(addr) __asm__ __volatile__ ( "ocbi %0, 0\n\t" : : "r" (addr))
#include <linux/bug.h>
#ifdef CONFIG_UNCACHED_MAPPING
+extern unsigned long cached_to_uncached;
+extern unsigned long uncached_size;
extern unsigned long uncached_start, uncached_end;
extern int virt_addr_uncached(unsigned long kaddr);
extern void uncached_init(void);
extern void uncached_resize(unsigned long size);
+
+/*
+ * Jump to uncached area.
+ * When handling TLB or caches, we need to do it from an uncached area.
+ */
+#define jump_to_uncached() \
+do { \
+ unsigned long __dummy; \
+ \
+ __asm__ __volatile__( \
+ "mova 1f, %0\n\t" \
+ "add %1, %0\n\t" \
+ "jmp @%0\n\t" \
+ " nop\n\t" \
+ ".balign 4\n" \
+ "1:" \
+ : "=&z" (__dummy) \
+ : "r" (cached_to_uncached)); \
+} while (0)
+
+/*
+ * Back to cached area.
+ */
+#define back_to_cached() \
+do { \
+ unsigned long __dummy; \
+ ctrl_barrier(); \
+ __asm__ __volatile__( \
+ "mov.l 1f, %0\n\t" \
+ "jmp @%0\n\t" \
+ " nop\n\t" \
+ ".balign 4\n" \
+ "1: .long 2f\n" \
+ "2:" \
+ : "=&r" (__dummy)); \
+} while (0)
#else
#define virt_addr_uncached(kaddr) (0)
#define uncached_init() do { } while (0)
#define uncached_resize(size) BUG()
+#define jump_to_uncached() do { } while (0)
+#define back_to_cached() do { } while (0)
#endif
#endif /* __ASM_SH_UNCACHED_H */
+++ /dev/null
-#ifndef __ASM_SH_EDOSK7705_H
-#define __ASM_SH_EDOSK7705_H
-
-#define __IO_PREFIX sh_edosk7705
-#include <asm/io_generic.h>
-
-#endif /* __ASM_SH_EDOSK7705_H */
#define __IO_PREFIX microdev
#include <asm/io_generic.h>
-#if defined(CONFIG_PCI)
-unsigned char microdev_pci_inb(unsigned long port);
-unsigned short microdev_pci_inw(unsigned long port);
-unsigned long microdev_pci_inl(unsigned long port);
-void microdev_pci_outb(unsigned char data, unsigned long port);
-void microdev_pci_outw(unsigned short data, unsigned long port);
-void microdev_pci_outl(unsigned long data, unsigned long port);
-#endif
-
#endif /* __ASM_SH_MICRODEV_H */
#ifndef _ASM_SH_IO_SNAPGEAR_H
#define _ASM_SH_IO_SNAPGEAR_H
-#if defined(CONFIG_CPU_SH4)
-/*
- * The external interrupt lines, these take up ints 0 - 15 inclusive
- * depending on the priority for the interrupt. In fact the priority
- * is the interrupt :-)
- */
-
-#define IRL0_IRQ 2
-#define IRL0_PRIORITY 13
-
-#define IRL1_IRQ 5
-#define IRL1_PRIORITY 10
-
-#define IRL2_IRQ 8
-#define IRL2_PRIORITY 7
-
-#define IRL3_IRQ 11
-#define IRL3_PRIORITY 4
-#endif
-
#define __IO_PREFIX snapgear
#include <asm/io_generic.h>
-#ifdef CONFIG_SH_SECUREEDGE5410
/*
* We need to remember what was written to the ioport as some bits
* are shared with other functions and you cannot read back what was
((secureedge5410_ioport & ~(mask)) | ((val) & (mask)))))
#define SECUREEDGE_READ_IOPORT() \
((*SECUREEDGE_IOPORT_ADDR&0x0817) | (secureedge5410_ioport&~0x0817))
-#endif
#endif /* _ASM_SH_IO_SNAPGEAR_H */
+++ /dev/null
-#ifndef __ASM_SH_SYSTEMH_7751SYSTEMH_H
-#define __ASM_SH_SYSTEMH_7751SYSTEMH_H
-
-/*
- * linux/include/asm-sh/systemh/7751systemh.h
- *
- * Copyright (C) 2000 Kazumoto Kojima
- *
- * Hitachi SystemH support
-
- * Modified for 7751 SystemH by
- * Jonathan Short, 2002.
- */
-
-/* Box specific addresses. */
-
-#define PA_ROM 0x00000000 /* EPROM */
-#define PA_ROM_SIZE 0x00400000 /* EPROM size 4M byte */
-#define PA_FROM 0x01000000 /* EPROM */
-#define PA_FROM_SIZE 0x00400000 /* EPROM size 4M byte */
-#define PA_EXT1 0x04000000
-#define PA_EXT1_SIZE 0x04000000
-#define PA_EXT2 0x08000000
-#define PA_EXT2_SIZE 0x04000000
-#define PA_SDRAM 0x0c000000
-#define PA_SDRAM_SIZE 0x04000000
-
-#define PA_EXT4 0x12000000
-#define PA_EXT4_SIZE 0x02000000
-#define PA_EXT5 0x14000000
-#define PA_EXT5_SIZE 0x04000000
-#define PA_PCIC 0x18000000 /* MR-SHPC-01 PCMCIA */
-
-#define PA_DIPSW0 0xb9000000 /* Dip switch 5,6 */
-#define PA_DIPSW1 0xb9000002 /* Dip switch 7,8 */
-#define PA_LED 0xba000000 /* LED */
-#define PA_BCR 0xbb000000 /* FPGA on the MS7751SE01 */
-
-#define PA_MRSHPC 0xb83fffe0 /* MR-SHPC-01 PCMCIA controller */
-#define PA_MRSHPC_MW1 0xb8400000 /* MR-SHPC-01 memory window base */
-#define PA_MRSHPC_MW2 0xb8500000 /* MR-SHPC-01 attribute window base */
-#define PA_MRSHPC_IO 0xb8600000 /* MR-SHPC-01 I/O window base */
-#define MRSHPC_MODE (PA_MRSHPC + 4)
-#define MRSHPC_OPTION (PA_MRSHPC + 6)
-#define MRSHPC_CSR (PA_MRSHPC + 8)
-#define MRSHPC_ISR (PA_MRSHPC + 10)
-#define MRSHPC_ICR (PA_MRSHPC + 12)
-#define MRSHPC_CPWCR (PA_MRSHPC + 14)
-#define MRSHPC_MW0CR1 (PA_MRSHPC + 16)
-#define MRSHPC_MW1CR1 (PA_MRSHPC + 18)
-#define MRSHPC_IOWCR1 (PA_MRSHPC + 20)
-#define MRSHPC_MW0CR2 (PA_MRSHPC + 22)
-#define MRSHPC_MW1CR2 (PA_MRSHPC + 24)
-#define MRSHPC_IOWCR2 (PA_MRSHPC + 26)
-#define MRSHPC_CDCR (PA_MRSHPC + 28)
-#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
-
-#define BCR_ILCRA (PA_BCR + 0)
-#define BCR_ILCRB (PA_BCR + 2)
-#define BCR_ILCRC (PA_BCR + 4)
-#define BCR_ILCRD (PA_BCR + 6)
-#define BCR_ILCRE (PA_BCR + 8)
-#define BCR_ILCRF (PA_BCR + 10)
-#define BCR_ILCRG (PA_BCR + 12)
-
-#define IRQ_79C973 13
-
-#define __IO_PREFIX sh7751systemh
-#include <asm/io_generic.h>
-
-#endif /* __ASM_SH_SYSTEMH_7751SYSTEMH_H */
* Default rate for the root input clock, reset this with clk_set_rate()
* from the platform code.
*/
-struct clk extal_clk = {
+static struct clk extal_clk = {
.rate = 33333333,
};
.parent = &pll_clk,
};
-struct clk *main_clks[] = {
+static struct clk *main_clks[] = {
&r_clk,
&extal_clk,
&fll_clk,
enum { DIV6_V, DIV6_FA, DIV6_FB, DIV6_I, DIV6_S, DIV6_NR };
-struct clk div6_clks[DIV6_NR] = {
+static struct clk div6_clks[DIV6_NR] = {
[DIV6_V] = SH_CLK_DIV6(&div3_clk, VCLKCR, 0),
[DIV6_FA] = SH_CLK_DIV6(&div3_clk, FCLKACR, 0),
[DIV6_FB] = SH_CLK_DIV6(&div3_clk, FCLKBCR, 0),
config 32BIT
bool
- default y if CPU_SH5
+ default y if CPU_SH5 || !MMU
config PMB
bool "Support 32-bit physical addressing through PMB"
void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
enum dma_data_direction direction)
{
-#if defined(CONFIG_CPU_SH5) || defined(CONFIG_PMB)
- void *p1addr = vaddr;
-#else
- void *p1addr = (void*) P1SEGADDR((unsigned long)vaddr);
-#endif
+ void *addr;
+
+ addr = __in_29bit_mode() ?
+ (void *)P1SEGADDR((unsigned long)vaddr) : vaddr;
switch (direction) {
case DMA_FROM_DEVICE: /* invalidate only */
- __flush_invalidate_region(p1addr, size);
+ __flush_invalidate_region(addr, size);
break;
case DMA_TO_DEVICE: /* writeback only */
- __flush_wback_region(p1addr, size);
+ __flush_wback_region(addr, size);
break;
case DMA_BIDIRECTIONAL: /* writeback and invalidate */
- __flush_purge_region(p1addr, size);
+ __flush_purge_region(addr, size);
break;
default:
BUG();
void __init uncached_init(void)
{
-#ifdef CONFIG_29BIT
+#if defined(CONFIG_29BIT) || !defined(CONFIG_MMU)
uncached_start = P2SEG;
#else
uncached_start = memory_end;
7724SE SH_7724_SOLUTION_ENGINE
7751SE SH_7751_SOLUTION_ENGINE
7780SE SH_7780_SOLUTION_ENGINE
-7751SYSTEMH SH_7751_SYSTEMH
HP6XX SH_HP6XX
DREAMCAST SH_DREAMCAST
SNAPGEAR SH_SECUREEDGE5410
#include <linux/interrupt.h>
#include <linux/threads.h>
-#include <asm/kmap_types.h>
#include <asm/tlbflush.h>
#include <asm/homecache.h>
#define _ASM_TILE_KMAP_TYPES_H
/*
- * In TILE Linux each set of four of these uses another 16MB chunk of
- * address space, given 64 tiles and 64KB pages, so we only enable
- * ones that are required by the kernel configuration.
+ * In 32-bit TILE Linux we have to balance the desire to have a lot of
+ * nested atomic mappings with the fact that large page sizes and many
+ * processors chew up address space quickly. In a typical
+ * 64-processor, 64KB-page layout build, making KM_TYPE_NR one larger
+ * adds 4MB of required address-space. For now we leave KM_TYPE_NR
+ * set to depth 8.
*/
enum km_type {
+ KM_TYPE_NR = 8
+};
+
+/*
+ * We provide dummy definitions of all the stray values that used to be
+ * required for kmap_atomic() and no longer are.
+ */
+enum {
KM_BOUNCE_READ,
KM_SKB_SUNRPC_DATA,
KM_SKB_DATA_SOFTIRQ,
KM_USER0,
KM_USER1,
KM_BIO_SRC_IRQ,
+ KM_BIO_DST_IRQ,
+ KM_PTE0,
+ KM_PTE1,
KM_IRQ0,
KM_IRQ1,
KM_SOFTIRQ0,
KM_SOFTIRQ1,
- KM_MEMCPY0,
- KM_MEMCPY1,
-#if defined(CONFIG_HIGHPTE)
- KM_PTE0,
- KM_PTE1,
-#endif
- KM_TYPE_NR
+ KM_SYNC_ICACHE,
+ KM_SYNC_DCACHE,
+ KM_UML_USERCOPY,
+ KM_IRQ_PTE,
+ KM_NMI,
+ KM_NMI_PTE,
+ KM_KDB
};
#endif /* _ASM_TILE_KMAP_TYPES_H */
#define pgd_offset_k(address) pgd_offset(&init_mm, address)
#if defined(CONFIG_HIGHPTE)
-extern pte_t *_pte_offset_map(pmd_t *, unsigned long address, enum km_type);
-#define pte_offset_map(dir, address) \
- _pte_offset_map(dir, address, KM_PTE0)
-#define pte_unmap(pte) kunmap_atomic(pte, KM_PTE0)
+extern pte_t *pte_offset_map(pmd_t *, unsigned long address);
+#define pte_unmap(pte) kunmap_atomic(pte)
#else
#define pte_offset_map(dir, address) pte_offset_kernel(dir, address)
#define pte_unmap(pte) do { } while (0)
+#ifdef CONFIG_COMPAT
+#define __ARCH_WANT_STAT64 /* Used for compat_sys_stat64() etc. */
+#endif
#include <asm-generic/stat.h>
#ifdef CONFIG_COMPAT
#define __ARCH_WANT_SYS_LLSEEK
#endif
+#define __ARCH_WANT_SYS_NEWFSTATAT
#endif
#endif /* _ASM_TILE_UNISTD_H */
#define compat_sys_readahead sys32_readahead
#define compat_sys_sync_file_range compat_sys_sync_file_range2
-/* The native 64-bit "struct stat" matches the 32-bit "struct stat64". */
-#define compat_sys_stat64 sys_newstat
-#define compat_sys_lstat64 sys_newlstat
-#define compat_sys_fstat64 sys_newfstat
-#define compat_sys_fstatat64 sys_newfstatat
+/* We leverage the "struct stat64" type for 32-bit time_t/nsec. */
+#define compat_sys_stat64 sys_stat64
+#define compat_sys_lstat64 sys_lstat64
+#define compat_sys_fstat64 sys_fstat64
+#define compat_sys_fstatat64 sys_fstatat64
/* The native sys_ptrace dynamically handles compat binaries. */
#define compat_sys_ptrace sys_ptrace
void early_panic(const char *fmt, ...)
{
va_list ap;
- raw_local_irq_disable_all();
+ arch_local_irq_disable_all();
va_start(ap, fmt);
early_printk("Kernel panic - not syncing: ");
early_vprintk(fmt, ap);
static void enable_firewall_interrupts(void)
{
- raw_local_irq_unmask_now(INT_UDN_FIREWALL);
+ arch_local_irq_unmask_now(INT_UDN_FIREWALL);
}
static void disable_firewall_interrupts(void)
{
- raw_local_irq_mask_now(INT_UDN_FIREWALL);
+ arch_local_irq_mask_now(INT_UDN_FIREWALL);
}
/* Set up hardwall on this cpu based on the passed hardwall_info. */
}
static const struct file_operations dev_hardwall_fops = {
+ .open = nonseekable_open,
.unlocked_ioctl = hardwall_ioctl,
#ifdef CONFIG_COMPAT
.compat_ioctl = hardwall_compat_ioctl,
#endif
.flush = hardwall_flush,
.release = hardwall_release,
- .llseek = noop_llseek,
};
static struct cdev hardwall_dev;
#define IS_HW_CLEARED 1
/*
- * The set of interrupts we enable for raw_local_irq_enable().
+ * The set of interrupts we enable for arch_local_irq_enable().
* This is initialized to have just a single interrupt that the kernel
* doesn't actually use as a sentinel. During kernel init,
* interrupts are added as the kernel gets prepared to support them.
/* Enable interrupt delivery. */
unmask_irqs(~0UL);
#if CHIP_HAS_IPI()
- raw_local_irq_unmask(INT_IPI_K);
+ arch_local_irq_unmask(INT_IPI_K);
#endif
}
if ((entry & IND_SOURCE)) {
void *va =
- kmap_atomic_pfn(entry >> PAGE_SHIFT, KM_USER0);
+ kmap_atomic_pfn(entry >> PAGE_SHIFT);
r = kexec_bn2cl(va);
if (r) {
command_line = r;
break;
}
- kunmap_atomic(va, KM_USER0);
+ kunmap_atomic(va);
}
}
hverr = hv_set_command_line(
(HV_VirtAddr) command_line, strlen(command_line));
- kunmap_atomic(command_line, KM_USER0);
+ kunmap_atomic(command_line);
} else {
pr_info("%s: no command line found; making empty\n",
__func__);
panic("hv_register_message_state: error %d", rc);
/* Make sure downcall interrupts will be enabled. */
- raw_local_irq_unmask(INT_INTCTRL_K);
+ arch_local_irq_unmask(INT_INTCTRL_K);
}
void hv_message_intr(struct pt_regs *regs, int intnum)
{
unsigned long __user *datap = (long __user __force *)data;
unsigned long tmp;
- int i;
long ret = -EIO;
- unsigned long *childregs;
char *childreg;
+ struct pt_regs copyregs;
+ int ex1_offset;
switch (request) {
if (addr >= PTREGS_SIZE)
break;
childreg = (char *)task_pt_regs(child) + addr;
+
+ /* Guard against overwrites of the privilege level. */
+ ex1_offset = PTREGS_OFFSET_EX1;
+#if defined(CONFIG_COMPAT) && defined(__BIG_ENDIAN)
+ if (is_compat_task()) /* point at low word */
+ ex1_offset += sizeof(compat_long_t);
+#endif
+ if (addr == ex1_offset)
+ data = PL_ICS_EX1(USER_PL, EX1_ICS(data));
+
#ifdef CONFIG_COMPAT
if (is_compat_task()) {
if (addr & (sizeof(compat_long_t)-1))
break;
case PTRACE_GETREGS: /* Get all registers from the child. */
- if (!access_ok(VERIFY_WRITE, datap, PTREGS_SIZE))
- break;
- childregs = (long *)task_pt_regs(child);
- for (i = 0; i < sizeof(struct pt_regs)/sizeof(unsigned long);
- ++i) {
- ret = __put_user(childregs[i], &datap[i]);
- if (ret != 0)
- break;
+ if (copy_to_user(datap, task_pt_regs(child),
+ sizeof(struct pt_regs)) == 0) {
+ ret = 0;
}
break;
case PTRACE_SETREGS: /* Set all registers in the child. */
- if (!access_ok(VERIFY_READ, datap, PTREGS_SIZE))
- break;
- childregs = (long *)task_pt_regs(child);
- for (i = 0; i < sizeof(struct pt_regs)/sizeof(unsigned long);
- ++i) {
- ret = __get_user(childregs[i], &datap[i]);
- if (ret != 0)
- break;
+ if (copy_from_user(©regs, datap,
+ sizeof(struct pt_regs)) == 0) {
+ copyregs.ex1 =
+ PL_ICS_EX1(USER_PL, EX1_ICS(copyregs.ex1));
+ *task_pt_regs(child) = copyregs;
+ ret = 0;
}
break;
void machine_halt(void)
{
warn_early_printk();
- raw_local_irq_disable_all();
+ arch_local_irq_disable_all();
smp_send_stop();
hv_halt();
}
void machine_power_off(void)
{
warn_early_printk();
- raw_local_irq_disable_all();
+ arch_local_irq_disable_all();
smp_send_stop();
hv_power_off();
}
void machine_restart(char *cmd)
{
- raw_local_irq_disable_all();
+ arch_local_irq_disable_all();
smp_send_stop();
hv_restart((HV_VirtAddr) "vmlinux", (HV_VirtAddr) cmd);
}
/* Allow asynchronous TLB interrupts. */
#if CHIP_HAS_TILE_DMA()
- raw_local_irq_unmask(INT_DMATLB_MISS);
- raw_local_irq_unmask(INT_DMATLB_ACCESS);
+ arch_local_irq_unmask(INT_DMATLB_MISS);
+ arch_local_irq_unmask(INT_DMATLB_ACCESS);
#endif
#if CHIP_HAS_SN_PROC()
- raw_local_irq_unmask(INT_SNITLB_MISS);
+ arch_local_irq_unmask(INT_SNITLB_MISS);
#endif
#ifdef __tilegx__
- raw_local_irq_unmask(INT_SINGLE_STEP_K);
+ arch_local_irq_unmask(INT_SINGLE_STEP_K);
#endif
/*
for (i = 0; i < sizeof(struct pt_regs)/sizeof(long); ++i)
err |= __get_user(regs->regs[i], &sc->gregs[i]);
+ /* Ensure that the PL is always set to USER_PL. */
+ regs->ex1 = PL_ICS_EX1(USER_PL, EX1_ICS(regs->ex1));
+
regs->faultnum = INT_SWINT_1_SIGRETURN;
err |= __get_user(*pr0, &sc->gregs[0]);
current_thread_info()->status &= ~TS_RESTORE_SIGMASK;
}
- return;
+ goto done;
}
/* Did we come from a system call? */
current_thread_info()->status &= ~TS_RESTORE_SIGMASK;
sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL);
}
+
+done:
+ /* Avoid double syscall restart if there are nested signals. */
+ regs->faultnum = INT_SWINT_1_SIGRETURN;
}
static void smp_stop_cpu_interrupt(void)
{
set_cpu_online(smp_processor_id(), 0);
- raw_local_irq_disable_all();
+ arch_local_irq_disable_all();
for (;;)
asm("nap");
}
{
BUG_ON(ticks > MAX_TICK);
__insn_mtspr(SPR_TILE_TIMER_CONTROL, ticks);
- raw_local_irq_unmask_now(INT_TILE_TIMER);
+ arch_local_irq_unmask_now(INT_TILE_TIMER);
return 0;
}
static void tile_timer_set_mode(enum clock_event_mode mode,
struct clock_event_device *evt)
{
- raw_local_irq_mask_now(INT_TILE_TIMER);
+ arch_local_irq_mask_now(INT_TILE_TIMER);
}
/*
evt->cpumask = cpumask_of(smp_processor_id());
/* Start out with timer not firing. */
- raw_local_irq_mask_now(INT_TILE_TIMER);
+ arch_local_irq_mask_now(INT_TILE_TIMER);
/* Register tile timer. */
clockevents_register_device(evt);
* Mask the timer interrupt here, since we are a oneshot timer
* and there are now by definition no events pending.
*/
- raw_local_irq_mask(INT_TILE_TIMER);
+ arch_local_irq_mask(INT_TILE_TIMER);
/* Track time spent here in an interrupt context */
irq_enter();
* we must run with interrupts disabled to avoid the risk of some
* other code seeing the incoherent data in our cache. (Recall that
* our cache is indexed by PA, so even if the other code doesn't use
- * our KM_MEMCPY virtual addresses, they'll still hit in cache using
+ * our kmap_atomic virtual addresses, they'll still hit in cache using
* the normal VAs that aren't supposed to hit in cache.)
*/
static void memcpy_multicache(void *dest, const void *source,
unsigned long flags, newsrc, newdst;
pmd_t *pmdp;
pte_t *ptep;
+ int type0, type1;
int cpu = get_cpu();
/*
sim_allow_multiple_caching(1);
/* Set up the new dest mapping */
- idx = FIX_KMAP_BEGIN + (KM_TYPE_NR * cpu) + KM_MEMCPY0;
+ type0 = kmap_atomic_idx_push();
+ idx = FIX_KMAP_BEGIN + (KM_TYPE_NR * cpu) + type0;
newdst = __fix_to_virt(idx) + ((unsigned long)dest & (PAGE_SIZE-1));
pmdp = pmd_offset(pud_offset(pgd_offset_k(newdst), newdst), newdst);
ptep = pte_offset_kernel(pmdp, newdst);
}
/* Set up the new source mapping */
- idx += (KM_MEMCPY0 - KM_MEMCPY1);
+ type1 = kmap_atomic_idx_push();
+ idx += (type0 - type1);
src_pte = hv_pte_set_nc(src_pte);
src_pte = hv_pte_clear_writable(src_pte); /* be paranoid */
newsrc = __fix_to_virt(idx) + ((unsigned long)source & (PAGE_SIZE-1));
* We're done: notify the simulator that all is back to normal,
* and re-enable interrupts and pre-emption.
*/
+ kmap_atomic_idx_pop();
+ kmap_atomic_idx_pop();
sim_allow_multiple_caching(0);
local_irq_restore(flags);
put_cpu();
void *__kmap_atomic(struct page *page)
{
/* PAGE_NONE is a magic value that tells us to check immutability. */
- return kmap_atomic_prot(page, type, PAGE_NONE);
+ return kmap_atomic_prot(page, PAGE_NONE);
}
EXPORT_SYMBOL(__kmap_atomic);
/* Select whether to free (1) or mark unusable (0) the __init pages. */
static int __init set_initfree(char *str)
{
- strict_strtol(str, 0, &initfree);
- pr_info("initfree: %s free init pages\n", initfree ? "will" : "won't");
+ long val;
+ if (strict_strtol(str, 0, &val)) {
+ initfree = val;
+ pr_info("initfree: %s free init pages\n",
+ initfree ? "will" : "won't");
+ }
return 1;
}
__setup("initfree=", set_initfree);
}
#if defined(CONFIG_HIGHPTE)
-pte_t *_pte_offset_map(pmd_t *dir, unsigned long address, enum km_type type)
+pte_t *_pte_offset_map(pmd_t *dir, unsigned long address)
{
- pte_t *pte = kmap_atomic(pmd_page(*dir), type) +
+ pte_t *pte = kmap_atomic(pmd_page(*dir)) +
(pmd_ptfn(*dir) << HV_LOG2_PAGE_TABLE_ALIGN) & ~PAGE_MASK;
return &pte[pte_index(address)];
}
struct task_struct;
-extern long subarch_ptrace(struct task_struct *child, long request, long addr,
- long data);
+extern long subarch_ptrace(struct task_struct *child, long request,
+ unsigned long addr, unsigned long data);
extern unsigned long getreg(struct task_struct *child, int regno);
extern int putreg(struct task_struct *child, int regno, unsigned long value);
extern int get_fpregs(struct user_i387_struct __user *buf,
break;
case PTRACE_SET_THREAD_AREA:
- ret = ptrace_set_thread_area(child, addr, datavp);
+ ret = ptrace_set_thread_area(child, addr, vp);
break;
case PTRACE_FAULTINFO: {
struct amd_nb *nb;
int i;
- nb = kmalloc(sizeof(struct amd_nb), GFP_KERNEL);
+ nb = kmalloc_node(sizeof(struct amd_nb), GFP_KERNEL | __GFP_ZERO,
+ cpu_to_node(cpu));
if (!nb)
return NULL;
- memset(nb, 0, sizeof(*nb));
nb->nb_id = nb_id;
/*
}
}
-static void set_spte_track_bits(u64 *sptep, u64 new_spte)
+static int set_spte_track_bits(u64 *sptep, u64 new_spte)
{
pfn_t pfn;
u64 old_spte = *sptep;
old_spte = __xchg_spte(sptep, new_spte);
if (!is_rmap_spte(old_spte))
- return;
+ return 0;
pfn = spte_to_pfn(old_spte);
if (!shadow_accessed_mask || old_spte & shadow_accessed_mask)
kvm_set_pfn_accessed(pfn);
if (!shadow_dirty_mask || (old_spte & shadow_dirty_mask))
kvm_set_pfn_dirty(pfn);
+ return 1;
}
static void drop_spte(struct kvm *kvm, u64 *sptep, u64 new_spte)
{
- set_spte_track_bits(sptep, new_spte);
- rmap_remove(kvm, sptep);
+ if (set_spte_track_bits(sptep, new_spte))
+ rmap_remove(kvm, sptep);
}
static u64 *rmap_next(struct kvm *kvm, unsigned long *rmapp, u64 *spte)
!kvm_exception_is_soft(vcpu->arch.exception.nr);
events->exception.nr = vcpu->arch.exception.nr;
events->exception.has_error_code = vcpu->arch.exception.has_error_code;
+ events->exception.pad = 0;
events->exception.error_code = vcpu->arch.exception.error_code;
events->interrupt.injected =
events->nmi.injected = vcpu->arch.nmi_injected;
events->nmi.pending = vcpu->arch.nmi_pending;
events->nmi.masked = kvm_x86_ops->get_nmi_mask(vcpu);
+ events->nmi.pad = 0;
events->sipi_vector = vcpu->arch.sipi_vector;
events->flags = (KVM_VCPUEVENT_VALID_NMI_PENDING
| KVM_VCPUEVENT_VALID_SIPI_VECTOR
| KVM_VCPUEVENT_VALID_SHADOW);
+ memset(&events->reserved, 0, sizeof(events->reserved));
}
static int kvm_vcpu_ioctl_x86_set_vcpu_events(struct kvm_vcpu *vcpu,
dbgregs->dr6 = vcpu->arch.dr6;
dbgregs->dr7 = vcpu->arch.dr7;
dbgregs->flags = 0;
+ memset(&dbgregs->reserved, 0, sizeof(dbgregs->reserved));
}
static int kvm_vcpu_ioctl_x86_set_debugregs(struct kvm_vcpu *vcpu,
sizeof(ps->channels));
ps->flags = kvm->arch.vpit->pit_state.flags;
mutex_unlock(&kvm->arch.vpit->pit_state.lock);
+ memset(&ps->reserved, 0, sizeof(ps->reserved));
return r;
}
struct kvm_memslots *slots, *old_slots;
unsigned long *dirty_bitmap;
- spin_lock(&kvm->mmu_lock);
- kvm_mmu_slot_remove_write_access(kvm, log->slot);
- spin_unlock(&kvm->mmu_lock);
-
r = -ENOMEM;
dirty_bitmap = vmalloc(n);
if (!dirty_bitmap)
dirty_bitmap = old_slots->memslots[log->slot].dirty_bitmap;
kfree(old_slots);
+ spin_lock(&kvm->mmu_lock);
+ kvm_mmu_slot_remove_write_access(kvm, log->slot);
+ spin_unlock(&kvm->mmu_lock);
+
r = -EFAULT;
if (copy_to_user(log->dirty_bitmap, dirty_bitmap, n)) {
vfree(dirty_bitmap);
user_ns.clock = kvm->arch.kvmclock_offset + now_ns;
local_irq_enable();
user_ns.flags = 0;
+ memset(&user_ns.pad, 0, sizeof(user_ns.pad));
r = -EFAULT;
if (copy_to_user(argp, &user_ns, sizeof(user_ns)))
return X86EMUL_CONTINUE;
if (kvm_x86_ops->has_wbinvd_exit()) {
+ preempt_disable();
smp_call_function_many(vcpu->arch.wbinvd_dirty_mask,
wbinvd_ipi, NULL, 1);
+ preempt_enable();
cpumask_clear(vcpu->arch.wbinvd_dirty_mask);
}
wbinvd();
static void pcrypt_fini_padata(struct padata_pcrypt *pcrypt)
{
- kobject_put(&pcrypt->pinst->kobj);
free_cpumask_var(pcrypt->cb_cpumask->mask);
kfree(pcrypt->cb_cpumask);
# char/ comes before serial/ etc so that the VT console is the boot-time
# default.
+obj-y += tty/
obj-y += char/
# gpu/ comes after char for AGP vs DRM startup
out_put_disk:
while (dr--) {
del_timer(&motor_off_timer[dr]);
- put_disk(disks[dr]);
if (disks[dr]->queue)
blk_cleanup_queue(disks[dr]->queue);
+ put_disk(disks[dr]);
}
return err;
}
device_remove_file(&floppy_device[drive].dev, &dev_attr_cmos);
platform_device_unregister(&floppy_device[drive]);
}
- put_disk(disks[drive]);
blk_cleanup_queue(disks[drive]->queue);
+ put_disk(disks[drive]);
}
del_timer_sync(&fd_timeout);
# Makefile for the kernel character device drivers.
#
-#
-# This file contains the font map for the default (hardware) font
-#
-FONTMAPFILE = cp437.uni
-
-obj-y += mem.o random.o tty_io.o n_tty.o tty_ioctl.o tty_ldisc.o tty_buffer.o tty_port.o
-
-obj-y += tty_mutex.o
-obj-$(CONFIG_LEGACY_PTYS) += pty.o
-obj-$(CONFIG_UNIX98_PTYS) += pty.o
+obj-y += mem.o random.o
obj-$(CONFIG_TTY_PRINTK) += ttyprintk.o
obj-y += misc.o
-obj-$(CONFIG_VT) += vt_ioctl.o vc_screen.o selection.o keyboard.o
obj-$(CONFIG_BFIN_JTAG_COMM) += bfin_jtag_comm.o
-obj-$(CONFIG_CONSOLE_TRANSLATIONS) += consolemap.o consolemap_deftbl.o
-obj-$(CONFIG_HW_CONSOLE) += vt.o defkeymap.o
-obj-$(CONFIG_AUDIT) += tty_audit.o
-obj-$(CONFIG_MAGIC_SYSRQ) += sysrq.o
obj-$(CONFIG_MVME147_SCC) += generic_serial.o vme_scc.o
obj-$(CONFIG_MVME162_SCC) += generic_serial.o vme_scc.o
obj-$(CONFIG_BVME6000_SCC) += generic_serial.o vme_scc.o
obj-$(CONFIG_SYNCLINK) += synclink.o
obj-$(CONFIG_SYNCLINKMP) += synclinkmp.o
obj-$(CONFIG_SYNCLINK_GT) += synclink_gt.o
-obj-$(CONFIG_N_HDLC) += n_hdlc.o
-obj-$(CONFIG_N_GSM) += n_gsm.o
obj-$(CONFIG_AMIGA_BUILTIN_SERIAL) += amiserial.o
obj-$(CONFIG_SX) += sx.o generic_serial.o
obj-$(CONFIG_RIO) += rio/ generic_serial.o
obj-$(CONFIG_APM_EMULATION) += apm-emulation.o
obj-$(CONFIG_DTLK) += dtlk.o
-obj-$(CONFIG_R3964) += n_r3964.o
obj-$(CONFIG_APPLICOM) += applicom.o
obj-$(CONFIG_SONYPI) += sonypi.o
obj-$(CONFIG_RTC) += rtc.o
obj-$(CONFIG_JS_RTC) += js-rtc.o
js-rtc-y = rtc.o
-
-# Files generated that shall be removed upon make clean
-clean-files := consolemap_deftbl.c defkeymap.c
-
-quiet_cmd_conmk = CONMK $@
- cmd_conmk = scripts/conmakehash $< > $@
-
-$(obj)/consolemap_deftbl.c: $(src)/$(FONTMAPFILE)
- $(call cmd,conmk)
-
-$(obj)/defkeymap.o: $(obj)/defkeymap.c
-
-# Uncomment if you're changing the keymap and have an appropriate
-# loadkeys version for the map. By default, we'll use the shipped
-# versions.
-# GENERATE_KEYMAP := 1
-
-ifdef GENERATE_KEYMAP
-
-$(obj)/defkeymap.c: $(obj)/%.c: $(src)/%.map
- loadkeys --mktable $< > $@.tmp
- sed -e 's/^static *//' $@.tmp > $@
- rm $@.tmp
-
-endif
unsigned int gfdt = flags & AGP_USER_CACHED_MEMORY_GFDT;
u32 pte_flags;
- if (type_mask == AGP_USER_UNCACHED_MEMORY)
+ if (type_mask == AGP_USER_MEMORY)
pte_flags = GEN6_PTE_UNCACHED | I810_PTE_VALID;
else if (type_mask == AGP_USER_CACHED_MEMORY_LLC_MLC) {
- pte_flags = GEN6_PTE_LLC | I810_PTE_VALID;
+ pte_flags = GEN6_PTE_LLC_MLC | I810_PTE_VALID;
if (gfdt)
pte_flags |= GEN6_PTE_GFDT;
} else { /* set 'normal'/'cached' to LLC by default */
- pte_flags = GEN6_PTE_LLC_MLC | I810_PTE_VALID;
+ pte_flags = GEN6_PTE_LLC | I810_PTE_VALID;
if (gfdt)
pte_flags |= GEN6_PTE_GFDT;
}
/* get hold of clock */
p->clk = clk_get(&p->pdev->dev, "cmt_fck");
if (IS_ERR(p->clk)) {
- dev_warn(&p->pdev->dev, "using deprecated clock lookup\n");
- p->clk = clk_get(&p->pdev->dev, cfg->clk);
- if (IS_ERR(p->clk)) {
- dev_err(&p->pdev->dev, "cannot get clock\n");
- ret = PTR_ERR(p->clk);
- goto err1;
- }
+ dev_err(&p->pdev->dev, "cannot get clock\n");
+ ret = PTR_ERR(p->clk);
+ goto err1;
}
if (resource_size(res) == 6) {
/* get hold of clock */
p->clk = clk_get(&p->pdev->dev, "mtu2_fck");
if (IS_ERR(p->clk)) {
- dev_warn(&p->pdev->dev, "using deprecated clock lookup\n");
- p->clk = clk_get(&p->pdev->dev, cfg->clk);
- if (IS_ERR(p->clk)) {
- dev_err(&p->pdev->dev, "cannot get clock\n");
- ret = PTR_ERR(p->clk);
- goto err1;
- }
+ dev_err(&p->pdev->dev, "cannot get clock\n");
+ ret = PTR_ERR(p->clk);
+ goto err1;
}
return sh_mtu2_register(p, (char *)dev_name(&p->pdev->dev),
/* get hold of clock */
p->clk = clk_get(&p->pdev->dev, "tmu_fck");
if (IS_ERR(p->clk)) {
- dev_warn(&p->pdev->dev, "using deprecated clock lookup\n");
- p->clk = clk_get(&p->pdev->dev, cfg->clk);
- if (IS_ERR(p->clk)) {
- dev_err(&p->pdev->dev, "cannot get clock\n");
- ret = PTR_ERR(p->clk);
- goto err1;
- }
+ dev_err(&p->pdev->dev, "cannot get clock\n");
+ ret = PTR_ERR(p->clk);
+ goto err1;
}
return sh_tmu_register(p, (char *)dev_name(&p->pdev->dev),
return ret;
}
-static int ar_context_add_page(struct ar_context *ctx)
+static void ar_context_link_page(struct ar_context *ctx,
+ struct ar_buffer *ab, dma_addr_t ab_bus)
{
- struct device *dev = ctx->ohci->card.device;
- struct ar_buffer *ab;
- dma_addr_t uninitialized_var(ab_bus);
size_t offset;
- ab = dma_alloc_coherent(dev, PAGE_SIZE, &ab_bus, GFP_ATOMIC);
- if (ab == NULL)
- return -ENOMEM;
-
ab->next = NULL;
memset(&ab->descriptor, 0, sizeof(ab->descriptor));
ab->descriptor.control = cpu_to_le16(DESCRIPTOR_INPUT_MORE |
reg_write(ctx->ohci, CONTROL_SET(ctx->regs), CONTEXT_WAKE);
flush_writes(ctx->ohci);
+}
+
+static int ar_context_add_page(struct ar_context *ctx)
+{
+ struct device *dev = ctx->ohci->card.device;
+ struct ar_buffer *ab;
+ dma_addr_t uninitialized_var(ab_bus);
+
+ ab = dma_alloc_coherent(dev, PAGE_SIZE, &ab_bus, GFP_ATOMIC);
+ if (ab == NULL)
+ return -ENOMEM;
+
+ ar_context_link_page(ctx, ab, ab_bus);
return 0;
}
static void ar_context_tasklet(unsigned long data)
{
struct ar_context *ctx = (struct ar_context *)data;
- struct fw_ohci *ohci = ctx->ohci;
struct ar_buffer *ab;
struct descriptor *d;
void *buffer, *end;
+ __le16 res_count;
ab = ctx->current_buffer;
d = &ab->descriptor;
- if (d->res_count == 0) {
- size_t size, rest, offset;
+ res_count = ACCESS_ONCE(d->res_count);
+ if (res_count == 0) {
+ size_t size, size2, rest, pktsize, size3, offset;
dma_addr_t start_bus;
void *start;
*/
offset = offsetof(struct ar_buffer, data);
- start = buffer = ab;
+ start = ab;
start_bus = le32_to_cpu(ab->descriptor.data_address) - offset;
+ buffer = ab->data;
ab = ab->next;
d = &ab->descriptor;
- size = buffer + PAGE_SIZE - ctx->pointer;
+ size = start + PAGE_SIZE - ctx->pointer;
+ /* valid buffer data in the next page */
rest = le16_to_cpu(d->req_count) - le16_to_cpu(d->res_count);
+ /* what actually fits in this page */
+ size2 = min(rest, (size_t)PAGE_SIZE - offset - size);
memmove(buffer, ctx->pointer, size);
- memcpy(buffer + size, ab->data, rest);
- ctx->current_buffer = ab;
- ctx->pointer = (void *) ab->data + rest;
- end = buffer + size + rest;
+ memcpy(buffer + size, ab->data, size2);
+
+ while (size > 0) {
+ void *next = handle_ar_packet(ctx, buffer);
+ pktsize = next - buffer;
+ if (pktsize >= size) {
+ /*
+ * We have handled all the data that was
+ * originally in this page, so we can now
+ * continue in the next page.
+ */
+ buffer = next;
+ break;
+ }
+ /* move the next packet to the start of the buffer */
+ memmove(buffer, next, size + size2 - pktsize);
+ size -= pktsize;
+ /* fill up this page again */
+ size3 = min(rest - size2,
+ (size_t)PAGE_SIZE - offset - size - size2);
+ memcpy(buffer + size + size2,
+ (void *) ab->data + size2, size3);
+ size2 += size3;
+ }
- while (buffer < end)
- buffer = handle_ar_packet(ctx, buffer);
+ if (rest > 0) {
+ /* handle the packets that are fully in the next page */
+ buffer = (void *) ab->data +
+ (buffer - (start + offset + size));
+ end = (void *) ab->data + rest;
+
+ while (buffer < end)
+ buffer = handle_ar_packet(ctx, buffer);
- dma_free_coherent(ohci->card.device, PAGE_SIZE,
- start, start_bus);
- ar_context_add_page(ctx);
+ ctx->current_buffer = ab;
+ ctx->pointer = end;
+
+ ar_context_link_page(ctx, start, start_bus);
+ } else {
+ ctx->pointer = start + PAGE_SIZE;
+ }
} else {
buffer = ctx->pointer;
ctx->pointer = end =
- (void *) ab + PAGE_SIZE - le16_to_cpu(d->res_count);
+ (void *) ab + PAGE_SIZE - le16_to_cpu(res_count);
while (buffer < end)
buffer = handle_ar_packet(ctx, buffer);
struct drm_crtc *tmp;
int crtc_mask = 1;
- WARN(!crtc, "checking null crtc?");
+ WARN(!crtc, "checking null crtc?\n");
dev = crtc->dev;
.addr = DDC_ADDR,
.flags = I2C_M_RD,
.len = len,
- .buf = buf + start,
+ .buf = buf,
}
};
static u8 *
drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter)
{
- int i, j = 0;
+ int i, j = 0, valid_extensions = 0;
u8 *block, *new;
if ((block = kmalloc(EDID_LENGTH, GFP_KERNEL)) == NULL)
for (j = 1; j <= block[0x7e]; j++) {
for (i = 0; i < 4; i++) {
- if (drm_do_probe_ddc_edid(adapter, block, j,
- EDID_LENGTH))
+ if (drm_do_probe_ddc_edid(adapter,
+ block + (valid_extensions + 1) * EDID_LENGTH,
+ j, EDID_LENGTH))
goto out;
- if (drm_edid_block_valid(block + j * EDID_LENGTH))
+ if (drm_edid_block_valid(block + (valid_extensions + 1) * EDID_LENGTH)) {
+ valid_extensions++;
break;
+ }
}
if (i == 4)
- goto carp;
+ dev_warn(connector->dev->dev,
+ "%s: Ignoring invalid EDID block %d.\n",
+ drm_get_connector_name(connector), j);
+ }
+
+ if (valid_extensions != block[0x7e]) {
+ block[EDID_LENGTH-1] += block[0x7e] - valid_extensions;
+ block[0x7e] = valid_extensions;
+ new = krealloc(block, (valid_extensions + 1) * EDID_LENGTH, GFP_KERNEL);
+ if (!new)
+ goto out;
+ block = new;
}
return block;
module_param_named(fbpercrtc, i915_fbpercrtc, int, 0400);
unsigned int i915_powersave = 1;
-module_param_named(powersave, i915_powersave, int, 0400);
+module_param_named(powersave, i915_powersave, int, 0600);
unsigned int i915_lvds_downclock = 0;
module_param_named(lvds_downclock, i915_lvds_downclock, int, 0400);
#define INTEL_PCH_TYPE(dev) (((struct drm_i915_private *)(dev)->dev_private)->pch_type)
#define HAS_PCH_CPT(dev) (INTEL_PCH_TYPE(dev) == PCH_CPT)
+#define HAS_PCH_IBX(dev) (INTEL_PCH_TYPE(dev) == PCH_IBX)
#define PRIMARY_RINGBUFFER_SIZE (128*1024)
static int i915_ring_idle(struct drm_device *dev,
struct intel_ring_buffer *ring)
{
- if (list_empty(&ring->gpu_write_list))
+ if (list_empty(&ring->gpu_write_list) && list_empty(&ring->active_list))
return 0;
i915_gem_flush_ring(dev, NULL, ring,
int ret;
lists_empty = (list_empty(&dev_priv->mm.flushing_list) &&
- list_empty(&dev_priv->render_ring.active_list) &&
- list_empty(&dev_priv->bsd_ring.active_list) &&
- list_empty(&dev_priv->blt_ring.active_list));
+ list_empty(&dev_priv->mm.active_list));
if (lists_empty)
return 0;
* write domain
*/
if (obj->write_domain &&
- obj->write_domain != obj->pending_read_domains) {
+ (obj->write_domain != obj->pending_read_domains ||
+ obj_priv->ring != ring)) {
flush_domains |= obj->write_domain;
invalidate_domains |=
obj->pending_read_domains & ~obj->write_domain;
return 0;
}
+static int
+i915_gem_execbuffer_move_to_gpu(struct drm_device *dev,
+ struct drm_file *file,
+ struct intel_ring_buffer *ring,
+ struct drm_gem_object **objects,
+ int count)
+{
+ struct drm_i915_private *dev_priv = dev->dev_private;
+ int ret, i;
+
+ /* Zero the global flush/invalidate flags. These
+ * will be modified as new domains are computed
+ * for each object
+ */
+ dev->invalidate_domains = 0;
+ dev->flush_domains = 0;
+ dev_priv->mm.flush_rings = 0;
+ for (i = 0; i < count; i++)
+ i915_gem_object_set_to_gpu_domain(objects[i], ring);
+
+ if (dev->invalidate_domains | dev->flush_domains) {
+#if WATCH_EXEC
+ DRM_INFO("%s: invalidate_domains %08x flush_domains %08x\n",
+ __func__,
+ dev->invalidate_domains,
+ dev->flush_domains);
+#endif
+ i915_gem_flush(dev, file,
+ dev->invalidate_domains,
+ dev->flush_domains,
+ dev_priv->mm.flush_rings);
+ }
+
+ for (i = 0; i < count; i++) {
+ struct drm_i915_gem_object *obj = to_intel_bo(objects[i]);
+ /* XXX replace with semaphores */
+ if (obj->ring && ring != obj->ring) {
+ ret = i915_gem_object_wait_rendering(&obj->base, true);
+ if (ret)
+ return ret;
+ }
+ }
+
+ return 0;
+}
+
/* Throttle our rendering by waiting until the ring has completed our requests
* emitted over 20 msec ago.
*
goto err;
}
- /* Zero the global flush/invalidate flags. These
- * will be modified as new domains are computed
- * for each object
- */
- dev->invalidate_domains = 0;
- dev->flush_domains = 0;
- dev_priv->mm.flush_rings = 0;
-
- for (i = 0; i < args->buffer_count; i++) {
- struct drm_gem_object *obj = object_list[i];
-
- /* Compute new gpu domains and update invalidate/flush */
- i915_gem_object_set_to_gpu_domain(obj, ring);
- }
-
- if (dev->invalidate_domains | dev->flush_domains) {
-#if WATCH_EXEC
- DRM_INFO("%s: invalidate_domains %08x flush_domains %08x\n",
- __func__,
- dev->invalidate_domains,
- dev->flush_domains);
-#endif
- i915_gem_flush(dev, file,
- dev->invalidate_domains,
- dev->flush_domains,
- dev_priv->mm.flush_rings);
- }
+ ret = i915_gem_execbuffer_move_to_gpu(dev, file, ring,
+ object_list, args->buffer_count);
+ if (ret)
+ goto err;
for (i = 0; i < args->buffer_count; i++) {
struct drm_gem_object *obj = object_list[i];
alignment = i915_gem_get_gtt_alignment(obj);
if (obj_priv->gtt_offset & (alignment - 1)) {
WARN(obj_priv->pin_count,
- "bo is already pinned with incorrect alignment:"
- " offset=%x, req.alignment=%x\n",
+ "bo is already pinned with incorrect alignment: offset=%x, req.alignment=%x\n",
obj_priv->gtt_offset, alignment);
ret = i915_gem_object_unbind(obj);
if (ret)
struct drm_file *file_priv)
{
struct drm_i915_gem_object *obj_priv = to_intel_bo(obj);
- void *obj_addr;
- int ret;
- char __user *user_data;
+ void *vaddr = obj_priv->phys_obj->handle->vaddr + args->offset;
+ char __user *user_data = (char __user *) (uintptr_t) args->data_ptr;
- user_data = (char __user *) (uintptr_t) args->data_ptr;
- obj_addr = obj_priv->phys_obj->handle->vaddr + args->offset;
+ DRM_DEBUG_DRIVER("vaddr %p, %lld\n", vaddr, args->size);
- DRM_DEBUG_DRIVER("obj_addr %p, %lld\n", obj_addr, args->size);
- ret = copy_from_user(obj_addr, user_data, args->size);
- if (ret)
- return -EFAULT;
+ if (__copy_from_user_inatomic_nocache(vaddr, user_data, args->size)) {
+ unsigned long unwritten;
+
+ /* The physical object once assigned is fixed for the lifetime
+ * of the obj, so we can safely drop the lock and continue
+ * to access vaddr.
+ */
+ mutex_unlock(&dev->struct_mutex);
+ unwritten = copy_from_user(vaddr, user_data, args->size);
+ mutex_lock(&dev->struct_mutex);
+ if (unwritten)
+ return -EFAULT;
+ }
drm_agp_chipset_flush(dev);
return 0;
int lists_empty;
lists_empty = list_empty(&dev_priv->mm.flushing_list) &&
- list_empty(&dev_priv->render_ring.active_list) &&
- list_empty(&dev_priv->bsd_ring.active_list) &&
- list_empty(&dev_priv->blt_ring.active_list);
+ list_empty(&dev_priv->mm.active_list);
return !lists_empty;
}
lists_empty = (list_empty(&dev_priv->mm.inactive_list) &&
list_empty(&dev_priv->mm.flushing_list) &&
- list_empty(&dev_priv->render_ring.active_list) &&
- list_empty(&dev_priv->bsd_ring.active_list) &&
- list_empty(&dev_priv->blt_ring.active_list));
+ list_empty(&dev_priv->mm.active_list));
if (lists_empty)
return -ENOSPC;
lists_empty = (list_empty(&dev_priv->mm.inactive_list) &&
list_empty(&dev_priv->mm.flushing_list) &&
- list_empty(&dev_priv->render_ring.active_list) &&
- list_empty(&dev_priv->bsd_ring.active_list) &&
- list_empty(&dev_priv->blt_ring.active_list));
+ list_empty(&dev_priv->mm.active_list));
BUG_ON(!lists_empty);
return 0;
/* Clock gating state */
intel_init_clock_gating(dev);
- if (HAS_PCH_SPLIT(dev))
+ if (HAS_PCH_SPLIT(dev)) {
ironlake_enable_drps(dev);
+ intel_init_emon(dev);
+ }
/* Cache mode state */
I915_WRITE (CACHE_MODE_0, dev_priv->saveCACHE_MODE_0 | 0xffff0000);
udelay(500);
}
+static void intel_fdi_normal_train(struct drm_crtc *crtc)
+{
+ struct drm_device *dev = crtc->dev;
+ struct drm_i915_private *dev_priv = dev->dev_private;
+ struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+ int pipe = intel_crtc->pipe;
+ u32 reg, temp;
+
+ /* enable normal train */
+ reg = FDI_TX_CTL(pipe);
+ temp = I915_READ(reg);
+ temp &= ~FDI_LINK_TRAIN_NONE;
+ temp |= FDI_LINK_TRAIN_NONE | FDI_TX_ENHANCE_FRAME_ENABLE;
+ I915_WRITE(reg, temp);
+
+ reg = FDI_RX_CTL(pipe);
+ temp = I915_READ(reg);
+ if (HAS_PCH_CPT(dev)) {
+ temp &= ~FDI_LINK_TRAIN_PATTERN_MASK_CPT;
+ temp |= FDI_LINK_TRAIN_NORMAL_CPT;
+ } else {
+ temp &= ~FDI_LINK_TRAIN_NONE;
+ temp |= FDI_LINK_TRAIN_NONE;
+ }
+ I915_WRITE(reg, temp | FDI_RX_ENHANCE_FRAME_ENABLE);
+
+ /* wait one idle pattern time */
+ POSTING_READ(reg);
+ udelay(1000);
+}
+
/* The FDI link training functions for ILK/Ibexpeak. */
static void ironlake_fdi_link_train(struct drm_crtc *crtc)
{
DRM_DEBUG_KMS("FDI train done\n");
- /* enable normal train */
- reg = FDI_TX_CTL(pipe);
- temp = I915_READ(reg);
- temp &= ~FDI_LINK_TRAIN_NONE;
- temp |= FDI_LINK_TRAIN_NONE | FDI_TX_ENHANCE_FRAME_ENABLE;
- I915_WRITE(reg, temp);
-
- reg = FDI_RX_CTL(pipe);
- temp = I915_READ(reg);
- if (HAS_PCH_CPT(dev)) {
- temp &= ~FDI_LINK_TRAIN_PATTERN_MASK_CPT;
- temp |= FDI_LINK_TRAIN_NORMAL_CPT;
- } else {
- temp &= ~FDI_LINK_TRAIN_NONE;
- temp |= FDI_LINK_TRAIN_NONE;
- }
- I915_WRITE(reg, temp | FDI_RX_ENHANCE_FRAME_ENABLE);
-
- /* wait one idle pattern time */
- POSTING_READ(reg);
- udelay(1000);
}
static const int const snb_b_fdi_train_param [] = {
I915_WRITE(TRANS_VBLANK(pipe), I915_READ(VBLANK(pipe)));
I915_WRITE(TRANS_VSYNC(pipe), I915_READ(VSYNC(pipe)));
+ intel_fdi_normal_train(crtc);
+
/* For PCH DP, enable TRANS_DP_CTL */
if (HAS_PCH_CPT(dev) &&
intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT)) {
udelay(100);
/* Ironlake workaround, disable clock pointer after downing FDI */
- I915_WRITE(FDI_RX_CHICKEN(pipe),
- I915_READ(FDI_RX_CHICKEN(pipe) &
- ~FDI_RX_PHASE_SYNC_POINTER_ENABLE));
+ if (HAS_PCH_IBX(dev))
+ I915_WRITE(FDI_RX_CHICKEN(pipe),
+ I915_READ(FDI_RX_CHICKEN(pipe) &
+ ~FDI_RX_PHASE_SYNC_POINTER_ENABLE));
/* still set train pattern 1 */
reg = FDI_TX_CTL(pipe);
fmin = (rgvmodectl & MEMMODE_FMIN_MASK);
fstart = (rgvmodectl & MEMMODE_FSTART_MASK) >>
MEMMODE_FSTART_SHIFT;
- fstart = fmax;
vstart = (I915_READ(PXVFREQ_BASE + (fstart * 4)) & PXVFREQ_PX_MASK) >>
PXVFREQ_PX_SHIFT;
- dev_priv->fmax = fstart; /* IPS callback will increase this */
+ dev_priv->fmax = fmax; /* IPS callback will increase this */
dev_priv->fstart = fstart;
- dev_priv->max_delay = fmax;
+ dev_priv->max_delay = fstart;
dev_priv->min_delay = fmin;
dev_priv->cur_delay = fstart;
- DRM_DEBUG_DRIVER("fmax: %d, fmin: %d, fstart: %d\n", fmax, fmin,
- fstart);
+ DRM_DEBUG_DRIVER("fmax: %d, fmin: %d, fstart: %d\n",
+ fmax, fmin, fstart);
I915_WRITE(MEMINTREN, MEMINT_CX_SUPR_EN | MEMINT_EVAL_CHG_EN);
status = connector_status_connected;
}
- return bit;
+ return status;
}
/**
extern void intel_init_clock_gating(struct drm_device *dev);
extern void ironlake_enable_drps(struct drm_device *dev);
extern void ironlake_disable_drps(struct drm_device *dev);
+extern void intel_init_emon(struct drm_device *dev);
extern int intel_pin_and_fence_fb_obj(struct drm_device *dev,
struct drm_gem_object *obj,
struct drm_device *dev = connector->dev;
struct drm_display_mode *mode;
- if (intel_lvds->edid) {
- drm_mode_connector_update_edid_property(connector,
- intel_lvds->edid);
+ if (intel_lvds->edid)
return drm_add_edid_modes(connector, intel_lvds->edid);
- }
mode = drm_mode_duplicate(dev, intel_lvds->fixed_mode);
if (mode == 0)
*/
intel_lvds->edid = drm_get_edid(connector,
&dev_priv->gmbus[pin].adapter);
-
+ if (intel_lvds->edid) {
+ if (drm_add_edid_modes(connector,
+ intel_lvds->edid)) {
+ drm_mode_connector_update_edid_property(connector,
+ intel_lvds->edid);
+ } else {
+ kfree(intel_lvds->edid);
+ intel_lvds->edid = NULL;
+ }
+ }
if (!intel_lvds->edid) {
/* Didn't get an EDID, so
* Set wide sync ranges so we get all modes
return 0;
err_out:
- iounmap(opregion->header);
+ iounmap(base);
return err;
}
{
int uv_hscale = uv_hsubsampling(rec->flags);
int uv_vscale = uv_vsubsampling(rec->flags);
- u32 stride_mask, depth, tmp;
+ u32 stride_mask;
+ int depth;
+ u32 tmp;
/* check src dimensions */
if (IS_845G(dev) || IS_I830(dev)) {
I915_WRITE_CTL(ring,
((ring->gem_object->size - PAGE_SIZE) & RING_NR_PAGES)
- | RING_NO_REPORT | RING_VALID);
+ | RING_REPORT_64K | RING_VALID);
head = I915_READ_HEAD(ring) & HEAD_ADDR;
/* If the head is still not zero, the ring is dead */
i915_gem_object_unpin(ring->gem_object);
drm_gem_object_unreference(ring->gem_object);
ring->gem_object = NULL;
+
+ if (ring->cleanup)
+ ring->cleanup(ring);
+
cleanup_status_page(dev, ring);
}
{
unsigned long end;
drm_i915_private_t *dev_priv = dev->dev_private;
+ u32 head;
+
+ head = intel_read_status_page(ring, 4);
+ if (head) {
+ ring->head = head & HEAD_ADDR;
+ ring->space = ring->head - (ring->tail + 8);
+ if (ring->space < 0)
+ ring->space += ring->size;
+ if (ring->space >= n)
+ return 0;
+ }
trace_i915_ring_wait_begin (dev);
end = jiffies + 3 * HZ;
/* do nothing */
}
+
+/* Workaround for some stepping of SNB,
+ * each time when BLT engine ring tail moved,
+ * the first command in the ring to be parsed
+ * should be MI_BATCH_BUFFER_START
+ */
+#define NEED_BLT_WORKAROUND(dev) \
+ (IS_GEN6(dev) && (dev->pdev->revision < 8))
+
+static inline struct drm_i915_gem_object *
+to_blt_workaround(struct intel_ring_buffer *ring)
+{
+ return ring->private;
+}
+
+static int blt_ring_init(struct drm_device *dev,
+ struct intel_ring_buffer *ring)
+{
+ if (NEED_BLT_WORKAROUND(dev)) {
+ struct drm_i915_gem_object *obj;
+ u32 __iomem *ptr;
+ int ret;
+
+ obj = to_intel_bo(i915_gem_alloc_object(dev, 4096));
+ if (obj == NULL)
+ return -ENOMEM;
+
+ ret = i915_gem_object_pin(&obj->base, 4096);
+ if (ret) {
+ drm_gem_object_unreference(&obj->base);
+ return ret;
+ }
+
+ ptr = kmap(obj->pages[0]);
+ iowrite32(MI_BATCH_BUFFER_END, ptr);
+ iowrite32(MI_NOOP, ptr+1);
+ kunmap(obj->pages[0]);
+
+ ret = i915_gem_object_set_to_gtt_domain(&obj->base, false);
+ if (ret) {
+ i915_gem_object_unpin(&obj->base);
+ drm_gem_object_unreference(&obj->base);
+ return ret;
+ }
+
+ ring->private = obj;
+ }
+
+ return init_ring_common(dev, ring);
+}
+
+static void blt_ring_begin(struct drm_device *dev,
+ struct intel_ring_buffer *ring,
+ int num_dwords)
+{
+ if (ring->private) {
+ intel_ring_begin(dev, ring, num_dwords+2);
+ intel_ring_emit(dev, ring, MI_BATCH_BUFFER_START);
+ intel_ring_emit(dev, ring, to_blt_workaround(ring)->gtt_offset);
+ } else
+ intel_ring_begin(dev, ring, 4);
+}
+
+static void blt_ring_flush(struct drm_device *dev,
+ struct intel_ring_buffer *ring,
+ u32 invalidate_domains,
+ u32 flush_domains)
+{
+ blt_ring_begin(dev, ring, 4);
+ intel_ring_emit(dev, ring, MI_FLUSH_DW);
+ intel_ring_emit(dev, ring, 0);
+ intel_ring_emit(dev, ring, 0);
+ intel_ring_emit(dev, ring, 0);
+ intel_ring_advance(dev, ring);
+}
+
+static u32
+blt_ring_add_request(struct drm_device *dev,
+ struct intel_ring_buffer *ring,
+ u32 flush_domains)
+{
+ u32 seqno = i915_gem_get_seqno(dev);
+
+ blt_ring_begin(dev, ring, 4);
+ intel_ring_emit(dev, ring, MI_STORE_DWORD_INDEX);
+ intel_ring_emit(dev, ring,
+ I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
+ intel_ring_emit(dev, ring, seqno);
+ intel_ring_emit(dev, ring, MI_USER_INTERRUPT);
+ intel_ring_advance(dev, ring);
+
+ DRM_DEBUG_DRIVER("%s %d\n", ring->name, seqno);
+ return seqno;
+}
+
+static void blt_ring_cleanup(struct intel_ring_buffer *ring)
+{
+ if (!ring->private)
+ return;
+
+ i915_gem_object_unpin(ring->private);
+ drm_gem_object_unreference(ring->private);
+ ring->private = NULL;
+}
+
static const struct intel_ring_buffer gen6_blt_ring = {
.name = "blt ring",
.id = RING_BLT,
.mmio_base = BLT_RING_BASE,
.size = 32 * PAGE_SIZE,
- .init = init_ring_common,
+ .init = blt_ring_init,
.write_tail = ring_write_tail,
- .flush = gen6_ring_flush,
- .add_request = ring_add_request,
+ .flush = blt_ring_flush,
+ .add_request = blt_ring_add_request,
.get_seqno = ring_status_page_get_seqno,
.user_irq_get = blt_ring_get_user_irq,
.user_irq_put = blt_ring_put_user_irq,
.dispatch_gem_execbuffer = gen6_ring_dispatch_gem_execbuffer,
+ .cleanup = blt_ring_cleanup,
};
int intel_init_render_ring_buffer(struct drm_device *dev)
struct drm_i915_gem_execbuffer2 *exec,
struct drm_clip_rect *cliprects,
uint64_t exec_offset);
+ void (*cleanup)(struct intel_ring_buffer *ring);
/**
* List of objects currently involved in rendering from the
wait_queue_head_t irq_queue;
drm_local_map_t map;
+
+ void *private;
};
static inline u32
u32 grbm_int_cntl = 0;
if (!rdev->irq.installed) {
- WARN(1, "Can't enable IRQ/MSI because no handler is installed.\n");
+ WARN(1, "Can't enable IRQ/MSI because no handler is installed\n");
return -EINVAL;
}
/* don't enable anything if the ih is disabled */
case 0: /* D1 vblank */
if (disp_int & LB_D1_VBLANK_INTERRUPT) {
drm_handle_vblank(rdev->ddev, 0);
+ rdev->pm.vblank_sync = true;
wake_up(&rdev->irq.vblank_queue);
disp_int &= ~LB_D1_VBLANK_INTERRUPT;
DRM_DEBUG("IH: D1 vblank\n");
case 0: /* D2 vblank */
if (disp_int_cont & LB_D2_VBLANK_INTERRUPT) {
drm_handle_vblank(rdev->ddev, 1);
+ rdev->pm.vblank_sync = true;
wake_up(&rdev->irq.vblank_queue);
disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
DRM_DEBUG("IH: D2 vblank\n");
case 0: /* D3 vblank */
if (disp_int_cont2 & LB_D3_VBLANK_INTERRUPT) {
drm_handle_vblank(rdev->ddev, 2);
+ rdev->pm.vblank_sync = true;
wake_up(&rdev->irq.vblank_queue);
disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
DRM_DEBUG("IH: D3 vblank\n");
case 0: /* D4 vblank */
if (disp_int_cont3 & LB_D4_VBLANK_INTERRUPT) {
drm_handle_vblank(rdev->ddev, 3);
+ rdev->pm.vblank_sync = true;
wake_up(&rdev->irq.vblank_queue);
disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
DRM_DEBUG("IH: D4 vblank\n");
case 0: /* D5 vblank */
if (disp_int_cont4 & LB_D5_VBLANK_INTERRUPT) {
drm_handle_vblank(rdev->ddev, 4);
+ rdev->pm.vblank_sync = true;
wake_up(&rdev->irq.vblank_queue);
disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
DRM_DEBUG("IH: D5 vblank\n");
case 0: /* D6 vblank */
if (disp_int_cont5 & LB_D6_VBLANK_INTERRUPT) {
drm_handle_vblank(rdev->ddev, 5);
+ rdev->pm.vblank_sync = true;
wake_up(&rdev->irq.vblank_queue);
disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
DRM_DEBUG("IH: D6 vblank\n");
int r;
if (rdev->gart.table.ram.ptr) {
- WARN(1, "R100 PCI GART already initialized.\n");
+ WARN(1, "R100 PCI GART already initialized\n");
return 0;
}
/* Initialize common gart structure */
uint32_t tmp = 0;
if (!rdev->irq.installed) {
- WARN(1, "Can't enable IRQ/MSI because no handler is installed.\n");
+ WARN(1, "Can't enable IRQ/MSI because no handler is installed\n");
WREG32(R_000040_GEN_INT_CNTL, 0);
return -EINVAL;
}
int r;
if (rdev->gart.table.vram.robj) {
- WARN(1, "RV370 PCIE GART already initialized.\n");
+ WARN(1, "RV370 PCIE GART already initialized\n");
return 0;
}
/* Initialize common gart structure */
{
u32 temp = (RREG32(CG_THERMAL_STATUS) & ASIC_T_MASK) >>
ASIC_T_SHIFT;
- u32 actual_temp = 0;
- if ((temp >> 7) & 1)
- actual_temp = 0;
- else
- actual_temp = (temp >> 1) & 0xff;
-
- return actual_temp * 1000;
+ return temp * 1000;
}
void r600_pm_get_dynpm_state(struct radeon_device *rdev)
int r;
if (rdev->gart.table.vram.robj) {
- WARN(1, "R600 PCIE GART already initialized.\n");
+ WARN(1, "R600 PCIE GART already initialized\n");
return 0;
}
/* Initialize common gart structure */
u32 hdmi1, hdmi2;
if (!rdev->irq.installed) {
- WARN(1, "Can't enable IRQ/MSI because no handler is installed.\n");
+ WARN(1, "Can't enable IRQ/MSI because no handler is installed\n");
return -EINVAL;
}
/* don't enable anything if the ih is disabled */
if (crev < 2)
return false;
- router.valid = false;
-
obj_header = (ATOM_OBJECT_HEADER *) (ctx->bios + data_offset);
path_obj = (ATOM_DISPLAY_OBJECT_PATH_TABLE *)
(ctx->bios + data_offset +
if (connector_type == DRM_MODE_CONNECTOR_Unknown)
continue;
+ router.ddc_valid = false;
+ router.cd_valid = false;
for (j = 0; j < ((le16_to_cpu(path->usSize) - 8) / 2); j++) {
uint8_t grph_obj_id, grph_obj_num, grph_obj_type;
usDeviceTag));
} else if (grph_obj_type == GRAPH_OBJECT_TYPE_ROUTER) {
- router.valid = false;
for (k = 0; k < router_obj->ucNumberOfObjects; k++) {
- u16 router_obj_id = le16_to_cpu(router_obj->asObjects[j].usObjectID);
+ u16 router_obj_id = le16_to_cpu(router_obj->asObjects[k].usObjectID);
if (le16_to_cpu(path->usGraphicObjIds[j]) == router_obj_id) {
ATOM_COMMON_RECORD_HEADER *record = (ATOM_COMMON_RECORD_HEADER *)
(ctx->bios + data_offset +
ATOM_I2C_RECORD *i2c_record;
ATOM_I2C_ID_CONFIG_ACCESS *i2c_config;
ATOM_ROUTER_DDC_PATH_SELECT_RECORD *ddc_path;
+ ATOM_ROUTER_DATA_CLOCK_PATH_SELECT_RECORD *cd_path;
ATOM_SRC_DST_TABLE_FOR_ONE_OBJECT *router_src_dst_table =
(ATOM_SRC_DST_TABLE_FOR_ONE_OBJECT *)
(ctx->bios + data_offset +
case ATOM_ROUTER_DDC_PATH_SELECT_RECORD_TYPE:
ddc_path = (ATOM_ROUTER_DDC_PATH_SELECT_RECORD *)
record;
- router.valid = true;
- router.mux_type = ddc_path->ucMuxType;
- router.mux_control_pin = ddc_path->ucMuxControlPin;
- router.mux_state = ddc_path->ucMuxState[enum_id];
+ router.ddc_valid = true;
+ router.ddc_mux_type = ddc_path->ucMuxType;
+ router.ddc_mux_control_pin = ddc_path->ucMuxControlPin;
+ router.ddc_mux_state = ddc_path->ucMuxState[enum_id];
+ break;
+ case ATOM_ROUTER_DATA_CLOCK_PATH_SELECT_RECORD_TYPE:
+ cd_path = (ATOM_ROUTER_DATA_CLOCK_PATH_SELECT_RECORD *)
+ record;
+ router.cd_valid = true;
+ router.cd_mux_type = cd_path->ucMuxType;
+ router.cd_mux_control_pin = cd_path->ucMuxControlPin;
+ router.cd_mux_state = cd_path->ucMuxState[enum_id];
break;
}
record = (ATOM_COMMON_RECORD_HEADER *)
size_t bc_size = sizeof(*bios_connectors) * ATOM_MAX_SUPPORTED_DEVICE;
struct radeon_router router;
- router.valid = false;
+ router.ddc_valid = false;
+ router.cd_valid = false;
bios_connectors = kzalloc(bc_size, GFP_KERNEL);
if (!bios_connectors)
continue;
if (priority == true) {
- DRM_INFO("1: conflicting encoders switching off %s\n", drm_get_connector_name(conflict));
- DRM_INFO("in favor of %s\n", drm_get_connector_name(connector));
+ DRM_DEBUG_KMS("1: conflicting encoders switching off %s\n", drm_get_connector_name(conflict));
+ DRM_DEBUG_KMS("in favor of %s\n", drm_get_connector_name(connector));
conflict->status = connector_status_disconnected;
radeon_connector_update_scratch_regs(conflict, connector_status_disconnected);
} else {
- DRM_INFO("2: conflicting encoders switching off %s\n", drm_get_connector_name(connector));
- DRM_INFO("in favor of %s\n", drm_get_connector_name(conflict));
+ DRM_DEBUG_KMS("2: conflicting encoders switching off %s\n", drm_get_connector_name(connector));
+ DRM_DEBUG_KMS("in favor of %s\n", drm_get_connector_name(conflict));
current_status = connector_status_disconnected;
}
break;
mode->vdisplay == native_mode->vdisplay) {
*native_mode = *mode;
drm_mode_set_crtcinfo(native_mode, CRTC_INTERLACE_HALVE_V);
- DRM_INFO("Determined LVDS native mode details from EDID\n");
+ DRM_DEBUG_KMS("Determined LVDS native mode details from EDID\n");
break;
}
}
}
if (!native_mode->clock) {
- DRM_INFO("No LVDS native mode details, disabling RMX\n");
+ DRM_DEBUG_KMS("No LVDS native mode details, disabling RMX\n");
radeon_encoder->rmx_type = RMX_OFF;
}
}
radeon_connector->shared_ddc = true;
shared_ddc = true;
}
- if (radeon_connector->router_bus && router->valid &&
+ if (radeon_connector->router_bus && router->ddc_valid &&
(radeon_connector->router.router_id == router->router_id)) {
radeon_connector->shared_ddc = false;
shared_ddc = false;
radeon_connector->connector_object_id = connector_object_id;
radeon_connector->hpd = *hpd;
radeon_connector->router = *router;
- if (router->valid) {
+ if (router->ddc_valid || router->cd_valid) {
radeon_connector->router_bus = radeon_i2c_lookup(rdev, &router->i2c_info);
if (!radeon_connector->router_bus)
goto failed;
radeon_connector->ddc_bus->rec.en_data_reg,
radeon_connector->ddc_bus->rec.y_clk_reg,
radeon_connector->ddc_bus->rec.y_data_reg);
- if (radeon_connector->router_bus)
+ if (radeon_connector->router.ddc_valid)
DRM_INFO(" DDC Router 0x%x/0x%x\n",
- radeon_connector->router.mux_control_pin,
- radeon_connector->router.mux_state);
+ radeon_connector->router.ddc_mux_control_pin,
+ radeon_connector->router.ddc_mux_state);
+ if (radeon_connector->router.cd_valid)
+ DRM_INFO(" Clock/Data Router 0x%x/0x%x\n",
+ radeon_connector->router.cd_mux_control_pin,
+ radeon_connector->router.cd_mux_state);
} else {
if (connector->connector_type == DRM_MODE_CONNECTOR_VGA ||
connector->connector_type == DRM_MODE_CONNECTOR_DVII ||
int ret = 0;
/* on hw with routers, select right port */
- if (radeon_connector->router.valid)
- radeon_router_select_port(radeon_connector);
+ if (radeon_connector->router.ddc_valid)
+ radeon_router_select_ddc_port(radeon_connector);
if ((radeon_connector->base.connector_type == DRM_MODE_CONNECTOR_DisplayPort) ||
(radeon_connector->base.connector_type == DRM_MODE_CONNECTOR_eDP)) {
int ret = 0;
/* on hw with routers, select right port */
- if (radeon_connector->router.valid)
- radeon_router_select_port(radeon_connector);
+ if (radeon_connector->router.ddc_valid)
+ radeon_router_select_ddc_port(radeon_connector);
if (!radeon_connector->ddc_bus)
return -1;
static void radeon_atom_encoder_prepare(struct drm_encoder *encoder)
{
struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
+ struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
if (radeon_encoder->active_device &
(ATOM_DEVICE_DFP_SUPPORT | ATOM_DEVICE_LCD_SUPPORT)) {
radeon_atom_output_lock(encoder, true);
radeon_atom_encoder_dpms(encoder, DRM_MODE_DPMS_OFF);
+ /* select the clock/data port if it uses a router */
+ if (connector) {
+ struct radeon_connector *radeon_connector = to_radeon_connector(connector);
+ if (radeon_connector->router.cd_valid)
+ radeon_router_select_cd_port(radeon_connector);
+ }
+
/* this is needed for the pll/ss setup to work correctly in some cases */
atombios_set_encoder_crtc_source(encoder);
}
struct radeon_device *rdev = dev->dev_private;
struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
struct radeon_encoder_atom_dig *dig;
+
+ /* check for pre-DCE3 cards with shared encoders;
+ * can't really use the links individually, so don't disable
+ * the encoder if it's in use by another connector
+ */
+ if (!ASIC_IS_DCE3(rdev)) {
+ struct drm_encoder *other_encoder;
+ struct radeon_encoder *other_radeon_encoder;
+
+ list_for_each_entry(other_encoder, &dev->mode_config.encoder_list, head) {
+ other_radeon_encoder = to_radeon_encoder(other_encoder);
+ if ((radeon_encoder->encoder_id == other_radeon_encoder->encoder_id) &&
+ drm_helper_encoder_in_use(other_encoder))
+ goto disable_done;
+ }
+ }
+
radeon_atom_encoder_dpms(encoder, DRM_MODE_DPMS_OFF);
switch (radeon_encoder->encoder_id) {
break;
}
+disable_done:
if (radeon_encoder_is_digital(encoder)) {
if (atombios_get_encoder_mode(encoder) == ATOM_ENCODER_MODE_HDMI)
r600_hdmi_disable(encoder);
*/
if (seq == rdev->fence_drv.last_seq && radeon_gpu_is_lockup(rdev)) {
/* good news we believe it's a lockup */
- WARN(1, "GPU lockup (waiting for 0x%08X last fence id 0x%08X)\n", fence->seq, seq);
+ WARN(1, "GPU lockup (waiting for 0x%08X last fence id 0x%08X)\n",
+ fence->seq, seq);
/* FIXME: what should we do ? marking everyone
* as signaled for now
*/
};
/* on hw with routers, select right port */
- if (radeon_connector->router.valid)
- radeon_router_select_port(radeon_connector);
+ if (radeon_connector->router.ddc_valid)
+ radeon_router_select_ddc_port(radeon_connector);
ret = i2c_transfer(&radeon_connector->ddc_bus->adapter, msgs, 2);
if (ret == 2)
addr, val);
}
-/* router switching */
-void radeon_router_select_port(struct radeon_connector *radeon_connector)
+/* ddc router switching */
+void radeon_router_select_ddc_port(struct radeon_connector *radeon_connector)
{
u8 val;
- if (!radeon_connector->router.valid)
+ if (!radeon_connector->router.ddc_valid)
return;
radeon_i2c_get_byte(radeon_connector->router_bus,
radeon_connector->router.i2c_addr,
0x3, &val);
- val &= radeon_connector->router.mux_control_pin;
+ val &= ~radeon_connector->router.ddc_mux_control_pin;
radeon_i2c_put_byte(radeon_connector->router_bus,
radeon_connector->router.i2c_addr,
0x3, val);
radeon_i2c_get_byte(radeon_connector->router_bus,
radeon_connector->router.i2c_addr,
0x1, &val);
- val &= radeon_connector->router.mux_control_pin;
- val |= radeon_connector->router.mux_state;
+ val &= ~radeon_connector->router.ddc_mux_control_pin;
+ val |= radeon_connector->router.ddc_mux_state;
+ radeon_i2c_put_byte(radeon_connector->router_bus,
+ radeon_connector->router.i2c_addr,
+ 0x1, val);
+}
+
+/* clock/data router switching */
+void radeon_router_select_cd_port(struct radeon_connector *radeon_connector)
+{
+ u8 val;
+
+ if (!radeon_connector->router.cd_valid)
+ return;
+
+ radeon_i2c_get_byte(radeon_connector->router_bus,
+ radeon_connector->router.i2c_addr,
+ 0x3, &val);
+ val &= ~radeon_connector->router.cd_mux_control_pin;
+ radeon_i2c_put_byte(radeon_connector->router_bus,
+ radeon_connector->router.i2c_addr,
+ 0x3, val);
+ radeon_i2c_get_byte(radeon_connector->router_bus,
+ radeon_connector->router.i2c_addr,
+ 0x1, &val);
+ val &= ~radeon_connector->router.cd_mux_control_pin;
+ val |= radeon_connector->router.cd_mux_state;
radeon_i2c_put_byte(radeon_connector->router_bus,
radeon_connector->router.i2c_addr,
0x1, val);
};
struct radeon_router {
- bool valid;
u32 router_id;
struct radeon_i2c_bus_rec i2c_info;
u8 i2c_addr;
- u8 mux_type;
- u8 mux_control_pin;
- u8 mux_state;
+ /* i2c mux */
+ bool ddc_valid;
+ u8 ddc_mux_type;
+ u8 ddc_mux_control_pin;
+ u8 ddc_mux_state;
+ /* clock/data mux */
+ bool cd_valid;
+ u8 cd_mux_type;
+ u8 cd_mux_control_pin;
+ u8 cd_mux_state;
};
struct radeon_connector {
u8 slave_addr,
u8 addr,
u8 val);
-extern void radeon_router_select_port(struct radeon_connector *radeon_connector);
+extern void radeon_router_select_ddc_port(struct radeon_connector *radeon_connector);
+extern void radeon_router_select_cd_port(struct radeon_connector *radeon_connector);
extern bool radeon_ddc_probe(struct radeon_connector *radeon_connector);
extern int radeon_ddc_get_modes(struct radeon_connector *radeon_connector);
type = ttm_bo_type_device;
}
*bo_ptr = NULL;
+
+retry:
bo = kzalloc(sizeof(struct radeon_bo), GFP_KERNEL);
if (bo == NULL)
return -ENOMEM;
bo->gobj = gobj;
bo->surface_reg = -1;
INIT_LIST_HEAD(&bo->list);
-
-retry:
radeon_ttm_placement_from_domain(bo, domain);
/* Kernel allocation are uninterruptible */
mutex_lock(&rdev->vram_mutex);
gtt = container_of(backend, struct radeon_ttm_backend, backend);
gtt->offset = bo_mem->start << PAGE_SHIFT;
if (!gtt->num_pages) {
- WARN(1, "nothing to bind %lu pages for mreg %p back %p!\n", gtt->num_pages, bo_mem, backend);
+ WARN(1, "nothing to bind %lu pages for mreg %p back %p!\n",
+ gtt->num_pages, bo_mem, backend);
}
r = radeon_gart_bind(gtt->rdev, gtt->offset,
gtt->num_pages, gtt->pages);
int r;
if (rdev->gart.table.ram.ptr) {
- WARN(1, "RS400 GART already initialized.\n");
+ WARN(1, "RS400 GART already initialized\n");
return 0;
}
/* Check gart size */
int r;
if (rdev->gart.table.vram.robj) {
- WARN(1, "RS600 GART already initialized.\n");
+ WARN(1, "RS600 GART already initialized\n");
return 0;
}
/* Initialize common gart structure */
~S_007D18_DC_HOT_PLUG_DETECT2_INT_EN(1);
if (!rdev->irq.installed) {
- WARN(1, "Can't enable IRQ/MSI because no handler is installed.\n");
+ WARN(1, "Can't enable IRQ/MSI because no handler is installed\n");
WREG32(R_000040_GEN_INT_CNTL, 0);
return -EINVAL;
}
/*
* Authors: Thomas Hellstrom <thellstrom-at-vmware-dot-com>
*/
-/* Notes:
- *
- * We store bo pointer in drm_mm_node struct so we know which bo own a
- * specific node. There is no protection on the pointer, thus to make
- * sure things don't go berserk you have to access this pointer while
- * holding the global lru lock and make sure anytime you free a node you
- * reset the pointer to NULL.
- */
#include "ttm/ttm_module.h"
#include "ttm/ttm_bo_driver.h"
#include <linux/mm.h>
#include <linux/file.h>
#include <linux/module.h>
+#include <asm/atomic.h>
#define TTM_ASSERT_LOCKED(param)
#define TTM_DEBUG(fmt, arg...)
ttm_bo_mem_put(bo, &bo->mem);
atomic_set(&bo->reserved, 0);
+
+ /*
+ * Make processes trying to reserve really pick it up.
+ */
+ smp_mb__after_atomic_dec();
wake_up_all(&bo->event_queue);
}
struct ttm_bo_device *bdev = bo->bdev;
struct ttm_bo_global *glob = bo->glob;
struct ttm_bo_driver *driver;
- void *sync_obj;
+ void *sync_obj = NULL;
void *sync_obj_arg;
int put_count;
int ret;
spin_lock(&glob->lru_lock);
}
queue:
- sync_obj = bo->sync_obj;
- sync_obj_arg = bo->sync_obj_arg;
driver = bdev->driver;
+ if (bo->sync_obj)
+ sync_obj = driver->sync_obj_ref(bo->sync_obj);
+ sync_obj_arg = bo->sync_obj_arg;
kref_get(&bo->list_kref);
list_add_tail(&bo->ddestroy, &bdev->ddestroy);
spin_unlock(&glob->lru_lock);
spin_unlock(&bo->lock);
- if (sync_obj)
+ if (sync_obj) {
driver->sync_obj_flush(sync_obj, sync_obj_arg);
+ driver->sync_obj_unref(&sync_obj);
+ }
schedule_delayed_work(&bdev->wq,
((HZ / 100) < 1) ? 1 : HZ / 100);
}
bool no_wait_gpu)
{
struct ttm_bo_device *bdev = bo->bdev;
- struct ttm_bo_global *glob = bdev->glob;
struct ttm_mem_type_manager *man = &bdev->man[mem_type];
int ret;
return ret;
if (mem->mm_node)
break;
- spin_lock(&glob->lru_lock);
- if (list_empty(&man->lru)) {
- spin_unlock(&glob->lru_lock);
- break;
- }
- spin_unlock(&glob->lru_lock);
ret = ttm_mem_evict_first(bdev, mem_type, interruptible,
no_wait_reserve, no_wait_gpu);
if (unlikely(ret != 0))
int ttm_bo_check_placement(struct ttm_buffer_object *bo,
struct ttm_placement *placement)
{
- int i;
+ BUG_ON((placement->fpfn || placement->lpfn) &&
+ (bo->mem.num_pages > (placement->lpfn - placement->fpfn)));
- if (placement->fpfn || placement->lpfn) {
- if (bo->mem.num_pages > (placement->lpfn - placement->fpfn)) {
- printk(KERN_ERR TTM_PFX "Page number range to small "
- "Need %lu pages, range is [%u, %u]\n",
- bo->mem.num_pages, placement->fpfn,
- placement->lpfn);
- return -EINVAL;
- }
- }
- for (i = 0; i < placement->num_placement; i++) {
- if (!capable(CAP_SYS_ADMIN)) {
- if (placement->placement[i] & TTM_PL_FLAG_NO_EVICT) {
- printk(KERN_ERR TTM_PFX "Need to be root to "
- "modify NO_EVICT status.\n");
- return -EINVAL;
- }
- }
- }
- for (i = 0; i < placement->num_busy_placement; i++) {
- if (!capable(CAP_SYS_ADMIN)) {
- if (placement->busy_placement[i] & TTM_PL_FLAG_NO_EVICT) {
- printk(KERN_ERR TTM_PFX "Need to be root to "
- "modify NO_EVICT status.\n");
- return -EINVAL;
- }
- }
- }
return 0;
}
num_pages = (size + PAGE_SIZE - 1) >> PAGE_SHIFT;
if (num_pages == 0) {
printk(KERN_ERR TTM_PFX "Illegal buffer object size.\n");
+ if (destroy)
+ (*destroy)(bo);
+ else
+ kfree(bo);
return -EINVAL;
}
bo->destroy = destroy;
int ret = -EINVAL;
struct ttm_mem_type_manager *man;
- if (type >= TTM_NUM_MEM_TYPES) {
- printk(KERN_ERR TTM_PFX "Illegal memory type %d\n", type);
- return ret;
- }
-
+ BUG_ON(type >= TTM_NUM_MEM_TYPES);
man = &bdev->man[type];
- if (man->has_type) {
- printk(KERN_ERR TTM_PFX
- "Memory manager already initialized for type %d\n",
- type);
- return ret;
- }
+ BUG_ON(man->has_type);
ret = bdev->driver->init_mem_type(bdev, type, man);
if (ret)
ret = 0;
if (type != TTM_PL_SYSTEM) {
- if (!p_size) {
- printk(KERN_ERR TTM_PFX
- "Zero size memory manager type %d\n",
- type);
- return ret;
- }
-
ret = (*man->func->init)(man, p_size);
if (ret)
return ret;
/**************************************************************************
*
- * Copyright (c) 2007-2009 VMware, Inc., Palo Alto, CA., USA
+ * Copyright (c) 2007-2010 VMware, Inc., Palo Alto, CA., USA
* All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a
#include "ttm/ttm_module.h"
#include "ttm/ttm_bo_driver.h"
#include "ttm/ttm_placement.h"
-#include <linux/jiffies.h>
+#include "drm_mm.h"
#include <linux/slab.h>
-#include <linux/sched.h>
-#include <linux/mm.h>
-#include <linux/file.h>
+#include <linux/spinlock.h>
#include <linux/module.h>
+/**
+ * Currently we use a spinlock for the lock, but a mutex *may* be
+ * more appropriate to reduce scheduling latency if the range manager
+ * ends up with very fragmented allocation patterns.
+ */
+
+struct ttm_range_manager {
+ struct drm_mm mm;
+ spinlock_t lock;
+};
+
static int ttm_bo_man_get_node(struct ttm_mem_type_manager *man,
struct ttm_buffer_object *bo,
struct ttm_placement *placement,
struct ttm_mem_reg *mem)
{
- struct ttm_bo_global *glob = man->bdev->glob;
- struct drm_mm *mm = man->priv;
+ struct ttm_range_manager *rman = (struct ttm_range_manager *) man->priv;
+ struct drm_mm *mm = &rman->mm;
struct drm_mm_node *node = NULL;
unsigned long lpfn;
int ret;
if (unlikely(ret))
return ret;
- spin_lock(&glob->lru_lock);
+ spin_lock(&rman->lock);
node = drm_mm_search_free_in_range(mm,
mem->num_pages, mem->page_alignment,
placement->fpfn, lpfn, 1);
if (unlikely(node == NULL)) {
- spin_unlock(&glob->lru_lock);
+ spin_unlock(&rman->lock);
return 0;
}
node = drm_mm_get_block_atomic_range(node, mem->num_pages,
- mem->page_alignment,
- placement->fpfn,
- lpfn);
- spin_unlock(&glob->lru_lock);
+ mem->page_alignment,
+ placement->fpfn,
+ lpfn);
+ spin_unlock(&rman->lock);
} while (node == NULL);
mem->mm_node = node;
static void ttm_bo_man_put_node(struct ttm_mem_type_manager *man,
struct ttm_mem_reg *mem)
{
- struct ttm_bo_global *glob = man->bdev->glob;
+ struct ttm_range_manager *rman = (struct ttm_range_manager *) man->priv;
if (mem->mm_node) {
- spin_lock(&glob->lru_lock);
+ spin_lock(&rman->lock);
drm_mm_put_block(mem->mm_node);
- spin_unlock(&glob->lru_lock);
+ spin_unlock(&rman->lock);
mem->mm_node = NULL;
}
}
static int ttm_bo_man_init(struct ttm_mem_type_manager *man,
unsigned long p_size)
{
- struct drm_mm *mm;
+ struct ttm_range_manager *rman;
int ret;
- mm = kzalloc(sizeof(*mm), GFP_KERNEL);
- if (!mm)
+ rman = kzalloc(sizeof(*rman), GFP_KERNEL);
+ if (!rman)
return -ENOMEM;
- ret = drm_mm_init(mm, 0, p_size);
+ ret = drm_mm_init(&rman->mm, 0, p_size);
if (ret) {
- kfree(mm);
+ kfree(rman);
return ret;
}
- man->priv = mm;
+ spin_lock_init(&rman->lock);
+ man->priv = rman;
return 0;
}
static int ttm_bo_man_takedown(struct ttm_mem_type_manager *man)
{
- struct ttm_bo_global *glob = man->bdev->glob;
- struct drm_mm *mm = man->priv;
- int ret = 0;
+ struct ttm_range_manager *rman = (struct ttm_range_manager *) man->priv;
+ struct drm_mm *mm = &rman->mm;
- spin_lock(&glob->lru_lock);
+ spin_lock(&rman->lock);
if (drm_mm_clean(mm)) {
drm_mm_takedown(mm);
- kfree(mm);
+ spin_unlock(&rman->lock);
+ kfree(rman);
man->priv = NULL;
- } else
- ret = -EBUSY;
- spin_unlock(&glob->lru_lock);
- return ret;
+ return 0;
+ }
+ spin_unlock(&rman->lock);
+ return -EBUSY;
}
static void ttm_bo_man_debug(struct ttm_mem_type_manager *man,
const char *prefix)
{
- struct ttm_bo_global *glob = man->bdev->glob;
- struct drm_mm *mm = man->priv;
+ struct ttm_range_manager *rman = (struct ttm_range_manager *) man->priv;
- spin_lock(&glob->lru_lock);
- drm_mm_debug_table(mm, prefix);
- spin_unlock(&glob->lru_lock);
+ spin_lock(&rman->lock);
+ drm_mm_debug_table(&rman->mm, prefix);
+ spin_unlock(&rman->lock);
}
const struct ttm_mem_type_manager_func ttm_bo_manager_func = {
return ret;
ret = be->func->bind(be, bo_mem);
- if (ret) {
- printk(KERN_ERR TTM_PFX "Couldn't bind backend.\n");
+ if (unlikely(ret != 0))
return ret;
- }
ttm->state = tt_bound;
vsg->num_pages = VIA_PFN(xfer->mem_addr + (xfer->num_lines * xfer->mem_stride - 1)) -
first_pfn + 1;
- if (NULL == (vsg->pages = vmalloc(sizeof(struct page *) * vsg->num_pages)))
+ vsg->pages = vzalloc(sizeof(struct page *) * vsg->num_pages);
+ if (NULL == vsg->pages)
return -ENOMEM;
- memset(vsg->pages, 0, sizeof(struct page *) * vsg->num_pages);
down_read(¤t->mm->mmap_sem);
ret = get_user_pages(current, current->mm,
(unsigned long)xfer->mem_addr,
fence_rep.error = ret;
fence_rep.fence_seq = (uint64_t) sequence;
+ fence_rep.pad64 = 0;
user_fence_rep = (struct drm_vmw_fence_rep __user *)
(unsigned long)arg->fence_rep;
&vmw_vram_ne_placement,
false, &vmw_dmabuf_bo_free);
vmw_overlay_resume_all(dev_priv);
+ if (unlikely(ret != 0))
+ vfbs->buffer = NULL;
return ret;
}
struct vmw_framebuffer_surface *vfbs =
vmw_framebuffer_to_vfbs(&vfb->base);
+ if (unlikely(vfbs->buffer == NULL))
+ return 0;
+
bo = &vfbs->buffer->base;
ttm_bo_unref(&bo);
vfbs->buffer = NULL;
return -EINVAL;
}
- dev_priv->ldu_priv = kmalloc(GFP_KERNEL, sizeof(*dev_priv->ldu_priv));
+ dev_priv->ldu_priv = kmalloc(sizeof(*dev_priv->ldu_priv), GFP_KERNEL);
if (!dev_priv->ldu_priv)
return -ENOMEM;
return -ENOSYS;
}
- overlay = kmalloc(GFP_KERNEL, sizeof(*overlay));
+ overlay = kmalloc(sizeof(*overlay), GFP_KERNEL);
if (!overlay)
return -ENOMEM;
depends on PCI
# Poulsbo stub depends on ACPI_VIDEO when ACPI is enabled
# but for select to work, need to select ACPI_VIDEO's dependencies, ick
+ select VIDEO_OUTPUT_CONTROL if ACPI
+ select BACKLIGHT_CLASS_DEVICE if ACPI
+ select INPUT if ACPI
select ACPI_VIDEO if ACPI
help
Choose this option if you have a system that has Intel GMA500
val = i2c_smbus_read_byte_data(client, i);
if (unlikely(val < 0)) {
dev_dbg(dev,
- "Failed to read ADC value: error %d",
+ "Failed to read ADC value: error %d\n",
val);
ret = ERR_PTR(val);
goto abort;
return -ENODEV;
if (i2c_smbus_read_byte_data(client, LTC4261_STATUS) < 0) {
- dev_err(&client->dev, "Failed to read register %d:%02x:%02x\n",
- adapter->id, client->addr, LTC4261_STATUS);
+ dev_err(&client->dev, "Failed to read status register\n");
return -ENODEV;
}
ICH10
5/3400 Series (PCH)
Cougar Point (PCH)
+ Patsburg (PCH)
This driver can also be built as a module. If so, the module
will be called i2c-i801.
Philip Edelbrock <phil@netroedge.com>, and Mark D. Studebaker
<mdsxyz123@yahoo.com>
Copyright (C) 2007, 2008 Jean Delvare <khali@linux-fr.org>
+ Copyright (C) 2010 Intel Corporation,
+ David Woodhouse <dwmw2@infradead.org>
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
ICH10 0x3a60 32 hard yes yes yes
5/3400 Series (PCH) 0x3b30 32 hard yes yes yes
Cougar Point (PCH) 0x1c22 32 hard yes yes yes
+ Patsburg (PCH) 0x1d22 32 hard yes yes yes
+ Patsburg (PCH) IDF 0x1d70 32 hard yes yes yes
+ Patsburg (PCH) IDF 0x1d71 32 hard yes yes yes
+ Patsburg (PCH) IDF 0x1d72 32 hard yes yes yes
Features supported by this driver:
Software PEC no
Block buffer yes
Block process call transaction no
I2C block read transaction yes (doesn't use the block buffer)
+ Slave mode no
See the file Documentation/i2c/busses/i2c-i801 for details.
*/
-/* Note: we assume there can only be one I801, with one SMBus interface */
-
#include <linux/module.h>
#include <linux/pci.h>
#include <linux/kernel.h>
#include <linux/dmi.h>
/* I801 SMBus address offsets */
-#define SMBHSTSTS (0 + i801_smba)
-#define SMBHSTCNT (2 + i801_smba)
-#define SMBHSTCMD (3 + i801_smba)
-#define SMBHSTADD (4 + i801_smba)
-#define SMBHSTDAT0 (5 + i801_smba)
-#define SMBHSTDAT1 (6 + i801_smba)
-#define SMBBLKDAT (7 + i801_smba)
-#define SMBPEC (8 + i801_smba) /* ICH3 and later */
-#define SMBAUXSTS (12 + i801_smba) /* ICH4 and later */
-#define SMBAUXCTL (13 + i801_smba) /* ICH4 and later */
+#define SMBHSTSTS(p) (0 + (p)->smba)
+#define SMBHSTCNT(p) (2 + (p)->smba)
+#define SMBHSTCMD(p) (3 + (p)->smba)
+#define SMBHSTADD(p) (4 + (p)->smba)
+#define SMBHSTDAT0(p) (5 + (p)->smba)
+#define SMBHSTDAT1(p) (6 + (p)->smba)
+#define SMBBLKDAT(p) (7 + (p)->smba)
+#define SMBPEC(p) (8 + (p)->smba) /* ICH3 and later */
+#define SMBAUXSTS(p) (12 + (p)->smba) /* ICH4 and later */
+#define SMBAUXCTL(p) (13 + (p)->smba) /* ICH4 and later */
/* PCI Address Constants */
#define SMBBAR 4
SMBHSTSTS_BUS_ERR | SMBHSTSTS_DEV_ERR | \
SMBHSTSTS_INTR)
-static unsigned long i801_smba;
-static unsigned char i801_original_hstcfg;
+/* Patsburg also has three 'Integrated Device Function' SMBus controllers */
+#define PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS_IDF0 0x1d70
+#define PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS_IDF1 0x1d71
+#define PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS_IDF2 0x1d72
+
+struct i801_priv {
+ struct i2c_adapter adapter;
+ unsigned long smba;
+ unsigned char original_hstcfg;
+ struct pci_dev *pci_dev;
+ unsigned int features;
+};
+
static struct pci_driver i801_driver;
-static struct pci_dev *I801_dev;
#define FEATURE_SMBUS_PEC (1 << 0)
#define FEATURE_BLOCK_BUFFER (1 << 1)
#define FEATURE_BLOCK_PROC (1 << 2)
#define FEATURE_I2C_BLOCK_READ (1 << 3)
-static unsigned int i801_features;
static const char *i801_feature_names[] = {
"SMBus PEC",
/* Make sure the SMBus host is ready to start transmitting.
Return 0 if it is, -EBUSY if it is not. */
-static int i801_check_pre(void)
+static int i801_check_pre(struct i801_priv *priv)
{
int status;
- status = inb_p(SMBHSTSTS);
+ status = inb_p(SMBHSTSTS(priv));
if (status & SMBHSTSTS_HOST_BUSY) {
- dev_err(&I801_dev->dev, "SMBus is busy, can't use it!\n");
+ dev_err(&priv->pci_dev->dev, "SMBus is busy, can't use it!\n");
return -EBUSY;
}
status &= STATUS_FLAGS;
if (status) {
- dev_dbg(&I801_dev->dev, "Clearing status flags (%02x)\n",
+ dev_dbg(&priv->pci_dev->dev, "Clearing status flags (%02x)\n",
status);
- outb_p(status, SMBHSTSTS);
- status = inb_p(SMBHSTSTS) & STATUS_FLAGS;
+ outb_p(status, SMBHSTSTS(priv));
+ status = inb_p(SMBHSTSTS(priv)) & STATUS_FLAGS;
if (status) {
- dev_err(&I801_dev->dev,
+ dev_err(&priv->pci_dev->dev,
"Failed clearing status flags (%02x)\n",
status);
return -EBUSY;
}
/* Convert the status register to an error code, and clear it. */
-static int i801_check_post(int status, int timeout)
+static int i801_check_post(struct i801_priv *priv, int status, int timeout)
{
int result = 0;
/* If the SMBus is still busy, we give up */
if (timeout) {
- dev_err(&I801_dev->dev, "Transaction timeout\n");
+ dev_err(&priv->pci_dev->dev, "Transaction timeout\n");
/* try to stop the current command */
- dev_dbg(&I801_dev->dev, "Terminating the current operation\n");
- outb_p(inb_p(SMBHSTCNT) | SMBHSTCNT_KILL, SMBHSTCNT);
+ dev_dbg(&priv->pci_dev->dev, "Terminating the current operation\n");
+ outb_p(inb_p(SMBHSTCNT(priv)) | SMBHSTCNT_KILL,
+ SMBHSTCNT(priv));
msleep(1);
- outb_p(inb_p(SMBHSTCNT) & (~SMBHSTCNT_KILL), SMBHSTCNT);
+ outb_p(inb_p(SMBHSTCNT(priv)) & (~SMBHSTCNT_KILL),
+ SMBHSTCNT(priv));
/* Check if it worked */
- status = inb_p(SMBHSTSTS);
+ status = inb_p(SMBHSTSTS(priv));
if ((status & SMBHSTSTS_HOST_BUSY) ||
!(status & SMBHSTSTS_FAILED))
- dev_err(&I801_dev->dev,
+ dev_err(&priv->pci_dev->dev,
"Failed terminating the transaction\n");
- outb_p(STATUS_FLAGS, SMBHSTSTS);
+ outb_p(STATUS_FLAGS, SMBHSTSTS(priv));
return -ETIMEDOUT;
}
if (status & SMBHSTSTS_FAILED) {
result = -EIO;
- dev_err(&I801_dev->dev, "Transaction failed\n");
+ dev_err(&priv->pci_dev->dev, "Transaction failed\n");
}
if (status & SMBHSTSTS_DEV_ERR) {
result = -ENXIO;
- dev_dbg(&I801_dev->dev, "No response\n");
+ dev_dbg(&priv->pci_dev->dev, "No response\n");
}
if (status & SMBHSTSTS_BUS_ERR) {
result = -EAGAIN;
- dev_dbg(&I801_dev->dev, "Lost arbitration\n");
+ dev_dbg(&priv->pci_dev->dev, "Lost arbitration\n");
}
if (result) {
/* Clear error flags */
- outb_p(status & STATUS_FLAGS, SMBHSTSTS);
- status = inb_p(SMBHSTSTS) & STATUS_FLAGS;
+ outb_p(status & STATUS_FLAGS, SMBHSTSTS(priv));
+ status = inb_p(SMBHSTSTS(priv)) & STATUS_FLAGS;
if (status) {
- dev_warn(&I801_dev->dev, "Failed clearing status "
+ dev_warn(&priv->pci_dev->dev, "Failed clearing status "
"flags at end of transaction (%02x)\n",
status);
}
return result;
}
-static int i801_transaction(int xact)
+static int i801_transaction(struct i801_priv *priv, int xact)
{
int status;
int result;
int timeout = 0;
- result = i801_check_pre();
+ result = i801_check_pre(priv);
if (result < 0)
return result;
/* the current contents of SMBHSTCNT can be overwritten, since PEC,
* INTREN, SMBSCMD are passed in xact */
- outb_p(xact | I801_START, SMBHSTCNT);
+ outb_p(xact | I801_START, SMBHSTCNT(priv));
/* We will always wait for a fraction of a second! */
do {
msleep(1);
- status = inb_p(SMBHSTSTS);
+ status = inb_p(SMBHSTSTS(priv));
} while ((status & SMBHSTSTS_HOST_BUSY) && (timeout++ < MAX_TIMEOUT));
- result = i801_check_post(status, timeout > MAX_TIMEOUT);
+ result = i801_check_post(priv, status, timeout > MAX_TIMEOUT);
if (result < 0)
return result;
- outb_p(SMBHSTSTS_INTR, SMBHSTSTS);
+ outb_p(SMBHSTSTS_INTR, SMBHSTSTS(priv));
return 0;
}
/* wait for INTR bit as advised by Intel */
-static void i801_wait_hwpec(void)
+static void i801_wait_hwpec(struct i801_priv *priv)
{
int timeout = 0;
int status;
do {
msleep(1);
- status = inb_p(SMBHSTSTS);
+ status = inb_p(SMBHSTSTS(priv));
} while ((!(status & SMBHSTSTS_INTR))
&& (timeout++ < MAX_TIMEOUT));
if (timeout > MAX_TIMEOUT)
- dev_dbg(&I801_dev->dev, "PEC Timeout!\n");
+ dev_dbg(&priv->pci_dev->dev, "PEC Timeout!\n");
- outb_p(status, SMBHSTSTS);
+ outb_p(status, SMBHSTSTS(priv));
}
-static int i801_block_transaction_by_block(union i2c_smbus_data *data,
+static int i801_block_transaction_by_block(struct i801_priv *priv,
+ union i2c_smbus_data *data,
char read_write, int hwpec)
{
int i, len;
int status;
- inb_p(SMBHSTCNT); /* reset the data buffer index */
+ inb_p(SMBHSTCNT(priv)); /* reset the data buffer index */
/* Use 32-byte buffer to process this transaction */
if (read_write == I2C_SMBUS_WRITE) {
len = data->block[0];
- outb_p(len, SMBHSTDAT0);
+ outb_p(len, SMBHSTDAT0(priv));
for (i = 0; i < len; i++)
- outb_p(data->block[i+1], SMBBLKDAT);
+ outb_p(data->block[i+1], SMBBLKDAT(priv));
}
- status = i801_transaction(I801_BLOCK_DATA | ENABLE_INT9 |
+ status = i801_transaction(priv, I801_BLOCK_DATA | ENABLE_INT9 |
I801_PEC_EN * hwpec);
if (status)
return status;
if (read_write == I2C_SMBUS_READ) {
- len = inb_p(SMBHSTDAT0);
+ len = inb_p(SMBHSTDAT0(priv));
if (len < 1 || len > I2C_SMBUS_BLOCK_MAX)
return -EPROTO;
data->block[0] = len;
for (i = 0; i < len; i++)
- data->block[i + 1] = inb_p(SMBBLKDAT);
+ data->block[i + 1] = inb_p(SMBBLKDAT(priv));
}
return 0;
}
-static int i801_block_transaction_byte_by_byte(union i2c_smbus_data *data,
+static int i801_block_transaction_byte_by_byte(struct i801_priv *priv,
+ union i2c_smbus_data *data,
char read_write, int command,
int hwpec)
{
int result;
int timeout;
- result = i801_check_pre();
+ result = i801_check_pre(priv);
if (result < 0)
return result;
len = data->block[0];
if (read_write == I2C_SMBUS_WRITE) {
- outb_p(len, SMBHSTDAT0);
- outb_p(data->block[1], SMBBLKDAT);
+ outb_p(len, SMBHSTDAT0(priv));
+ outb_p(data->block[1], SMBBLKDAT(priv));
}
for (i = 1; i <= len; i++) {
else
smbcmd = I801_BLOCK_DATA;
}
- outb_p(smbcmd | ENABLE_INT9, SMBHSTCNT);
+ outb_p(smbcmd | ENABLE_INT9, SMBHSTCNT(priv));
if (i == 1)
- outb_p(inb(SMBHSTCNT) | I801_START, SMBHSTCNT);
+ outb_p(inb(SMBHSTCNT(priv)) | I801_START,
+ SMBHSTCNT(priv));
/* We will always wait for a fraction of a second! */
timeout = 0;
do {
msleep(1);
- status = inb_p(SMBHSTSTS);
+ status = inb_p(SMBHSTSTS(priv));
} while ((!(status & SMBHSTSTS_BYTE_DONE))
&& (timeout++ < MAX_TIMEOUT));
- result = i801_check_post(status, timeout > MAX_TIMEOUT);
+ result = i801_check_post(priv, status, timeout > MAX_TIMEOUT);
if (result < 0)
return result;
if (i == 1 && read_write == I2C_SMBUS_READ
&& command != I2C_SMBUS_I2C_BLOCK_DATA) {
- len = inb_p(SMBHSTDAT0);
+ len = inb_p(SMBHSTDAT0(priv));
if (len < 1 || len > I2C_SMBUS_BLOCK_MAX) {
- dev_err(&I801_dev->dev,
+ dev_err(&priv->pci_dev->dev,
"Illegal SMBus block read size %d\n",
len);
/* Recover */
- while (inb_p(SMBHSTSTS) & SMBHSTSTS_HOST_BUSY)
- outb_p(SMBHSTSTS_BYTE_DONE, SMBHSTSTS);
- outb_p(SMBHSTSTS_INTR, SMBHSTSTS);
+ while (inb_p(SMBHSTSTS(priv)) &
+ SMBHSTSTS_HOST_BUSY)
+ outb_p(SMBHSTSTS_BYTE_DONE,
+ SMBHSTSTS(priv));
+ outb_p(SMBHSTSTS_INTR, SMBHSTSTS(priv));
return -EPROTO;
}
data->block[0] = len;
/* Retrieve/store value in SMBBLKDAT */
if (read_write == I2C_SMBUS_READ)
- data->block[i] = inb_p(SMBBLKDAT);
+ data->block[i] = inb_p(SMBBLKDAT(priv));
if (read_write == I2C_SMBUS_WRITE && i+1 <= len)
- outb_p(data->block[i+1], SMBBLKDAT);
+ outb_p(data->block[i+1], SMBBLKDAT(priv));
/* signals SMBBLKDAT ready */
- outb_p(SMBHSTSTS_BYTE_DONE | SMBHSTSTS_INTR, SMBHSTSTS);
+ outb_p(SMBHSTSTS_BYTE_DONE | SMBHSTSTS_INTR, SMBHSTSTS(priv));
}
return 0;
}
-static int i801_set_block_buffer_mode(void)
+static int i801_set_block_buffer_mode(struct i801_priv *priv)
{
- outb_p(inb_p(SMBAUXCTL) | SMBAUXCTL_E32B, SMBAUXCTL);
- if ((inb_p(SMBAUXCTL) & SMBAUXCTL_E32B) == 0)
+ outb_p(inb_p(SMBAUXCTL(priv)) | SMBAUXCTL_E32B, SMBAUXCTL(priv));
+ if ((inb_p(SMBAUXCTL(priv)) & SMBAUXCTL_E32B) == 0)
return -EIO;
return 0;
}
/* Block transaction function */
-static int i801_block_transaction(union i2c_smbus_data *data, char read_write,
+static int i801_block_transaction(struct i801_priv *priv,
+ union i2c_smbus_data *data, char read_write,
int command, int hwpec)
{
int result = 0;
if (command == I2C_SMBUS_I2C_BLOCK_DATA) {
if (read_write == I2C_SMBUS_WRITE) {
/* set I2C_EN bit in configuration register */
- pci_read_config_byte(I801_dev, SMBHSTCFG, &hostc);
- pci_write_config_byte(I801_dev, SMBHSTCFG,
+ pci_read_config_byte(priv->pci_dev, SMBHSTCFG, &hostc);
+ pci_write_config_byte(priv->pci_dev, SMBHSTCFG,
hostc | SMBHSTCFG_I2C_EN);
- } else if (!(i801_features & FEATURE_I2C_BLOCK_READ)) {
- dev_err(&I801_dev->dev,
+ } else if (!(priv->features & FEATURE_I2C_BLOCK_READ)) {
+ dev_err(&priv->pci_dev->dev,
"I2C block read is unsupported!\n");
return -EOPNOTSUPP;
}
/* Experience has shown that the block buffer can only be used for
SMBus (not I2C) block transactions, even though the datasheet
doesn't mention this limitation. */
- if ((i801_features & FEATURE_BLOCK_BUFFER)
+ if ((priv->features & FEATURE_BLOCK_BUFFER)
&& command != I2C_SMBUS_I2C_BLOCK_DATA
- && i801_set_block_buffer_mode() == 0)
- result = i801_block_transaction_by_block(data, read_write,
- hwpec);
+ && i801_set_block_buffer_mode(priv) == 0)
+ result = i801_block_transaction_by_block(priv, data,
+ read_write, hwpec);
else
- result = i801_block_transaction_byte_by_byte(data, read_write,
+ result = i801_block_transaction_byte_by_byte(priv, data,
+ read_write,
command, hwpec);
if (result == 0 && hwpec)
- i801_wait_hwpec();
+ i801_wait_hwpec(priv);
if (command == I2C_SMBUS_I2C_BLOCK_DATA
&& read_write == I2C_SMBUS_WRITE) {
/* restore saved configuration register value */
- pci_write_config_byte(I801_dev, SMBHSTCFG, hostc);
+ pci_write_config_byte(priv->pci_dev, SMBHSTCFG, hostc);
}
return result;
}
int hwpec;
int block = 0;
int ret, xact = 0;
+ struct i801_priv *priv = i2c_get_adapdata(adap);
- hwpec = (i801_features & FEATURE_SMBUS_PEC) && (flags & I2C_CLIENT_PEC)
+ hwpec = (priv->features & FEATURE_SMBUS_PEC) && (flags & I2C_CLIENT_PEC)
&& size != I2C_SMBUS_QUICK
&& size != I2C_SMBUS_I2C_BLOCK_DATA;
switch (size) {
case I2C_SMBUS_QUICK:
outb_p(((addr & 0x7f) << 1) | (read_write & 0x01),
- SMBHSTADD);
+ SMBHSTADD(priv));
xact = I801_QUICK;
break;
case I2C_SMBUS_BYTE:
outb_p(((addr & 0x7f) << 1) | (read_write & 0x01),
- SMBHSTADD);
+ SMBHSTADD(priv));
if (read_write == I2C_SMBUS_WRITE)
- outb_p(command, SMBHSTCMD);
+ outb_p(command, SMBHSTCMD(priv));
xact = I801_BYTE;
break;
case I2C_SMBUS_BYTE_DATA:
outb_p(((addr & 0x7f) << 1) | (read_write & 0x01),
- SMBHSTADD);
- outb_p(command, SMBHSTCMD);
+ SMBHSTADD(priv));
+ outb_p(command, SMBHSTCMD(priv));
if (read_write == I2C_SMBUS_WRITE)
- outb_p(data->byte, SMBHSTDAT0);
+ outb_p(data->byte, SMBHSTDAT0(priv));
xact = I801_BYTE_DATA;
break;
case I2C_SMBUS_WORD_DATA:
outb_p(((addr & 0x7f) << 1) | (read_write & 0x01),
- SMBHSTADD);
- outb_p(command, SMBHSTCMD);
+ SMBHSTADD(priv));
+ outb_p(command, SMBHSTCMD(priv));
if (read_write == I2C_SMBUS_WRITE) {
- outb_p(data->word & 0xff, SMBHSTDAT0);
- outb_p((data->word & 0xff00) >> 8, SMBHSTDAT1);
+ outb_p(data->word & 0xff, SMBHSTDAT0(priv));
+ outb_p((data->word & 0xff00) >> 8, SMBHSTDAT1(priv));
}
xact = I801_WORD_DATA;
break;
case I2C_SMBUS_BLOCK_DATA:
outb_p(((addr & 0x7f) << 1) | (read_write & 0x01),
- SMBHSTADD);
- outb_p(command, SMBHSTCMD);
+ SMBHSTADD(priv));
+ outb_p(command, SMBHSTCMD(priv));
block = 1;
break;
case I2C_SMBUS_I2C_BLOCK_DATA:
/* NB: page 240 of ICH5 datasheet shows that the R/#W
* bit should be cleared here, even when reading */
- outb_p((addr & 0x7f) << 1, SMBHSTADD);
+ outb_p((addr & 0x7f) << 1, SMBHSTADD(priv));
if (read_write == I2C_SMBUS_READ) {
/* NB: page 240 of ICH5 datasheet also shows
* that DATA1 is the cmd field when reading */
- outb_p(command, SMBHSTDAT1);
+ outb_p(command, SMBHSTDAT1(priv));
} else
- outb_p(command, SMBHSTCMD);
+ outb_p(command, SMBHSTCMD(priv));
block = 1;
break;
default:
- dev_err(&I801_dev->dev, "Unsupported transaction %d\n", size);
+ dev_err(&priv->pci_dev->dev, "Unsupported transaction %d\n",
+ size);
return -EOPNOTSUPP;
}
if (hwpec) /* enable/disable hardware PEC */
- outb_p(inb_p(SMBAUXCTL) | SMBAUXCTL_CRC, SMBAUXCTL);
+ outb_p(inb_p(SMBAUXCTL(priv)) | SMBAUXCTL_CRC, SMBAUXCTL(priv));
else
- outb_p(inb_p(SMBAUXCTL) & (~SMBAUXCTL_CRC), SMBAUXCTL);
+ outb_p(inb_p(SMBAUXCTL(priv)) & (~SMBAUXCTL_CRC),
+ SMBAUXCTL(priv));
if (block)
- ret = i801_block_transaction(data, read_write, size, hwpec);
+ ret = i801_block_transaction(priv, data, read_write, size,
+ hwpec);
else
- ret = i801_transaction(xact | ENABLE_INT9);
+ ret = i801_transaction(priv, xact | ENABLE_INT9);
/* Some BIOSes don't like it when PEC is enabled at reboot or resume
time, so we forcibly disable it after every transaction. Turn off
E32B for the same reason. */
if (hwpec || block)
- outb_p(inb_p(SMBAUXCTL) & ~(SMBAUXCTL_CRC | SMBAUXCTL_E32B),
- SMBAUXCTL);
+ outb_p(inb_p(SMBAUXCTL(priv)) &
+ ~(SMBAUXCTL_CRC | SMBAUXCTL_E32B), SMBAUXCTL(priv));
if (block)
return ret;
switch (xact & 0x7f) {
case I801_BYTE: /* Result put in SMBHSTDAT0 */
case I801_BYTE_DATA:
- data->byte = inb_p(SMBHSTDAT0);
+ data->byte = inb_p(SMBHSTDAT0(priv));
break;
case I801_WORD_DATA:
- data->word = inb_p(SMBHSTDAT0) + (inb_p(SMBHSTDAT1) << 8);
+ data->word = inb_p(SMBHSTDAT0(priv)) +
+ (inb_p(SMBHSTDAT1(priv)) << 8);
break;
}
return 0;
static u32 i801_func(struct i2c_adapter *adapter)
{
+ struct i801_priv *priv = i2c_get_adapdata(adapter);
+
return I2C_FUNC_SMBUS_QUICK | I2C_FUNC_SMBUS_BYTE |
I2C_FUNC_SMBUS_BYTE_DATA | I2C_FUNC_SMBUS_WORD_DATA |
I2C_FUNC_SMBUS_BLOCK_DATA | I2C_FUNC_SMBUS_WRITE_I2C_BLOCK |
- ((i801_features & FEATURE_SMBUS_PEC) ? I2C_FUNC_SMBUS_PEC : 0) |
- ((i801_features & FEATURE_I2C_BLOCK_READ) ?
+ ((priv->features & FEATURE_SMBUS_PEC) ? I2C_FUNC_SMBUS_PEC : 0) |
+ ((priv->features & FEATURE_I2C_BLOCK_READ) ?
I2C_FUNC_SMBUS_READ_I2C_BLOCK : 0);
}
.functionality = i801_func,
};
-static struct i2c_adapter i801_adapter = {
- .owner = THIS_MODULE,
- .class = I2C_CLASS_HWMON | I2C_CLASS_SPD,
- .algo = &smbus_algorithm,
-};
-
static const struct pci_device_id i801_ids[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801AA_3) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801AB_3) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH10_5) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_5_3400_SERIES_SMBUS) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_COUGARPOINT_SMBUS) },
+ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS) },
+ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS_IDF0) },
+ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS_IDF1) },
+ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS_IDF2) },
{ 0, }
};
{
unsigned char temp;
int err, i;
+ struct i801_priv *priv;
+
+ priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+ if (!priv)
+ return -ENOMEM;
+
+ i2c_set_adapdata(&priv->adapter, priv);
+ priv->adapter.owner = THIS_MODULE;
+ priv->adapter.class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
+ priv->adapter.algo = &smbus_algorithm;
- I801_dev = dev;
- i801_features = 0;
+ priv->pci_dev = dev;
switch (dev->device) {
default:
- i801_features |= FEATURE_I2C_BLOCK_READ;
+ priv->features |= FEATURE_I2C_BLOCK_READ;
/* fall through */
case PCI_DEVICE_ID_INTEL_82801DB_3:
- i801_features |= FEATURE_SMBUS_PEC;
- i801_features |= FEATURE_BLOCK_BUFFER;
+ priv->features |= FEATURE_SMBUS_PEC;
+ priv->features |= FEATURE_BLOCK_BUFFER;
/* fall through */
case PCI_DEVICE_ID_INTEL_82801CA_3:
case PCI_DEVICE_ID_INTEL_82801BA_2:
/* Disable features on user request */
for (i = 0; i < ARRAY_SIZE(i801_feature_names); i++) {
- if (i801_features & disable_features & (1 << i))
+ if (priv->features & disable_features & (1 << i))
dev_notice(&dev->dev, "%s disabled by user\n",
i801_feature_names[i]);
}
- i801_features &= ~disable_features;
+ priv->features &= ~disable_features;
err = pci_enable_device(dev);
if (err) {
}
/* Determine the address of the SMBus area */
- i801_smba = pci_resource_start(dev, SMBBAR);
- if (!i801_smba) {
+ priv->smba = pci_resource_start(dev, SMBBAR);
+ if (!priv->smba) {
dev_err(&dev->dev, "SMBus base address uninitialized, "
"upgrade BIOS\n");
err = -ENODEV;
err = pci_request_region(dev, SMBBAR, i801_driver.name);
if (err) {
dev_err(&dev->dev, "Failed to request SMBus region "
- "0x%lx-0x%Lx\n", i801_smba,
+ "0x%lx-0x%Lx\n", priv->smba,
(unsigned long long)pci_resource_end(dev, SMBBAR));
goto exit;
}
- pci_read_config_byte(I801_dev, SMBHSTCFG, &temp);
- i801_original_hstcfg = temp;
+ pci_read_config_byte(priv->pci_dev, SMBHSTCFG, &temp);
+ priv->original_hstcfg = temp;
temp &= ~SMBHSTCFG_I2C_EN; /* SMBus timing */
if (!(temp & SMBHSTCFG_HST_EN)) {
dev_info(&dev->dev, "Enabling SMBus device\n");
temp |= SMBHSTCFG_HST_EN;
}
- pci_write_config_byte(I801_dev, SMBHSTCFG, temp);
+ pci_write_config_byte(priv->pci_dev, SMBHSTCFG, temp);
if (temp & SMBHSTCFG_SMB_SMI_EN)
dev_dbg(&dev->dev, "SMBus using interrupt SMI#\n");
dev_dbg(&dev->dev, "SMBus using PCI Interrupt\n");
/* Clear special mode bits */
- if (i801_features & (FEATURE_SMBUS_PEC | FEATURE_BLOCK_BUFFER))
- outb_p(inb_p(SMBAUXCTL) & ~(SMBAUXCTL_CRC | SMBAUXCTL_E32B),
- SMBAUXCTL);
+ if (priv->features & (FEATURE_SMBUS_PEC | FEATURE_BLOCK_BUFFER))
+ outb_p(inb_p(SMBAUXCTL(priv)) &
+ ~(SMBAUXCTL_CRC | SMBAUXCTL_E32B), SMBAUXCTL(priv));
/* set up the sysfs linkage to our parent device */
- i801_adapter.dev.parent = &dev->dev;
+ priv->adapter.dev.parent = &dev->dev;
/* Retry up to 3 times on lost arbitration */
- i801_adapter.retries = 3;
+ priv->adapter.retries = 3;
- snprintf(i801_adapter.name, sizeof(i801_adapter.name),
- "SMBus I801 adapter at %04lx", i801_smba);
- err = i2c_add_adapter(&i801_adapter);
+ snprintf(priv->adapter.name, sizeof(priv->adapter.name),
+ "SMBus I801 adapter at %04lx", priv->smba);
+ err = i2c_add_adapter(&priv->adapter);
if (err) {
dev_err(&dev->dev, "Failed to add SMBus adapter\n");
goto exit_release;
memset(&info, 0, sizeof(struct i2c_board_info));
info.addr = apanel_addr;
strlcpy(info.type, "fujitsu_apanel", I2C_NAME_SIZE);
- i2c_new_device(&i801_adapter, &info);
+ i2c_new_device(&priv->adapter, &info);
}
#endif
#if defined CONFIG_SENSORS_FSCHMD || defined CONFIG_SENSORS_FSCHMD_MODULE
if (dmi_name_in_vendors("FUJITSU"))
- dmi_walk(dmi_check_onboard_devices, &i801_adapter);
+ dmi_walk(dmi_check_onboard_devices, &priv->adapter);
#endif
+ pci_set_drvdata(dev, priv);
return 0;
exit_release:
pci_release_region(dev, SMBBAR);
exit:
+ kfree(priv);
return err;
}
static void __devexit i801_remove(struct pci_dev *dev)
{
- i2c_del_adapter(&i801_adapter);
- pci_write_config_byte(I801_dev, SMBHSTCFG, i801_original_hstcfg);
+ struct i801_priv *priv = pci_get_drvdata(dev);
+
+ i2c_del_adapter(&priv->adapter);
+ pci_write_config_byte(dev, SMBHSTCFG, priv->original_hstcfg);
pci_release_region(dev, SMBBAR);
+ pci_set_drvdata(dev, NULL);
+ kfree(priv);
/*
* do not call pci_disable_device(dev) since it can cause hard hangs on
* some systems during power-off (eg. Fujitsu-Siemens Lifebook E8010)
#ifdef CONFIG_PM
static int i801_suspend(struct pci_dev *dev, pm_message_t mesg)
{
+ struct i801_priv *priv = pci_get_drvdata(dev);
+
pci_save_state(dev);
- pci_write_config_byte(dev, SMBHSTCFG, i801_original_hstcfg);
+ pci_write_config_byte(dev, SMBHSTCFG, priv->original_hstcfg);
pci_set_power_state(dev, pci_choose_state(dev, mesg));
return 0;
}
/* Just update the base values (i.e. touchpad in untouched state) */
if (dev->data[dev->info->datalen - 1] & ATP_STATUS_BASE_UPDATE) {
- dprintk(KERN_DEBUG "appletouch: updated base values\n");
+ dprintk("appletouch: updated base values\n");
memcpy(dev->xy_old, dev->xy_cur, sizeof(dev->xy_old));
goto exit;
u16 cmd_crtl1;
u16 cmd_crtl2;
u16 cmd_crtl3;
+ int x;
+ int y;
+ int Rt;
};
static int ad7879_read(struct ad7879 *ts, u8 reg)
Rt /= z1;
Rt = (Rt + 2047) >> 12;
- if (!timer_pending(&ts->timer))
+ /*
+ * Sample found inconsistent, pressure is beyond
+ * the maximum. Don't report it to user space.
+ */
+ if (Rt > ts->pressure_max)
+ return -EINVAL;
+
+ /*
+ * Note that we delay reporting events by one sample.
+ * This is done to avoid reporting last sample of the
+ * touch sequence, which may be incomplete if finger
+ * leaves the surface before last reading is taken.
+ */
+ if (timer_pending(&ts->timer)) {
+ /* Touch continues */
input_report_key(input_dev, BTN_TOUCH, 1);
+ input_report_abs(input_dev, ABS_X, ts->x);
+ input_report_abs(input_dev, ABS_Y, ts->y);
+ input_report_abs(input_dev, ABS_PRESSURE, ts->Rt);
+ input_sync(input_dev);
+ }
+
+ ts->x = x;
+ ts->y = y;
+ ts->Rt = Rt;
- input_report_abs(input_dev, ABS_X, x);
- input_report_abs(input_dev, ABS_Y, y);
- input_report_abs(input_dev, ABS_PRESSURE, Rt);
- input_sync(input_dev);
return 0;
}
err_cs_disable:
pdata->cs_dis(pdata->cs_pin);
err_free_mem:
- input_free_device(bu21013_data->in_dev);
+ input_free_device(in_dev);
kfree(bu21013_data);
return error;
&bcs->hw.isar.reg->Flags))
bcs->hw.isar.dpath = 1;
else {
- printk(KERN_WARNING"isar modeisar analog funktions only with DP1\n");
- debugl1(cs, "isar modeisar analog funktions only with DP1");
+ printk(KERN_WARNING"isar modeisar analog functions only with DP1\n");
+ debugl1(cs, "isar modeisar analog functions only with DP1");
return(1);
}
break;
if NEW_LEDS
config LEDS_CLASS
- tristate "LED Class Support"
+ bool "LED Class Support"
help
This option enables the led sysfs class in /sys/class/leds. You'll
need this to do anything useful with LEDs. If unsure, say N.
To compile this driver as a module, choose M here: the
module will be called leds-lp3944.
+config LEDS_LP5521
+ tristate "LED Support for N.S. LP5521 LED driver chip"
+ depends on LEDS_CLASS && I2C
+ help
+ If you say yes here you get support for the National Semiconductor
+ LP5521 LED driver. It is 3 channel chip with programmable engines.
+ Driver provides direct control via LED class and interface for
+ programming the engines.
+
+config LEDS_LP5523
+ tristate "LED Support for N.S. LP5523 LED driver chip"
+ depends on LEDS_CLASS && I2C
+ help
+ If you say yes here you get support for the National Semiconductor
+ LP5523 LED driver. It is 9 channel chip with programmable engines.
+ Driver provides direct control via LED class and interface for
+ programming the engines.
+
config LEDS_CLEVO_MAIL
tristate "Mail LED on Clevo notebook"
depends on X86 && SERIO_I8042 && DMI
obj-$(CONFIG_LEDS_PCA9532) += leds-pca9532.o
obj-$(CONFIG_LEDS_GPIO) += leds-gpio.o
obj-$(CONFIG_LEDS_LP3944) += leds-lp3944.o
+obj-$(CONFIG_LEDS_LP5521) += leds-lp5521.o
+obj-$(CONFIG_LEDS_LP5523) += leds-lp5523.o
obj-$(CONFIG_LEDS_CLEVO_MAIL) += leds-clevo-mail.o
obj-$(CONFIG_LEDS_HP6XX) += leds-hp6xx.o
obj-$(CONFIG_LEDS_FSG) += leds-fsg.o
__ATTR_NULL,
};
+static void led_timer_function(unsigned long data)
+{
+ struct led_classdev *led_cdev = (void *)data;
+ unsigned long brightness;
+ unsigned long delay;
+
+ if (!led_cdev->blink_delay_on || !led_cdev->blink_delay_off) {
+ led_set_brightness(led_cdev, LED_OFF);
+ return;
+ }
+
+ brightness = led_get_brightness(led_cdev);
+ if (!brightness) {
+ /* Time to switch the LED on. */
+ brightness = led_cdev->blink_brightness;
+ delay = led_cdev->blink_delay_on;
+ } else {
+ /* Store the current brightness value to be able
+ * to restore it when the delay_off period is over.
+ */
+ led_cdev->blink_brightness = brightness;
+ brightness = LED_OFF;
+ delay = led_cdev->blink_delay_off;
+ }
+
+ led_set_brightness(led_cdev, brightness);
+
+ mod_timer(&led_cdev->blink_timer, jiffies + msecs_to_jiffies(delay));
+}
+
+static void led_stop_software_blink(struct led_classdev *led_cdev)
+{
+ /* deactivate previous settings */
+ del_timer_sync(&led_cdev->blink_timer);
+ led_cdev->blink_delay_on = 0;
+ led_cdev->blink_delay_off = 0;
+}
+
+static void led_set_software_blink(struct led_classdev *led_cdev,
+ unsigned long delay_on,
+ unsigned long delay_off)
+{
+ int current_brightness;
+
+ current_brightness = led_get_brightness(led_cdev);
+ if (current_brightness)
+ led_cdev->blink_brightness = current_brightness;
+ if (!led_cdev->blink_brightness)
+ led_cdev->blink_brightness = led_cdev->max_brightness;
+
+ if (delay_on == led_cdev->blink_delay_on &&
+ delay_off == led_cdev->blink_delay_off)
+ return;
+
+ led_stop_software_blink(led_cdev);
+
+ led_cdev->blink_delay_on = delay_on;
+ led_cdev->blink_delay_off = delay_off;
+
+ /* never on - don't blink */
+ if (!delay_on)
+ return;
+
+ /* never off - just set to brightness */
+ if (!delay_off) {
+ led_set_brightness(led_cdev, led_cdev->blink_brightness);
+ return;
+ }
+
+ mod_timer(&led_cdev->blink_timer, jiffies + 1);
+}
+
+
/**
* led_classdev_suspend - suspend an led_classdev.
* @led_cdev: the led_classdev to suspend.
led_update_brightness(led_cdev);
+ init_timer(&led_cdev->blink_timer);
+ led_cdev->blink_timer.function = led_timer_function;
+ led_cdev->blink_timer.data = (unsigned long)led_cdev;
+
#ifdef CONFIG_LEDS_TRIGGERS
led_trigger_set_default(led_cdev);
#endif
return 0;
}
-
EXPORT_SYMBOL_GPL(led_classdev_register);
/**
up_write(&led_cdev->trigger_lock);
#endif
+ /* Stop blinking */
+ led_brightness_set(led_cdev, LED_OFF);
+
device_unregister(led_cdev->dev);
down_write(&leds_list_lock);
}
EXPORT_SYMBOL_GPL(led_classdev_unregister);
+void led_blink_set(struct led_classdev *led_cdev,
+ unsigned long *delay_on,
+ unsigned long *delay_off)
+{
+ if (led_cdev->blink_set &&
+ led_cdev->blink_set(led_cdev, delay_on, delay_off))
+ return;
+
+ /* blink with 1 Hz as default if nothing specified */
+ if (!*delay_on && !*delay_off)
+ *delay_on = *delay_off = 500;
+
+ led_set_software_blink(led_cdev, *delay_on, *delay_off);
+}
+EXPORT_SYMBOL(led_blink_set);
+
+void led_brightness_set(struct led_classdev *led_cdev,
+ enum led_brightness brightness)
+{
+ led_stop_software_blink(led_cdev);
+ led_cdev->brightness_set(led_cdev, brightness);
+}
+EXPORT_SYMBOL(led_brightness_set);
+
static int __init leds_init(void)
{
leds_class = class_create(THIS_MODULE, "leds");
if (led_cdev->trigger->deactivate)
led_cdev->trigger->deactivate(led_cdev);
led_cdev->trigger = NULL;
- led_set_brightness(led_cdev, LED_OFF);
+ led_brightness_set(led_cdev, LED_OFF);
}
if (trigger) {
write_lock_irqsave(&trigger->leddev_list_lock, flags);
static int __init gpio_led_init(void)
{
- int ret;
+ int ret = 0;
#ifdef CONFIG_LEDS_GPIO_PLATFORM
ret = platform_driver_register(&gpio_led_driver);
--- /dev/null
+/*
+ * LP5521 LED chip driver.
+ *
+ * Copyright (C) 2010 Nokia Corporation
+ *
+ * Contact: Samu Onkalo <samu.p.onkalo@nokia.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
+ * 02110-1301 USA
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/i2c.h>
+#include <linux/mutex.h>
+#include <linux/gpio.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/ctype.h>
+#include <linux/spinlock.h>
+#include <linux/wait.h>
+#include <linux/leds.h>
+#include <linux/leds-lp5521.h>
+#include <linux/workqueue.h>
+#include <linux/slab.h>
+
+#define LP5521_PROGRAM_LENGTH 32 /* in bytes */
+
+#define LP5521_MAX_LEDS 3 /* Maximum number of LEDs */
+#define LP5521_MAX_ENGINES 3 /* Maximum number of engines */
+
+#define LP5521_ENG_MASK_BASE 0x30 /* 00110000 */
+#define LP5521_ENG_STATUS_MASK 0x07 /* 00000111 */
+
+#define LP5521_CMD_LOAD 0x15 /* 00010101 */
+#define LP5521_CMD_RUN 0x2a /* 00101010 */
+#define LP5521_CMD_DIRECT 0x3f /* 00111111 */
+#define LP5521_CMD_DISABLED 0x00 /* 00000000 */
+
+/* Registers */
+#define LP5521_REG_ENABLE 0x00
+#define LP5521_REG_OP_MODE 0x01
+#define LP5521_REG_R_PWM 0x02
+#define LP5521_REG_G_PWM 0x03
+#define LP5521_REG_B_PWM 0x04
+#define LP5521_REG_R_CURRENT 0x05
+#define LP5521_REG_G_CURRENT 0x06
+#define LP5521_REG_B_CURRENT 0x07
+#define LP5521_REG_CONFIG 0x08
+#define LP5521_REG_R_CHANNEL_PC 0x09
+#define LP5521_REG_G_CHANNEL_PC 0x0A
+#define LP5521_REG_B_CHANNEL_PC 0x0B
+#define LP5521_REG_STATUS 0x0C
+#define LP5521_REG_RESET 0x0D
+#define LP5521_REG_GPO 0x0E
+#define LP5521_REG_R_PROG_MEM 0x10
+#define LP5521_REG_G_PROG_MEM 0x30
+#define LP5521_REG_B_PROG_MEM 0x50
+
+#define LP5521_PROG_MEM_BASE LP5521_REG_R_PROG_MEM
+#define LP5521_PROG_MEM_SIZE 0x20
+
+/* Base register to set LED current */
+#define LP5521_REG_LED_CURRENT_BASE LP5521_REG_R_CURRENT
+
+/* Base register to set the brightness */
+#define LP5521_REG_LED_PWM_BASE LP5521_REG_R_PWM
+
+/* Bits in ENABLE register */
+#define LP5521_MASTER_ENABLE 0x40 /* Chip master enable */
+#define LP5521_LOGARITHMIC_PWM 0x80 /* Logarithmic PWM adjustment */
+#define LP5521_EXEC_RUN 0x2A
+
+/* Bits in CONFIG register */
+#define LP5521_PWM_HF 0x40 /* PWM: 0 = 256Hz, 1 = 558Hz */
+#define LP5521_PWRSAVE_EN 0x20 /* 1 = Power save mode */
+#define LP5521_CP_MODE_OFF 0 /* Charge pump (CP) off */
+#define LP5521_CP_MODE_BYPASS 8 /* CP forced to bypass mode */
+#define LP5521_CP_MODE_1X5 0x10 /* CP forced to 1.5x mode */
+#define LP5521_CP_MODE_AUTO 0x18 /* Automatic mode selection */
+#define LP5521_R_TO_BATT 4 /* R out: 0 = CP, 1 = Vbat */
+#define LP5521_CLK_SRC_EXT 0 /* Ext-clk source (CLK_32K) */
+#define LP5521_CLK_INT 1 /* Internal clock */
+#define LP5521_CLK_AUTO 2 /* Automatic clock selection */
+
+/* Status */
+#define LP5521_EXT_CLK_USED 0x08
+
+struct lp5521_engine {
+ const struct attribute_group *attributes;
+ int id;
+ u8 mode;
+ u8 prog_page;
+ u8 engine_mask;
+};
+
+struct lp5521_led {
+ int id;
+ u8 chan_nr;
+ u8 led_current;
+ u8 max_current;
+ struct led_classdev cdev;
+ struct work_struct brightness_work;
+ u8 brightness;
+};
+
+struct lp5521_chip {
+ struct lp5521_platform_data *pdata;
+ struct mutex lock; /* Serialize control */
+ struct i2c_client *client;
+ struct lp5521_engine engines[LP5521_MAX_ENGINES];
+ struct lp5521_led leds[LP5521_MAX_LEDS];
+ u8 num_channels;
+ u8 num_leds;
+};
+
+#define cdev_to_led(c) container_of(c, struct lp5521_led, cdev)
+#define engine_to_lp5521(eng) container_of((eng), struct lp5521_chip, \
+ engines[(eng)->id - 1])
+#define led_to_lp5521(led) container_of((led), struct lp5521_chip, \
+ leds[(led)->id])
+
+static void lp5521_led_brightness_work(struct work_struct *work);
+
+static inline int lp5521_write(struct i2c_client *client, u8 reg, u8 value)
+{
+ return i2c_smbus_write_byte_data(client, reg, value);
+}
+
+static int lp5521_read(struct i2c_client *client, u8 reg, u8 *buf)
+{
+ s32 ret;
+
+ ret = i2c_smbus_read_byte_data(client, reg);
+ if (ret < 0)
+ return -EIO;
+
+ *buf = ret;
+ return 0;
+}
+
+static int lp5521_set_engine_mode(struct lp5521_engine *engine, u8 mode)
+{
+ struct lp5521_chip *chip = engine_to_lp5521(engine);
+ struct i2c_client *client = chip->client;
+ int ret;
+ u8 engine_state;
+
+ /* Only transition between RUN and DIRECT mode are handled here */
+ if (mode == LP5521_CMD_LOAD)
+ return 0;
+
+ if (mode == LP5521_CMD_DISABLED)
+ mode = LP5521_CMD_DIRECT;
+
+ ret = lp5521_read(client, LP5521_REG_OP_MODE, &engine_state);
+
+ /* set mode only for this engine */
+ engine_state &= ~(engine->engine_mask);
+ mode &= engine->engine_mask;
+ engine_state |= mode;
+ ret |= lp5521_write(client, LP5521_REG_OP_MODE, engine_state);
+
+ return ret;
+}
+
+static int lp5521_load_program(struct lp5521_engine *eng, const u8 *pattern)
+{
+ struct lp5521_chip *chip = engine_to_lp5521(eng);
+ struct i2c_client *client = chip->client;
+ int ret;
+ int addr;
+ u8 mode;
+
+ /* move current engine to direct mode and remember the state */
+ ret = lp5521_set_engine_mode(eng, LP5521_CMD_DIRECT);
+ usleep_range(1000, 10000);
+ ret |= lp5521_read(client, LP5521_REG_OP_MODE, &mode);
+
+ /* For loading, all the engines to load mode */
+ lp5521_write(client, LP5521_REG_OP_MODE, LP5521_CMD_DIRECT);
+ usleep_range(1000, 10000);
+ lp5521_write(client, LP5521_REG_OP_MODE, LP5521_CMD_LOAD);
+ usleep_range(1000, 10000);
+
+ addr = LP5521_PROG_MEM_BASE + eng->prog_page * LP5521_PROG_MEM_SIZE;
+ i2c_smbus_write_i2c_block_data(client,
+ addr,
+ LP5521_PROG_MEM_SIZE,
+ pattern);
+
+ ret |= lp5521_write(client, LP5521_REG_OP_MODE, mode);
+ return ret;
+}
+
+static int lp5521_set_led_current(struct lp5521_chip *chip, int led, u8 curr)
+{
+ return lp5521_write(chip->client,
+ LP5521_REG_LED_CURRENT_BASE + chip->leds[led].chan_nr,
+ curr);
+}
+
+static void lp5521_init_engine(struct lp5521_chip *chip,
+ const struct attribute_group *attr_group)
+{
+ int i;
+ for (i = 0; i < ARRAY_SIZE(chip->engines); i++) {
+ chip->engines[i].id = i + 1;
+ chip->engines[i].engine_mask = LP5521_ENG_MASK_BASE >> (i * 2);
+ chip->engines[i].prog_page = i;
+ chip->engines[i].attributes = &attr_group[i];
+ }
+}
+
+static int lp5521_configure(struct i2c_client *client,
+ const struct attribute_group *attr_group)
+{
+ struct lp5521_chip *chip = i2c_get_clientdata(client);
+ int ret;
+
+ lp5521_init_engine(chip, attr_group);
+
+ lp5521_write(client, LP5521_REG_RESET, 0xff);
+
+ usleep_range(10000, 20000);
+
+ /* Set all PWMs to direct control mode */
+ ret = lp5521_write(client, LP5521_REG_OP_MODE, 0x3F);
+
+ /* Enable auto-powersave, set charge pump to auto, red to battery */
+ ret |= lp5521_write(client, LP5521_REG_CONFIG,
+ LP5521_PWRSAVE_EN | LP5521_CP_MODE_AUTO | LP5521_R_TO_BATT);
+
+ /* Initialize all channels PWM to zero -> leds off */
+ ret |= lp5521_write(client, LP5521_REG_R_PWM, 0);
+ ret |= lp5521_write(client, LP5521_REG_G_PWM, 0);
+ ret |= lp5521_write(client, LP5521_REG_B_PWM, 0);
+
+ /* Set engines are set to run state when OP_MODE enables engines */
+ ret |= lp5521_write(client, LP5521_REG_ENABLE,
+ LP5521_MASTER_ENABLE | LP5521_LOGARITHMIC_PWM |
+ LP5521_EXEC_RUN);
+ /* enable takes 500us */
+ usleep_range(500, 20000);
+
+ return ret;
+}
+
+static int lp5521_run_selftest(struct lp5521_chip *chip, char *buf)
+{
+ int ret;
+ u8 status;
+
+ ret = lp5521_read(chip->client, LP5521_REG_STATUS, &status);
+ if (ret < 0)
+ return ret;
+
+ /* Check that ext clock is really in use if requested */
+ if (chip->pdata && chip->pdata->clock_mode == LP5521_CLOCK_EXT)
+ if ((status & LP5521_EXT_CLK_USED) == 0)
+ return -EIO;
+ return 0;
+}
+
+static void lp5521_set_brightness(struct led_classdev *cdev,
+ enum led_brightness brightness)
+{
+ struct lp5521_led *led = cdev_to_led(cdev);
+ led->brightness = (u8)brightness;
+ schedule_work(&led->brightness_work);
+}
+
+static void lp5521_led_brightness_work(struct work_struct *work)
+{
+ struct lp5521_led *led = container_of(work,
+ struct lp5521_led,
+ brightness_work);
+ struct lp5521_chip *chip = led_to_lp5521(led);
+ struct i2c_client *client = chip->client;
+
+ mutex_lock(&chip->lock);
+ lp5521_write(client, LP5521_REG_LED_PWM_BASE + led->chan_nr,
+ led->brightness);
+ mutex_unlock(&chip->lock);
+}
+
+/* Detect the chip by setting its ENABLE register and reading it back. */
+static int lp5521_detect(struct i2c_client *client)
+{
+ int ret;
+ u8 buf;
+
+ ret = lp5521_write(client, LP5521_REG_ENABLE,
+ LP5521_MASTER_ENABLE | LP5521_LOGARITHMIC_PWM);
+ if (ret)
+ return ret;
+ usleep_range(1000, 10000);
+ ret = lp5521_read(client, LP5521_REG_ENABLE, &buf);
+ if (ret)
+ return ret;
+ if (buf != (LP5521_MASTER_ENABLE | LP5521_LOGARITHMIC_PWM))
+ return -ENODEV;
+
+ return 0;
+}
+
+/* Set engine mode and create appropriate sysfs attributes, if required. */
+static int lp5521_set_mode(struct lp5521_engine *engine, u8 mode)
+{
+ struct lp5521_chip *chip = engine_to_lp5521(engine);
+ struct i2c_client *client = chip->client;
+ struct device *dev = &client->dev;
+ int ret = 0;
+
+ /* if in that mode already do nothing, except for run */
+ if (mode == engine->mode && mode != LP5521_CMD_RUN)
+ return 0;
+
+ if (mode == LP5521_CMD_RUN) {
+ ret = lp5521_set_engine_mode(engine, LP5521_CMD_RUN);
+ } else if (mode == LP5521_CMD_LOAD) {
+ lp5521_set_engine_mode(engine, LP5521_CMD_DISABLED);
+ lp5521_set_engine_mode(engine, LP5521_CMD_LOAD);
+
+ ret = sysfs_create_group(&dev->kobj, engine->attributes);
+ if (ret)
+ return ret;
+ } else if (mode == LP5521_CMD_DISABLED) {
+ lp5521_set_engine_mode(engine, LP5521_CMD_DISABLED);
+ }
+
+ /* remove load attribute from sysfs if not in load mode */
+ if (engine->mode == LP5521_CMD_LOAD && mode != LP5521_CMD_LOAD)
+ sysfs_remove_group(&dev->kobj, engine->attributes);
+
+ engine->mode = mode;
+
+ return ret;
+}
+
+static int lp5521_do_store_load(struct lp5521_engine *engine,
+ const char *buf, size_t len)
+{
+ struct lp5521_chip *chip = engine_to_lp5521(engine);
+ struct i2c_client *client = chip->client;
+ int ret, nrchars, offset = 0, i = 0;
+ char c[3];
+ unsigned cmd;
+ u8 pattern[LP5521_PROGRAM_LENGTH] = {0};
+
+ while ((offset < len - 1) && (i < LP5521_PROGRAM_LENGTH)) {
+ /* separate sscanfs because length is working only for %s */
+ ret = sscanf(buf + offset, "%2s%n ", c, &nrchars);
+ ret = sscanf(c, "%2x", &cmd);
+ if (ret != 1)
+ goto fail;
+ pattern[i] = (u8)cmd;
+
+ offset += nrchars;
+ i++;
+ }
+
+ /* Each instruction is 16bit long. Check that length is even */
+ if (i % 2)
+ goto fail;
+
+ mutex_lock(&chip->lock);
+ ret = lp5521_load_program(engine, pattern);
+ mutex_unlock(&chip->lock);
+
+ if (ret) {
+ dev_err(&client->dev, "failed loading pattern\n");
+ return ret;
+ }
+
+ return len;
+fail:
+ dev_err(&client->dev, "wrong pattern format\n");
+ return -EINVAL;
+}
+
+static ssize_t store_engine_load(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t len, int nr)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct lp5521_chip *chip = i2c_get_clientdata(client);
+ return lp5521_do_store_load(&chip->engines[nr - 1], buf, len);
+}
+
+#define store_load(nr) \
+static ssize_t store_engine##nr##_load(struct device *dev, \
+ struct device_attribute *attr, \
+ const char *buf, size_t len) \
+{ \
+ return store_engine_load(dev, attr, buf, len, nr); \
+}
+store_load(1)
+store_load(2)
+store_load(3)
+
+static ssize_t show_engine_mode(struct device *dev,
+ struct device_attribute *attr,
+ char *buf, int nr)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct lp5521_chip *chip = i2c_get_clientdata(client);
+ switch (chip->engines[nr - 1].mode) {
+ case LP5521_CMD_RUN:
+ return sprintf(buf, "run\n");
+ case LP5521_CMD_LOAD:
+ return sprintf(buf, "load\n");
+ case LP5521_CMD_DISABLED:
+ return sprintf(buf, "disabled\n");
+ default:
+ return sprintf(buf, "disabled\n");
+ }
+}
+
+#define show_mode(nr) \
+static ssize_t show_engine##nr##_mode(struct device *dev, \
+ struct device_attribute *attr, \
+ char *buf) \
+{ \
+ return show_engine_mode(dev, attr, buf, nr); \
+}
+show_mode(1)
+show_mode(2)
+show_mode(3)
+
+static ssize_t store_engine_mode(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t len, int nr)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct lp5521_chip *chip = i2c_get_clientdata(client);
+ struct lp5521_engine *engine = &chip->engines[nr - 1];
+ mutex_lock(&chip->lock);
+
+ if (!strncmp(buf, "run", 3))
+ lp5521_set_mode(engine, LP5521_CMD_RUN);
+ else if (!strncmp(buf, "load", 4))
+ lp5521_set_mode(engine, LP5521_CMD_LOAD);
+ else if (!strncmp(buf, "disabled", 8))
+ lp5521_set_mode(engine, LP5521_CMD_DISABLED);
+
+ mutex_unlock(&chip->lock);
+ return len;
+}
+
+#define store_mode(nr) \
+static ssize_t store_engine##nr##_mode(struct device *dev, \
+ struct device_attribute *attr, \
+ const char *buf, size_t len) \
+{ \
+ return store_engine_mode(dev, attr, buf, len, nr); \
+}
+store_mode(1)
+store_mode(2)
+store_mode(3)
+
+static ssize_t show_max_current(struct device *dev,
+ struct device_attribute *attr,
+ char *buf)
+{
+ struct led_classdev *led_cdev = dev_get_drvdata(dev);
+ struct lp5521_led *led = cdev_to_led(led_cdev);
+
+ return sprintf(buf, "%d\n", led->max_current);
+}
+
+static ssize_t show_current(struct device *dev,
+ struct device_attribute *attr,
+ char *buf)
+{
+ struct led_classdev *led_cdev = dev_get_drvdata(dev);
+ struct lp5521_led *led = cdev_to_led(led_cdev);
+
+ return sprintf(buf, "%d\n", led->led_current);
+}
+
+static ssize_t store_current(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t len)
+{
+ struct led_classdev *led_cdev = dev_get_drvdata(dev);
+ struct lp5521_led *led = cdev_to_led(led_cdev);
+ struct lp5521_chip *chip = led_to_lp5521(led);
+ ssize_t ret;
+ unsigned long curr;
+
+ if (strict_strtoul(buf, 0, &curr))
+ return -EINVAL;
+
+ if (curr > led->max_current)
+ return -EINVAL;
+
+ mutex_lock(&chip->lock);
+ ret = lp5521_set_led_current(chip, led->id, curr);
+ mutex_unlock(&chip->lock);
+
+ if (ret < 0)
+ return ret;
+
+ led->led_current = (u8)curr;
+
+ return len;
+}
+
+static ssize_t lp5521_selftest(struct device *dev,
+ struct device_attribute *attr,
+ char *buf)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct lp5521_chip *chip = i2c_get_clientdata(client);
+ int ret;
+
+ mutex_lock(&chip->lock);
+ ret = lp5521_run_selftest(chip, buf);
+ mutex_unlock(&chip->lock);
+ return sprintf(buf, "%s\n", ret ? "FAIL" : "OK");
+}
+
+/* led class device attributes */
+static DEVICE_ATTR(led_current, S_IRUGO | S_IWUGO, show_current, store_current);
+static DEVICE_ATTR(max_current, S_IRUGO , show_max_current, NULL);
+
+static struct attribute *lp5521_led_attributes[] = {
+ &dev_attr_led_current.attr,
+ &dev_attr_max_current.attr,
+ NULL,
+};
+
+static struct attribute_group lp5521_led_attribute_group = {
+ .attrs = lp5521_led_attributes
+};
+
+/* device attributes */
+static DEVICE_ATTR(engine1_mode, S_IRUGO | S_IWUGO,
+ show_engine1_mode, store_engine1_mode);
+static DEVICE_ATTR(engine2_mode, S_IRUGO | S_IWUGO,
+ show_engine2_mode, store_engine2_mode);
+static DEVICE_ATTR(engine3_mode, S_IRUGO | S_IWUGO,
+ show_engine3_mode, store_engine3_mode);
+static DEVICE_ATTR(engine1_load, S_IWUGO, NULL, store_engine1_load);
+static DEVICE_ATTR(engine2_load, S_IWUGO, NULL, store_engine2_load);
+static DEVICE_ATTR(engine3_load, S_IWUGO, NULL, store_engine3_load);
+static DEVICE_ATTR(selftest, S_IRUGO, lp5521_selftest, NULL);
+
+static struct attribute *lp5521_attributes[] = {
+ &dev_attr_engine1_mode.attr,
+ &dev_attr_engine2_mode.attr,
+ &dev_attr_engine3_mode.attr,
+ &dev_attr_selftest.attr,
+ NULL
+};
+
+static struct attribute *lp5521_engine1_attributes[] = {
+ &dev_attr_engine1_load.attr,
+ NULL
+};
+
+static struct attribute *lp5521_engine2_attributes[] = {
+ &dev_attr_engine2_load.attr,
+ NULL
+};
+
+static struct attribute *lp5521_engine3_attributes[] = {
+ &dev_attr_engine3_load.attr,
+ NULL
+};
+
+static const struct attribute_group lp5521_group = {
+ .attrs = lp5521_attributes,
+};
+
+static const struct attribute_group lp5521_engine_group[] = {
+ {.attrs = lp5521_engine1_attributes },
+ {.attrs = lp5521_engine2_attributes },
+ {.attrs = lp5521_engine3_attributes },
+};
+
+static int lp5521_register_sysfs(struct i2c_client *client)
+{
+ struct device *dev = &client->dev;
+ return sysfs_create_group(&dev->kobj, &lp5521_group);
+}
+
+static void lp5521_unregister_sysfs(struct i2c_client *client)
+{
+ struct lp5521_chip *chip = i2c_get_clientdata(client);
+ struct device *dev = &client->dev;
+ int i;
+
+ sysfs_remove_group(&dev->kobj, &lp5521_group);
+
+ for (i = 0; i < ARRAY_SIZE(chip->engines); i++) {
+ if (chip->engines[i].mode == LP5521_CMD_LOAD)
+ sysfs_remove_group(&dev->kobj,
+ chip->engines[i].attributes);
+ }
+
+ for (i = 0; i < chip->num_leds; i++)
+ sysfs_remove_group(&chip->leds[i].cdev.dev->kobj,
+ &lp5521_led_attribute_group);
+}
+
+static int __init lp5521_init_led(struct lp5521_led *led,
+ struct i2c_client *client,
+ int chan, struct lp5521_platform_data *pdata)
+{
+ struct device *dev = &client->dev;
+ char name[32];
+ int res;
+
+ if (chan >= LP5521_MAX_LEDS)
+ return -EINVAL;
+
+ if (pdata->led_config[chan].led_current == 0)
+ return 0;
+
+ led->led_current = pdata->led_config[chan].led_current;
+ led->max_current = pdata->led_config[chan].max_current;
+ led->chan_nr = pdata->led_config[chan].chan_nr;
+
+ if (led->chan_nr >= LP5521_MAX_LEDS) {
+ dev_err(dev, "Use channel numbers between 0 and %d\n",
+ LP5521_MAX_LEDS - 1);
+ return -EINVAL;
+ }
+
+ snprintf(name, sizeof(name), "%s:channel%d", client->name, chan);
+ led->cdev.brightness_set = lp5521_set_brightness;
+ led->cdev.name = name;
+ res = led_classdev_register(dev, &led->cdev);
+ if (res < 0) {
+ dev_err(dev, "couldn't register led on channel %d\n", chan);
+ return res;
+ }
+
+ res = sysfs_create_group(&led->cdev.dev->kobj,
+ &lp5521_led_attribute_group);
+ if (res < 0) {
+ dev_err(dev, "couldn't register current attribute\n");
+ led_classdev_unregister(&led->cdev);
+ return res;
+ }
+ return 0;
+}
+
+static int lp5521_probe(struct i2c_client *client,
+ const struct i2c_device_id *id)
+{
+ struct lp5521_chip *chip;
+ struct lp5521_platform_data *pdata;
+ int ret, i, led;
+
+ chip = kzalloc(sizeof(*chip), GFP_KERNEL);
+ if (!chip)
+ return -ENOMEM;
+
+ i2c_set_clientdata(client, chip);
+ chip->client = client;
+
+ pdata = client->dev.platform_data;
+
+ if (!pdata) {
+ dev_err(&client->dev, "no platform data\n");
+ ret = -EINVAL;
+ goto fail1;
+ }
+
+ mutex_init(&chip->lock);
+
+ chip->pdata = pdata;
+
+ if (pdata->setup_resources) {
+ ret = pdata->setup_resources();
+ if (ret < 0)
+ goto fail1;
+ }
+
+ if (pdata->enable) {
+ pdata->enable(0);
+ usleep_range(1000, 10000);
+ pdata->enable(1);
+ usleep_range(1000, 10000); /* Spec says min 500us */
+ }
+
+ ret = lp5521_detect(client);
+
+ if (ret) {
+ dev_err(&client->dev, "Chip not found\n");
+ goto fail2;
+ }
+
+ dev_info(&client->dev, "%s programmable led chip found\n", id->name);
+
+ ret = lp5521_configure(client, lp5521_engine_group);
+ if (ret < 0) {
+ dev_err(&client->dev, "error configuring chip\n");
+ goto fail2;
+ }
+
+ /* Initialize leds */
+ chip->num_channels = pdata->num_channels;
+ chip->num_leds = 0;
+ led = 0;
+ for (i = 0; i < pdata->num_channels; i++) {
+ /* Do not initialize channels that are not connected */
+ if (pdata->led_config[i].led_current == 0)
+ continue;
+
+ ret = lp5521_init_led(&chip->leds[led], client, i, pdata);
+ if (ret) {
+ dev_err(&client->dev, "error initializing leds\n");
+ goto fail3;
+ }
+ chip->num_leds++;
+
+ chip->leds[led].id = led;
+ /* Set initial LED current */
+ lp5521_set_led_current(chip, led,
+ chip->leds[led].led_current);
+
+ INIT_WORK(&(chip->leds[led].brightness_work),
+ lp5521_led_brightness_work);
+
+ led++;
+ }
+
+ ret = lp5521_register_sysfs(client);
+ if (ret) {
+ dev_err(&client->dev, "registering sysfs failed\n");
+ goto fail3;
+ }
+ return ret;
+fail3:
+ for (i = 0; i < chip->num_leds; i++) {
+ led_classdev_unregister(&chip->leds[i].cdev);
+ cancel_work_sync(&chip->leds[i].brightness_work);
+ }
+fail2:
+ if (pdata->enable)
+ pdata->enable(0);
+ if (pdata->release_resources)
+ pdata->release_resources();
+fail1:
+ kfree(chip);
+ return ret;
+}
+
+static int lp5521_remove(struct i2c_client *client)
+{
+ struct lp5521_chip *chip = i2c_get_clientdata(client);
+ int i;
+
+ lp5521_unregister_sysfs(client);
+
+ for (i = 0; i < chip->num_leds; i++) {
+ led_classdev_unregister(&chip->leds[i].cdev);
+ cancel_work_sync(&chip->leds[i].brightness_work);
+ }
+
+ if (chip->pdata->enable)
+ chip->pdata->enable(0);
+ if (chip->pdata->release_resources)
+ chip->pdata->release_resources();
+ kfree(chip);
+ return 0;
+}
+
+static const struct i2c_device_id lp5521_id[] = {
+ { "lp5521", 0 }, /* Three channel chip */
+ { }
+};
+MODULE_DEVICE_TABLE(i2c, lp5521_id);
+
+static struct i2c_driver lp5521_driver = {
+ .driver = {
+ .name = "lp5521",
+ },
+ .probe = lp5521_probe,
+ .remove = lp5521_remove,
+ .id_table = lp5521_id,
+};
+
+static int __init lp5521_init(void)
+{
+ int ret;
+
+ ret = i2c_add_driver(&lp5521_driver);
+
+ if (ret < 0)
+ printk(KERN_ALERT "Adding lp5521 driver failed\n");
+
+ return ret;
+}
+
+static void __exit lp5521_exit(void)
+{
+ i2c_del_driver(&lp5521_driver);
+}
+
+module_init(lp5521_init);
+module_exit(lp5521_exit);
+
+MODULE_AUTHOR("Mathias Nyman, Yuri Zaporozhets, Samu Onkalo");
+MODULE_DESCRIPTION("LP5521 LED engine");
+MODULE_LICENSE("GPL v2");
--- /dev/null
+/*
+ * lp5523.c - LP5523 LED Driver
+ *
+ * Copyright (C) 2010 Nokia Corporation
+ *
+ * Contact: Samu Onkalo <samu.p.onkalo@nokia.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
+ * 02110-1301 USA
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/i2c.h>
+#include <linux/mutex.h>
+#include <linux/gpio.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/ctype.h>
+#include <linux/spinlock.h>
+#include <linux/wait.h>
+#include <linux/leds.h>
+#include <linux/leds-lp5523.h>
+#include <linux/workqueue.h>
+#include <linux/slab.h>
+
+#define LP5523_REG_ENABLE 0x00
+#define LP5523_REG_OP_MODE 0x01
+#define LP5523_REG_RATIOMETRIC_MSB 0x02
+#define LP5523_REG_RATIOMETRIC_LSB 0x03
+#define LP5523_REG_ENABLE_LEDS_MSB 0x04
+#define LP5523_REG_ENABLE_LEDS_LSB 0x05
+#define LP5523_REG_LED_CNTRL_BASE 0x06
+#define LP5523_REG_LED_PWM_BASE 0x16
+#define LP5523_REG_LED_CURRENT_BASE 0x26
+#define LP5523_REG_CONFIG 0x36
+#define LP5523_REG_CHANNEL1_PC 0x37
+#define LP5523_REG_CHANNEL2_PC 0x38
+#define LP5523_REG_CHANNEL3_PC 0x39
+#define LP5523_REG_STATUS 0x3a
+#define LP5523_REG_GPO 0x3b
+#define LP5523_REG_VARIABLE 0x3c
+#define LP5523_REG_RESET 0x3d
+#define LP5523_REG_TEMP_CTRL 0x3e
+#define LP5523_REG_TEMP_READ 0x3f
+#define LP5523_REG_TEMP_WRITE 0x40
+#define LP5523_REG_LED_TEST_CTRL 0x41
+#define LP5523_REG_LED_TEST_ADC 0x42
+#define LP5523_REG_ENG1_VARIABLE 0x45
+#define LP5523_REG_ENG2_VARIABLE 0x46
+#define LP5523_REG_ENG3_VARIABLE 0x47
+#define LP5523_REG_MASTER_FADER1 0x48
+#define LP5523_REG_MASTER_FADER2 0x49
+#define LP5523_REG_MASTER_FADER3 0x4a
+#define LP5523_REG_CH1_PROG_START 0x4c
+#define LP5523_REG_CH2_PROG_START 0x4d
+#define LP5523_REG_CH3_PROG_START 0x4e
+#define LP5523_REG_PROG_PAGE_SEL 0x4f
+#define LP5523_REG_PROG_MEM 0x50
+
+#define LP5523_CMD_LOAD 0x15 /* 00010101 */
+#define LP5523_CMD_RUN 0x2a /* 00101010 */
+#define LP5523_CMD_DISABLED 0x00 /* 00000000 */
+
+#define LP5523_ENABLE 0x40
+#define LP5523_AUTO_INC 0x40
+#define LP5523_PWR_SAVE 0x20
+#define LP5523_PWM_PWR_SAVE 0x04
+#define LP5523_CP_1 0x08
+#define LP5523_CP_1_5 0x10
+#define LP5523_CP_AUTO 0x18
+#define LP5523_INT_CLK 0x01
+#define LP5523_AUTO_CLK 0x02
+#define LP5523_EN_LEDTEST 0x80
+#define LP5523_LEDTEST_DONE 0x80
+
+#define LP5523_DEFAULT_CURRENT 50 /* microAmps */
+#define LP5523_PROGRAM_LENGTH 32 /* in bytes */
+#define LP5523_PROGRAM_PAGES 6
+#define LP5523_ADC_SHORTCIRC_LIM 80
+
+#define LP5523_LEDS 9
+#define LP5523_ENGINES 3
+
+#define LP5523_ENG_MASK_BASE 0x30 /* 00110000 */
+
+#define LP5523_ENG_STATUS_MASK 0x07 /* 00000111 */
+
+#define LP5523_IRQ_FLAGS IRQF_TRIGGER_FALLING
+
+#define LP5523_EXT_CLK_USED 0x08
+
+#define LED_ACTIVE(mux, led) (!!(mux & (0x0001 << led)))
+#define SHIFT_MASK(id) (((id) - 1) * 2)
+
+struct lp5523_engine {
+ const struct attribute_group *attributes;
+ int id;
+ u8 mode;
+ u8 prog_page;
+ u8 mux_page;
+ u16 led_mux;
+ u8 engine_mask;
+};
+
+struct lp5523_led {
+ int id;
+ u8 chan_nr;
+ u8 led_current;
+ u8 max_current;
+ struct led_classdev cdev;
+ struct work_struct brightness_work;
+ u8 brightness;
+};
+
+struct lp5523_chip {
+ struct mutex lock; /* Serialize control */
+ struct i2c_client *client;
+ struct lp5523_engine engines[LP5523_ENGINES];
+ struct lp5523_led leds[LP5523_LEDS];
+ struct lp5523_platform_data *pdata;
+ u8 num_channels;
+ u8 num_leds;
+};
+
+#define cdev_to_led(c) container_of(c, struct lp5523_led, cdev)
+
+static struct lp5523_chip *engine_to_lp5523(struct lp5523_engine *engine)
+{
+ return container_of(engine, struct lp5523_chip,
+ engines[engine->id - 1]);
+}
+
+static struct lp5523_chip *led_to_lp5523(struct lp5523_led *led)
+{
+ return container_of(led, struct lp5523_chip,
+ leds[led->id]);
+}
+
+static int lp5523_set_mode(struct lp5523_engine *engine, u8 mode);
+static int lp5523_set_engine_mode(struct lp5523_engine *engine, u8 mode);
+static int lp5523_load_program(struct lp5523_engine *engine, u8 *pattern);
+
+static void lp5523_led_brightness_work(struct work_struct *work);
+
+static int lp5523_write(struct i2c_client *client, u8 reg, u8 value)
+{
+ return i2c_smbus_write_byte_data(client, reg, value);
+}
+
+static int lp5523_read(struct i2c_client *client, u8 reg, u8 *buf)
+{
+ s32 ret = i2c_smbus_read_byte_data(client, reg);
+
+ if (ret < 0)
+ return -EIO;
+
+ *buf = ret;
+ return 0;
+}
+
+static int lp5523_detect(struct i2c_client *client)
+{
+ int ret;
+ u8 buf;
+
+ ret = lp5523_write(client, LP5523_REG_ENABLE, 0x40);
+ if (ret)
+ return ret;
+ ret = lp5523_read(client, LP5523_REG_ENABLE, &buf);
+ if (ret)
+ return ret;
+ if (buf == 0x40)
+ return 0;
+ else
+ return -ENODEV;
+}
+
+static int lp5523_configure(struct i2c_client *client)
+{
+ struct lp5523_chip *chip = i2c_get_clientdata(client);
+ int ret = 0;
+ u8 status;
+
+ /* one pattern per engine setting led mux start and stop addresses */
+ u8 pattern[][LP5523_PROGRAM_LENGTH] = {
+ { 0x9c, 0x30, 0x9c, 0xb0, 0x9d, 0x80, 0xd8, 0x00, 0},
+ { 0x9c, 0x40, 0x9c, 0xc0, 0x9d, 0x80, 0xd8, 0x00, 0},
+ { 0x9c, 0x50, 0x9c, 0xd0, 0x9d, 0x80, 0xd8, 0x00, 0},
+ };
+
+ lp5523_write(client, LP5523_REG_RESET, 0xff);
+
+ usleep_range(10000, 100000);
+
+ ret |= lp5523_write(client, LP5523_REG_ENABLE, LP5523_ENABLE);
+ /* Chip startup time after reset is 500 us */
+ usleep_range(1000, 10000);
+
+ ret |= lp5523_write(client, LP5523_REG_CONFIG,
+ LP5523_AUTO_INC | LP5523_PWR_SAVE |
+ LP5523_CP_AUTO | LP5523_AUTO_CLK |
+ LP5523_PWM_PWR_SAVE);
+
+ /* turn on all leds */
+ ret |= lp5523_write(client, LP5523_REG_ENABLE_LEDS_MSB, 0x01);
+ ret |= lp5523_write(client, LP5523_REG_ENABLE_LEDS_LSB, 0xff);
+
+ /* hardcode 32 bytes of memory for each engine from program memory */
+ ret |= lp5523_write(client, LP5523_REG_CH1_PROG_START, 0x00);
+ ret |= lp5523_write(client, LP5523_REG_CH2_PROG_START, 0x10);
+ ret |= lp5523_write(client, LP5523_REG_CH3_PROG_START, 0x20);
+
+ /* write led mux address space for each channel */
+ ret |= lp5523_load_program(&chip->engines[0], pattern[0]);
+ ret |= lp5523_load_program(&chip->engines[1], pattern[1]);
+ ret |= lp5523_load_program(&chip->engines[2], pattern[2]);
+
+ if (ret) {
+ dev_err(&client->dev, "could not load mux programs\n");
+ return -1;
+ }
+
+ /* set all engines exec state and mode to run 00101010 */
+ ret |= lp5523_write(client, LP5523_REG_ENABLE,
+ (LP5523_CMD_RUN | LP5523_ENABLE));
+
+ ret |= lp5523_write(client, LP5523_REG_OP_MODE, LP5523_CMD_RUN);
+
+ if (ret) {
+ dev_err(&client->dev, "could not start mux programs\n");
+ return -1;
+ }
+
+ /* Wait 3ms and check the engine status */
+ usleep_range(3000, 20000);
+ lp5523_read(client, LP5523_REG_STATUS, &status);
+ status &= LP5523_ENG_STATUS_MASK;
+
+ if (status == LP5523_ENG_STATUS_MASK) {
+ dev_dbg(&client->dev, "all engines configured\n");
+ } else {
+ dev_info(&client->dev, "status == %x\n", status);
+ dev_err(&client->dev, "cound not configure LED engine\n");
+ return -1;
+ }
+
+ dev_info(&client->dev, "disabling engines\n");
+
+ ret |= lp5523_write(client, LP5523_REG_OP_MODE, LP5523_CMD_DISABLED);
+
+ return ret;
+}
+
+static int lp5523_set_engine_mode(struct lp5523_engine *engine, u8 mode)
+{
+ struct lp5523_chip *chip = engine_to_lp5523(engine);
+ struct i2c_client *client = chip->client;
+ int ret;
+ u8 engine_state;
+
+ ret = lp5523_read(client, LP5523_REG_OP_MODE, &engine_state);
+ if (ret)
+ goto fail;
+
+ engine_state &= ~(engine->engine_mask);
+
+ /* set mode only for this engine */
+ mode &= engine->engine_mask;
+
+ engine_state |= mode;
+
+ ret |= lp5523_write(client, LP5523_REG_OP_MODE, engine_state);
+fail:
+ return ret;
+}
+
+static int lp5523_load_mux(struct lp5523_engine *engine, u16 mux)
+{
+ struct lp5523_chip *chip = engine_to_lp5523(engine);
+ struct i2c_client *client = chip->client;
+ int ret = 0;
+
+ ret |= lp5523_set_engine_mode(engine, LP5523_CMD_LOAD);
+
+ ret |= lp5523_write(client, LP5523_REG_PROG_PAGE_SEL, engine->mux_page);
+ ret |= lp5523_write(client, LP5523_REG_PROG_MEM,
+ (u8)(mux >> 8));
+ ret |= lp5523_write(client, LP5523_REG_PROG_MEM + 1, (u8)(mux));
+ engine->led_mux = mux;
+
+ return ret;
+}
+
+static int lp5523_load_program(struct lp5523_engine *engine, u8 *pattern)
+{
+ struct lp5523_chip *chip = engine_to_lp5523(engine);
+ struct i2c_client *client = chip->client;
+
+ int ret = 0;
+
+ ret |= lp5523_set_engine_mode(engine, LP5523_CMD_LOAD);
+
+ ret |= lp5523_write(client, LP5523_REG_PROG_PAGE_SEL,
+ engine->prog_page);
+ ret |= i2c_smbus_write_i2c_block_data(client, LP5523_REG_PROG_MEM,
+ LP5523_PROGRAM_LENGTH, pattern);
+
+ return ret;
+}
+
+static int lp5523_run_program(struct lp5523_engine *engine)
+{
+ struct lp5523_chip *chip = engine_to_lp5523(engine);
+ struct i2c_client *client = chip->client;
+ int ret;
+
+ ret = lp5523_write(client, LP5523_REG_ENABLE,
+ LP5523_CMD_RUN | LP5523_ENABLE);
+ if (ret)
+ goto fail;
+
+ ret = lp5523_set_engine_mode(engine, LP5523_CMD_RUN);
+fail:
+ return ret;
+}
+
+static int lp5523_mux_parse(const char *buf, u16 *mux, size_t len)
+{
+ int i;
+ u16 tmp_mux = 0;
+ len = len < LP5523_LEDS ? len : LP5523_LEDS;
+ for (i = 0; i < len; i++) {
+ switch (buf[i]) {
+ case '1':
+ tmp_mux |= (1 << i);
+ break;
+ case '0':
+ break;
+ case '\n':
+ i = len;
+ break;
+ default:
+ return -1;
+ }
+ }
+ *mux = tmp_mux;
+
+ return 0;
+}
+
+static void lp5523_mux_to_array(u16 led_mux, char *array)
+{
+ int i, pos = 0;
+ for (i = 0; i < LP5523_LEDS; i++)
+ pos += sprintf(array + pos, "%x", LED_ACTIVE(led_mux, i));
+
+ array[pos] = '\0';
+}
+
+/*--------------------------------------------------------------*/
+/* Sysfs interface */
+/*--------------------------------------------------------------*/
+
+static ssize_t show_engine_leds(struct device *dev,
+ struct device_attribute *attr,
+ char *buf, int nr)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct lp5523_chip *chip = i2c_get_clientdata(client);
+ char mux[LP5523_LEDS + 1];
+
+ lp5523_mux_to_array(chip->engines[nr - 1].led_mux, mux);
+
+ return sprintf(buf, "%s\n", mux);
+}
+
+#define show_leds(nr) \
+static ssize_t show_engine##nr##_leds(struct device *dev, \
+ struct device_attribute *attr, \
+ char *buf) \
+{ \
+ return show_engine_leds(dev, attr, buf, nr); \
+}
+show_leds(1)
+show_leds(2)
+show_leds(3)
+
+static ssize_t store_engine_leds(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t len, int nr)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct lp5523_chip *chip = i2c_get_clientdata(client);
+ u16 mux = 0;
+
+ if (lp5523_mux_parse(buf, &mux, len))
+ return -EINVAL;
+
+ if (lp5523_load_mux(&chip->engines[nr - 1], mux))
+ return -EINVAL;
+
+ return len;
+}
+
+#define store_leds(nr) \
+static ssize_t store_engine##nr##_leds(struct device *dev, \
+ struct device_attribute *attr, \
+ const char *buf, size_t len) \
+{ \
+ return store_engine_leds(dev, attr, buf, len, nr); \
+}
+store_leds(1)
+store_leds(2)
+store_leds(3)
+
+static ssize_t lp5523_selftest(struct device *dev,
+ struct device_attribute *attr,
+ char *buf)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct lp5523_chip *chip = i2c_get_clientdata(client);
+ int i, ret, pos = 0;
+ int led = 0;
+ u8 status, adc, vdd;
+
+ mutex_lock(&chip->lock);
+
+ ret = lp5523_read(chip->client, LP5523_REG_STATUS, &status);
+ if (ret < 0)
+ goto fail;
+
+ /* Check that ext clock is really in use if requested */
+ if ((chip->pdata) && (chip->pdata->clock_mode == LP5523_CLOCK_EXT))
+ if ((status & LP5523_EXT_CLK_USED) == 0)
+ goto fail;
+
+ /* Measure VDD (i.e. VBAT) first (channel 16 corresponds to VDD) */
+ lp5523_write(chip->client, LP5523_REG_LED_TEST_CTRL,
+ LP5523_EN_LEDTEST | 16);
+ usleep_range(3000, 10000);
+ ret = lp5523_read(chip->client, LP5523_REG_STATUS, &status);
+ if (!(status & LP5523_LEDTEST_DONE))
+ usleep_range(3000, 10000);
+
+ ret |= lp5523_read(chip->client, LP5523_REG_LED_TEST_ADC, &vdd);
+ vdd--; /* There may be some fluctuation in measurement */
+
+ for (i = 0; i < LP5523_LEDS; i++) {
+ /* Skip non-existing channels */
+ if (chip->pdata->led_config[i].led_current == 0)
+ continue;
+
+ /* Set default current */
+ lp5523_write(chip->client,
+ LP5523_REG_LED_CURRENT_BASE + i,
+ chip->pdata->led_config[i].led_current);
+
+ lp5523_write(chip->client, LP5523_REG_LED_PWM_BASE + i, 0xff);
+ /* let current stabilize 2ms before measurements start */
+ usleep_range(2000, 10000);
+ lp5523_write(chip->client,
+ LP5523_REG_LED_TEST_CTRL,
+ LP5523_EN_LEDTEST | i);
+ /* ledtest takes 2.7ms */
+ usleep_range(3000, 10000);
+ ret = lp5523_read(chip->client, LP5523_REG_STATUS, &status);
+ if (!(status & LP5523_LEDTEST_DONE))
+ usleep_range(3000, 10000);
+ ret |= lp5523_read(chip->client, LP5523_REG_LED_TEST_ADC, &adc);
+
+ if (adc >= vdd || adc < LP5523_ADC_SHORTCIRC_LIM)
+ pos += sprintf(buf + pos, "LED %d FAIL\n", i);
+
+ lp5523_write(chip->client, LP5523_REG_LED_PWM_BASE + i, 0x00);
+
+ /* Restore current */
+ lp5523_write(chip->client,
+ LP5523_REG_LED_CURRENT_BASE + i,
+ chip->leds[led].led_current);
+ led++;
+ }
+ if (pos == 0)
+ pos = sprintf(buf, "OK\n");
+ goto release_lock;
+fail:
+ pos = sprintf(buf, "FAIL\n");
+
+release_lock:
+ mutex_unlock(&chip->lock);
+
+ return pos;
+}
+
+static void lp5523_set_brightness(struct led_classdev *cdev,
+ enum led_brightness brightness)
+{
+ struct lp5523_led *led = cdev_to_led(cdev);
+
+ led->brightness = (u8)brightness;
+
+ schedule_work(&led->brightness_work);
+}
+
+static void lp5523_led_brightness_work(struct work_struct *work)
+{
+ struct lp5523_led *led = container_of(work,
+ struct lp5523_led,
+ brightness_work);
+ struct lp5523_chip *chip = led_to_lp5523(led);
+ struct i2c_client *client = chip->client;
+
+ mutex_lock(&chip->lock);
+
+ lp5523_write(client, LP5523_REG_LED_PWM_BASE + led->chan_nr,
+ led->brightness);
+
+ mutex_unlock(&chip->lock);
+}
+
+static int lp5523_do_store_load(struct lp5523_engine *engine,
+ const char *buf, size_t len)
+{
+ struct lp5523_chip *chip = engine_to_lp5523(engine);
+ struct i2c_client *client = chip->client;
+ int ret, nrchars, offset = 0, i = 0;
+ char c[3];
+ unsigned cmd;
+ u8 pattern[LP5523_PROGRAM_LENGTH] = {0};
+
+ while ((offset < len - 1) && (i < LP5523_PROGRAM_LENGTH)) {
+ /* separate sscanfs because length is working only for %s */
+ ret = sscanf(buf + offset, "%2s%n ", c, &nrchars);
+ ret = sscanf(c, "%2x", &cmd);
+ if (ret != 1)
+ goto fail;
+ pattern[i] = (u8)cmd;
+
+ offset += nrchars;
+ i++;
+ }
+
+ /* Each instruction is 16bit long. Check that length is even */
+ if (i % 2)
+ goto fail;
+
+ mutex_lock(&chip->lock);
+
+ ret = lp5523_load_program(engine, pattern);
+ mutex_unlock(&chip->lock);
+
+ if (ret) {
+ dev_err(&client->dev, "failed loading pattern\n");
+ return ret;
+ }
+
+ return len;
+fail:
+ dev_err(&client->dev, "wrong pattern format\n");
+ return -EINVAL;
+}
+
+static ssize_t store_engine_load(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t len, int nr)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct lp5523_chip *chip = i2c_get_clientdata(client);
+ return lp5523_do_store_load(&chip->engines[nr - 1], buf, len);
+}
+
+#define store_load(nr) \
+static ssize_t store_engine##nr##_load(struct device *dev, \
+ struct device_attribute *attr, \
+ const char *buf, size_t len) \
+{ \
+ return store_engine_load(dev, attr, buf, len, nr); \
+}
+store_load(1)
+store_load(2)
+store_load(3)
+
+static ssize_t show_engine_mode(struct device *dev,
+ struct device_attribute *attr,
+ char *buf, int nr)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct lp5523_chip *chip = i2c_get_clientdata(client);
+ switch (chip->engines[nr - 1].mode) {
+ case LP5523_CMD_RUN:
+ return sprintf(buf, "run\n");
+ case LP5523_CMD_LOAD:
+ return sprintf(buf, "load\n");
+ case LP5523_CMD_DISABLED:
+ return sprintf(buf, "disabled\n");
+ default:
+ return sprintf(buf, "disabled\n");
+ }
+}
+
+#define show_mode(nr) \
+static ssize_t show_engine##nr##_mode(struct device *dev, \
+ struct device_attribute *attr, \
+ char *buf) \
+{ \
+ return show_engine_mode(dev, attr, buf, nr); \
+}
+show_mode(1)
+show_mode(2)
+show_mode(3)
+
+static ssize_t store_engine_mode(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t len, int nr)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct lp5523_chip *chip = i2c_get_clientdata(client);
+ struct lp5523_engine *engine = &chip->engines[nr - 1];
+ mutex_lock(&chip->lock);
+
+ if (!strncmp(buf, "run", 3))
+ lp5523_set_mode(engine, LP5523_CMD_RUN);
+ else if (!strncmp(buf, "load", 4))
+ lp5523_set_mode(engine, LP5523_CMD_LOAD);
+ else if (!strncmp(buf, "disabled", 8))
+ lp5523_set_mode(engine, LP5523_CMD_DISABLED);
+
+ mutex_unlock(&chip->lock);
+ return len;
+}
+
+#define store_mode(nr) \
+static ssize_t store_engine##nr##_mode(struct device *dev, \
+ struct device_attribute *attr, \
+ const char *buf, size_t len) \
+{ \
+ return store_engine_mode(dev, attr, buf, len, nr); \
+}
+store_mode(1)
+store_mode(2)
+store_mode(3)
+
+static ssize_t show_max_current(struct device *dev,
+ struct device_attribute *attr,
+ char *buf)
+{
+ struct led_classdev *led_cdev = dev_get_drvdata(dev);
+ struct lp5523_led *led = cdev_to_led(led_cdev);
+
+ return sprintf(buf, "%d\n", led->max_current);
+}
+
+static ssize_t show_current(struct device *dev,
+ struct device_attribute *attr,
+ char *buf)
+{
+ struct led_classdev *led_cdev = dev_get_drvdata(dev);
+ struct lp5523_led *led = cdev_to_led(led_cdev);
+
+ return sprintf(buf, "%d\n", led->led_current);
+}
+
+static ssize_t store_current(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t len)
+{
+ struct led_classdev *led_cdev = dev_get_drvdata(dev);
+ struct lp5523_led *led = cdev_to_led(led_cdev);
+ struct lp5523_chip *chip = led_to_lp5523(led);
+ ssize_t ret;
+ unsigned long curr;
+
+ if (strict_strtoul(buf, 0, &curr))
+ return -EINVAL;
+
+ if (curr > led->max_current)
+ return -EINVAL;
+
+ mutex_lock(&chip->lock);
+ ret = lp5523_write(chip->client,
+ LP5523_REG_LED_CURRENT_BASE + led->chan_nr,
+ (u8)curr);
+ mutex_unlock(&chip->lock);
+
+ if (ret < 0)
+ return ret;
+
+ led->led_current = (u8)curr;
+
+ return len;
+}
+
+/* led class device attributes */
+static DEVICE_ATTR(led_current, S_IRUGO | S_IWUGO, show_current, store_current);
+static DEVICE_ATTR(max_current, S_IRUGO , show_max_current, NULL);
+
+static struct attribute *lp5523_led_attributes[] = {
+ &dev_attr_led_current.attr,
+ &dev_attr_max_current.attr,
+ NULL,
+};
+
+static struct attribute_group lp5523_led_attribute_group = {
+ .attrs = lp5523_led_attributes
+};
+
+/* device attributes */
+static DEVICE_ATTR(engine1_mode, S_IRUGO | S_IWUGO,
+ show_engine1_mode, store_engine1_mode);
+static DEVICE_ATTR(engine2_mode, S_IRUGO | S_IWUGO,
+ show_engine2_mode, store_engine2_mode);
+static DEVICE_ATTR(engine3_mode, S_IRUGO | S_IWUGO,
+ show_engine3_mode, store_engine3_mode);
+static DEVICE_ATTR(engine1_leds, S_IRUGO | S_IWUGO,
+ show_engine1_leds, store_engine1_leds);
+static DEVICE_ATTR(engine2_leds, S_IRUGO | S_IWUGO,
+ show_engine2_leds, store_engine2_leds);
+static DEVICE_ATTR(engine3_leds, S_IRUGO | S_IWUGO,
+ show_engine3_leds, store_engine3_leds);
+static DEVICE_ATTR(engine1_load, S_IWUGO, NULL, store_engine1_load);
+static DEVICE_ATTR(engine2_load, S_IWUGO, NULL, store_engine2_load);
+static DEVICE_ATTR(engine3_load, S_IWUGO, NULL, store_engine3_load);
+static DEVICE_ATTR(selftest, S_IRUGO, lp5523_selftest, NULL);
+
+static struct attribute *lp5523_attributes[] = {
+ &dev_attr_engine1_mode.attr,
+ &dev_attr_engine2_mode.attr,
+ &dev_attr_engine3_mode.attr,
+ &dev_attr_selftest.attr,
+ NULL
+};
+
+static struct attribute *lp5523_engine1_attributes[] = {
+ &dev_attr_engine1_load.attr,
+ &dev_attr_engine1_leds.attr,
+ NULL
+};
+
+static struct attribute *lp5523_engine2_attributes[] = {
+ &dev_attr_engine2_load.attr,
+ &dev_attr_engine2_leds.attr,
+ NULL
+};
+
+static struct attribute *lp5523_engine3_attributes[] = {
+ &dev_attr_engine3_load.attr,
+ &dev_attr_engine3_leds.attr,
+ NULL
+};
+
+static const struct attribute_group lp5523_group = {
+ .attrs = lp5523_attributes,
+};
+
+static const struct attribute_group lp5523_engine_group[] = {
+ {.attrs = lp5523_engine1_attributes },
+ {.attrs = lp5523_engine2_attributes },
+ {.attrs = lp5523_engine3_attributes },
+};
+
+static int lp5523_register_sysfs(struct i2c_client *client)
+{
+ struct device *dev = &client->dev;
+ int ret;
+
+ ret = sysfs_create_group(&dev->kobj, &lp5523_group);
+ if (ret < 0)
+ return ret;
+
+ return 0;
+}
+
+static void lp5523_unregister_sysfs(struct i2c_client *client)
+{
+ struct lp5523_chip *chip = i2c_get_clientdata(client);
+ struct device *dev = &client->dev;
+ int i;
+
+ sysfs_remove_group(&dev->kobj, &lp5523_group);
+
+ for (i = 0; i < ARRAY_SIZE(chip->engines); i++)
+ if (chip->engines[i].mode == LP5523_CMD_LOAD)
+ sysfs_remove_group(&dev->kobj, &lp5523_engine_group[i]);
+
+ for (i = 0; i < chip->num_leds; i++)
+ sysfs_remove_group(&chip->leds[i].cdev.dev->kobj,
+ &lp5523_led_attribute_group);
+}
+
+/*--------------------------------------------------------------*/
+/* Set chip operating mode */
+/*--------------------------------------------------------------*/
+static int lp5523_set_mode(struct lp5523_engine *engine, u8 mode)
+{
+ /* engine to chip */
+ struct lp5523_chip *chip = engine_to_lp5523(engine);
+ struct i2c_client *client = chip->client;
+ struct device *dev = &client->dev;
+ int ret = 0;
+
+ /* if in that mode already do nothing, except for run */
+ if (mode == engine->mode && mode != LP5523_CMD_RUN)
+ return 0;
+
+ if (mode == LP5523_CMD_RUN) {
+ ret = lp5523_run_program(engine);
+ } else if (mode == LP5523_CMD_LOAD) {
+ lp5523_set_engine_mode(engine, LP5523_CMD_DISABLED);
+ lp5523_set_engine_mode(engine, LP5523_CMD_LOAD);
+
+ ret = sysfs_create_group(&dev->kobj, engine->attributes);
+ if (ret)
+ return ret;
+ } else if (mode == LP5523_CMD_DISABLED) {
+ lp5523_set_engine_mode(engine, LP5523_CMD_DISABLED);
+ }
+
+ /* remove load attribute from sysfs if not in load mode */
+ if (engine->mode == LP5523_CMD_LOAD && mode != LP5523_CMD_LOAD)
+ sysfs_remove_group(&dev->kobj, engine->attributes);
+
+ engine->mode = mode;
+
+ return ret;
+}
+
+/*--------------------------------------------------------------*/
+/* Probe, Attach, Remove */
+/*--------------------------------------------------------------*/
+static int __init lp5523_init_engine(struct lp5523_engine *engine, int id)
+{
+ if (id < 1 || id > LP5523_ENGINES)
+ return -1;
+ engine->id = id;
+ engine->engine_mask = LP5523_ENG_MASK_BASE >> SHIFT_MASK(id);
+ engine->prog_page = id - 1;
+ engine->mux_page = id + 2;
+ engine->attributes = &lp5523_engine_group[id - 1];
+
+ return 0;
+}
+
+static int __init lp5523_init_led(struct lp5523_led *led, struct device *dev,
+ int chan, struct lp5523_platform_data *pdata)
+{
+ char name[32];
+ int res;
+
+ if (chan >= LP5523_LEDS)
+ return -EINVAL;
+
+ if (pdata->led_config[chan].led_current) {
+ led->led_current = pdata->led_config[chan].led_current;
+ led->max_current = pdata->led_config[chan].max_current;
+ led->chan_nr = pdata->led_config[chan].chan_nr;
+
+ if (led->chan_nr >= LP5523_LEDS) {
+ dev_err(dev, "Use channel numbers between 0 and %d\n",
+ LP5523_LEDS - 1);
+ return -EINVAL;
+ }
+
+ snprintf(name, 32, "lp5523:channel%d", chan);
+
+ led->cdev.name = name;
+ led->cdev.brightness_set = lp5523_set_brightness;
+ res = led_classdev_register(dev, &led->cdev);
+ if (res < 0) {
+ dev_err(dev, "couldn't register led on channel %d\n",
+ chan);
+ return res;
+ }
+ res = sysfs_create_group(&led->cdev.dev->kobj,
+ &lp5523_led_attribute_group);
+ if (res < 0) {
+ dev_err(dev, "couldn't register current attribute\n");
+ led_classdev_unregister(&led->cdev);
+ return res;
+ }
+ } else {
+ led->led_current = 0;
+ }
+ return 0;
+}
+
+static struct i2c_driver lp5523_driver;
+
+static int lp5523_probe(struct i2c_client *client,
+ const struct i2c_device_id *id)
+{
+ struct lp5523_chip *chip;
+ struct lp5523_platform_data *pdata;
+ int ret, i, led;
+
+ chip = kzalloc(sizeof(*chip), GFP_KERNEL);
+ if (!chip)
+ return -ENOMEM;
+
+ i2c_set_clientdata(client, chip);
+ chip->client = client;
+
+ pdata = client->dev.platform_data;
+
+ if (!pdata) {
+ dev_err(&client->dev, "no platform data\n");
+ ret = -EINVAL;
+ goto fail1;
+ }
+
+ mutex_init(&chip->lock);
+
+ chip->pdata = pdata;
+
+ if (pdata->setup_resources) {
+ ret = pdata->setup_resources();
+ if (ret < 0)
+ goto fail1;
+ }
+
+ if (pdata->enable) {
+ pdata->enable(0);
+ usleep_range(1000, 10000);
+ pdata->enable(1);
+ usleep_range(1000, 10000); /* Spec says min 500us */
+ }
+
+ ret = lp5523_detect(client);
+ if (ret)
+ goto fail2;
+
+ dev_info(&client->dev, "LP5523 Programmable led chip found\n");
+
+ /* Initialize engines */
+ for (i = 0; i < ARRAY_SIZE(chip->engines); i++) {
+ ret = lp5523_init_engine(&chip->engines[i], i + 1);
+ if (ret) {
+ dev_err(&client->dev, "error initializing engine\n");
+ goto fail2;
+ }
+ }
+ ret = lp5523_configure(client);
+ if (ret < 0) {
+ dev_err(&client->dev, "error configuring chip\n");
+ goto fail2;
+ }
+
+ /* Initialize leds */
+ chip->num_channels = pdata->num_channels;
+ chip->num_leds = 0;
+ led = 0;
+ for (i = 0; i < pdata->num_channels; i++) {
+ /* Do not initialize channels that are not connected */
+ if (pdata->led_config[i].led_current == 0)
+ continue;
+
+ ret = lp5523_init_led(&chip->leds[led], &client->dev, i, pdata);
+ if (ret) {
+ dev_err(&client->dev, "error initializing leds\n");
+ goto fail3;
+ }
+ chip->num_leds++;
+
+ chip->leds[led].id = led;
+ /* Set LED current */
+ lp5523_write(client,
+ LP5523_REG_LED_CURRENT_BASE + chip->leds[led].chan_nr,
+ chip->leds[led].led_current);
+
+ INIT_WORK(&(chip->leds[led].brightness_work),
+ lp5523_led_brightness_work);
+
+ led++;
+ }
+
+ ret = lp5523_register_sysfs(client);
+ if (ret) {
+ dev_err(&client->dev, "registering sysfs failed\n");
+ goto fail3;
+ }
+ return ret;
+fail3:
+ for (i = 0; i < chip->num_leds; i++) {
+ led_classdev_unregister(&chip->leds[i].cdev);
+ cancel_work_sync(&chip->leds[i].brightness_work);
+ }
+fail2:
+ if (pdata->enable)
+ pdata->enable(0);
+ if (pdata->release_resources)
+ pdata->release_resources();
+fail1:
+ kfree(chip);
+ return ret;
+}
+
+static int lp5523_remove(struct i2c_client *client)
+{
+ struct lp5523_chip *chip = i2c_get_clientdata(client);
+ int i;
+
+ lp5523_unregister_sysfs(client);
+
+ for (i = 0; i < chip->num_leds; i++) {
+ led_classdev_unregister(&chip->leds[i].cdev);
+ cancel_work_sync(&chip->leds[i].brightness_work);
+ }
+
+ if (chip->pdata->enable)
+ chip->pdata->enable(0);
+ if (chip->pdata->release_resources)
+ chip->pdata->release_resources();
+ kfree(chip);
+ return 0;
+}
+
+static const struct i2c_device_id lp5523_id[] = {
+ { "lp5523", 0 },
+ { }
+};
+
+MODULE_DEVICE_TABLE(i2c, lp5523_id);
+
+static struct i2c_driver lp5523_driver = {
+ .driver = {
+ .name = "lp5523",
+ },
+ .probe = lp5523_probe,
+ .remove = lp5523_remove,
+ .id_table = lp5523_id,
+};
+
+static int __init lp5523_init(void)
+{
+ int ret;
+
+ ret = i2c_add_driver(&lp5523_driver);
+
+ if (ret < 0)
+ printk(KERN_ALERT "Adding lp5523 driver failed\n");
+
+ return ret;
+}
+
+static void __exit lp5523_exit(void)
+{
+ i2c_del_driver(&lp5523_driver);
+}
+
+module_init(lp5523_init);
+module_exit(lp5523_exit);
+
+MODULE_AUTHOR("Mathias Nyman <mathias.nyman@nokia.com>");
+MODULE_DESCRIPTION("LP5523 LED engine");
+MODULE_LICENSE("GPL");
}
arch_initcall(soekris_init);
+
+MODULE_LICENSE("GPL");
*/
#include <linux/module.h>
-#include <linux/jiffies.h>
#include <linux/kernel.h>
#include <linux/init.h>
-#include <linux/list.h>
-#include <linux/spinlock.h>
#include <linux/device.h>
-#include <linux/sysdev.h>
-#include <linux/timer.h>
#include <linux/ctype.h>
#include <linux/leds.h>
-#include <linux/slab.h>
#include "leds.h"
-struct timer_trig_data {
- int brightness_on; /* LED brightness during "on" period.
- * (LED_OFF < brightness_on <= LED_FULL)
- */
- unsigned long delay_on; /* milliseconds on */
- unsigned long delay_off; /* milliseconds off */
- struct timer_list timer;
-};
-
-static void led_timer_function(unsigned long data)
-{
- struct led_classdev *led_cdev = (struct led_classdev *) data;
- struct timer_trig_data *timer_data = led_cdev->trigger_data;
- unsigned long brightness;
- unsigned long delay;
-
- if (!timer_data->delay_on || !timer_data->delay_off) {
- led_set_brightness(led_cdev, LED_OFF);
- return;
- }
-
- brightness = led_get_brightness(led_cdev);
- if (!brightness) {
- /* Time to switch the LED on. */
- brightness = timer_data->brightness_on;
- delay = timer_data->delay_on;
- } else {
- /* Store the current brightness value to be able
- * to restore it when the delay_off period is over.
- */
- timer_data->brightness_on = brightness;
- brightness = LED_OFF;
- delay = timer_data->delay_off;
- }
-
- led_set_brightness(led_cdev, brightness);
-
- mod_timer(&timer_data->timer, jiffies + msecs_to_jiffies(delay));
-}
-
static ssize_t led_delay_on_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct led_classdev *led_cdev = dev_get_drvdata(dev);
- struct timer_trig_data *timer_data = led_cdev->trigger_data;
- return sprintf(buf, "%lu\n", timer_data->delay_on);
+ return sprintf(buf, "%lu\n", led_cdev->blink_delay_on);
}
static ssize_t led_delay_on_store(struct device *dev,
struct device_attribute *attr, const char *buf, size_t size)
{
struct led_classdev *led_cdev = dev_get_drvdata(dev);
- struct timer_trig_data *timer_data = led_cdev->trigger_data;
int ret = -EINVAL;
char *after;
unsigned long state = simple_strtoul(buf, &after, 10);
count++;
if (count == size) {
- if (timer_data->delay_on != state) {
- /* the new value differs from the previous */
- timer_data->delay_on = state;
-
- /* deactivate previous settings */
- del_timer_sync(&timer_data->timer);
-
- /* try to activate hardware acceleration, if any */
- if (!led_cdev->blink_set ||
- led_cdev->blink_set(led_cdev,
- &timer_data->delay_on, &timer_data->delay_off)) {
- /* no hardware acceleration, blink via timer */
- mod_timer(&timer_data->timer, jiffies + 1);
- }
- }
+ led_blink_set(led_cdev, &state, &led_cdev->blink_delay_off);
ret = count;
}
struct device_attribute *attr, char *buf)
{
struct led_classdev *led_cdev = dev_get_drvdata(dev);
- struct timer_trig_data *timer_data = led_cdev->trigger_data;
- return sprintf(buf, "%lu\n", timer_data->delay_off);
+ return sprintf(buf, "%lu\n", led_cdev->blink_delay_off);
}
static ssize_t led_delay_off_store(struct device *dev,
struct device_attribute *attr, const char *buf, size_t size)
{
struct led_classdev *led_cdev = dev_get_drvdata(dev);
- struct timer_trig_data *timer_data = led_cdev->trigger_data;
int ret = -EINVAL;
char *after;
unsigned long state = simple_strtoul(buf, &after, 10);
count++;
if (count == size) {
- if (timer_data->delay_off != state) {
- /* the new value differs from the previous */
- timer_data->delay_off = state;
-
- /* deactivate previous settings */
- del_timer_sync(&timer_data->timer);
-
- /* try to activate hardware acceleration, if any */
- if (!led_cdev->blink_set ||
- led_cdev->blink_set(led_cdev,
- &timer_data->delay_on, &timer_data->delay_off)) {
- /* no hardware acceleration, blink via timer */
- mod_timer(&timer_data->timer, jiffies + 1);
- }
- }
+ led_blink_set(led_cdev, &led_cdev->blink_delay_on, &state);
ret = count;
}
static void timer_trig_activate(struct led_classdev *led_cdev)
{
- struct timer_trig_data *timer_data;
int rc;
- timer_data = kzalloc(sizeof(struct timer_trig_data), GFP_KERNEL);
- if (!timer_data)
- return;
-
- timer_data->brightness_on = led_get_brightness(led_cdev);
- if (timer_data->brightness_on == LED_OFF)
- timer_data->brightness_on = led_cdev->max_brightness;
- led_cdev->trigger_data = timer_data;
-
- init_timer(&timer_data->timer);
- timer_data->timer.function = led_timer_function;
- timer_data->timer.data = (unsigned long) led_cdev;
+ led_cdev->trigger_data = NULL;
rc = device_create_file(led_cdev->dev, &dev_attr_delay_on);
if (rc)
- goto err_out;
+ return;
rc = device_create_file(led_cdev->dev, &dev_attr_delay_off);
if (rc)
goto err_out_delayon;
- /* If there is hardware support for blinking, start one
- * user friendly blink rate chosen by the driver.
- */
- if (led_cdev->blink_set)
- led_cdev->blink_set(led_cdev,
- &timer_data->delay_on, &timer_data->delay_off);
+ led_cdev->trigger_data = (void *)1;
return;
err_out_delayon:
device_remove_file(led_cdev->dev, &dev_attr_delay_on);
-err_out:
- led_cdev->trigger_data = NULL;
- kfree(timer_data);
}
static void timer_trig_deactivate(struct led_classdev *led_cdev)
{
- struct timer_trig_data *timer_data = led_cdev->trigger_data;
- unsigned long on = 0, off = 0;
-
- if (timer_data) {
+ if (led_cdev->trigger_data) {
device_remove_file(led_cdev->dev, &dev_attr_delay_on);
device_remove_file(led_cdev->dev, &dev_attr_delay_off);
- del_timer_sync(&timer_data->timer);
- kfree(timer_data);
}
- /* If there is hardware support for blinking, stop it */
- if (led_cdev->blink_set)
- led_cdev->blink_set(led_cdev, &on, &off);
+ /* Stop blinking */
+ led_brightness_set(led_cdev, LED_OFF);
}
static struct led_trigger timer_led_trigger = {
static void adb_iop_complete(struct iop_msg *msg)
{
struct adb_request *req;
- uint flags;
+ unsigned long flags;
local_irq_save(flags);
{
struct adb_iopmsg *amsg = (struct adb_iopmsg *) msg->message;
struct adb_request *req;
- uint flags;
+ unsigned long flags;
#ifdef DEBUG_ADB_IOP
int i;
#endif
ke->len = sizeof(entry->scancode);
memcpy(ke->scancode, &entry->scancode, sizeof(entry->scancode));
+ retval = 0;
+
out:
spin_unlock_irqrestore(&rc_tab->lock, flags);
return retval;
{
struct i2c_client *client = to_i2c_client(dev);
struct als_data *data = i2c_get_clientdata(client);
- unsigned int ret_val;
+ int ret_val;
unsigned long val;
if (strict_strtoul(buf, 10, &val))
{
struct bh1770_chip *chip = dev_get_drvdata(dev);
unsigned long value;
- size_t ret;
+ ssize_t ret;
if (strict_strtoul(buf, 0, &value))
return -EINVAL;
pm_runtime_get_sync(dev);
ret = bh1770_lux_rate(chip, chip->lux_rate_index);
- ret |= bh1770_lux_interrupt_control(chip, BH1770_ENABLE);
+ if (ret < 0) {
+ pm_runtime_put(dev);
+ goto leave;
+ }
+ ret = bh1770_lux_interrupt_control(chip, BH1770_ENABLE);
if (ret < 0) {
pm_runtime_put(dev);
goto leave;
struct device_attribute *attr, const char *buf, size_t count)
{
struct i2c_client *client = to_i2c_client(dev);
- unsigned int ret_val;
+ int ret_val;
unsigned long val;
if (strict_strtoul(buf, 10, &val))
val = 4;
ret_val = i2c_smbus_read_byte_data(client, 0x00);
+ if (ret_val < 0)
+ return ret_val;
ret_val &= 0xFC; /*reset the bit before setting them */
ret_val |= val - 1;
atl1_pcie_patch(adapter);
/* assume we have no link for now */
netif_carrier_off(netdev);
- netif_stop_queue(netdev);
setup_timer(&adapter->phy_config_timer, atl1_phy_config,
(unsigned long)adapter);
* (you will need to reboot afterwards) */
/* #define BNX2X_STOP_ON_ERROR */
-#define DRV_MODULE_VERSION "1.60.00-3"
-#define DRV_MODULE_RELDATE "2010/10/19"
+#define DRV_MODULE_VERSION "1.60.00-4"
+#define DRV_MODULE_RELDATE "2010/11/01"
#define BNX2X_BC_VER 0x040200
#define BNX2X_MULTI_QUEUE
u16 xgxs_config_tx[4]; /* 0x1A0 */
- u32 Reserved1[57]; /* 0x1A8 */
+ u32 Reserved1[56]; /* 0x1A8 */
+ u32 default_cfg; /* 0x288 */
+ /* Enable BAM on KR */
+#define PORT_HW_CFG_ENABLE_BAM_ON_KR_MASK 0x00100000
+#define PORT_HW_CFG_ENABLE_BAM_ON_KR_SHIFT 20
+#define PORT_HW_CFG_ENABLE_BAM_ON_KR_DISABLED 0x00000000
+#define PORT_HW_CFG_ENABLE_BAM_ON_KR_ENABLED 0x00100000
+
u32 speed_capability_mask2; /* 0x28C */
#define PORT_HW_CFG_SPEED_CAPABILITY2_D3_MASK 0x0000FFFF
#define PORT_HW_CFG_SPEED_CAPABILITY2_D3_SHIFT 0
/* reset and unreset the BigMac */
REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR,
(MISC_REGISTERS_RESET_REG_2_RST_BMAC0 << port));
- udelay(10);
+ msleep(1);
REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_SET,
(MISC_REGISTERS_RESET_REG_2_RST_BMAC0 << port));
DP(NETIF_MSG_LINK, "Before rom RX_ALARM(port1): 0x%x\n", tmp1);
/* Enable CL37 BAM */
- bnx2x_cl45_read(bp, phy,
- MDIO_AN_DEVAD,
- MDIO_AN_REG_8073_BAM, &val);
- bnx2x_cl45_write(bp, phy,
- MDIO_AN_DEVAD,
- MDIO_AN_REG_8073_BAM, val | 1);
+ if (REG_RD(bp, params->shmem_base +
+ offsetof(struct shmem_region, dev_info.
+ port_hw_config[params->port].default_cfg)) &
+ PORT_HW_CFG_ENABLE_BAM_ON_KR_ENABLED) {
+ bnx2x_cl45_read(bp, phy,
+ MDIO_AN_DEVAD,
+ MDIO_AN_REG_8073_BAM, &val);
+ bnx2x_cl45_write(bp, phy,
+ MDIO_AN_DEVAD,
+ MDIO_AN_REG_8073_BAM, val | 1);
+ DP(NETIF_MSG_LINK, "Enable CL37 BAM on KR\n");
+ }
if (params->loopback_mode == LOOPBACK_EXT) {
bnx2x_807x_force_10G(bp, phy);
DP(NETIF_MSG_LINK, "Forced speed 10G on 807X\n");
{
struct bnx2x *bp = params->bp;
u16 autoneg_val, an_1000_val, an_10_100_val;
- bnx2x_wait_reset_complete(bp, phy);
+
bnx2x_bits_en(bp, NIG_REG_LATCH_BC_0 + params->port*4,
1 << NIG_LATCH_BC_ENABLE_MI_INT);
/* HW reset */
bnx2x_ext_phy_hw_reset(bp, params->port);
+ bnx2x_wait_reset_complete(bp, phy);
bnx2x_cl45_write(bp, phy, MDIO_PMA_DEVAD, MDIO_PMA_REG_CTRL, 1<<15);
return bnx2x_848xx_cmn_config_init(phy, params, vars);
struct link_vars *vars)
{
struct bnx2x *bp = params->bp;
- u8 port = params->port, initialize = 1;
+ u8 port, initialize = 1;
u16 val;
u16 temp;
u32 actual_phy_selection;
/* This is just for MDIO_CTL_REG_84823_MEDIA register. */
msleep(1);
+ if (CHIP_IS_E2(bp))
+ port = BP_PATH(bp);
+ else
+ port = params->port;
bnx2x_set_gpio(bp, MISC_REGISTERS_GPIO_3,
MISC_REGISTERS_GPIO_OUTPUT_HIGH,
port);
- msleep(200); /* 100 is not enough */
-
+ bnx2x_wait_reset_complete(bp, phy);
+ /* Wait for GPHY to come out of reset */
+ msleep(50);
/* BCM84823 requires that XGXS links up first @ 10G for normal
behavior */
temp = vars->line_speed;
struct link_params *params)
{
struct bnx2x *bp = params->bp;
- u8 port = params->port;
+ u8 port;
+ if (CHIP_IS_E2(bp))
+ port = BP_PATH(bp);
+ else
+ port = params->port;
bnx2x_set_gpio(bp, MISC_REGISTERS_GPIO_3,
MISC_REGISTERS_GPIO_OUTPUT_LOW,
port);
u8 reset_ext_phy)
{
struct bnx2x *bp = params->bp;
- u8 phy_index, port = params->port;
+ u8 phy_index, port = params->port, clear_latch_ind = 0;
DP(NETIF_MSG_LINK, "Resetting the link of port %d\n", port);
/* disable attentions */
vars->link_status = 0;
params->phy[phy_index].link_reset(
¶ms->phy[phy_index],
params);
+ if (params->phy[phy_index].flags &
+ FLAGS_REARM_LATCH_SIGNAL)
+ clear_latch_ind = 1;
}
}
+ if (clear_latch_ind) {
+ /* Clear latching indication */
+ bnx2x_rearm_latch_signal(bp, port, 0);
+ bnx2x_bits_dis(bp, NIG_REG_LATCH_BC_0 + port*4,
+ 1 << NIG_LATCH_BC_ENABLE_MI_INT);
+ }
if (params->phy[INT_PHY].link_reset)
params->phy[INT_PHY].link_reset(
¶ms->phy[INT_PHY], params);
s8 port;
s8 port_of_path = 0;
+ bnx2x_ext_phy_hw_reset(bp, 0);
/* PART1 - Reset both phys */
for (port = PORT_MAX - 1; port >= PORT_0; port--) {
u32 shmem_base, shmem2_base;
return -EINVAL;
}
/* disable attentions */
- bnx2x_bits_dis(bp, NIG_REG_MASK_INTERRUPT_PORT0 + port*4,
+ bnx2x_bits_dis(bp, NIG_REG_MASK_INTERRUPT_PORT0 +
+ port_of_path*4,
(NIG_MASK_XGXS0_LINK_STATUS |
NIG_MASK_XGXS0_LINK10G |
NIG_MASK_SERDES0_LINK_STATUS |
(1<<(MISC_REGISTERS_GPIO_3 + MISC_REGISTERS_GPIO_PORT_SHIFT)));
REG_WR(bp, MISC_REG_GPIO_EVENT_EN, val);
- bnx2x_ext_phy_hw_reset(bp, 1);
+ bnx2x_ext_phy_hw_reset(bp, 0);
msleep(5);
for (port = 0; port < PORT_MAX; port++) {
u32 shmem_base, shmem2_base;
MODULE_AUTHOR("Daniel Martensson<daniel.martensson@stericsson.com>");
MODULE_DESCRIPTION("CAIF SPI driver");
+/* Returns the number of padding bytes for alignment. */
+#define PAD_POW2(x, pow) ((((x)&((pow)-1))==0) ? 0 : (((pow)-((x)&((pow)-1)))))
+
static int spi_loop;
module_param(spi_loop, bool, S_IRUGO);
MODULE_PARM_DESC(spi_loop, "SPI running in loopback mode.");
module_param(spi_frm_align, int, S_IRUGO);
MODULE_PARM_DESC(spi_frm_align, "SPI frame alignment.");
-/* SPI padding options. */
+/*
+ * SPI padding options.
+ * Warning: must be a base of 2 (& operation used) and can not be zero !
+ */
module_param(spi_up_head_align, int, S_IRUGO);
MODULE_PARM_DESC(spi_up_head_align, "SPI uplink head alignment.");
static const struct file_operations dbgfs_state_fops = {
.open = dbgfs_open,
.read = dbgfs_state,
- .owner = THIS_MODULE,
- .llseek = default_llseek,
+ .owner = THIS_MODULE
};
static const struct file_operations dbgfs_frame_fops = {
.open = dbgfs_open,
.read = dbgfs_frame,
- .owner = THIS_MODULE,
- .llseek = default_llseek,
+ .owner = THIS_MODULE
};
static inline void dev_debugfs_add(struct cfspi *cfspi)
u8 *dst = buf;
caif_assert(buf);
+ if (cfspi->slave && !cfspi->slave_talked)
+ cfspi->slave_talked = true;
+
do {
struct sk_buff *skb;
struct caif_payload_info *info;
* Compute head offset i.e. number of bytes to add to
* get the start of the payload aligned.
*/
- if (spi_up_head_align) {
- spad = 1 + ((info->hdr_len + 1) & spi_up_head_align);
+ if (spi_up_head_align > 1) {
+ spad = 1 + PAD_POW2((info->hdr_len + 1), spi_up_head_align);
*dst = (u8)(spad - 1);
dst += spad;
}
* Compute tail offset i.e. number of bytes to add to
* get the complete CAIF frame aligned.
*/
- epad = (skb->len + spad) & spi_up_tail_align;
+ epad = PAD_POW2((skb->len + spad), spi_up_tail_align);
dst += epad;
dev_kfree_skb(skb);
* Compute head offset i.e. number of bytes to add to
* get the start of the payload aligned.
*/
- if (spi_up_head_align)
- spad = 1 + ((info->hdr_len + 1) & spi_up_head_align);
+ if (spi_up_head_align > 1)
+ spad = 1 + PAD_POW2((info->hdr_len + 1), spi_up_head_align);
/*
* Compute tail offset i.e. number of bytes to add to
* get the complete CAIF frame aligned.
*/
- epad = (skb->len + spad) & spi_up_tail_align;
+ epad = PAD_POW2((skb->len + spad), spi_up_tail_align);
if ((skb->len + spad + epad + frm_len) <= CAIF_MAX_SPI_FRAME) {
skb_queue_tail(&cfspi->chead, skb);
} else {
/* Put back packet. */
skb_queue_head(&cfspi->qhead, skb);
+ break;
}
} while (pkts <= CAIF_MAX_SPI_PKTS);
{
struct cfspi *cfspi = (struct cfspi *)ifc->priv;
+ /*
+ * The slave device is the master on the link. Interrupts before the
+ * slave has transmitted are considered spurious.
+ */
+ if (cfspi->slave && !cfspi->slave_talked) {
+ printk(KERN_WARNING "CFSPI: Spurious SS interrupt.\n");
+ return;
+ }
+
if (!in_interrupt())
spin_lock(&cfspi->lock);
if (assert) {
spin_unlock(&cfspi->lock);
/* Wake up the xfer thread. */
- wake_up_interruptible(&cfspi->wait);
+ if (assert)
+ wake_up_interruptible(&cfspi->wait);
}
static void cfspi_xfer_done_cb(struct cfspi_ifc *ifc)
* Compute head offset i.e. number of bytes added to
* get the start of the payload aligned.
*/
- if (spi_down_head_align) {
+ if (spi_down_head_align > 1) {
spad = 1 + *src;
src += spad;
}
* Compute tail offset i.e. number of bytes added to
* get the complete CAIF frame aligned.
*/
- epad = (pkt_len + spad) & spi_down_tail_align;
+ epad = PAD_POW2((pkt_len + spad), spi_down_tail_align);
src += epad;
} while ((src - buf) < len);
cfspi->ndev = ndev;
cfspi->pdev = pdev;
- /* Set flow info */
+ /* Set flow info. */
cfspi->flow_off_sent = 0;
cfspi->qd_low_mark = LOW_WATER_MARK;
cfspi->qd_high_mark = HIGH_WATER_MARK;
+ /* Set slave info. */
+ if (!strncmp(cfspi_spi_driver.driver.name, "cfspi_sspi", 10)) {
+ cfspi->slave = true;
+ cfspi->slave_talked = false;
+ } else {
+ cfspi->slave = false;
+ cfspi->slave_talked = false;
+ }
+
/* Assign the SPI device. */
cfspi->dev = dev;
/* Assign the device ifc to this SPI interface. */
#endif
int spi_frm_align = 2;
-int spi_up_head_align = 1;
-int spi_up_tail_align;
-int spi_down_head_align = 3;
-int spi_down_tail_align = 1;
+
+/*
+ * SPI padding options.
+ * Warning: must be a base of 2 (& operation used) and can not be zero !
+ */
+int spi_up_head_align = 1 << 1;
+int spi_up_tail_align = 1 << 0;
+int spi_down_head_align = 1 << 2;
+int spi_down_tail_align = 1 << 1;
#ifdef CONFIG_DEBUG_FS
static inline void debugfs_store_prev(struct cfspi *cfspi)
adapter->name = adapter->port[i]->name;
__set_bit(i, &adapter->registered_device_map);
- netif_tx_stop_all_queues(adapter->port[i]);
}
}
if (!adapter->registered_device_map) {
__set_bit(i, &adapter->registered_device_map);
adapter->chan_map[adap2pinfo(adapter, i)->tx_chan] = i;
- netif_tx_stop_all_queues(adapter->port[i]);
}
}
if (!adapter->registered_device_map) {
pi->xact_addr_filt = -1;
pi->rx_offload = RX_CSO;
netif_carrier_off(netdev);
- netif_tx_stop_all_queues(netdev);
netdev->irq = pdev->irq;
netdev->features = (NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 |
SET_ETHTOOL_OPS(ndev, &emac_ethtool_ops);
netif_carrier_off(ndev);
- netif_stop_queue(ndev);
err = register_netdev(ndev);
if (err) {
* Tell stack that we are not ready to work until open()
*/
netif_carrier_off(netdev);
- netif_stop_queue(netdev);
- /*
- * Register netdev
- */
rc = register_netdev(netdev);
if (rc) {
pr_err("Cannot register net device\n");
MODULE_DESCRIPTION("QLogic/NetXen (1/10) GbE Converged Ethernet Driver");
MODULE_LICENSE("GPL");
MODULE_VERSION(NETXEN_NIC_LINUX_VERSIONID);
-MODULE_FIRMWARE(NX_P2_MN_ROMIMAGE_NAME);
-MODULE_FIRMWARE(NX_P3_CT_ROMIMAGE_NAME);
-MODULE_FIRMWARE(NX_P3_MN_ROMIMAGE_NAME);
MODULE_FIRMWARE(NX_UNIFIED_ROMIMAGE_NAME);
char netxen_nic_driver_name[] = "netxen_nic";
netdev->irq = adapter->msix_entries[0].vector;
netif_carrier_off(netdev);
- netif_stop_queue(netdev);
err = register_netdev(netdev);
if (err) {
#define __SMSC911X_H__
#define TX_FIFO_LOW_THRESHOLD ((u32)1600)
-#define SMSC911X_EEPROM_SIZE ((u32)7)
+#define SMSC911X_EEPROM_SIZE ((u32)128)
#define USE_DEBUG 0
/* This is the maximum number of packets to be received every
de->media_timer.data = (unsigned long) de;
netif_carrier_off(dev);
- netif_stop_queue(dev);
/* wake up device, assign resources */
rc = pci_enable_device(pdev);
#include <linux/usb/usbnet.h>
#include <linux/slab.h>
#include <linux/kernel.h>
+#include <linux/pm_runtime.h>
#define DRIVER_VERSION "22-Aug-2005"
struct usb_device *xdev;
int status;
const char *name;
+ struct usb_driver *driver = to_usb_driver(udev->dev.driver);
+
+ /* usbnet already took usb runtime pm, so have to enable the feature
+ * for usb interface, otherwise usb_autopm_get_interface may return
+ * failure if USB_SUSPEND(RUNTIME_PM) is enabled.
+ */
+ if (!driver->supports_autosuspend) {
+ driver->supports_autosuspend = 1;
+ pm_runtime_enable(&udev->dev);
+ }
name = udev->dev.driver->name;
info = (struct driver_info *) prod->driver_info;
#define DRV_DESCRIPTION "802.11 data/management/control stack"
#define DRV_NAME "libipw"
+#define DRV_PROCNAME "ieee80211"
#define DRV_VERSION LIBIPW_VERSION
#define DRV_COPYRIGHT "Copyright (C) 2004-2005 Intel Corporation <jketreno@linux.intel.com>"
struct proc_dir_entry *e;
libipw_debug_level = debug;
- libipw_proc = proc_mkdir("ieee80211", init_net.proc_net);
+ libipw_proc = proc_mkdir(DRV_PROCNAME, init_net.proc_net);
if (libipw_proc == NULL) {
- LIBIPW_ERROR("Unable to create " DRV_NAME
+ LIBIPW_ERROR("Unable to create " DRV_PROCNAME
" proc directory\n");
return -EIO;
}
e = proc_create("debug_level", S_IRUGO | S_IWUSR, libipw_proc,
&debug_level_proc_fops);
if (!e) {
- remove_proc_entry(DRV_NAME, init_net.proc_net);
+ remove_proc_entry(DRV_PROCNAME, init_net.proc_net);
libipw_proc = NULL;
return -EIO;
}
#ifdef CONFIG_LIBIPW_DEBUG
if (libipw_proc) {
remove_proc_entry("debug_level", libipw_proc);
- remove_proc_entry(DRV_NAME, init_net.proc_net);
+ remove_proc_entry(DRV_PROCNAME, init_net.proc_net);
libipw_proc = NULL;
}
#endif /* CONFIG_LIBIPW_DEBUG */
boolean
default y if (RT2X00_LIB=y && LEDS_CLASS=y) || (RT2X00_LIB=m && LEDS_CLASS!=n)
-comment "rt2x00 leds support disabled due to modularized LEDS_CLASS and built-in rt2x00"
- depends on RT2X00_LIB=y && LEDS_CLASS=m
-
config RT2X00_LIB_DEBUGFS
bool "Ralink debugfs support"
depends on RT2X00_LIB && MAC80211_DEBUGFS
list_for_each_entry(port, &rio_mports, node) {
if (!request_mem_region(port->iores.start,
- port->iores.end - port->iores.start,
+ resource_size(&port->iores),
port->name)) {
printk(KERN_ERR
"RIO: Error requesting master port region 0x%016llx-0x%016llx\n",
- (u64)port->iores.start, (u64)port->iores.end - 1);
+ (u64)port->iores.start, (u64)port->iores.end);
rc = -ENOMEM;
goto out;
}
#ifdef CONFIG_SH_SECUREEDGE5410
#include <asm/rtc.h>
-#include <mach/snapgear.h>
+#include <mach/secureedge5410.h>
#define RTC_RESET 0x1000
#define RTC_IODATA 0x0800
if (unlikely(rsp_flags & FCP_SNS_LEN_VAL)) {
sense = (char *) &fcp_rsp[1];
if (rsp_flags & FCP_RSP_LEN_VAL)
- sense += fcp_rsp->ext.fr_sns_len;
+ sense += fcp_rsp->ext.fr_rsp_len;
sense_len = min(fcp_rsp->ext.fr_sns_len,
(u32) SCSI_SENSE_BUFFERSIZE);
memcpy(scsi->sense_buffer, sense, sense_len);
fc_host_port_type(shost) = FC_PORTTYPE_UNKNOWN;
adapter->hydra_version = 0;
- atomic_set_mask(ZFCP_STATUS_ADAPTER_XCONFIG_OK,
- &adapter->status);
-
zfcp_fsf_link_down_info_eval(req,
&qtcb->header.fsf_status_qual.link_down_info);
break;
return -ENOMEM;
}
+ get_device(&port->dev);
+
if (device_register(&unit->dev)) {
put_device(&unit->dev);
return -ENOMEM;
return -EINVAL;
}
- get_device(&port->dev);
-
write_lock_irq(&port->unit_list_lock);
list_add_tail(&unit->list, &port->unit_list);
write_unlock_irq(&port->unit_list_lock);
typedef void (*bfa_isr_func_t) (struct bfa_s *bfa, struct bfi_msg_s *m);
typedef void (*bfa_cb_cbfn_t) (void *cbarg, bfa_boolean_t complete);
-/**
+/*
* Interrupt message handlers
*/
void bfa_isr_unhandled(struct bfa_s *bfa, struct bfi_msg_s *m);
void bfa_isr_bind(enum bfi_mclass mc, bfa_isr_func_t isr_func);
-/**
+/*
* Request and response queue related defines
*/
#define BFA_REQQ_NELEMS_MIN (4)
#define bfa_reqq_produce(__bfa, __reqq) do { \
(__bfa)->iocfc.req_cq_pi[__reqq]++; \
(__bfa)->iocfc.req_cq_pi[__reqq] &= \
- ((__bfa)->iocfc.cfg.drvcfg.num_reqq_elems - 1); \
- bfa_reg_write((__bfa)->iocfc.bfa_regs.cpe_q_pi[__reqq], \
- (__bfa)->iocfc.req_cq_pi[__reqq]); \
+ ((__bfa)->iocfc.cfg.drvcfg.num_reqq_elems - 1); \
+ writel((__bfa)->iocfc.req_cq_pi[__reqq], \
+ (__bfa)->iocfc.bfa_regs.cpe_q_pi[__reqq]); \
mmiowb(); \
} while (0)
(__index) &= ((__size) - 1); \
} while (0)
-/**
+/*
* Queue element to wait for room in request queue. FIFO order is
* maintained when fullfilling requests.
*/
void *cbarg;
};
-/**
+/*
* Circular queue usage assignments
*/
enum {
#define bfa_reqq(__bfa, __reqq) (&(__bfa)->reqq_waitq[__reqq])
-/**
+/*
* static inline void
* bfa_reqq_wait(struct bfa_s *bfa, int reqq, struct bfa_reqq_wait_s *wqe)
*/
#define bfa_reqq_wcancel(__wqe) list_del(&(__wqe)->qe)
-/**
+/*
* Generic BFA callback element.
*/
struct bfa_cb_qe_s {
} while (0)
-/**
+/*
* PCI devices supported by the current BFA
*/
struct bfa_pciid_s {
extern char bfa_version[];
-/**
+/*
* BFA memory resources
*/
enum bfa_mem_type {
((_m)->meminfo[BFA_MEM_TYPE_DMA - 1].dma_curp)
struct bfa_iocfc_regs_s {
- bfa_os_addr_t intr_status;
- bfa_os_addr_t intr_mask;
- bfa_os_addr_t cpe_q_pi[BFI_IOC_MAX_CQS];
- bfa_os_addr_t cpe_q_ci[BFI_IOC_MAX_CQS];
- bfa_os_addr_t cpe_q_depth[BFI_IOC_MAX_CQS];
- bfa_os_addr_t cpe_q_ctrl[BFI_IOC_MAX_CQS];
- bfa_os_addr_t rme_q_ci[BFI_IOC_MAX_CQS];
- bfa_os_addr_t rme_q_pi[BFI_IOC_MAX_CQS];
- bfa_os_addr_t rme_q_depth[BFI_IOC_MAX_CQS];
- bfa_os_addr_t rme_q_ctrl[BFI_IOC_MAX_CQS];
+ void __iomem *intr_status;
+ void __iomem *intr_mask;
+ void __iomem *cpe_q_pi[BFI_IOC_MAX_CQS];
+ void __iomem *cpe_q_ci[BFI_IOC_MAX_CQS];
+ void __iomem *cpe_q_depth[BFI_IOC_MAX_CQS];
+ void __iomem *cpe_q_ctrl[BFI_IOC_MAX_CQS];
+ void __iomem *rme_q_ci[BFI_IOC_MAX_CQS];
+ void __iomem *rme_q_pi[BFI_IOC_MAX_CQS];
+ void __iomem *rme_q_depth[BFI_IOC_MAX_CQS];
+ void __iomem *rme_q_ctrl[BFI_IOC_MAX_CQS];
};
-/**
+/*
* MSIX vector handlers
*/
#define BFA_MSIX_MAX_VECTORS 22
bfa_msix_handler_t handler[BFA_MSIX_MAX_VECTORS];
};
-/**
+/*
* Chip specific interfaces
*/
struct bfa_hwif_s {
struct bfi_pbc_vport_s *pbc_vport);
-/**
+/*
*----------------------------------------------------------------------
* BFA public interfaces
*----------------------------------------------------------------------
} lun;
lun.bfa_lun = 0;
- lun.scsi_lun[0] = bfa_os_htons(luno);
+ lun.scsi_lun[0] = cpu_to_be16(luno);
return lun.bfa_lun;
}
-/**
+/*
* Get LUN for the I/O request
*/
#define bfa_cb_ioim_get_lun(__dio) \
bfad_int_to_lun(((struct scsi_cmnd *)__dio)->device->lun)
-/**
+/*
* Get CDB for the I/O request
*/
static inline u8 *
return (u8 *) cmnd->cmnd;
}
-/**
+/*
* Get I/O direction (read/write) for the I/O request
*/
static inline enum fcp_iodir
return FCP_IODIR_NONE;
}
-/**
+/*
* Get IO size in bytes for the I/O request
*/
static inline u32
return scsi_bufflen(cmnd);
}
-/**
+/*
* Get timeout for the I/O request
*/
static inline u8
return 0;
}
-/**
+/*
* Get Command Reference Number for the I/O request. 0 if none.
*/
static inline u8
return 0;
}
-/**
+/*
* Get SAM-3 priority for the I/O request. 0 is default.
*/
static inline u8
return 0;
}
-/**
+/*
* Get task attributes for the I/O request. Default is FCP_TASK_ATTR_SIMPLE(0).
*/
static inline u8
return task_attr;
}
-/**
+/*
* Get CDB length in bytes for the I/O request. Default is FCP_CMND_CDB_LEN(16).
*/
static inline u8
return cmnd->cmd_len;
}
-/**
+/*
* Assign queue to be used for the I/O request. This value depends on whether
* the driver wants to use the queues via any specific algorithm. Currently,
* this is not supported.
BFA_TRC_FILE(HAL, CORE);
-/**
+/*
* BFA IOC FC related definitions
*/
-/**
+/*
* IOC local definitions
*/
#define BFA_IOCFC_TOV 5000 /* msecs */
#define DEF_CFG_NUM_SBOOT_TGTS 16
#define DEF_CFG_NUM_SBOOT_LUNS 16
-/**
+/*
* forward declaration for IOC FC functions
*/
static void bfa_iocfc_enable_cbfn(void *bfa_arg, enum bfa_status status);
static void bfa_iocfc_reset_cbfn(void *bfa_arg);
static struct bfa_ioc_cbfn_s bfa_iocfc_cbfn;
-/**
+/*
* BFA Interrupt handling functions
*/
static void
waitq = bfa_reqq(bfa, qid);
list_for_each_safe(qe, qen, waitq) {
- /**
+ /*
* Callback only as long as there is room in request queue
*/
if (bfa_reqq_full(bfa, qid))
bfa_intx(bfa);
}
-/**
+/*
* hal_intr_api
*/
bfa_boolean_t
u32 intr, qintr;
int queue;
- intr = bfa_reg_read(bfa->iocfc.bfa_regs.intr_status);
+ intr = readl(bfa->iocfc.bfa_regs.intr_status);
if (!intr)
return BFA_FALSE;
- /**
+ /*
* RME completion queue interrupt
*/
qintr = intr & __HFN_INT_RME_MASK;
- bfa_reg_write(bfa->iocfc.bfa_regs.intr_status, qintr);
+ writel(qintr, bfa->iocfc.bfa_regs.intr_status);
for (queue = 0; queue < BFI_IOC_MAX_CQS_ASIC; queue++) {
if (intr & (__HFN_INT_RME_Q0 << queue))
if (!intr)
return BFA_TRUE;
- /**
+ /*
* CPE completion queue interrupt
*/
qintr = intr & __HFN_INT_CPE_MASK;
- bfa_reg_write(bfa->iocfc.bfa_regs.intr_status, qintr);
+ writel(qintr, bfa->iocfc.bfa_regs.intr_status);
for (queue = 0; queue < BFI_IOC_MAX_CQS_ASIC; queue++) {
if (intr & (__HFN_INT_CPE_Q0 << queue))
void
bfa_intx_enable(struct bfa_s *bfa)
{
- bfa_reg_write(bfa->iocfc.bfa_regs.intr_mask, bfa->iocfc.intr_mask);
+ writel(bfa->iocfc.intr_mask, bfa->iocfc.bfa_regs.intr_mask);
}
void
bfa_intx_disable(struct bfa_s *bfa)
{
- bfa_reg_write(bfa->iocfc.bfa_regs.intr_mask, -1L);
+ writel(-1L, bfa->iocfc.bfa_regs.intr_mask);
}
void
__HFN_INT_RME_Q6 | __HFN_INT_RME_Q7 |
__HFN_INT_MBOX_LPU1);
- bfa_reg_write(bfa->iocfc.bfa_regs.intr_status, intr_unmask);
- bfa_reg_write(bfa->iocfc.bfa_regs.intr_mask, ~intr_unmask);
+ writel(intr_unmask, bfa->iocfc.bfa_regs.intr_status);
+ writel(~intr_unmask, bfa->iocfc.bfa_regs.intr_mask);
bfa->iocfc.intr_mask = ~intr_unmask;
bfa_isr_mode_set(bfa, bfa->msix.nvecs != 0);
}
bfa_isr_disable(struct bfa_s *bfa)
{
bfa_isr_mode_set(bfa, BFA_FALSE);
- bfa_reg_write(bfa->iocfc.bfa_regs.intr_mask, -1L);
+ writel(-1L, bfa->iocfc.bfa_regs.intr_mask);
bfa_msix_uninstall(bfa);
}
bfa->iocfc.hwif.hw_reqq_ack(bfa, qid);
- /**
+ /*
* Resume any pending requests in the corresponding reqq.
*/
waitq = bfa_reqq(bfa, qid);
}
}
- /**
+ /*
* update CI
*/
bfa_rspq_ci(bfa, qid) = pi;
- bfa_reg_write(bfa->iocfc.bfa_regs.rme_q_ci[qid], pi);
+ writel(pi, bfa->iocfc.bfa_regs.rme_q_ci[qid]);
mmiowb();
- /**
+ /*
* Resume any pending requests in the corresponding reqq.
*/
waitq = bfa_reqq(bfa, qid);
{
u32 intr, curr_value;
- intr = bfa_reg_read(bfa->iocfc.bfa_regs.intr_status);
+ intr = readl(bfa->iocfc.bfa_regs.intr_status);
if (intr & (__HFN_INT_MBOX_LPU0 | __HFN_INT_MBOX_LPU1))
bfa_msix_lpu(bfa);
if (intr) {
if (intr & __HFN_INT_LL_HALT) {
- /**
+ /*
* If LL_HALT bit is set then FW Init Halt LL Port
* Register needs to be cleared as well so Interrupt
* Status Register will be cleared.
*/
- curr_value = bfa_reg_read(bfa->ioc.ioc_regs.ll_halt);
+ curr_value = readl(bfa->ioc.ioc_regs.ll_halt);
curr_value &= ~__FW_INIT_HALT_P;
- bfa_reg_write(bfa->ioc.ioc_regs.ll_halt, curr_value);
+ writel(curr_value, bfa->ioc.ioc_regs.ll_halt);
}
if (intr & __HFN_INT_ERR_PSS) {
- /**
+ /*
* ERR_PSS bit needs to be cleared as well in case
* interrups are shared so driver's interrupt handler is
* still called eventhough it is already masked out.
*/
- curr_value = bfa_reg_read(
+ curr_value = readl(
bfa->ioc.ioc_regs.pss_err_status_reg);
curr_value &= __PSS_ERR_STATUS_SET;
- bfa_reg_write(bfa->ioc.ioc_regs.pss_err_status_reg,
- curr_value);
+ writel(curr_value,
+ bfa->ioc.ioc_regs.pss_err_status_reg);
}
- bfa_reg_write(bfa->iocfc.bfa_regs.intr_status, intr);
+ writel(intr, bfa->iocfc.bfa_regs.intr_status);
bfa_msix_errint(bfa, intr);
}
}
bfa_isrs[mc] = isr_func;
}
-/**
+/*
* BFA IOC FC related functions
*/
-/**
+/*
* hal_ioc_pvt BFA IOC private functions
*/
BFA_CACHELINE_SZ);
}
-/**
+/*
* Use the Mailbox interface to send BFI_IOCFC_H2I_CFG_REQ
*/
static void
bfa_iocfc_reset_queues(bfa);
- /**
+ /*
* initialize IOC configuration info
*/
cfg_info->endian_sig = BFI_IOC_ENDIAN_SIG;
cfg_info->num_cqs = cfg->fwcfg.num_cqs;
bfa_dma_be_addr_set(cfg_info->cfgrsp_addr, iocfc->cfgrsp_dma.pa);
- /**
+ /*
* dma map REQ and RSP circular queues and shadow pointers
*/
for (i = 0; i < cfg->fwcfg.num_cqs; i++) {
bfa_dma_be_addr_set(cfg_info->req_shadow_ci[i],
iocfc->req_cq_shadow_ci[i].pa);
cfg_info->req_cq_elems[i] =
- bfa_os_htons(cfg->drvcfg.num_reqq_elems);
+ cpu_to_be16(cfg->drvcfg.num_reqq_elems);
bfa_dma_be_addr_set(cfg_info->rsp_cq_ba[i],
iocfc->rsp_cq_ba[i].pa);
bfa_dma_be_addr_set(cfg_info->rsp_shadow_pi[i],
iocfc->rsp_cq_shadow_pi[i].pa);
cfg_info->rsp_cq_elems[i] =
- bfa_os_htons(cfg->drvcfg.num_rspq_elems);
+ cpu_to_be16(cfg->drvcfg.num_rspq_elems);
}
- /**
+ /*
* Enable interrupt coalescing if it is driver init path
* and not ioc disable/enable path.
*/
iocfc->cfgdone = BFA_FALSE;
- /**
+ /*
* dma map IOC configuration itself
*/
bfi_h2i_set(cfg_req.mh, BFI_MC_IOCFC, BFI_IOCFC_H2I_CFG_REQ,
iocfc->bfa = bfa;
iocfc->action = BFA_IOCFC_ACT_NONE;
- bfa_os_assign(iocfc->cfg, *cfg);
+ iocfc->cfg = *cfg;
- /**
+ /*
* Initialize chip specific handlers.
*/
if (bfa_asic_id_ct(bfa_ioc_devid(&bfa->ioc))) {
for (i = 0; i < cfg->fwcfg.num_cqs; i++) {
iocfc->req_cq_ba[i].kva = dm_kva;
iocfc->req_cq_ba[i].pa = dm_pa;
- bfa_os_memset(dm_kva, 0, per_reqq_sz);
+ memset(dm_kva, 0, per_reqq_sz);
dm_kva += per_reqq_sz;
dm_pa += per_reqq_sz;
iocfc->rsp_cq_ba[i].kva = dm_kva;
iocfc->rsp_cq_ba[i].pa = dm_pa;
- bfa_os_memset(dm_kva, 0, per_rspq_sz);
+ memset(dm_kva, 0, per_rspq_sz);
dm_kva += per_rspq_sz;
dm_pa += per_rspq_sz;
}
}
}
-/**
+/*
* Start BFA submodules.
*/
static void
hal_mods[i]->start(bfa);
}
-/**
+/*
* Disable BFA submodules.
*/
static void
complete(&bfad->disable_comp);
}
-/**
+/*
* Update BFA configuration from firmware configuration.
*/
static void
struct bfa_iocfc_fwcfg_s *fwcfg = &cfgrsp->fwcfg;
fwcfg->num_cqs = fwcfg->num_cqs;
- fwcfg->num_ioim_reqs = bfa_os_ntohs(fwcfg->num_ioim_reqs);
- fwcfg->num_tskim_reqs = bfa_os_ntohs(fwcfg->num_tskim_reqs);
- fwcfg->num_fcxp_reqs = bfa_os_ntohs(fwcfg->num_fcxp_reqs);
- fwcfg->num_uf_bufs = bfa_os_ntohs(fwcfg->num_uf_bufs);
- fwcfg->num_rports = bfa_os_ntohs(fwcfg->num_rports);
+ fwcfg->num_ioim_reqs = be16_to_cpu(fwcfg->num_ioim_reqs);
+ fwcfg->num_tskim_reqs = be16_to_cpu(fwcfg->num_tskim_reqs);
+ fwcfg->num_fcxp_reqs = be16_to_cpu(fwcfg->num_fcxp_reqs);
+ fwcfg->num_uf_bufs = be16_to_cpu(fwcfg->num_uf_bufs);
+ fwcfg->num_rports = be16_to_cpu(fwcfg->num_rports);
iocfc->cfgdone = BFA_TRUE;
- /**
+ /*
* Configuration is complete - initialize/start submodules
*/
bfa_fcport_init(bfa);
}
}
-/**
+/*
* IOC enable request is complete
*/
static void
bfa_iocfc_send_cfg(bfa);
}
-/**
+/*
* IOC disable request is complete
*/
static void
}
}
-/**
+/*
* Notify sub-modules of hardware failure.
*/
static void
bfa);
}
-/**
+/*
* Actions on chip-reset completion.
*/
static void
bfa_isr_enable(bfa);
}
-/**
+/*
* hal_ioc_public
*/
-/**
+/*
* Query IOC memory requirement information.
*/
void
*km_len += bfa_ioc_debug_trcsz(bfa_auto_recover);
}
-/**
+/*
* Query IOC memory requirement information.
*/
void
ioc->trcmod = bfa->trcmod;
bfa_ioc_attach(&bfa->ioc, bfa, &bfa_iocfc_cbfn, &bfa->timer_mod);
- /**
+ /*
* Set FC mode for BFA_PCI_DEVICE_ID_CT_FC.
*/
if (pcidev->device_id == BFA_PCI_DEVICE_ID_CT_FC)
INIT_LIST_HEAD(&bfa->reqq_waitq[i]);
}
-/**
+/*
* Query IOC memory requirement information.
*/
void
bfa_ioc_detach(&bfa->ioc);
}
-/**
+/*
* Query IOC memory requirement information.
*/
void
bfa_ioc_enable(&bfa->ioc);
}
-/**
+/*
* IOC start called from bfa_start(). Called to start IOC operations
* at driver instantiation for this instance.
*/
bfa_iocfc_start_submod(bfa);
}
-/**
+/*
* IOC stop called from bfa_stop(). Called only when driver is unloaded
* for this instance.
*/
attr->intr_attr.coalesce = iocfc->cfginfo->intr_attr.coalesce;
attr->intr_attr.delay = iocfc->cfginfo->intr_attr.delay ?
- bfa_os_ntohs(iocfc->cfginfo->intr_attr.delay) :
- bfa_os_ntohs(iocfc->cfgrsp->intr_attr.delay);
+ be16_to_cpu(iocfc->cfginfo->intr_attr.delay) :
+ be16_to_cpu(iocfc->cfgrsp->intr_attr.delay);
attr->intr_attr.latency = iocfc->cfginfo->intr_attr.latency ?
- bfa_os_ntohs(iocfc->cfginfo->intr_attr.latency) :
- bfa_os_ntohs(iocfc->cfgrsp->intr_attr.latency);
+ be16_to_cpu(iocfc->cfginfo->intr_attr.latency) :
+ be16_to_cpu(iocfc->cfgrsp->intr_attr.latency);
attr->config = iocfc->cfg;
}
struct bfi_iocfc_set_intr_req_s *m;
iocfc->cfginfo->intr_attr.coalesce = attr->coalesce;
- iocfc->cfginfo->intr_attr.delay = bfa_os_htons(attr->delay);
- iocfc->cfginfo->intr_attr.latency = bfa_os_htons(attr->latency);
+ iocfc->cfginfo->intr_attr.delay = cpu_to_be16(attr->delay);
+ iocfc->cfginfo->intr_attr.latency = cpu_to_be16(attr->latency);
if (!bfa_iocfc_is_operational(bfa))
return BFA_STATUS_OK;
iocfc->cfginfo->sense_buf_len = (BFI_IOIM_SNSLEN - 1);
bfa_dma_be_addr_set(iocfc->cfginfo->ioim_snsbase, snsbase_pa);
}
-/**
+/*
* Enable IOC after it is disabled.
*/
void
return bfa_ioc_is_operational(&bfa->ioc) && bfa->iocfc.cfgdone;
}
-/**
+/*
* Return boot target port wwns -- read from boot information in flash.
*/
void
return cfgrsp->pbc_cfg.nvports;
}
-/**
+/*
* hal_api
*/
-/**
+/*
* Use this function query the memory requirement of the BFA library.
* This function needs to be called before bfa_attach() to get the
* memory required of the BFA layer for a given driver configuration.
bfa_assert((cfg != NULL) && (meminfo != NULL));
- bfa_os_memset((void *)meminfo, 0, sizeof(struct bfa_meminfo_s));
+ memset((void *)meminfo, 0, sizeof(struct bfa_meminfo_s));
meminfo->meminfo[BFA_MEM_TYPE_KVA - 1].mem_type =
BFA_MEM_TYPE_KVA;
meminfo->meminfo[BFA_MEM_TYPE_DMA - 1].mem_type =
meminfo->meminfo[BFA_MEM_TYPE_DMA - 1].mem_len = dm_len;
}
-/**
+/*
* Use this function to do attach the driver instance with the BFA
* library. This function will not trigger any HW initialization
* process (which will be done in bfa_init() call)
bfa_assert((cfg != NULL) && (meminfo != NULL));
- /**
+ /*
* initialize all memory pointers for iterative allocation
*/
for (i = 0; i < BFA_MEM_TYPE_MAX; i++) {
bfa_com_port_attach(bfa, meminfo);
}
-/**
+/*
* Use this function to delete a BFA IOC. IOC should be stopped (by
* calling bfa_stop()) before this function call.
*
bfa->plog = plog;
}
-/**
+/*
* Initialize IOC.
*
* This function will return immediately, when the IOC initialization is
bfa_iocfc_init(bfa);
}
-/**
+/*
* Use this function initiate the IOC configuration setup. This function
* will return immediately.
*
bfa_iocfc_start(bfa);
}
-/**
+/*
* Use this function quiese the IOC. This function will return immediately,
* when the IOC is actually stopped, the bfad->comp will be set.
*
bfa->fcs = BFA_TRUE;
}
-/**
+/*
* Periodic timer heart beat from driver
*/
void
bfa_timer_beat(&bfa->timer_mod);
}
-/**
+/*
* Return the list of PCI vendor/device id lists supported by this
* BFA instance.
*/
*pciids = __pciids;
}
-/**
+/*
* Use this function query the default struct bfa_iocfc_cfg_s value (compiled
* into BFA layer). The OS driver can then turn back and overwrite entries that
* have been configured by the user.
bfa_ioc_get_attr(&bfa->ioc, ioc_attr);
}
-/**
+/*
* Retrieve firmware trace information on IOC failure.
*/
bfa_status_t
return bfa_ioc_debug_fwsave(&bfa->ioc, trcdata, trclen);
}
-/**
+/*
* Clear the saved firmware trace information of an IOC.
*/
void
bfa_ioc_debug_fwsave_clear(&bfa->ioc);
}
-/**
+/*
* Fetch firmware trace data.
*
* @param[in] bfa BFA instance
return bfa_ioc_debug_fwtrc(&bfa->ioc, trcdata, trclen);
}
-/**
+/*
* Dump firmware memory.
*
* @param[in] bfa BFA instance
{
return bfa_ioc_debug_fwcore(&bfa->ioc, buf, offset, buflen);
}
-/**
+/*
* Reset hw semaphore & usage cnt regs and initialize.
*/
void
bfa_ioc_pll_init(&bfa->ioc);
}
-/**
+/*
* Fetch firmware statistics data.
*
* @param[in] bfa BFA instance
* General Public License for more details.
*/
-/**
+/*
* bfa_cs.h BFA common services
*/
#include "bfa_os_inc.h"
-/**
+/*
* BFA TRC
*/
#define BFA_TRC_MOD_SH 10
#define BFA_TRC_MOD(__mod) ((BFA_TRC_ ## __mod) << BFA_TRC_MOD_SH)
-/**
+/*
* Define a new tracing file (module). Module should match one defined above.
*/
#define BFA_TRC_FILE(__mod, __submod) \
#define bfa_trc_fp(_trcp, _data)
#endif
-/**
+/*
* @ BFA LOG interfaces
*/
#define bfa_assert(__cond) do { \
#define bfa_q_is_on_q(_q, _qe) \
bfa_q_is_on_q_func(_q, (struct list_head *)(_qe))
-/**
+/*
* @ BFA state machine interfaces
*/
typedef void (*bfa_sm_t)(void *sm, int event);
-/**
+/*
* oc - object class eg. bfa_ioc
* st - state, eg. reset
* otype - object type, eg. struct bfa_ioc_s
#define bfa_sm_get_state(_sm) ((_sm)->sm)
#define bfa_sm_cmp_state(_sm, _state) ((_sm)->sm == (bfa_sm_t)(_state))
-/**
+/*
* For converting from state machine function to state encoding.
*/
struct bfa_sm_table_s {
};
#define BFA_SM(_sm) ((bfa_sm_t)(_sm))
-/**
+/*
* State machine with entry actions.
*/
typedef void (*bfa_fsm_t)(void *fsm, int event);
-/**
+/*
* oc - object class eg. bfa_ioc
* st - state, eg. reset
* otype - object type, eg. struct bfa_ioc_s
return smt[i].state;
}
-/**
+/*
* @ Generic wait counter.
*/
wc->wc_resume(wc->wc_cbarg);
}
-/**
+/*
* Initialize a waiting counter.
*/
static inline void
bfa_wc_up(wc);
}
-/**
+/*
* Wait for counter to reach zero
*/
static inline void
#define BFA_MFG_SERIALNUM_SIZE 11
#define STRSZ(_n) (((_n) + 4) & ~3)
-/**
+/*
* Manufacturing card type
*/
enum {
#pragma pack(1)
-/**
+/*
* Check if Mezz card
*/
#define bfa_mfg_is_mezz(type) (( \
(type) == BFA_MFG_TYPE_LIGHTNING_P0 || \
(type) == BFA_MFG_TYPE_LIGHTNING))
-/**
+/*
* Check if the card having old wwn/mac handling
*/
#define bfa_mfg_is_old_wwn_mac_model(type) (( \
(m)[2] = t & 0xFF; \
} while (0)
-/**
+/*
* VPD data length
*/
#define BFA_MFG_VPD_LEN 512
-/**
+/*
* VPD vendor tag
*/
enum {
BFA_MFG_VPD_PCI_BRCD = 0xf8, /* PCI VPD Brocade */
};
-/**
+/*
* All numerical fields are in big-endian format.
*/
struct bfa_mfg_vpd_s {
#pragma pack()
-/**
+/*
* Status return values
*/
enum bfa_status {
#define BFA_STRING_32 32
#define BFA_VERSION_LEN 64
-/**
+/*
* ---------------------- adapter definitions ------------
*/
-/**
+/*
* BFA adapter level attributes.
*/
enum {
u8 trunk_capable;
};
-/**
+/*
* ---------------------- IOC definitions ------------
*/
BFA_IOC_CHIP_REV_LEN = 8,
};
-/**
+/*
* Driver and firmware versions.
*/
struct bfa_ioc_driver_attr_s {
char ob_ver[BFA_VERSION_LEN]; /* openboot version */
};
-/**
+/*
* IOC PCI device attributes
*/
struct bfa_ioc_pci_attr_s {
char chip_rev[BFA_IOC_CHIP_REV_LEN]; /* chip revision */
};
-/**
+/*
* IOC states
*/
enum bfa_ioc_state {
BFA_IOC_ENABLING = 12, /* IOC is being enabled */
};
-/**
+/*
* IOC firmware stats
*/
struct bfa_fw_ioc_stats_s {
u32 unknown_reqs;
};
-/**
+/*
* IOC driver stats
*/
struct bfa_ioc_drv_stats_s {
u32 enable_replies;
};
-/**
+/*
* IOC statistics
*/
struct bfa_ioc_stats_s {
BFA_IOC_TYPE_LL = 3,
};
-/**
+/*
* IOC attributes returned in queries
*/
struct bfa_ioc_attr_s {
u8 rsvd[7]; /* 64bit align */
};
-/**
+/*
* ---------------------- mfg definitions ------------
*/
-/**
+/*
* Checksum size
*/
#define BFA_MFG_CHKSUM_SIZE 16
#pragma pack(1)
-/**
+/*
* All numerical fields are in big-endian format.
*/
struct bfa_mfg_block_s {
#pragma pack()
-/**
+/*
* ---------------------- pci definitions ------------
*/
-/**
+/*
* PCI device and vendor ID information
*/
enum {
((devid) == BFA_PCI_DEVICE_ID_CT || \
(devid) == BFA_PCI_DEVICE_ID_CT_FC)
-/**
+/*
* PCI sub-system device and vendor ID information
*/
enum {
BFA_PCI_FCOE_SSDEVICE_ID = 0x14,
};
-/**
+/*
* Maximum number of device address ranges mapped through different BAR(s)
*/
#define BFA_PCI_ACCESS_RANGES 1
#define BOOT_CFG_REV1 1
#define BOOT_CFG_VLAN 1
-/**
+/*
* Boot options setting. Boot options setting determines from where
* to get the boot lun information
*/
};
#pragma pack(1)
-/**
+/*
* Boot lun information.
*/
struct bfa_boot_bootlun_s {
};
#pragma pack()
-/**
+/*
* BOOT boot configuraton
*/
struct bfa_boot_pbc_s {
#include "bfa_fc.h"
#include "bfa_defs_svc.h"
-/**
+/*
* VF states
*/
enum bfa_vf_state {
BFA_VF_ISOLATED = 7, /* port isolated due to vf_id mismatch */
};
-/**
+/*
* VF statistics
*/
struct bfa_vf_stats_s {
u32 resvd; /* padding for 64 bit alignment */
};
-/**
+/*
* VF attributes returned in queries
*/
struct bfa_vf_attr_s {
#define BFA_FCS_MAX_LPORTS 256
#define BFA_FCS_FABRIC_IPADDR_SZ 16
-/**
+/*
* symbolic names for base port/virtual port
*/
#define BFA_SYMNAME_MAXLEN 128 /* 128 bytes */
char symname[BFA_SYMNAME_MAXLEN];
};
-/**
+/*
* Roles of FCS port:
* - FCP IM and FCP TM roles cannot be enabled together for a FCS port
* - Create multiple ports if both IM and TM functions required.
BFA_LPORT_ROLE_FCP_MAX = BFA_LPORT_ROLE_FCP_IM,
};
-/**
+/*
* FCS port configuration.
*/
struct bfa_lport_cfg_s {
wwn_t pwwn; /* port wwn */
wwn_t nwwn; /* node wwn */
struct bfa_lport_symname_s sym_name; /* vm port symbolic name */
- bfa_boolean_t preboot_vp; /* vport created from PBC */
+ bfa_boolean_t preboot_vp; /* vport created from PBC */
enum bfa_lport_role roles; /* FCS port roles */
u8 tag[16]; /* opaque tag from application */
};
-/**
+/*
* FCS port states
*/
enum bfa_lport_state {
BFA_LPORT_OFFLINE = 3, /* No login to fabric */
};
-/**
+/*
* FCS port type.
*/
enum bfa_lport_type {
BFA_LPORT_TYPE_VIRTUAL,
};
-/**
+/*
* FCS port offline reason.
*/
enum bfa_lport_offline_reason {
BFA_LPORT_OFFLINE_FAB_LOGOUT,
};
-/**
+/*
* FCS lport info.
*/
struct bfa_lport_info_s {
};
-/**
+/*
* FCS port statistics
*/
struct bfa_lport_stats_s {
* (max retry of plogi) */
};
-/**
+/*
* BFA port attribute returned in queries
*/
struct bfa_lport_attr_s {
};
-/**
+/*
* VPORT states
*/
enum bfa_vport_state {
BFA_FCS_VPORT_MAX_STATE,
};
-/**
+/*
* vport statistics
*/
struct bfa_vport_stats_s {
u32 rsvd;
};
-/**
+/*
* BFA vport attribute returned in queries
*/
struct bfa_vport_attr_s {
u32 rsvd;
};
-/**
+/*
* FCS remote port states
*/
enum bfa_rport_state {
BFA_RPORT_NSDISC = 9, /* re-discover rport */
};
-/**
+/*
* Rport Scsi Function : Initiator/Target.
*/
enum bfa_rport_function {
BFA_RPORT_TARGET = 0x02, /* SCSI Target */
};
-/**
+/*
* port/node symbolic names for rport
*/
#define BFA_RPORT_SYMNAME_MAXLEN 255
char symname[BFA_RPORT_SYMNAME_MAXLEN];
};
-/**
+/*
* FCS remote port statistics
*/
struct bfa_rport_stats_s {
struct bfa_rport_hal_stats_s hal_stats; /* BFA rport stats */
};
-/**
+/*
* FCS remote port attributes returned in queries
*/
struct bfa_rport_attr_s {
#define BFA_MAX_IO_INDEX 7
#define BFA_NO_IO_INDEX 9
-/**
+/*
* FCS itnim states
*/
enum bfa_itnim_state {
BFA_ITNIM_INITIATIOR = 7, /* initiator */
};
-/**
+/*
* FCS remote port statistics
*/
struct bfa_itnim_stats_s {
u32 rsvd; /* padding for 64 bit alignment */
};
-/**
+/*
* FCS itnim attributes returned in queries
*/
struct bfa_itnim_attr_s {
#define BFA_IOCFCOE_INTR_DELAY 25
#define BFA_IOCFCOE_INTR_LATENCY 5
-/**
+/*
* Interrupt coalescing configuration.
*/
#pragma pack(1)
u16 delay; /* delay in microseconds */
};
-/**
+/*
* IOC firmware configuraton
*/
struct bfa_iocfc_fwcfg_s {
u32 rsvd;
};
-/**
+/*
* IOC configuration
*/
struct bfa_iocfc_cfg_s {
struct bfa_iocfc_drvcfg_s drvcfg; /* driver side config */
};
-/**
+/*
* IOC firmware IO stats
*/
struct bfa_fw_io_stats_s {
*/
};
-/**
+/*
* IOC port firmware stats
*/
u32 mac_invalids; /* Invalid mac assigned */
};
-/**
+/*
* IOC firmware FCoE port stats
*/
struct bfa_fw_fcoe_port_stats_s {
struct bfa_fw_fip_stats_s fip_stats;
};
-/**
+/*
* IOC firmware FC uport stats
*/
struct bfa_fw_fc_uport_stats_s {
struct bfa_fw_port_lksm_stats_s lksm_stats;
};
-/**
+/*
* IOC firmware FC port stats
*/
union bfa_fw_fc_port_stats_s {
struct bfa_fw_fcoe_port_stats_s fcoe_stats;
};
-/**
+/*
* IOC firmware port stats
*/
struct bfa_fw_port_stats_s {
union bfa_fw_fc_port_stats_s fc_port;
};
-/**
+/*
* fcxchg module statistics
*/
struct bfa_fw_fcxchg_stats_s {
u32 cls_tx;
};
-/**
+/*
* Trunk statistics
*/
struct bfa_fw_trunk_stats_s {
u32 elp_dropped; /* ELP dropped */
};
-/**
+/*
* IOCFC firmware stats
*/
struct bfa_fw_iocfc_stats_s {
u32 set_intr_reqs; /* set interrupt reqs */
};
-/**
+/*
* IOC attributes returned in queries
*/
struct bfa_iocfc_attr_s {
struct bfa_iocfc_intr_attr_s intr_attr; /* interrupt attr */
};
-/**
+/*
* Eth_sndrcv mod stats
*/
struct bfa_fw_eth_sndrcv_stats_s {
u32 rsvd; /* 64bit align */
};
-/**
+/*
* CT MAC mod stats
*/
struct bfa_fw_mac_mod_stats_s {
u32 rsvd; /* 64bit align */
};
-/**
+/*
* CT MOD stats
*/
struct bfa_fw_ct_mod_stats_s {
u32 rsvd; /* 64bit align */
};
-/**
+/*
* IOC firmware stats
*/
struct bfa_fw_stats_s {
#define BFA_IOCFC_PATHTOV_MAX 60
#define BFA_IOCFC_QDEPTH_MAX 2000
-/**
+/*
* QoS states
*/
enum bfa_qos_state {
BFA_QOS_OFFLINE = 2, /* QoS is offline */
};
-/**
+/*
* QoS Priority levels.
*/
enum bfa_qos_priority {
BFA_QOS_LOW = 3, /* QoS Priority Level Low */
};
-/**
+/*
* QoS bandwidth allocation for each priority level
*/
enum bfa_qos_bw_alloc {
BFA_QOS_BW_LOW = 10, /* bandwidth allocation for Low */
};
#pragma pack(1)
-/**
+/*
* QoS attribute returned in QoS Query
*/
struct bfa_qos_attr_s {
u32 total_bb_cr; /* Total BB Credits */
};
-/**
+/*
* These fields should be displayed only from the CLI.
* There will be a separate BFAL API (get_qos_vc_attr ?)
* to retrieve this.
* total_vc_count */
};
-/**
+/*
* QoS statistics
*/
struct bfa_qos_stats_s {
u32 rsvd; /* padding for 64 bit alignment */
};
-/**
+/*
* FCoE statistics
*/
struct bfa_fcoe_stats_s {
u64 rxf_bcast_vlan; /* Rx FCoE broadcast vlan frames */
};
-/**
+/*
* QoS or FCoE stats (fcport stats excluding physical FC port stats)
*/
union bfa_fcport_stats_u {
BFA_PORT_ST_MAX_STATE,
};
-/**
+/*
* Port operational type (in sync with SNIA port type).
*/
enum bfa_port_type {
BFA_PORT_TYPE_VPORT = 22, /* NPIV - virtual port */
};
-/**
+/*
* Port topology setting. A port's topology and fabric login status
* determine its operational type.
*/
BFA_PORT_TOPOLOGY_AUTO = 3, /* auto topology selection */
};
-/**
+/*
* Physical port loopback types.
*/
enum bfa_port_opmode {
(_mode == BFA_PORT_OPMODE_LB_SLW) || \
(_mode == BFA_PORT_OPMODE_LB_EXT))
-/**
+/*
* Port link state
*/
enum bfa_port_linkstate {
BFA_PORT_LINKDOWN = 2, /* Physical port/Trunk link down */
};
-/**
+/*
* Port link state reason code
*/
enum bfa_port_linkstate_rsn {
CEE_ISCSI_PRI_OVERLAP_FCOE_PRI = 43
};
#pragma pack(1)
-/**
+/*
* Physical port configuration
*/
struct bfa_port_cfg_s {
};
#pragma pack()
-/**
+/*
* Port attribute values.
*/
struct bfa_port_attr_s {
u8 rsvd1[6];
};
-/**
+/*
* Port FCP mappings.
*/
struct bfa_port_fcpmap_s {
char luid[256];
};
-/**
+/*
* Port RNID info.
*/
struct bfa_port_rnid_s {
mac_t mac; /* FCF mac */
};
-/**
+/*
* Trunk states for BCU/BFAL
*/
enum bfa_trunk_state {
BFA_TRUNK_OFFLINE = 2, /* Trunk is offline */
};
-/**
+/*
* VC attributes for trunked link
*/
struct bfa_trunk_vc_attr_s {
u16 vc_credits[8];
};
-/**
+/*
* Link state information
*/
struct bfa_port_link_s {
u32 rsvd;
};
#pragma pack(1)
-/**
+/*
* Rport's QoS attributes
*/
struct bfa_rport_qos_attr_s {
struct bfa_itnim_latency_s io_latency;
};
-/**
+/*
* FC physical port statistics.
*/
struct bfa_port_fc_stats_s {
u64 err_enc; /* Encoding err frame_8b10b */
};
-/**
+/*
* Eth Physical Port statistics.
*/
struct bfa_port_eth_stats_s {
u64 tx_iscsi_zero_pause; /* Tx iSCSI zero pause */
};
-/**
+/*
* Port statistics.
*/
union bfa_port_stats_u {
#include "bfa_modules.h"
-/**
+/*
* BFA module list terminated by NULL
*/
struct bfa_module_s *hal_mods[] = {
NULL
};
-/**
+/*
* Message handlers for various modules.
*/
bfa_isr_func_t bfa_isrs[BFI_MC_MAX] = {
};
-/**
+/*
* Message handlers for mailbox command classes
*/
bfa_ioc_mbox_mcfunc_t bfa_mbox_isrs[BFI_MC_MAX] = {
struct link_e2e_beacon_param_s beacon_parm;
};
-/**
+/*
* If RPSC request is sent to the Domain Controller, the request is for
* all the ports within that domain (TODO - I don't think FOS implements
* this...).
struct fc_rpsc_speed_info_s speed_info[1];
};
-/**
+/*
* If RPSC2 request is sent to the Domain Controller,
*/
#define FC_BRCD_TOKEN 0x42524344
struct fc_rpsc2_port_info_s port_info[1]; /* port information */
};
-/**
+/*
* bit fields so that multiple classes can be specified
*/
enum fc_cos {
#define FC_VF_ID_MAX 0xEFF
#define FC_VF_ID_CTL 0xFEF /* control VF_ID */
-/**
+/*
* Virtual Fabric Tagging header format
* @caution This is defined only in BIG ENDIAN format.
*/
u32 dap:24; /* port identifier */
};
-/**
+/*
* RFT_ID
*/
struct fcgs_rftid_req_s {
u32 fc4_type[8]; /* fc4 types */
};
-/**
+/*
* RFF_ID : Register FC4 features.
*/
u32 fc4_type:8; /* corresponding FC4 Type */
};
-/**
+/*
* GID_FT Request
*/
struct fcgs_gidft_req_s {
u8 fc4_type; /* FC_TYPE_FCP for SCSI devices */
}; /* GID_FT Request */
-/**
+/*
* GID_FT Response
*/
struct fcgs_gidft_resp_s {
u32 pid:24; /* port identifier */
}; /* GID_FT Response */
-/**
+/*
* RSPN_ID
*/
struct fcgs_rspnid_req_s {
u8 spn[256]; /* symbolic port name */
};
-/**
+/*
* RPN_ID
*/
struct fcgs_rpnid_req_s {
wwn_t port_name;
};
-/**
+/*
* RNN_ID
*/
struct fcgs_rnnid_req_s {
wwn_t node_name;
};
-/**
+/*
* RCS_ID
*/
struct fcgs_rcsid_req_s {
u32 cos;
};
-/**
+/*
* RPT_ID
*/
struct fcgs_rptid_req_s {
u32 rsvd1:24;
};
-/**
+/*
* GA_NXT Request
*/
struct fcgs_ganxt_req_s {
u32 port_id:24;
};
-/**
+/*
* GA_NXT Response
*/
struct fcgs_ganxt_rsp_s {
*/
plogi_tmpl.csp.verhi = FC_PH_VER_PH_3;
plogi_tmpl.csp.verlo = FC_PH_VER_4_3;
- plogi_tmpl.csp.bbcred = bfa_os_htons(0x0004);
+ plogi_tmpl.csp.bbcred = cpu_to_be16(0x0004);
plogi_tmpl.csp.ciro = 0x1;
plogi_tmpl.csp.cisc = 0x0;
plogi_tmpl.csp.altbbcred = 0x0;
- plogi_tmpl.csp.conseq = bfa_os_htons(0x00FF);
- plogi_tmpl.csp.ro_bitmap = bfa_os_htons(0x0002);
- plogi_tmpl.csp.e_d_tov = bfa_os_htonl(2000);
+ plogi_tmpl.csp.conseq = cpu_to_be16(0x00FF);
+ plogi_tmpl.csp.ro_bitmap = cpu_to_be16(0x0002);
+ plogi_tmpl.csp.e_d_tov = cpu_to_be32(2000);
plogi_tmpl.class3.class_valid = 1;
plogi_tmpl.class3.sequential = 1;
*/
prli_tmpl.command = FC_ELS_PRLI;
prli_tmpl.pglen = 0x10;
- prli_tmpl.pagebytes = bfa_os_htons(0x0014);
+ prli_tmpl.pagebytes = cpu_to_be16(0x0014);
prli_tmpl.parampage.type = FC_TYPE_FCP;
prli_tmpl.parampage.imagepair = 1;
prli_tmpl.parampage.servparams.rxrdisab = 1;
static void
fc_gs_fchdr_build(struct fchs_s *fchs, u32 d_id, u32 s_id, u32 ox_id)
{
- bfa_os_memset(fchs, 0, sizeof(struct fchs_s));
+ memset(fchs, 0, sizeof(struct fchs_s));
fchs->routing = FC_RTG_FC4_DEV_DATA;
fchs->cat_info = FC_CAT_UNSOLICIT_CTRL;
fchs->rx_id = FC_RXID_ANY;
fchs->d_id = (d_id);
fchs->s_id = (s_id);
- fchs->ox_id = bfa_os_htons(ox_id);
+ fchs->ox_id = cpu_to_be16(ox_id);
- /**
+ /*
* @todo no need to set ox_id for request
* no need to set rx_id for response
*/
void
fc_els_req_build(struct fchs_s *fchs, u32 d_id, u32 s_id, u16 ox_id)
{
- bfa_os_memcpy(fchs, &fc_els_req_tmpl, sizeof(struct fchs_s));
+ memcpy(fchs, &fc_els_req_tmpl, sizeof(struct fchs_s));
fchs->d_id = (d_id);
fchs->s_id = (s_id);
- fchs->ox_id = bfa_os_htons(ox_id);
+ fchs->ox_id = cpu_to_be16(ox_id);
}
static void
fc_els_rsp_build(struct fchs_s *fchs, u32 d_id, u32 s_id, u16 ox_id)
{
- bfa_os_memcpy(fchs, &fc_els_rsp_tmpl, sizeof(struct fchs_s));
+ memcpy(fchs, &fc_els_rsp_tmpl, sizeof(struct fchs_s));
fchs->d_id = d_id;
fchs->s_id = s_id;
fchs->ox_id = ox_id;
static void
fc_bls_rsp_build(struct fchs_s *fchs, u32 d_id, u32 s_id, u16 ox_id)
{
- bfa_os_memcpy(fchs, &fc_bls_rsp_tmpl, sizeof(struct fchs_s));
+ memcpy(fchs, &fc_bls_rsp_tmpl, sizeof(struct fchs_s));
fchs->d_id = d_id;
fchs->s_id = s_id;
fchs->ox_id = ox_id;
{
struct fc_logi_s *plogi = (struct fc_logi_s *) (pld);
- bfa_os_memcpy(plogi, &plogi_tmpl, sizeof(struct fc_logi_s));
+ memcpy(plogi, &plogi_tmpl, sizeof(struct fc_logi_s));
plogi->els_cmd.els_code = els_code;
if (els_code == FC_ELS_PLOGI)
else
fc_els_rsp_build(fchs, d_id, s_id, ox_id);
- plogi->csp.rxsz = plogi->class3.rxsz = bfa_os_htons(pdu_size);
+ plogi->csp.rxsz = plogi->class3.rxsz = cpu_to_be16(pdu_size);
- bfa_os_memcpy(&plogi->port_name, &port_name, sizeof(wwn_t));
- bfa_os_memcpy(&plogi->node_name, &node_name, sizeof(wwn_t));
+ memcpy(&plogi->port_name, &port_name, sizeof(wwn_t));
+ memcpy(&plogi->node_name, &node_name, sizeof(wwn_t));
return sizeof(struct fc_logi_s);
}
u32 d_id = bfa_os_hton3b(FC_FABRIC_PORT);
u32 *vvl_info;
- bfa_os_memcpy(flogi, &plogi_tmpl, sizeof(struct fc_logi_s));
+ memcpy(flogi, &plogi_tmpl, sizeof(struct fc_logi_s));
flogi->els_cmd.els_code = FC_ELS_FLOGI;
fc_els_req_build(fchs, d_id, s_id, ox_id);
- flogi->csp.rxsz = flogi->class3.rxsz = bfa_os_htons(pdu_size);
+ flogi->csp.rxsz = flogi->class3.rxsz = cpu_to_be16(pdu_size);
flogi->port_name = port_name;
flogi->node_name = node_name;
/* set AUTH capability */
flogi->csp.security = set_auth;
- flogi->csp.bbcred = bfa_os_htons(local_bb_credits);
+ flogi->csp.bbcred = cpu_to_be16(local_bb_credits);
/* Set brcd token in VVL */
vvl_info = (u32 *)&flogi->vvl[0];
/* set the flag to indicate the presence of VVL */
flogi->csp.npiv_supp = 1; /* @todo. field name is not correct */
- vvl_info[0] = bfa_os_htonl(FLOGI_VVL_BRCD);
+ vvl_info[0] = cpu_to_be32(FLOGI_VVL_BRCD);
return sizeof(struct fc_logi_s);
}
{
u32 d_id = 0;
- bfa_os_memcpy(flogi, &plogi_tmpl, sizeof(struct fc_logi_s));
+ memcpy(flogi, &plogi_tmpl, sizeof(struct fc_logi_s));
fc_els_rsp_build(fchs, d_id, s_id, ox_id);
flogi->els_cmd.els_code = FC_ELS_ACC;
- flogi->csp.rxsz = flogi->class3.rxsz = bfa_os_htons(pdu_size);
+ flogi->csp.rxsz = flogi->class3.rxsz = cpu_to_be16(pdu_size);
flogi->port_name = port_name;
flogi->node_name = node_name;
- flogi->csp.bbcred = bfa_os_htons(local_bb_credits);
+ flogi->csp.bbcred = cpu_to_be16(local_bb_credits);
return sizeof(struct fc_logi_s);
}
{
u32 d_id = bfa_os_hton3b(FC_FABRIC_PORT);
- bfa_os_memcpy(flogi, &plogi_tmpl, sizeof(struct fc_logi_s));
+ memcpy(flogi, &plogi_tmpl, sizeof(struct fc_logi_s));
flogi->els_cmd.els_code = FC_ELS_FDISC;
fc_els_req_build(fchs, d_id, s_id, ox_id);
- flogi->csp.rxsz = flogi->class3.rxsz = bfa_os_htons(pdu_size);
+ flogi->csp.rxsz = flogi->class3.rxsz = cpu_to_be16(pdu_size);
flogi->port_name = port_name;
flogi->node_name = node_name;
if (!plogi->class3.class_valid)
return FC_PARSE_FAILURE;
- if (bfa_os_ntohs(plogi->class3.rxsz) < (FC_MIN_PDUSZ))
+ if (be16_to_cpu(plogi->class3.rxsz) < (FC_MIN_PDUSZ))
return FC_PARSE_FAILURE;
return FC_PARSE_OK;
if (plogi->class3.class_valid != 1)
return FC_PARSE_FAILURE;
- if ((bfa_os_ntohs(plogi->class3.rxsz) < FC_MIN_PDUSZ)
- || (bfa_os_ntohs(plogi->class3.rxsz) > FC_MAX_PDUSZ)
+ if ((be16_to_cpu(plogi->class3.rxsz) < FC_MIN_PDUSZ)
+ || (be16_to_cpu(plogi->class3.rxsz) > FC_MAX_PDUSZ)
|| (plogi->class3.rxsz == 0))
return FC_PARSE_FAILURE;
struct fc_prli_s *prli = (struct fc_prli_s *) (pld);
fc_els_req_build(fchs, d_id, s_id, ox_id);
- bfa_os_memcpy(prli, &prli_tmpl, sizeof(struct fc_prli_s));
+ memcpy(prli, &prli_tmpl, sizeof(struct fc_prli_s));
prli->command = FC_ELS_PRLI;
prli->parampage.servparams.initiator = 1;
struct fc_prli_s *prli = (struct fc_prli_s *) (pld);
fc_els_rsp_build(fchs, d_id, s_id, ox_id);
- bfa_os_memcpy(prli, &prli_tmpl, sizeof(struct fc_prli_s));
+ memcpy(prli, &prli_tmpl, sizeof(struct fc_prli_s));
prli->command = FC_ELS_ACC;
{
fc_els_req_build(fchs, d_id, s_id, ox_id);
- bfa_os_memset(logo, '\0', sizeof(struct fc_logo_s));
+ memset(logo, '\0', sizeof(struct fc_logo_s));
logo->els_cmd.els_code = FC_ELS_LOGO;
logo->nport_id = (s_id);
logo->orig_port_name = port_name;
u32 s_id, u16 ox_id, wwn_t port_name,
wwn_t node_name, u8 els_code)
{
- bfa_os_memset(adisc, '\0', sizeof(struct fc_adisc_s));
+ memset(adisc, '\0', sizeof(struct fc_adisc_s));
adisc->els_cmd.els_code = els_code;
if (pdisc->class3.class_valid != 1)
return FC_PARSE_FAILURE;
- if ((bfa_os_ntohs(pdisc->class3.rxsz) <
+ if ((be16_to_cpu(pdisc->class3.rxsz) <
(FC_MIN_PDUSZ - sizeof(struct fchs_s)))
|| (pdisc->class3.rxsz == 0))
return FC_PARSE_FAILURE;
u16
fc_abts_build(struct fchs_s *fchs, u32 d_id, u32 s_id, u16 ox_id)
{
- bfa_os_memcpy(fchs, &fc_bls_req_tmpl, sizeof(struct fchs_s));
+ memcpy(fchs, &fc_bls_req_tmpl, sizeof(struct fchs_s));
fchs->cat_info = FC_CAT_ABTS;
fchs->d_id = (d_id);
fchs->s_id = (s_id);
- fchs->ox_id = bfa_os_htons(ox_id);
+ fchs->ox_id = cpu_to_be16(ox_id);
return sizeof(struct fchs_s);
}
/*
* build rrq payload
*/
- bfa_os_memcpy(rrq, &rrq_tmpl, sizeof(struct fc_rrq_s));
+ memcpy(rrq, &rrq_tmpl, sizeof(struct fc_rrq_s));
rrq->s_id = (s_id);
- rrq->ox_id = bfa_os_htons(rrq_oxid);
+ rrq->ox_id = cpu_to_be16(rrq_oxid);
rrq->rx_id = FC_RXID_ANY;
return sizeof(struct fc_rrq_s);
fc_els_rsp_build(fchs, d_id, s_id, ox_id);
- bfa_os_memset(acc, 0, sizeof(struct fc_els_cmd_s));
+ memset(acc, 0, sizeof(struct fc_els_cmd_s));
acc->els_code = FC_ELS_ACC;
return sizeof(struct fc_els_cmd_s);
u8 reason_code_expl)
{
fc_els_rsp_build(fchs, d_id, s_id, ox_id);
- bfa_os_memset(ls_rjt, 0, sizeof(struct fc_ls_rjt_s));
+ memset(ls_rjt, 0, sizeof(struct fc_ls_rjt_s));
ls_rjt->els_cmd.els_code = FC_ELS_LS_RJT;
ls_rjt->reason_code = reason_code;
{
fc_bls_rsp_build(fchs, d_id, s_id, ox_id);
- bfa_os_memcpy(ba_acc, &ba_acc_tmpl, sizeof(struct fc_ba_acc_s));
+ memcpy(ba_acc, &ba_acc_tmpl, sizeof(struct fc_ba_acc_s));
fchs->rx_id = rx_id;
u32 s_id, u16 ox_id)
{
fc_els_rsp_build(fchs, d_id, s_id, ox_id);
- bfa_os_memset(els_cmd, 0, sizeof(struct fc_els_cmd_s));
+ memset(els_cmd, 0, sizeof(struct fc_els_cmd_s));
els_cmd->els_code = FC_ELS_ACC;
return sizeof(struct fc_els_cmd_s);
if (els_code == FC_ELS_PRLO) {
prlo = (struct fc_prlo_s *) (fc_frame + 1);
- num_pages = (bfa_os_ntohs(prlo->payload_len) - 4) / 16;
+ num_pages = (be16_to_cpu(prlo->payload_len) - 4) / 16;
} else {
tprlo = (struct fc_tprlo_s *) (fc_frame + 1);
- num_pages = (bfa_os_ntohs(tprlo->payload_len) - 4) / 16;
+ num_pages = (be16_to_cpu(tprlo->payload_len) - 4) / 16;
}
return num_pages;
}
fc_els_rsp_build(fchs, d_id, s_id, ox_id);
- bfa_os_memset(tprlo_acc, 0, (num_pages * 16) + 4);
+ memset(tprlo_acc, 0, (num_pages * 16) + 4);
tprlo_acc->command = FC_ELS_ACC;
tprlo_acc->page_len = 0x10;
- tprlo_acc->payload_len = bfa_os_htons((num_pages * 16) + 4);
+ tprlo_acc->payload_len = cpu_to_be16((num_pages * 16) + 4);
for (page = 0; page < num_pages; page++) {
tprlo_acc->tprlo_acc_params[page].opa_valid = 0;
tprlo_acc->tprlo_acc_params[page].orig_process_assc = 0;
tprlo_acc->tprlo_acc_params[page].resp_process_assc = 0;
}
- return bfa_os_ntohs(tprlo_acc->payload_len);
+ return be16_to_cpu(tprlo_acc->payload_len);
}
u16
fc_els_rsp_build(fchs, d_id, s_id, ox_id);
- bfa_os_memset(prlo_acc, 0, (num_pages * 16) + 4);
+ memset(prlo_acc, 0, (num_pages * 16) + 4);
prlo_acc->command = FC_ELS_ACC;
prlo_acc->page_len = 0x10;
- prlo_acc->payload_len = bfa_os_htons((num_pages * 16) + 4);
+ prlo_acc->payload_len = cpu_to_be16((num_pages * 16) + 4);
for (page = 0; page < num_pages; page++) {
prlo_acc->prlo_acc_params[page].opa_valid = 0;
prlo_acc->prlo_acc_params[page].resp_process_assc = 0;
}
- return bfa_os_ntohs(prlo_acc->payload_len);
+ return be16_to_cpu(prlo_acc->payload_len);
}
u16
{
fc_els_req_build(fchs, d_id, s_id, ox_id);
- bfa_os_memset(rnid, 0, sizeof(struct fc_rnid_cmd_s));
+ memset(rnid, 0, sizeof(struct fc_rnid_cmd_s));
rnid->els_cmd.els_code = FC_ELS_RNID;
rnid->node_id_data_format = data_format;
struct fc_rnid_common_id_data_s *common_id_data,
struct fc_rnid_general_topology_data_s *gen_topo_data)
{
- bfa_os_memset(rnid_acc, 0, sizeof(struct fc_rnid_acc_s));
+ memset(rnid_acc, 0, sizeof(struct fc_rnid_acc_s));
fc_els_rsp_build(fchs, d_id, s_id, ox_id);
if (data_format == RNID_NODEID_DATA_FORMAT_DISCOVERY) {
rnid_acc->specific_id_data_length =
sizeof(struct fc_rnid_general_topology_data_s);
- bfa_os_assign(rnid_acc->gen_topology_data, *gen_topo_data);
+ rnid_acc->gen_topology_data = *gen_topo_data;
return sizeof(struct fc_rnid_acc_s);
} else {
return sizeof(struct fc_rnid_acc_s) -
{
fc_els_req_build(fchs, d_id, s_id, ox_id);
- bfa_os_memset(rpsc, 0, sizeof(struct fc_rpsc_cmd_s));
+ memset(rpsc, 0, sizeof(struct fc_rpsc_cmd_s));
rpsc->els_cmd.els_code = FC_ELS_RPSC;
return sizeof(struct fc_rpsc_cmd_s);
fc_els_req_build(fchs, bfa_os_hton3b(dctlr_id), s_id, 0);
- bfa_os_memset(rpsc2, 0, sizeof(struct fc_rpsc2_cmd_s));
+ memset(rpsc2, 0, sizeof(struct fc_rpsc2_cmd_s));
rpsc2->els_cmd.els_code = FC_ELS_RPSC;
- rpsc2->token = bfa_os_htonl(FC_BRCD_TOKEN);
- rpsc2->num_pids = bfa_os_htons(npids);
+ rpsc2->token = cpu_to_be32(FC_BRCD_TOKEN);
+ rpsc2->num_pids = cpu_to_be16(npids);
for (i = 0; i < npids; i++)
rpsc2->pid_list[i].pid = pid_list[i];
u32 d_id, u32 s_id, u16 ox_id,
struct fc_rpsc_speed_info_s *oper_speed)
{
- bfa_os_memset(rpsc_acc, 0, sizeof(struct fc_rpsc_acc_s));
+ memset(rpsc_acc, 0, sizeof(struct fc_rpsc_acc_s));
fc_els_rsp_build(fchs, d_id, s_id, ox_id);
rpsc_acc->command = FC_ELS_ACC;
- rpsc_acc->num_entries = bfa_os_htons(1);
+ rpsc_acc->num_entries = cpu_to_be16(1);
rpsc_acc->speed_info[0].port_speed_cap =
- bfa_os_htons(oper_speed->port_speed_cap);
+ cpu_to_be16(oper_speed->port_speed_cap);
rpsc_acc->speed_info[0].port_op_speed =
- bfa_os_htons(oper_speed->port_op_speed);
+ cpu_to_be16(oper_speed->port_op_speed);
return sizeof(struct fc_rpsc_acc_s);
}
{
struct fc_logi_s *pdisc = (struct fc_logi_s *) (fchs + 1);
- bfa_os_memcpy(pdisc, &plogi_tmpl, sizeof(struct fc_logi_s));
+ memcpy(pdisc, &plogi_tmpl, sizeof(struct fc_logi_s));
pdisc->els_cmd.els_code = FC_ELS_PDISC;
fc_els_req_build(fchs, d_id, s_id, ox_id);
- pdisc->csp.rxsz = pdisc->class3.rxsz = bfa_os_htons(pdu_size);
+ pdisc->csp.rxsz = pdisc->class3.rxsz = cpu_to_be16(pdu_size);
pdisc->port_name = port_name;
pdisc->node_name = node_name;
if (!pdisc->class3.class_valid)
return FC_PARSE_NWWN_NOT_EQUAL;
- if (bfa_os_ntohs(pdisc->class3.rxsz) < (FC_MIN_PDUSZ))
+ if (be16_to_cpu(pdisc->class3.rxsz) < (FC_MIN_PDUSZ))
return FC_PARSE_RXSZ_INVAL;
return FC_PARSE_OK;
int page;
fc_els_req_build(fchs, d_id, s_id, ox_id);
- bfa_os_memset(prlo, 0, (num_pages * 16) + 4);
+ memset(prlo, 0, (num_pages * 16) + 4);
prlo->command = FC_ELS_PRLO;
prlo->page_len = 0x10;
- prlo->payload_len = bfa_os_htons((num_pages * 16) + 4);
+ prlo->payload_len = cpu_to_be16((num_pages * 16) + 4);
for (page = 0; page < num_pages; page++) {
prlo->prlo_params[page].type = FC_TYPE_FCP;
prlo->prlo_params[page].resp_process_assc = 0;
}
- return bfa_os_ntohs(prlo->payload_len);
+ return be16_to_cpu(prlo->payload_len);
}
u16
if (prlo->command != FC_ELS_ACC)
return FC_PARSE_FAILURE;
- num_pages = ((bfa_os_ntohs(prlo->payload_len)) - 4) / 16;
+ num_pages = ((be16_to_cpu(prlo->payload_len)) - 4) / 16;
for (page = 0; page < num_pages; page++) {
if (prlo->prlo_acc_params[page].type != FC_TYPE_FCP)
int page;
fc_els_req_build(fchs, d_id, s_id, ox_id);
- bfa_os_memset(tprlo, 0, (num_pages * 16) + 4);
+ memset(tprlo, 0, (num_pages * 16) + 4);
tprlo->command = FC_ELS_TPRLO;
tprlo->page_len = 0x10;
- tprlo->payload_len = bfa_os_htons((num_pages * 16) + 4);
+ tprlo->payload_len = cpu_to_be16((num_pages * 16) + 4);
for (page = 0; page < num_pages; page++) {
tprlo->tprlo_params[page].type = FC_TYPE_FCP;
}
}
- return bfa_os_ntohs(tprlo->payload_len);
+ return be16_to_cpu(tprlo->payload_len);
}
u16
if (tprlo->command != FC_ELS_ACC)
return FC_PARSE_ACC_INVAL;
- num_pages = (bfa_os_ntohs(tprlo->payload_len) - 4) / 16;
+ num_pages = (be16_to_cpu(tprlo->payload_len) - 4) / 16;
for (page = 0; page < num_pages; page++) {
if (tprlo->tprlo_acc_params[page].type != FC_TYPE_FCP)
static void
fc_gs_cthdr_build(struct ct_hdr_s *cthdr, u32 s_id, u16 cmd_code)
{
- bfa_os_memset(cthdr, 0, sizeof(struct ct_hdr_s));
+ memset(cthdr, 0, sizeof(struct ct_hdr_s));
cthdr->rev_id = CT_GS3_REVISION;
cthdr->gs_type = CT_GSTYPE_DIRSERVICE;
cthdr->gs_sub_type = CT_GSSUBTYPE_NAMESERVER;
- cthdr->cmd_rsp_code = bfa_os_htons(cmd_code);
+ cthdr->cmd_rsp_code = cpu_to_be16(cmd_code);
}
static void
fc_gs_fdmi_cthdr_build(struct ct_hdr_s *cthdr, u32 s_id, u16 cmd_code)
{
- bfa_os_memset(cthdr, 0, sizeof(struct ct_hdr_s));
+ memset(cthdr, 0, sizeof(struct ct_hdr_s));
cthdr->rev_id = CT_GS3_REVISION;
cthdr->gs_type = CT_GSTYPE_MGMTSERVICE;
cthdr->gs_sub_type = CT_GSSUBTYPE_HBA_MGMTSERVER;
- cthdr->cmd_rsp_code = bfa_os_htons(cmd_code);
+ cthdr->cmd_rsp_code = cpu_to_be16(cmd_code);
}
static void
fc_gs_ms_cthdr_build(struct ct_hdr_s *cthdr, u32 s_id, u16 cmd_code,
u8 sub_type)
{
- bfa_os_memset(cthdr, 0, sizeof(struct ct_hdr_s));
+ memset(cthdr, 0, sizeof(struct ct_hdr_s));
cthdr->rev_id = CT_GS3_REVISION;
cthdr->gs_type = CT_GSTYPE_MGMTSERVICE;
cthdr->gs_sub_type = sub_type;
- cthdr->cmd_rsp_code = bfa_os_htons(cmd_code);
+ cthdr->cmd_rsp_code = cpu_to_be16(cmd_code);
}
u16
fc_gs_fchdr_build(fchs, d_id, s_id, ox_id);
fc_gs_cthdr_build(cthdr, s_id, GS_GID_PN);
- bfa_os_memset(gidpn, 0, sizeof(struct fcgs_gidpn_req_s));
+ memset(gidpn, 0, sizeof(struct fcgs_gidpn_req_s));
gidpn->port_name = port_name;
return sizeof(struct fcgs_gidpn_req_s) + sizeof(struct ct_hdr_s);
}
fc_gs_fchdr_build(fchs, d_id, s_id, ox_id);
fc_gs_cthdr_build(cthdr, s_id, GS_GPN_ID);
- bfa_os_memset(gpnid, 0, sizeof(fcgs_gpnid_req_t));
+ memset(gpnid, 0, sizeof(fcgs_gpnid_req_t));
gpnid->dap = port_id;
return sizeof(fcgs_gpnid_req_t) + sizeof(struct ct_hdr_s);
}
fc_gs_fchdr_build(fchs, d_id, s_id, ox_id);
fc_gs_cthdr_build(cthdr, s_id, GS_GNN_ID);
- bfa_os_memset(gnnid, 0, sizeof(fcgs_gnnid_req_t));
+ memset(gnnid, 0, sizeof(fcgs_gnnid_req_t));
gnnid->dap = port_id;
return sizeof(fcgs_gnnid_req_t) + sizeof(struct ct_hdr_s);
}
u16
fc_ct_rsp_parse(struct ct_hdr_s *cthdr)
{
- if (bfa_os_ntohs(cthdr->cmd_rsp_code) != CT_RSP_ACCEPT) {
+ if (be16_to_cpu(cthdr->cmd_rsp_code) != CT_RSP_ACCEPT) {
if (cthdr->reason_code == CT_RSN_LOGICAL_BUSY)
return FC_PARSE_BUSY;
else
fc_els_req_build(fchs, d_id, s_id, ox_id);
- bfa_os_memset(scr, 0, sizeof(struct fc_scr_s));
+ memset(scr, 0, sizeof(struct fc_scr_s));
scr->command = FC_ELS_SCR;
scr->reg_func = FC_SCR_REG_FUNC_FULL;
if (set_br_reg)
rscn->pagelen = sizeof(rscn->event[0]);
payldlen = sizeof(u32) + rscn->pagelen;
- rscn->payldlen = bfa_os_htons(payldlen);
+ rscn->payldlen = cpu_to_be16(payldlen);
rscn->event[0].format = FC_RSCN_FORMAT_PORTID;
rscn->event[0].portid = s_id;
fc_gs_fchdr_build(fchs, d_id, s_id, ox_id);
fc_gs_cthdr_build(cthdr, s_id, GS_RFT_ID);
- bfa_os_memset(rftid, 0, sizeof(struct fcgs_rftid_req_s));
+ memset(rftid, 0, sizeof(struct fcgs_rftid_req_s));
rftid->dap = s_id;
/* By default, FCP FC4 Type is registered */
index = FC_TYPE_FCP >> 5;
type_value = 1 << (FC_TYPE_FCP % 32);
- rftid->fc4_type[index] = bfa_os_htonl(type_value);
+ rftid->fc4_type[index] = cpu_to_be32(type_value);
return sizeof(struct fcgs_rftid_req_s) + sizeof(struct ct_hdr_s);
}
fc_gs_fchdr_build(fchs, d_id, s_id, ox_id);
fc_gs_cthdr_build(cthdr, s_id, GS_RFT_ID);
- bfa_os_memset(rftid, 0, sizeof(struct fcgs_rftid_req_s));
+ memset(rftid, 0, sizeof(struct fcgs_rftid_req_s));
rftid->dap = s_id;
- bfa_os_memcpy((void *)rftid->fc4_type, (void *)fc4_bitmap,
+ memcpy((void *)rftid->fc4_type, (void *)fc4_bitmap,
(bitmap_size < 32 ? bitmap_size : 32));
return sizeof(struct fcgs_rftid_req_s) + sizeof(struct ct_hdr_s);
fc_gs_fchdr_build(fchs, d_id, s_id, ox_id);
fc_gs_cthdr_build(cthdr, s_id, GS_RFF_ID);
- bfa_os_memset(rffid, 0, sizeof(struct fcgs_rffid_req_s));
+ memset(rffid, 0, sizeof(struct fcgs_rffid_req_s));
rffid->dap = s_id;
rffid->fc4ftr_bits = fc4_ftrs;
fc_gs_fchdr_build(fchs, d_id, s_id, ox_id);
fc_gs_cthdr_build(cthdr, s_id, GS_RSPN_ID);
- bfa_os_memset(rspnid, 0, sizeof(struct fcgs_rspnid_req_s));
+ memset(rspnid, 0, sizeof(struct fcgs_rspnid_req_s));
rspnid->dap = s_id;
rspnid->spn_len = (u8) strlen((char *)name);
fc_gs_cthdr_build(cthdr, s_id, GS_GID_FT);
- bfa_os_memset(gidft, 0, sizeof(struct fcgs_gidft_req_s));
+ memset(gidft, 0, sizeof(struct fcgs_gidft_req_s));
gidft->fc4_type = fc4_type;
gidft->domain_id = 0;
gidft->area_id = 0;
fc_gs_fchdr_build(fchs, d_id, s_id, 0);
fc_gs_cthdr_build(cthdr, s_id, GS_RPN_ID);
- bfa_os_memset(rpnid, 0, sizeof(struct fcgs_rpnid_req_s));
+ memset(rpnid, 0, sizeof(struct fcgs_rpnid_req_s));
rpnid->port_id = port_id;
rpnid->port_name = port_name;
fc_gs_fchdr_build(fchs, d_id, s_id, 0);
fc_gs_cthdr_build(cthdr, s_id, GS_RNN_ID);
- bfa_os_memset(rnnid, 0, sizeof(struct fcgs_rnnid_req_s));
+ memset(rnnid, 0, sizeof(struct fcgs_rnnid_req_s));
rnnid->port_id = port_id;
rnnid->node_name = node_name;
fc_gs_fchdr_build(fchs, d_id, s_id, 0);
fc_gs_cthdr_build(cthdr, s_id, GS_RCS_ID);
- bfa_os_memset(rcsid, 0, sizeof(struct fcgs_rcsid_req_s));
+ memset(rcsid, 0, sizeof(struct fcgs_rcsid_req_s));
rcsid->port_id = port_id;
rcsid->cos = cos;
fc_gs_fchdr_build(fchs, d_id, s_id, 0);
fc_gs_cthdr_build(cthdr, s_id, GS_RPT_ID);
- bfa_os_memset(rptid, 0, sizeof(struct fcgs_rptid_req_s));
+ memset(rptid, 0, sizeof(struct fcgs_rptid_req_s));
rptid->port_id = port_id;
rptid->port_type = port_type;
fc_gs_fchdr_build(fchs, d_id, s_id, 0);
fc_gs_cthdr_build(cthdr, s_id, GS_GA_NXT);
- bfa_os_memset(ganxt, 0, sizeof(struct fcgs_ganxt_req_s));
+ memset(ganxt, 0, sizeof(struct fcgs_ganxt_req_s));
ganxt->port_id = port_id;
return sizeof(struct ct_hdr_s) + sizeof(struct fcgs_ganxt_req_s);
index = fc4_type >> 5;
type_value = 1 << (fc4_type % 32);
- ptr[index] = bfa_os_htonl(type_value);
+ ptr[index] = cpu_to_be32(type_value);
}
fc_gs_ms_cthdr_build(cthdr, s_id, GS_FC_GMAL_CMD,
CT_GSSUBTYPE_CFGSERVER);
- bfa_os_memset(gmal, 0, sizeof(fcgs_gmal_req_t));
+ memset(gmal, 0, sizeof(fcgs_gmal_req_t));
gmal->wwn = wwn;
return sizeof(struct ct_hdr_s) + sizeof(fcgs_gmal_req_t);
fc_gs_ms_cthdr_build(cthdr, s_id, GS_FC_GFN_CMD,
CT_GSSUBTYPE_CFGSERVER);
- bfa_os_memset(gfn, 0, sizeof(fcgs_gfn_req_t));
+ memset(gfn, 0, sizeof(fcgs_gfn_req_t));
gfn->wwn = wwn;
return sizeof(struct ct_hdr_s) + sizeof(fcgs_gfn_req_t);
(__l->__stats += __r->__stats)
-/**
+/*
* BFA ITNIM Related definitions
*/
static void bfa_itnim_update_del_itn_stats(struct bfa_itnim_s *itnim);
} \
} while (0)
-/**
+/*
* bfa_itnim_sm BFA itnim state machine
*/
BFA_ITNIM_SM_QRESUME = 9, /* queue space available */
};
-/**
+/*
* BFA IOIM related definitions
*/
#define bfa_ioim_move_to_comp_q(__ioim) do { \
if ((__fcpim)->profile_start) \
(__fcpim)->profile_start(__ioim); \
} while (0)
-/**
+/*
* hal_ioim_sm
*/
-/**
+/*
* IO state machine events
*/
enum bfa_ioim_event {
};
-/**
+/*
* BFA TSKIM related definitions
*/
-/**
+/*
* task management completion handling
*/
#define bfa_tskim_qcomp(__tskim, __cbfn) do { \
BFA_TSKIM_SM_CLEANUP_DONE = 9, /* TM abort completion */
};
-/**
+/*
* forward declaration for BFA ITNIM functions
*/
static void bfa_itnim_iocdisable_cleanup(struct bfa_itnim_s *itnim);
static void bfa_itnim_iotov_stop(struct bfa_itnim_s *itnim);
static void bfa_itnim_iotov_delete(struct bfa_itnim_s *itnim);
-/**
+/*
* forward declaration of ITNIM state machine
*/
static void bfa_itnim_sm_uninit(struct bfa_itnim_s *itnim,
static void bfa_itnim_sm_deleting_qfull(struct bfa_itnim_s *itnim,
enum bfa_itnim_event event);
-/**
+/*
* forward declaration for BFA IOIM functions
*/
static bfa_boolean_t bfa_ioim_send_ioreq(struct bfa_ioim_s *ioim);
static bfa_boolean_t bfa_ioim_is_abortable(struct bfa_ioim_s *ioim);
-/**
+/*
* forward declaration of BFA IO state machine
*/
static void bfa_ioim_sm_uninit(struct bfa_ioim_s *ioim,
static void bfa_ioim_sm_cmnd_retry(struct bfa_ioim_s *ioim,
enum bfa_ioim_event event);
-/**
+/*
* forward declaration for BFA TSKIM functions
*/
static void __bfa_cb_tskim_done(void *cbarg, bfa_boolean_t complete);
static void bfa_tskim_iocdisable_ios(struct bfa_tskim_s *tskim);
-/**
+/*
* forward declaration of BFA TSKIM state machine
*/
static void bfa_tskim_sm_uninit(struct bfa_tskim_s *tskim,
static void bfa_tskim_sm_hcb(struct bfa_tskim_s *tskim,
enum bfa_tskim_event event);
-/**
+/*
* hal_fcpim_mod BFA FCP Initiator Mode module
*/
-/**
+/*
* Compute and return memory needed by FCP(im) module.
*/
static void
{
bfa_itnim_meminfo(cfg, km_len, dm_len);
- /**
+ /*
* IO memory
*/
if (cfg->fwcfg.num_ioim_reqs < BFA_IOIM_MIN)
*dm_len += cfg->fwcfg.num_ioim_reqs * BFI_IOIM_SNSLEN;
- /**
+ /*
* task management command memory
*/
if (cfg->fwcfg.num_tskim_reqs < BFA_TSKIM_MIN)
struct bfa_itnim_s *itnim;
/* accumulate IO stats from itnim */
- bfa_os_memset(stats, 0, sizeof(struct bfa_itnim_iostats_s));
+ memset(stats, 0, sizeof(struct bfa_itnim_iostats_s));
list_for_each_safe(qe, qen, &fcpim->itnim_q) {
itnim = (struct bfa_itnim_s *) qe;
if (itnim->rport->rport_info.lp_tag != lp_tag)
struct bfa_itnim_s *itnim;
/* accumulate IO stats from itnim */
- bfa_os_memset(modstats, 0, sizeof(struct bfa_itnim_iostats_s));
+ memset(modstats, 0, sizeof(struct bfa_itnim_iostats_s));
list_for_each_safe(qe, qen, &fcpim->itnim_q) {
itnim = (struct bfa_itnim_s *) qe;
bfa_fcpim_add_stats(modstats, &(itnim->stats));
itnim = (struct bfa_itnim_s *) qe;
bfa_itnim_clear_stats(itnim);
}
- bfa_os_memset(&fcpim->del_itn_stats, 0,
+ memset(&fcpim->del_itn_stats, 0,
sizeof(struct bfa_fcpim_del_itn_stats_s));
return BFA_STATUS_OK;
-/**
+/*
* BFA ITNIM module state machine functions
*/
-/**
+/*
* Beginning/unallocated state - no events expected.
*/
static void
}
}
-/**
+/*
* Beginning state, only online event expected.
*/
static void
}
}
-/**
+/*
* Waiting for itnim create response from firmware.
*/
static void
}
}
-/**
+/*
* Waiting for itnim create response from firmware, a delete is pending.
*/
static void
}
}
-/**
+/*
* Online state - normal parking state.
*/
static void
}
}
-/**
+/*
* Second level error recovery need.
*/
static void
}
}
-/**
+/*
* Going offline. Waiting for active IO cleanup.
*/
static void
}
}
-/**
+/*
* Deleting itnim. Waiting for active IO cleanup.
*/
static void
}
}
-/**
+/*
* Rport offline. Fimrware itnim is being deleted - awaiting f/w response.
*/
static void
}
}
-/**
+/*
* Offline state.
*/
static void
}
}
-/**
+/*
* IOC h/w failed state.
*/
static void
}
}
-/**
+/*
* Itnim is deleted, waiting for firmware response to delete.
*/
static void
}
}
-/**
+/*
* Initiate cleanup of all IOs on an IOC failure.
*/
static void
bfa_ioim_iocdisable(ioim);
}
- /**
+ /*
* For IO request in pending queue, we pretend an early timeout.
*/
list_for_each_safe(qe, qen, &itnim->pending_q) {
}
}
-/**
+/*
* IO cleanup completion
*/
static void
bfa_sm_send_event(itnim, BFA_ITNIM_SM_CLEANUP);
}
-/**
+/*
* Initiate cleanup of all IOs.
*/
static void
list_for_each_safe(qe, qen, &itnim->io_q) {
ioim = (struct bfa_ioim_s *) qe;
- /**
+ /*
* Move IO to a cleanup queue from active queue so that a later
* TM will not pickup this IO.
*/
bfa_cb_itnim_sler(itnim->ditn);
}
-/**
+/*
* Call to resume any I/O requests waiting for room in request queue.
*/
static void
-/**
+/*
* bfa_itnim_public
*/
bfa_itnim_meminfo(struct bfa_iocfc_cfg_s *cfg, u32 *km_len,
u32 *dm_len)
{
- /**
+ /*
* ITN memory
*/
*km_len += cfg->fwcfg.num_rports * sizeof(struct bfa_itnim_s);
fcpim->itnim_arr = itnim;
for (i = 0; i < fcpim->num_itnims; i++, itnim++) {
- bfa_os_memset(itnim, 0, sizeof(struct bfa_itnim_s));
+ memset(itnim, 0, sizeof(struct bfa_itnim_s));
itnim->bfa = bfa;
itnim->fcpim = fcpim;
itnim->reqq = BFA_REQQ_QOS_LO;
itnim->msg_no++;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(itnim->bfa, itnim->reqq);
m->msg_no = itnim->msg_no;
bfa_stats(itnim, fw_create);
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(itnim->bfa, itnim->reqq);
{
struct bfi_itnim_delete_req_s *m;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(itnim->bfa, itnim->reqq);
m->fw_handle = itnim->rport->fw_handle;
bfa_stats(itnim, fw_delete);
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(itnim->bfa, itnim->reqq);
return BFA_TRUE;
}
-/**
+/*
* Cleanup all pending failed inflight requests.
*/
static void
}
}
-/**
+/*
* Start all pending IO requests.
*/
static void
bfa_itnim_iotov_stop(itnim);
- /**
+ /*
* Abort all inflight IO requests in the queue
*/
bfa_itnim_delayed_comp(itnim, BFA_FALSE);
- /**
+ /*
* Start all pending IO requests.
*/
while (!list_empty(&itnim->pending_q)) {
}
}
-/**
+/*
* Fail all pending IO requests
*/
static void
{
struct bfa_ioim_s *ioim;
- /**
+ /*
* Fail all inflight IO requests in the queue
*/
bfa_itnim_delayed_comp(itnim, BFA_TRUE);
- /**
+ /*
* Fail any pending IO requests.
*/
while (!list_empty(&itnim->pending_q)) {
}
}
-/**
+/*
* IO TOV timer callback. Fail any pending IO requests.
*/
static void
bfa_cb_itnim_tov(itnim->ditn);
}
-/**
+/*
* Start IO TOV timer for failing back pending IO requests in offline state.
*/
static void
}
}
-/**
+/*
* Stop IO TOV timer.
*/
static void
}
}
-/**
+/*
* Stop IO TOV timer.
*/
static void
-/**
+/*
* bfa_itnim_public
*/
-/**
+/*
* Itnim interrupt processing.
*/
void
-/**
+/*
* bfa_itnim_api
*/
bfa_sm_send_event(itnim, BFA_ITNIM_SM_OFFLINE);
}
-/**
+/*
* Return true if itnim is considered offline for holding off IO request.
* IO is not held if itnim is being deleted.
*/
bfa_itnim_clear_stats(struct bfa_itnim_s *itnim)
{
int j;
- bfa_os_memset(&itnim->stats, 0, sizeof(itnim->stats));
- bfa_os_memset(&itnim->ioprofile, 0, sizeof(itnim->ioprofile));
+ memset(&itnim->stats, 0, sizeof(itnim->stats));
+ memset(&itnim->ioprofile, 0, sizeof(itnim->ioprofile));
for (j = 0; j < BFA_IOBUCKET_MAX; j++)
itnim->ioprofile.io_latency.min[j] = ~0;
}
-/**
+/*
* BFA IO module state machine functions
*/
-/**
+/*
* IO is not started (unallocated).
*/
static void
break;
case BFA_IOIM_SM_ABORT:
- /**
+ /*
* IO in pending queue can get abort requests. Complete abort
* requests immediately.
*/
}
}
-/**
+/*
* IO is waiting for SG pages.
*/
static void
}
}
-/**
+/*
* IO is active.
*/
static void
}
}
-/**
+/*
* IO is retried with new tag.
*/
static void
break;
case BFA_IOIM_SM_ABORT:
- /** in this state IO abort is done.
+ /* in this state IO abort is done.
* Waiting for IO tag resource free.
*/
bfa_sm_set_state(ioim, bfa_ioim_sm_hcb_free);
}
}
-/**
+/*
* IO is being aborted, waiting for completion from firmware.
*/
static void
}
}
-/**
+/*
* IO is being cleaned up (implicit abort), waiting for completion from
* firmware.
*/
break;
case BFA_IOIM_SM_ABORT:
- /**
+ /*
* IO is already being aborted implicitly
*/
ioim->io_cbfn = __bfa_cb_ioim_abort;
break;
case BFA_IOIM_SM_CLEANUP:
- /**
+ /*
* IO can be in cleanup state already due to TM command.
* 2nd cleanup request comes from ITN offline event.
*/
}
}
-/**
+/*
* IO is waiting for room in request CQ
*/
static void
}
}
-/**
+/*
* Active IO is being aborted, waiting for room in request CQ.
*/
static void
}
}
-/**
+/*
* Active IO is being cleaned up, waiting for room in request CQ.
*/
static void
break;
case BFA_IOIM_SM_ABORT:
- /**
+ /*
* IO is alraedy being cleaned up implicitly
*/
ioim->io_cbfn = __bfa_cb_ioim_abort;
}
}
-/**
+/*
* IO bfa callback is pending.
*/
static void
}
}
-/**
+/*
* IO bfa callback is pending. IO resource cannot be freed.
*/
static void
}
}
-/**
+/*
* IO is completed, waiting resource free from firmware.
*/
static void
-/**
+/*
* hal_ioim_private
*/
m = (struct bfi_ioim_rsp_s *) &ioim->iosp->comp_rspmsg;
if (m->io_status == BFI_IOIM_STS_OK) {
- /**
+ /*
* setup sense information, if present
*/
if ((m->scsi_status == SCSI_STATUS_CHECK_CONDITION) &&
snsinfo = ioim->iosp->snsinfo;
}
- /**
+ /*
* setup residue value correctly for normal completions
*/
if (m->resid_flags == FCP_RESID_UNDER) {
- residue = bfa_os_ntohl(m->residue);
+ residue = be32_to_cpu(m->residue);
bfa_stats(ioim->itnim, iocomp_underrun);
}
if (m->resid_flags == FCP_RESID_OVER) {
- residue = bfa_os_ntohl(m->residue);
+ residue = be32_to_cpu(m->residue);
residue = -residue;
bfa_stats(ioim->itnim, iocomp_overrun);
}
bfa_sm_send_event(ioim, BFA_IOIM_SM_SGALLOCED);
}
-/**
+/*
* Send I/O request to firmware.
*/
static bfa_boolean_t
struct scatterlist *sg;
struct scsi_cmnd *cmnd = (struct scsi_cmnd *) ioim->dio;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(ioim->bfa, ioim->reqq);
return BFA_FALSE;
}
- /**
+ /*
* build i/o request message next
*/
- m->io_tag = bfa_os_htons(ioim->iotag);
+ m->io_tag = cpu_to_be16(ioim->iotag);
m->rport_hdl = ioim->itnim->rport->fw_handle;
m->io_timeout = bfa_cb_ioim_get_timeout(ioim->dio);
- /**
+ /*
* build inline IO SG element here
*/
sge = &m->sges[0];
sge->flags = BFI_SGE_PGDLEN;
bfa_sge_to_be(sge);
- /**
+ /*
* set up I/O command parameters
*/
- bfa_os_assign(m->cmnd, cmnd_z0);
+ m->cmnd = cmnd_z0;
m->cmnd.lun = bfa_cb_ioim_get_lun(ioim->dio);
m->cmnd.iodir = bfa_cb_ioim_get_iodir(ioim->dio);
- bfa_os_assign(m->cmnd.cdb,
- *(scsi_cdb_t *)bfa_cb_ioim_get_cdb(ioim->dio));
+ m->cmnd.cdb = *(scsi_cdb_t *)bfa_cb_ioim_get_cdb(ioim->dio);
fcp_dl = bfa_cb_ioim_get_size(ioim->dio);
- m->cmnd.fcp_dl = bfa_os_htonl(fcp_dl);
+ m->cmnd.fcp_dl = cpu_to_be32(fcp_dl);
- /**
+ /*
* set up I/O message header
*/
switch (m->cmnd.iodir) {
m->cmnd.priority = bfa_cb_ioim_get_priority(ioim->dio);
m->cmnd.taskattr = bfa_cb_ioim_get_taskattr(ioim->dio);
- /**
+ /*
* Handle large CDB (>16 bytes).
*/
m->cmnd.addl_cdb_len = (bfa_cb_ioim_get_cdblen(ioim->dio) -
FCP_CMND_CDB_LEN) / sizeof(u32);
if (m->cmnd.addl_cdb_len) {
- bfa_os_memcpy(&m->cmnd.cdb + 1, (scsi_cdb_t *)
+ memcpy(&m->cmnd.cdb + 1, (scsi_cdb_t *)
bfa_cb_ioim_get_cdb(ioim->dio) + 1,
m->cmnd.addl_cdb_len * sizeof(u32));
fcp_cmnd_fcpdl(&m->cmnd) =
- bfa_os_htonl(bfa_cb_ioim_get_size(ioim->dio));
+ cpu_to_be32(bfa_cb_ioim_get_size(ioim->dio));
}
#endif
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(ioim->bfa, ioim->reqq);
return BFA_TRUE;
}
-/**
+/*
* Setup any additional SG pages needed.Inline SG element is setup
* at queuing time.
*/
bfa_assert(ioim->nsges > BFI_SGE_INLINE);
- /**
+ /*
* allocate SG pages needed
*/
nsgpgs = BFA_SGPG_NPAGE(ioim->nsges);
sge->sg_len = sg_dma_len(sg);
pgcumsz += sge->sg_len;
- /**
+ /*
* set flags
*/
if (i < (nsges - 1))
sgpg = (struct bfa_sgpg_s *) bfa_q_next(sgpg);
- /**
+ /*
* set the link element of each page
*/
if (sgeid == ioim->nsges) {
} while (sgeid < ioim->nsges);
}
-/**
+/*
* Send I/O abort request to firmware.
*/
static bfa_boolean_t
struct bfi_ioim_abort_req_s *m;
enum bfi_ioim_h2i msgop;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(ioim->bfa, ioim->reqq);
if (!m)
return BFA_FALSE;
- /**
+ /*
* build i/o request message next
*/
if (ioim->iosp->abort_explicit)
msgop = BFI_IOIM_H2I_IOCLEANUP_REQ;
bfi_h2i_set(m->mh, BFI_MC_IOIM, msgop, bfa_lpuid(ioim->bfa));
- m->io_tag = bfa_os_htons(ioim->iotag);
+ m->io_tag = cpu_to_be16(ioim->iotag);
m->abort_tag = ++ioim->abort_tag;
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(ioim->bfa, ioim->reqq);
return BFA_TRUE;
}
-/**
+/*
* Call to resume any I/O requests waiting for room in request queue.
*/
static void
static void
bfa_ioim_notify_cleanup(struct bfa_ioim_s *ioim)
{
- /**
+ /*
* Move IO from itnim queue to fcpim global queue since itnim will be
* freed.
*/
return BFA_TRUE;
}
-/**
+/*
* or after the link comes back.
*/
void
bfa_ioim_delayed_comp(struct bfa_ioim_s *ioim, bfa_boolean_t iotov)
{
- /**
+ /*
* If path tov timer expired, failback with PATHTOV status - these
* IO requests are not normally retried by IO stack.
*
}
bfa_cb_queue(ioim->bfa, &ioim->hcb_qe, ioim->io_cbfn, ioim);
- /**
+ /*
* Move IO to fcpim global queue since itnim will be
* freed.
*/
-/**
+/*
* hal_ioim_friend
*/
-/**
+/*
* Memory allocation and initialization.
*/
void
u8 *snsinfo;
u32 snsbufsz;
- /**
+ /*
* claim memory first
*/
ioim = (struct bfa_ioim_s *) bfa_meminfo_kva(minfo);
fcpim->ioim_sp_arr = iosp;
bfa_meminfo_kva(minfo) = (u8 *) (iosp + fcpim->num_ioim_reqs);
- /**
+ /*
* Claim DMA memory for per IO sense data.
*/
snsbufsz = fcpim->num_ioim_reqs * BFI_IOIM_SNSLEN;
snsinfo = fcpim->snsbase.kva;
bfa_iocfc_set_snsbase(fcpim->bfa, fcpim->snsbase.pa);
- /**
+ /*
* Initialize ioim free queues
*/
INIT_LIST_HEAD(&fcpim->ioim_free_q);
/*
* initialize IOIM
*/
- bfa_os_memset(ioim, 0, sizeof(struct bfa_ioim_s));
+ memset(ioim, 0, sizeof(struct bfa_ioim_s));
ioim->iotag = i;
ioim->bfa = fcpim->bfa;
ioim->fcpim = fcpim;
}
}
-/**
+/*
* Driver detach time call.
*/
void
u16 iotag;
enum bfa_ioim_event evt = BFA_IOIM_SM_COMP;
- iotag = bfa_os_ntohs(rsp->io_tag);
+ iotag = be16_to_cpu(rsp->io_tag);
ioim = BFA_IOIM_FROM_TAG(fcpim, iotag);
bfa_assert(ioim->iotag == iotag);
bfa_trc(ioim->bfa, rsp->reuse_io_tag);
if (bfa_sm_cmp_state(ioim, bfa_ioim_sm_active))
- bfa_os_assign(ioim->iosp->comp_rspmsg, *m);
+ ioim->iosp->comp_rspmsg = *m;
switch (rsp->io_status) {
case BFI_IOIM_STS_OK:
struct bfa_ioim_s *ioim;
u16 iotag;
- iotag = bfa_os_ntohs(rsp->io_tag);
+ iotag = be16_to_cpu(rsp->io_tag);
ioim = BFA_IOIM_FROM_TAG(fcpim, iotag);
bfa_assert(ioim->iotag == iotag);
void
bfa_ioim_profile_start(struct bfa_ioim_s *ioim)
{
- ioim->start_time = bfa_os_get_clock();
+ ioim->start_time = jiffies;
}
void
{
u32 fcp_dl = bfa_cb_ioim_get_size(ioim->dio);
u32 index = bfa_ioim_get_index(fcp_dl);
- u64 end_time = bfa_os_get_clock();
+ u64 end_time = jiffies;
struct bfa_itnim_latency_s *io_lat =
&(ioim->itnim->ioprofile.io_latency);
u32 val = (u32)(end_time - ioim->start_time);
io_lat->max[index] : val;
io_lat->avg[index] += val;
}
-/**
+/*
* Called by itnim to clean up IO while going offline.
*/
void
bfa_sm_send_event(ioim, BFA_IOIM_SM_CLEANUP);
}
-/**
+/*
* IOC failure handling.
*/
void
bfa_sm_send_event(ioim, BFA_IOIM_SM_HWFAIL);
}
-/**
+/*
* IO offline TOV popped. Fail the pending IO.
*/
void
-/**
+/*
* hal_ioim_api
*/
-/**
+/*
* Allocate IOIM resource for initiator mode I/O request.
*/
struct bfa_ioim_s *
struct bfa_fcpim_mod_s *fcpim = BFA_FCPIM_MOD(bfa);
struct bfa_ioim_s *ioim;
- /**
+ /*
* alocate IOIM resource
*/
bfa_q_deq(&fcpim->ioim_free_q, &ioim);
bfa_ioim_cb_profile_start(ioim->fcpim, ioim);
- /**
+ /*
* Obtain the queue over which this request has to be issued
*/
ioim->reqq = bfa_fcpim_ioredirect_enabled(ioim->bfa) ?
bfa_sm_send_event(ioim, BFA_IOIM_SM_START);
}
-/**
+/*
* Driver I/O abort request.
*/
bfa_status_t
}
-/**
+/*
* BFA TSKIM state machine functions
*/
-/**
+/*
* Task management command beginning state.
*/
static void
bfa_sm_set_state(tskim, bfa_tskim_sm_active);
bfa_tskim_gather_ios(tskim);
- /**
+ /*
* If device is offline, do not send TM on wire. Just cleanup
* any pending IO requests and complete TM request.
*/
}
}
-/**
+/*
* brief
* TM command is active, awaiting completion from firmware to
* cleanup IO requests in TM scope.
}
}
-/**
+/*
* An active TM is being cleaned up since ITN is offline. Awaiting cleanup
* completion event from firmware.
*/
switch (event) {
case BFA_TSKIM_SM_DONE:
- /**
+ /*
* Ignore and wait for ABORT completion from firmware.
*/
break;
break;
case BFA_TSKIM_SM_CLEANUP:
- /**
+ /*
* Ignore, TM command completed on wire.
* Notify TM conmpletion on IO cleanup completion.
*/
}
}
-/**
+/*
* Task management command is waiting for room in request CQ
*/
static void
break;
case BFA_TSKIM_SM_CLEANUP:
- /**
+ /*
* No need to send TM on wire since ITN is offline.
*/
bfa_sm_set_state(tskim, bfa_tskim_sm_iocleanup);
}
}
-/**
+/*
* Task management command is active, awaiting for room in request CQ
* to send clean up request.
*/
switch (event) {
case BFA_TSKIM_SM_DONE:
bfa_reqq_wcancel(&tskim->reqq_wait);
- /**
+ /*
*
* Fall through !!!
*/
}
}
-/**
+/*
* BFA callback is pending
*/
static void
-/**
+/*
* hal_tskim_private
*/
return BFA_FALSE;
}
-/**
+/*
* Gather affected IO requests and task management commands.
*/
static void
INIT_LIST_HEAD(&tskim->io_q);
- /**
+ /*
* Gather any active IO requests first.
*/
list_for_each_safe(qe, qen, &itnim->io_q) {
}
}
- /**
+ /*
* Failback any pending IO requests immediately.
*/
list_for_each_safe(qe, qen, &itnim->pending_q) {
}
}
-/**
+/*
* IO cleanup completion
*/
static void
bfa_sm_send_event(tskim, BFA_TSKIM_SM_IOS_DONE);
}
-/**
+/*
* Gather affected IO requests and task management commands.
*/
static void
bfa_wc_wait(&tskim->wc);
}
-/**
+/*
* Send task management request to firmware.
*/
static bfa_boolean_t
struct bfa_itnim_s *itnim = tskim->itnim;
struct bfi_tskim_req_s *m;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(tskim->bfa, itnim->reqq);
if (!m)
return BFA_FALSE;
- /**
+ /*
* build i/o request message next
*/
bfi_h2i_set(m->mh, BFI_MC_TSKIM, BFI_TSKIM_H2I_TM_REQ,
bfa_lpuid(tskim->bfa));
- m->tsk_tag = bfa_os_htons(tskim->tsk_tag);
+ m->tsk_tag = cpu_to_be16(tskim->tsk_tag);
m->itn_fhdl = tskim->itnim->rport->fw_handle;
m->t_secs = tskim->tsecs;
m->lun = tskim->lun;
m->tm_flags = tskim->tm_cmnd;
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(tskim->bfa, itnim->reqq);
return BFA_TRUE;
}
-/**
+/*
* Send abort request to cleanup an active TM to firmware.
*/
static bfa_boolean_t
struct bfa_itnim_s *itnim = tskim->itnim;
struct bfi_tskim_abortreq_s *m;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(tskim->bfa, itnim->reqq);
if (!m)
return BFA_FALSE;
- /**
+ /*
* build i/o request message next
*/
bfi_h2i_set(m->mh, BFI_MC_TSKIM, BFI_TSKIM_H2I_ABORT_REQ,
bfa_lpuid(tskim->bfa));
- m->tsk_tag = bfa_os_htons(tskim->tsk_tag);
+ m->tsk_tag = cpu_to_be16(tskim->tsk_tag);
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(tskim->bfa, itnim->reqq);
return BFA_TRUE;
}
-/**
+/*
* Call to resume task management cmnd waiting for room in request queue.
*/
static void
bfa_sm_send_event(tskim, BFA_TSKIM_SM_QRESUME);
}
-/**
+/*
* Cleanup IOs associated with a task mangement command on IOC failures.
*/
static void
-/**
+/*
* hal_tskim_friend
*/
-/**
+/*
* Notification on completions from related ioim.
*/
void
bfa_wc_down(&tskim->wc);
}
-/**
+/*
* Handle IOC h/w failure notification from itnim.
*/
void
bfa_sm_send_event(tskim, BFA_TSKIM_SM_HWFAIL);
}
-/**
+/*
* Cleanup TM command and associated IOs as part of ITNIM offline.
*/
void
bfa_sm_send_event(tskim, BFA_TSKIM_SM_CLEANUP);
}
-/**
+/*
* Memory allocation and initialization.
*/
void
/*
* initialize TSKIM
*/
- bfa_os_memset(tskim, 0, sizeof(struct bfa_tskim_s));
+ memset(tskim, 0, sizeof(struct bfa_tskim_s));
tskim->tsk_tag = i;
tskim->bfa = fcpim->bfa;
tskim->fcpim = fcpim;
void
bfa_tskim_detach(struct bfa_fcpim_mod_s *fcpim)
{
- /**
+ /*
* @todo
*/
}
struct bfa_fcpim_mod_s *fcpim = BFA_FCPIM_MOD(bfa);
struct bfi_tskim_rsp_s *rsp = (struct bfi_tskim_rsp_s *) m;
struct bfa_tskim_s *tskim;
- u16 tsk_tag = bfa_os_ntohs(rsp->tsk_tag);
+ u16 tsk_tag = be16_to_cpu(rsp->tsk_tag);
tskim = BFA_TSKIM_FROM_TAG(fcpim, tsk_tag);
bfa_assert(tskim->tsk_tag == tsk_tag);
tskim->tsk_status = rsp->tsk_status;
- /**
+ /*
* Firmware sends BFI_TSKIM_STS_ABORTED status for abort
* requests. All other statuses are for normal completions.
*/
-/**
+/*
* hal_tskim_api
*/
list_add_tail(&tskim->qe, &tskim->fcpim->tskim_free_q);
}
-/**
+/*
* Start a task management command.
*
* @param[in] tskim BFA task management command instance
bfa_fcpim_profile_t profile_start;
};
-/**
+/*
* BFA IO (initiator mode)
*/
struct bfa_ioim_s {
struct bfa_tskim_s *tskim; /* Relevant TM cmd */
};
-/**
+/*
* BFA Task management command (initiator mode)
*/
struct bfa_tskim_s {
};
-/**
+/*
* BFA i-t-n (initiator mode)
*/
struct bfa_itnim_s {
struct bfa_itnim_ioprofile_s *ioprofile);
#define bfa_itnim_get_reqq(__ioim) (((struct bfa_ioim_s *)__ioim)->itnim->reqq)
-/**
+/*
* BFA completion callback for bfa_itnim_online().
*
* @param[in] itnim FCS or driver itnim instance
*/
void bfa_cb_itnim_online(void *itnim);
-/**
+/*
* BFA completion callback for bfa_itnim_offline().
*
* @param[in] itnim FCS or driver itnim instance
void bfa_cb_itnim_tov_begin(void *itnim);
void bfa_cb_itnim_tov(void *itnim);
-/**
+/*
* BFA notification to FCS/driver for second level error recovery.
*
* Atleast one I/O request has timedout and target is unresponsive to
bfa_boolean_t iotov);
-/**
+/*
* I/O completion notification.
*
* @param[in] dio driver IO structure
u8 scsi_status, int sns_len,
u8 *sns_info, s32 residue);
-/**
+/*
* I/O good completion notification.
*
* @param[in] dio driver IO structure
*/
void bfa_cb_ioim_good_comp(void *bfad, struct bfad_ioim_s *dio);
-/**
+/*
* I/O abort completion notification
*
* @param[in] dio driver IO that was aborted
* General Public License for more details.
*/
-/**
+/*
* bfa_fcs.c BFA FCS main
*/
BFA_TRC_FILE(FCS, FCS);
-/**
+/*
* FCS sub-modules
*/
struct bfa_fcs_mod_s {
bfa_fcs_fabric_modexit },
};
-/**
+/*
* fcs_api BFA FCS API
*/
-/**
+/*
* fcs_api BFA FCS API
*/
-/**
+/*
* fcs attach -- called once to initialize data structures at driver attach time
*/
void
}
}
-/**
+/*
* fcs initialization, called once after bfa initialization is complete
*/
void
}
}
-/**
+/*
* Start FCS operations.
*/
void
bfa_fcs_fabric_modstart(fcs);
}
-/**
+/*
* brief
* FCS driver details initialization.
*
bfa_fcs_fabric_psymb_init(&fcs->fabric);
}
-/**
+/*
* brief
* FCS FDMI Driver Parameter Initialization
*
fcs->fdmi_enabled = fdmi_enable;
}
-/**
+/*
* brief
* FCS instance cleanup and exit.
*
bfa_wc_down(&fcs->wc);
}
-/**
+/*
* Fabric module implementation.
*/
u32 rsp_len,
u32 resid_len,
struct fchs_s *rspfchs);
-/**
+/*
* fcs_fabric_sm fabric state machine functions
*/
-/**
+/*
* Fabric state machine events
*/
enum bfa_fcs_fabric_event {
enum bfa_fcs_fabric_event event);
static void bfa_fcs_fabric_sm_deleting(struct bfa_fcs_fabric_s *fabric,
enum bfa_fcs_fabric_event event);
-/**
+/*
* Beginning state before fabric creation.
*/
static void
}
}
-/**
+/*
* Beginning state before fabric creation.
*/
static void
}
}
-/**
+/*
* Link is down, awaiting LINK UP event from port. This is also the
* first state at fabric creation.
*/
}
}
-/**
+/*
* FLOGI is in progress, awaiting FLOGI reply.
*/
static void
}
}
-/**
+/*
* Authentication is in progress, awaiting authentication results.
*/
static void
}
}
-/**
+/*
* Authentication failed
*/
static void
}
}
-/**
+/*
* Port is in loopback mode.
*/
static void
}
}
-/**
+/*
* There is no attached fabric - private loop or NPort-to-NPort topology.
*/
static void
}
}
-/**
+/*
* Fabric is online - normal operating state.
*/
static void
}
}
-/**
+/*
* Exchanging virtual fabric parameters.
*/
static void
}
}
-/**
+/*
* EVFP exchange complete and VFT tagging is enabled.
*/
static void
bfa_trc(fabric->fcs, event);
}
-/**
+/*
* Port is isolated after EVFP exchange due to VF_ID mismatch (N and F).
*/
static void
fabric->event_arg.swp_vfid);
}
-/**
+/*
* Fabric is being deleted, awaiting vport delete completions.
*/
static void
-/**
+/*
* fcs_fabric_private fabric private functions
*/
port_cfg->pwwn = bfa_ioc_get_pwwn(&fabric->fcs->bfa->ioc);
}
-/**
+/*
* Port Symbolic Name Creation for base port.
*/
void
port_cfg->sym_name.symname[BFA_SYMNAME_MAXLEN - 1] = 0;
}
-/**
+/*
* bfa lps login completion callback
*/
void
bfa_trc(fabric->fcs, fabric->is_npiv);
bfa_trc(fabric->fcs, fabric->is_auth);
}
-/**
+/*
* Allocate and send FLOGI.
*/
static void
bfa_fcs_fabric_set_opertype(fabric);
fabric->stats.fabric_onlines++;
- /**
+ /*
* notify online event to base and then virtual ports
*/
bfa_fcs_lport_online(&fabric->bport);
bfa_trc(fabric->fcs, fabric->fabric_name);
fabric->stats.fabric_offlines++;
- /**
+ /*
* notify offline event first to vports and then base port.
*/
list_for_each_safe(qe, qen, &fabric->vport_q) {
bfa_sm_send_event(fabric, BFA_FCS_FABRIC_SM_DELAYED);
}
-/**
+/*
* Delete all vports and wait for vport delete completions.
*/
static void
bfa_sm_send_event(fabric, BFA_FCS_FABRIC_SM_DELCOMP);
}
-/**
+/*
* fcs_fabric_public fabric public functions
*/
-/**
+/*
* Attach time initialization.
*/
void
struct bfa_fcs_fabric_s *fabric;
fabric = &fcs->fabric;
- bfa_os_memset(fabric, 0, sizeof(struct bfa_fcs_fabric_s));
+ memset(fabric, 0, sizeof(struct bfa_fcs_fabric_s));
- /**
+ /*
* Initialize base fabric.
*/
fabric->fcs = fcs;
fabric->lps = bfa_lps_alloc(fcs->bfa);
bfa_assert(fabric->lps);
- /**
+ /*
* Initialize fabric delete completion handler. Fabric deletion is
* complete when the last vport delete is complete.
*/
bfa_trc(fcs, 0);
}
-/**
+/*
* Module cleanup
*/
void
bfa_trc(fcs, 0);
- /**
+ /*
* Cleanup base fabric.
*/
fabric = &fcs->fabric;
bfa_sm_send_event(fabric, BFA_FCS_FABRIC_SM_DELETE);
}
-/**
+/*
* Fabric module start -- kick starts FCS actions
*/
void
bfa_sm_send_event(fabric, BFA_FCS_FABRIC_SM_START);
}
-/**
+/*
* Suspend fabric activity as part of driver suspend.
*/
void
return fabric->oper_type;
}
-/**
+/*
* Link up notification from BFA physical port module.
*/
void
bfa_sm_send_event(fabric, BFA_FCS_FABRIC_SM_LINK_UP);
}
-/**
+/*
* Link down notification from BFA physical port module.
*/
void
bfa_sm_send_event(fabric, BFA_FCS_FABRIC_SM_LINK_DOWN);
}
-/**
+/*
* A child vport is being created in the fabric.
*
* Call from vport module at vport creation. A list of base port and vports
bfa_fcs_fabric_addvport(struct bfa_fcs_fabric_s *fabric,
struct bfa_fcs_vport_s *vport)
{
- /**
+ /*
* - add vport to fabric's vport_q
*/
bfa_trc(fabric->fcs, fabric->vf_id);
bfa_wc_up(&fabric->wc);
}
-/**
+/*
* A child vport is being deleted from fabric.
*
* Vport is being deleted.
bfa_wc_down(&fabric->wc);
}
-/**
+/*
* Base port is deleted.
*/
void
}
-/**
+/*
* Check if fabric is online.
*
* param[in] fabric - Fabric instance. This can be a base fabric or vf.
return bfa_sm_cmp_state(fabric, bfa_fcs_fabric_sm_online);
}
-/**
+/*
* brief
*
*/
return BFA_STATUS_OK;
}
-/**
+/*
* Lookup for a vport withing a fabric given its pwwn
*/
struct bfa_fcs_vport_s *
return NULL;
}
-/**
+/*
* In a given fabric, return the number of lports.
*
* param[in] fabric - Fabric instance. This can be a base fabric or vf.
return oui;
}
-/**
+/*
* Unsolicited frame receive handling.
*/
void
bfa_trc(fabric->fcs, len);
bfa_trc(fabric->fcs, pid);
- /**
+ /*
* Look for our own FLOGI frames being looped back. This means an
* external loopback cable is in place. Our own FLOGI frames are
* sometimes looped back when switch port gets temporarily bypassed.
return;
}
- /**
+ /*
* FLOGI/EVFP exchanges should be consumed by base fabric.
*/
if (fchs->d_id == bfa_os_hton3b(FC_FABRIC_PORT)) {
}
if (fabric->bport.pid == pid) {
- /**
+ /*
* All authentication frames should be routed to auth
*/
bfa_trc(fabric->fcs, els_cmd->els_code);
return;
}
- /**
+ /*
* look for a matching local port ID
*/
list_for_each(qe, &fabric->vport_q) {
bfa_fcs_lport_uf_recv(&fabric->bport, fchs, len);
}
-/**
+/*
* Unsolicited frames to be processed by fabric.
*/
static void
}
}
-/**
+/*
* Process incoming FLOGI
*/
static void
return;
}
- fabric->bb_credit = bfa_os_ntohs(flogi->csp.bbcred);
+ fabric->bb_credit = be16_to_cpu(flogi->csp.bbcred);
bport->port_topo.pn2n.rem_port_wwn = flogi->port_name;
bport->port_topo.pn2n.reply_oxid = fchs->ox_id;
struct fchs_s fchs;
fcxp = bfa_fcs_fcxp_alloc(fabric->fcs);
- /**
+ /*
* Do not expect this failure -- expect remote node to retry
*/
if (!fcxp)
FC_MAX_PDUSZ, 0);
}
-/**
+/*
* Flogi Acc completion callback.
*/
static void
}
}
-/**
- * fcs_vf_api virtual fabrics API
- */
-
-/**
- * Enable VF mode.
- *
- * @param[in] fcs fcs module instance
- * @param[in] vf_id default vf_id of port, FC_VF_ID_NULL
- * to use standard default vf_id of 1.
- *
- * @retval BFA_STATUS_OK vf mode is enabled
- * @retval BFA_STATUS_BUSY Port is active. Port must be disabled
- * before VF mode can be enabled.
- */
-bfa_status_t
-bfa_fcs_vf_mode_enable(struct bfa_fcs_s *fcs, u16 vf_id)
-{
- return BFA_STATUS_OK;
-}
-
-/**
- * Disable VF mode.
- *
- * @param[in] fcs fcs module instance
- *
- * @retval BFA_STATUS_OK vf mode is disabled
- * @retval BFA_STATUS_BUSY VFs are present and being used. All
- * VFs must be deleted before disabling
- * VF mode.
- */
-bfa_status_t
-bfa_fcs_vf_mode_disable(struct bfa_fcs_s *fcs)
-{
- return BFA_STATUS_OK;
-}
-
-/**
- * Create a new VF instance.
- *
- * A new VF is created using the given VF configuration. A VF is identified
- * by VF id. No duplicate VF creation is allowed with the same VF id. Once
- * a VF is created, VF is automatically started after link initialization
- * and EVFP exchange is completed.
- *
- * param[in] vf - FCS vf data structure. Memory is
- * allocated by caller (driver)
- * param[in] fcs - FCS module
- * param[in] vf_cfg - VF configuration
- * param[in] vf_drv - Opaque handle back to the driver's
- * virtual vf structure
- *
- * retval BFA_STATUS_OK VF creation is successful
- * retval BFA_STATUS_FAILED VF creation failed
- * retval BFA_STATUS_EEXIST A VF exists with the given vf_id
- */
-bfa_status_t
-bfa_fcs_vf_create(bfa_fcs_vf_t *vf, struct bfa_fcs_s *fcs, u16 vf_id,
- struct bfa_lport_cfg_s *port_cfg, struct bfad_vf_s *vf_drv)
-{
- bfa_trc(fcs, vf_id);
- return BFA_STATUS_OK;
-}
-
-/**
- * Use this function to delete a BFA VF object. VF object should
- * be stopped before this function call.
- *
- * param[in] vf - pointer to bfa_vf_t.
- *
- * retval BFA_STATUS_OK On vf deletion success
- * retval BFA_STATUS_BUSY VF is not in a stopped state
- * retval BFA_STATUS_INPROGRESS VF deletion in in progress
- */
-bfa_status_t
-bfa_fcs_vf_delete(bfa_fcs_vf_t *vf)
-{
- bfa_trc(vf->fcs, vf->vf_id);
- return BFA_STATUS_OK;
-}
-
-
-/**
- * Returns attributes of the given VF.
- *
- * param[in] vf pointer to bfa_vf_t.
- * param[out] vf_attr vf attributes returned
- *
- * return None
- */
-void
-bfa_fcs_vf_get_attr(bfa_fcs_vf_t *vf, struct bfa_vf_attr_s *vf_attr)
-{
- bfa_trc(vf->fcs, vf->vf_id);
-}
-
-/**
- * Return statistics associated with the given vf.
- *
- * param[in] vf pointer to bfa_vf_t.
- * param[out] vf_stats vf statistics returned
- *
- * @return None
- */
-void
-bfa_fcs_vf_get_stats(bfa_fcs_vf_t *vf, struct bfa_vf_stats_s *vf_stats)
-{
- bfa_os_memcpy(vf_stats, &vf->stats, sizeof(struct bfa_vf_stats_s));
-}
-
-/**
- * clear statistics associated with the given vf.
- *
- * param[in] vf pointer to bfa_vf_t.
- *
- * @return None
- */
-void
-bfa_fcs_vf_clear_stats(bfa_fcs_vf_t *vf)
-{
- bfa_os_memset(&vf->stats, 0, sizeof(struct bfa_vf_stats_s));
-}
-
-/**
+/*
* Returns FCS vf structure for a given vf_id.
*
* param[in] vf_id - VF_ID
return NULL;
}
-/**
- * Return the list of VFs configured.
- *
- * param[in] fcs fcs module instance
- * param[out] vf_ids returned list of vf_ids
- * param[in,out] nvfs in:size of vf_ids array,
- * out:total elements present,
- * actual elements returned is limited by the size
- *
- * return Driver VF structure
- */
-void
-bfa_fcs_vf_list(struct bfa_fcs_s *fcs, u16 *vf_ids, int *nvfs)
-{
- bfa_trc(fcs, *nvfs);
-}
-
-/**
- * Return the list of all VFs visible from fabric.
- *
- * param[in] fcs fcs module instance
- * param[out] vf_ids returned list of vf_ids
- * param[in,out] nvfs in:size of vf_ids array,
- * out:total elements present,
- * actual elements returned is limited by the size
- *
- * return Driver VF structure
- */
-void
-bfa_fcs_vf_list_all(struct bfa_fcs_s *fcs, u16 *vf_ids, int *nvfs)
-{
- bfa_trc(fcs, *nvfs);
-}
-
-/**
- * Return the list of local logical ports present in the given VF.
- *
- * param[in] vf vf for which logical ports are returned
- * param[out] lpwwn returned logical port wwn list
- * param[in,out] nlports in:size of lpwwn list;
- * out:total elements present,
- * actual elements returned is limited by the size
- */
-void
-bfa_fcs_vf_get_ports(bfa_fcs_vf_t *vf, wwn_t lpwwn[], int *nlports)
-{
- struct list_head *qe;
- struct bfa_fcs_vport_s *vport;
- int i;
- struct bfa_fcs_s *fcs;
-
- if (vf == NULL || lpwwn == NULL || *nlports == 0)
- return;
-
- fcs = vf->fcs;
-
- bfa_trc(fcs, vf->vf_id);
- bfa_trc(fcs, (u32) *nlports);
-
- i = 0;
- lpwwn[i++] = vf->bport.port_cfg.pwwn;
-
- list_for_each(qe, &vf->vport_q) {
- if (i >= *nlports)
- break;
-
- vport = (struct bfa_fcs_vport_s *) qe;
- lpwwn[i++] = vport->lport.port_cfg.pwwn;
- }
-
- bfa_trc(fcs, i);
- *nlports = i;
-}
-
-/**
+/*
* BFA FCS PPORT ( physical port)
*/
static void
bfa_fcport_event_register(fcs->bfa, bfa_fcs_port_event_handler, fcs);
}
-/**
+/*
* BFA FCS UF ( Unsolicited Frames)
*/
-/**
+/*
* BFA callback for unsolicited frame receive handler.
*
* @param[in] cbarg callback arg for receive handler
struct fc_vft_s *vft;
struct bfa_fcs_fabric_s *fabric;
- /**
+ /*
* check for VFT header
*/
if (fchs->routing == FC_RTG_EXT_HDR &&
else
fabric = bfa_fcs_vf_lookup(fcs, (u16) vft->vf_id);
- /**
+ /*
* drop frame if vfid is unknown
*/
if (!fabric) {
return;
}
- /**
+ /*
* skip vft header
*/
fchs = (struct fchs_s *) (vft + 1);
#define bfa_fcs_fabric_is_switched(__f) \
((__f)->fab_type == BFA_FCS_FABRIC_SWITCHED)
-/**
+/*
* The design calls for a single implementation of base fabric and vf.
*/
#define bfa_fcs_vf_t struct bfa_fcs_fabric_s
#define bfa_fcs_lport_t struct bfa_fcs_lport_s
-/**
+/*
* Symbolic Name related defines
* Total bytes 255.
* Physical Port's symbolic name 128 bytes.
#define BFA_FCS_PORT_SYMBNAME_OSINFO_SZ 48
#define BFA_FCS_PORT_SYMBNAME_OSPATCH_SZ 16
-/**
+/*
* Get FC port ID for a logical port.
*/
#define bfa_fcs_lport_get_fcid(_lport) ((_lport)->pid)
#define bfa_fcs_lport_get_fabric_ipaddr(_lport) \
((_lport)->fabric->fabric_ip_addr)
-/**
+/*
* bfa fcs port public functions
*/
#define bfa_fcs_vport_get_port(vport) \
((struct bfa_fcs_lport_s *)(&vport->port))
-/**
+/*
* bfa fcs vport public functions
*/
bfa_status_t bfa_fcs_vport_create(struct bfa_fcs_vport_s *vport,
enum bfa_port_speed rpsc_speed;
/* Current Speed from RPSC. O if RPSC fails */
enum bfa_port_speed assigned_speed;
- /**
+ /*
* Speed assigned by the user. will be used if RPSC is
* not supported by the rport.
*/
return rport->bfa_rport;
}
-/**
+/*
* bfa fcs rport API functions
*/
bfa_status_t bfa_fcs_rport_add(struct bfa_fcs_lport_s *port, wwn_t *pwwn,
return itnim->bfa_itnim;
}
-/**
+/*
* bfa fcs FCP Initiator mode API functions
*/
void bfa_fcs_itnim_get_attr(struct bfa_fcs_itnim_s *itnim,
void bfa_fcs_trc_init(struct bfa_fcs_s *fcs, struct bfa_trc_mod_s *trcmod);
void bfa_fcs_start(struct bfa_fcs_s *fcs);
-/**
+/*
* bfa fcs vf public functions
*/
-bfa_status_t bfa_fcs_vf_mode_enable(struct bfa_fcs_s *fcs, u16 vf_id);
-bfa_status_t bfa_fcs_vf_mode_disable(struct bfa_fcs_s *fcs);
-bfa_status_t bfa_fcs_vf_create(bfa_fcs_vf_t *vf, struct bfa_fcs_s *fcs,
- u16 vf_id, struct bfa_lport_cfg_s *port_cfg,
- struct bfad_vf_s *vf_drv);
-bfa_status_t bfa_fcs_vf_delete(bfa_fcs_vf_t *vf);
-void bfa_fcs_vf_list(struct bfa_fcs_s *fcs, u16 *vf_ids, int *nvfs);
-void bfa_fcs_vf_list_all(struct bfa_fcs_s *fcs, u16 *vf_ids, int *nvfs);
-void bfa_fcs_vf_get_attr(bfa_fcs_vf_t *vf, struct bfa_vf_attr_s *vf_attr);
-void bfa_fcs_vf_get_stats(bfa_fcs_vf_t *vf,
- struct bfa_vf_stats_s *vf_stats);
-void bfa_fcs_vf_clear_stats(bfa_fcs_vf_t *vf);
-void bfa_fcs_vf_get_ports(bfa_fcs_vf_t *vf, wwn_t vpwwn[], int *nports);
bfa_fcs_vf_t *bfa_fcs_vf_lookup(struct bfa_fcs_s *fcs, u16 vf_id);
u16 bfa_fcs_fabric_vport_count(struct bfa_fcs_fabric_s *fabric);
void bfa_fcs_uf_attach(struct bfa_fcs_s *fcs);
void bfa_fcs_port_attach(struct bfa_fcs_s *fcs);
-/**
+/*
* BFA FCS callback interfaces
*/
-/**
+/*
* fcb Main fcs callbacks
*/
struct bfad_vport_s;
struct bfad_rport_s;
-/**
+/*
* lport callbacks
*/
struct bfad_port_s *bfa_fcb_lport_new(struct bfad_s *bfad,
struct bfad_vf_s *vf_drv,
struct bfad_vport_s *vp_drv);
-/**
+/*
* vport callbacks
*/
void bfa_fcb_pbc_vport_create(struct bfad_s *bfad, struct bfi_pbc_vport_s);
-/**
+/*
* rport callbacks
*/
bfa_status_t bfa_fcb_rport_alloc(struct bfad_s *bfad,
struct bfa_fcs_rport_s **rport,
struct bfad_rport_s **rport_drv);
-/**
+/*
* itnim callbacks
*/
void bfa_fcb_itnim_alloc(struct bfad_s *bfad, struct bfa_fcs_itnim_s **itnim,
* General Public License for more details.
*/
-/**
+/*
* fcpim.c - FCP initiator mode i-t nexus state machine
*/
bfa_status_t req_status, u32 rsp_len,
u32 resid_len, struct fchs_s *rsp_fchs);
-/**
+/*
* fcs_itnim_sm FCS itnim state machine events
*/
{BFA_SM(bfa_fcs_itnim_sm_initiator), BFA_ITNIM_INITIATIOR},
};
-/**
+/*
* fcs_itnim_sm FCS itnim state machine
*/
-/**
+/*
* itnim_public FCS ITNIM public interfaces
*/
-/**
+/*
* Called by rport when a new rport is created.
*
* @param[in] rport - remote port.
return itnim;
}
-/**
+/*
* Called by rport to delete the instance of FCPIM.
*
* @param[in] rport - remote port.
bfa_sm_send_event(itnim, BFA_FCS_ITNIM_SM_DELETE);
}
-/**
+/*
* Notification from rport that PLOGI is complete to initiate FC-4 session.
*/
void
}
}
-/**
+/*
* Called by rport to handle a remote device offline.
*/
void
bfa_sm_send_event(itnim, BFA_FCS_ITNIM_SM_OFFLINE);
}
-/**
+/*
* Called by rport when remote port is known to be an initiator from
* PRLI received.
*/
bfa_sm_send_event(itnim, BFA_FCS_ITNIM_SM_INITIATOR);
}
-/**
+/*
* Called by rport to check if the itnim is online.
*/
bfa_status_t
}
}
-/**
+/*
* BFA completion callback for bfa_itnim_online().
*/
void
bfa_sm_send_event(itnim, BFA_FCS_ITNIM_SM_HCB_ONLINE);
}
-/**
+/*
* BFA completion callback for bfa_itnim_offline().
*/
void
bfa_sm_send_event(itnim, BFA_FCS_ITNIM_SM_HCB_OFFLINE);
}
-/**
+/*
* Mark the beginning of PATH TOV handling. IO completion callbacks
* are still pending.
*/
bfa_trc(itnim->fcs, itnim->rport->pwwn);
}
-/**
+/*
* Mark the end of PATH TOV handling. All pending IOs are already cleaned up.
*/
void
itnim_drv->state = ITNIM_STATE_TIMEOUT;
}
-/**
+/*
* BFA notification to FCS/driver for second level error recovery.
*
* Atleast one I/O request has timedout and target is unresponsive to
if (itnim == NULL)
return BFA_STATUS_NO_FCPIM_NEXUS;
- bfa_os_memcpy(stats, &itnim->stats, sizeof(struct bfa_itnim_stats_s));
+ memcpy(stats, &itnim->stats, sizeof(struct bfa_itnim_stats_s));
return BFA_STATUS_OK;
}
if (itnim == NULL)
return BFA_STATUS_NO_FCPIM_NEXUS;
- bfa_os_memset(&itnim->stats, 0, sizeof(struct bfa_itnim_stats_s));
+ memset(&itnim->stats, 0, sizeof(struct bfa_itnim_stats_s));
return BFA_STATUS_OK;
}
* General Public License for more details.
*/
-/**
- * bfa_fcs_lport.c BFA FCS port
- */
-
#include "bfa_fcs.h"
#include "bfa_fcbuild.h"
#include "bfa_fc.h"
BFA_TRC_FILE(FCS, PORT);
-/**
- * Forward declarations
- */
-
static void bfa_fcs_lport_send_ls_rjt(struct bfa_fcs_lport_s *port,
struct fchs_s *rx_fchs, u8 reason_code,
u8 reason_code_expl);
bfa_fcs_lport_n2n_offline},
};
-/**
+/*
* fcs_port_sm FCS logical port state machine
*/
}
}
-/**
+/*
* fcs_port_pvt
*/
FC_MAX_PDUSZ, 0);
}
-/**
+/*
* Process incoming plogi from a remote port.
*/
static void
return;
}
- /**
+ /*
* Direct Attach P2P mode : verify address assigned by the r-port.
*/
if ((!bfa_fcs_fabric_is_switched(port->fabric)) &&
port->pid = rx_fchs->d_id;
}
- /**
+ /*
* First, check if we know the device by pwwn.
*/
rport = bfa_fcs_lport_get_rport_by_pwwn(port, plogi->port_name);
if (rport) {
- /**
+ /*
* Direct Attach P2P mode : handle address assigned by r-port.
*/
if ((!bfa_fcs_fabric_is_switched(port->fabric)) &&
return;
}
- /**
+ /*
* Next, lookup rport by PID.
*/
rport = bfa_fcs_lport_get_rport_by_pid(port, rx_fchs->s_id);
if (!rport) {
- /**
+ /*
* Inbound PLOGI from a new device.
*/
bfa_fcs_rport_plogi_create(port, rx_fchs, plogi);
return;
}
- /**
+ /*
* Rport is known only by PID.
*/
if (rport->pwwn) {
- /**
+ /*
* This is a different device with the same pid. Old device
* disappeared. Send implicit LOGO to old device.
*/
bfa_assert(rport->pwwn != plogi->port_name);
bfa_fcs_rport_logo_imp(rport);
- /**
+ /*
* Inbound PLOGI from a new device (with old PID).
*/
bfa_fcs_rport_plogi_create(port, rx_fchs, plogi);
return;
}
- /**
+ /*
* PLOGI crossing each other.
*/
bfa_assert(rport->pwwn == WWN_NULL);
bfa_fs_port_get_gen_topo_data(struct bfa_fcs_lport_s *port,
struct fc_rnid_general_topology_data_s *gen_topo_data)
{
- bfa_os_memset(gen_topo_data, 0,
+ memset(gen_topo_data, 0,
sizeof(struct fc_rnid_general_topology_data_s));
- gen_topo_data->asso_type = bfa_os_htonl(RNID_ASSOCIATED_TYPE_HOST);
+ gen_topo_data->asso_type = cpu_to_be32(RNID_ASSOCIATED_TYPE_HOST);
gen_topo_data->phy_port_num = 0; /* @todo */
- gen_topo_data->num_attached_nodes = bfa_os_htonl(1);
+ gen_topo_data->num_attached_nodes = cpu_to_be32(1);
}
static void
-/**
+/*
* fcs_lport_api BFA FCS port API
*/
-/**
+/*
* Module initialization
*/
void
}
-/**
+/*
* Module cleanup
*/
void
bfa_fcs_modexit_comp(fcs);
}
-/**
+/*
* Unsolicited frame receive handling.
*/
void
return;
}
- /**
+ /*
* First, handle ELSs that donot require a login.
*/
/*
bfa_fcs_lport_abts_acc(lport, fchs);
return;
}
- /**
+ /*
* look for a matching remote port ID
*/
rport = bfa_fcs_lport_get_rport_by_pid(lport, pid);
return;
}
- /**
+ /*
* Only handles ELS frames for now.
*/
if (fchs->type != FC_TYPE_ELS) {
}
if (els_cmd->els_code == FC_ELS_LOGO) {
- /**
+ /*
* @todo Handle LOGO frames received.
*/
return;
}
if (els_cmd->els_code == FC_ELS_PRLI) {
- /**
+ /*
* @todo Handle PRLI frames received.
*/
return;
}
- /**
+ /*
* Unhandled ELS frames. Send a LS_RJT.
*/
bfa_fcs_lport_send_ls_rjt(lport, fchs, FC_LS_RJT_RSN_CMD_NOT_SUPP,
}
-/**
+/*
* PID based Lookup for a R-Port in the Port R-Port Queue
*/
struct bfa_fcs_rport_s *
return NULL;
}
-/**
+/*
* PWWN based Lookup for a R-Port in the Port R-Port Queue
*/
struct bfa_fcs_rport_s *
return NULL;
}
-/**
+/*
* NWWN based Lookup for a R-Port in the Port R-Port Queue
*/
struct bfa_fcs_rport_s *
return NULL;
}
-/**
+/*
* Called by rport module when new rports are discovered.
*/
void
port->num_rports++;
}
-/**
+/*
* Called by rport module to when rports are deleted.
*/
void
bfa_sm_send_event(port, BFA_FCS_PORT_SM_DELRPORT);
}
-/**
+/*
* Called by fabric for base port when fabric login is complete.
* Called by vport for virtual ports when FDISC is complete.
*/
bfa_sm_send_event(port, BFA_FCS_PORT_SM_ONLINE);
}
-/**
+/*
* Called by fabric for base port when fabric goes offline.
* Called by vport for virtual ports when virtual port becomes offline.
*/
bfa_sm_send_event(port, BFA_FCS_PORT_SM_OFFLINE);
}
-/**
+/*
* Called by fabric to delete base lport and associated resources.
*
* Called by vport to delete lport and associated resources. Should call
bfa_sm_send_event(port, BFA_FCS_PORT_SM_DELETE);
}
-/**
+/*
* Return TRUE if port is online, else return FALSE
*/
bfa_boolean_t
return bfa_sm_cmp_state(port, bfa_fcs_lport_sm_online);
}
-/**
+/*
* Attach time initialization of logical ports.
*/
void
lport->num_rports = 0;
}
-/**
+/*
* Logical port initialization of base or virtual port.
* Called by fabric for base port or by vport for virtual ports.
*/
struct bfad_s *bfad = (struct bfad_s *)lport->fcs->bfad;
char lpwwn_buf[BFA_STRING_32];
- bfa_os_assign(lport->port_cfg, *port_cfg);
+ lport->port_cfg = *port_cfg;
lport->bfad_port = bfa_fcb_lport_new(lport->fcs->bfad, lport,
lport->port_cfg.roles,
bfa_sm_send_event(lport, BFA_FCS_PORT_SM_CREATE);
}
-/**
+/*
* fcs_lport_api
*/
}
}
-/**
+/*
* bfa_fcs_lport_fab port fab functions
*/
-/**
+/*
* Called by port to initialize fabric services of the base port.
*/
static void
bfa_fcs_lport_ms_init(port);
}
-/**
+/*
* Called by port to notify transition to online state.
*/
static void
bfa_fcs_lport_scn_online(port);
}
-/**
+/*
* Called by port to notify transition to offline state.
*/
static void
bfa_fcs_lport_ms_offline(port);
}
-/**
+/*
* bfa_fcs_lport_n2n functions
*/
-/**
+/*
* Called by fcs/port to initialize N2N topology.
*/
static void
{
}
-/**
+/*
* Called by fcs/port to notify transition to online state.
*/
static void
((void *)&pcfg->pwwn, (void *)&n2n_port->rem_port_wwn,
sizeof(wwn_t)) > 0) {
port->pid = N2N_LOCAL_PID;
- /**
+ /*
* First, check if we know the device by pwwn.
*/
rport = bfa_fcs_lport_get_rport_by_pwwn(port,
}
}
-/**
+/*
* Called by fcs/port to notify transition to offline state.
*/
static void
struct bfa_fcs_fdmi_hba_attr_s *hba_attr);
static void bfa_fcs_fdmi_get_portattr(struct bfa_fcs_lport_fdmi_s *fdmi,
struct bfa_fcs_fdmi_port_attr_s *port_attr);
-/**
+/*
* fcs_fdmi_sm FCS FDMI state machine
*/
-/**
+/*
* FDMI State Machine events
*/
enum port_fdmi_event {
static void bfa_fcs_lport_fdmi_sm_disabled(
struct bfa_fcs_lport_fdmi_s *fdmi,
enum port_fdmi_event event);
-/**
+/*
* Start in offline state - awaiting MS to send start.
*/
static void
bfa_sm_fault(port->fcs, event);
}
}
-/**
+/*
* FDMI is disabled state.
*/
static void
/* No op State. It can only be enabled at Driver Init. */
}
-/**
+/*
* RHBA : Register HBA Attributes.
*/
static void
fdmi->fcxp = fcxp;
pyld = bfa_fcxp_get_reqbuf(fcxp);
- bfa_os_memset(pyld, 0, FC_MAX_PDUSZ);
+ memset(pyld, 0, FC_MAX_PDUSZ);
len = fc_fdmi_reqhdr_build(&fchs, pyld, bfa_fcs_lport_get_fcid(port),
FDMI_RHBA);
bfa_fcs_fdmi_get_hbaattr(fdmi, fcs_hba_attr);
rhba->hba_id = bfa_fcs_lport_get_pwwn(port);
- rhba->port_list.num_ports = bfa_os_htonl(1);
+ rhba->port_list.num_ports = cpu_to_be32(1);
rhba->port_list.port_entry = bfa_fcs_lport_get_pwwn(port);
len = sizeof(rhba->hba_id) + sizeof(rhba->port_list);
* Node Name
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_HBA_ATTRIB_NODENAME);
+ attr->type = cpu_to_be16(FDMI_HBA_ATTRIB_NODENAME);
attr->len = sizeof(wwn_t);
memcpy(attr->value, &bfa_fcs_lport_get_nwwn(port), attr->len);
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
count++;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
* Manufacturer
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_HBA_ATTRIB_MANUFACTURER);
+ attr->type = cpu_to_be16(FDMI_HBA_ATTRIB_MANUFACTURER);
attr->len = (u16) strlen(fcs_hba_attr->manufacturer);
memcpy(attr->value, fcs_hba_attr->manufacturer, attr->len);
- attr->len = fc_roundup(attr->len, sizeof(u32)); /* variable
- *fields need
- *to be 4 byte
- *aligned */
+ attr->len = fc_roundup(attr->len, sizeof(u32));
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
count++;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
* Serial Number
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_HBA_ATTRIB_SERIALNUM);
+ attr->type = cpu_to_be16(FDMI_HBA_ATTRIB_SERIALNUM);
attr->len = (u16) strlen(fcs_hba_attr->serial_num);
memcpy(attr->value, fcs_hba_attr->serial_num, attr->len);
- attr->len = fc_roundup(attr->len, sizeof(u32)); /* variable
- *fields need
- *to be 4 byte
- *aligned */
+ attr->len = fc_roundup(attr->len, sizeof(u32));
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
count++;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
* Model
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_HBA_ATTRIB_MODEL);
+ attr->type = cpu_to_be16(FDMI_HBA_ATTRIB_MODEL);
attr->len = (u16) strlen(fcs_hba_attr->model);
memcpy(attr->value, fcs_hba_attr->model, attr->len);
- attr->len = fc_roundup(attr->len, sizeof(u32)); /* variable
- *fields need
- *to be 4 byte
- *aligned */
+ attr->len = fc_roundup(attr->len, sizeof(u32));
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
count++;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
* Model Desc
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_HBA_ATTRIB_MODEL_DESC);
+ attr->type = cpu_to_be16(FDMI_HBA_ATTRIB_MODEL_DESC);
attr->len = (u16) strlen(fcs_hba_attr->model_desc);
memcpy(attr->value, fcs_hba_attr->model_desc, attr->len);
- attr->len = fc_roundup(attr->len, sizeof(u32)); /* variable
- *fields need
- *to be 4 byte
- *aligned */
+ attr->len = fc_roundup(attr->len, sizeof(u32));
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
count++;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
*/
if (fcs_hba_attr->hw_version[0] != '\0') {
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_HBA_ATTRIB_HW_VERSION);
+ attr->type = cpu_to_be16(FDMI_HBA_ATTRIB_HW_VERSION);
attr->len = (u16) strlen(fcs_hba_attr->hw_version);
memcpy(attr->value, fcs_hba_attr->hw_version, attr->len);
- attr->len = fc_roundup(attr->len, sizeof(u32)); /* variable
- *fields need
- *to be 4 byte
- *aligned */
+ attr->len = fc_roundup(attr->len, sizeof(u32));
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
count++;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
}
* Driver Version
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_HBA_ATTRIB_DRIVER_VERSION);
+ attr->type = cpu_to_be16(FDMI_HBA_ATTRIB_DRIVER_VERSION);
attr->len = (u16) strlen(fcs_hba_attr->driver_version);
memcpy(attr->value, fcs_hba_attr->driver_version, attr->len);
- attr->len = fc_roundup(attr->len, sizeof(u32)); /* variable
- *fields need
- *to be 4 byte
- *aligned */
+ attr->len = fc_roundup(attr->len, sizeof(u32));
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;;
count++;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
*/
if (fcs_hba_attr->option_rom_ver[0] != '\0') {
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_HBA_ATTRIB_ROM_VERSION);
+ attr->type = cpu_to_be16(FDMI_HBA_ATTRIB_ROM_VERSION);
attr->len = (u16) strlen(fcs_hba_attr->option_rom_ver);
memcpy(attr->value, fcs_hba_attr->option_rom_ver, attr->len);
- attr->len = fc_roundup(attr->len, sizeof(u32)); /* variable
- *fields need
- *to be 4 byte
- *aligned */
+ attr->len = fc_roundup(attr->len, sizeof(u32));
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
count++;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
}
* f/w Version = driver version
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_HBA_ATTRIB_FW_VERSION);
+ attr->type = cpu_to_be16(FDMI_HBA_ATTRIB_FW_VERSION);
attr->len = (u16) strlen(fcs_hba_attr->driver_version);
memcpy(attr->value, fcs_hba_attr->driver_version, attr->len);
- attr->len = fc_roundup(attr->len, sizeof(u32)); /* variable
- *fields need
- *to be 4 byte
- *aligned */
+ attr->len = fc_roundup(attr->len, sizeof(u32));
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
count++;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
*/
if (fcs_hba_attr->os_name[0] != '\0') {
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_HBA_ATTRIB_OS_NAME);
+ attr->type = cpu_to_be16(FDMI_HBA_ATTRIB_OS_NAME);
attr->len = (u16) strlen(fcs_hba_attr->os_name);
memcpy(attr->value, fcs_hba_attr->os_name, attr->len);
- attr->len = fc_roundup(attr->len, sizeof(u32)); /* variable
- *fields need
- *to be 4 byte
- *aligned */
+ attr->len = fc_roundup(attr->len, sizeof(u32));
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
count++;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
}
* MAX_CT_PAYLOAD
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_HBA_ATTRIB_MAX_CT);
+ attr->type = cpu_to_be16(FDMI_HBA_ATTRIB_MAX_CT);
attr->len = sizeof(fcs_hba_attr->max_ct_pyld);
memcpy(attr->value, &fcs_hba_attr->max_ct_pyld, attr->len);
len += attr->len;
count++;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
* Update size of payload
*/
- len += ((sizeof(attr->type) +
- sizeof(attr->len)) * count);
+ len += ((sizeof(attr->type) + sizeof(attr->len)) * count);
- rhba->hba_attr_blk.attr_count = bfa_os_htonl(count);
+ rhba->hba_attr_blk.attr_count = cpu_to_be32(count);
return len;
}
}
cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp);
- cthdr->cmd_rsp_code = bfa_os_ntohs(cthdr->cmd_rsp_code);
+ cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code);
if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) {
bfa_sm_send_event(fdmi, FDMISM_EVENT_RSP_OK);
bfa_sm_send_event(fdmi, FDMISM_EVENT_RSP_ERROR);
}
-/**
+/*
* RPRT : Register Port
*/
static void
fdmi->fcxp = fcxp;
pyld = bfa_fcxp_get_reqbuf(fcxp);
- bfa_os_memset(pyld, 0, FC_MAX_PDUSZ);
+ memset(pyld, 0, FC_MAX_PDUSZ);
len = fc_fdmi_reqhdr_build(&fchs, pyld, bfa_fcs_lport_get_fcid(port),
FDMI_RPRT);
bfa_sm_send_event(fdmi, FDMISM_EVENT_RPRT_SENT);
}
-/**
+/*
* This routine builds Port Attribute Block that used in RPA, RPRT commands.
*/
static u16
* FC4 Types
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_PORT_ATTRIB_FC4_TYPES);
+ attr->type = cpu_to_be16(FDMI_PORT_ATTRIB_FC4_TYPES);
attr->len = sizeof(fcs_port_attr.supp_fc4_types);
memcpy(attr->value, fcs_port_attr.supp_fc4_types, attr->len);
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
++count;
attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
* Supported Speed
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_PORT_ATTRIB_SUPP_SPEED);
+ attr->type = cpu_to_be16(FDMI_PORT_ATTRIB_SUPP_SPEED);
attr->len = sizeof(fcs_port_attr.supp_speed);
memcpy(attr->value, &fcs_port_attr.supp_speed, attr->len);
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
++count;
attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
* current Port Speed
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_PORT_ATTRIB_PORT_SPEED);
+ attr->type = cpu_to_be16(FDMI_PORT_ATTRIB_PORT_SPEED);
attr->len = sizeof(fcs_port_attr.curr_speed);
memcpy(attr->value, &fcs_port_attr.curr_speed, attr->len);
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
++count;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
* max frame size
*/
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_PORT_ATTRIB_FRAME_SIZE);
+ attr->type = cpu_to_be16(FDMI_PORT_ATTRIB_FRAME_SIZE);
attr->len = sizeof(fcs_port_attr.max_frm_size);
memcpy(attr->value, &fcs_port_attr.max_frm_size, attr->len);
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
++count;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
/*
*/
if (fcs_port_attr.os_device_name[0] != '\0') {
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_PORT_ATTRIB_DEV_NAME);
+ attr->type = cpu_to_be16(FDMI_PORT_ATTRIB_DEV_NAME);
attr->len = (u16) strlen(fcs_port_attr.os_device_name);
memcpy(attr->value, fcs_port_attr.os_device_name, attr->len);
- attr->len = fc_roundup(attr->len, sizeof(u32)); /* variable
- *fields need
- *to be 4 byte
- *aligned */
+ attr->len = fc_roundup(attr->len, sizeof(u32));
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
++count;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
}
/*
*/
if (fcs_port_attr.host_name[0] != '\0') {
attr = (struct fdmi_attr_s *) curr_ptr;
- attr->type = bfa_os_htons(FDMI_PORT_ATTRIB_HOST_NAME);
+ attr->type = cpu_to_be16(FDMI_PORT_ATTRIB_HOST_NAME);
attr->len = (u16) strlen(fcs_port_attr.host_name);
memcpy(attr->value, fcs_port_attr.host_name, attr->len);
- attr->len = fc_roundup(attr->len, sizeof(u32)); /* variable
- *fields need
- *to be 4 byte
- *aligned */
+ attr->len = fc_roundup(attr->len, sizeof(u32));
curr_ptr += sizeof(attr->type) + sizeof(attr->len) + attr->len;
len += attr->len;
++count;
- attr->len =
- bfa_os_htons(attr->len + sizeof(attr->type) +
+ attr->len = cpu_to_be16(attr->len + sizeof(attr->type) +
sizeof(attr->len));
}
/*
* Update size of payload
*/
- port_attrib->attr_count = bfa_os_htonl(count);
- len += ((sizeof(attr->type) +
- sizeof(attr->len)) * count);
+ port_attrib->attr_count = cpu_to_be32(count);
+ len += ((sizeof(attr->type) + sizeof(attr->len)) * count);
return len;
}
}
cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp);
- cthdr->cmd_rsp_code = bfa_os_ntohs(cthdr->cmd_rsp_code);
+ cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code);
if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) {
bfa_sm_send_event(fdmi, FDMISM_EVENT_RSP_OK);
bfa_sm_send_event(fdmi, FDMISM_EVENT_RSP_ERROR);
}
-/**
+/*
* RPA : Register Port Attributes.
*/
static void
fdmi->fcxp = fcxp;
pyld = bfa_fcxp_get_reqbuf(fcxp);
- bfa_os_memset(pyld, 0, FC_MAX_PDUSZ);
+ memset(pyld, 0, FC_MAX_PDUSZ);
len = fc_fdmi_reqhdr_build(&fchs, pyld, bfa_fcs_lport_get_fcid(port),
FDMI_RPA);
- attr_len =
- bfa_fcs_lport_fdmi_build_rpa_pyld(fdmi,
- (u8 *) ((struct ct_hdr_s *) pyld
- + 1));
+ attr_len = bfa_fcs_lport_fdmi_build_rpa_pyld(fdmi,
+ (u8 *) ((struct ct_hdr_s *) pyld + 1));
bfa_fcxp_send(fcxp, NULL, port->fabric->vf_id, port->lp_tag, BFA_FALSE,
FC_CLASS_3, len + attr_len, &fchs,
}
cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp);
- cthdr->cmd_rsp_code = bfa_os_ntohs(cthdr->cmd_rsp_code);
+ cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code);
if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) {
bfa_sm_send_event(fdmi, FDMISM_EVENT_RSP_OK);
struct bfa_fcs_lport_s *port = fdmi->ms->port;
struct bfa_fcs_driver_info_s *driver_info = &port->fcs->driver_info;
- bfa_os_memset(hba_attr, 0, sizeof(struct bfa_fcs_fdmi_hba_attr_s));
+ memset(hba_attr, 0, sizeof(struct bfa_fcs_fdmi_hba_attr_s));
bfa_ioc_get_adapter_manufacturer(&port->fcs->bfa->ioc,
hba_attr->manufacturer);
sizeof(driver_info->host_os_patch));
}
- hba_attr->max_ct_pyld = bfa_os_htonl(FC_MAX_PDUSZ);
+ hba_attr->max_ct_pyld = cpu_to_be32(FC_MAX_PDUSZ);
}
void
struct bfa_fcs_driver_info_s *driver_info = &port->fcs->driver_info;
struct bfa_port_attr_s pport_attr;
- bfa_os_memset(port_attr, 0, sizeof(struct bfa_fcs_fdmi_port_attr_s));
+ memset(port_attr, 0, sizeof(struct bfa_fcs_fdmi_port_attr_s));
/*
* get pport attributes from hal
/*
* Supported Speeds
*/
- port_attr->supp_speed = bfa_os_htonl(BFA_FCS_FDMI_SUPORTED_SPEEDS);
+ port_attr->supp_speed = cpu_to_be32(BFA_FCS_FDMI_SUPORTED_SPEEDS);
/*
* Current Speed
*/
- port_attr->curr_speed = bfa_os_htonl(pport_attr.speed);
+ port_attr->curr_speed = cpu_to_be32(pport_attr.speed);
/*
* Max PDU Size.
*/
- port_attr->max_frm_size = bfa_os_htonl(FC_MAX_PDUSZ);
+ port_attr->max_frm_size = cpu_to_be32(FC_MAX_PDUSZ);
/*
* OS device Name
u32 rsp_len,
u32 resid_len,
struct fchs_s *rsp_fchs);
-/**
+/*
* fcs_ms_sm FCS MS state machine
*/
-/**
+/*
* MS State Machine events
*/
enum port_ms_event {
enum port_ms_event event);
static void bfa_fcs_lport_ms_sm_online(struct bfa_fcs_lport_ms_s *ms,
enum port_ms_event event);
-/**
+/*
* Start in offline state - awaiting NS to send start.
*/
static void
*/
bfa_fcs_lport_fdmi_online(ms);
- /**
+ /*
* if this is a Vport, go to online state.
*/
if (ms->port->vport) {
bfa_sm_fault(ms->port->fcs, event);
}
}
-/**
+/*
* ms_pvt MS local functions
*/
}
cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp);
- cthdr->cmd_rsp_code = bfa_os_ntohs(cthdr->cmd_rsp_code);
+ cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code);
if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) {
gmal_resp = (struct fcgs_gmal_resp_s *)(cthdr + 1);
- num_entries = bfa_os_ntohl(gmal_resp->ms_len);
+ num_entries = be32_to_cpu(gmal_resp->ms_len);
if (num_entries == 0) {
bfa_sm_send_event(ms, MSSM_EVENT_RSP_ERROR);
return;
bfa_sm_fault(ms->port->fcs, event);
}
}
-/**
+/*
* ms_pvt MS local functions
*/
}
cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp);
- cthdr->cmd_rsp_code = bfa_os_ntohs(cthdr->cmd_rsp_code);
+ cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code);
if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) {
gfn_resp = (wwn_t *)(cthdr + 1);
bfa_sm_send_event(ms, MSSM_EVENT_RSP_ERROR);
}
-/**
+/*
* ms_pvt MS local functions
*/
bfa_sm_send_event(ms, MSSM_EVENT_PORT_FABRIC_RSCN);
}
-/**
+/*
* @page ns_sm_info VPORT NS State Machine
*
* @section ns_sm_interactions VPORT NS State Machine Interactions
u32 *pid_buf, u32 n_pids);
static void bfa_fcs_lport_ns_boot_target_disc(bfa_fcs_lport_t *port);
-/**
+/*
* fcs_ns_sm FCS nameserver interface state machine
*/
-/**
+/*
* VPort NS State Machine events
*/
enum vport_ns_event {
enum vport_ns_event event);
static void bfa_fcs_lport_ns_sm_online(struct bfa_fcs_lport_ns_s *ns,
enum vport_ns_event event);
-/**
+/*
* Start in offline state - awaiting linkup
*/
static void
-/**
+/*
* ns_pvt Nameserver local functions
*/
}
}
-/**
+/*
* Register the symbolic port name.
*/
static void
u8 symbl[256];
u8 *psymbl = &symbl[0];
- bfa_os_memset(symbl, 0, sizeof(symbl));
+ memset(symbl, 0, sizeof(symbl));
bfa_trc(port->fcs, port->port_cfg.pwwn);
* for V-Port, form a Port Symbolic Name
*/
if (port->vport) {
- /**
+ /*
* For Vports, we append the vport's port symbolic name
* to that of the base port.
*/
}
cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp);
- cthdr->cmd_rsp_code = bfa_os_ntohs(cthdr->cmd_rsp_code);
+ cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code);
if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) {
port->stats.ns_rspnid_accepts++;
bfa_sm_send_event(ns, NSSM_EVENT_RSP_ERROR);
}
-/**
+/*
* Register FC4-Types
*/
static void
}
cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp);
- cthdr->cmd_rsp_code = bfa_os_ntohs(cthdr->cmd_rsp_code);
+ cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code);
if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) {
port->stats.ns_rftid_accepts++;
bfa_sm_send_event(ns, NSSM_EVENT_RSP_ERROR);
}
-/**
+/*
* Register FC4-Features : Should be done after RFT_ID
*/
static void
}
cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp);
- cthdr->cmd_rsp_code = bfa_os_ntohs(cthdr->cmd_rsp_code);
+ cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code);
if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) {
port->stats.ns_rffid_accepts++;
} else
bfa_sm_send_event(ns, NSSM_EVENT_RSP_ERROR);
}
-/**
+/*
* Query Fabric for FC4-Types Devices.
*
* TBD : Need to use a local (FCS private) response buffer, since the response
}
cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp);
- cthdr->cmd_rsp_code = bfa_os_ntohs(cthdr->cmd_rsp_code);
+ cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code);
switch (cthdr->cmd_rsp_code) {
}
}
-/**
+/*
* This routine will be called by bfa_timer on timer timeouts.
*
* param[in] port - pointer to bfa_fcs_lport_t.
}
}
-/**
+/*
* fcs_ns_public FCS nameserver public interfaces
*/
}
}
-/**
+/*
* FCS SCN
*/
struct fchs_s *rx_fchs);
static void bfa_fcs_lport_scn_timeout(void *arg);
-/**
+/*
* fcs_scm_sm FCS SCN state machine
*/
-/**
+/*
* VPort SCN State Machine events
*/
enum port_scn_event {
static void bfa_fcs_lport_scn_sm_online(struct bfa_fcs_lport_scn_s *scn,
enum port_scn_event event);
-/**
+/*
* Starting state - awaiting link up.
*/
static void
-/**
+/*
* fcs_scn_private FCS SCN private functions
*/
-/**
+/*
* This routine will be called to send a SCR command.
*/
static void
FC_MAX_PDUSZ, 0);
}
-/**
+/*
* This routine will be called by bfa_timer on timer timeouts.
*
* param[in] vport - pointer to bfa_fcs_lport_t.
-/**
+/*
* fcs_scn_public FCS state change notification public interfaces
*/
bfa_trc(port->fcs, rpid);
- /**
+ /*
* If this is an unknown device, then it just came online.
* Otherwise let rport handle the RSCN event.
*/
bfa_fcs_rport_scn(rport);
}
-/**
+/*
* rscn format based PID comparison
*/
#define __fc_pid_match(__c0, __c1, __fmt) \
int i = 0, j;
num_entries =
- (bfa_os_ntohs(rscn->payldlen) -
+ (be16_to_cpu(rscn->payldlen) -
sizeof(u32)) / sizeof(rscn->event[0]);
bfa_trc(port->fcs, num_entries);
}
}
- /**
- * If any of area, domain or fabric RSCN is received, do a fresh discovery
- * to find new devices.
+ /*
+ * If any of area, domain or fabric RSCN is received, do a fresh
+ * discovery to find new devices.
*/
if (nsquery)
bfa_fcs_lport_ns_query(port);
}
-/**
+/*
* BFA FCS port
*/
-/**
+/*
* fcs_port_api BFA FCS port API
*/
struct bfa_fcs_lport_s *
void
bfa_fcs_lport_clear_stats(struct bfa_fcs_lport_s *fcs_port)
{
- bfa_os_memset(&fcs_port->stats, 0, sizeof(struct bfa_lport_stats_s));
+ memset(&fcs_port->stats, 0, sizeof(struct bfa_lport_stats_s));
}
-/**
+/*
* FCS virtual port state machine
*/
static void bfa_fcs_vport_do_logo(struct bfa_fcs_vport_s *vport);
static void bfa_fcs_vport_free(struct bfa_fcs_vport_s *vport);
-/**
+/*
* fcs_vport_sm FCS virtual port state machine
*/
-/**
+/*
* VPort State Machine events
*/
enum bfa_fcs_vport_event {
{BFA_SM(bfa_fcs_vport_sm_error), BFA_FCS_VPORT_ERROR}
};
-/**
+/*
* Beginning state.
*/
static void
}
}
-/**
+/*
* Created state - a start event is required to start up the state machine.
*/
static void
bfa_sm_set_state(vport, bfa_fcs_vport_sm_fdisc);
bfa_fcs_vport_do_fdisc(vport);
} else {
- /**
+ /*
* Fabric is offline or not NPIV capable, stay in
* offline state.
*/
case BFA_FCS_VPORT_SM_ONLINE:
case BFA_FCS_VPORT_SM_OFFLINE:
- /**
+ /*
* Ignore ONLINE/OFFLINE events from fabric
* till vport is started.
*/
}
}
-/**
+/*
* Offline state - awaiting ONLINE event from fabric SM.
*/
static void
}
-/**
+/*
* FDISC is sent and awaiting reply from fabric.
*/
static void
}
}
-/**
+/*
* FDISC attempt failed - a timer is active to retry FDISC.
*/
static void
}
}
-/**
+/*
* Vport is online (FDISC is complete).
*/
static void
}
}
-/**
+/*
* Vport is being deleted - awaiting lport delete completion to send
* LOGO to fabric.
*/
}
}
-/**
+/*
* Error State.
* This state will be set when the Vport Creation fails due
* to errors like Dup WWN. In this state only operation allowed
}
}
-/**
+/*
* Lport cleanup is in progress since vport is being deleted. Fabric is
* offline, so no LOGO is needed to complete vport deletion.
*/
}
}
-/**
+/*
* LOGO is sent to fabric. Vport delete is in progress. Lport delete cleanup
* is done.
*/
-/**
+/*
* fcs_vport_private FCS virtual port private functions
*/
-/**
+/*
* This routine will be called to send a FDISC command.
*/
static void
}
}
-/**
+/*
* Called to send a logout to the fabric. Used when a V-Port is
* deleted/stopped.
*/
}
-/**
+/*
* This routine will be called by bfa_timer on timer timeouts.
*
* param[in] vport - pointer to bfa_fcs_vport_t.
-/**
+/*
* fcs_vport_public FCS virtual port public interfaces
*/
-/**
+/*
* Online notification from fabric SM.
*/
void
bfa_sm_send_event(vport, BFA_FCS_VPORT_SM_ONLINE);
}
-/**
+/*
* Offline notification from fabric SM.
*/
void
bfa_sm_send_event(vport, BFA_FCS_VPORT_SM_OFFLINE);
}
-/**
+/*
* Cleanup notification from fabric SM on link timer expiry.
*/
void
{
vport->vport_stats.fab_cleanup++;
}
-/**
+/*
* delete notification from fabric SM. To be invoked from within FCS.
*/
void
bfa_sm_send_event(vport, BFA_FCS_VPORT_SM_DELETE);
}
-/**
+/*
* Delete completion callback from associated lport
*/
void
-/**
+/*
* fcs_vport_api Virtual port API
*/
-/**
+/*
* Use this function to instantiate a new FCS vport object. This
* function will not trigger any HW initialization process (which will be
* done in vport_start() call)
return BFA_STATUS_OK;
}
-/**
+/*
* Use this function to instantiate a new FCS PBC vport object. This
* function will not trigger any HW initialization process (which will be
* done in vport_start() call)
return rc;
}
-/**
+/*
* Use this function to findout if this is a pbc vport or not.
*
* @param[in] vport - pointer to bfa_fcs_vport_t.
}
-/**
+/*
* Use this function initialize the vport.
*
* @param[in] vport - pointer to bfa_fcs_vport_t.
return BFA_STATUS_OK;
}
-/**
+/*
* Use this function quiese the vport object. This function will return
* immediately, when the vport is actually stopped, the
* bfa_drv_vport_stop_cb() will be called.
return BFA_STATUS_OK;
}
-/**
+/*
* Use this function to delete a vport object. Fabric object should
* be stopped before this function call.
*
return BFA_STATUS_OK;
}
-/**
+/*
* Use this function to get vport's current status info.
*
* param[in] vport pointer to bfa_fcs_vport_t.
if (vport == NULL || attr == NULL)
return;
- bfa_os_memset(attr, 0, sizeof(struct bfa_vport_attr_s));
+ memset(attr, 0, sizeof(struct bfa_vport_attr_s));
bfa_fcs_lport_get_attr(&vport->lport, &attr->port_attr);
attr->vport_state = bfa_sm_to_state(vport_sm_table, vport->sm);
}
-/**
+/*
* Use this function to get vport's statistics.
*
* param[in] vport pointer to bfa_fcs_vport_t.
*stats = vport->vport_stats;
}
-/**
+/*
* Use this function to clear vport's statistics.
*
* param[in] vport pointer to bfa_fcs_vport_t.
void
bfa_fcs_vport_clr_stats(struct bfa_fcs_vport_s *vport)
{
- bfa_os_memset(&vport->vport_stats, 0, sizeof(struct bfa_vport_stats_s));
+ memset(&vport->vport_stats, 0, sizeof(struct bfa_vport_stats_s));
}
-/**
+/*
* Lookup a virtual port. Excludes base port from lookup.
*/
struct bfa_fcs_vport_s *
return vport;
}
-/**
+/*
* FDISC Response
*/
void
}
}
-/**
+/*
* LOGO response
*/
void
bfa_sm_send_event(vport, BFA_FCS_VPORT_SM_RSP_OK);
}
-/**
+/*
* Received clear virtual link
*/
void
* General Public License for more details.
*/
-/**
+/*
* rport.c Remote port implementation.
*/
static void bfa_fcs_rport_process_adisc(struct bfa_fcs_rport_s *rport,
struct fchs_s *rx_fchs, u16 len);
static void bfa_fcs_rport_send_prlo_acc(struct bfa_fcs_rport_s *rport);
-/**
+/*
* fcs_rport_sm FCS rport state machine events
*/
{BFA_SM(bfa_fcs_rport_sm_nsdisc_sent), BFA_RPORT_NSDISC},
};
-/**
+/*
* Beginning state.
*/
static void
}
}
-/**
+/*
* PLOGI is being sent.
*/
static void
}
}
-/**
+/*
* PLOGI is being sent.
*/
static void
case RPSM_EVENT_PLOGI_RCVD:
case RPSM_EVENT_SCN:
- /**
+ /*
* Ignore, SCN is possibly online notification.
*/
break;
break;
case RPSM_EVENT_HCB_OFFLINE:
- /**
+ /*
* Ignore BFA callback, on a PLOGI receive we call bfa offline.
*/
break;
}
}
-/**
+/*
* PLOGI is sent.
*/
static void
}
}
-/**
+/*
* PLOGI is sent.
*/
static void
}
}
-/**
+/*
* PLOGI is complete. Awaiting BFA rport online callback. FC-4s
* are offline.
*/
break;
case RPSM_EVENT_SCN:
- /**
+ /*
* @todo
* Ignore SCN - PLOGI just completed, FC-4 login should detect
* device failures.
}
}
-/**
+/*
* Rport is ONLINE. FC-4s active.
*/
static void
}
}
-/**
+/*
* An SCN event is received in ONLINE state. NS query is being sent
* prior to ADISC authentication with rport. FC-4s are paused.
*/
break;
case RPSM_EVENT_SCN:
- /**
+ /*
* ignore SCN, wait for response to query itself
*/
break;
}
}
-/**
+/*
* An SCN event is received in ONLINE state. NS query is sent to rport.
* FC-4s are paused.
*/
}
}
-/**
+/*
* An SCN event is received in ONLINE state. ADISC is being sent for
* authenticating with rport. FC-4s are paused.
*/
}
}
-/**
+/*
* An SCN event is received in ONLINE state. ADISC is to rport.
* FC-4s are paused.
*/
break;
case RPSM_EVENT_PLOGI_RCVD:
- /**
+ /*
* Too complex to cleanup FC-4 & rport and then acc to PLOGI.
* At least go offline when a PLOGI is received.
*/
break;
case RPSM_EVENT_SCN:
- /**
+ /*
* already processing RSCN
*/
break;
}
}
-/**
+/*
* Rport has sent LOGO. Awaiting FC-4 offline completion callback.
*/
static void
}
}
-/**
+/*
* LOGO needs to be sent to rport. Awaiting FC-4 offline completion
* callback.
*/
}
}
-/**
+/*
* Rport is going offline. Awaiting FC-4 offline completion callback.
*/
static void
case RPSM_EVENT_LOGO_RCVD:
case RPSM_EVENT_PRLO_RCVD:
case RPSM_EVENT_ADDRESS_CHANGE:
- /**
+ /*
* rport is already going offline.
* SCN - ignore and wait till transitioning to offline state
*/
}
}
-/**
+/*
* Rport is offline. FC-4s are offline. Awaiting BFA rport offline
* callback.
*/
case RPSM_EVENT_SCN:
case RPSM_EVENT_LOGO_RCVD:
case RPSM_EVENT_PRLO_RCVD:
- /**
+ /*
* Ignore, already offline.
*/
break;
}
}
-/**
+/*
* Rport is offline. FC-4s are offline. Awaiting BFA rport offline
* callback to send LOGO accept.
*/
case RPSM_EVENT_LOGO_RCVD:
case RPSM_EVENT_PRLO_RCVD:
- /**
+ /*
* Ignore - already processing a LOGO.
*/
break;
}
}
-/**
+/*
* Rport is being deleted. FC-4s are offline.
* Awaiting BFA rport offline
* callback to send LOGO.
}
}
-/**
+/*
* Rport is being deleted. FC-4s are offline. LOGO is being sent.
*/
static void
}
}
-/**
+/*
* Rport is offline. FC-4s are offline. BFA rport is offline.
* Timer active to delete stale rport.
*/
}
}
-/**
+/*
* Rport address has changed. Nameserver discovery request is being sent.
*/
static void
}
}
-/**
+/*
* Nameserver discovery failed. Waiting for timeout to retry.
*/
static void
}
}
-/**
+/*
* Rport address has changed. Nameserver discovery request is sent.
*/
static void
bfa_fcs_rport_send_prlo_acc(rport);
break;
case RPSM_EVENT_SCN:
- /**
+ /*
* ignore, wait for NS query response
*/
break;
case RPSM_EVENT_LOGO_RCVD:
- /**
+ /*
* Not logged-in yet. Accept LOGO.
*/
bfa_fcs_rport_send_logo_acc(rport);
-/**
+/*
* fcs_rport_private FCS RPORT provate functions
*/
plogi_rsp = (struct fc_logi_s *) BFA_FCXP_RSP_PLD(fcxp);
- /**
+ /*
* Check for failure first.
*/
if (plogi_rsp->els_cmd.els_code != FC_ELS_ACC) {
return;
}
- /**
+ /*
* PLOGI is complete. Make sure this device is not one of the known
* device with a new FC port address.
*/
}
}
- /**
+ /*
* Normal login path -- no evil twins.
*/
rport->stats.plogi_accs++;
bfa_trc(rport->fcs, rport->pwwn);
cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp);
- cthdr->cmd_rsp_code = bfa_os_ntohs(cthdr->cmd_rsp_code);
+ cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code);
if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) {
/* Check if the pid is the same as before. */
bfa_trc(rport->fcs, rport->pwwn);
cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp);
- cthdr->cmd_rsp_code = bfa_os_ntohs(cthdr->cmd_rsp_code);
+ cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code);
if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) {
bfa_sm_send_event(rport, RPSM_EVENT_ACCEPTED);
}
}
-/**
+/*
* Called to send a logout to the rport.
*/
static void
bfa_sm_send_event(rport, RPSM_EVENT_FCXP_SENT);
}
-/**
+/*
* Send ACC for a LOGO received.
*/
static void
FC_CLASS_3, len, &fchs, NULL, NULL, FC_MAX_PDUSZ, 0);
}
-/**
+/*
* brief
* This routine will be called by bfa_timer on timer timeouts.
*
struct bfa_fcs_rport_s *rport;
struct bfad_rport_s *rport_drv;
- /**
+ /*
* allocate rport
*/
if (bfa_fcb_rport_alloc(fcs->bfad, &rport, &rport_drv)
rport->pid = rpid;
rport->pwwn = pwwn;
- /**
+ /*
* allocate BFA rport
*/
rport->bfa_rport = bfa_rport_create(port->fcs->bfa, rport);
return NULL;
}
- /**
+ /*
* allocate FC-4s
*/
bfa_assert(bfa_fcs_lport_is_initiator(port));
{
struct bfa_fcs_lport_s *port = rport->port;
- /**
+ /*
* - delete FC-4s
* - delete BFA rport
* - remove from queue of rports
}
}
-/**
+/*
* Update rport parameters from PLOGI or PLOGI accept.
*/
static void
{
bfa_fcs_lport_t *port = rport->port;
- /**
+ /*
* - port name
* - node name
*/
rport->pwwn = plogi->port_name;
rport->nwwn = plogi->node_name;
- /**
+ /*
* - class of service
*/
rport->fc_cos = 0;
if (plogi->class2.class_valid)
rport->fc_cos |= FC_CLASS_2;
- /**
+ /*
* - CISC
* - MAX receive frame size
*/
rport->cisc = plogi->csp.cisc;
- rport->maxfrsize = bfa_os_ntohs(plogi->class3.rxsz);
+ rport->maxfrsize = be16_to_cpu(plogi->class3.rxsz);
- bfa_trc(port->fcs, bfa_os_ntohs(plogi->csp.bbcred));
+ bfa_trc(port->fcs, be16_to_cpu(plogi->csp.bbcred));
bfa_trc(port->fcs, port->fabric->bb_credit);
- /**
+ /*
* Direct Attach P2P mode :
* This is to handle a bug (233476) in IBM targets in Direct Attach
* Mode. Basically, in FLOGI Accept the target would have
* in PLOGI.
*/
if ((!bfa_fcs_fabric_is_switched(port->fabric)) &&
- (bfa_os_ntohs(plogi->csp.bbcred) < port->fabric->bb_credit)) {
+ (be16_to_cpu(plogi->csp.bbcred) < port->fabric->bb_credit)) {
- bfa_trc(port->fcs, bfa_os_ntohs(plogi->csp.bbcred));
+ bfa_trc(port->fcs, be16_to_cpu(plogi->csp.bbcred));
bfa_trc(port->fcs, port->fabric->bb_credit);
- port->fabric->bb_credit = bfa_os_ntohs(plogi->csp.bbcred);
+ port->fabric->bb_credit = be16_to_cpu(plogi->csp.bbcred);
bfa_fcport_set_tx_bbcredit(port->fcs->bfa,
port->fabric->bb_credit);
}
}
-/**
+/*
* Called to handle LOGO received from an existing remote port.
*/
static void
-/**
+/*
* fcs_rport_public FCS rport public interfaces
*/
-/**
+/*
* Called by bport/vport to create a remote port instance for a discovered
* remote device.
*
return rport;
}
-/**
+/*
* Called to create a rport for which only the wwn is known.
*
* @param[in] port - base port
bfa_sm_send_event(rport, RPSM_EVENT_ADDRESS_DISC);
return rport;
}
-/**
+/*
* Called by bport in private loop topology to indicate that a
* rport has been discovered and plogi has been completed.
*
bfa_sm_send_event(rport, RPSM_EVENT_PLOGI_COMP);
}
-/**
+/*
* Called by bport/vport to handle PLOGI received from a new remote port.
* If an existing rport does a plogi, it will be handled separately.
*/
return 0;
}
-/**
+/*
* Called by bport/vport to handle PLOGI received from an existing
* remote port.
*/
bfa_fcs_rport_plogi(struct bfa_fcs_rport_s *rport, struct fchs_s *rx_fchs,
struct fc_logi_s *plogi)
{
- /**
+ /*
* @todo Handle P2P and initiator-initiator.
*/
rport->reply_oxid = rx_fchs->ox_id;
bfa_trc(rport->fcs, rport->reply_oxid);
- /**
+ /*
* In Switched fabric topology,
* PLOGI to each other. If our pwwn is smaller, ignore it,
* if it is not a well known address.
bfa_sm_send_event(rport, RPSM_EVENT_PLOGI_RCVD);
}
-/**
+/*
* Called by bport/vport to delete a remote port instance.
*
* Rport delete is called under the following conditions:
bfa_sm_send_event(rport, RPSM_EVENT_DELETE);
}
-/**
+/*
* Called by bport/vport to when a target goes offline.
*
*/
bfa_sm_send_event(rport, RPSM_EVENT_LOGO_IMP);
}
-/**
+/*
* Called by bport in n2n when a target (attached port) becomes online.
*
*/
{
bfa_sm_send_event(rport, RPSM_EVENT_PLOGI_SEND);
}
-/**
+/*
* Called by bport/vport to notify SCN for the remote port
*/
void
bfa_sm_send_event(rport, RPSM_EVENT_SCN);
}
-/**
+/*
* Called by fcpim to notify that the ITN cleanup is done.
*/
void
bfa_sm_send_event(rport, RPSM_EVENT_FC4_OFFLINE);
}
-/**
+/*
* Called by fcptm to notify that the ITN cleanup is done.
*/
void
bfa_sm_send_event(rport, RPSM_EVENT_FC4_OFFLINE);
}
-/**
+/*
* brief
* This routine BFA callback for bfa_rport_online() call.
*
bfa_sm_send_event(rport, RPSM_EVENT_HCB_ONLINE);
}
-/**
+/*
* brief
* This routine BFA callback for bfa_rport_offline() call.
*
bfa_sm_send_event(rport, RPSM_EVENT_HCB_OFFLINE);
}
-/**
+/*
* brief
* This routine is a static BFA callback when there is a QoS flow_id
* change notification
bfa_trc(rport->fcs, rport->pwwn);
}
-/**
+/*
* brief
* This routine is a static BFA callback when there is a QoS priority
* change notification
bfa_trc(rport->fcs, rport->pwwn);
}
-/**
+/*
* Called to process any unsolicted frames from this remote port
*/
void
bfa_sm_send_event(rport, RPSM_EVENT_LOGO_IMP);
}
-/**
+/*
* Called to process any unsolicted frames from this remote port
*/
void
FC_MAX_PDUSZ, 0);
}
-/**
+/*
* Return state of rport.
*/
int
return bfa_sm_to_state(rport_sm_table, rport->sm);
}
-/**
+/*
* brief
* Called by the Driver to set rport delete/ageout timeout
*
-/**
+/*
* Remote port implementation.
*/
-/**
+/*
* fcs_rport_api FCS rport API.
*/
-/**
+/*
* Direct API to add a target by port wwn. This interface is used, for
* example, by bios when target pwwn is known from boot lun configuration.
*/
return BFA_STATUS_OK;
}
-/**
+/*
* Direct API to remove a target and its associated resources. This
* interface is used, for example, by driver to remove target
* ports from the target list for a VM.
}
-/**
+/*
* Remote device status for display/debug.
*/
void
bfa_fcs_lport_t *port = rport->port;
bfa_port_speed_t rport_speed = rport->rpf.rpsc_speed;
- bfa_os_memset(rport_attr, 0, sizeof(struct bfa_rport_attr_s));
+ memset(rport_attr, 0, sizeof(struct bfa_rport_attr_s));
rport_attr->pid = rport->pid;
rport_attr->pwwn = rport->pwwn;
}
}
-/**
+/*
* Per remote device statistics.
*/
void
void
bfa_fcs_rport_clear_stats(struct bfa_fcs_rport_s *rport)
{
- bfa_os_memset((char *)&rport->stats, 0,
+ memset((char *)&rport->stats, 0,
sizeof(struct bfa_rport_stats_s));
}
-/**
+/*
* Remote port features (RPF) implementation.
*/
static void bfa_fcs_rpf_timeout(void *arg);
-/**
+/*
* fcs_rport_ftrs_sm FCS rport state machine events
*/
bfa_sm_fault(rport->fcs, event);
}
}
-/**
+/*
* Called when Rport is created.
*/
void
bfa_sm_set_state(rpf, bfa_fcs_rpf_sm_uninit);
}
-/**
+/*
* Called when Rport becomes online
*/
void
bfa_sm_send_event(&rport->rpf, RPFSM_EVENT_RPORT_ONLINE);
}
-/**
+/*
* Called when Rport becomes offline
*/
void
rpsc2_acc = (struct fc_rpsc2_acc_s *) BFA_FCXP_RSP_PLD(fcxp);
if (rpsc2_acc->els_cmd == FC_ELS_ACC) {
rport->stats.rpsc_accs++;
- num_ents = bfa_os_ntohs(rpsc2_acc->num_pids);
+ num_ents = be16_to_cpu(rpsc2_acc->num_pids);
bfa_trc(rport->fcs, num_ents);
if (num_ents > 0) {
bfa_assert(rpsc2_acc->port_info[0].pid != rport->pid);
bfa_trc(rport->fcs,
- bfa_os_ntohs(rpsc2_acc->port_info[0].pid));
+ be16_to_cpu(rpsc2_acc->port_info[0].pid));
bfa_trc(rport->fcs,
- bfa_os_ntohs(rpsc2_acc->port_info[0].speed));
+ be16_to_cpu(rpsc2_acc->port_info[0].speed));
bfa_trc(rport->fcs,
- bfa_os_ntohs(rpsc2_acc->port_info[0].index));
+ be16_to_cpu(rpsc2_acc->port_info[0].index));
bfa_trc(rport->fcs,
rpsc2_acc->port_info[0].type);
}
rpf->rpsc_speed = fc_rpsc_operspeed_to_bfa_speed(
- bfa_os_ntohs(rpsc2_acc->port_info[0].speed));
+ be16_to_cpu(rpsc2_acc->port_info[0].speed));
bfa_sm_send_event(rpf, RPFSM_EVENT_RPSC_COMP);
}
bfa_hwcb_reginit(struct bfa_s *bfa)
{
struct bfa_iocfc_regs_s *bfa_regs = &bfa->iocfc.bfa_regs;
- bfa_os_addr_t kva = bfa_ioc_bar0(&bfa->ioc);
+ void __iomem *kva = bfa_ioc_bar0(&bfa->ioc);
int i, q, fn = bfa_ioc_pcifn(&bfa->ioc);
if (fn == 0) {
static void
bfa_hwcb_reqq_ack_msix(struct bfa_s *bfa, int reqq)
{
- bfa_reg_write(bfa->iocfc.bfa_regs.intr_status,
- __HFN_INT_CPE_Q0 << CPE_Q_NUM(bfa_ioc_pcifn(&bfa->ioc), reqq));
+ writel(__HFN_INT_CPE_Q0 << CPE_Q_NUM(bfa_ioc_pcifn(&bfa->ioc), reqq),
+ bfa->iocfc.bfa_regs.intr_status);
}
void
static void
bfa_hwcb_rspq_ack_msix(struct bfa_s *bfa, int rspq)
{
- bfa_reg_write(bfa->iocfc.bfa_regs.intr_status,
- __HFN_INT_RME_Q0 << RME_Q_NUM(bfa_ioc_pcifn(&bfa->ioc), rspq));
+ writel(__HFN_INT_RME_Q0 << RME_Q_NUM(bfa_ioc_pcifn(&bfa->ioc), rspq),
+ bfa->iocfc.bfa_regs.intr_status);
}
void
*num_vecs = __HFN_NUMINTS;
}
-/**
+/*
* No special setup required for crossbow -- vector assignments are implicit.
*/
void
bfa->msix.handler[i] = bfa_msix_lpu_err;
}
-/**
+/*
* Crossbow -- dummy, interrupts are masked
*/
void
{
}
-/**
+/*
* No special enable/disable -- vector assignments are implicit.
*/
void
bfa_hwct_msix_lpu_err_set(struct bfa_s *bfa, bfa_boolean_t msix, int vec)
{
int fn = bfa_ioc_pcifn(&bfa->ioc);
- bfa_os_addr_t kva = bfa_ioc_bar0(&bfa->ioc);
+ void __iomem *kva = bfa_ioc_bar0(&bfa->ioc);
if (msix)
- bfa_reg_write(kva + __ct_msix_err_vec_reg[fn], vec);
+ writel(vec, kva + __ct_msix_err_vec_reg[fn]);
else
- bfa_reg_write(kva + __ct_msix_err_vec_reg[fn], 0);
+ writel(0, kva + __ct_msix_err_vec_reg[fn]);
}
-/**
+/*
* Dummy interrupt handler for handling spurious interrupt during chip-reinit.
*/
static void
bfa_hwct_reginit(struct bfa_s *bfa)
{
struct bfa_iocfc_regs_s *bfa_regs = &bfa->iocfc.bfa_regs;
- bfa_os_addr_t kva = bfa_ioc_bar0(&bfa->ioc);
+ void __iomem *kva = bfa_ioc_bar0(&bfa->ioc);
int i, q, fn = bfa_ioc_pcifn(&bfa->ioc);
if (fn == 0) {
{
u32 r32;
- r32 = bfa_reg_read(bfa->iocfc.bfa_regs.cpe_q_ctrl[reqq]);
- bfa_reg_write(bfa->iocfc.bfa_regs.cpe_q_ctrl[reqq], r32);
+ r32 = readl(bfa->iocfc.bfa_regs.cpe_q_ctrl[reqq]);
+ writel(r32, bfa->iocfc.bfa_regs.cpe_q_ctrl[reqq]);
}
void
{
u32 r32;
- r32 = bfa_reg_read(bfa->iocfc.bfa_regs.rme_q_ctrl[rspq]);
- bfa_reg_write(bfa->iocfc.bfa_regs.rme_q_ctrl[rspq], r32);
+ r32 = readl(bfa->iocfc.bfa_regs.rme_q_ctrl[rspq]);
+ writel(r32, bfa->iocfc.bfa_regs.rme_q_ctrl[rspq]);
}
void
*num_vecs = BFA_MSIX_CT_MAX;
}
-/**
+/*
* Setup MSI-X vector for catapult
*/
void
bfa->msix.handler[i] = bfa_hwct_msix_dummy;
}
-/**
+/*
* Enable MSI-X vectors
*/
void
BFA_TRC_FILE(CNA, IOC);
-/**
+/*
* IOC local definitions
*/
#define BFA_IOC_TOV 3000 /* msecs */
BFA_TRC_MAX * sizeof(struct bfa_trc_s)))
#define BFA_DBG_FWTRC_OFF(_fn) (BFI_IOC_TRC_OFF + BFA_DBG_FWTRC_LEN * (_fn))
-/**
+/*
* Asic specific macros : see bfa_hw_cb.c and bfa_hw_ct.c for details.
*/
#define bfa_ioc_mbox_cmd_pending(__ioc) \
(!list_empty(&((__ioc)->mbox_mod.cmd_q)) || \
- bfa_reg_read((__ioc)->ioc_regs.hfn_mbox_cmd))
+ readl((__ioc)->ioc_regs.hfn_mbox_cmd))
bfa_boolean_t bfa_auto_recover = BFA_TRUE;
static void bfa_ioc_pf_failed(struct bfa_ioc_s *ioc);
static void bfa_ioc_pf_fwmismatch(struct bfa_ioc_s *ioc);
-/**
+/*
* hal_ioc_sm
*/
-/**
+/*
* IOC state machine definitions/declarations
*/
enum ioc_event {
{BFA_SM(bfa_ioc_sm_disabled), BFA_IOC_DISABLED},
};
-/**
+/*
* IOCPF state machine definitions/declarations
*/
static void bfa_iocpf_timeout(void *ioc_arg);
static void bfa_iocpf_sem_timeout(void *ioc_arg);
-/**
+/*
* IOCPF state machine events
*/
enum iocpf_event {
IOCPF_E_TIMEOUT = 11, /* f/w response timeout */
};
-/**
+/*
* IOCPF states
*/
enum bfa_iocpf_state {
{BFA_SM(bfa_iocpf_sm_disabled), BFA_IOCPF_DISABLED},
};
-/**
+/*
* IOC State Machine
*/
-/**
+/*
* Beginning state. IOC uninit state.
*/
{
}
-/**
+/*
* IOC is in uninit state.
*/
static void
bfa_sm_fault(ioc, event);
}
}
-/**
+/*
* Reset entry actions -- initialize state machine
*/
static void
bfa_fsm_set_state(&ioc->iocpf, bfa_iocpf_sm_reset);
}
-/**
+/*
* IOC is in reset state.
*/
static void
bfa_iocpf_enable(ioc);
}
-/**
+/*
* Host IOC function is being enabled, awaiting response from firmware.
* Semaphore is acquired.
*/
bfa_ioc_send_getattr(ioc);
}
-/**
+/*
* IOC configuration in progress. Timer is active.
*/
static void
BFA_LOG(KERN_INFO, bfad, log_level, "IOC disabled\n");
}
-/**
+/*
* IOC is being disabled
*/
static void
}
}
-/**
+/*
* IOC disable completion entry.
*/
static void
ioc->cbfn->enable_cbfn(ioc->bfa, BFA_STATUS_IOC_FAILURE);
}
-/**
+/*
* Hardware initialization failed.
*/
static void
break;
case IOC_E_FAILED:
- /**
+ /*
* Initialization failure during iocpf init retry.
*/
ioc->cbfn->enable_cbfn(ioc->bfa, BFA_STATUS_IOC_FAILURE);
struct bfa_ioc_hbfail_notify_s *notify;
struct bfad_s *bfad = (struct bfad_s *)ioc->bfa->bfad;
- /**
+ /*
* Notify driver and common modules registered for notification.
*/
ioc->cbfn->hbfail_cbfn(ioc->bfa);
"Heart Beat of IOC has failed\n");
}
-/**
+/*
* IOC failure.
*/
static void
switch (event) {
case IOC_E_FAILED:
- /**
+ /*
* Initialization failure during iocpf recovery.
* !!! Fall through !!!
*/
-/**
+/*
* IOCPF State Machine
*/
-/**
+/*
* Reset entry actions -- initialize state machine
*/
static void
iocpf->auto_recover = bfa_auto_recover;
}
-/**
+/*
* Beginning state. IOC is in reset state.
*/
static void
}
}
-/**
+/*
* Semaphore should be acquired for version check.
*/
static void
bfa_ioc_hw_sem_get(iocpf->ioc);
}
-/**
+/*
* Awaiting h/w semaphore to continue with version check.
*/
static void
}
}
-/**
+/*
* Notify enable completion callback.
*/
static void
bfa_iocpf_timer_start(iocpf->ioc);
}
-/**
+/*
* Awaiting firmware version match.
*/
static void
}
}
-/**
+/*
* Request for semaphore.
*/
static void
bfa_ioc_hw_sem_get(iocpf->ioc);
}
-/**
+/*
* Awaiting semaphore for h/w initialzation.
*/
static void
bfa_ioc_reset(iocpf->ioc, BFA_FALSE);
}
-/**
+/*
* Hardware is being initialized. Interrupts are enabled.
* Holding hardware semaphore lock.
*/
bfa_ioc_send_enable(iocpf->ioc);
}
-/**
+/*
* Host IOC function is being enabled, awaiting response from firmware.
* Semaphore is acquired.
*/
case IOCPF_E_TIMEOUT:
iocpf->retry_count++;
if (iocpf->retry_count < BFA_IOC_HWINIT_MAX) {
- bfa_reg_write(ioc->ioc_regs.ioc_fwstate,
- BFI_IOC_UNINIT);
+ writel(BFI_IOC_UNINIT, ioc->ioc_regs.ioc_fwstate);
bfa_fsm_set_state(iocpf, bfa_iocpf_sm_hwinit);
break;
}
bfa_ioc_send_disable(iocpf->ioc);
}
-/**
+/*
* IOC is being disabled
*/
static void
*/
case IOCPF_E_TIMEOUT:
- bfa_reg_write(ioc->ioc_regs.ioc_fwstate, BFI_IOC_FAIL);
+ writel(BFI_IOC_FAIL, ioc->ioc_regs.ioc_fwstate);
bfa_fsm_set_state(iocpf, bfa_iocpf_sm_disabled);
break;
}
}
-/**
+/*
* IOC disable completion entry.
*/
static void
bfa_iocpf_timer_start(iocpf->ioc);
}
-/**
+/*
* Hardware initialization failed.
*/
static void
static void
bfa_iocpf_sm_fail_entry(struct bfa_iocpf_s *iocpf)
{
- /**
+ /*
* Mark IOC as failed in hardware and stop firmware.
*/
bfa_ioc_lpu_stop(iocpf->ioc);
- bfa_reg_write(iocpf->ioc->ioc_regs.ioc_fwstate, BFI_IOC_FAIL);
+ writel(BFI_IOC_FAIL, iocpf->ioc->ioc_regs.ioc_fwstate);
- /**
+ /*
* Notify other functions on HB failure.
*/
bfa_ioc_notify_hbfail(iocpf->ioc);
- /**
+ /*
* Flush any queued up mailbox requests.
*/
bfa_ioc_mbox_hbfail(iocpf->ioc);
bfa_iocpf_recovery_timer_start(iocpf->ioc);
}
-/**
+/*
* IOC is in failed state.
*/
static void
-/**
+/*
* hal_ioc_pvt BFA IOC private functions
*/
ioc->cbfn->disable_cbfn(ioc->bfa);
- /**
+ /*
* Notify common modules registered for notification.
*/
list_for_each(qe, &ioc->hb_notify_q) {
}
bfa_boolean_t
-bfa_ioc_sem_get(bfa_os_addr_t sem_reg)
+bfa_ioc_sem_get(void __iomem *sem_reg)
{
u32 r32;
int cnt = 0;
#define BFA_SEM_SPINCNT 3000
- r32 = bfa_reg_read(sem_reg);
+ r32 = readl(sem_reg);
while (r32 && (cnt < BFA_SEM_SPINCNT)) {
cnt++;
- bfa_os_udelay(2);
- r32 = bfa_reg_read(sem_reg);
+ udelay(2);
+ r32 = readl(sem_reg);
}
if (r32 == 0)
}
void
-bfa_ioc_sem_release(bfa_os_addr_t sem_reg)
+bfa_ioc_sem_release(void __iomem *sem_reg)
{
- bfa_reg_write(sem_reg, 1);
+ writel(1, sem_reg);
}
static void
{
u32 r32;
- /**
+ /*
* First read to the semaphore register will return 0, subsequent reads
* will return 1. Semaphore is released by writing 1 to the register
*/
- r32 = bfa_reg_read(ioc->ioc_regs.ioc_sem_reg);
+ r32 = readl(ioc->ioc_regs.ioc_sem_reg);
if (r32 == 0) {
bfa_fsm_send_event(&ioc->iocpf, IOCPF_E_SEMLOCKED);
return;
void
bfa_ioc_hw_sem_release(struct bfa_ioc_s *ioc)
{
- bfa_reg_write(ioc->ioc_regs.ioc_sem_reg, 1);
+ writel(1, ioc->ioc_regs.ioc_sem_reg);
}
static void
bfa_sem_timer_stop(ioc);
}
-/**
+/*
* Initialize LPU local memory (aka secondary memory / SRAM)
*/
static void
int i;
#define PSS_LMEM_INIT_TIME 10000
- pss_ctl = bfa_reg_read(ioc->ioc_regs.pss_ctl_reg);
+ pss_ctl = readl(ioc->ioc_regs.pss_ctl_reg);
pss_ctl &= ~__PSS_LMEM_RESET;
pss_ctl |= __PSS_LMEM_INIT_EN;
* i2c workaround 12.5khz clock
*/
pss_ctl |= __PSS_I2C_CLK_DIV(3UL);
- bfa_reg_write(ioc->ioc_regs.pss_ctl_reg, pss_ctl);
+ writel(pss_ctl, ioc->ioc_regs.pss_ctl_reg);
- /**
+ /*
* wait for memory initialization to be complete
*/
i = 0;
do {
- pss_ctl = bfa_reg_read(ioc->ioc_regs.pss_ctl_reg);
+ pss_ctl = readl(ioc->ioc_regs.pss_ctl_reg);
i++;
} while (!(pss_ctl & __PSS_LMEM_INIT_DONE) && (i < PSS_LMEM_INIT_TIME));
- /**
+ /*
* If memory initialization is not successful, IOC timeout will catch
* such failures.
*/
bfa_trc(ioc, pss_ctl);
pss_ctl &= ~(__PSS_LMEM_INIT_DONE | __PSS_LMEM_INIT_EN);
- bfa_reg_write(ioc->ioc_regs.pss_ctl_reg, pss_ctl);
+ writel(pss_ctl, ioc->ioc_regs.pss_ctl_reg);
}
static void
{
u32 pss_ctl;
- /**
+ /*
* Take processor out of reset.
*/
- pss_ctl = bfa_reg_read(ioc->ioc_regs.pss_ctl_reg);
+ pss_ctl = readl(ioc->ioc_regs.pss_ctl_reg);
pss_ctl &= ~__PSS_LPU0_RESET;
- bfa_reg_write(ioc->ioc_regs.pss_ctl_reg, pss_ctl);
+ writel(pss_ctl, ioc->ioc_regs.pss_ctl_reg);
}
static void
{
u32 pss_ctl;
- /**
+ /*
* Put processors in reset.
*/
- pss_ctl = bfa_reg_read(ioc->ioc_regs.pss_ctl_reg);
+ pss_ctl = readl(ioc->ioc_regs.pss_ctl_reg);
pss_ctl |= (__PSS_LPU0_RESET | __PSS_LPU1_RESET);
- bfa_reg_write(ioc->ioc_regs.pss_ctl_reg, pss_ctl);
+ writel(pss_ctl, ioc->ioc_regs.pss_ctl_reg);
}
-/**
+/*
* Get driver and firmware versions.
*/
void
pgnum = bfa_ioc_smem_pgnum(ioc, loff);
pgoff = bfa_ioc_smem_pgoff(ioc, loff);
- bfa_reg_write(ioc->ioc_regs.host_page_num_fn, pgnum);
+ writel(pgnum, ioc->ioc_regs.host_page_num_fn);
for (i = 0; i < (sizeof(struct bfi_ioc_image_hdr_s) / sizeof(u32));
i++) {
}
}
-/**
+/*
* Returns TRUE if same.
*/
bfa_boolean_t
return BFA_TRUE;
}
-/**
+/*
* Return true if current running version is valid. Firmware signature and
* execution context (driver/bios) must match.
*/
{
struct bfi_ioc_image_hdr_s fwhdr, *drv_fwhdr;
- /**
+ /*
* If bios/efi boot (flash based) -- return true
*/
if (bfa_ioc_is_bios_optrom(ioc))
return BFA_FALSE;
}
- if (bfa_os_swap32(fwhdr.param) != boot_env) {
+ if (swab32(fwhdr.param) != boot_env) {
bfa_trc(ioc, fwhdr.param);
bfa_trc(ioc, boot_env);
return BFA_FALSE;
return bfa_ioc_fwver_cmp(ioc, &fwhdr);
}
-/**
+/*
* Conditionally flush any pending message from firmware at start.
*/
static void
{
u32 r32;
- r32 = bfa_reg_read(ioc->ioc_regs.lpu_mbox_cmd);
+ r32 = readl(ioc->ioc_regs.lpu_mbox_cmd);
if (r32)
- bfa_reg_write(ioc->ioc_regs.lpu_mbox_cmd, 1);
+ writel(1, ioc->ioc_regs.lpu_mbox_cmd);
}
u32 boot_type;
u32 boot_env;
- ioc_fwstate = bfa_reg_read(ioc->ioc_regs.ioc_fwstate);
+ ioc_fwstate = readl(ioc->ioc_regs.ioc_fwstate);
if (force)
ioc_fwstate = BFI_IOC_UNINIT;
boot_type = BFI_BOOT_TYPE_NORMAL;
boot_env = BFI_BOOT_LOADER_OS;
- /**
+ /*
* Flash based firmware boot BIOS env.
*/
if (bfa_ioc_is_bios_optrom(ioc)) {
boot_env = BFI_BOOT_LOADER_BIOS;
}
- /**
+ /*
* Flash based firmware boot UEFI env.
*/
if (bfa_ioc_is_uefi(ioc)) {
boot_env = BFI_BOOT_LOADER_UEFI;
}
- /**
+ /*
* check if firmware is valid
*/
fwvalid = (ioc_fwstate == BFI_IOC_UNINIT) ?
return;
}
- /**
+ /*
* If hardware initialization is in progress (initialized by other IOC),
* just wait for an initialization completion interrupt.
*/
return;
}
- /**
+ /*
* If IOC function is disabled and firmware version is same,
* just re-enable IOC.
*
if (ioc_fwstate == BFI_IOC_DISABLED ||
(!bfa_ioc_is_bios_optrom(ioc) && ioc_fwstate == BFI_IOC_OP)) {
- /**
+ /*
* When using MSI-X any pending firmware ready event should
* be flushed. Otherwise MSI-X interrupts are not delivered.
*/
return;
}
- /**
+ /*
* Initialize the h/w for any other states.
*/
bfa_ioc_boot(ioc, boot_type, boot_env);
* first write msg to mailbox registers
*/
for (i = 0; i < len / sizeof(u32); i++)
- bfa_reg_write(ioc->ioc_regs.hfn_mbox + i * sizeof(u32),
- bfa_os_wtole(msgp[i]));
+ writel(cpu_to_le32(msgp[i]),
+ ioc->ioc_regs.hfn_mbox + i * sizeof(u32));
for (; i < BFI_IOC_MSGLEN_MAX / sizeof(u32); i++)
- bfa_reg_write(ioc->ioc_regs.hfn_mbox + i * sizeof(u32), 0);
+ writel(0, ioc->ioc_regs.hfn_mbox + i * sizeof(u32));
/*
* write 1 to mailbox CMD to trigger LPU event
*/
- bfa_reg_write(ioc->ioc_regs.hfn_mbox_cmd, 1);
- (void) bfa_reg_read(ioc->ioc_regs.hfn_mbox_cmd);
+ writel(1, ioc->ioc_regs.hfn_mbox_cmd);
+ (void) readl(ioc->ioc_regs.hfn_mbox_cmd);
}
static void
bfa_ioc_portid(ioc));
enable_req.ioc_class = ioc->ioc_mc;
bfa_os_gettimeofday(&tv);
- enable_req.tv_sec = bfa_os_ntohl(tv.tv_sec);
+ enable_req.tv_sec = be32_to_cpu(tv.tv_sec);
bfa_ioc_mbox_send(ioc, &enable_req, sizeof(struct bfi_ioc_ctrl_req_s));
}
struct bfa_ioc_s *ioc = cbarg;
u32 hb_count;
- hb_count = bfa_reg_read(ioc->ioc_regs.heartbeat);
+ hb_count = readl(ioc->ioc_regs.heartbeat);
if (ioc->hb_count == hb_count) {
printk(KERN_CRIT "Firmware heartbeat failure at %d", hb_count);
bfa_ioc_recover(ioc);
static void
bfa_ioc_hb_monitor(struct bfa_ioc_s *ioc)
{
- ioc->hb_count = bfa_reg_read(ioc->ioc_regs.heartbeat);
+ ioc->hb_count = readl(ioc->ioc_regs.heartbeat);
bfa_hb_timer_start(ioc);
}
}
-/**
+/*
* Initiate a full firmware download.
*/
static void
u32 chunkno = 0;
u32 i;
- /**
+ /*
* Initialize LMEM first before code download
*/
bfa_ioc_lmem_init(ioc);
pgnum = bfa_ioc_smem_pgnum(ioc, loff);
pgoff = bfa_ioc_smem_pgoff(ioc, loff);
- bfa_reg_write(ioc->ioc_regs.host_page_num_fn, pgnum);
+ writel(pgnum, ioc->ioc_regs.host_page_num_fn);
for (i = 0; i < bfa_cb_image_get_size(BFA_IOC_FWIMG_TYPE(ioc)); i++) {
BFA_IOC_FLASH_CHUNK_ADDR(chunkno));
}
- /**
+ /*
* write smem
*/
bfa_mem_write(ioc->ioc_regs.smem_page_start, loff,
loff += sizeof(u32);
- /**
+ /*
* handle page offset wrap around
*/
loff = PSS_SMEM_PGOFF(loff);
if (loff == 0) {
pgnum++;
- bfa_reg_write(ioc->ioc_regs.host_page_num_fn,
- pgnum);
+ writel(pgnum, ioc->ioc_regs.host_page_num_fn);
}
}
- bfa_reg_write(ioc->ioc_regs.host_page_num_fn,
- bfa_ioc_smem_pgnum(ioc, 0));
+ writel(bfa_ioc_smem_pgnum(ioc, 0), ioc->ioc_regs.host_page_num_fn);
/*
* Set boot type and boot param at the end.
*/
bfa_mem_write(ioc->ioc_regs.smem_page_start, BFI_BOOT_TYPE_OFF,
- bfa_os_swap32(boot_type));
+ swab32(boot_type));
bfa_mem_write(ioc->ioc_regs.smem_page_start, BFI_BOOT_LOADER_OFF,
- bfa_os_swap32(boot_env));
+ swab32(boot_env));
}
static void
bfa_ioc_hwinit(ioc, force);
}
-/**
+/*
* Update BFA configuration from firmware configuration.
*/
static void
{
struct bfi_ioc_attr_s *attr = ioc->attr;
- attr->adapter_prop = bfa_os_ntohl(attr->adapter_prop);
- attr->card_type = bfa_os_ntohl(attr->card_type);
- attr->maxfrsize = bfa_os_ntohs(attr->maxfrsize);
+ attr->adapter_prop = be32_to_cpu(attr->adapter_prop);
+ attr->card_type = be32_to_cpu(attr->card_type);
+ attr->maxfrsize = be16_to_cpu(attr->maxfrsize);
bfa_fsm_send_event(ioc, IOC_E_FWRSP_GETATTR);
}
-/**
+/*
* Attach time initialization of mbox logic.
*/
static void
}
}
-/**
+/*
* Mbox poll timer -- restarts any pending mailbox requests.
*/
static void
struct bfa_mbox_cmd_s *cmd;
u32 stat;
- /**
+ /*
* If no command pending, do nothing
*/
if (list_empty(&mod->cmd_q))
return;
- /**
+ /*
* If previous command is not yet fetched by firmware, do nothing
*/
- stat = bfa_reg_read(ioc->ioc_regs.hfn_mbox_cmd);
+ stat = readl(ioc->ioc_regs.hfn_mbox_cmd);
if (stat)
return;
- /**
+ /*
* Enqueue command to firmware.
*/
bfa_q_deq(&mod->cmd_q, &cmd);
bfa_ioc_mbox_send(ioc, cmd->msg, sizeof(cmd->msg));
}
-/**
+/*
* Cleanup any pending requests.
*/
static void
bfa_q_deq(&mod->cmd_q, &cmd);
}
-/**
+/*
* Read data from SMEM to host through PCI memmap
*
* @param[in] ioc memory for IOC
return BFA_STATUS_FAILED;
}
- bfa_reg_write(ioc->ioc_regs.host_page_num_fn, pgnum);
+ writel(pgnum, ioc->ioc_regs.host_page_num_fn);
len = sz/sizeof(u32);
bfa_trc(ioc, len);
for (i = 0; i < len; i++) {
r32 = bfa_mem_read(ioc->ioc_regs.smem_page_start, loff);
- buf[i] = bfa_os_ntohl(r32);
+ buf[i] = be32_to_cpu(r32);
loff += sizeof(u32);
- /**
+ /*
* handle page offset wrap around
*/
loff = PSS_SMEM_PGOFF(loff);
if (loff == 0) {
pgnum++;
- bfa_reg_write(ioc->ioc_regs.host_page_num_fn, pgnum);
+ writel(pgnum, ioc->ioc_regs.host_page_num_fn);
}
}
- bfa_reg_write(ioc->ioc_regs.host_page_num_fn,
- bfa_ioc_smem_pgnum(ioc, 0));
+ writel(bfa_ioc_smem_pgnum(ioc, 0), ioc->ioc_regs.host_page_num_fn);
/*
* release semaphore.
*/
return BFA_STATUS_OK;
}
-/**
+/*
* Clear SMEM data from host through PCI memmap
*
* @param[in] ioc memory for IOC
return BFA_STATUS_FAILED;
}
- bfa_reg_write(ioc->ioc_regs.host_page_num_fn, pgnum);
+ writel(pgnum, ioc->ioc_regs.host_page_num_fn);
len = sz/sizeof(u32); /* len in words */
bfa_trc(ioc, len);
bfa_mem_write(ioc->ioc_regs.smem_page_start, loff, 0);
loff += sizeof(u32);
- /**
+ /*
* handle page offset wrap around
*/
loff = PSS_SMEM_PGOFF(loff);
if (loff == 0) {
pgnum++;
- bfa_reg_write(ioc->ioc_regs.host_page_num_fn, pgnum);
+ writel(pgnum, ioc->ioc_regs.host_page_num_fn);
}
}
- bfa_reg_write(ioc->ioc_regs.host_page_num_fn,
- bfa_ioc_smem_pgnum(ioc, 0));
+ writel(bfa_ioc_smem_pgnum(ioc, 0), ioc->ioc_regs.host_page_num_fn);
/*
* release semaphore.
return BFA_STATUS_OK;
}
-/**
+/*
* hal iocpf to ioc interface
*/
static void
bfa_ioc_pf_fwmismatch(struct bfa_ioc_s *ioc)
{
struct bfad_s *bfad = (struct bfad_s *)ioc->bfa->bfad;
- /**
+ /*
* Provide enable completion callback.
*/
ioc->cbfn->enable_cbfn(ioc->bfa, BFA_STATUS_IOC_FAILURE);
-/**
+/*
* hal_ioc_public
*/
return BFA_STATUS_OK;
}
-/**
+/*
* Interface used by diag module to do firmware boot with memory test
* as the entry vector.
*/
void
bfa_ioc_boot(struct bfa_ioc_s *ioc, u32 boot_type, u32 boot_env)
{
- bfa_os_addr_t rb;
+ void __iomem *rb;
bfa_ioc_stats(ioc, ioc_boots);
if (bfa_ioc_pll_init(ioc) != BFA_STATUS_OK)
return;
- /**
+ /*
* Initialize IOC state of all functions on a chip reset.
*/
rb = ioc->pcidev.pci_bar_kva;
if (boot_type == BFI_BOOT_TYPE_MEMTEST) {
- bfa_reg_write((rb + BFA_IOC0_STATE_REG), BFI_IOC_MEMTEST);
- bfa_reg_write((rb + BFA_IOC1_STATE_REG), BFI_IOC_MEMTEST);
+ writel(BFI_IOC_MEMTEST, (rb + BFA_IOC0_STATE_REG));
+ writel(BFI_IOC_MEMTEST, (rb + BFA_IOC1_STATE_REG));
} else {
- bfa_reg_write((rb + BFA_IOC0_STATE_REG), BFI_IOC_INITING);
- bfa_reg_write((rb + BFA_IOC1_STATE_REG), BFI_IOC_INITING);
+ writel(BFI_IOC_INITING, (rb + BFA_IOC0_STATE_REG));
+ writel(BFI_IOC_INITING, (rb + BFA_IOC1_STATE_REG));
}
bfa_ioc_msgflush(ioc);
bfa_ioc_download_fw(ioc, boot_type, boot_env);
- /**
+ /*
* Enable interrupts just before starting LPU
*/
ioc->cbfn->reset_cbfn(ioc->bfa);
bfa_ioc_lpu_start(ioc);
}
-/**
+/*
* Enable/disable IOC failure auto recovery.
*/
void
bfa_boolean_t
bfa_ioc_is_initialized(struct bfa_ioc_s *ioc)
{
- u32 r32 = bfa_reg_read(ioc->ioc_regs.ioc_fwstate);
+ u32 r32 = readl(ioc->ioc_regs.ioc_fwstate);
return ((r32 != BFI_IOC_UNINIT) &&
(r32 != BFI_IOC_INITING) &&
u32 r32;
int i;
- /**
+ /*
* read the MBOX msg
*/
for (i = 0; i < (sizeof(union bfi_ioc_i2h_msg_u) / sizeof(u32));
i++) {
- r32 = bfa_reg_read(ioc->ioc_regs.lpu_mbox +
+ r32 = readl(ioc->ioc_regs.lpu_mbox +
i * sizeof(u32));
- msgp[i] = bfa_os_htonl(r32);
+ msgp[i] = cpu_to_be32(r32);
}
- /**
+ /*
* turn off mailbox interrupt by clearing mailbox status
*/
- bfa_reg_write(ioc->ioc_regs.lpu_mbox_cmd, 1);
- bfa_reg_read(ioc->ioc_regs.lpu_mbox_cmd);
+ writel(1, ioc->ioc_regs.lpu_mbox_cmd);
+ readl(ioc->ioc_regs.lpu_mbox_cmd);
}
void
}
}
-/**
+/*
* IOC attach time initialization and setup.
*
* @param[in] ioc memory for IOC
bfa_fsm_send_event(ioc, IOC_E_RESET);
}
-/**
+/*
* Driver detach time IOC cleanup.
*/
void
bfa_fsm_send_event(ioc, IOC_E_DETACH);
}
-/**
+/*
* Setup IOC PCI properties.
*
* @param[in] pcidev PCI device information for this IOC
ioc->ctdev = bfa_asic_id_ct(ioc->pcidev.device_id);
ioc->cna = ioc->ctdev && !ioc->fcmode;
- /**
+ /*
* Set asic specific interfaces. See bfa_ioc_cb.c and bfa_ioc_ct.c
*/
if (ioc->ctdev)
bfa_ioc_reg_init(ioc);
}
-/**
+/*
* Initialize IOC dma memory
*
* @param[in] dm_kva kernel virtual address of IOC dma memory
void
bfa_ioc_mem_claim(struct bfa_ioc_s *ioc, u8 *dm_kva, u64 dm_pa)
{
- /**
+ /*
* dma memory for firmware attribute
*/
ioc->attr_dma.kva = dm_kva;
ioc->attr = (struct bfi_ioc_attr_s *) dm_kva;
}
-/**
+/*
* Return size of dma memory required.
*/
u32
bfa_fsm_send_event(ioc, IOC_E_DISABLE);
}
-/**
+/*
* Returns memory required for saving firmware trace in case of crash.
* Driver must call this interface to allocate memory required for
* automatic saving of firmware trace. Driver should call
return (auto_recover) ? BFA_DBG_FWTRC_LEN : 0;
}
-/**
+/*
* Initialize memory for saving firmware trace. Driver must initialize
* trace memory before call bfa_ioc_enable().
*/
return PSS_SMEM_PGOFF(fmaddr);
}
-/**
+/*
* Register mailbox message handler functions
*
* @param[in] ioc IOC instance
mod->mbhdlr[mc].cbfn = mcfuncs[mc];
}
-/**
+/*
* Register mailbox message handler function, to be called by common modules
*/
void
mod->mbhdlr[mc].cbarg = cbarg;
}
-/**
+/*
* Queue a mailbox command request to firmware. Waits if mailbox is busy.
* Responsibility of caller to serialize
*
struct bfa_ioc_mbox_mod_s *mod = &ioc->mbox_mod;
u32 stat;
- /**
+ /*
* If a previous command is pending, queue new command
*/
if (!list_empty(&mod->cmd_q)) {
return;
}
- /**
+ /*
* If mailbox is busy, queue command for poll timer
*/
- stat = bfa_reg_read(ioc->ioc_regs.hfn_mbox_cmd);
+ stat = readl(ioc->ioc_regs.hfn_mbox_cmd);
if (stat) {
list_add_tail(&cmd->qe, &mod->cmd_q);
return;
}
- /**
+ /*
* mailbox is free -- queue command to firmware
*/
bfa_ioc_mbox_send(ioc, cmd->msg, sizeof(cmd->msg));
}
-/**
+/*
* Handle mailbox interrupts
*/
void
bfa_ioc_msgget(ioc, &m);
- /**
+ /*
* Treat IOC message class as special.
*/
mc = m.mh.msg_class;
ioc->port_id = bfa_ioc_pcifn(ioc);
}
-/**
+/*
* return true if IOC is disabled
*/
bfa_boolean_t
bfa_fsm_cmp_state(ioc, bfa_ioc_sm_disabled);
}
-/**
+/*
* return true if IOC firmware is different.
*/
bfa_boolean_t
((__sm) == BFI_IOC_FAIL) || \
((__sm) == BFI_IOC_CFG_DISABLED))
-/**
+/*
* Check if adapter is disabled -- both IOCs should be in a disabled
* state.
*/
bfa_ioc_adapter_is_disabled(struct bfa_ioc_s *ioc)
{
u32 ioc_state;
- bfa_os_addr_t rb = ioc->pcidev.pci_bar_kva;
+ void __iomem *rb = ioc->pcidev.pci_bar_kva;
if (!bfa_fsm_cmp_state(ioc, bfa_ioc_sm_disabled))
return BFA_FALSE;
- ioc_state = bfa_reg_read(rb + BFA_IOC0_STATE_REG);
+ ioc_state = readl(rb + BFA_IOC0_STATE_REG);
if (!bfa_ioc_state_disabled(ioc_state))
return BFA_FALSE;
if (ioc->pcidev.device_id != BFA_PCI_DEVICE_ID_FC_8G1P) {
- ioc_state = bfa_reg_read(rb + BFA_IOC1_STATE_REG);
+ ioc_state = readl(rb + BFA_IOC1_STATE_REG);
if (!bfa_ioc_state_disabled(ioc_state))
return BFA_FALSE;
}
return BFA_TRUE;
}
-/**
+/*
* Add to IOC heartbeat failure notification queue. To be used by common
* modules such as cee, port, diag.
*/
bfa_ioc_get_adapter_fw_ver(ioc, ad_attr->fw_ver);
bfa_ioc_get_adapter_optrom_ver(ioc, ad_attr->optrom_ver);
bfa_ioc_get_adapter_manufacturer(ioc, ad_attr->manufacturer);
- bfa_os_memcpy(&ad_attr->vpd, &ioc_attr->vpd,
+ memcpy(&ad_attr->vpd, &ioc_attr->vpd,
sizeof(struct bfa_mfg_vpd_s));
ad_attr->nports = bfa_ioc_get_nports(ioc);
void
bfa_ioc_get_adapter_serial_num(struct bfa_ioc_s *ioc, char *serial_num)
{
- bfa_os_memset((void *)serial_num, 0, BFA_ADAPTER_SERIAL_NUM_LEN);
- bfa_os_memcpy((void *)serial_num,
+ memset((void *)serial_num, 0, BFA_ADAPTER_SERIAL_NUM_LEN);
+ memcpy((void *)serial_num,
(void *)ioc->attr->brcd_serialnum,
BFA_ADAPTER_SERIAL_NUM_LEN);
}
void
bfa_ioc_get_adapter_fw_ver(struct bfa_ioc_s *ioc, char *fw_ver)
{
- bfa_os_memset((void *)fw_ver, 0, BFA_VERSION_LEN);
- bfa_os_memcpy(fw_ver, ioc->attr->fw_version, BFA_VERSION_LEN);
+ memset((void *)fw_ver, 0, BFA_VERSION_LEN);
+ memcpy(fw_ver, ioc->attr->fw_version, BFA_VERSION_LEN);
}
void
{
bfa_assert(chip_rev);
- bfa_os_memset((void *)chip_rev, 0, BFA_IOC_CHIP_REV_LEN);
+ memset((void *)chip_rev, 0, BFA_IOC_CHIP_REV_LEN);
chip_rev[0] = 'R';
chip_rev[1] = 'e';
void
bfa_ioc_get_adapter_optrom_ver(struct bfa_ioc_s *ioc, char *optrom_ver)
{
- bfa_os_memset((void *)optrom_ver, 0, BFA_VERSION_LEN);
- bfa_os_memcpy(optrom_ver, ioc->attr->optrom_version,
+ memset((void *)optrom_ver, 0, BFA_VERSION_LEN);
+ memcpy(optrom_ver, ioc->attr->optrom_version,
BFA_VERSION_LEN);
}
void
bfa_ioc_get_adapter_manufacturer(struct bfa_ioc_s *ioc, char *manufacturer)
{
- bfa_os_memset((void *)manufacturer, 0, BFA_ADAPTER_MFG_NAME_LEN);
- bfa_os_memcpy(manufacturer, BFA_MFG_NAME, BFA_ADAPTER_MFG_NAME_LEN);
+ memset((void *)manufacturer, 0, BFA_ADAPTER_MFG_NAME_LEN);
+ memcpy(manufacturer, BFA_MFG_NAME, BFA_ADAPTER_MFG_NAME_LEN);
}
void
struct bfi_ioc_attr_s *ioc_attr;
bfa_assert(model);
- bfa_os_memset((void *)model, 0, BFA_ADAPTER_MODEL_NAME_LEN);
+ memset((void *)model, 0, BFA_ADAPTER_MODEL_NAME_LEN);
ioc_attr = ioc->attr;
- /**
+ /*
* model name
*/
- bfa_os_snprintf(model, BFA_ADAPTER_MODEL_NAME_LEN, "%s-%u",
+ snprintf(model, BFA_ADAPTER_MODEL_NAME_LEN, "%s-%u",
BFA_MFG_NAME, ioc_attr->card_type);
}
void
bfa_ioc_get_attr(struct bfa_ioc_s *ioc, struct bfa_ioc_attr_s *ioc_attr)
{
- bfa_os_memset((void *)ioc_attr, 0, sizeof(struct bfa_ioc_attr_s));
+ memset((void *)ioc_attr, 0, sizeof(struct bfa_ioc_attr_s));
ioc_attr->state = bfa_ioc_get_state(ioc);
ioc_attr->port_id = ioc->port_id;
bfa_ioc_get_pci_chip_rev(ioc, ioc_attr->pci_attr.chip_rev);
}
-/**
+/*
* hal_wwn_public
*/
wwn_t
return ioc->fcmode || !bfa_asic_id_ct(ioc->pcidev.device_id);
}
-/**
+/*
* Retrieve saved firmware trace from a prior IOC failure.
*/
bfa_status_t
if (tlen > ioc->dbg_fwsave_len)
tlen = ioc->dbg_fwsave_len;
- bfa_os_memcpy(trcdata, ioc->dbg_fwsave, tlen);
+ memcpy(trcdata, ioc->dbg_fwsave, tlen);
*trclen = tlen;
return BFA_STATUS_OK;
}
-/**
+/*
* Clear saved firmware trace
*/
void
ioc->dbg_fwsave_once = BFA_TRUE;
}
-/**
+/*
* Retrieve saved firmware trace from a prior IOC failure.
*/
bfa_status_t
bfa_ioc_send_fwsync(ioc);
- /**
+ /*
* After sending a fw sync mbox command wait for it to
* take effect. We will not wait for a response because
* 1. fw_sync mbox cmd doesn't have a response.
fwsync_iter--;
}
-/**
+/*
* Dump firmware smem
*/
bfa_status_t
loff = *offset;
dlen = *buflen;
- /**
+ /*
* First smem read, sync smem before proceeding
* No need to sync before reading every chunk.
*/
return status;
}
-/**
+/*
* Firmware statistics
*/
bfa_status_t
return status;
}
-/**
+/*
* Save firmware trace if configured.
*/
static void
}
}
-/**
+/*
* Firmware failure detected. Start recovery actions.
*/
static void
return;
}
-/**
+/*
* hal_iocpf_pvt BFA IOC PF private functions
*/
bfa_ioc_hw_sem_get(ioc);
}
-/**
+/*
* bfa timer function
*/
void
}
}
-/**
+/*
* Should be called with lock protection
*/
void
list_add_tail(&timer->qe, &mod->timer_q);
}
-/**
+/*
* Should be called with lock protection
*/
void
#include "bfa_cs.h"
#include "bfi.h"
-/**
+/*
* BFA timer declarations
*/
typedef void (*bfa_timer_cbfn_t)(void *);
-/**
+/*
* BFA timer data structure
*/
struct bfa_timer_s {
struct list_head qe;
bfa_timer_cbfn_t timercb;
void *arg;
- int timeout; /**< in millisecs. */
+ int timeout; /* in millisecs */
};
-/**
+/*
* Timer module structure
*/
struct bfa_timer_mod_s {
struct list_head timer_q;
};
-#define BFA_TIMER_FREQ 200 /**< specified in millisecs */
+#define BFA_TIMER_FREQ 200 /* specified in millisecs */
void bfa_timer_beat(struct bfa_timer_mod_s *mod);
void bfa_timer_init(struct bfa_timer_mod_s *mod);
unsigned int timeout);
void bfa_timer_stop(struct bfa_timer_s *timer);
-/**
+/*
* Generic Scatter Gather Element used by driver
*/
struct bfa_sge_s {
};
#define bfa_sge_word_swap(__sge) do { \
- ((u32 *)(__sge))[0] = bfa_os_swap32(((u32 *)(__sge))[0]); \
- ((u32 *)(__sge))[1] = bfa_os_swap32(((u32 *)(__sge))[1]); \
- ((u32 *)(__sge))[2] = bfa_os_swap32(((u32 *)(__sge))[2]); \
+ ((u32 *)(__sge))[0] = swab32(((u32 *)(__sge))[0]); \
+ ((u32 *)(__sge))[1] = swab32(((u32 *)(__sge))[1]); \
+ ((u32 *)(__sge))[2] = swab32(((u32 *)(__sge))[2]); \
} while (0)
#define bfa_swap_words(_x) ( \
#define bfa_sgaddr_le(_x) (_x)
#endif
-/**
+/*
* PCI device information required by IOC
*/
struct bfa_pcidev_s {
int pci_slot;
u8 pci_func;
- u16 device_id;
- bfa_os_addr_t pci_bar_kva;
+ u16 device_id;
+ void __iomem *pci_bar_kva;
};
-/**
+/*
* Structure used to remember the DMA-able memory block's KVA and Physical
* Address
*/
#define BFA_DMA_ALIGN_SZ 256
#define BFA_ROUNDUP(_l, _s) (((_l) + ((_s) - 1)) & ~((_s) - 1))
-/**
+/*
* smem size for Crossbow and Catapult
*/
#define BFI_SMEM_CB_SIZE 0x200000U /* ! 2MB for crossbow */
static inline void
__bfa_dma_be_addr_set(union bfi_addr_u *dma_addr, u64 pa)
{
- dma_addr->a32.addr_lo = (u32) bfa_os_htonl(pa);
- dma_addr->a32.addr_hi = (u32) bfa_os_htonl(bfa_os_u32(pa));
+ dma_addr->a32.addr_lo = (u32) cpu_to_be32(pa);
+ dma_addr->a32.addr_hi = (u32) cpu_to_be32(bfa_os_u32(pa));
}
struct bfa_ioc_regs_s {
- bfa_os_addr_t hfn_mbox_cmd;
- bfa_os_addr_t hfn_mbox;
- bfa_os_addr_t lpu_mbox_cmd;
- bfa_os_addr_t lpu_mbox;
- bfa_os_addr_t pss_ctl_reg;
- bfa_os_addr_t pss_err_status_reg;
- bfa_os_addr_t app_pll_fast_ctl_reg;
- bfa_os_addr_t app_pll_slow_ctl_reg;
- bfa_os_addr_t ioc_sem_reg;
- bfa_os_addr_t ioc_usage_sem_reg;
- bfa_os_addr_t ioc_init_sem_reg;
- bfa_os_addr_t ioc_usage_reg;
- bfa_os_addr_t host_page_num_fn;
- bfa_os_addr_t heartbeat;
- bfa_os_addr_t ioc_fwstate;
- bfa_os_addr_t ll_halt;
- bfa_os_addr_t err_set;
- bfa_os_addr_t shirq_isr_next;
- bfa_os_addr_t shirq_msk_next;
- bfa_os_addr_t smem_page_start;
+ void __iomem *hfn_mbox_cmd;
+ void __iomem *hfn_mbox;
+ void __iomem *lpu_mbox_cmd;
+ void __iomem *lpu_mbox;
+ void __iomem *pss_ctl_reg;
+ void __iomem *pss_err_status_reg;
+ void __iomem *app_pll_fast_ctl_reg;
+ void __iomem *app_pll_slow_ctl_reg;
+ void __iomem *ioc_sem_reg;
+ void __iomem *ioc_usage_sem_reg;
+ void __iomem *ioc_init_sem_reg;
+ void __iomem *ioc_usage_reg;
+ void __iomem *host_page_num_fn;
+ void __iomem *heartbeat;
+ void __iomem *ioc_fwstate;
+ void __iomem *ll_halt;
+ void __iomem *err_set;
+ void __iomem *shirq_isr_next;
+ void __iomem *shirq_msk_next;
+ void __iomem *smem_page_start;
u32 smem_pg0;
};
-#define bfa_reg_read(_raddr) bfa_os_reg_read(_raddr)
-#define bfa_reg_write(_raddr, _val) bfa_os_reg_write(_raddr, _val)
-#define bfa_mem_read(_raddr, _off) bfa_os_mem_read(_raddr, _off)
+#define bfa_mem_read(_raddr, _off) swab32(readl(((_raddr) + (_off))))
#define bfa_mem_write(_raddr, _off, _val) \
- bfa_os_mem_write(_raddr, _off, _val)
-/**
+ writel(swab32((_val)), ((_raddr) + (_off)))
+/*
* IOC Mailbox structures
*/
struct bfa_mbox_cmd_s {
u32 msg[BFI_IOC_MSGSZ];
};
-/**
+/*
* IOC mailbox module
*/
typedef void (*bfa_ioc_mbox_mcfunc_t)(void *cbarg, struct bfi_mbmsg_s *m);
} mbhdlr[BFI_MC_MAX];
};
-/**
+/*
* IOC callback function interfaces
*/
typedef void (*bfa_ioc_enable_cbfn_t)(void *bfa, enum bfa_status status);
bfa_ioc_reset_cbfn_t reset_cbfn;
};
-/**
+/*
* Heartbeat failure notification queue element.
*/
struct bfa_ioc_hbfail_notify_s {
void *cbarg;
};
-/**
+/*
* Initialize a heartbeat failure notification structure
*/
#define bfa_ioc_hbfail_init(__notify, __cbfn, __cbarg) do { \
};
struct bfa_ioc_hwif_s {
- bfa_status_t (*ioc_pll_init) (bfa_os_addr_t rb, bfa_boolean_t fcmode);
+ bfa_status_t (*ioc_pll_init) (void __iomem *rb, bfa_boolean_t fcmode);
bfa_boolean_t (*ioc_firmware_lock) (struct bfa_ioc_s *ioc);
void (*ioc_firmware_unlock) (struct bfa_ioc_s *ioc);
void (*ioc_reg_init) (struct bfa_ioc_s *ioc);
#define bfa_ioc_fetch_stats(__ioc, __stats) \
(((__stats)->drv_stats) = (__ioc)->stats)
#define bfa_ioc_clr_stats(__ioc) \
- bfa_os_memset(&(__ioc)->stats, 0, sizeof((__ioc)->stats))
+ memset(&(__ioc)->stats, 0, sizeof((__ioc)->stats))
#define bfa_ioc_maxfrsize(__ioc) ((__ioc)->attr->maxfrsize)
#define bfa_ioc_rx_bbcredit(__ioc) ((__ioc)->attr->rx_bbcredit)
#define bfa_ioc_speed_sup(__ioc) \
#define BFA_IOC_FLASH_OFFSET_IN_CHUNK(off) (off % BFI_FLASH_CHUNK_SZ_WORDS)
#define BFA_IOC_FLASH_CHUNK_ADDR(chunkno) (chunkno * BFI_FLASH_CHUNK_SZ_WORDS)
-/**
+/*
* IOC mailbox interface
*/
void bfa_ioc_mbox_queue(struct bfa_ioc_s *ioc, struct bfa_mbox_cmd_s *cmd);
void bfa_ioc_mbox_regisr(struct bfa_ioc_s *ioc, enum bfi_mclass mc,
bfa_ioc_mbox_mcfunc_t cbfn, void *cbarg);
-/**
+/*
* IOC interfaces
*/
(__ioc)->fcmode))
bfa_status_t bfa_ioc_pll_init(struct bfa_ioc_s *ioc);
-bfa_status_t bfa_ioc_cb_pll_init(bfa_os_addr_t rb, bfa_boolean_t fcmode);
-bfa_boolean_t bfa_ioc_ct_pll_init_complete(bfa_os_addr_t rb);
-bfa_status_t bfa_ioc_ct_pll_init(bfa_os_addr_t rb, bfa_boolean_t fcmode);
+bfa_status_t bfa_ioc_cb_pll_init(void __iomem *rb, bfa_boolean_t fcmode);
+bfa_boolean_t bfa_ioc_ct_pll_init_complete(void __iomem *rb);
+bfa_status_t bfa_ioc_ct_pll_init(void __iomem *rb, bfa_boolean_t fcmode);
#define bfa_ioc_isr_mode_set(__ioc, __msix) \
((__ioc)->ioc_hwif->ioc_isr_mode_set(__ioc, __msix))
bfa_boolean_t bfa_ioc_get_fcmode(struct bfa_ioc_s *ioc);
void bfa_ioc_hbfail_register(struct bfa_ioc_s *ioc,
struct bfa_ioc_hbfail_notify_s *notify);
-bfa_boolean_t bfa_ioc_sem_get(bfa_os_addr_t sem_reg);
-void bfa_ioc_sem_release(bfa_os_addr_t sem_reg);
+bfa_boolean_t bfa_ioc_sem_get(void __iomem *sem_reg);
+void bfa_ioc_sem_release(void __iomem *sem_reg);
void bfa_ioc_hw_sem_release(struct bfa_ioc_s *ioc);
void bfa_ioc_fwver_get(struct bfa_ioc_s *ioc,
struct bfi_ioc_image_hdr_s *fwhdr);
}
}
-/**
+/*
* CNA TRCMOD declaration
*/
/*
struct bfa_ioc_hwif_s hwif_cb;
-/**
+/*
* Called from bfa_ioc_attach() to map asic specific calls.
*/
void
ioc->ioc_hwif = &hwif_cb;
}
-/**
+/*
* Return true if firmware of current driver matches the running firmware.
*/
static bfa_boolean_t
{
}
-/**
+/*
* Notify other functions on HB failure.
*/
static void
bfa_ioc_cb_notify_hbfail(struct bfa_ioc_s *ioc)
{
- bfa_reg_write(ioc->ioc_regs.err_set, __PSS_ERR_STATUS_SET);
- bfa_reg_read(ioc->ioc_regs.err_set);
+ writel(__PSS_ERR_STATUS_SET, ioc->ioc_regs.err_set);
+ readl(ioc->ioc_regs.err_set);
}
-/**
+/*
* Host to LPU mailbox message addresses
*/
static struct { u32 hfn_mbox, lpu_mbox, hfn_pgn; } iocreg_fnreg[] = {
{ HOSTFN1_LPU_MBOX0_8, LPU_HOSTFN1_MBOX0_8, HOST_PAGE_NUM_FN1 }
};
-/**
+/*
* Host <-> LPU mailbox command/status registers
*/
static struct { u32 hfn, lpu; } iocreg_mbcmd[] = {
static void
bfa_ioc_cb_reg_init(struct bfa_ioc_s *ioc)
{
- bfa_os_addr_t rb;
+ void __iomem *rb;
int pcifn = bfa_ioc_pcifn(ioc);
rb = bfa_ioc_bar0(ioc);
ioc->ioc_regs.ioc_fwstate = (rb + BFA_IOC1_STATE_REG);
}
- /**
+ /*
* Host <-> LPU mailbox command/status registers
*/
ioc->ioc_regs.hfn_mbox_cmd = rb + iocreg_mbcmd[pcifn].hfn;
ioc->ioc_regs.ioc_sem_reg = (rb + HOST_SEM0_REG);
ioc->ioc_regs.ioc_init_sem_reg = (rb + HOST_SEM2_REG);
- /**
+ /*
* sram memory access
*/
ioc->ioc_regs.smem_page_start = (rb + PSS_SMEM_PAGE_START);
ioc->ioc_regs.err_set = (rb + ERR_SET_REG);
}
-/**
+/*
* Initialize IOC to port mapping.
*/
static void
bfa_ioc_cb_map_port(struct bfa_ioc_s *ioc)
{
- /**
+ /*
* For crossbow, port id is same as pci function.
*/
ioc->port_id = bfa_ioc_pcifn(ioc);
bfa_trc(ioc, ioc->port_id);
}
-/**
+/*
* Set interrupt mode for a function: INTX or MSIX
*/
static void
{
}
-/**
+/*
* Cleanup hw semaphore and usecnt registers
*/
static void
* before we clear it. If it is not locked, writing 1
* will lock it instead of clearing it.
*/
- bfa_reg_read(ioc->ioc_regs.ioc_sem_reg);
+ readl(ioc->ioc_regs.ioc_sem_reg);
bfa_ioc_hw_sem_release(ioc);
}
bfa_status_t
-bfa_ioc_cb_pll_init(bfa_os_addr_t rb, bfa_boolean_t fcmode)
+bfa_ioc_cb_pll_init(void __iomem *rb, bfa_boolean_t fcmode)
{
u32 pll_sclk, pll_fclk;
__APP_PLL_400_RSEL200500 | __APP_PLL_400_P0_1(3U) |
__APP_PLL_400_JITLMT0_1(3U) |
__APP_PLL_400_CNTLMT0_1(3U);
- bfa_reg_write((rb + BFA_IOC0_STATE_REG), BFI_IOC_UNINIT);
- bfa_reg_write((rb + BFA_IOC1_STATE_REG), BFI_IOC_UNINIT);
- bfa_reg_write((rb + HOSTFN0_INT_MSK), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN1_INT_MSK), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN0_INT_STATUS), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN1_INT_STATUS), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN0_INT_MSK), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN1_INT_MSK), 0xffffffffU);
- bfa_reg_write(rb + APP_PLL_212_CTL_REG,
- __APP_PLL_212_LOGIC_SOFT_RESET);
- bfa_reg_write(rb + APP_PLL_212_CTL_REG,
- __APP_PLL_212_BYPASS |
- __APP_PLL_212_LOGIC_SOFT_RESET);
- bfa_reg_write(rb + APP_PLL_400_CTL_REG,
- __APP_PLL_400_LOGIC_SOFT_RESET);
- bfa_reg_write(rb + APP_PLL_400_CTL_REG,
- __APP_PLL_400_BYPASS |
- __APP_PLL_400_LOGIC_SOFT_RESET);
- bfa_os_udelay(2);
- bfa_reg_write(rb + APP_PLL_212_CTL_REG,
- __APP_PLL_212_LOGIC_SOFT_RESET);
- bfa_reg_write(rb + APP_PLL_400_CTL_REG,
- __APP_PLL_400_LOGIC_SOFT_RESET);
- bfa_reg_write(rb + APP_PLL_212_CTL_REG,
- pll_sclk | __APP_PLL_212_LOGIC_SOFT_RESET);
- bfa_reg_write(rb + APP_PLL_400_CTL_REG,
- pll_fclk | __APP_PLL_400_LOGIC_SOFT_RESET);
- bfa_os_udelay(2000);
- bfa_reg_write((rb + HOSTFN0_INT_STATUS), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN1_INT_STATUS), 0xffffffffU);
- bfa_reg_write((rb + APP_PLL_212_CTL_REG), pll_sclk);
- bfa_reg_write((rb + APP_PLL_400_CTL_REG), pll_fclk);
+ writel(BFI_IOC_UNINIT, (rb + BFA_IOC0_STATE_REG));
+ writel(BFI_IOC_UNINIT, (rb + BFA_IOC1_STATE_REG));
+ writel(0xffffffffU, (rb + HOSTFN0_INT_MSK));
+ writel(0xffffffffU, (rb + HOSTFN1_INT_MSK));
+ writel(0xffffffffU, (rb + HOSTFN0_INT_STATUS));
+ writel(0xffffffffU, (rb + HOSTFN1_INT_STATUS));
+ writel(0xffffffffU, (rb + HOSTFN0_INT_MSK));
+ writel(0xffffffffU, (rb + HOSTFN1_INT_MSK));
+ writel(__APP_PLL_212_LOGIC_SOFT_RESET, rb + APP_PLL_212_CTL_REG);
+ writel(__APP_PLL_212_BYPASS | __APP_PLL_212_LOGIC_SOFT_RESET,
+ rb + APP_PLL_212_CTL_REG);
+ writel(__APP_PLL_400_LOGIC_SOFT_RESET, rb + APP_PLL_400_CTL_REG);
+ writel(__APP_PLL_400_BYPASS | __APP_PLL_400_LOGIC_SOFT_RESET,
+ rb + APP_PLL_400_CTL_REG);
+ udelay(2);
+ writel(__APP_PLL_212_LOGIC_SOFT_RESET, rb + APP_PLL_212_CTL_REG);
+ writel(__APP_PLL_400_LOGIC_SOFT_RESET, rb + APP_PLL_400_CTL_REG);
+ writel(pll_sclk | __APP_PLL_212_LOGIC_SOFT_RESET,
+ rb + APP_PLL_212_CTL_REG);
+ writel(pll_fclk | __APP_PLL_400_LOGIC_SOFT_RESET,
+ rb + APP_PLL_400_CTL_REG);
+ udelay(2000);
+ writel(0xffffffffU, (rb + HOSTFN0_INT_STATUS));
+ writel(0xffffffffU, (rb + HOSTFN1_INT_STATUS));
+ writel(pll_sclk, (rb + APP_PLL_212_CTL_REG));
+ writel(pll_fclk, (rb + APP_PLL_400_CTL_REG));
return BFA_STATUS_OK;
}
struct bfa_ioc_hwif_s hwif_ct;
-/**
+/*
* Called from bfa_ioc_attach() to map asic specific calls.
*/
void
ioc->ioc_hwif = &hwif_ct;
}
-/**
+/*
* Return true if firmware of current driver matches the running firmware.
*/
static bfa_boolean_t
u32 usecnt;
struct bfi_ioc_image_hdr_s fwhdr;
- /**
+ /*
* Firmware match check is relevant only for CNA.
*/
if (!ioc->cna)
return BFA_TRUE;
- /**
+ /*
* If bios boot (flash based) -- do not increment usage count
*/
if (bfa_cb_image_get_size(BFA_IOC_FWIMG_TYPE(ioc)) <
return BFA_TRUE;
bfa_ioc_sem_get(ioc->ioc_regs.ioc_usage_sem_reg);
- usecnt = bfa_reg_read(ioc->ioc_regs.ioc_usage_reg);
+ usecnt = readl(ioc->ioc_regs.ioc_usage_reg);
- /**
+ /*
* If usage count is 0, always return TRUE.
*/
if (usecnt == 0) {
- bfa_reg_write(ioc->ioc_regs.ioc_usage_reg, 1);
+ writel(1, ioc->ioc_regs.ioc_usage_reg);
bfa_ioc_sem_release(ioc->ioc_regs.ioc_usage_sem_reg);
bfa_trc(ioc, usecnt);
return BFA_TRUE;
}
- ioc_fwstate = bfa_reg_read(ioc->ioc_regs.ioc_fwstate);
+ ioc_fwstate = readl(ioc->ioc_regs.ioc_fwstate);
bfa_trc(ioc, ioc_fwstate);
- /**
+ /*
* Use count cannot be non-zero and chip in uninitialized state.
*/
bfa_assert(ioc_fwstate != BFI_IOC_UNINIT);
- /**
+ /*
* Check if another driver with a different firmware is active
*/
bfa_ioc_fwver_get(ioc, &fwhdr);
return BFA_FALSE;
}
- /**
+ /*
* Same firmware version. Increment the reference count.
*/
usecnt++;
- bfa_reg_write(ioc->ioc_regs.ioc_usage_reg, usecnt);
+ writel(usecnt, ioc->ioc_regs.ioc_usage_reg);
bfa_ioc_sem_release(ioc->ioc_regs.ioc_usage_sem_reg);
bfa_trc(ioc, usecnt);
return BFA_TRUE;
{
u32 usecnt;
- /**
+ /*
* Firmware lock is relevant only for CNA.
*/
if (!ioc->cna)
return;
- /**
+ /*
* If bios boot (flash based) -- do not decrement usage count
*/
if (bfa_cb_image_get_size(BFA_IOC_FWIMG_TYPE(ioc)) <
BFA_IOC_FWIMG_MINSZ)
return;
- /**
+ /*
* decrement usage count
*/
bfa_ioc_sem_get(ioc->ioc_regs.ioc_usage_sem_reg);
- usecnt = bfa_reg_read(ioc->ioc_regs.ioc_usage_reg);
+ usecnt = readl(ioc->ioc_regs.ioc_usage_reg);
bfa_assert(usecnt > 0);
usecnt--;
- bfa_reg_write(ioc->ioc_regs.ioc_usage_reg, usecnt);
+ writel(usecnt, ioc->ioc_regs.ioc_usage_reg);
bfa_trc(ioc, usecnt);
bfa_ioc_sem_release(ioc->ioc_regs.ioc_usage_sem_reg);
}
-/**
+/*
* Notify other functions on HB failure.
*/
static void
bfa_ioc_ct_notify_hbfail(struct bfa_ioc_s *ioc)
{
if (ioc->cna) {
- bfa_reg_write(ioc->ioc_regs.ll_halt, __FW_INIT_HALT_P);
+ writel(__FW_INIT_HALT_P, ioc->ioc_regs.ll_halt);
/* Wait for halt to take effect */
- bfa_reg_read(ioc->ioc_regs.ll_halt);
+ readl(ioc->ioc_regs.ll_halt);
} else {
- bfa_reg_write(ioc->ioc_regs.err_set, __PSS_ERR_STATUS_SET);
- bfa_reg_read(ioc->ioc_regs.err_set);
+ writel(__PSS_ERR_STATUS_SET, ioc->ioc_regs.err_set);
+ readl(ioc->ioc_regs.err_set);
}
}
-/**
+/*
* Host to LPU mailbox message addresses
*/
static struct { u32 hfn_mbox, lpu_mbox, hfn_pgn; } iocreg_fnreg[] = {
{ HOSTFN3_LPU_MBOX0_8, LPU_HOSTFN3_MBOX0_8, HOST_PAGE_NUM_FN3 }
};
-/**
+/*
* Host <-> LPU mailbox command/status registers - port 0
*/
static struct { u32 hfn, lpu; } iocreg_mbcmd_p0[] = {
{ HOSTFN3_LPU0_MBOX0_CMD_STAT, LPU0_HOSTFN3_MBOX0_CMD_STAT }
};
-/**
+/*
* Host <-> LPU mailbox command/status registers - port 1
*/
static struct { u32 hfn, lpu; } iocreg_mbcmd_p1[] = {
static void
bfa_ioc_ct_reg_init(struct bfa_ioc_s *ioc)
{
- bfa_os_addr_t rb;
+ void __iomem *rb;
int pcifn = bfa_ioc_pcifn(ioc);
rb = bfa_ioc_bar0(ioc);
ioc->ioc_regs.ioc_init_sem_reg = (rb + HOST_SEM2_REG);
ioc->ioc_regs.ioc_usage_reg = (rb + BFA_FW_USE_COUNT);
- /**
+ /*
* sram memory access
*/
ioc->ioc_regs.smem_page_start = (rb + PSS_SMEM_PAGE_START);
ioc->ioc_regs.err_set = (rb + ERR_SET_REG);
}
-/**
+/*
* Initialize IOC to port mapping.
*/
static void
bfa_ioc_ct_map_port(struct bfa_ioc_s *ioc)
{
- bfa_os_addr_t rb = ioc->pcidev.pci_bar_kva;
+ void __iomem *rb = ioc->pcidev.pci_bar_kva;
u32 r32;
- /**
+ /*
* For catapult, base port id on personality register and IOC type
*/
- r32 = bfa_reg_read(rb + FNC_PERS_REG);
+ r32 = readl(rb + FNC_PERS_REG);
r32 >>= FNC_PERS_FN_SHIFT(bfa_ioc_pcifn(ioc));
ioc->port_id = (r32 & __F0_PORT_MAP_MK) >> __F0_PORT_MAP_SH;
bfa_trc(ioc, ioc->port_id);
}
-/**
+/*
* Set interrupt mode for a function: INTX or MSIX
*/
static void
bfa_ioc_ct_isr_mode_set(struct bfa_ioc_s *ioc, bfa_boolean_t msix)
{
- bfa_os_addr_t rb = ioc->pcidev.pci_bar_kva;
+ void __iomem *rb = ioc->pcidev.pci_bar_kva;
u32 r32, mode;
- r32 = bfa_reg_read(rb + FNC_PERS_REG);
+ r32 = readl(rb + FNC_PERS_REG);
bfa_trc(ioc, r32);
mode = (r32 >> FNC_PERS_FN_SHIFT(bfa_ioc_pcifn(ioc))) &
__F0_INTX_STATUS;
- /**
+ /*
* If already in desired mode, do not change anything
*/
if (!msix && mode)
r32 |= (mode << FNC_PERS_FN_SHIFT(bfa_ioc_pcifn(ioc)));
bfa_trc(ioc, r32);
- bfa_reg_write(rb + FNC_PERS_REG, r32);
+ writel(r32, rb + FNC_PERS_REG);
}
-/**
+/*
* Cleanup hw semaphore and usecnt registers
*/
static void
if (ioc->cna) {
bfa_ioc_sem_get(ioc->ioc_regs.ioc_usage_sem_reg);
- bfa_reg_write(ioc->ioc_regs.ioc_usage_reg, 0);
+ writel(0, ioc->ioc_regs.ioc_usage_reg);
bfa_ioc_sem_release(ioc->ioc_regs.ioc_usage_sem_reg);
}
* before we clear it. If it is not locked, writing 1
* will lock it instead of clearing it.
*/
- bfa_reg_read(ioc->ioc_regs.ioc_sem_reg);
+ readl(ioc->ioc_regs.ioc_sem_reg);
bfa_ioc_hw_sem_release(ioc);
}
* Check the firmware state to know if pll_init has been completed already
*/
bfa_boolean_t
-bfa_ioc_ct_pll_init_complete(bfa_os_addr_t rb)
+bfa_ioc_ct_pll_init_complete(void __iomem *rb)
{
- if ((bfa_reg_read(rb + BFA_IOC0_STATE_REG) == BFI_IOC_OP) ||
- (bfa_reg_read(rb + BFA_IOC1_STATE_REG) == BFI_IOC_OP))
+ if ((readl(rb + BFA_IOC0_STATE_REG) == BFI_IOC_OP) ||
+ (readl(rb + BFA_IOC1_STATE_REG) == BFI_IOC_OP))
return BFA_TRUE;
return BFA_FALSE;
}
bfa_status_t
-bfa_ioc_ct_pll_init(bfa_os_addr_t rb, bfa_boolean_t fcmode)
+bfa_ioc_ct_pll_init(void __iomem *rb, bfa_boolean_t fcmode)
{
u32 pll_sclk, pll_fclk, r32;
__APP_PLL_425_JITLMT0_1(3U) |
__APP_PLL_425_CNTLMT0_1(1U);
if (fcmode) {
- bfa_reg_write((rb + OP_MODE), 0);
- bfa_reg_write((rb + ETH_MAC_SER_REG),
- __APP_EMS_CMLCKSEL |
- __APP_EMS_REFCKBUFEN2 |
- __APP_EMS_CHANNEL_SEL);
+ writel(0, (rb + OP_MODE));
+ writel(__APP_EMS_CMLCKSEL | __APP_EMS_REFCKBUFEN2 |
+ __APP_EMS_CHANNEL_SEL, (rb + ETH_MAC_SER_REG));
} else {
- bfa_reg_write((rb + OP_MODE), __GLOBAL_FCOE_MODE);
- bfa_reg_write((rb + ETH_MAC_SER_REG),
- __APP_EMS_REFCKBUFEN1);
+ writel(__GLOBAL_FCOE_MODE, (rb + OP_MODE));
+ writel(__APP_EMS_REFCKBUFEN1, (rb + ETH_MAC_SER_REG));
}
- bfa_reg_write((rb + BFA_IOC0_STATE_REG), BFI_IOC_UNINIT);
- bfa_reg_write((rb + BFA_IOC1_STATE_REG), BFI_IOC_UNINIT);
- bfa_reg_write((rb + HOSTFN0_INT_MSK), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN1_INT_MSK), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN0_INT_STATUS), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN1_INT_STATUS), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN0_INT_MSK), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN1_INT_MSK), 0xffffffffU);
- bfa_reg_write(rb + APP_PLL_312_CTL_REG, pll_sclk |
- __APP_PLL_312_LOGIC_SOFT_RESET);
- bfa_reg_write(rb + APP_PLL_425_CTL_REG, pll_fclk |
- __APP_PLL_425_LOGIC_SOFT_RESET);
- bfa_reg_write(rb + APP_PLL_312_CTL_REG, pll_sclk |
- __APP_PLL_312_LOGIC_SOFT_RESET | __APP_PLL_312_ENABLE);
- bfa_reg_write(rb + APP_PLL_425_CTL_REG, pll_fclk |
- __APP_PLL_425_LOGIC_SOFT_RESET | __APP_PLL_425_ENABLE);
- bfa_reg_read(rb + HOSTFN0_INT_MSK);
- bfa_os_udelay(2000);
- bfa_reg_write((rb + HOSTFN0_INT_STATUS), 0xffffffffU);
- bfa_reg_write((rb + HOSTFN1_INT_STATUS), 0xffffffffU);
- bfa_reg_write(rb + APP_PLL_312_CTL_REG, pll_sclk |
- __APP_PLL_312_ENABLE);
- bfa_reg_write(rb + APP_PLL_425_CTL_REG, pll_fclk |
- __APP_PLL_425_ENABLE);
+ writel(BFI_IOC_UNINIT, (rb + BFA_IOC0_STATE_REG));
+ writel(BFI_IOC_UNINIT, (rb + BFA_IOC1_STATE_REG));
+ writel(0xffffffffU, (rb + HOSTFN0_INT_MSK));
+ writel(0xffffffffU, (rb + HOSTFN1_INT_MSK));
+ writel(0xffffffffU, (rb + HOSTFN0_INT_STATUS));
+ writel(0xffffffffU, (rb + HOSTFN1_INT_STATUS));
+ writel(0xffffffffU, (rb + HOSTFN0_INT_MSK));
+ writel(0xffffffffU, (rb + HOSTFN1_INT_MSK));
+ writel(pll_sclk | __APP_PLL_312_LOGIC_SOFT_RESET,
+ rb + APP_PLL_312_CTL_REG);
+ writel(pll_fclk | __APP_PLL_425_LOGIC_SOFT_RESET,
+ rb + APP_PLL_425_CTL_REG);
+ writel(pll_sclk | __APP_PLL_312_LOGIC_SOFT_RESET | __APP_PLL_312_ENABLE,
+ rb + APP_PLL_312_CTL_REG);
+ writel(pll_fclk | __APP_PLL_425_LOGIC_SOFT_RESET | __APP_PLL_425_ENABLE,
+ rb + APP_PLL_425_CTL_REG);
+ readl(rb + HOSTFN0_INT_MSK);
+ udelay(2000);
+ writel(0xffffffffU, (rb + HOSTFN0_INT_STATUS));
+ writel(0xffffffffU, (rb + HOSTFN1_INT_STATUS));
+ writel(pll_sclk | __APP_PLL_312_ENABLE, rb + APP_PLL_312_CTL_REG);
+ writel(pll_fclk | __APP_PLL_425_ENABLE, rb + APP_PLL_425_CTL_REG);
if (!fcmode) {
- bfa_reg_write((rb + PMM_1T_RESET_REG_P0), __PMM_1T_RESET_P);
- bfa_reg_write((rb + PMM_1T_RESET_REG_P1), __PMM_1T_RESET_P);
+ writel(__PMM_1T_RESET_P, (rb + PMM_1T_RESET_REG_P0));
+ writel(__PMM_1T_RESET_P, (rb + PMM_1T_RESET_REG_P1));
}
- r32 = bfa_reg_read((rb + PSS_CTL_REG));
+ r32 = readl((rb + PSS_CTL_REG));
r32 &= ~__PSS_LMEM_RESET;
- bfa_reg_write((rb + PSS_CTL_REG), r32);
- bfa_os_udelay(1000);
+ writel(r32, (rb + PSS_CTL_REG));
+ udelay(1000);
if (!fcmode) {
- bfa_reg_write((rb + PMM_1T_RESET_REG_P0), 0);
- bfa_reg_write((rb + PMM_1T_RESET_REG_P1), 0);
+ writel(0, (rb + PMM_1T_RESET_REG_P0));
+ writel(0, (rb + PMM_1T_RESET_REG_P1));
}
- bfa_reg_write((rb + MBIST_CTL_REG), __EDRAM_BISTR_START);
- bfa_os_udelay(1000);
- r32 = bfa_reg_read((rb + MBIST_STAT_REG));
- bfa_reg_write((rb + MBIST_CTL_REG), 0);
+ writel(__EDRAM_BISTR_START, (rb + MBIST_CTL_REG));
+ udelay(1000);
+ r32 = readl((rb + MBIST_STAT_REG));
+ writel(0, (rb + MBIST_CTL_REG));
return BFA_STATUS_OK;
}
* General Public License for more details.
*/
-/**
+/*
* bfa_modules.h BFA modules
*/
};
-/**
+/*
* Macro to define a new BFA module
*/
#define BFA_MODULE(__mod) \
#define BFA_CACHELINE_SZ (256)
-/**
+/*
* Structure used to interact between different BFA sub modules
*
* Each sub module needs to implement only the entry points relevant to it (and
* General Public License for more details.
*/
-/**
- * Contains declarations all OS Specific files needed for BFA layer
- */
-
#ifndef __BFA_OS_INC_H__
#define __BFA_OS_INC_H__
#define __BIGENDIAN
#endif
-static inline u64 bfa_os_get_clock(void)
-{
- return jiffies;
-}
-
static inline u64 bfa_os_get_log_time(void)
{
u64 system_time = 0;
#define bfa_io_lat_clock_res_div HZ
#define bfa_io_lat_clock_res_mul 1000
-#define BFA_ASSERT(p) do { \
- if (!(p)) { \
- printk(KERN_ERR "assert(%s) failed at %s:%d\n", \
- #p, __FILE__, __LINE__); \
- } \
-} while (0)
-
#define BFA_LOG(level, bfad, mask, fmt, arg...) \
do { \
if (((mask) == 4) || (level[1] <= '4')) \
((_x) & 0x00ff00) | \
(((_x) & 0xff0000) >> 16))
-#define bfa_swap_8b(_x) \
- ((((_x) & 0xff00000000000000ull) >> 56) \
- | (((_x) & 0x00ff000000000000ull) >> 40) \
- | (((_x) & 0x0000ff0000000000ull) >> 24) \
- | (((_x) & 0x000000ff00000000ull) >> 8) \
- | (((_x) & 0x00000000ff000000ull) << 8) \
- | (((_x) & 0x0000000000ff0000ull) << 24) \
- | (((_x) & 0x000000000000ff00ull) << 40) \
- | (((_x) & 0x00000000000000ffull) << 56))
-
-#define bfa_os_swap32(_x) \
- ((((_x) & 0xff) << 24) | \
- (((_x) & 0x0000ff00) << 8) | \
- (((_x) & 0x00ff0000) >> 8) | \
- (((_x) & 0xff000000) >> 24))
-
#define bfa_os_swap_sgaddr(_x) ((u64)( \
(((u64)(_x) & (u64)0x00000000000000ffull) << 32) | \
(((u64)(_x) & (u64)0x000000000000ff00ull) << 32) | \
(((u64)(_x) & (u64)0xff00000000000000ull) >> 32)))
#ifndef __BIGENDIAN
-#define bfa_os_htons(_x) ((u16)((((_x) & 0xff00) >> 8) | \
- (((_x) & 0x00ff) << 8)))
-#define bfa_os_htonl(_x) bfa_os_swap32(_x)
-#define bfa_os_htonll(_x) bfa_swap_8b(_x)
-#define bfa_os_hton3b(_x) bfa_swap_3b(_x)
-#define bfa_os_wtole(_x) (_x)
+#define bfa_os_hton3b(_x) bfa_swap_3b(_x)
#define bfa_os_sgaddr(_x) (_x)
-
#else
-
-#define bfa_os_htons(_x) (_x)
-#define bfa_os_htonl(_x) (_x)
#define bfa_os_hton3b(_x) (_x)
-#define bfa_os_htonll(_x) (_x)
-#define bfa_os_wtole(_x) bfa_os_swap32(_x)
#define bfa_os_sgaddr(_x) bfa_os_swap_sgaddr(_x)
-
#endif
-#define bfa_os_ntohs(_x) bfa_os_htons(_x)
-#define bfa_os_ntohl(_x) bfa_os_htonl(_x)
-#define bfa_os_ntohll(_x) bfa_os_htonll(_x)
#define bfa_os_ntoh3b(_x) bfa_os_hton3b(_x)
-
#define bfa_os_u32(__pa64) ((__pa64) >> 32)
-#define bfa_os_memset memset
-#define bfa_os_memcpy memcpy
-#define bfa_os_udelay udelay
-#define bfa_os_vsprintf vsprintf
-#define bfa_os_snprintf snprintf
-
-#define bfa_os_assign(__t, __s) __t = __s
-#define bfa_os_addr_t void __iomem *
-
-#define bfa_os_reg_read(_raddr) readl(_raddr)
-#define bfa_os_reg_write(_raddr, _val) writel((_val), (_raddr))
-#define bfa_os_mem_read(_raddr, _off) \
- bfa_os_swap32(readl(((_raddr) + (_off))))
-#define bfa_os_mem_write(_raddr, _off, _val) \
- writel(bfa_os_swap32((_val)), ((_raddr) + (_off)))
-
-#define BFA_TRC_TS(_trcm) \
- ({ \
- struct timeval tv; \
- \
- do_gettimeofday(&tv); \
- (tv.tv_sec*1000000+tv.tv_usec); \
- })
+#define BFA_TRC_TS(_trcm) \
+ ({ \
+ struct timeval tv; \
+ \
+ do_gettimeofday(&tv); \
+ (tv.tv_sec*1000000+tv.tv_usec); \
+ })
#define boolean_t int
-/**
+/*
* For current time stamp, OS API will fill-in
*/
struct bfa_timeval_s {
t0 = dip[i];
t1 = dip[i + 1];
#ifdef __BIGENDIAN
- dip[i] = bfa_os_ntohl(t0);
- dip[i + 1] = bfa_os_ntohl(t1);
+ dip[i] = be32_to_cpu(t0);
+ dip[i + 1] = be32_to_cpu(t1);
#else
- dip[i] = bfa_os_ntohl(t1);
- dip[i + 1] = bfa_os_ntohl(t0);
+ dip[i] = be32_to_cpu(t1);
+ dip[i + 1] = be32_to_cpu(t0);
#endif
}
}
-/**
+/*
* bfa_port_enable_isr()
*
*
port->endis_cbfn(port->endis_cbarg, status);
}
-/**
+/*
* bfa_port_disable_isr()
*
*
port->endis_cbfn(port->endis_cbarg, status);
}
-/**
+/*
* bfa_port_get_stats_isr()
*
*
}
}
-/**
+/*
* bfa_port_clear_stats_isr()
*
*
port->stats_status = status;
port->stats_busy = BFA_FALSE;
- /**
+ /*
* re-initialize time stamp for stats reset
*/
bfa_os_gettimeofday(&tv);
}
}
-/**
+/*
* bfa_port_isr()
*
*
}
}
-/**
+/*
* bfa_port_meminfo()
*
*
return BFA_ROUNDUP(sizeof(union bfa_port_stats_u), BFA_DMA_ALIGN_SZ);
}
-/**
+/*
* bfa_port_mem_claim()
*
*
port->stats_dma.pa = dma_pa;
}
-/**
+/*
* bfa_port_enable()
*
* Send the Port enable request to the f/w
return BFA_STATUS_OK;
}
-/**
+/*
* bfa_port_disable()
*
* Send the Port disable request to the f/w
return BFA_STATUS_OK;
}
-/**
+/*
* bfa_port_get_stats()
*
* Send the request to the f/w to fetch Port statistics.
return BFA_STATUS_OK;
}
-/**
+/*
* bfa_port_clear_stats()
*
*
return BFA_STATUS_OK;
}
-/**
+/*
* bfa_port_hbfail()
*
*
}
}
-/**
+/*
* bfa_port_attach()
*
*
bfa_ioc_hbfail_init(&port->hbfail, bfa_port_hbfail, port);
bfa_ioc_hbfail_register(port->ioc, &port->hbfail);
- /**
+ /*
* initialize time stamp for stats reset
*/
bfa_os_gettimeofday(&tv);
bfa_trc(port, 0);
}
-/**
+/*
* bfa_port_detach()
*
*
BFA_MODULE(rport);
BFA_MODULE(uf);
-/**
+/*
* LPS related definitions
*/
#define BFA_LPS_MIN_LPORTS (1)
#define BFA_LPS_MAX_VPORTS_SUPP_CB 255
#define BFA_LPS_MAX_VPORTS_SUPP_CT 190
-/**
+/*
* lps_pvt BFA LPS private functions
*/
BFA_LPS_SM_RX_CVL = 7, /* Rx clear virtual link */
};
-/**
+/*
* FC PORT related definitions
*/
/*
(bfa_ioc_is_disabled(&bfa->ioc) == BFA_TRUE))
-/**
+/*
* BFA port state machine events
*/
enum bfa_fcport_sm_event {
BFA_FCPORT_SM_HWFAIL = 9, /* IOC h/w failure */
};
-/**
+/*
* BFA port link notification state machine events
*/
BFA_FCPORT_LN_SM_NOTIFICATION = 3 /* done notification */
};
-/**
+/*
* RPORT related definitions
*/
#define bfa_rport_offline_cb(__rp) do { \
BFA_RPORT_SM_QRESUME = 9, /* space in requeue queue */
};
-/**
+/*
* forward declarations FCXP related functions
*/
static void __bfa_fcxp_send_cbfn(void *cbarg, bfa_boolean_t complete);
static void bfa_fcxp_queue(struct bfa_fcxp_s *fcxp,
struct bfi_fcxp_send_req_s *send_req);
-/**
+/*
* forward declarations for LPS functions
*/
static void bfa_lps_meminfo(struct bfa_iocfc_cfg_s *cfg, u32 *ndm_len,
static void bfa_lps_logout_comp(struct bfa_lps_s *lps);
static void bfa_lps_cvl_event(struct bfa_lps_s *lps);
-/**
+/*
* forward declaration for LPS state machine
*/
static void bfa_lps_sm_init(struct bfa_lps_s *lps, enum bfa_lps_event event);
static void bfa_lps_sm_logowait(struct bfa_lps_s *lps, enum bfa_lps_event
event);
-/**
+/*
* forward declaration for FC Port functions
*/
static bfa_boolean_t bfa_fcport_send_enable(struct bfa_fcport_s *fcport);
static void bfa_fcport_stats_clr_timeout(void *cbarg);
static void bfa_trunk_iocdisable(struct bfa_s *bfa);
-/**
+/*
* forward declaration for FC PORT state machine
*/
static void bfa_fcport_sm_uninit(struct bfa_fcport_s *fcport,
};
-/**
+/*
* forward declaration for RPORT related functions
*/
static struct bfa_rport_s *bfa_rport_alloc(struct bfa_rport_mod_s *rp_mod);
static void __bfa_cb_rport_offline(void *cbarg,
bfa_boolean_t complete);
-/**
+/*
* forward declaration for RPORT state machine
*/
static void bfa_rport_sm_uninit(struct bfa_rport_s *rp,
static void bfa_rport_sm_deleting_qfull(struct bfa_rport_s *rp,
enum bfa_rport_event event);
-/**
+/*
* PLOG related definitions
*/
static int
pl_recp = &(plog->plog_recs[tail]);
- bfa_os_memcpy(pl_recp, pl_rec, sizeof(struct bfa_plog_rec_s));
+ memcpy(pl_recp, pl_rec, sizeof(struct bfa_plog_rec_s));
pl_recp->tv = bfa_os_get_log_time();
BFA_PL_LOG_REC_INCR(plog->tail);
void
bfa_plog_init(struct bfa_plog_s *plog)
{
- bfa_os_memset((char *)plog, 0, sizeof(struct bfa_plog_s));
+ memset((char *)plog, 0, sizeof(struct bfa_plog_s));
- bfa_os_memcpy(plog->plog_sig, BFA_PL_SIG_STR, BFA_PL_SIG_LEN);
+ memcpy(plog->plog_sig, BFA_PL_SIG_STR, BFA_PL_SIG_LEN);
plog->head = plog->tail = 0;
plog->plog_enabled = 1;
}
struct bfa_plog_rec_s lp;
if (plog->plog_enabled) {
- bfa_os_memset(&lp, 0, sizeof(struct bfa_plog_rec_s));
+ memset(&lp, 0, sizeof(struct bfa_plog_rec_s));
lp.mid = mid;
lp.eid = event;
lp.log_type = BFA_PL_LOG_TYPE_STRING;
num_ints = BFA_PL_INT_LOG_SZ;
if (plog->plog_enabled) {
- bfa_os_memset(&lp, 0, sizeof(struct bfa_plog_rec_s));
+ memset(&lp, 0, sizeof(struct bfa_plog_rec_s));
lp.mid = mid;
lp.eid = event;
lp.log_type = BFA_PL_LOG_TYPE_INT;
lp.misc = misc;
for (i = 0; i < num_ints; i++)
- bfa_os_assign(lp.log_entry.int_log[i],
- intarr[i]);
+ lp.log_entry.int_log[i] = intarr[i];
lp.log_num_ints = (u8) num_ints;
u32 ints[BFA_PL_INT_LOG_SZ];
if (plog->plog_enabled) {
- bfa_os_memset(&lp, 0, sizeof(struct bfa_plog_rec_s));
+ memset(&lp, 0, sizeof(struct bfa_plog_rec_s));
ints[0] = tmp_int[0];
ints[1] = tmp_int[1];
u32 ints[BFA_PL_INT_LOG_SZ];
if (plog->plog_enabled) {
- bfa_os_memset(&lp, 0, sizeof(struct bfa_plog_rec_s));
+ memset(&lp, 0, sizeof(struct bfa_plog_rec_s));
ints[0] = tmp_int[0];
ints[1] = tmp_int[1];
return (bfa_boolean_t)plog->plog_enabled;
}
-/**
+/*
* fcxp_pvt BFA FCXP private functions
*/
mod->req_pld_list_pa = dm_pa;
dm_kva += buf_pool_sz;
dm_pa += buf_pool_sz;
- bfa_os_memset(mod->req_pld_list_kva, 0, buf_pool_sz);
+ memset(mod->req_pld_list_kva, 0, buf_pool_sz);
/*
* Initialize the fcxp rsp payload list
mod->rsp_pld_list_pa = dm_pa;
dm_kva += buf_pool_sz;
dm_pa += buf_pool_sz;
- bfa_os_memset(mod->rsp_pld_list_kva, 0, buf_pool_sz);
+ memset(mod->rsp_pld_list_kva, 0, buf_pool_sz);
bfa_meminfo_dma_virt(mi) = dm_kva;
bfa_meminfo_dma_phys(mi) = dm_pa;
struct bfa_fcxp_s *fcxp;
fcxp = (struct bfa_fcxp_s *) bfa_meminfo_kva(mi);
- bfa_os_memset(fcxp, 0, sizeof(struct bfa_fcxp_s) * mod->num_fcxps);
+ memset(fcxp, 0, sizeof(struct bfa_fcxp_s) * mod->num_fcxps);
INIT_LIST_HEAD(&mod->fcxp_free_q);
INIT_LIST_HEAD(&mod->fcxp_active_q);
{
struct bfa_fcxp_mod_s *mod = BFA_FCXP_MOD(bfa);
- bfa_os_memset(mod, 0, sizeof(struct bfa_fcxp_mod_s));
+ memset(mod, 0, sizeof(struct bfa_fcxp_mod_s));
mod->bfa = bfa;
mod->num_fcxps = cfg->fwcfg.num_fcxp_reqs;
- /**
+ /*
* Initialize FCXP request and response payload sizes.
*/
mod->req_pld_sz = mod->rsp_pld_sz = BFA_FCXP_MAX_IBUF_SZ;
{
struct bfa_fcxp_mod_s *mod = BFA_FCXP_MOD(bfa);
struct bfa_fcxp_s *fcxp;
- u16 fcxp_tag = bfa_os_ntohs(fcxp_rsp->fcxp_tag);
+ u16 fcxp_tag = be16_to_cpu(fcxp_rsp->fcxp_tag);
bfa_trc(bfa, fcxp_tag);
- fcxp_rsp->rsp_len = bfa_os_ntohl(fcxp_rsp->rsp_len);
+ fcxp_rsp->rsp_len = be32_to_cpu(fcxp_rsp->rsp_len);
- /**
+ /*
* @todo f/w should not set residue to non-0 when everything
* is received.
*/
if (fcxp_rsp->req_status == BFA_STATUS_OK)
fcxp_rsp->residue_len = 0;
else
- fcxp_rsp->residue_len = bfa_os_ntohl(fcxp_rsp->residue_len);
+ fcxp_rsp->residue_len = be32_to_cpu(fcxp_rsp->residue_len);
fcxp = BFA_FCXP_FROM_TAG(mod, fcxp_tag);
}
}
-/**
+/*
* Handler to resume sending fcxp when space in available in cpe queue.
*/
static void
bfa_fcxp_queue(fcxp, send_req);
}
-/**
+/*
* Queue fcxp send request to foimrware.
*/
static void
bfi_h2i_set(send_req->mh, BFI_MC_FCXP, BFI_FCXP_H2I_SEND_REQ,
bfa_lpuid(bfa));
- send_req->fcxp_tag = bfa_os_htons(fcxp->fcxp_tag);
+ send_req->fcxp_tag = cpu_to_be16(fcxp->fcxp_tag);
if (rport) {
send_req->rport_fw_hndl = rport->fw_handle;
- send_req->max_frmsz = bfa_os_htons(rport->rport_info.max_frmsz);
+ send_req->max_frmsz = cpu_to_be16(rport->rport_info.max_frmsz);
if (send_req->max_frmsz == 0)
- send_req->max_frmsz = bfa_os_htons(FC_MAX_PDUSZ);
+ send_req->max_frmsz = cpu_to_be16(FC_MAX_PDUSZ);
} else {
send_req->rport_fw_hndl = 0;
- send_req->max_frmsz = bfa_os_htons(FC_MAX_PDUSZ);
+ send_req->max_frmsz = cpu_to_be16(FC_MAX_PDUSZ);
}
- send_req->vf_id = bfa_os_htons(reqi->vf_id);
+ send_req->vf_id = cpu_to_be16(reqi->vf_id);
send_req->lp_tag = reqi->lp_tag;
send_req->class = reqi->class;
send_req->rsp_timeout = rspi->rsp_timeout;
send_req->cts = reqi->cts;
send_req->fchs = reqi->fchs;
- send_req->req_len = bfa_os_htonl(reqi->req_tot_len);
- send_req->rsp_maxlen = bfa_os_htonl(rspi->rsp_maxlen);
+ send_req->req_len = cpu_to_be32(reqi->req_tot_len);
+ send_req->rsp_maxlen = cpu_to_be32(rspi->rsp_maxlen);
/*
* setup req sgles
bfa_trc(bfa, bfa_reqq_ci(bfa, BFA_REQQ_FCXP));
}
-/**
+/*
* hal_fcxp_api BFA FCXP API
*/
-/**
+/*
* Allocate an FCXP instance to send a response or to send a request
* that has a response. Request/response buffers are allocated by caller.
*
return fcxp;
}
-/**
+/*
* Get the internal request buffer pointer
*
* @param[in] fcxp BFA fcxp pointer
return mod->req_pld_sz;
}
-/**
+/*
* Get the internal response buffer pointer
*
* @param[in] fcxp BFA fcxp pointer
return rspbuf;
}
-/**
+/*
* Free the BFA FCXP
*
* @param[in] fcxp BFA fcxp pointer
bfa_fcxp_put(fcxp);
}
-/**
+/*
* Send a FCXP request
*
* @param[in] fcxp BFA fcxp pointer
bfa_trc(bfa, fcxp->fcxp_tag);
- /**
+ /*
* setup request/response info
*/
reqi->bfa_rport = rport;
fcxp->send_cbfn = cbfn ? cbfn : bfa_fcxp_null_comp;
fcxp->send_cbarg = cbarg;
- /**
+ /*
* If no room in CPE queue, wait for space in request queue
*/
send_req = bfa_reqq_next(bfa, BFA_REQQ_FCXP);
bfa_fcxp_queue(fcxp, send_req);
}
-/**
+/*
* Abort a BFA FCXP
*
* @param[in] fcxp BFA fcxp pointer
void
bfa_fcxp_discard(struct bfa_fcxp_s *fcxp)
{
- /**
+ /*
* If waiting for room in request queue, cancel reqq wait
* and free fcxp.
*/
-/**
+/*
* hal_fcxp_public BFA FCXP public functions
*/
}
-/**
+/*
* BFA LPS state machine functions
*/
-/**
+/*
* Init state -- no login
*/
static void
}
}
-/**
+/*
* login is in progress -- awaiting response from firmware
*/
static void
}
}
-/**
+/*
* login pending - awaiting space in request queue
*/
static void
}
}
-/**
+/*
* login complete
*/
static void
}
}
-/**
+/*
* logout in progress - awaiting firmware response
*/
static void
}
}
-/**
+/*
* logout pending -- awaiting space in request queue
*/
static void
-/**
+/*
* lps_pvt BFA LPS private functions
*/
-/**
+/*
* return memory requirement
*/
static void
*ndm_len += sizeof(struct bfa_lps_s) * BFA_LPS_MAX_LPORTS;
}
-/**
+/*
* bfa module attach at initialization time
*/
static void
struct bfa_lps_s *lps;
int i;
- bfa_os_memset(mod, 0, sizeof(struct bfa_lps_mod_s));
+ memset(mod, 0, sizeof(struct bfa_lps_mod_s));
mod->num_lps = BFA_LPS_MAX_LPORTS;
if (cfg->drvcfg.min_cfg)
mod->num_lps = BFA_LPS_MIN_LPORTS;
{
}
-/**
+/*
* IOC in disabled state -- consider all lps offline
*/
static void
}
}
-/**
+/*
* Firmware login response
*/
static void
lps->fport = rsp->f_port;
lps->npiv_en = rsp->npiv_en;
lps->lp_pid = rsp->lp_pid;
- lps->pr_bbcred = bfa_os_ntohs(rsp->bb_credit);
+ lps->pr_bbcred = be16_to_cpu(rsp->bb_credit);
lps->pr_pwwn = rsp->port_name;
lps->pr_nwwn = rsp->node_name;
lps->auth_req = rsp->auth_req;
bfa_sm_send_event(lps, BFA_LPS_SM_FWRSP);
}
-/**
+/*
* Firmware logout response
*/
static void
bfa_sm_send_event(lps, BFA_LPS_SM_FWRSP);
}
-/**
+/*
* Firmware received a Clear virtual link request (for FCoE)
*/
static void
bfa_sm_send_event(lps, BFA_LPS_SM_RX_CVL);
}
-/**
+/*
* Space is available in request queue, resume queueing request to firmware.
*/
static void
bfa_sm_send_event(lps, BFA_LPS_SM_RESUME);
}
-/**
+/*
* lps is freed -- triggered by vport delete
*/
static void
list_add_tail(&lps->qe, &mod->lps_free_q);
}
-/**
+/*
* send login request to firmware
*/
static void
m->lp_tag = lps->lp_tag;
m->alpa = lps->alpa;
- m->pdu_size = bfa_os_htons(lps->pdusz);
+ m->pdu_size = cpu_to_be16(lps->pdusz);
m->pwwn = lps->pwwn;
m->nwwn = lps->nwwn;
m->fdisc = lps->fdisc;
bfa_reqq_produce(lps->bfa, lps->reqq);
}
-/**
+/*
* send logout request to firmware
*/
static void
bfa_reqq_produce(lps->bfa, lps->reqq);
}
-/**
+/*
* Indirect login completion handler for non-fcs
*/
static void
bfa_cb_lps_flogi_comp(lps->bfa->bfad, lps->uarg, lps->status);
}
-/**
+/*
* Login completion handler -- direct call for fcs, queue for others
*/
static void
bfa_cb_lps_flogi_comp(lps->bfa->bfad, lps->uarg, lps->status);
}
-/**
+/*
* Indirect logout completion handler for non-fcs
*/
static void
bfa_cb_lps_fdisclogo_comp(lps->bfa->bfad, lps->uarg);
}
-/**
+/*
* Logout completion handler -- direct call for fcs, queue for others
*/
static void
bfa_cb_lps_fdisclogo_comp(lps->bfa->bfad, lps->uarg);
}
-/**
+/*
* Clear virtual link completion handler for non-fcs
*/
static void
bfa_cb_lps_cvl_event(lps->bfa->bfad, lps->uarg);
}
-/**
+/*
* Received Clear virtual link event --direct call for fcs,
* queue for others
*/
-/**
+/*
* lps_public BFA LPS public functions
*/
return BFA_LPS_MAX_VPORTS_SUPP_CB;
}
-/**
+/*
* Allocate a lport srvice tag.
*/
struct bfa_lps_s *
return lps;
}
-/**
+/*
* Free lport service tag. This can be called anytime after an alloc.
* No need to wait for any pending login/logout completions.
*/
bfa_sm_send_event(lps, BFA_LPS_SM_DELETE);
}
-/**
+/*
* Initiate a lport login.
*/
void
bfa_sm_send_event(lps, BFA_LPS_SM_LOGIN);
}
-/**
+/*
* Initiate a lport fdisc login.
*/
void
bfa_sm_send_event(lps, BFA_LPS_SM_LOGIN);
}
-/**
+/*
* Initiate a lport logout (flogi).
*/
void
bfa_sm_send_event(lps, BFA_LPS_SM_LOGOUT);
}
-/**
+/*
* Initiate a lport FDSIC logout.
*/
void
bfa_sm_send_event(lps, BFA_LPS_SM_LOGOUT);
}
-/**
+/*
* Discard a pending login request -- should be called only for
* link down handling.
*/
bfa_sm_send_event(lps, BFA_LPS_SM_OFFLINE);
}
-/**
+/*
* Return lport services tag
*/
u8
return lps->lp_tag;
}
-/**
+/*
* Return lport services tag given the pid
*/
u8
return 0;
}
-/**
+/*
* return if fabric login indicates support for NPIV
*/
bfa_boolean_t
return lps->npiv_en;
}
-/**
+/*
* Return TRUE if attached to F-Port, else return FALSE
*/
bfa_boolean_t
return lps->fport;
}
-/**
+/*
* Return TRUE if attached to a Brocade Fabric
*/
bfa_boolean_t
{
return lps->brcd_switch;
}
-/**
+/*
* return TRUE if authentication is required
*/
bfa_boolean_t
return lps->ext_status;
}
-/**
+/*
* return port id assigned to the lport
*/
u32
return lps->lp_pid;
}
-/**
+/*
* return port id assigned to the base lport
*/
u32
return BFA_LPS_FROM_TAG(mod, 0)->lp_pid;
}
-/**
+/*
* Return bb_credit assigned in FLOGI response
*/
u16
return lps->pr_bbcred;
}
-/**
+/*
* Return peer port name
*/
wwn_t
return lps->pr_pwwn;
}
-/**
+/*
* Return peer node name
*/
wwn_t
return lps->pr_nwwn;
}
-/**
+/*
* return reason code if login request is rejected
*/
u8
return lps->lsrjt_rsn;
}
-/**
+/*
* return explanation code if login request is rejected
*/
u8
return lps->lsrjt_expl;
}
-/**
+/*
* Return fpma/spma MAC for lport
*/
mac_t
return lps->lp_mac;
}
-/**
+/*
* LPS firmware message class handler.
*/
void
}
}
-/**
+/*
* FC PORT state machine functions
*/
static void
switch (event) {
case BFA_FCPORT_SM_START:
- /**
+ /*
* Start event after IOC is configured and BFA is started.
*/
if (bfa_fcport_send_enable(fcport)) {
break;
case BFA_FCPORT_SM_ENABLE:
- /**
+ /*
* Port is persistently configured to be in enabled state. Do
* not change state. Port enabling is done when START event is
* received.
break;
case BFA_FCPORT_SM_DISABLE:
- /**
+ /*
* If a port is persistently configured to be disabled, the
* first event will a port disable request.
*/
break;
case BFA_FCPORT_SM_ENABLE:
- /**
+ /*
* Already enable is in progress.
*/
break;
case BFA_FCPORT_SM_DISABLE:
- /**
+ /*
* Just send disable request to firmware when room becomes
* available in request queue.
*/
case BFA_FCPORT_SM_LINKUP:
case BFA_FCPORT_SM_LINKDOWN:
- /**
+ /*
* Possible to get link events when doing back-to-back
* enable/disables.
*/
break;
case BFA_FCPORT_SM_ENABLE:
- /**
+ /*
* Already being enabled.
*/
break;
break;
case BFA_FCPORT_SM_LINKDOWN:
- /**
+ /*
* Possible to get link down event.
*/
break;
case BFA_FCPORT_SM_ENABLE:
- /**
+ /*
* Already enabled.
*/
break;
switch (event) {
case BFA_FCPORT_SM_ENABLE:
- /**
+ /*
* Already enabled.
*/
break;
break;
case BFA_FCPORT_SM_DISABLE:
- /**
+ /*
* Already being disabled.
*/
break;
case BFA_FCPORT_SM_LINKUP:
case BFA_FCPORT_SM_LINKDOWN:
- /**
+ /*
* Possible to get link events when doing back-to-back
* enable/disables.
*/
case BFA_FCPORT_SM_LINKUP:
case BFA_FCPORT_SM_LINKDOWN:
- /**
+ /*
* Possible to get link events when doing back-to-back
* enable/disables.
*/
break;
case BFA_FCPORT_SM_DISABLE:
- /**
+ /*
* Already being disabled.
*/
break;
case BFA_FCPORT_SM_LINKUP:
case BFA_FCPORT_SM_LINKDOWN:
- /**
+ /*
* Possible to get link events when doing back-to-back
* enable/disables.
*/
switch (event) {
case BFA_FCPORT_SM_START:
- /**
+ /*
* Ignore start event for a port that is disabled.
*/
break;
break;
case BFA_FCPORT_SM_DISABLE:
- /**
+ /*
* Already disabled.
*/
break;
break;
default:
- /**
+ /*
* Ignore all other events.
*/
;
}
}
-/**
+/*
* Port is enabled. IOC is down/failed.
*/
static void
break;
default:
- /**
+ /*
* Ignore all events.
*/
;
}
}
-/**
+/*
* Port is disabled. IOC is down/failed.
*/
static void
break;
default:
- /**
+ /*
* Ignore all events.
*/
;
}
}
-/**
+/*
* Link state is down
*/
static void
}
}
-/**
+/*
* Link state is waiting for down notification
*/
static void
}
}
-/**
+/*
* Link state is waiting for down notification and there is a pending up
*/
static void
}
}
-/**
+/*
* Link state is up
*/
static void
}
}
-/**
+/*
* Link state is waiting for up notification
*/
static void
}
}
-/**
+/*
* Link state is waiting for up notification and there is a pending down
*/
static void
}
}
-/**
+/*
* Link state is waiting for up notification and there are pending down and up
*/
static void
-/**
+/*
* hal_port_private
*/
bfa_sm_send_event(ln, BFA_FCPORT_LN_SM_NOTIFICATION);
}
-/**
+/*
* Send SCN notification to upper layers.
* trunk - false if caller is fcport to ignore fcport event in trunked mode
*/
bfa_meminfo_dma_phys(meminfo) = dm_pa;
}
-/**
+/*
* Memory initialization.
*/
static void
struct bfa_fcport_ln_s *ln = &fcport->ln;
struct bfa_timeval_s tv;
- bfa_os_memset(fcport, 0, sizeof(struct bfa_fcport_s));
+ memset(fcport, 0, sizeof(struct bfa_fcport_s));
fcport->bfa = bfa;
ln->fcport = fcport;
bfa_sm_set_state(fcport, bfa_fcport_sm_uninit);
bfa_sm_set_state(ln, bfa_fcport_ln_sm_dn);
- /**
+ /*
* initialize time stamp for stats reset
*/
bfa_os_gettimeofday(&tv);
fcport->stats_reset_time = tv.tv_sec;
- /**
+ /*
* initialize and set default configuration
*/
port_cfg->topology = BFA_PORT_TOPOLOGY_P2P;
{
}
-/**
+/*
* Called when IOC is ready.
*/
static void
bfa_sm_send_event(BFA_FCPORT_MOD(bfa), BFA_FCPORT_SM_START);
}
-/**
+/*
* Called before IOC is stopped.
*/
static void
bfa_trunk_iocdisable(bfa);
}
-/**
+/*
* Called when IOC failure is detected.
*/
static void
fcport->myalpa = 0;
/* QoS Details */
- bfa_os_assign(fcport->qos_attr, pevent->link_state.qos_attr);
- bfa_os_assign(fcport->qos_vc_attr,
- pevent->link_state.vc_fcf.qos_vc_attr);
+ fcport->qos_attr = pevent->link_state.qos_attr;
+ fcport->qos_vc_attr = pevent->link_state.vc_fcf.qos_vc_attr;
- /**
+ /*
* update trunk state if applicable
*/
if (!fcport->cfg.trunked)
trunk->attr.state = BFA_TRUNK_DISABLED;
/* update FCoE specific */
- fcport->fcoe_vlan = bfa_os_ntohs(pevent->link_state.vc_fcf.fcf.vlan);
+ fcport->fcoe_vlan = be16_to_cpu(pevent->link_state.vc_fcf.fcf.vlan);
bfa_trc(fcport->bfa, fcport->speed);
bfa_trc(fcport->bfa, fcport->topology);
fcport->topology = BFA_PORT_TOPOLOGY_NONE;
}
-/**
+/*
* Send port enable message to firmware.
*/
static bfa_boolean_t
{
struct bfi_fcport_enable_req_s *m;
- /**
+ /*
* Increment message tag before queue check, so that responses to old
* requests are discarded.
*/
fcport->msgtag++;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(fcport->bfa, BFA_REQQ_PORT);
m->pwwn = fcport->pwwn;
m->port_cfg = fcport->cfg;
m->msgtag = fcport->msgtag;
- m->port_cfg.maxfrsize = bfa_os_htons(fcport->cfg.maxfrsize);
+ m->port_cfg.maxfrsize = cpu_to_be16(fcport->cfg.maxfrsize);
bfa_dma_be_addr_set(m->stats_dma_addr, fcport->stats_pa);
bfa_trc(fcport->bfa, m->stats_dma_addr.a32.addr_lo);
bfa_trc(fcport->bfa, m->stats_dma_addr.a32.addr_hi);
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(fcport->bfa, BFA_REQQ_PORT);
return BFA_TRUE;
}
-/**
+/*
* Send port disable message to firmware.
*/
static bfa_boolean_t
{
struct bfi_fcport_req_s *m;
- /**
+ /*
* Increment message tag before queue check, so that responses to old
* requests are discarded.
*/
fcport->msgtag++;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(fcport->bfa, BFA_REQQ_PORT);
bfa_lpuid(fcport->bfa));
m->msgtag = fcport->msgtag;
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(fcport->bfa, BFA_REQQ_PORT);
struct bfa_fcport_s *fcport = port_cbarg;
struct bfi_fcport_set_svc_params_req_s *m;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(fcport->bfa, BFA_REQQ_PORT);
bfi_h2i_set(m->mh, BFI_MC_FCPORT, BFI_FCPORT_H2I_SET_SVC_PARAMS_REQ,
bfa_lpuid(fcport->bfa));
- m->tx_bbcredit = bfa_os_htons((u16)fcport->cfg.tx_bbcredit);
+ m->tx_bbcredit = cpu_to_be16((u16)fcport->cfg.tx_bbcredit);
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(fcport->bfa, BFA_REQQ_PORT);
/* Now swap the 32 bit fields */
for (i = 0; i < (sizeof(struct bfa_qos_stats_s)/sizeof(u32)); ++i)
- dip[i] = bfa_os_ntohl(sip[i]);
+ dip[i] = be32_to_cpu(sip[i]);
}
static void
for (i = 0; i < ((sizeof(struct bfa_fcoe_stats_s))/sizeof(u32));
i = i + 2) {
#ifdef __BIGENDIAN
- dip[i] = bfa_os_ntohl(sip[i]);
- dip[i + 1] = bfa_os_ntohl(sip[i + 1]);
+ dip[i] = be32_to_cpu(sip[i]);
+ dip[i + 1] = be32_to_cpu(sip[i + 1]);
#else
- dip[i] = bfa_os_ntohl(sip[i + 1]);
- dip[i + 1] = bfa_os_ntohl(sip[i]);
+ dip[i] = be32_to_cpu(sip[i + 1]);
+ dip[i + 1] = be32_to_cpu(sip[i]);
#endif
}
}
}
fcport->stats_qfull = BFA_FALSE;
- bfa_os_memset(msg, 0, sizeof(struct bfi_fcport_req_s));
+ memset(msg, 0, sizeof(struct bfi_fcport_req_s));
bfi_h2i_set(msg->mh, BFI_MC_FCPORT, BFI_FCPORT_H2I_STATS_GET_REQ,
bfa_lpuid(fcport->bfa));
bfa_reqq_produce(fcport->bfa, BFA_REQQ_PORT);
if (complete) {
struct bfa_timeval_s tv;
- /**
+ /*
* re-initialize time stamp for stats reset
*/
bfa_os_gettimeofday(&tv);
}
fcport->stats_qfull = BFA_FALSE;
- bfa_os_memset(msg, 0, sizeof(struct bfi_fcport_req_s));
+ memset(msg, 0, sizeof(struct bfi_fcport_req_s));
bfi_h2i_set(msg->mh, BFI_MC_FCPORT, BFI_FCPORT_H2I_STATS_CLEAR_REQ,
bfa_lpuid(fcport->bfa));
bfa_reqq_produce(fcport->bfa, BFA_REQQ_PORT);
}
-/**
+/*
* Handle trunk SCN event from firmware.
*/
static void
bfa_trc(fcport->bfa, scn->trunk_state);
bfa_trc(fcport->bfa, scn->trunk_speed);
- /**
+ /*
* Save off new state for trunk attribute query
*/
state_prev = trunk->attr.state;
lattr->trunk_wwn = tlink->trunk_wwn;
lattr->fctl = tlink->fctl;
lattr->speed = tlink->speed;
- lattr->deskew = bfa_os_ntohl(tlink->deskew);
+ lattr->deskew = be32_to_cpu(tlink->deskew);
if (tlink->state == BFA_TRUNK_LINK_STATE_UP) {
fcport->speed = tlink->speed;
BFA_PL_EID_TRUNK_SCN, 0, "Trunk down");
}
- /**
+ /*
* Notify upper layers if trunk state changed.
*/
if ((state_prev != trunk->attr.state) ||
struct bfa_fcport_s *fcport = BFA_FCPORT_MOD(bfa);
int i = 0;
- /**
+ /*
* In trunked mode, notify upper layers that link is down
*/
if (fcport->cfg.trunked) {
-/**
+/*
* hal_port_public
*/
-/**
+/*
* Called to initialize port attributes
*/
void
{
struct bfa_fcport_s *fcport = BFA_FCPORT_MOD(bfa);
- /**
+ /*
* Initialize port attributes from IOC hardware data.
*/
bfa_fcport_set_wwns(fcport);
bfa_assert(fcport->speed_sup);
}
-/**
+/*
* Firmware message handler.
*/
void
-/**
+/*
* hal_port_api
*/
-/**
+/*
* Registered callback for port events.
*/
void
return BFA_STATUS_OK;
}
-/**
+/*
* Configure port speed.
*/
bfa_status_t
return BFA_STATUS_OK;
}
-/**
+/*
* Get current speed.
*/
enum bfa_port_speed
return fcport->speed;
}
-/**
+/*
* Configure port topology.
*/
bfa_status_t
return BFA_STATUS_OK;
}
-/**
+/*
* Get current topology.
*/
enum bfa_port_topology
bfa_fcport_send_txcredit(fcport);
}
-/**
+/*
* Get port attributes.
*/
{
struct bfa_fcport_s *fcport = BFA_FCPORT_MOD(bfa);
- bfa_os_memset(attr, 0, sizeof(struct bfa_port_attr_s));
+ memset(attr, 0, sizeof(struct bfa_port_attr_s));
attr->nwwn = fcport->nwwn;
attr->pwwn = fcport->pwwn;
attr->factorypwwn = bfa_ioc_get_mfg_pwwn(&bfa->ioc);
attr->factorynwwn = bfa_ioc_get_mfg_nwwn(&bfa->ioc);
- bfa_os_memcpy(&attr->pport_cfg, &fcport->cfg,
+ memcpy(&attr->pport_cfg, &fcport->cfg,
sizeof(struct bfa_port_cfg_s));
/* speed attributes */
attr->pport_cfg.speed = fcport->cfg.speed;
#define BFA_FCPORT_STATS_TOV 1000
-/**
+/*
* Fetch port statistics (FCQoS or FCoE).
*/
bfa_status_t
return BFA_STATUS_OK;
}
-/**
+/*
* Reset port statistics (FCQoS or FCoE).
*/
bfa_status_t
return BFA_STATUS_OK;
}
-/**
+/*
* Fetch FCQoS port statistics
*/
bfa_status_t
return bfa_fcport_get_stats(bfa, stats, cbfn, cbarg);
}
-/**
+/*
* Reset FCoE port statistics
*/
bfa_status_t
return bfa_fcport_clear_stats(bfa, cbfn, cbarg);
}
-/**
+/*
* Fetch FCQoS port statistics
*/
bfa_status_t
return bfa_fcport_get_stats(bfa, stats, cbfn, cbarg);
}
-/**
+/*
* Reset FCoE port statistics
*/
bfa_status_t
struct bfa_fcport_s *fcport = BFA_FCPORT_MOD(bfa);
qos_attr->state = fcport->qos_attr.state;
- qos_attr->total_bb_cr = bfa_os_ntohl(fcport->qos_attr.total_bb_cr);
+ qos_attr->total_bb_cr = be32_to_cpu(fcport->qos_attr.total_bb_cr);
}
void
struct bfa_qos_vc_attr_s *bfa_vc_attr = &fcport->qos_vc_attr;
u32 i = 0;
- qos_vc_attr->total_vc_count = bfa_os_ntohs(bfa_vc_attr->total_vc_count);
- qos_vc_attr->shared_credit = bfa_os_ntohs(bfa_vc_attr->shared_credit);
+ qos_vc_attr->total_vc_count = be16_to_cpu(bfa_vc_attr->total_vc_count);
+ qos_vc_attr->shared_credit = be16_to_cpu(bfa_vc_attr->shared_credit);
qos_vc_attr->elp_opmode_flags =
- bfa_os_ntohl(bfa_vc_attr->elp_opmode_flags);
+ be32_to_cpu(bfa_vc_attr->elp_opmode_flags);
/* Individual VC info */
while (i < qos_vc_attr->total_vc_count) {
}
}
-/**
+/*
* Fetch port attributes.
*/
bfa_boolean_t
if (ioc_type == BFA_IOC_TYPE_FC) {
fcport->cfg.qos_enabled = on_off;
- /**
+ /*
* Notify fcpim of the change in QoS state
*/
bfa_fcpim_update_ioredirect(bfa);
fcport->cfg.trl_def_speed = BFA_PORT_SPEED_1GBPS;
}
-/**
+/*
* Configure default minimum ratelim speed
*/
bfa_status_t
return BFA_STATUS_OK;
}
-/**
+/*
* Get default minimum ratelim speed
*/
enum bfa_port_speed
}
-/**
+/*
* Rport State machine functions
*/
-/**
+/*
* Beginning state, only online event expected.
*/
static void
}
}
-/**
+/*
* Waiting for rport create response from firmware.
*/
static void
}
}
-/**
+/*
* Request queue is full, awaiting queue resume to send create request.
*/
static void
}
}
-/**
+/*
* Online state - normal parking state.
*/
static void
bfa_trc(rp->bfa, qos_scn->new_qos_attr.qos_priority);
qos_scn->old_qos_attr.qos_flow_id =
- bfa_os_ntohl(qos_scn->old_qos_attr.qos_flow_id);
+ be32_to_cpu(qos_scn->old_qos_attr.qos_flow_id);
qos_scn->new_qos_attr.qos_flow_id =
- bfa_os_ntohl(qos_scn->new_qos_attr.qos_flow_id);
+ be32_to_cpu(qos_scn->new_qos_attr.qos_flow_id);
if (qos_scn->old_qos_attr.qos_flow_id !=
qos_scn->new_qos_attr.qos_flow_id)
}
}
-/**
+/*
* Firmware rport is being deleted - awaiting f/w response.
*/
static void
}
}
-/**
+/*
* Offline state.
*/
static void
}
}
-/**
+/*
* Rport is deleted, waiting for firmware response to delete.
*/
static void
}
}
-/**
+/*
* Waiting for rport create response from firmware. A delete is pending.
*/
static void
}
}
-/**
+/*
* Waiting for rport create response from firmware. Rport offline is pending.
*/
static void
}
}
-/**
+/*
* IOC h/w failed.
*/
static void
-/**
+/*
* bfa_rport_private BFA rport private functions
*/
!(mod->num_rports & (mod->num_rports - 1)));
for (i = 0; i < mod->num_rports; i++, rp++) {
- bfa_os_memset(rp, 0, sizeof(struct bfa_rport_s));
+ memset(rp, 0, sizeof(struct bfa_rport_s));
rp->bfa = bfa;
rp->rport_tag = i;
bfa_sm_set_state(rp, bfa_rport_sm_uninit);
- /**
+ /*
* - is unused
*/
if (i)
bfa_reqq_winit(&rp->reqq_wait, bfa_rport_qresume, rp);
}
- /**
+ /*
* consume memory
*/
bfa_meminfo_kva(meminfo) = (u8 *) rp;
{
struct bfi_rport_create_req_s *m;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(rp->bfa, BFA_REQQ_RPORT);
bfi_h2i_set(m->mh, BFI_MC_RPORT, BFI_RPORT_H2I_CREATE_REQ,
bfa_lpuid(rp->bfa));
m->bfa_handle = rp->rport_tag;
- m->max_frmsz = bfa_os_htons(rp->rport_info.max_frmsz);
+ m->max_frmsz = cpu_to_be16(rp->rport_info.max_frmsz);
m->pid = rp->rport_info.pid;
m->lp_tag = rp->rport_info.lp_tag;
m->local_pid = rp->rport_info.local_pid;
m->vf_id = rp->rport_info.vf_id;
m->cisc = rp->rport_info.cisc;
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(rp->bfa, BFA_REQQ_RPORT);
{
struct bfi_rport_delete_req_s *m;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(rp->bfa, BFA_REQQ_RPORT);
bfa_lpuid(rp->bfa));
m->fw_handle = rp->fw_handle;
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(rp->bfa, BFA_REQQ_RPORT);
{
struct bfa_rport_speed_req_s *m;
- /**
+ /*
* check for room in queue to send request now
*/
m = bfa_reqq_next(rp->bfa, BFA_REQQ_RPORT);
m->fw_handle = rp->fw_handle;
m->speed = (u8)rp->rport_info.speed;
- /**
+ /*
* queue I/O message to firmware
*/
bfa_reqq_produce(rp->bfa, BFA_REQQ_RPORT);
-/**
+/*
* bfa_rport_public
*/
-/**
+/*
* Rport interrupt processing.
*/
void
-/**
+/*
* bfa_rport_api
*/
{
bfa_assert(rport_info->max_frmsz != 0);
- /**
+ /*
* Some JBODs are seen to be not setting PDU size correctly in PLOGI
* responses. Default to minimum size.
*/
rport_info->max_frmsz = FC_MIN_PDUSZ;
}
- bfa_os_assign(rport->rport_info, *rport_info);
+ rport->rport_info = *rport_info;
bfa_sm_send_event(rport, BFA_RPORT_SM_ONLINE);
}
struct bfa_rport_qos_attr_s *qos_attr)
{
qos_attr->qos_priority = rport->qos_attr.qos_priority;
- qos_attr->qos_flow_id = bfa_os_ntohl(rport->qos_attr.qos_flow_id);
+ qos_attr->qos_flow_id = be32_to_cpu(rport->qos_attr.qos_flow_id);
}
void
bfa_rport_clear_stats(struct bfa_rport_s *rport)
{
- bfa_os_memset(&rport->stats, 0, sizeof(rport->stats));
+ memset(&rport->stats, 0, sizeof(rport->stats));
}
-/**
+/*
* SGPG related functions
*/
-/**
+/*
* Compute and return memory needed by FCP(im) module.
*/
static void
bfa_assert(!(sgpg_pa.pa & (sizeof(struct bfi_sgpg_s) - 1)));
for (i = 0; i < mod->num_sgpgs; i++) {
- bfa_os_memset(hsgpg, 0, sizeof(*hsgpg));
- bfa_os_memset(sgpg, 0, sizeof(*sgpg));
+ memset(hsgpg, 0, sizeof(*hsgpg));
+ memset(sgpg, 0, sizeof(*sgpg));
hsgpg->sgpg = sgpg;
sgpg_pa_tmp.pa = bfa_sgaddr_le(sgpg_pa.pa);
-/**
+/*
* hal_sgpg_public BFA SGPG public functions
*/
if (list_empty(&mod->sgpg_wait_q))
return;
- /**
+ /*
* satisfy as many waiting requests as possible
*/
do {
wqe->nsgpg_total = wqe->nsgpg = nsgpg;
- /**
+ /*
* allocate any left to this one first
*/
if (mod->free_sgpgs) {
- /**
+ /*
* no one else is waiting for SGPG
*/
bfa_assert(list_empty(&mod->sgpg_wait_q));
wqe->cbarg = cbarg;
}
-/**
+/*
* UF related functions
*/
/*
bfa_meminfo_dma_virt(mi) += uf_pb_tot_sz;
bfa_meminfo_dma_phys(mi) += uf_pb_tot_sz;
- bfa_os_memset((void *)ufm->uf_pbs_kva, 0, uf_pb_tot_sz);
+ memset((void *)ufm->uf_pbs_kva, 0, uf_pb_tot_sz);
}
static void
for (i = 0, uf_bp_msg = ufm->uf_buf_posts; i < ufm->num_ufs;
i++, uf_bp_msg++) {
- bfa_os_memset(uf_bp_msg, 0, sizeof(struct bfi_uf_buf_post_s));
+ memset(uf_bp_msg, 0, sizeof(struct bfi_uf_buf_post_s));
uf_bp_msg->buf_tag = i;
buf_len = sizeof(struct bfa_uf_buf_s);
- uf_bp_msg->buf_len = bfa_os_htons(buf_len);
+ uf_bp_msg->buf_len = cpu_to_be16(buf_len);
bfi_h2i_set(uf_bp_msg->mh, BFI_MC_UF, BFI_UF_H2I_BUF_POST,
bfa_lpuid(ufm->bfa));
bfa_sge_to_be(&sge[1]);
}
- /**
+ /*
* advance pointer beyond consumed memory
*/
bfa_meminfo_kva(mi) = (u8 *) uf_bp_msg;
* Initialize UFs and queue it in UF free queue
*/
for (i = 0, uf = ufm->uf_list; i < ufm->num_ufs; i++, uf++) {
- bfa_os_memset(uf, 0, sizeof(struct bfa_uf_s));
+ memset(uf, 0, sizeof(struct bfa_uf_s));
uf->bfa = ufm->bfa;
uf->uf_tag = i;
uf->pb_len = sizeof(struct bfa_uf_buf_s);
list_add_tail(&uf->qe, &ufm->uf_free_q);
}
- /**
+ /*
* advance memory pointer
*/
bfa_meminfo_kva(mi) = (u8 *) uf;
{
struct bfa_uf_mod_s *ufm = BFA_UF_MOD(bfa);
- bfa_os_memset(ufm, 0, sizeof(struct bfa_uf_mod_s));
+ memset(ufm, 0, sizeof(struct bfa_uf_mod_s));
ufm->bfa = bfa;
ufm->num_ufs = cfg->fwcfg.num_uf_bufs;
INIT_LIST_HEAD(&ufm->uf_free_q);
if (!uf_post_msg)
return BFA_STATUS_FAILED;
- bfa_os_memcpy(uf_post_msg, &ufm->uf_buf_posts[uf->uf_tag],
+ memcpy(uf_post_msg, &ufm->uf_buf_posts[uf->uf_tag],
sizeof(struct bfi_uf_buf_post_s));
bfa_reqq_produce(ufm->bfa, BFA_REQQ_FCXP);
u8 *buf = &uf_buf->d[0];
struct fchs_s *fchs;
- m->frm_len = bfa_os_ntohs(m->frm_len);
- m->xfr_len = bfa_os_ntohs(m->xfr_len);
+ m->frm_len = be16_to_cpu(m->frm_len);
+ m->xfr_len = be16_to_cpu(m->xfr_len);
fchs = (struct fchs_s *)uf_buf;
-/**
+/*
* hal_uf_api
*/
-/**
+/*
* Register handler for all unsolicted recieve frames.
*
* @param[in] bfa BFA instance
ufm->cbarg = cbarg;
}
-/**
+/*
* Free an unsolicited frame back to BFA.
*
* @param[in] uf unsolicited frame to be freed
-/**
+/*
* uf_pub BFA uf module public functions
*/
void
#include "bfi_ms.h"
-/**
+/*
* Scatter-gather DMA related defines
*/
#define BFA_SGPG_MIN (16)
-/**
+/*
* Alignment macro for SG page allocation
*/
#define BFA_SGPG_ROUNDUP(_l) (((_l) + (sizeof(struct bfi_sgpg_s) - 1)) \
union bfi_addr_u sgpg_pa; /* pa of SG page */
};
-/**
+/*
* Given number of SG elements, BFA_SGPG_NPAGE() returns the number of
* SG pages required.
*/
void bfa_sgpg_wcancel(struct bfa_s *bfa, struct bfa_sgpg_wqe_s *wqe);
-/**
+/*
* FCXP related defines
*/
#define BFA_FCXP_MIN (1)
-/**
+/*
* Information needed for a FCXP request
*/
struct bfa_fcxp_req_info_s {
struct bfa_rport_s *bfa_rport;
- /** Pointer to the bfa rport that was
+ /* Pointer to the bfa rport that was
* returned from bfa_rport_create().
* This could be left NULL for WKA or
* for FCXP interactions before the
struct bfa_fcxp_rsp_info_s {
struct fchs_s rsp_fchs;
- /** !< Response frame's FC header will
+ /* Response frame's FC header will
* be sent back in this field */
u8 rsp_timeout;
- /** !< timeout in seconds, 0-no response
- */
+ /* timeout in seconds, 0-no response */
u8 rsvd2[3];
u32 rsp_maxlen; /* max response length expected */
};
void bfa_fcxp_isr(struct bfa_s *bfa, struct bfi_msg_s *msg);
-/**
+/*
* RPORT related defines
*/
#define BFA_RPORT_MIN 4
#define BFA_RPORT_MOD(__bfa) (&(__bfa)->modules.rport_mod)
-/**
+/*
* Convert rport tag to RPORT
*/
#define BFA_RPORT_FROM_TAG(__bfa, _tag) \
*/
void bfa_rport_isr(struct bfa_s *bfa, struct bfi_msg_s *msg);
-/**
+/*
* BFA rport information.
*/
struct bfa_rport_info_s {
enum bfa_port_speed speed; /* Rport's current speed */
};
-/**
+/*
* BFA rport data structure
*/
struct bfa_rport_s {
#define BFA_RPORT_FC_COS(_rport) ((_rport)->rport_info.fc_class)
-/**
+/*
* UF - unsolicited receive related defines
*/
struct bfa_sge_s sges[BFI_SGE_INLINE_MAX];
};
-/**
+/*
* Callback prototype for unsolicited frame receive handler.
*
* @param[in] cbarg callback arg for receive handler
#define BFA_UF_BUFSZ (2 * 1024 + 256)
-/**
+/*
* @todo private
*/
struct bfa_uf_buf_s {
};
-/**
+/*
* LPS - bfa lport login/logout service interface
*/
struct bfa_lps_s {
void bfa_lps_isr(struct bfa_s *bfa, struct bfi_msg_s *msg);
-/**
+/*
* FCPORT related defines
*/
#define BFA_FCPORT(_bfa) (&((_bfa)->modules.port))
typedef void (*bfa_cb_port_t) (void *cbarg, enum bfa_status status);
-/**
+/*
* Link notification data structure
*/
struct bfa_fcport_ln_s {
struct bfa_trunk_attr_s attr;
};
-/**
+/*
* BFA FC port data structure
*/
struct bfa_fcport_s {
void *cbarg);
void bfa_uf_free(struct bfa_uf_s *uf);
-/**
+/*
* bfa lport service api
*/
* General Public License for more details.
*/
-/**
+/*
* bfad.c Linux driver PCI interface module.
*/
#include <linux/module.h>
static void
bfad_sm_fcs_exit(struct bfad_s *bfad, enum bfad_sm_event event);
-/**
+/*
* Beginning state for the driver instance, awaiting the pci_probe event
*/
static void
}
}
-/**
+/*
* Driver Instance is created, awaiting event INIT to initialize the bfad
*/
static void
}
}
-/**
+/*
* BFA callbacks
*/
void
complete(&fcomp->comp);
}
-/**
+/*
* bfa_init callback
*/
void
complete(&bfad->comp);
}
-/**
+/*
* BFA_FCS callbacks
*/
struct bfad_port_s *
}
}
-/**
+/*
* FCS RPORT alloc callback, after successful PLOGI by FCS
*/
bfa_status_t
return rc;
}
-/**
+/*
* FCS PBC VPORT Create
*/
void
return rc;
}
-/**
+/*
* Create a vport under a vf.
*/
bfa_status_t
return rc;
}
-/**
- * Create a vf and its base vport implicitely.
- */
-bfa_status_t
-bfad_vf_create(struct bfad_s *bfad, u16 vf_id,
- struct bfa_lport_cfg_s *port_cfg)
-{
- struct bfad_vf_s *vf;
- int rc = BFA_STATUS_OK;
-
- vf = kzalloc(sizeof(struct bfad_vf_s), GFP_KERNEL);
- if (!vf) {
- rc = BFA_STATUS_FAILED;
- goto ext;
- }
-
- rc = bfa_fcs_vf_create(&vf->fcs_vf, &bfad->bfa_fcs, vf_id, port_cfg,
- vf);
- if (rc != BFA_STATUS_OK)
- kfree(vf);
-ext:
- return rc;
-}
-
void
bfad_bfa_tmo(unsigned long data)
{
pci_set_drvdata(pdev, NULL);
}
-void
-bfad_fcs_port_cfg(struct bfad_s *bfad)
-{
- struct bfa_lport_cfg_s port_cfg;
- struct bfa_port_attr_s attr;
- char symname[BFA_SYMNAME_MAXLEN];
-
- sprintf(symname, "%s-%d", BFAD_DRIVER_NAME, bfad->inst_no);
- memcpy(port_cfg.sym_name.symname, symname, strlen(symname));
- bfa_fcport_get_attr(&bfad->bfa, &attr);
- port_cfg.nwwn = attr.nwwn;
- port_cfg.pwwn = attr.pwwn;
-}
-
bfa_status_t
bfad_drv_init(struct bfad_s *bfad)
{
bfa_fcs_init(&bfad->bfa_fcs);
spin_unlock_irqrestore(&bfad->bfad_lock, flags);
- /* PPORT FCS config */
- bfad_fcs_port_cfg(bfad);
-
retval = bfad_cfg_pport(bfad, BFA_LPORT_ROLE_FCP_IM);
if (retval != BFA_STATUS_OK) {
if (bfa_sm_cmp_state(bfad, bfad_sm_initializing))
return 0;
}
-/**
+/*
* BFA driver interrupt functions
*/
irqreturn_t
return IRQ_HANDLED;
}
-/**
+/*
* Initialize the MSIX entry table.
*/
static void
return 0;
}
-/**
+/*
* Setup MSIX based interrupt.
*/
int
}
}
-/**
+/*
* PCI probe entry.
*/
int
return error;
}
-/**
+/*
* PCI remove entry.
*/
void
.remove = __devexit_p(bfad_pci_remove),
};
-/**
+/*
* Driver module init.
*/
static int __init
return error;
}
-/**
+/*
* Driver module exit.
*/
static void __exit
* General Public License for more details.
*/
-/**
+/*
* bfa_attr.c Linux driver configuration interface module.
*/
#include "bfad_drv.h"
#include "bfad_im.h"
-/**
+/*
* FC transport template entry, get SCSI target port ID.
*/
void
spin_unlock_irqrestore(&bfad->bfad_lock, flags);
}
-/**
+/*
* FC transport template entry, get SCSI target nwwn.
*/
void
if (itnim)
node_name = bfa_fcs_itnim_get_nwwn(&itnim->fcs_itnim);
- fc_starget_node_name(starget) = bfa_os_htonll(node_name);
+ fc_starget_node_name(starget) = cpu_to_be64(node_name);
spin_unlock_irqrestore(&bfad->bfad_lock, flags);
}
-/**
+/*
* FC transport template entry, get SCSI target pwwn.
*/
void
if (itnim)
port_name = bfa_fcs_itnim_get_pwwn(&itnim->fcs_itnim);
- fc_starget_port_name(starget) = bfa_os_htonll(port_name);
+ fc_starget_port_name(starget) = cpu_to_be64(port_name);
spin_unlock_irqrestore(&bfad->bfad_lock, flags);
}
-/**
+/*
* FC transport template entry, get SCSI host port ID.
*/
void
bfa_os_hton3b(bfa_fcs_lport_get_fcid(port->fcs_port));
}
-/**
+/*
* FC transport template entry, get SCSI host port type.
*/
static void
}
}
-/**
+/*
* FC transport template entry, get SCSI host port state.
*/
static void
}
}
-/**
+/*
* FC transport template entry, get SCSI host active fc4s.
*/
static void
fc_host_active_fc4s(shost)[7] = 1;
}
-/**
+/*
* FC transport template entry, get SCSI host link speed.
*/
static void
}
}
-/**
+/*
* FC transport template entry, get SCSI host port type.
*/
static void
fabric_nwwn = bfa_fcs_lport_get_fabric_name(port->fcs_port);
- fc_host_fabric_name(shost) = bfa_os_htonll(fabric_nwwn);
+ fc_host_fabric_name(shost) = cpu_to_be64(fabric_nwwn);
}
-/**
+/*
* FC transport template entry, get BFAD statistics.
*/
static struct fc_host_statistics *
return hstats;
}
-/**
+/*
* FC transport template entry, reset BFAD statistics.
*/
static void
return;
}
-/**
+/*
* FC transport template entry, get rport loss timeout.
*/
static void
spin_unlock_irqrestore(&bfad->bfad_lock, flags);
}
-/**
+/*
* FC transport template entry, set rport loss timeout.
*/
static void
.set_rport_dev_loss_tmo = bfad_im_set_rport_loss_tmo,
};
-/**
+/*
* Scsi_Host_attrs SCSI host attributes
*/
static ssize_t
u64 nwwn;
nwwn = bfa_fcs_lport_get_nwwn(port->fcs_port);
- return snprintf(buf, PAGE_SIZE, "0x%llx\n", bfa_os_htonll(nwwn));
+ return snprintf(buf, PAGE_SIZE, "0x%llx\n", cpu_to_be64(nwwn));
}
static ssize_t
regbuf = (u32 *)bfad->regdata;
spin_lock_irqsave(&bfad->bfad_lock, flags);
for (i = 0; i < len; i++) {
- *regbuf = bfa_reg_read(reg_addr);
+ *regbuf = readl(reg_addr);
regbuf++;
reg_addr += sizeof(u32);
}
reg_addr = (u32 *) ((u8 *) bfa_ioc_bar0(ioc) + addr);
spin_lock_irqsave(&bfad->bfad_lock, flags);
- bfa_reg_write(reg_addr, val);
+ writel(val, reg_addr);
spin_unlock_irqrestore(&bfad->bfad_lock, flags);
return nbytes;
* General Public License for more details.
*/
-/**
+/*
* Contains base driver definitions.
*/
-/**
+/*
* bfa_drv.h Linux driver data structures.
*/
void bfad_init_timer(struct bfad_s *bfad);
int bfad_pci_init(struct pci_dev *pdev, struct bfad_s *bfad);
void bfad_pci_uninit(struct pci_dev *pdev, struct bfad_s *bfad);
-void bfad_fcs_port_cfg(struct bfad_s *bfad);
void bfad_drv_uninit(struct bfad_s *bfad);
int bfad_worker(void *ptr);
void bfad_debugfs_init(struct bfad_port_s *port);
* General Public License for more details.
*/
-/**
+/*
* bfad_im.c Linux driver IM module.
*/
wake_up(wq);
}
-/**
+/*
* Scsi_Host_template SCSI host template
*/
-/**
+/*
* Scsi_Host template entry, returns BFAD PCI info.
*/
static const char *
return bfa_buf;
}
-/**
+/*
* Scsi_Host template entry, aborts the specified SCSI command.
*
* Returns: SUCCESS or FAILED.
return rc;
}
-/**
+/*
* Scsi_Host template entry, resets a LUN and abort its all commands.
*
* Returns: SUCCESS or FAILED.
goto out;
}
- /**
+ /*
* Set host_scribble to NULL to avoid aborting a task command
* if happens.
*/
return rc;
}
-/**
+/*
* Scsi_Host template entry, resets the bus and abort all commands.
*/
static int
return SUCCESS;
}
-/**
+/*
* Scsi_Host template entry slave_destroy.
*/
static void
return;
}
-/**
+/*
* BFA FCS itnim callbacks
*/
-/**
+/*
* BFA FCS itnim alloc callback, after successful PRLI
* Context: Interrupt
*/
bfad->bfad_flags |= BFAD_RPORT_ONLINE;
}
-/**
+/*
* BFA FCS itnim free callback.
* Context: Interrupt. bfad_lock is held
*/
queue_work(im->drv_workq, &itnim_drv->itnim_work);
}
-/**
+/*
* BFA FCS itnim online callback.
* Context: Interrupt. bfad_lock is held
*/
queue_work(im->drv_workq, &itnim_drv->itnim_work);
}
-/**
+/*
* BFA FCS itnim offline callback.
* Context: Interrupt. bfad_lock is held
*/
queue_work(im->drv_workq, &itnim_drv->itnim_work);
}
-/**
+/*
* Allocate a Scsi_Host for a port.
*/
int
return BFA_STATUS_OK;
}
-/**
+/*
* Scsi_Host template entry.
*
* Description:
return NULL;
}
-/**
+/*
* Scsi_Host template entry slave_alloc
*/
static int
static u32
bfad_im_supported_speeds(struct bfa_s *bfa)
{
- struct bfa_ioc_attr_s ioc_attr;
+ struct bfa_ioc_attr_s *ioc_attr;
u32 supported_speed = 0;
- bfa_get_attr(bfa, &ioc_attr);
- if (ioc_attr.adapter_attr.max_speed == BFA_PORT_SPEED_8GBPS) {
- if (ioc_attr.adapter_attr.is_mezz) {
+ ioc_attr = kzalloc(sizeof(struct bfa_ioc_attr_s), GFP_KERNEL);
+ if (!ioc_attr)
+ return 0;
+
+ bfa_get_attr(bfa, ioc_attr);
+ if (ioc_attr->adapter_attr.max_speed == BFA_PORT_SPEED_8GBPS) {
+ if (ioc_attr->adapter_attr.is_mezz) {
supported_speed |= FC_PORTSPEED_8GBIT |
FC_PORTSPEED_4GBIT |
FC_PORTSPEED_2GBIT | FC_PORTSPEED_1GBIT;
FC_PORTSPEED_4GBIT |
FC_PORTSPEED_2GBIT;
}
- } else if (ioc_attr.adapter_attr.max_speed == BFA_PORT_SPEED_4GBPS) {
+ } else if (ioc_attr->adapter_attr.max_speed == BFA_PORT_SPEED_4GBPS) {
supported_speed |= FC_PORTSPEED_4GBIT | FC_PORTSPEED_2GBIT |
FC_PORTSPEED_1GBIT;
- } else if (ioc_attr.adapter_attr.max_speed == BFA_PORT_SPEED_10GBPS) {
+ } else if (ioc_attr->adapter_attr.max_speed == BFA_PORT_SPEED_10GBPS) {
supported_speed |= FC_PORTSPEED_10GBIT;
}
+ kfree(ioc_attr);
return supported_speed;
}
struct Scsi_Host *host = im_port->shost;
struct bfad_s *bfad = im_port->bfad;
struct bfad_port_s *port = im_port->port;
- struct bfa_port_attr_s pattr;
- struct bfa_lport_attr_s port_attr;
char symname[BFA_SYMNAME_MAXLEN];
+ struct bfa_fcport_s *fcport = BFA_FCPORT_MOD(&bfad->bfa);
fc_host_node_name(host) =
- bfa_os_htonll((bfa_fcs_lport_get_nwwn(port->fcs_port)));
+ cpu_to_be64((bfa_fcs_lport_get_nwwn(port->fcs_port)));
fc_host_port_name(host) =
- bfa_os_htonll((bfa_fcs_lport_get_pwwn(port->fcs_port)));
+ cpu_to_be64((bfa_fcs_lport_get_pwwn(port->fcs_port)));
fc_host_max_npiv_vports(host) = bfa_lps_get_max_vport(&bfad->bfa);
fc_host_supported_classes(host) = FC_COS_CLASS3;
/* For fibre channel services type 0x20 */
fc_host_supported_fc4s(host)[7] = 1;
- bfa_fcs_lport_get_attr(&bfad->bfa_fcs.fabric.bport, &port_attr);
- strncpy(symname, port_attr.port_cfg.sym_name.symname,
+ strncpy(symname, bfad->bfa_fcs.fabric.bport.port_cfg.sym_name.symname,
BFA_SYMNAME_MAXLEN);
sprintf(fc_host_symbolic_name(host), "%s", symname);
fc_host_supported_speeds(host) = bfad_im_supported_speeds(&bfad->bfa);
-
- bfa_fcport_get_attr(&bfad->bfa, &pattr);
- fc_host_maxframe_size(host) = pattr.pport_cfg.maxfrsize;
+ fc_host_maxframe_size(host) = fcport->cfg.maxfrsize;
}
static void
struct bfad_itnim_data_s *itnim_data;
rport_ids.node_name =
- bfa_os_htonll(bfa_fcs_itnim_get_nwwn(&itnim->fcs_itnim));
+ cpu_to_be64(bfa_fcs_itnim_get_nwwn(&itnim->fcs_itnim));
rport_ids.port_name =
- bfa_os_htonll(bfa_fcs_itnim_get_pwwn(&itnim->fcs_itnim));
+ cpu_to_be64(bfa_fcs_itnim_get_pwwn(&itnim->fcs_itnim));
rport_ids.port_id =
bfa_os_hton3b(bfa_fcs_itnim_get_fcid(&itnim->fcs_itnim));
rport_ids.roles = FC_RPORT_ROLE_UNKNOWN;
return;
}
-/**
+/*
* Work queue handler using FC transport service
* Context: kernel
*/
spin_unlock_irqrestore(&bfad->bfad_lock, flags);
}
-/**
+/*
* Scsi_Host template entry, queue a SCSI command to the BFAD.
*/
static int
#pragma pack(1)
-/**
+/*
* BFI FW image type
*/
#define BFI_FLASH_CHUNK_SZ 256 /* Flash chunk size */
BFI_IMAGE_MAX,
};
-/**
+/*
* Msg header common to all msgs
*/
struct bfi_mhdr_s {
#define BFI_I2H_OPCODE_BASE 128
#define BFA_I2HM(_x) ((_x) + BFI_I2H_OPCODE_BASE)
-/**
+/*
****************************************************************************
*
* Scatter Gather Element and Page definition
#define BFI_SGE_INLINE 1
#define BFI_SGE_INLINE_MAX (BFI_SGE_INLINE + 1)
-/**
+/*
* SG Flags
*/
enum {
BFI_SGE_PGDLEN = 2, /* cumulative data length for page */
};
-/**
+/*
* DMA addresses
*/
union bfi_addr_u {
} a32;
};
-/**
+/*
* Scatter Gather Element
*/
struct bfi_sge_s {
union bfi_addr_u sga;
};
-/**
+/*
* Scatter Gather Page
*/
#define BFI_SGPG_DATA_SGES 7
u32 pl[BFI_LMSG_PL_WSZ];
};
-/**
+/*
* Mailbox message structure
*/
#define BFI_MBMSG_SZ 7
u32 pl[BFI_MBMSG_SZ];
};
-/**
+/*
* Message Classes
*/
enum bfi_mclass {
#define BFI_BOOT_LOADER_BIOS 1
#define BFI_BOOT_LOADER_UEFI 2
-/**
+/*
*----------------------------------------------------------------------
* IOC
*----------------------------------------------------------------------
BFI_IOC_I2H_HBEAT = BFA_I2HM(5),
};
-/**
+/*
* BFI_IOC_H2I_GETATTR_REQ message
*/
struct bfi_ioc_getattr_req_s {
u32 card_type; /* card type */
};
-/**
+/*
* BFI_IOC_I2H_GETATTR_REPLY message
*/
struct bfi_ioc_getattr_reply_s {
u8 rsvd[3];
};
-/**
+/*
* Firmware memory page offsets
*/
#define BFI_IOC_SMEM_PG0_CB (0x40)
#define BFI_IOC_SMEM_PG0_CT (0x180)
-/**
+/*
* Firmware statistic offset
*/
#define BFI_IOC_FWSTATS_OFF (0x6B40)
#define BFI_IOC_FWSTATS_SZ (4096)
-/**
+/*
* Firmware trace offset
*/
#define BFI_IOC_TRC_OFF (0x4b00)
u32 md5sum[BFI_IOC_MD5SUM_SZ];
};
-/**
+/*
* BFI_IOC_I2H_READY_EVENT message
*/
struct bfi_ioc_rdy_event_s {
u32 hb_count; /* current heart beat count */
};
-/**
+/*
* IOC hardware/firmware state
*/
enum bfi_ioc_state {
((__adap_type) & (BFI_ADAPTER_TTV | BFI_ADAPTER_PROTO | \
BFI_ADAPTER_UNSUPP))
-/**
+/*
* BFI_IOC_H2I_ENABLE_REQ & BFI_IOC_H2I_DISABLE_REQ messages
*/
struct bfi_ioc_ctrl_req_s {
#define bfi_ioc_enable_req_t struct bfi_ioc_ctrl_req_s;
#define bfi_ioc_disable_req_t struct bfi_ioc_ctrl_req_s;
-/**
+/*
* BFI_IOC_I2H_ENABLE_REPLY & BFI_IOC_I2H_DISABLE_REPLY messages
*/
struct bfi_ioc_ctrl_reply_s {
#define bfi_ioc_disable_reply_t struct bfi_ioc_ctrl_reply_s;
#define BFI_IOC_MSGSZ 8
-/**
+/*
* H2I Messages
*/
union bfi_ioc_h2i_msg_u {
u32 mboxmsg[BFI_IOC_MSGSZ];
};
-/**
+/*
* I2H Messages
*/
union bfi_ioc_i2h_msg_u {
};
-/**
+/*
*----------------------------------------------------------------------
* PBC
*----------------------------------------------------------------------
#define BFI_PBC_MAX_BLUNS 8
#define BFI_PBC_MAX_VPORTS 16
-/**
+/*
* PBC boot lun configuration
*/
struct bfi_pbc_blun_s {
lun_t tgt_lun;
};
-/**
+/*
* PBC virtual port configuration
*/
struct bfi_pbc_vport_s {
wwn_t vp_nwwn;
};
-/**
+/*
* BFI pre-boot configuration information
*/
struct bfi_pbc_s {
struct bfi_pbc_vport_s vport[BFI_PBC_MAX_VPORTS];
};
-/**
+/*
*----------------------------------------------------------------------
* MSGQ
*----------------------------------------------------------------------
BFI_PORT_I2H_CLEAR_STATS_RSP = BFA_I2HM(4),
};
-/**
+/*
* Generic REQ type
*/
struct bfi_port_generic_req_s {
u32 rsvd;
};
-/**
+/*
* Generic RSP type
*/
struct bfi_port_generic_rsp_s {
u32 msgtag; /* msgtag for reply */
};
-/**
+/*
* BFI_PORT_H2I_GET_STATS_REQ
*/
struct bfi_port_get_stats_req_s {
u16 rsvd_1;
u32 endian_sig; /* endian signature of host */
- /**
+ /*
* Request and response circular queue base addresses, size and
* shadow index pointers.
*/
struct bfa_iocfc_intr_attr_s intr_attr; /* IOC interrupt attributes */
};
-/**
+/*
* Boot target wwn information for this port. This contains either the stored
* or discovered boot target port wwns for the port.
*/
struct bfi_pbc_s pbc_cfg;
};
-/**
+/*
* BFI_IOCFC_H2I_CFG_REQ message
*/
struct bfi_iocfc_cfg_req_s {
};
-/**
+/*
* BFI_IOCFC_I2H_CFG_REPLY message
*/
struct bfi_iocfc_cfg_reply_s {
};
-/**
+/*
* BFI_IOCFC_H2I_SET_INTR_REQ message
*/
struct bfi_iocfc_set_intr_req_s {
};
-/**
+/*
* BFI_IOCFC_H2I_UPDATEQ_REQ message
*/
struct bfi_iocfc_updateq_req_s {
};
-/**
+/*
* BFI_IOCFC_I2H_UPDATEQ_RSP message
*/
struct bfi_iocfc_updateq_rsp_s {
};
-/**
+/*
* H2I Messages
*/
union bfi_iocfc_h2i_msg_u {
};
-/**
+/*
* I2H Messages
*/
union bfi_iocfc_i2h_msg_u {
};
-/**
+/*
* Generic REQ type
*/
struct bfi_fcport_req_s {
u32 msgtag; /* msgtag for reply */
};
-/**
+/*
* Generic RSP type
*/
struct bfi_fcport_rsp_s {
u32 msgtag; /* msgtag for reply */
};
-/**
+/*
* BFI_FCPORT_H2I_ENABLE_REQ
*/
struct bfi_fcport_enable_req_s {
u32 rsvd2;
};
-/**
+/*
* BFI_FCPORT_H2I_SET_SVC_PARAMS_REQ
*/
struct bfi_fcport_set_svc_params_req_s {
u16 rsvd;
};
-/**
+/*
* BFI_FCPORT_I2H_EVENT
*/
struct bfi_fcport_event_s {
struct bfa_port_link_s link_state;
};
-/**
+/*
* BFI_FCPORT_I2H_TRUNK_SCN
*/
struct bfi_fcport_trunk_link_s {
struct bfi_fcport_trunk_link_s tlink[BFI_FCPORT_MAX_LINKS];
};
-/**
+/*
* fcport H2I message
*/
union bfi_fcport_h2i_msg_u {
struct bfi_fcport_req_s *pstatsclear;
};
-/**
+/*
* fcport I2H message
*/
union bfi_fcport_i2h_msg_u {
#define BFA_FCXP_MAX_SGES 2
-/**
+/*
* FCXP send request structure
*/
struct bfi_fcxp_send_req_s {
struct bfi_sge_s rsp_sge[BFA_FCXP_MAX_SGES]; /* response buf */
};
-/**
+/*
* FCXP send response structure
*/
struct bfi_fcxp_send_rsp_s {
BFI_IOIM_I2H_IOABORT_RSP = BFA_I2HM(2), /* ABORT rsp */
};
-/**
+/*
* IO command DIF info
*/
struct bfi_ioim_dif_s {
u32 dif_info[4];
};
-/**
+/*
* FCP IO messages overview
*
* @note
u16 rport_hdl; /* itnim/rport firmware handle */
struct fcp_cmnd_s cmnd; /* IO request info */
- /**
+ /*
* SG elements array within the IO request must be double word
* aligned. This aligment is required to optimize SGM setup for the IO.
*/
struct bfi_ioim_dif_s dif;
};
-/**
+/*
* This table shows various IO status codes from firmware and their
* meaning. Host driver can use these status codes to further process
* IO completions.
};
#define BFI_IOIM_SNSLEN (256)
-/**
+/*
* I/O response message
*/
struct bfi_ioim_rsp_s {
BFI_TSKIM_STS_NOT_SUPP = 4,
BFI_TSKIM_STS_FAILED = 5,
- /**
+ /*
* Defined by BFA
*/
BFI_TSKIM_STS_TIMEOUT = 10, /* TM request timedout */
&csk->daddr.sin_addr.s_addr, ntohs(csk->daddr.sin_port),
atid, tid, status, csk, csk->state, csk->flags);
+ if (status == CPL_ERR_RTX_NEG_ADVICE)
+ goto rel_skb;
+
if (status && status != CPL_ERR_TCAM_FULL &&
status != CPL_ERR_CONN_EXIST &&
status != CPL_ERR_ARP_MISS)
{"ENGENIO", "INF-01-00"},
{"STK", "FLEXLINE 380"},
{"SUN", "CSM100_R_FC"},
+ {"SUN", "STK6580_6780"},
+ {"SUN", "SUN_6180"},
{NULL, NULL},
};
static void fcoe_get_lesb(struct fc_lport *, struct fc_els_lesb *);
-module_param_call(create, fcoe_create, NULL, (void *)FIP_MODE_AUTO, S_IWUSR);
+module_param_call(create, fcoe_create, NULL, (void *)FIP_MODE_FABRIC, S_IWUSR);
__MODULE_PARM_TYPE(create, "string");
MODULE_PARM_DESC(create, " Creates fcoe instance on a ethernet interface");
module_param_call(create_vn2vn, fcoe_create, NULL,
struct fcoe_interface *fcoe;
struct fc_frame_header *fh;
struct fcoe_percpu_s *fps;
- struct fcoe_port *port;
struct ethhdr *eh;
unsigned int cpu;
skb_tail_pointer(skb), skb_end_pointer(skb),
skb->csum, skb->dev ? skb->dev->name : "<NULL>");
- /* check for mac addresses */
eh = eth_hdr(skb);
- port = lport_priv(lport);
- if (compare_ether_addr(eh->h_dest, port->data_src_addr) &&
- compare_ether_addr(eh->h_dest, fcoe->ctlr.ctl_src_addr) &&
- compare_ether_addr(eh->h_dest, (u8[6])FC_FCOE_FLOGI_MAC)) {
- FCOE_NETDEV_DBG(netdev, "wrong destination mac address:%pM\n",
- eh->h_dest);
- goto err;
- }
if (is_fip_mode(&fcoe->ctlr) &&
compare_ether_addr(eh->h_source, fcoe->ctlr.dest_addr)) {
skb_set_transport_header(skb, sizeof(struct fcoe_hdr));
fh = (struct fc_frame_header *) skb_transport_header(skb);
+ if (ntoh24(&eh->h_dest[3]) != ntoh24(fh->fh_d_id)) {
+ FCOE_NETDEV_DBG(netdev, "FC frame d_id mismatch with MAC:%pM\n",
+ eh->h_dest);
+ goto err;
+ }
+
fr = fcoe_dev_from_skb(skb);
fr->fr_dev = lport;
fr->ptype = ptype;
{
struct fip_header *fiph;
enum fip_vn2vn_subcode sub;
- union {
+ struct {
struct fc_rport_priv rdata;
struct fcoe_rport frport;
} buf;
ha = gdth_find_ha(gen.ionode);
if (!ha)
return -EFAULT;
+
+ if (gen.data_len > INT_MAX)
+ return -EINVAL;
+ if (gen.sense_len > INT_MAX)
+ return -EINVAL;
+ if (gen.data_len + gen.sense_len > INT_MAX)
+ return -EINVAL;
+
if (gen.data_len + gen.sense_len != 0) {
if (!(buf = gdth_ioctl_alloc(ha, gen.data_len + gen.sense_len,
FALSE, &paddr)))
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_574D, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROC_FPGA_E2,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57B2, 0, 0, 0 },
+ { PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROC_FPGA_E2,
+ PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57C4, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROC_ASIC_E2,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57B4, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROC_ASIC_E2,
#define IPR_SUBS_DEV_ID_57B4 0x033B
#define IPR_SUBS_DEV_ID_57B2 0x035F
+#define IPR_SUBS_DEV_ID_57C4 0x0354
#define IPR_SUBS_DEV_ID_57C6 0x0357
#define IPR_SUBS_DEV_ID_57CC 0x035C
{
struct fc_disc *disc = &lport->disc;
- if (disc) {
+ if (disc->pending)
cancel_delayed_work_sync(&disc->disc_work);
- fc_disc_stop_rports(disc);
- }
+ fc_disc_stop_rports(disc);
}
/**
#define FC_SRB_WRITE (1 << 0)
/*
- * The SCp.ptr should be tested and set under the host lock. NULL indicates
- * that the command has been retruned to the scsi layer.
+ * The SCp.ptr should be tested and set under the scsi_pkt_queue lock
*/
#define CMD_SP(Cmnd) ((struct fc_fcp_pkt *)(Cmnd)->SCp.ptr)
#define CMD_ENTRY_STATUS(Cmnd) ((Cmnd)->SCp.have_data_in)
lport = fsp->lp;
si = fc_get_scsi_internal(lport);
- if (!fsp->cmd)
- return;
/*
* if can_queue ramp down is done then try can_queue ramp up
fc_fcp_can_queue_ramp_up(lport);
sc_cmd = fsp->cmd;
- fsp->cmd = NULL;
-
- if (!sc_cmd->SCp.ptr)
- return;
-
CMD_SCSI_STATUS(sc_cmd) = fsp->cdb_status;
switch (fsp->status_code) {
case FC_COMPLETE:
break;
}
- if (lport->state != LPORT_ST_READY && fsp->status_code != FC_COMPLETE) {
- sc_cmd->result = (DID_REQUEUE << 16);
- FC_FCP_DBG(fsp, "Returning DID_REQUEUE to scsi-ml\n");
- }
+ if (lport->state != LPORT_ST_READY && fsp->status_code != FC_COMPLETE)
+ sc_cmd->result = (DID_TRANSPORT_DISRUPTED << 16);
spin_lock_irqsave(&si->scsi_queue_lock, flags);
list_del(&fsp->list);
- spin_unlock_irqrestore(&si->scsi_queue_lock, flags);
sc_cmd->SCp.ptr = NULL;
+ spin_unlock_irqrestore(&si->scsi_queue_lock, flags);
sc_cmd->scsi_done(sc_cmd);
/* release ref from initial allocation in queue command */
{
struct fc_fcp_pkt *fsp;
struct fc_lport *lport;
+ struct fc_fcp_internal *si;
int rc = FAILED;
unsigned long flags;
else if (!lport->link_up)
return rc;
- spin_lock_irqsave(lport->host->host_lock, flags);
+ si = fc_get_scsi_internal(lport);
+ spin_lock_irqsave(&si->scsi_queue_lock, flags);
fsp = CMD_SP(sc_cmd);
if (!fsp) {
/* command completed while scsi eh was setting up */
}
/* grab a ref so the fsp and sc_cmd cannot be relased from under us */
fc_fcp_pkt_hold(fsp);
- spin_unlock_irqrestore(lport->host->host_lock, flags);
+ spin_unlock_irqrestore(&si->scsi_queue_lock, flags);
if (fc_fcp_lock_pkt(fsp)) {
/* completed while we were waiting for timer to be deleted */
}
did = fc_frame_did(fp);
-
- if (!did) {
- FC_LPORT_DBG(lport, "Bad FLOGI response\n");
- goto out;
- }
-
- if (fc_frame_payload_op(fp) == ELS_LS_ACC) {
+ if (fc_frame_payload_op(fp) == ELS_LS_ACC && did) {
flp = fc_frame_payload_get(fp, sizeof(*flp));
if (flp) {
mfs = ntohs(flp->fl_csp.sp_bb_data) &
fc_lport_enter_dns(lport);
}
}
- } else
+ } else {
+ FC_LPORT_DBG(lport, "FLOGI RJT or bad response\n");
fc_lport_error(lport, fp);
+ }
out:
fc_frame_free(fp);
void fc_set_rport_loss_tmo(struct fc_rport *rport, u32 timeout)
{
if (timeout)
- rport->dev_loss_tmo = timeout + 5;
+ rport->dev_loss_tmo = timeout;
else
- rport->dev_loss_tmo = 30;
+ rport->dev_loss_tmo = 1;
}
EXPORT_SYMBOL(fc_set_rport_loss_tmo);
uint32_t elsRcvPRLO;
uint32_t elsRcvPRLI;
uint32_t elsRcvLIRR;
+ uint32_t elsRcvRLS;
uint32_t elsRcvRPS;
uint32_t elsRcvRPL;
uint32_t elsRcvRRQ;
+ uint32_t elsRcvRTV;
+ uint32_t elsRcvECHO;
uint32_t elsXmitFLOGI;
uint32_t elsXmitFDISC;
uint32_t elsXmitPLOGI;
#define ELS_XRI_ABORT_EVENT 0x40
#define ASYNC_EVENT 0x80
#define LINK_DISABLED 0x100 /* Link disabled by user */
-#define FCF_DISC_INPROGRESS 0x200 /* FCF discovery in progress */
-#define HBA_FIP_SUPPORT 0x400 /* FIP support in HBA */
-#define HBA_AER_ENABLED 0x800 /* AER enabled with HBA */
+#define FCF_TS_INPROG 0x200 /* FCF table scan in progress */
+#define FCF_RR_INPROG 0x400 /* FCF roundrobin flogi in progress */
+#define HBA_FIP_SUPPORT 0x800 /* FIP support in HBA */
+#define HBA_AER_ENABLED 0x1000 /* AER enabled with HBA */
+#define HBA_DEVLOSS_TMO 0x2000 /* HBA in devloss timeout */
uint32_t fcp_ring_in_use; /* When polling test if intr-hndlr active*/
struct lpfc_dmabuf slim2p;
/* These fields used to be binfo */
uint32_t fc_pref_DID; /* preferred D_ID */
uint8_t fc_pref_ALPA; /* preferred AL_PA */
+ uint32_t fc_edtovResol; /* E_D_TOV timer resolution */
uint32_t fc_edtov; /* E_D_TOV timer value */
uint32_t fc_arbtov; /* ARB_TOV timer value */
uint32_t fc_ratov; /* R_A_TOV timer value */
break;
case MBX_SECURITY_MGMT:
case MBX_AUTH_PORT:
- if (phba->pci_dev_grp == LPFC_PCI_DEV_OC)
+ if (phba->pci_dev_grp == LPFC_PCI_DEV_OC) {
+ printk(KERN_WARNING "mbox_read:Command 0x%x "
+ "is not permitted\n", pmb->mbxCommand);
+ sysfs_mbox_idle(phba);
+ spin_unlock_irq(&phba->hbalock);
return -EPERM;
+ }
break;
case MBX_READ_SPARM64:
case MBX_READ_LA:
job = menlo->set_job;
job->dd_data = NULL; /* so timeout handler does not reply */
- spin_lock_irqsave(&phba->hbalock, flags);
+ spin_lock(&phba->hbalock);
cmdiocbq->iocb_flag |= LPFC_IO_WAKE;
if (cmdiocbq->context2 && rspiocbq)
memcpy(&((struct lpfc_iocbq *)cmdiocbq->context2)->iocb,
&rspiocbq->iocb, sizeof(IOCB_t));
- spin_unlock_irqrestore(&phba->hbalock, flags);
+ spin_unlock(&phba->hbalock);
bmp = menlo->bmp;
rspiocbq = menlo->rspiocbq;
void lpfc_set_var(struct lpfc_hba *, LPFC_MBOXQ_t *, uint32_t, uint32_t);
void lpfc_unreg_login(struct lpfc_hba *, uint16_t, uint32_t, LPFC_MBOXQ_t *);
void lpfc_unreg_did(struct lpfc_hba *, uint16_t, uint32_t, LPFC_MBOXQ_t *);
+void lpfc_sli4_unreg_all_rpis(struct lpfc_vport *);
+
void lpfc_reg_vpi(struct lpfc_vport *, LPFC_MBOXQ_t *);
void lpfc_register_new_vport(struct lpfc_hba *, struct lpfc_vport *,
struct lpfc_nodelist *);
uint16_t lpfc_sli4_fcf_rr_next_index_get(struct lpfc_hba *);
int lpfc_sli4_fcf_rr_index_set(struct lpfc_hba *, uint16_t);
void lpfc_sli4_fcf_rr_index_clear(struct lpfc_hba *, uint16_t);
+int lpfc_sli4_fcf_rr_next_proc(struct lpfc_vport *, uint16_t);
int lpfc_mem_alloc(struct lpfc_hba *, int align);
void lpfc_mem_free(struct lpfc_hba *);
void lpfc_sli_pcimem_bcopy(void *, void *, uint32_t);
void lpfc_sli_bemem_bcopy(void *, void *, uint32_t);
void lpfc_sli_abort_iocb_ring(struct lpfc_hba *, struct lpfc_sli_ring *);
+void lpfc_sli_hba_iocb_abort(struct lpfc_hba *);
void lpfc_sli_flush_fcp_rings(struct lpfc_hba *);
int lpfc_sli_ringpostbuf_put(struct lpfc_hba *, struct lpfc_sli_ring *,
struct lpfc_dmabuf *);
(elscmd == ELS_CMD_LOGO)))
switch (elscmd) {
case ELS_CMD_FLOGI:
- elsiocb->iocb_flag |= ((ELS_ID_FLOGI << LPFC_FIP_ELS_ID_SHIFT)
+ elsiocb->iocb_flag |=
+ ((LPFC_ELS_ID_FLOGI << LPFC_FIP_ELS_ID_SHIFT)
& LPFC_FIP_ELS_ID_MASK);
break;
case ELS_CMD_FDISC:
- elsiocb->iocb_flag |= ((ELS_ID_FDISC << LPFC_FIP_ELS_ID_SHIFT)
+ elsiocb->iocb_flag |=
+ ((LPFC_ELS_ID_FDISC << LPFC_FIP_ELS_ID_SHIFT)
& LPFC_FIP_ELS_ID_MASK);
break;
case ELS_CMD_LOGO:
- elsiocb->iocb_flag |= ((ELS_ID_LOGO << LPFC_FIP_ELS_ID_SHIFT)
+ elsiocb->iocb_flag |=
+ ((LPFC_ELS_ID_LOGO << LPFC_FIP_ELS_ID_SHIFT)
& LPFC_FIP_ELS_ID_MASK);
break;
}
if (sp->cmn.edtovResolution) /* E_D_TOV ticks are in nanoseconds */
phba->fc_edtov = (phba->fc_edtov + 999999) / 1000000;
+ phba->fc_edtovResol = sp->cmn.edtovResolution;
phba->fc_ratov = (be32_to_cpu(sp->cmn.w2.r_a_tov) + 999) / 1000;
if (phba->fc_topology == TOPOLOGY_LOOP) {
spin_lock_irq(shost->host_lock);
vport->fc_flag |= FC_PUBLIC_LOOP;
spin_unlock_irq(shost->host_lock);
- } else {
- /*
- * If we are a N-port connected to a Fabric, fixup sparam's so
- * logins to devices on remote loops work.
- */
- vport->fc_sparam.cmn.altBbCredit = 1;
}
vport->fc_myDID = irsp->un.ulpWord[4] & Mask_DID;
lpfc_unreg_rpi(vport, np);
}
lpfc_cleanup_pending_mbox(vport);
+
+ if (phba->sli_rev == LPFC_SLI_REV4)
+ lpfc_sli4_unreg_all_rpis(vport);
+
if (phba->sli3_options & LPFC_SLI3_NPIV_ENABLED) {
lpfc_mbx_unreg_vpi(vport);
spin_lock_irq(shost->host_lock);
if (irsp->ulpStatus) {
/*
- * In case of FIP mode, perform round robin FCF failover
+ * In case of FIP mode, perform roundrobin FCF failover
* due to new FCF discovery
*/
if ((phba->hba_flag & HBA_FIP_SUPPORT) &&
(irsp->ulpStatus != IOSTAT_LOCAL_REJECT) &&
(irsp->un.ulpWord[4] != IOERR_SLI_ABORTED)) {
lpfc_printf_log(phba, KERN_WARNING, LOG_FIP | LOG_ELS,
- "2611 FLOGI failed on registered "
- "FCF record fcf_index(%d), status: "
- "x%x/x%x, tmo:x%x, trying to perform "
- "round robin failover\n",
+ "2611 FLOGI failed on FCF (x%x), "
+ "status:x%x/x%x, tmo:x%x, perform "
+ "roundrobin FCF failover\n",
phba->fcf.current_rec.fcf_indx,
irsp->ulpStatus, irsp->un.ulpWord[4],
irsp->ulpTimeout);
fcf_index = lpfc_sli4_fcf_rr_next_index_get(phba);
- if (fcf_index == LPFC_FCOE_FCF_NEXT_NONE) {
- /*
- * Exhausted the eligible FCF record list,
- * fail through to retry FLOGI on current
- * FCF record.
- */
- lpfc_printf_log(phba, KERN_WARNING,
- LOG_FIP | LOG_ELS,
- "2760 Completed one round "
- "of FLOGI FCF round robin "
- "failover list, retry FLOGI "
- "on currently registered "
- "FCF index:%d\n",
- phba->fcf.current_rec.fcf_indx);
- } else {
- lpfc_printf_log(phba, KERN_INFO,
- LOG_FIP | LOG_ELS,
- "2794 FLOGI FCF round robin "
- "failover to FCF index x%x\n",
- fcf_index);
- rc = lpfc_sli4_fcf_rr_read_fcf_rec(phba,
- fcf_index);
- if (rc)
- lpfc_printf_log(phba, KERN_WARNING,
- LOG_FIP | LOG_ELS,
- "2761 FLOGI round "
- "robin FCF failover "
- "read FCF failed "
- "rc:x%x, fcf_index:"
- "%d\n", rc,
- phba->fcf.current_rec.fcf_indx);
- else
- goto out;
- }
+ rc = lpfc_sli4_fcf_rr_next_proc(vport, fcf_index);
+ if (rc)
+ goto out;
}
/* FLOGI failure */
lpfc_nlp_put(ndlp);
spin_lock_irq(&phba->hbalock);
phba->fcf.fcf_flag &= ~FCF_DISCOVERY;
+ phba->hba_flag &= ~(FCF_RR_INPROG | HBA_DEVLOSS_TMO);
spin_unlock_irq(&phba->hbalock);
goto out;
}
if (phba->hba_flag & HBA_FIP_SUPPORT)
lpfc_printf_vlog(vport, KERN_INFO, LOG_FIP |
LOG_ELS,
- "2769 FLOGI successful on FCF "
- "record: current_fcf_index:"
- "x%x, terminate FCF round "
- "robin failover process\n",
+ "2769 FLOGI to FCF (x%x) "
+ "completed successfully\n",
phba->fcf.current_rec.fcf_indx);
spin_lock_irq(&phba->hbalock);
phba->fcf.fcf_flag &= ~FCF_DISCOVERY;
+ phba->hba_flag &= ~(FCF_RR_INPROG | HBA_DEVLOSS_TMO);
spin_unlock_irq(&phba->hbalock);
goto out;
}
return 0;
}
- if (lpfc_issue_els_flogi(vport, ndlp, 0))
+ if (lpfc_issue_els_flogi(vport, ndlp, 0)) {
/* This decrement of reference count to node shall kick off
* the release of the node.
*/
lpfc_nlp_put(ndlp);
-
+ return 0;
+ }
return 1;
}
memcpy(pcmd, &vport->fc_sparam, sizeof(struct serv_parm));
sp = (struct serv_parm *) pcmd;
+ /*
+ * If we are a N-port connected to a Fabric, fix-up paramm's so logins
+ * to device on remote loops work.
+ */
+ if ((vport->fc_flag & FC_FABRIC) && !(vport->fc_flag & FC_PUBLIC_LOOP))
+ sp->cmn.altBbCredit = 1;
+
if (sp->cmn.fcphLow < FC_PH_4_3)
sp->cmn.fcphLow = FC_PH_4_3;
return 0;
}
+/**
+ * lpfc_els_rsp_echo_acc - Issue echo acc response
+ * @vport: pointer to a virtual N_Port data structure.
+ * @data: pointer to echo data to return in the accept.
+ * @oldiocb: pointer to the original lpfc command iocb data structure.
+ * @ndlp: pointer to a node-list data structure.
+ *
+ * Return code
+ * 0 - Successfully issued acc echo response
+ * 1 - Failed to issue acc echo response
+ **/
+static int
+lpfc_els_rsp_echo_acc(struct lpfc_vport *vport, uint8_t *data,
+ struct lpfc_iocbq *oldiocb, struct lpfc_nodelist *ndlp)
+{
+ struct lpfc_hba *phba = vport->phba;
+ struct lpfc_iocbq *elsiocb;
+ struct lpfc_sli *psli;
+ uint8_t *pcmd;
+ uint16_t cmdsize;
+ int rc;
+
+ psli = &phba->sli;
+ cmdsize = oldiocb->iocb.unsli3.rcvsli3.acc_len;
+
+ elsiocb = lpfc_prep_els_iocb(vport, 0, cmdsize, oldiocb->retry, ndlp,
+ ndlp->nlp_DID, ELS_CMD_ACC);
+ if (!elsiocb)
+ return 1;
+
+ elsiocb->iocb.ulpContext = oldiocb->iocb.ulpContext; /* Xri */
+ /* Xmit ECHO ACC response tag <ulpIoTag> */
+ lpfc_printf_vlog(vport, KERN_INFO, LOG_ELS,
+ "2876 Xmit ECHO ACC response tag x%x xri x%x\n",
+ elsiocb->iotag, elsiocb->iocb.ulpContext);
+ pcmd = (uint8_t *) (((struct lpfc_dmabuf *) elsiocb->context2)->virt);
+ *((uint32_t *) (pcmd)) = ELS_CMD_ACC;
+ pcmd += sizeof(uint32_t);
+ memcpy(pcmd, data, cmdsize - sizeof(uint32_t));
+
+ lpfc_debugfs_disc_trc(vport, LPFC_DISC_TRC_ELS_RSP,
+ "Issue ACC ECHO: did:x%x flg:x%x",
+ ndlp->nlp_DID, ndlp->nlp_flag, 0);
+
+ phba->fc_stat.elsXmitACC++;
+ elsiocb->iocb_cmpl = lpfc_cmpl_els_rsp;
+ lpfc_nlp_put(ndlp);
+ elsiocb->context1 = NULL; /* Don't need ndlp for cmpl,
+ * it could be freed */
+
+ rc = lpfc_sli_issue_iocb(phba, LPFC_ELS_RING, elsiocb, 0);
+ if (rc == IOCB_ERROR) {
+ lpfc_els_free_iocb(phba, elsiocb);
+ return 1;
+ }
+ return 0;
+}
+
/**
* lpfc_els_disc_adisc - Issue remaining adisc iocbs to npr nodes of a vport
* @vport: pointer to a host virtual N_Port data structure.
return 0;
}
+/**
+ * lpfc_els_rcv_echo - Process an unsolicited echo iocb
+ * @vport: pointer to a host virtual N_Port data structure.
+ * @cmdiocb: pointer to lpfc command iocb data structure.
+ * @ndlp: pointer to a node-list data structure.
+ *
+ * Return code
+ * 0 - Successfully processed echo iocb (currently always return 0)
+ **/
+static int
+lpfc_els_rcv_echo(struct lpfc_vport *vport, struct lpfc_iocbq *cmdiocb,
+ struct lpfc_nodelist *ndlp)
+{
+ uint8_t *pcmd;
+
+ pcmd = (uint8_t *) (((struct lpfc_dmabuf *) cmdiocb->context2)->virt);
+
+ /* skip over first word of echo command to find echo data */
+ pcmd += sizeof(uint32_t);
+
+ lpfc_els_rsp_echo_acc(vport, pcmd, cmdiocb, ndlp);
+ return 0;
+}
+
/**
* lpfc_els_rcv_lirr - Process an unsolicited lirr iocb
* @vport: pointer to a host virtual N_Port data structure.
lpfc_els_rsp_acc(vport, ELS_CMD_ACC, cmdiocb, ndlp, NULL);
}
+/**
+ * lpfc_els_rsp_rls_acc - Completion callbk func for MBX_READ_LNK_STAT mbox cmd
+ * @phba: pointer to lpfc hba data structure.
+ * @pmb: pointer to the driver internal queue element for mailbox command.
+ *
+ * This routine is the completion callback function for the MBX_READ_LNK_STAT
+ * mailbox command. This callback function is to actually send the Accept
+ * (ACC) response to a Read Port Status (RPS) unsolicited IOCB event. It
+ * collects the link statistics from the completion of the MBX_READ_LNK_STAT
+ * mailbox command, constructs the RPS response with the link statistics
+ * collected, and then invokes the lpfc_sli_issue_iocb() routine to send ACC
+ * response to the RPS.
+ *
+ * Note that, in lpfc_prep_els_iocb() routine, the reference count of ndlp
+ * will be incremented by 1 for holding the ndlp and the reference to ndlp
+ * will be stored into the context1 field of the IOCB for the completion
+ * callback function to the RPS Accept Response ELS IOCB command.
+ *
+ **/
+static void
+lpfc_els_rsp_rls_acc(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
+{
+ MAILBOX_t *mb;
+ IOCB_t *icmd;
+ struct RLS_RSP *rls_rsp;
+ uint8_t *pcmd;
+ struct lpfc_iocbq *elsiocb;
+ struct lpfc_nodelist *ndlp;
+ uint16_t xri;
+ uint32_t cmdsize;
+
+ mb = &pmb->u.mb;
+
+ ndlp = (struct lpfc_nodelist *) pmb->context2;
+ xri = (uint16_t) ((unsigned long)(pmb->context1));
+ pmb->context1 = NULL;
+ pmb->context2 = NULL;
+
+ if (mb->mbxStatus) {
+ mempool_free(pmb, phba->mbox_mem_pool);
+ return;
+ }
+
+ cmdsize = sizeof(struct RLS_RSP) + sizeof(uint32_t);
+ mempool_free(pmb, phba->mbox_mem_pool);
+ elsiocb = lpfc_prep_els_iocb(phba->pport, 0, cmdsize,
+ lpfc_max_els_tries, ndlp,
+ ndlp->nlp_DID, ELS_CMD_ACC);
+
+ /* Decrement the ndlp reference count from previous mbox command */
+ lpfc_nlp_put(ndlp);
+
+ if (!elsiocb)
+ return;
+
+ icmd = &elsiocb->iocb;
+ icmd->ulpContext = xri;
+
+ pcmd = (uint8_t *) (((struct lpfc_dmabuf *) elsiocb->context2)->virt);
+ *((uint32_t *) (pcmd)) = ELS_CMD_ACC;
+ pcmd += sizeof(uint32_t); /* Skip past command */
+ rls_rsp = (struct RLS_RSP *)pcmd;
+
+ rls_rsp->linkFailureCnt = cpu_to_be32(mb->un.varRdLnk.linkFailureCnt);
+ rls_rsp->lossSyncCnt = cpu_to_be32(mb->un.varRdLnk.lossSyncCnt);
+ rls_rsp->lossSignalCnt = cpu_to_be32(mb->un.varRdLnk.lossSignalCnt);
+ rls_rsp->primSeqErrCnt = cpu_to_be32(mb->un.varRdLnk.primSeqErrCnt);
+ rls_rsp->invalidXmitWord = cpu_to_be32(mb->un.varRdLnk.invalidXmitWord);
+ rls_rsp->crcCnt = cpu_to_be32(mb->un.varRdLnk.crcCnt);
+
+ /* Xmit ELS RLS ACC response tag <ulpIoTag> */
+ lpfc_printf_vlog(ndlp->vport, KERN_INFO, LOG_ELS,
+ "2874 Xmit ELS RLS ACC response tag x%x xri x%x, "
+ "did x%x, nlp_flag x%x, nlp_state x%x, rpi x%x\n",
+ elsiocb->iotag, elsiocb->iocb.ulpContext,
+ ndlp->nlp_DID, ndlp->nlp_flag, ndlp->nlp_state,
+ ndlp->nlp_rpi);
+ elsiocb->iocb_cmpl = lpfc_cmpl_els_rsp;
+ phba->fc_stat.elsXmitACC++;
+ if (lpfc_sli_issue_iocb(phba, LPFC_ELS_RING, elsiocb, 0) == IOCB_ERROR)
+ lpfc_els_free_iocb(phba, elsiocb);
+}
+
/**
* lpfc_els_rsp_rps_acc - Completion callbk func for MBX_READ_LNK_STAT mbox cmd
* @phba: pointer to lpfc hba data structure.
}
/**
- * lpfc_els_rcv_rps - Process an unsolicited rps iocb
+ * lpfc_els_rcv_rls - Process an unsolicited rls iocb
+ * @vport: pointer to a host virtual N_Port data structure.
+ * @cmdiocb: pointer to lpfc command iocb data structure.
+ * @ndlp: pointer to a node-list data structure.
+ *
+ * This routine processes Read Port Status (RPL) IOCB received as an
+ * ELS unsolicited event. It first checks the remote port state. If the
+ * remote port is not in NLP_STE_UNMAPPED_NODE state or NLP_STE_MAPPED_NODE
+ * state, it invokes the lpfc_els_rsl_reject() routine to send the reject
+ * response. Otherwise, it issue the MBX_READ_LNK_STAT mailbox command
+ * for reading the HBA link statistics. It is for the callback function,
+ * lpfc_els_rsp_rls_acc(), set to the MBX_READ_LNK_STAT mailbox command
+ * to actually sending out RPL Accept (ACC) response.
+ *
+ * Return codes
+ * 0 - Successfully processed rls iocb (currently always return 0)
+ **/
+static int
+lpfc_els_rcv_rls(struct lpfc_vport *vport, struct lpfc_iocbq *cmdiocb,
+ struct lpfc_nodelist *ndlp)
+{
+ struct lpfc_hba *phba = vport->phba;
+ LPFC_MBOXQ_t *mbox;
+ struct lpfc_dmabuf *pcmd;
+ struct ls_rjt stat;
+
+ if ((ndlp->nlp_state != NLP_STE_UNMAPPED_NODE) &&
+ (ndlp->nlp_state != NLP_STE_MAPPED_NODE))
+ /* reject the unsolicited RPS request and done with it */
+ goto reject_out;
+
+ pcmd = (struct lpfc_dmabuf *) cmdiocb->context2;
+
+ mbox = mempool_alloc(phba->mbox_mem_pool, GFP_ATOMIC);
+ if (mbox) {
+ lpfc_read_lnk_stat(phba, mbox);
+ mbox->context1 =
+ (void *)((unsigned long) cmdiocb->iocb.ulpContext);
+ mbox->context2 = lpfc_nlp_get(ndlp);
+ mbox->vport = vport;
+ mbox->mbox_cmpl = lpfc_els_rsp_rls_acc;
+ if (lpfc_sli_issue_mbox(phba, mbox, MBX_NOWAIT)
+ != MBX_NOT_FINISHED)
+ /* Mbox completion will send ELS Response */
+ return 0;
+ /* Decrement reference count used for the failed mbox
+ * command.
+ */
+ lpfc_nlp_put(ndlp);
+ mempool_free(mbox, phba->mbox_mem_pool);
+ }
+reject_out:
+ /* issue rejection response */
+ stat.un.b.lsRjtRsvd0 = 0;
+ stat.un.b.lsRjtRsnCode = LSRJT_UNABLE_TPC;
+ stat.un.b.lsRjtRsnCodeExp = LSEXP_CANT_GIVE_DATA;
+ stat.un.b.vendorUnique = 0;
+ lpfc_els_rsp_reject(vport, stat.un.lsRjtError, cmdiocb, ndlp, NULL);
+ return 0;
+}
+
+/**
+ * lpfc_els_rcv_rtv - Process an unsolicited rtv iocb
+ * @vport: pointer to a host virtual N_Port data structure.
+ * @cmdiocb: pointer to lpfc command iocb data structure.
+ * @ndlp: pointer to a node-list data structure.
+ *
+ * This routine processes Read Timout Value (RTV) IOCB received as an
+ * ELS unsolicited event. It first checks the remote port state. If the
+ * remote port is not in NLP_STE_UNMAPPED_NODE state or NLP_STE_MAPPED_NODE
+ * state, it invokes the lpfc_els_rsl_reject() routine to send the reject
+ * response. Otherwise, it sends the Accept(ACC) response to a Read Timeout
+ * Value (RTV) unsolicited IOCB event.
+ *
+ * Note that, in lpfc_prep_els_iocb() routine, the reference count of ndlp
+ * will be incremented by 1 for holding the ndlp and the reference to ndlp
+ * will be stored into the context1 field of the IOCB for the completion
+ * callback function to the RPS Accept Response ELS IOCB command.
+ *
+ * Return codes
+ * 0 - Successfully processed rtv iocb (currently always return 0)
+ **/
+static int
+lpfc_els_rcv_rtv(struct lpfc_vport *vport, struct lpfc_iocbq *cmdiocb,
+ struct lpfc_nodelist *ndlp)
+{
+ struct lpfc_hba *phba = vport->phba;
+ struct ls_rjt stat;
+ struct RTV_RSP *rtv_rsp;
+ uint8_t *pcmd;
+ struct lpfc_iocbq *elsiocb;
+ uint32_t cmdsize;
+
+
+ if ((ndlp->nlp_state != NLP_STE_UNMAPPED_NODE) &&
+ (ndlp->nlp_state != NLP_STE_MAPPED_NODE))
+ /* reject the unsolicited RPS request and done with it */
+ goto reject_out;
+
+ cmdsize = sizeof(struct RTV_RSP) + sizeof(uint32_t);
+ elsiocb = lpfc_prep_els_iocb(phba->pport, 0, cmdsize,
+ lpfc_max_els_tries, ndlp,
+ ndlp->nlp_DID, ELS_CMD_ACC);
+
+ if (!elsiocb)
+ return 1;
+
+ pcmd = (uint8_t *) (((struct lpfc_dmabuf *) elsiocb->context2)->virt);
+ *((uint32_t *) (pcmd)) = ELS_CMD_ACC;
+ pcmd += sizeof(uint32_t); /* Skip past command */
+
+ /* use the command's xri in the response */
+ elsiocb->iocb.ulpContext = cmdiocb->iocb.ulpContext;
+
+ rtv_rsp = (struct RTV_RSP *)pcmd;
+
+ /* populate RTV payload */
+ rtv_rsp->ratov = cpu_to_be32(phba->fc_ratov * 1000); /* report msecs */
+ rtv_rsp->edtov = cpu_to_be32(phba->fc_edtov);
+ bf_set(qtov_edtovres, rtv_rsp, phba->fc_edtovResol ? 1 : 0);
+ bf_set(qtov_rttov, rtv_rsp, 0); /* Field is for FC ONLY */
+ rtv_rsp->qtov = cpu_to_be32(rtv_rsp->qtov);
+
+ /* Xmit ELS RLS ACC response tag <ulpIoTag> */
+ lpfc_printf_vlog(ndlp->vport, KERN_INFO, LOG_ELS,
+ "2875 Xmit ELS RTV ACC response tag x%x xri x%x, "
+ "did x%x, nlp_flag x%x, nlp_state x%x, rpi x%x, "
+ "Data: x%x x%x x%x\n",
+ elsiocb->iotag, elsiocb->iocb.ulpContext,
+ ndlp->nlp_DID, ndlp->nlp_flag, ndlp->nlp_state,
+ ndlp->nlp_rpi,
+ rtv_rsp->ratov, rtv_rsp->edtov, rtv_rsp->qtov);
+ elsiocb->iocb_cmpl = lpfc_cmpl_els_rsp;
+ phba->fc_stat.elsXmitACC++;
+ if (lpfc_sli_issue_iocb(phba, LPFC_ELS_RING, elsiocb, 0) == IOCB_ERROR)
+ lpfc_els_free_iocb(phba, elsiocb);
+ return 0;
+
+reject_out:
+ /* issue rejection response */
+ stat.un.b.lsRjtRsvd0 = 0;
+ stat.un.b.lsRjtRsnCode = LSRJT_UNABLE_TPC;
+ stat.un.b.lsRjtRsnCodeExp = LSEXP_CANT_GIVE_DATA;
+ stat.un.b.vendorUnique = 0;
+ lpfc_els_rsp_reject(vport, stat.un.lsRjtError, cmdiocb, ndlp, NULL);
+ return 0;
+}
+
+/* lpfc_els_rcv_rps - Process an unsolicited rps iocb
* @vport: pointer to a host virtual N_Port data structure.
* @cmdiocb: pointer to lpfc command iocb data structure.
* @ndlp: pointer to a node-list data structure.
pcmd = (struct lpfc_dmabuf *) cmdiocb->context2;
lp = (uint32_t *) pcmd->virt;
rpl = (RPL *) (lp + 1);
-
maxsize = be32_to_cpu(rpl->maxsize);
/* We support only one port */
if (newnode)
lpfc_nlp_put(ndlp);
break;
+ case ELS_CMD_RLS:
+ lpfc_debugfs_disc_trc(vport, LPFC_DISC_TRC_ELS_UNSOL,
+ "RCV RLS: did:x%x/ste:x%x flg:x%x",
+ did, vport->port_state, ndlp->nlp_flag);
+
+ phba->fc_stat.elsRcvRLS++;
+ lpfc_els_rcv_rls(vport, elsiocb, ndlp);
+ if (newnode)
+ lpfc_nlp_put(ndlp);
+ break;
case ELS_CMD_RPS:
lpfc_debugfs_disc_trc(vport, LPFC_DISC_TRC_ELS_UNSOL,
"RCV RPS: did:x%x/ste:x%x flg:x%x",
if (newnode)
lpfc_nlp_put(ndlp);
break;
+ case ELS_CMD_RTV:
+ lpfc_debugfs_disc_trc(vport, LPFC_DISC_TRC_ELS_UNSOL,
+ "RCV RTV: did:x%x/ste:x%x flg:x%x",
+ did, vport->port_state, ndlp->nlp_flag);
+ phba->fc_stat.elsRcvRTV++;
+ lpfc_els_rcv_rtv(vport, elsiocb, ndlp);
+ if (newnode)
+ lpfc_nlp_put(ndlp);
+ break;
case ELS_CMD_RRQ:
lpfc_debugfs_disc_trc(vport, LPFC_DISC_TRC_ELS_UNSOL,
"RCV RRQ: did:x%x/ste:x%x flg:x%x",
if (newnode)
lpfc_nlp_put(ndlp);
break;
+ case ELS_CMD_ECHO:
+ lpfc_debugfs_disc_trc(vport, LPFC_DISC_TRC_ELS_UNSOL,
+ "RCV ECHO: did:x%x/ste:x%x flg:x%x",
+ did, vport->port_state, ndlp->nlp_flag);
+
+ phba->fc_stat.elsRcvECHO++;
+ lpfc_els_rcv_echo(vport, elsiocb, ndlp);
+ if (newnode)
+ lpfc_nlp_put(ndlp);
+ break;
default:
lpfc_debugfs_disc_trc(vport, LPFC_DISC_TRC_ELS_UNSOL,
"RCV ELS cmd: cmd:x%x did:x%x/ste:x%x",
default:
/* Try to recover from this error */
+ if (phba->sli_rev == LPFC_SLI_REV4)
+ lpfc_sli4_unreg_all_rpis(vport);
lpfc_mbx_unreg_vpi(vport);
spin_lock_irq(shost->host_lock);
vport->fc_flag |= FC_VPORT_NEEDS_REG_VPI;
lpfc_unreg_rpi(vport, np);
}
lpfc_cleanup_pending_mbox(vport);
+
+ if (phba->sli_rev == LPFC_SLI_REV4)
+ lpfc_sli4_unreg_all_rpis(vport);
+
lpfc_mbx_unreg_vpi(vport);
spin_lock_irq(shost->host_lock);
vport->fc_flag |= FC_VPORT_NEEDS_REG_VPI;
* to update the MAC address.
*/
lpfc_register_new_vport(phba, vport, ndlp);
- return ;
+ goto out;
}
if (vport->fc_flag & FC_VPORT_NEEDS_INIT_VPI)
*******************************************************************/
#include <linux/blkdev.h>
+#include <linux/delay.h>
#include <linux/slab.h>
#include <linux/pci.h>
#include <linux/kthread.h>
static void lpfc_disc_timeout_handler(struct lpfc_vport *);
static void lpfc_disc_flush_list(struct lpfc_vport *vport);
static void lpfc_unregister_fcfi_cmpl(struct lpfc_hba *, LPFC_MBOXQ_t *);
+static int lpfc_fcf_inuse(struct lpfc_hba *);
void
lpfc_terminate_rport_io(struct fc_rport *rport)
return;
}
-/*
- * This function is called from the worker thread when dev_loss_tmo
- * expire.
- */
-static void
+/**
+ * lpfc_dev_loss_tmo_handler - Remote node devloss timeout handler
+ * @ndlp: Pointer to remote node object.
+ *
+ * This function is called from the worker thread when devloss timeout timer
+ * expires. For SLI4 host, this routine shall return 1 when at lease one
+ * remote node, including this @ndlp, is still in use of FCF; otherwise, this
+ * routine shall return 0 when there is no remote node is still in use of FCF
+ * when devloss timeout happened to this @ndlp.
+ **/
+static int
lpfc_dev_loss_tmo_handler(struct lpfc_nodelist *ndlp)
{
struct lpfc_rport_data *rdata;
int put_node;
int put_rport;
int warn_on = 0;
+ int fcf_inuse = 0;
rport = ndlp->rport;
if (!rport)
- return;
+ return fcf_inuse;
rdata = rport->dd_data;
name = (uint8_t *) &ndlp->nlp_portname;
vport = ndlp->vport;
phba = vport->phba;
+ if (phba->sli_rev == LPFC_SLI_REV4)
+ fcf_inuse = lpfc_fcf_inuse(phba);
+
lpfc_debugfs_disc_trc(vport, LPFC_DISC_TRC_RPORT,
"rport devlosstmo:did:x%x type:x%x id:x%x",
ndlp->nlp_DID, ndlp->nlp_type, rport->scsi_target_id);
lpfc_nlp_put(ndlp);
if (put_rport)
put_device(&rport->dev);
- return;
+ return fcf_inuse;
}
if (ndlp->nlp_state == NLP_STE_MAPPED_NODE) {
*name, *(name+1), *(name+2), *(name+3),
*(name+4), *(name+5), *(name+6), *(name+7),
ndlp->nlp_DID);
- return;
+ return fcf_inuse;
}
if (ndlp->nlp_type & NLP_FABRIC) {
lpfc_nlp_put(ndlp);
if (put_rport)
put_device(&rport->dev);
- return;
+ return fcf_inuse;
}
if (ndlp->nlp_sid != NLP_NO_SID) {
(ndlp->nlp_state != NLP_STE_PRLI_ISSUE))
lpfc_disc_state_machine(vport, ndlp, NULL, NLP_EVT_DEVICE_RM);
+ return fcf_inuse;
+}
+
+/**
+ * lpfc_sli4_post_dev_loss_tmo_handler - SLI4 post devloss timeout handler
+ * @phba: Pointer to hba context object.
+ * @fcf_inuse: SLI4 FCF in-use state reported from devloss timeout handler.
+ * @nlp_did: remote node identifer with devloss timeout.
+ *
+ * This function is called from the worker thread after invoking devloss
+ * timeout handler and releasing the reference count for the ndlp with
+ * which the devloss timeout was handled for SLI4 host. For the devloss
+ * timeout of the last remote node which had been in use of FCF, when this
+ * routine is invoked, it shall be guaranteed that none of the remote are
+ * in-use of FCF. When devloss timeout to the last remote using the FCF,
+ * if the FIP engine is neither in FCF table scan process nor roundrobin
+ * failover process, the in-use FCF shall be unregistered. If the FIP
+ * engine is in FCF discovery process, the devloss timeout state shall
+ * be set for either the FCF table scan process or roundrobin failover
+ * process to unregister the in-use FCF.
+ **/
+static void
+lpfc_sli4_post_dev_loss_tmo_handler(struct lpfc_hba *phba, int fcf_inuse,
+ uint32_t nlp_did)
+{
+ /* If devloss timeout happened to a remote node when FCF had no
+ * longer been in-use, do nothing.
+ */
+ if (!fcf_inuse)
+ return;
+
+ if ((phba->hba_flag & HBA_FIP_SUPPORT) && !lpfc_fcf_inuse(phba)) {
+ spin_lock_irq(&phba->hbalock);
+ if (phba->fcf.fcf_flag & FCF_DISCOVERY) {
+ if (phba->hba_flag & HBA_DEVLOSS_TMO) {
+ spin_unlock_irq(&phba->hbalock);
+ return;
+ }
+ phba->hba_flag |= HBA_DEVLOSS_TMO;
+ lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
+ "2847 Last remote node (x%x) using "
+ "FCF devloss tmo\n", nlp_did);
+ }
+ if (phba->fcf.fcf_flag & FCF_REDISC_PROG) {
+ spin_unlock_irq(&phba->hbalock);
+ lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
+ "2868 Devloss tmo to FCF rediscovery "
+ "in progress\n");
+ return;
+ }
+ if (!(phba->hba_flag & (FCF_TS_INPROG | FCF_RR_INPROG))) {
+ spin_unlock_irq(&phba->hbalock);
+ lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
+ "2869 Devloss tmo to idle FIP engine, "
+ "unreg in-use FCF and rescan.\n");
+ /* Unregister in-use FCF and rescan */
+ lpfc_unregister_fcf_rescan(phba);
+ return;
+ }
+ spin_unlock_irq(&phba->hbalock);
+ if (phba->hba_flag & FCF_TS_INPROG)
+ lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
+ "2870 FCF table scan in progress\n");
+ if (phba->hba_flag & FCF_RR_INPROG)
+ lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
+ "2871 FLOGI roundrobin FCF failover "
+ "in progress\n");
+ }
lpfc_unregister_unused_fcf(phba);
}
struct lpfc_work_evt *evtp = NULL;
struct lpfc_nodelist *ndlp;
int free_evt;
+ int fcf_inuse;
+ uint32_t nlp_did;
spin_lock_irq(&phba->hbalock);
while (!list_empty(&phba->work_list)) {
break;
case LPFC_EVT_DEV_LOSS:
ndlp = (struct lpfc_nodelist *)(evtp->evt_arg1);
- lpfc_dev_loss_tmo_handler(ndlp);
+ fcf_inuse = lpfc_dev_loss_tmo_handler(ndlp);
free_evt = 0;
/* decrement the node reference count held for
* this queued work
*/
+ nlp_did = ndlp->nlp_DID;
lpfc_nlp_put(ndlp);
+ if (phba->sli_rev == LPFC_SLI_REV4)
+ lpfc_sli4_post_dev_loss_tmo_handler(phba,
+ fcf_inuse,
+ nlp_did);
break;
case LPFC_EVT_ONLINE:
if (phba->link_state < LPFC_LINK_DOWN)
: NLP_EVT_DEVICE_RECOVERY);
}
if (phba->sli3_options & LPFC_SLI3_VPORT_TEARDOWN) {
+ if (phba->sli_rev == LPFC_SLI_REV4)
+ lpfc_sli4_unreg_all_rpis(vport);
lpfc_mbx_unreg_vpi(vport);
spin_lock_irq(shost->host_lock);
vport->fc_flag |= FC_VPORT_NEEDS_REG_VPI;
"2017 REG_FCFI mbxStatus error x%x "
"HBA state x%x\n",
mboxq->u.mb.mbxStatus, vport->port_state);
- mempool_free(mboxq, phba->mbox_mem_pool);
- return;
+ goto fail_out;
}
/* Start FCoE discovery by sending a FLOGI. */
spin_lock_irq(&phba->hbalock);
phba->fcf.fcf_flag |= FCF_REGISTERED;
spin_unlock_irq(&phba->hbalock);
+
/* If there is a pending FCoE event, restart FCF table scan. */
- if (lpfc_check_pending_fcoe_event(phba, 1)) {
- mempool_free(mboxq, phba->mbox_mem_pool);
- return;
- }
+ if (lpfc_check_pending_fcoe_event(phba, LPFC_UNREG_FCF))
+ goto fail_out;
+
+ /* Mark successful completion of FCF table scan */
spin_lock_irq(&phba->hbalock);
phba->fcf.fcf_flag |= (FCF_SCAN_DONE | FCF_IN_USE);
- phba->hba_flag &= ~FCF_DISC_INPROGRESS;
- spin_unlock_irq(&phba->hbalock);
- if (vport->port_state != LPFC_FLOGI)
+ phba->hba_flag &= ~FCF_TS_INPROG;
+ if (vport->port_state != LPFC_FLOGI) {
+ phba->hba_flag |= FCF_RR_INPROG;
+ spin_unlock_irq(&phba->hbalock);
lpfc_initial_flogi(vport);
+ goto out;
+ }
+ spin_unlock_irq(&phba->hbalock);
+ goto out;
+fail_out:
+ spin_lock_irq(&phba->hbalock);
+ phba->hba_flag &= ~FCF_RR_INPROG;
+ spin_unlock_irq(&phba->hbalock);
+out:
mempool_free(mboxq, phba->mbox_mem_pool);
- return;
}
/**
int rc;
spin_lock_irq(&phba->hbalock);
-
/* If the FCF is not availabe do nothing. */
if (!(phba->fcf.fcf_flag & FCF_AVAILABLE)) {
- phba->hba_flag &= ~FCF_DISC_INPROGRESS;
+ phba->hba_flag &= ~(FCF_TS_INPROG | FCF_RR_INPROG);
spin_unlock_irq(&phba->hbalock);
return;
}
/* The FCF is already registered, start discovery */
if (phba->fcf.fcf_flag & FCF_REGISTERED) {
phba->fcf.fcf_flag |= (FCF_SCAN_DONE | FCF_IN_USE);
- phba->hba_flag &= ~FCF_DISC_INPROGRESS;
- spin_unlock_irq(&phba->hbalock);
- if (phba->pport->port_state != LPFC_FLOGI)
+ phba->hba_flag &= ~FCF_TS_INPROG;
+ if (phba->pport->port_state != LPFC_FLOGI) {
+ phba->hba_flag |= FCF_RR_INPROG;
+ spin_unlock_irq(&phba->hbalock);
lpfc_initial_flogi(phba->pport);
+ return;
+ }
+ spin_unlock_irq(&phba->hbalock);
return;
}
spin_unlock_irq(&phba->hbalock);
- fcf_mbxq = mempool_alloc(phba->mbox_mem_pool,
- GFP_KERNEL);
+ fcf_mbxq = mempool_alloc(phba->mbox_mem_pool, GFP_KERNEL);
if (!fcf_mbxq) {
spin_lock_irq(&phba->hbalock);
- phba->hba_flag &= ~FCF_DISC_INPROGRESS;
+ phba->hba_flag &= ~(FCF_TS_INPROG | FCF_RR_INPROG);
spin_unlock_irq(&phba->hbalock);
return;
}
rc = lpfc_sli_issue_mbox(phba, fcf_mbxq, MBX_NOWAIT);
if (rc == MBX_NOT_FINISHED) {
spin_lock_irq(&phba->hbalock);
- phba->hba_flag &= ~FCF_DISC_INPROGRESS;
+ phba->hba_flag &= ~(FCF_TS_INPROG | FCF_RR_INPROG);
spin_unlock_irq(&phba->hbalock);
mempool_free(fcf_mbxq, phba->mbox_mem_pool);
}
* FCF discovery, no need to restart FCF discovery.
*/
if ((phba->link_state >= LPFC_LINK_UP) &&
- (phba->fcoe_eventtag == phba->fcoe_eventtag_at_fcf_scan))
+ (phba->fcoe_eventtag == phba->fcoe_eventtag_at_fcf_scan))
return 0;
lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
lpfc_sli4_fcf_scan_read_fcf_rec(phba, LPFC_FCOE_FCF_GET_FIRST);
} else {
/*
- * Do not continue FCF discovery and clear FCF_DISC_INPROGRESS
+ * Do not continue FCF discovery and clear FCF_TS_INPROG
* flag
*/
lpfc_printf_log(phba, KERN_INFO, LOG_FIP | LOG_DISCOVERY,
"2833 Stop FCF discovery process due to link "
"state change (x%x)\n", phba->link_state);
spin_lock_irq(&phba->hbalock);
- phba->hba_flag &= ~FCF_DISC_INPROGRESS;
+ phba->hba_flag &= ~(FCF_TS_INPROG | FCF_RR_INPROG);
phba->fcf.fcf_flag &= ~(FCF_REDISC_FOV | FCF_DISCOVERY);
spin_unlock_irq(&phba->hbalock);
}
return true;
}
+/**
+ * lpfc_sli4_fcf_rr_next_proc - processing next roundrobin fcf
+ * @vport: Pointer to vport object.
+ * @fcf_index: index to next fcf.
+ *
+ * This function processing the roundrobin fcf failover to next fcf index.
+ * When this function is invoked, there will be a current fcf registered
+ * for flogi.
+ * Return: 0 for continue retrying flogi on currently registered fcf;
+ * 1 for stop flogi on currently registered fcf;
+ */
+int lpfc_sli4_fcf_rr_next_proc(struct lpfc_vport *vport, uint16_t fcf_index)
+{
+ struct lpfc_hba *phba = vport->phba;
+ int rc;
+
+ if (fcf_index == LPFC_FCOE_FCF_NEXT_NONE) {
+ spin_lock_irq(&phba->hbalock);
+ if (phba->hba_flag & HBA_DEVLOSS_TMO) {
+ spin_unlock_irq(&phba->hbalock);
+ lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
+ "2872 Devloss tmo with no eligible "
+ "FCF, unregister in-use FCF (x%x) "
+ "and rescan FCF table\n",
+ phba->fcf.current_rec.fcf_indx);
+ lpfc_unregister_fcf_rescan(phba);
+ goto stop_flogi_current_fcf;
+ }
+ /* Mark the end to FLOGI roundrobin failover */
+ phba->hba_flag &= ~FCF_RR_INPROG;
+ /* Allow action to new fcf asynchronous event */
+ phba->fcf.fcf_flag &= ~(FCF_AVAILABLE | FCF_SCAN_DONE);
+ spin_unlock_irq(&phba->hbalock);
+ lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
+ "2865 No FCF available, stop roundrobin FCF "
+ "failover and change port state:x%x/x%x\n",
+ phba->pport->port_state, LPFC_VPORT_UNKNOWN);
+ phba->pport->port_state = LPFC_VPORT_UNKNOWN;
+ goto stop_flogi_current_fcf;
+ } else {
+ lpfc_printf_log(phba, KERN_INFO, LOG_FIP | LOG_ELS,
+ "2794 Try FLOGI roundrobin FCF failover to "
+ "(x%x)\n", fcf_index);
+ rc = lpfc_sli4_fcf_rr_read_fcf_rec(phba, fcf_index);
+ if (rc)
+ lpfc_printf_log(phba, KERN_WARNING, LOG_FIP | LOG_ELS,
+ "2761 FLOGI roundrobin FCF failover "
+ "failed (rc:x%x) to read FCF (x%x)\n",
+ rc, phba->fcf.current_rec.fcf_indx);
+ else
+ goto stop_flogi_current_fcf;
+ }
+ return 0;
+
+stop_flogi_current_fcf:
+ lpfc_can_disctmo(vport);
+ return 1;
+}
+
/**
* lpfc_mbx_cmpl_fcf_scan_read_fcf_rec - fcf scan read_fcf mbox cmpl handler.
* @phba: pointer to lpfc hba data structure.
int rc;
/* If there is pending FCoE event restart FCF table scan */
- if (lpfc_check_pending_fcoe_event(phba, 0)) {
+ if (lpfc_check_pending_fcoe_event(phba, LPFC_SKIP_UNREG_FCF)) {
lpfc_sli4_mbox_cmd_free(phba, mboxq);
return;
}
new_fcf_record = lpfc_sli4_fcf_rec_mbox_parse(phba, mboxq,
&next_fcf_index);
if (!new_fcf_record) {
- lpfc_printf_log(phba, KERN_WARNING, LOG_FIP,
+ lpfc_printf_log(phba, KERN_ERR, LOG_FIP,
"2765 Mailbox command READ_FCF_RECORD "
"failed to retrieve a FCF record.\n");
/* Let next new FCF event trigger fast failover */
spin_lock_irq(&phba->hbalock);
- phba->hba_flag &= ~FCF_DISC_INPROGRESS;
+ phba->hba_flag &= ~FCF_TS_INPROG;
spin_unlock_irq(&phba->hbalock);
lpfc_sli4_mbox_cmd_free(phba, mboxq);
return;
/*
* If the fcf record does not match with connect list entries
* read the next entry; otherwise, this is an eligible FCF
- * record for round robin FCF failover.
+ * record for roundrobin FCF failover.
*/
if (!rc) {
lpfc_printf_log(phba, KERN_WARNING, LOG_FIP,
- "2781 FCF record (x%x) failed FCF "
- "connection list check, fcf_avail:x%x, "
- "fcf_valid:x%x\n",
+ "2781 FCF (x%x) failed connection "
+ "list check: (x%x/x%x)\n",
bf_get(lpfc_fcf_record_fcf_index,
new_fcf_record),
bf_get(lpfc_fcf_record_fcf_avail,
if ((phba->fcf.fcf_flag & FCF_IN_USE) &&
lpfc_sli4_fcf_record_match(phba, &phba->fcf.current_rec,
new_fcf_record, LPFC_FCOE_IGNORE_VID)) {
+ if (bf_get(lpfc_fcf_record_fcf_index, new_fcf_record) !=
+ phba->fcf.current_rec.fcf_indx) {
+ lpfc_printf_log(phba, KERN_ERR, LOG_FIP,
+ "2862 FCF (x%x) matches property "
+ "of in-use FCF (x%x)\n",
+ bf_get(lpfc_fcf_record_fcf_index,
+ new_fcf_record),
+ phba->fcf.current_rec.fcf_indx);
+ goto read_next_fcf;
+ }
/*
* In case the current in-use FCF record becomes
* invalid/unavailable during FCF discovery that
!(phba->fcf.fcf_flag & FCF_REDISC_FOV)) {
lpfc_printf_log(phba, KERN_WARNING, LOG_FIP,
"2835 Invalid in-use FCF "
- "record (x%x) reported, "
- "entering fast FCF failover "
- "mode scanning.\n",
+ "(x%x), enter FCF failover "
+ "table scan.\n",
phba->fcf.current_rec.fcf_indx);
spin_lock_irq(&phba->hbalock);
phba->fcf.fcf_flag |= FCF_REDISC_FOV;
if (phba->fcf.fcf_flag & FCF_IN_USE) {
if (lpfc_sli4_fcf_record_match(phba, &phba->fcf.current_rec,
new_fcf_record, vlan_id)) {
- phba->fcf.fcf_flag |= FCF_AVAILABLE;
- if (phba->fcf.fcf_flag & FCF_REDISC_PEND)
- /* Stop FCF redisc wait timer if pending */
- __lpfc_sli4_stop_fcf_redisc_wait_timer(phba);
- else if (phba->fcf.fcf_flag & FCF_REDISC_FOV)
- /* If in fast failover, mark it's completed */
- phba->fcf.fcf_flag &= ~FCF_REDISC_FOV;
- spin_unlock_irq(&phba->hbalock);
- lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
- "2836 The new FCF record (x%x) "
- "matches the in-use FCF record "
- "(x%x)\n",
- phba->fcf.current_rec.fcf_indx,
+ if (bf_get(lpfc_fcf_record_fcf_index, new_fcf_record) ==
+ phba->fcf.current_rec.fcf_indx) {
+ phba->fcf.fcf_flag |= FCF_AVAILABLE;
+ if (phba->fcf.fcf_flag & FCF_REDISC_PEND)
+ /* Stop FCF redisc wait timer */
+ __lpfc_sli4_stop_fcf_redisc_wait_timer(
+ phba);
+ else if (phba->fcf.fcf_flag & FCF_REDISC_FOV)
+ /* Fast failover, mark completed */
+ phba->fcf.fcf_flag &= ~FCF_REDISC_FOV;
+ spin_unlock_irq(&phba->hbalock);
+ lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
+ "2836 New FCF matches in-use "
+ "FCF (x%x)\n",
+ phba->fcf.current_rec.fcf_indx);
+ goto out;
+ } else
+ lpfc_printf_log(phba, KERN_ERR, LOG_FIP,
+ "2863 New FCF (x%x) matches "
+ "property of in-use FCF (x%x)\n",
bf_get(lpfc_fcf_record_fcf_index,
- new_fcf_record));
- goto out;
+ new_fcf_record),
+ phba->fcf.current_rec.fcf_indx);
}
/*
* Read next FCF record from HBA searching for the matching
*/
if (fcf_rec) {
lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
- "2840 Update current FCF record "
- "with initial FCF record (x%x)\n",
+ "2840 Update initial FCF candidate "
+ "with FCF (x%x)\n",
bf_get(lpfc_fcf_record_fcf_index,
new_fcf_record));
__lpfc_update_fcf_record(phba, fcf_rec, new_fcf_record,
*/
if (!(phba->fcf.failover_rec.flag & RECORD_VALID)) {
lpfc_printf_log(phba, KERN_WARNING, LOG_FIP,
- "2782 No suitable FCF record "
- "found during this round of "
- "post FCF rediscovery scan: "
- "fcf_evt_tag:x%x, fcf_index: "
- "x%x\n",
+ "2782 No suitable FCF found: "
+ "(x%x/x%x)\n",
phba->fcoe_eventtag_at_fcf_scan,
bf_get(lpfc_fcf_record_fcf_index,
new_fcf_record));
+ spin_lock_irq(&phba->hbalock);
+ if (phba->hba_flag & HBA_DEVLOSS_TMO) {
+ phba->hba_flag &= ~FCF_TS_INPROG;
+ spin_unlock_irq(&phba->hbalock);
+ /* Unregister in-use FCF and rescan */
+ lpfc_printf_log(phba, KERN_INFO,
+ LOG_FIP,
+ "2864 On devloss tmo "
+ "unreg in-use FCF and "
+ "rescan FCF table\n");
+ lpfc_unregister_fcf_rescan(phba);
+ return;
+ }
/*
- * Let next new FCF event trigger fast
- * failover
+ * Let next new FCF event trigger fast failover
*/
- spin_lock_irq(&phba->hbalock);
- phba->hba_flag &= ~FCF_DISC_INPROGRESS;
+ phba->hba_flag &= ~FCF_TS_INPROG;
spin_unlock_irq(&phba->hbalock);
return;
}
/* Replace in-use record with the new record */
lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
- "2842 Replace the current in-use "
- "FCF record (x%x) with failover FCF "
- "record (x%x)\n",
+ "2842 Replace in-use FCF (x%x) "
+ "with failover FCF (x%x)\n",
phba->fcf.current_rec.fcf_indx,
phba->fcf.failover_rec.fcf_indx);
memcpy(&phba->fcf.current_rec,
* FCF failover.
*/
spin_lock_irq(&phba->hbalock);
- phba->fcf.fcf_flag &=
- ~(FCF_REDISC_FOV | FCF_REDISC_RRU);
+ phba->fcf.fcf_flag &= ~FCF_REDISC_FOV;
spin_unlock_irq(&phba->hbalock);
- /*
- * Set up the initial registered FCF index for FLOGI
- * round robin FCF failover.
- */
- phba->fcf.fcf_rr_init_indx =
- phba->fcf.failover_rec.fcf_indx;
/* Register to the new FCF record */
lpfc_register_fcf(phba);
} else {
LPFC_FCOE_FCF_GET_FIRST);
return;
}
-
- /*
- * Otherwise, initial scan or post linkdown rescan,
- * register with the best FCF record found so far
- * through the FCF scanning process.
- */
-
- /*
- * Mark the initial FCF discovery completed and
- * the start of the first round of the roundrobin
- * FCF failover.
- */
- spin_lock_irq(&phba->hbalock);
- phba->fcf.fcf_flag &=
- ~(FCF_INIT_DISC | FCF_REDISC_RRU);
- spin_unlock_irq(&phba->hbalock);
- /*
- * Set up the initial registered FCF index for FLOGI
- * round robin FCF failover
- */
- phba->fcf.fcf_rr_init_indx =
- phba->fcf.current_rec.fcf_indx;
/* Register to the new FCF record */
lpfc_register_fcf(phba);
}
}
/**
- * lpfc_mbx_cmpl_fcf_rr_read_fcf_rec - fcf round robin read_fcf mbox cmpl hdler
+ * lpfc_mbx_cmpl_fcf_rr_read_fcf_rec - fcf roundrobin read_fcf mbox cmpl hdler
* @phba: pointer to lpfc hba data structure.
* @mboxq: pointer to mailbox object.
*
- * This is the callback function for FLOGI failure round robin FCF failover
+ * This is the callback function for FLOGI failure roundrobin FCF failover
* read FCF record mailbox command from the eligible FCF record bmask for
* performing the failover. If the FCF read back is not valid/available, it
* fails through to retrying FLOGI to the currently registered FCF again.
{
struct fcf_record *new_fcf_record;
uint32_t boot_flag, addr_mode;
- uint16_t next_fcf_index;
+ uint16_t next_fcf_index, fcf_index;
uint16_t current_fcf_index;
uint16_t vlan_id;
+ int rc;
- /* If link state is not up, stop the round robin failover process */
+ /* If link state is not up, stop the roundrobin failover process */
if (phba->link_state < LPFC_LINK_UP) {
spin_lock_irq(&phba->hbalock);
phba->fcf.fcf_flag &= ~FCF_DISCOVERY;
+ phba->hba_flag &= ~FCF_RR_INPROG;
spin_unlock_irq(&phba->hbalock);
- lpfc_sli4_mbox_cmd_free(phba, mboxq);
- return;
+ goto out;
}
/* Parse the FCF record from the non-embedded mailbox command */
lpfc_printf_log(phba, KERN_WARNING, LOG_FIP,
"2766 Mailbox command READ_FCF_RECORD "
"failed to retrieve a FCF record.\n");
- goto out;
+ goto error_out;
}
/* Get the needed parameters from FCF record */
- lpfc_match_fcf_conn_list(phba, new_fcf_record, &boot_flag,
- &addr_mode, &vlan_id);
+ rc = lpfc_match_fcf_conn_list(phba, new_fcf_record, &boot_flag,
+ &addr_mode, &vlan_id);
/* Log the FCF record information if turned on */
lpfc_sli4_log_fcf_record_info(phba, new_fcf_record, vlan_id,
next_fcf_index);
+ fcf_index = bf_get(lpfc_fcf_record_fcf_index, new_fcf_record);
+ if (!rc) {
+ lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
+ "2848 Remove ineligible FCF (x%x) from "
+ "from roundrobin bmask\n", fcf_index);
+ /* Clear roundrobin bmask bit for ineligible FCF */
+ lpfc_sli4_fcf_rr_index_clear(phba, fcf_index);
+ /* Perform next round of roundrobin FCF failover */
+ fcf_index = lpfc_sli4_fcf_rr_next_index_get(phba);
+ rc = lpfc_sli4_fcf_rr_next_proc(phba->pport, fcf_index);
+ if (rc)
+ goto out;
+ goto error_out;
+ }
+
+ if (fcf_index == phba->fcf.current_rec.fcf_indx) {
+ lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
+ "2760 Perform FLOGI roundrobin FCF failover: "
+ "FCF (x%x) back to FCF (x%x)\n",
+ phba->fcf.current_rec.fcf_indx, fcf_index);
+ /* Wait 500 ms before retrying FLOGI to current FCF */
+ msleep(500);
+ lpfc_initial_flogi(phba->pport);
+ goto out;
+ }
+
/* Upload new FCF record to the failover FCF record */
lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
- "2834 Update the current FCF record (x%x) "
- "with the next FCF record (x%x)\n",
- phba->fcf.failover_rec.fcf_indx,
- bf_get(lpfc_fcf_record_fcf_index, new_fcf_record));
+ "2834 Update current FCF (x%x) with new FCF (x%x)\n",
+ phba->fcf.failover_rec.fcf_indx, fcf_index);
spin_lock_irq(&phba->hbalock);
__lpfc_update_fcf_record(phba, &phba->fcf.failover_rec,
new_fcf_record, addr_mode, vlan_id,
sizeof(struct lpfc_fcf_rec));
lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
- "2783 FLOGI round robin FCF failover from FCF "
- "(x%x) to FCF (x%x).\n",
- current_fcf_index,
- bf_get(lpfc_fcf_record_fcf_index, new_fcf_record));
+ "2783 Perform FLOGI roundrobin FCF failover: FCF "
+ "(x%x) to FCF (x%x)\n", current_fcf_index, fcf_index);
+error_out:
+ lpfc_register_fcf(phba);
out:
lpfc_sli4_mbox_cmd_free(phba, mboxq);
- lpfc_register_fcf(phba);
}
/**
* @mboxq: pointer to mailbox object.
*
* This is the callback function of read FCF record mailbox command for
- * updating the eligible FCF bmask for FLOGI failure round robin FCF
+ * updating the eligible FCF bmask for FLOGI failure roundrobin FCF
* failover when a new FCF event happened. If the FCF read back is
* valid/available and it passes the connection list check, it updates
- * the bmask for the eligible FCF record for round robin failover.
+ * the bmask for the eligible FCF record for roundrobin failover.
*/
void
lpfc_mbx_cmpl_read_fcf_rec(struct lpfc_hba *phba, LPFC_MBOXQ_t *mboxq)
* and get the FCF Table.
*/
spin_lock_irq(&phba->hbalock);
- if (phba->hba_flag & FCF_DISC_INPROGRESS) {
+ if (phba->hba_flag & FCF_TS_INPROG) {
spin_unlock_irq(&phba->hbalock);
return;
}
LPFC_MBOXQ_t *mbox;
int rc;
+ if (phba->sli_rev == LPFC_SLI_REV4) {
+ lpfc_sli4_unreg_all_rpis(vport);
+ return;
+ }
+
mbox = mempool_alloc(phba->mbox_mem_pool, GFP_KERNEL);
if (mbox) {
lpfc_unreg_login(phba, vport->vpi, 0xffff, mbox);
}
spin_lock_irq(&phba->hbalock);
+ /* Cleanup REG_LOGIN completions which are not yet processed */
+ list_for_each_entry(mb, &phba->sli.mboxq_cmpl, list) {
+ if ((mb->u.mb.mbxCommand != MBX_REG_LOGIN64) ||
+ (ndlp != (struct lpfc_nodelist *) mb->context2))
+ continue;
+
+ mb->context2 = NULL;
+ mb->mbox_cmpl = lpfc_sli_def_mbox_cmpl;
+ }
+
list_for_each_entry_safe(mb, nextmb, &phba->sli.mboxq, list) {
if ((mb->u.mb.mbxCommand == MBX_REG_LOGIN64) &&
(ndlp == (struct lpfc_nodelist *) mb->context2)) {
if (ndlp)
lpfc_cancel_retry_delay_tmo(vports[i], ndlp);
lpfc_cleanup_pending_mbox(vports[i]);
+ if (phba->sli_rev == LPFC_SLI_REV4)
+ lpfc_sli4_unreg_all_rpis(vports[i]);
lpfc_mbx_unreg_vpi(vports[i]);
shost = lpfc_shost_from_vport(vports[i]);
spin_lock_irq(shost->host_lock);
uint32_t crcCnt;
} RPS_RSP;
+struct RLS { /* Structure is in Big Endian format */
+ uint32_t rls;
+#define rls_rsvd_SHIFT 24
+#define rls_rsvd_MASK 0x000000ff
+#define rls_rsvd_WORD rls
+#define rls_did_SHIFT 0
+#define rls_did_MASK 0x00ffffff
+#define rls_did_WORD rls
+};
+
+struct RLS_RSP { /* Structure is in Big Endian format */
+ uint32_t linkFailureCnt;
+ uint32_t lossSyncCnt;
+ uint32_t lossSignalCnt;
+ uint32_t primSeqErrCnt;
+ uint32_t invalidXmitWord;
+ uint32_t crcCnt;
+};
+
+struct RTV_RSP { /* Structure is in Big Endian format */
+ uint32_t ratov;
+ uint32_t edtov;
+ uint32_t qtov;
+#define qtov_rsvd0_SHIFT 28
+#define qtov_rsvd0_MASK 0x0000000f
+#define qtov_rsvd0_WORD qtov /* reserved */
+#define qtov_edtovres_SHIFT 27
+#define qtov_edtovres_MASK 0x00000001
+#define qtov_edtovres_WORD qtov /* E_D_TOV Resolution */
+#define qtov__rsvd1_SHIFT 19
+#define qtov_rsvd1_MASK 0x0000003f
+#define qtov_rsvd1_WORD qtov /* reserved */
+#define qtov_rttov_SHIFT 18
+#define qtov_rttov_MASK 0x00000001
+#define qtov_rttov_WORD qtov /* R_T_TOV value */
+#define qtov_rsvd2_SHIFT 0
+#define qtov_rsvd2_MASK 0x0003ffff
+#define qtov_rsvd2_WORD qtov /* reserved */
+};
+
+
typedef struct _RPL { /* Structure is in Big Endian format */
uint32_t maxsize;
uint32_t index;
#define FCOE_SOFn3 0x36
};
-struct lpfc_wqe_generic{
- struct ulp_bde64 bde;
- uint32_t word3;
- uint32_t word4;
- uint32_t word5;
- uint32_t word6;
-#define lpfc_wqe_gen_context_SHIFT 16
-#define lpfc_wqe_gen_context_MASK 0x0000FFFF
-#define lpfc_wqe_gen_context_WORD word6
-#define lpfc_wqe_gen_xri_SHIFT 0
-#define lpfc_wqe_gen_xri_MASK 0x0000FFFF
-#define lpfc_wqe_gen_xri_WORD word6
- uint32_t word7;
-#define lpfc_wqe_gen_lnk_SHIFT 23
-#define lpfc_wqe_gen_lnk_MASK 0x00000001
-#define lpfc_wqe_gen_lnk_WORD word7
-#define lpfc_wqe_gen_erp_SHIFT 22
-#define lpfc_wqe_gen_erp_MASK 0x00000001
-#define lpfc_wqe_gen_erp_WORD word7
-#define lpfc_wqe_gen_pu_SHIFT 20
-#define lpfc_wqe_gen_pu_MASK 0x00000003
-#define lpfc_wqe_gen_pu_WORD word7
-#define lpfc_wqe_gen_class_SHIFT 16
-#define lpfc_wqe_gen_class_MASK 0x00000007
-#define lpfc_wqe_gen_class_WORD word7
-#define lpfc_wqe_gen_command_SHIFT 8
-#define lpfc_wqe_gen_command_MASK 0x000000FF
-#define lpfc_wqe_gen_command_WORD word7
-#define lpfc_wqe_gen_status_SHIFT 4
-#define lpfc_wqe_gen_status_MASK 0x0000000F
-#define lpfc_wqe_gen_status_WORD word7
-#define lpfc_wqe_gen_ct_SHIFT 2
-#define lpfc_wqe_gen_ct_MASK 0x00000003
-#define lpfc_wqe_gen_ct_WORD word7
- uint32_t abort_tag;
- uint32_t word9;
-#define lpfc_wqe_gen_request_tag_SHIFT 0
-#define lpfc_wqe_gen_request_tag_MASK 0x0000FFFF
-#define lpfc_wqe_gen_request_tag_WORD word9
- uint32_t word10;
-#define lpfc_wqe_gen_ccp_SHIFT 24
-#define lpfc_wqe_gen_ccp_MASK 0x000000FF
-#define lpfc_wqe_gen_ccp_WORD word10
-#define lpfc_wqe_gen_ccpe_SHIFT 23
-#define lpfc_wqe_gen_ccpe_MASK 0x00000001
-#define lpfc_wqe_gen_ccpe_WORD word10
-#define lpfc_wqe_gen_pv_SHIFT 19
-#define lpfc_wqe_gen_pv_MASK 0x00000001
-#define lpfc_wqe_gen_pv_WORD word10
-#define lpfc_wqe_gen_pri_SHIFT 16
-#define lpfc_wqe_gen_pri_MASK 0x00000007
-#define lpfc_wqe_gen_pri_WORD word10
- uint32_t word11;
-#define lpfc_wqe_gen_cq_id_SHIFT 16
-#define lpfc_wqe_gen_cq_id_MASK 0x0000FFFF
-#define lpfc_wqe_gen_cq_id_WORD word11
-#define LPFC_WQE_CQ_ID_DEFAULT 0xffff
-#define lpfc_wqe_gen_wqec_SHIFT 7
-#define lpfc_wqe_gen_wqec_MASK 0x00000001
-#define lpfc_wqe_gen_wqec_WORD word11
-#define ELS_ID_FLOGI 3
-#define ELS_ID_FDISC 2
-#define ELS_ID_LOGO 1
-#define ELS_ID_DEFAULT 0
-#define lpfc_wqe_gen_els_id_SHIFT 4
-#define lpfc_wqe_gen_els_id_MASK 0x00000003
-#define lpfc_wqe_gen_els_id_WORD word11
-#define lpfc_wqe_gen_cmd_type_SHIFT 0
-#define lpfc_wqe_gen_cmd_type_MASK 0x0000000F
-#define lpfc_wqe_gen_cmd_type_WORD word11
- uint32_t payload[4];
-};
-
struct lpfc_rqe {
uint32_t address_hi;
uint32_t address_lo;
#define wqe_reqtag_MASK 0x0000FFFF
#define wqe_reqtag_WORD word9
#define wqe_rcvoxid_SHIFT 16
-#define wqe_rcvoxid_MASK 0x0000FFFF
-#define wqe_rcvoxid_WORD word9
+#define wqe_rcvoxid_MASK 0x0000FFFF
+#define wqe_rcvoxid_WORD word9
uint32_t word10;
+#define wqe_ebde_cnt_SHIFT 0
+#define wqe_ebde_cnt_MASK 0x00000007
+#define wqe_ebde_cnt_WORD word10
+#define wqe_lenloc_SHIFT 7
+#define wqe_lenloc_MASK 0x00000003
+#define wqe_lenloc_WORD word10
+#define LPFC_WQE_LENLOC_NONE 0
+#define LPFC_WQE_LENLOC_WORD3 1
+#define LPFC_WQE_LENLOC_WORD12 2
+#define LPFC_WQE_LENLOC_WORD4 3
+#define wqe_qosd_SHIFT 9
+#define wqe_qosd_MASK 0x00000001
+#define wqe_qosd_WORD word10
+#define wqe_xbl_SHIFT 11
+#define wqe_xbl_MASK 0x00000001
+#define wqe_xbl_WORD word10
+#define wqe_iod_SHIFT 13
+#define wqe_iod_MASK 0x00000001
+#define wqe_iod_WORD word10
+#define LPFC_WQE_IOD_WRITE 0
+#define LPFC_WQE_IOD_READ 1
+#define wqe_dbde_SHIFT 14
+#define wqe_dbde_MASK 0x00000001
+#define wqe_dbde_WORD word10
+#define wqe_wqes_SHIFT 15
+#define wqe_wqes_MASK 0x00000001
+#define wqe_wqes_WORD word10
#define wqe_pri_SHIFT 16
#define wqe_pri_MASK 0x00000007
#define wqe_pri_WORD word10
#define wqe_ccpe_MASK 0x00000001
#define wqe_ccpe_WORD word10
#define wqe_ccp_SHIFT 24
-#define wqe_ccp_MASK 0x000000ff
-#define wqe_ccp_WORD word10
+#define wqe_ccp_MASK 0x000000ff
+#define wqe_ccp_WORD word10
uint32_t word11;
-#define wqe_cmd_type_SHIFT 0
-#define wqe_cmd_type_MASK 0x0000000f
-#define wqe_cmd_type_WORD word11
-#define wqe_wqec_SHIFT 7
-#define wqe_wqec_MASK 0x00000001
-#define wqe_wqec_WORD word11
-#define wqe_cqid_SHIFT 16
-#define wqe_cqid_MASK 0x0000ffff
-#define wqe_cqid_WORD word11
+#define wqe_cmd_type_SHIFT 0
+#define wqe_cmd_type_MASK 0x0000000f
+#define wqe_cmd_type_WORD word11
+#define wqe_els_id_SHIFT 4
+#define wqe_els_id_MASK 0x00000003
+#define wqe_els_id_WORD word11
+#define LPFC_ELS_ID_FLOGI 3
+#define LPFC_ELS_ID_FDISC 2
+#define LPFC_ELS_ID_LOGO 1
+#define LPFC_ELS_ID_DEFAULT 0
+#define wqe_wqec_SHIFT 7
+#define wqe_wqec_MASK 0x00000001
+#define wqe_wqec_WORD word11
+#define wqe_cqid_SHIFT 16
+#define wqe_cqid_MASK 0x0000ffff
+#define wqe_cqid_WORD word11
+#define LPFC_WQE_CQ_ID_DEFAULT 0xffff
};
struct wqe_did {
#define wqe_xmit_bls_xo_WORD word5
};
+struct lpfc_wqe_generic{
+ struct ulp_bde64 bde;
+ uint32_t word3;
+ uint32_t word4;
+ uint32_t word5;
+ struct wqe_common wqe_com;
+ uint32_t payload[4];
+};
+
struct els_request64_wqe {
struct ulp_bde64 bde;
uint32_t payload_len;
struct xmit_els_rsp64_wqe {
struct ulp_bde64 bde;
- uint32_t rsvd3;
+ uint32_t response_payload_len;
uint32_t rsvd4;
- struct wqe_did wqe_dest;
+ struct wqe_did wqe_dest;
struct wqe_common wqe_com; /* words 6-11 */
uint32_t rsvd_12_15[4];
};
struct xmit_seq64_wqe {
struct ulp_bde64 bde;
- uint32_t paylaod_offset;
+ uint32_t rsvd3;
uint32_t relative_offset;
struct wqe_rctl_dfctl wge_ctl;
struct wqe_common wqe_com; /* words 6-11 */
};
struct xmit_bcast64_wqe {
struct ulp_bde64 bde;
- uint32_t paylaod_len;
+ uint32_t seq_payload_len;
uint32_t rsvd4;
struct wqe_rctl_dfctl wge_ctl; /* word 5 */
struct wqe_common wqe_com; /* words 6-11 */
struct gen_req64_wqe {
struct ulp_bde64 bde;
- uint32_t command_len;
- uint32_t payload_len;
+ uint32_t request_payload_len;
+ uint32_t relative_offset;
struct wqe_rctl_dfctl wge_ctl; /* word 5 */
struct wqe_common wqe_com; /* words 6-11 */
uint32_t rsvd_12_15[4];
struct fcp_iwrite64_wqe {
struct ulp_bde64 bde;
- uint32_t payload_len;
+ uint32_t payload_offset_len;
uint32_t total_xfer_len;
uint32_t initial_xfer_len;
struct wqe_common wqe_com; /* words 6-11 */
struct fcp_iread64_wqe {
struct ulp_bde64 bde;
- uint32_t payload_len; /* word 3 */
+ uint32_t payload_offset_len; /* word 3 */
uint32_t total_xfer_len; /* word 4 */
uint32_t rsrvd5; /* word 5 */
struct wqe_common wqe_com; /* words 6-11 */
};
struct fcp_icmnd64_wqe {
- struct ulp_bde64 bde; /* words 0-2 */
- uint32_t rsrvd[3]; /* words 3-5 */
+ struct ulp_bde64 bde; /* words 0-2 */
+ uint32_t rsrvd3; /* word 3 */
+ uint32_t rsrvd4; /* word 4 */
+ uint32_t rsrvd5; /* word 5 */
struct wqe_common wqe_com; /* words 6-11 */
- uint32_t rsvd_12_15[4]; /* word 12-15 */
+ uint32_t rsvd_12_15[4]; /* word 12-15 */
};
return 0;
}
+
/**
* lpfc_hba_down_post_s4 - Perform lpfc uninitialization after HBA reset
* @phba: pointer to lpfc HBA data structure.
void
__lpfc_sli4_stop_fcf_redisc_wait_timer(struct lpfc_hba *phba)
{
- /* Clear pending FCF rediscovery wait and failover in progress flags */
- phba->fcf.fcf_flag &= ~(FCF_REDISC_PEND |
- FCF_DEAD_DISC |
- FCF_ACVL_DISC);
+ /* Clear pending FCF rediscovery wait flag */
+ phba->fcf.fcf_flag &= ~FCF_REDISC_PEND;
+
/* Now, try to stop the timer */
del_timer(&phba->fcf.redisc_wait);
}
return;
}
__lpfc_sli4_stop_fcf_redisc_wait_timer(phba);
+ /* Clear failover in progress flags */
+ phba->fcf.fcf_flag &= ~(FCF_DEAD_DISC | FCF_ACVL_DISC);
spin_unlock_irq(&phba->hbalock);
}
phba->fcf.fcf_flag |= FCF_REDISC_EVT;
spin_unlock_irq(&phba->hbalock);
lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
- "2776 FCF rediscover wait timer expired, post "
- "a worker thread event for FCF table scan\n");
+ "2776 FCF rediscover quiescent timer expired\n");
/* wake up worker thread */
lpfc_worker_wake_up(phba);
}
if (event_type == LPFC_FCOE_EVENT_TYPE_NEW_FCF)
lpfc_printf_log(phba, KERN_ERR, LOG_FIP |
LOG_DISCOVERY,
- "2546 New FCF found event: "
- "evt_tag:x%x, fcf_index:x%x\n",
+ "2546 New FCF event, evt_tag:x%x, "
+ "index:x%x\n",
acqe_fcoe->event_tag,
acqe_fcoe->index);
else
lpfc_printf_log(phba, KERN_WARNING, LOG_FIP |
LOG_DISCOVERY,
- "2788 FCF parameter modified event: "
- "evt_tag:x%x, fcf_index:x%x\n",
+ "2788 FCF param modified event, "
+ "evt_tag:x%x, index:x%x\n",
acqe_fcoe->event_tag,
acqe_fcoe->index);
if (phba->fcf.fcf_flag & FCF_DISCOVERY) {
/*
* During period of FCF discovery, read the FCF
* table record indexed by the event to update
- * FCF round robin failover eligible FCF bmask.
+ * FCF roundrobin failover eligible FCF bmask.
*/
lpfc_printf_log(phba, KERN_INFO, LOG_FIP |
LOG_DISCOVERY,
- "2779 Read new FCF record with "
- "fcf_index:x%x for updating FCF "
- "round robin failover bmask\n",
+ "2779 Read FCF (x%x) for updating "
+ "roundrobin FCF failover bmask\n",
acqe_fcoe->index);
rc = lpfc_sli4_read_fcf_rec(phba, acqe_fcoe->index);
}
/* If the FCF discovery is in progress, do nothing. */
spin_lock_irq(&phba->hbalock);
- if (phba->hba_flag & FCF_DISC_INPROGRESS) {
+ if (phba->hba_flag & FCF_TS_INPROG) {
spin_unlock_irq(&phba->hbalock);
break;
}
/* Otherwise, scan the entire FCF table and re-discover SAN */
lpfc_printf_log(phba, KERN_INFO, LOG_FIP | LOG_DISCOVERY,
- "2770 Start FCF table scan due to new FCF "
- "event: evt_tag:x%x, fcf_index:x%x\n",
+ "2770 Start FCF table scan per async FCF "
+ "event, evt_tag:x%x, index:x%x\n",
acqe_fcoe->event_tag, acqe_fcoe->index);
rc = lpfc_sli4_fcf_scan_read_fcf_rec(phba,
LPFC_FCOE_FCF_GET_FIRST);
if (rc)
lpfc_printf_log(phba, KERN_ERR, LOG_FIP | LOG_DISCOVERY,
"2547 Issue FCF scan read FCF mailbox "
- "command failed 0x%x\n", rc);
+ "command failed (x%x)\n", rc);
break;
case LPFC_FCOE_EVENT_TYPE_FCF_TABLE_FULL:
case LPFC_FCOE_EVENT_TYPE_FCF_DEAD:
lpfc_printf_log(phba, KERN_ERR, LOG_FIP | LOG_DISCOVERY,
- "2549 FCF disconnected from network index 0x%x"
- " tag 0x%x\n", acqe_fcoe->index,
- acqe_fcoe->event_tag);
+ "2549 FCF (x%x) disconnected from network, "
+ "tag:x%x\n", acqe_fcoe->index, acqe_fcoe->event_tag);
/*
* If we are in the middle of FCF failover process, clear
* the corresponding FCF bit in the roundrobin bitmap.
spin_unlock_irq(&phba->hbalock);
lpfc_printf_log(phba, KERN_INFO, LOG_FIP |
LOG_DISCOVERY,
- "2773 Start FCF fast failover due "
- "to CVL event: evt_tag:x%x\n",
- acqe_fcoe->event_tag);
+ "2773 Start FCF failover per CVL, "
+ "evt_tag:x%x\n", acqe_fcoe->event_tag);
rc = lpfc_sli4_redisc_fcf_table(phba);
if (rc) {
lpfc_printf_log(phba, KERN_ERR, LOG_FIP |
/* Scan FCF table from the first entry to re-discover SAN */
lpfc_printf_log(phba, KERN_INFO, LOG_FIP | LOG_DISCOVERY,
- "2777 Start FCF table scan after FCF "
- "rediscovery quiescent period over\n");
+ "2777 Start post-quiescent FCF table scan\n");
rc = lpfc_sli4_fcf_scan_read_fcf_rec(phba, LPFC_FCOE_FCF_GET_FIRST);
if (rc)
lpfc_printf_log(phba, KERN_ERR, LOG_FIP | LOG_DISCOVERY,
goto out_free_active_sgl;
}
- /* Allocate eligible FCF bmask memory for FCF round robin failover */
+ /* Allocate eligible FCF bmask memory for FCF roundrobin failover */
longs = (LPFC_SLI4_FCF_TBL_INDX_MAX + BITS_PER_LONG - 1)/BITS_PER_LONG;
phba->fcf.fcf_rr_bmask = kzalloc(longs * sizeof(unsigned long),
GFP_KERNEL);
return;
}
+/**
+ * lpfc_sli4_xri_exchange_busy_wait - Wait for device XRI exchange busy
+ * @phba: Pointer to HBA context object.
+ *
+ * This function is called in the SLI4 code path to wait for completion
+ * of device's XRIs exchange busy. It will check the XRI exchange busy
+ * on outstanding FCP and ELS I/Os every 10ms for up to 10 seconds; after
+ * that, it will check the XRI exchange busy on outstanding FCP and ELS
+ * I/Os every 30 seconds, log error message, and wait forever. Only when
+ * all XRI exchange busy complete, the driver unload shall proceed with
+ * invoking the function reset ioctl mailbox command to the CNA and the
+ * the rest of the driver unload resource release.
+ **/
+static void
+lpfc_sli4_xri_exchange_busy_wait(struct lpfc_hba *phba)
+{
+ int wait_time = 0;
+ int fcp_xri_cmpl = list_empty(&phba->sli4_hba.lpfc_abts_scsi_buf_list);
+ int els_xri_cmpl = list_empty(&phba->sli4_hba.lpfc_abts_els_sgl_list);
+
+ while (!fcp_xri_cmpl || !els_xri_cmpl) {
+ if (wait_time > LPFC_XRI_EXCH_BUSY_WAIT_TMO) {
+ if (!fcp_xri_cmpl)
+ lpfc_printf_log(phba, KERN_ERR, LOG_INIT,
+ "2877 FCP XRI exchange busy "
+ "wait time: %d seconds.\n",
+ wait_time/1000);
+ if (!els_xri_cmpl)
+ lpfc_printf_log(phba, KERN_ERR, LOG_INIT,
+ "2878 ELS XRI exchange busy "
+ "wait time: %d seconds.\n",
+ wait_time/1000);
+ msleep(LPFC_XRI_EXCH_BUSY_WAIT_T2);
+ wait_time += LPFC_XRI_EXCH_BUSY_WAIT_T2;
+ } else {
+ msleep(LPFC_XRI_EXCH_BUSY_WAIT_T1);
+ wait_time += LPFC_XRI_EXCH_BUSY_WAIT_T1;
+ }
+ fcp_xri_cmpl =
+ list_empty(&phba->sli4_hba.lpfc_abts_scsi_buf_list);
+ els_xri_cmpl =
+ list_empty(&phba->sli4_hba.lpfc_abts_els_sgl_list);
+ }
+}
+
/**
* lpfc_sli4_hba_unset - Unset the fcoe hba
* @phba: Pointer to HBA context object.
spin_unlock_irq(&phba->hbalock);
}
+ /* Abort all iocbs associated with the hba */
+ lpfc_sli_hba_iocb_abort(phba);
+
+ /* Wait for completion of device XRI exchange busy */
+ lpfc_sli4_xri_exchange_busy_wait(phba);
+
/* Disable PCI subsystem interrupt */
lpfc_sli4_disable_intr(phba);
return;
}
+/**
+ * lpfc_sli4_unreg_all_rpis - unregister all RPIs for a vport on SLI4 HBA.
+ * @vport: pointer to a vport object.
+ *
+ * This routine sends mailbox command to unregister all active RPIs for
+ * a vport.
+ **/
+void
+lpfc_sli4_unreg_all_rpis(struct lpfc_vport *vport)
+{
+ struct lpfc_hba *phba = vport->phba;
+ LPFC_MBOXQ_t *mbox;
+ int rc;
+
+ mbox = mempool_alloc(phba->mbox_mem_pool, GFP_KERNEL);
+ if (mbox) {
+ lpfc_unreg_login(phba, vport->vpi,
+ vport->vpi + phba->vpi_base, mbox);
+ mbox->u.mb.un.varUnregLogin.rsvd1 = 0x4000 ;
+ mbox->vport = vport;
+ mbox->mbox_cmpl = lpfc_sli_def_mbox_cmpl;
+ mbox->context1 = NULL;
+ rc = lpfc_sli_issue_mbox(phba, mbox, MBX_NOWAIT);
+ if (rc == MBX_NOT_FINISHED)
+ mempool_free(mbox, phba->mbox_mem_pool);
+ }
+}
+
/**
* lpfc_reg_vpi - Prepare a mailbox command for registering vport identifier
* @phba: pointer to lpfc hba data structure.
spin_lock_irqsave(shost->host_lock, flags);
if (!vport->stat_data_enabled ||
vport->stat_data_blocked ||
+ !pnode ||
!pnode->lat_data ||
(phba->bucket_type == LPFC_NO_BUCKET)) {
spin_unlock_irqrestore(shost->host_lock, flags);
struct lpfc_nodelist *pnode = lpfc_cmd->rdata->pnode;
unsigned long flags;
+ if (!pnode || !NLP_CHK_NODE_ACT(pnode))
+ return;
+
/* If there is queuefull or busy condition send a scsi event */
if ((cmnd->result == SAM_STAT_TASK_SET_FULL) ||
(cmnd->result == SAM_STAT_BUSY)) {
struct lpfc_scsi_buf *lpfc_cmd;
struct lpfc_iocbq *iocbq;
struct lpfc_iocbq *iocbqrsp;
+ struct lpfc_nodelist *pnode = rdata->pnode;
int ret;
int status;
- if (!rdata->pnode || !NLP_CHK_NODE_ACT(rdata->pnode))
+ if (!pnode || !NLP_CHK_NODE_ACT(pnode))
return FAILED;
lpfc_cmd = lpfc_get_scsi_buf(phba);
"0702 Issue %s to TGT %d LUN %d "
"rpi x%x nlp_flag x%x\n",
lpfc_taskmgmt_name(task_mgmt_cmd), tgt_id, lun_id,
- rdata->pnode->nlp_rpi, rdata->pnode->nlp_flag);
+ pnode->nlp_rpi, pnode->nlp_flag);
status = lpfc_sli_issue_iocb_wait(phba, LPFC_FCP_RING,
iocbq, iocbqrsp, lpfc_cmd->timeout);
return -ENOMEM;
/* set consumption flag every once in a while */
if (!((q->host_index + 1) % LPFC_RELEASE_NOTIFICATION_INTERVAL))
- bf_set(lpfc_wqe_gen_wqec, &wqe->generic, 1);
+ bf_set(wqe_wqec, &wqe->generic.wqe_com, 1);
lpfc_sli_pcimem_bcopy(wqe, temp_wqe, q->entry_size);
struct lpfc_vport *vport = pmb->vport;
struct lpfc_dmabuf *mp;
struct lpfc_nodelist *ndlp;
+ struct Scsi_Host *shost;
uint16_t rpi, vpi;
int rc;
}
if ((pmb->u.mb.mbxCommand == MBX_UNREG_LOGIN) &&
- (phba->sli_rev == LPFC_SLI_REV4))
+ (phba->sli_rev == LPFC_SLI_REV4) &&
+ (pmb->u.mb.un.varUnregLogin.rsvd1 == 0x0))
lpfc_sli4_free_rpi(phba, pmb->u.mb.un.varUnregLogin.rpi);
/*
return;
}
- /* Unreg VPI, if the REG_VPI succeed after VLink failure */
if ((pmb->u.mb.mbxCommand == MBX_REG_VPI) &&
!(phba->pport->load_flag & FC_UNLOADING) &&
!pmb->u.mb.mbxStatus) {
- lpfc_unreg_vpi(phba, pmb->u.mb.un.varRegVpi.vpi, pmb);
- pmb->vport = vport;
- pmb->mbox_cmpl = lpfc_sli_def_mbox_cmpl;
- rc = lpfc_sli_issue_mbox(phba, pmb, MBX_NOWAIT);
- if (rc != MBX_NOT_FINISHED)
- return;
+ shost = lpfc_shost_from_vport(vport);
+ spin_lock_irq(shost->host_lock);
+ vport->vpi_state |= LPFC_VPI_REGISTERED;
+ vport->fc_flag &= ~FC_VPORT_NEEDS_REG_VPI;
+ spin_unlock_irq(shost->host_lock);
}
if (pmb->u.mb.mbxCommand == MBX_REG_LOGIN64) {
* lpfc_sli4_scmd_to_wqidx_distr - scsi command to SLI4 WQ index distribution
* @phba: Pointer to HBA context object.
*
- * This routine performs a round robin SCSI command to SLI4 FCP WQ index
+ * This routine performs a roundrobin SCSI command to SLI4 FCP WQ index
* distribution. This is called by __lpfc_sli_issue_iocb_s4() with the hbalock
* held.
*
uint16_t abrt_iotag;
struct lpfc_iocbq *abrtiocbq;
struct ulp_bde64 *bpl = NULL;
- uint32_t els_id = ELS_ID_DEFAULT;
+ uint32_t els_id = LPFC_ELS_ID_DEFAULT;
int numBdes, i;
struct ulp_bde64 bde;
memcpy(wqe, &iocbq->iocb, sizeof(union lpfc_wqe));
abort_tag = (uint32_t) iocbq->iotag;
xritag = iocbq->sli4_xritag;
- wqe->words[7] = 0; /* The ct field has moved so reset */
+ wqe->generic.wqe_com.word7 = 0; /* The ct field has moved so reset */
/* words0-2 bpl convert bde */
if (iocbq->iocb.un.genreq64.bdl.bdeFlags == BUFF_TYPE_BLP_64) {
numBdes = iocbq->iocb.un.genreq64.bdl.bdeSize /
* contains the FCFI and remote N_Port_ID is
* in word 5.
*/
-
ct = ((iocbq->iocb.ulpCt_h << 1) | iocbq->iocb.ulpCt_l);
- bf_set(lpfc_wqe_gen_context, &wqe->generic,
- iocbq->iocb.ulpContext);
-
- bf_set(lpfc_wqe_gen_ct, &wqe->generic, ct);
- bf_set(lpfc_wqe_gen_pu, &wqe->generic, 0);
+ bf_set(wqe_ctxt_tag, &wqe->els_req.wqe_com,
+ iocbq->iocb.ulpContext);
+ bf_set(wqe_ct, &wqe->els_req.wqe_com, ct);
+ bf_set(wqe_pu, &wqe->els_req.wqe_com, 0);
/* CCP CCPE PV PRI in word10 were set in the memcpy */
-
if (command_type == ELS_COMMAND_FIP) {
els_id = ((iocbq->iocb_flag & LPFC_FIP_ELS_ID_MASK)
>> LPFC_FIP_ELS_ID_SHIFT);
}
- bf_set(lpfc_wqe_gen_els_id, &wqe->generic, els_id);
-
+ bf_set(wqe_els_id, &wqe->els_req.wqe_com, els_id);
+ bf_set(wqe_dbde, &wqe->els_req.wqe_com, 1);
+ bf_set(wqe_iod, &wqe->els_req.wqe_com, LPFC_WQE_IOD_READ);
+ bf_set(wqe_qosd, &wqe->els_req.wqe_com, 1);
+ bf_set(wqe_lenloc, &wqe->els_req.wqe_com, LPFC_WQE_LENLOC_NONE);
+ bf_set(wqe_ebde_cnt, &wqe->els_req.wqe_com, 0);
break;
case CMD_XMIT_SEQUENCE64_CX:
- bf_set(lpfc_wqe_gen_context, &wqe->generic,
- iocbq->iocb.un.ulpWord[3]);
- wqe->generic.word3 = 0;
- bf_set(wqe_rcvoxid, &wqe->generic, iocbq->iocb.ulpContext);
+ bf_set(wqe_ctxt_tag, &wqe->xmit_sequence.wqe_com,
+ iocbq->iocb.un.ulpWord[3]);
+ bf_set(wqe_rcvoxid, &wqe->xmit_sequence.wqe_com,
+ iocbq->iocb.ulpContext);
/* The entire sequence is transmitted for this IOCB */
xmit_len = total_len;
cmnd = CMD_XMIT_SEQUENCE64_CR;
case CMD_XMIT_SEQUENCE64_CR:
- /* word3 iocb=io_tag32 wqe=payload_offset */
- /* payload offset used for multilpe outstanding
- * sequences on the same exchange
- */
- wqe->words[3] = 0;
+ /* word3 iocb=io_tag32 wqe=reserved */
+ wqe->xmit_sequence.rsvd3 = 0;
/* word4 relative_offset memcpy */
/* word5 r_ctl/df_ctl memcpy */
- bf_set(lpfc_wqe_gen_pu, &wqe->generic, 0);
+ bf_set(wqe_pu, &wqe->xmit_sequence.wqe_com, 0);
+ bf_set(wqe_dbde, &wqe->xmit_sequence.wqe_com, 1);
+ bf_set(wqe_iod, &wqe->xmit_sequence.wqe_com,
+ LPFC_WQE_IOD_WRITE);
+ bf_set(wqe_lenloc, &wqe->xmit_sequence.wqe_com,
+ LPFC_WQE_LENLOC_WORD12);
+ bf_set(wqe_ebde_cnt, &wqe->xmit_sequence.wqe_com, 0);
wqe->xmit_sequence.xmit_len = xmit_len;
command_type = OTHER_COMMAND;
break;
case CMD_XMIT_BCAST64_CN:
- /* word3 iocb=iotag32 wqe=payload_len */
- wqe->words[3] = 0; /* no definition for this in wqe */
+ /* word3 iocb=iotag32 wqe=seq_payload_len */
+ wqe->xmit_bcast64.seq_payload_len = xmit_len;
/* word4 iocb=rsvd wqe=rsvd */
/* word5 iocb=rctl/type/df_ctl wqe=rctl/type/df_ctl memcpy */
/* word6 iocb=ctxt_tag/io_tag wqe=ctxt_tag/xri */
- bf_set(lpfc_wqe_gen_ct, &wqe->generic,
+ bf_set(wqe_ct, &wqe->xmit_bcast64.wqe_com,
((iocbq->iocb.ulpCt_h << 1) | iocbq->iocb.ulpCt_l));
+ bf_set(wqe_dbde, &wqe->xmit_bcast64.wqe_com, 1);
+ bf_set(wqe_iod, &wqe->xmit_bcast64.wqe_com, LPFC_WQE_IOD_WRITE);
+ bf_set(wqe_lenloc, &wqe->xmit_bcast64.wqe_com,
+ LPFC_WQE_LENLOC_WORD3);
+ bf_set(wqe_ebde_cnt, &wqe->xmit_bcast64.wqe_com, 0);
break;
case CMD_FCP_IWRITE64_CR:
command_type = FCP_COMMAND_DATA_OUT;
- /* The struct for wqe fcp_iwrite has 3 fields that are somewhat
- * confusing.
- * word3 is payload_len: byte offset to the sgl entry for the
- * fcp_command.
- * word4 is total xfer len, same as the IOCB->ulpParameter.
- * word5 is initial xfer len 0 = wait for xfer-ready
- */
-
- /* Always wait for xfer-ready before sending data */
- wqe->fcp_iwrite.initial_xfer_len = 0;
- /* word 4 (xfer length) should have been set on the memcpy */
-
- /* allow write to fall through to read */
+ /* word3 iocb=iotag wqe=payload_offset_len */
+ /* Add the FCP_CMD and FCP_RSP sizes to get the offset */
+ wqe->fcp_iwrite.payload_offset_len =
+ xmit_len + sizeof(struct fcp_rsp);
+ /* word4 iocb=parameter wqe=total_xfer_length memcpy */
+ /* word5 iocb=initial_xfer_len wqe=initial_xfer_len memcpy */
+ bf_set(wqe_erp, &wqe->fcp_iwrite.wqe_com,
+ iocbq->iocb.ulpFCP2Rcvy);
+ bf_set(wqe_lnk, &wqe->fcp_iwrite.wqe_com, iocbq->iocb.ulpXS);
+ /* Always open the exchange */
+ bf_set(wqe_xc, &wqe->fcp_iwrite.wqe_com, 0);
+ bf_set(wqe_dbde, &wqe->fcp_iwrite.wqe_com, 1);
+ bf_set(wqe_iod, &wqe->fcp_iwrite.wqe_com, LPFC_WQE_IOD_WRITE);
+ bf_set(wqe_lenloc, &wqe->fcp_iwrite.wqe_com,
+ LPFC_WQE_LENLOC_WORD4);
+ bf_set(wqe_ebde_cnt, &wqe->fcp_iwrite.wqe_com, 0);
+ bf_set(wqe_pu, &wqe->fcp_iwrite.wqe_com, iocbq->iocb.ulpPU);
+ break;
case CMD_FCP_IREAD64_CR:
- /* FCP_CMD is always the 1st sgl entry */
- wqe->fcp_iread.payload_len =
+ /* word3 iocb=iotag wqe=payload_offset_len */
+ /* Add the FCP_CMD and FCP_RSP sizes to get the offset */
+ wqe->fcp_iread.payload_offset_len =
xmit_len + sizeof(struct fcp_rsp);
-
- /* word 4 (xfer length) should have been set on the memcpy */
-
- bf_set(lpfc_wqe_gen_erp, &wqe->generic,
- iocbq->iocb.ulpFCP2Rcvy);
- bf_set(lpfc_wqe_gen_lnk, &wqe->generic, iocbq->iocb.ulpXS);
- /* The XC bit and the XS bit are similar. The driver never
- * tracked whether or not the exchange was previouslly open.
- * XC = Exchange create, 0 is create. 1 is already open.
- * XS = link cmd: 1 do not close the exchange after command.
- * XS = 0 close exchange when command completes.
- * The only time we would not set the XC bit is when the XS bit
- * is set and we are sending our 2nd or greater command on
- * this exchange.
- */
+ /* word4 iocb=parameter wqe=total_xfer_length memcpy */
+ /* word5 iocb=initial_xfer_len wqe=initial_xfer_len memcpy */
+ bf_set(wqe_erp, &wqe->fcp_iread.wqe_com,
+ iocbq->iocb.ulpFCP2Rcvy);
+ bf_set(wqe_lnk, &wqe->fcp_iread.wqe_com, iocbq->iocb.ulpXS);
/* Always open the exchange */
bf_set(wqe_xc, &wqe->fcp_iread.wqe_com, 0);
-
- wqe->words[10] &= 0xffff0000; /* zero out ebde count */
- bf_set(lpfc_wqe_gen_pu, &wqe->generic, iocbq->iocb.ulpPU);
- break;
+ bf_set(wqe_dbde, &wqe->fcp_iread.wqe_com, 1);
+ bf_set(wqe_iod, &wqe->fcp_iread.wqe_com, LPFC_WQE_IOD_READ);
+ bf_set(wqe_lenloc, &wqe->fcp_iread.wqe_com,
+ LPFC_WQE_LENLOC_WORD4);
+ bf_set(wqe_ebde_cnt, &wqe->fcp_iread.wqe_com, 0);
+ bf_set(wqe_pu, &wqe->fcp_iread.wqe_com, iocbq->iocb.ulpPU);
+ break;
case CMD_FCP_ICMND64_CR:
+ /* word3 iocb=IO_TAG wqe=reserved */
+ wqe->fcp_icmd.rsrvd3 = 0;
+ bf_set(wqe_pu, &wqe->fcp_icmd.wqe_com, 0);
/* Always open the exchange */
- bf_set(wqe_xc, &wqe->fcp_iread.wqe_com, 0);
-
- wqe->words[4] = 0;
- wqe->words[10] &= 0xffff0000; /* zero out ebde count */
- bf_set(lpfc_wqe_gen_pu, &wqe->generic, 0);
+ bf_set(wqe_xc, &wqe->fcp_icmd.wqe_com, 0);
+ bf_set(wqe_dbde, &wqe->fcp_icmd.wqe_com, 1);
+ bf_set(wqe_iod, &wqe->fcp_icmd.wqe_com, LPFC_WQE_IOD_WRITE);
+ bf_set(wqe_qosd, &wqe->fcp_icmd.wqe_com, 1);
+ bf_set(wqe_lenloc, &wqe->fcp_icmd.wqe_com,
+ LPFC_WQE_LENLOC_NONE);
+ bf_set(wqe_ebde_cnt, &wqe->fcp_icmd.wqe_com, 0);
break;
case CMD_GEN_REQUEST64_CR:
- /* word3 command length is described as byte offset to the
- * rsp_data. Would always be 16, sizeof(struct sli4_sge)
- * sgl[0] = cmnd
- * sgl[1] = rsp.
- *
- */
- wqe->gen_req.command_len = xmit_len;
- /* Word4 parameter copied in the memcpy */
- /* Word5 [rctl, type, df_ctl, la] copied in memcpy */
+ /* word3 iocb=IO_TAG wqe=request_payload_len */
+ wqe->gen_req.request_payload_len = xmit_len;
+ /* word4 iocb=parameter wqe=relative_offset memcpy */
+ /* word5 [rctl, type, df_ctl, la] copied in memcpy */
/* word6 context tag copied in memcpy */
if (iocbq->iocb.ulpCt_h || iocbq->iocb.ulpCt_l) {
ct = ((iocbq->iocb.ulpCt_h << 1) | iocbq->iocb.ulpCt_l);
ct, iocbq->iocb.ulpCommand);
return IOCB_ERROR;
}
- bf_set(lpfc_wqe_gen_ct, &wqe->generic, 0);
- bf_set(wqe_tmo, &wqe->gen_req.wqe_com,
- iocbq->iocb.ulpTimeout);
-
- bf_set(lpfc_wqe_gen_pu, &wqe->generic, iocbq->iocb.ulpPU);
+ bf_set(wqe_ct, &wqe->gen_req.wqe_com, 0);
+ bf_set(wqe_tmo, &wqe->gen_req.wqe_com, iocbq->iocb.ulpTimeout);
+ bf_set(wqe_pu, &wqe->gen_req.wqe_com, iocbq->iocb.ulpPU);
+ bf_set(wqe_dbde, &wqe->gen_req.wqe_com, 1);
+ bf_set(wqe_iod, &wqe->gen_req.wqe_com, LPFC_WQE_IOD_READ);
+ bf_set(wqe_qosd, &wqe->gen_req.wqe_com, 1);
+ bf_set(wqe_lenloc, &wqe->gen_req.wqe_com, LPFC_WQE_LENLOC_NONE);
+ bf_set(wqe_ebde_cnt, &wqe->gen_req.wqe_com, 0);
command_type = OTHER_COMMAND;
break;
case CMD_XMIT_ELS_RSP64_CX:
/* words0-2 BDE memcpy */
- /* word3 iocb=iotag32 wqe=rsvd */
- wqe->words[3] = 0;
+ /* word3 iocb=iotag32 wqe=response_payload_len */
+ wqe->xmit_els_rsp.response_payload_len = xmit_len;
/* word4 iocb=did wge=rsvd. */
- wqe->words[4] = 0;
+ wqe->xmit_els_rsp.rsvd4 = 0;
/* word5 iocb=rsvd wge=did */
bf_set(wqe_els_did, &wqe->xmit_els_rsp.wqe_dest,
iocbq->iocb.un.elsreq64.remoteID);
-
- bf_set(lpfc_wqe_gen_ct, &wqe->generic,
- ((iocbq->iocb.ulpCt_h << 1) | iocbq->iocb.ulpCt_l));
-
- bf_set(lpfc_wqe_gen_pu, &wqe->generic, iocbq->iocb.ulpPU);
- bf_set(wqe_rcvoxid, &wqe->generic, iocbq->iocb.ulpContext);
+ bf_set(wqe_ct, &wqe->xmit_els_rsp.wqe_com,
+ ((iocbq->iocb.ulpCt_h << 1) | iocbq->iocb.ulpCt_l));
+ bf_set(wqe_pu, &wqe->xmit_els_rsp.wqe_com, iocbq->iocb.ulpPU);
+ bf_set(wqe_rcvoxid, &wqe->xmit_els_rsp.wqe_com,
+ iocbq->iocb.ulpContext);
if (!iocbq->iocb.ulpCt_h && iocbq->iocb.ulpCt_l)
- bf_set(lpfc_wqe_gen_context, &wqe->generic,
+ bf_set(wqe_ctxt_tag, &wqe->xmit_els_rsp.wqe_com,
iocbq->vport->vpi + phba->vpi_base);
+ bf_set(wqe_dbde, &wqe->xmit_els_rsp.wqe_com, 1);
+ bf_set(wqe_iod, &wqe->xmit_els_rsp.wqe_com, LPFC_WQE_IOD_WRITE);
+ bf_set(wqe_qosd, &wqe->xmit_els_rsp.wqe_com, 1);
+ bf_set(wqe_lenloc, &wqe->xmit_els_rsp.wqe_com,
+ LPFC_WQE_LENLOC_WORD3);
+ bf_set(wqe_ebde_cnt, &wqe->xmit_els_rsp.wqe_com, 0);
command_type = OTHER_COMMAND;
break;
case CMD_CLOSE_XRI_CN:
else
bf_set(abort_cmd_ia, &wqe->abort_cmd, 0);
bf_set(abort_cmd_criteria, &wqe->abort_cmd, T_XRI_TAG);
- wqe->words[5] = 0;
- bf_set(lpfc_wqe_gen_ct, &wqe->generic,
+ /* word5 iocb=CONTEXT_TAG|IO_TAG wqe=reserved */
+ wqe->abort_cmd.rsrvd5 = 0;
+ bf_set(wqe_ct, &wqe->abort_cmd.wqe_com,
((iocbq->iocb.ulpCt_h << 1) | iocbq->iocb.ulpCt_l));
abort_tag = iocbq->iocb.un.acxri.abortIoTag;
/*
* The abort handler will send us CMD_ABORT_XRI_CN or
* CMD_CLOSE_XRI_CN and the fw only accepts CMD_ABORT_XRI_CX
*/
- bf_set(lpfc_wqe_gen_command, &wqe->generic, CMD_ABORT_XRI_CX);
+ bf_set(wqe_cmnd, &wqe->abort_cmd.wqe_com, CMD_ABORT_XRI_CX);
+ bf_set(wqe_qosd, &wqe->abort_cmd.wqe_com, 1);
+ bf_set(wqe_lenloc, &wqe->abort_cmd.wqe_com,
+ LPFC_WQE_LENLOC_NONE);
cmnd = CMD_ABORT_XRI_CX;
command_type = OTHER_COMMAND;
xritag = 0;
bf_set(wqe_xmit_bls_pt, &wqe->xmit_bls_rsp.wqe_dest, 0x1);
bf_set(wqe_ctxt_tag, &wqe->xmit_bls_rsp.wqe_com,
iocbq->iocb.ulpContext);
+ bf_set(wqe_qosd, &wqe->xmit_bls_rsp.wqe_com, 1);
+ bf_set(wqe_lenloc, &wqe->xmit_bls_rsp.wqe_com,
+ LPFC_WQE_LENLOC_NONE);
/* Overwrite the pre-set comnd type with OTHER_COMMAND */
command_type = OTHER_COMMAND;
break;
case CMD_XRI_ABORTED_CX:
case CMD_CREATE_XRI_CR: /* Do we expect to use this? */
- /* words0-2 are all 0's no bde */
- /* word3 and word4 are rsvrd */
- wqe->words[3] = 0;
- wqe->words[4] = 0;
- /* word5 iocb=rsvd wge=did */
- /* There is no remote port id in the IOCB? */
- /* Let this fall through and fail */
case CMD_IOCB_FCP_IBIDIR64_CR: /* bidirectional xfer */
case CMD_FCP_TSEND64_CX: /* Target mode send xfer-ready */
case CMD_FCP_TRSP64_CX: /* Target mode rcv */
iocbq->iocb.ulpCommand);
return IOCB_ERROR;
break;
-
}
- bf_set(lpfc_wqe_gen_xri, &wqe->generic, xritag);
- bf_set(lpfc_wqe_gen_request_tag, &wqe->generic, iocbq->iotag);
- wqe->generic.abort_tag = abort_tag;
- bf_set(lpfc_wqe_gen_cmd_type, &wqe->generic, command_type);
- bf_set(lpfc_wqe_gen_command, &wqe->generic, cmnd);
- bf_set(lpfc_wqe_gen_class, &wqe->generic, iocbq->iocb.ulpClass);
- bf_set(lpfc_wqe_gen_cq_id, &wqe->generic, LPFC_WQE_CQ_ID_DEFAULT);
-
+ bf_set(wqe_xri_tag, &wqe->generic.wqe_com, xritag);
+ bf_set(wqe_reqtag, &wqe->generic.wqe_com, iocbq->iotag);
+ wqe->generic.wqe_com.abort_tag = abort_tag;
+ bf_set(wqe_cmd_type, &wqe->generic.wqe_com, command_type);
+ bf_set(wqe_cmnd, &wqe->generic.wqe_com, cmnd);
+ bf_set(wqe_class, &wqe->generic.wqe_com, iocbq->iocb.ulpClass);
+ bf_set(wqe_cqid, &wqe->generic.wqe_com, LPFC_WQE_CQ_ID_DEFAULT);
return 0;
}
}
/**
- * lpfc_sli_issue_abort_iotag - Abort function for a command iocb
+ * lpfc_sli_abort_iotag_issue - Issue abort for a command iocb
* @phba: Pointer to HBA context object.
* @pring: Pointer to driver SLI ring object.
* @cmdiocb: Pointer to driver command iocb object.
*
- * This function issues an abort iocb for the provided command
- * iocb. This function is called with hbalock held.
- * The function returns 0 when it fails due to memory allocation
- * failure or when the command iocb is an abort request.
+ * This function issues an abort iocb for the provided command iocb down to
+ * the port. Other than the case the outstanding command iocb is an abort
+ * request, this function issues abort out unconditionally. This function is
+ * called with hbalock held. The function returns 0 when it fails due to
+ * memory allocation failure or when the command iocb is an abort request.
**/
-int
-lpfc_sli_issue_abort_iotag(struct lpfc_hba *phba, struct lpfc_sli_ring *pring,
+static int
+lpfc_sli_abort_iotag_issue(struct lpfc_hba *phba, struct lpfc_sli_ring *pring,
struct lpfc_iocbq *cmdiocb)
{
struct lpfc_vport *vport = cmdiocb->vport;
struct lpfc_iocbq *abtsiocbp;
IOCB_t *icmd = NULL;
IOCB_t *iabt = NULL;
- int retval = IOCB_ERROR;
+ int retval;
/*
* There are certain command types we don't want to abort. And we
(cmdiocb->iocb_flag & LPFC_DRIVER_ABORTED) != 0)
return 0;
- /* If we're unloading, don't abort iocb on the ELS ring, but change the
- * callback so that nothing happens when it finishes.
- */
- if ((vport->load_flag & FC_UNLOADING) &&
- (pring->ringno == LPFC_ELS_RING)) {
- if (cmdiocb->iocb_flag & LPFC_IO_FABRIC)
- cmdiocb->fabric_iocb_cmpl = lpfc_ignore_els_cmpl;
- else
- cmdiocb->iocb_cmpl = lpfc_ignore_els_cmpl;
- goto abort_iotag_exit;
- }
-
/* issue ABTS for this IOCB based on iotag */
abtsiocbp = __lpfc_sli_get_iocbq(phba);
if (abtsiocbp == NULL)
if (retval)
__lpfc_sli_release_iocbq(phba, abtsiocbp);
+
+ /*
+ * Caller to this routine should check for IOCB_ERROR
+ * and handle it properly. This routine no longer removes
+ * iocb off txcmplq and call compl in case of IOCB_ERROR.
+ */
+ return retval;
+}
+
+/**
+ * lpfc_sli_issue_abort_iotag - Abort function for a command iocb
+ * @phba: Pointer to HBA context object.
+ * @pring: Pointer to driver SLI ring object.
+ * @cmdiocb: Pointer to driver command iocb object.
+ *
+ * This function issues an abort iocb for the provided command iocb. In case
+ * of unloading, the abort iocb will not be issued to commands on the ELS
+ * ring. Instead, the callback function shall be changed to those commands
+ * so that nothing happens when them finishes. This function is called with
+ * hbalock held. The function returns 0 when the command iocb is an abort
+ * request.
+ **/
+int
+lpfc_sli_issue_abort_iotag(struct lpfc_hba *phba, struct lpfc_sli_ring *pring,
+ struct lpfc_iocbq *cmdiocb)
+{
+ struct lpfc_vport *vport = cmdiocb->vport;
+ int retval = IOCB_ERROR;
+ IOCB_t *icmd = NULL;
+
+ /*
+ * There are certain command types we don't want to abort. And we
+ * don't want to abort commands that are already in the process of
+ * being aborted.
+ */
+ icmd = &cmdiocb->iocb;
+ if (icmd->ulpCommand == CMD_ABORT_XRI_CN ||
+ icmd->ulpCommand == CMD_CLOSE_XRI_CN ||
+ (cmdiocb->iocb_flag & LPFC_DRIVER_ABORTED) != 0)
+ return 0;
+
+ /*
+ * If we're unloading, don't abort iocb on the ELS ring, but change
+ * the callback so that nothing happens when it finishes.
+ */
+ if ((vport->load_flag & FC_UNLOADING) &&
+ (pring->ringno == LPFC_ELS_RING)) {
+ if (cmdiocb->iocb_flag & LPFC_IO_FABRIC)
+ cmdiocb->fabric_iocb_cmpl = lpfc_ignore_els_cmpl;
+ else
+ cmdiocb->iocb_cmpl = lpfc_ignore_els_cmpl;
+ goto abort_iotag_exit;
+ }
+
+ /* Now, we try to issue the abort to the cmdiocb out */
+ retval = lpfc_sli_abort_iotag_issue(phba, pring, cmdiocb);
+
abort_iotag_exit:
/*
* Caller to this routine should check for IOCB_ERROR
return retval;
}
+/**
+ * lpfc_sli_iocb_ring_abort - Unconditionally abort all iocbs on an iocb ring
+ * @phba: Pointer to HBA context object.
+ * @pring: Pointer to driver SLI ring object.
+ *
+ * This function aborts all iocbs in the given ring and frees all the iocb
+ * objects in txq. This function issues abort iocbs unconditionally for all
+ * the iocb commands in txcmplq. The iocbs in the txcmplq is not guaranteed
+ * to complete before the return of this function. The caller is not required
+ * to hold any locks.
+ **/
+static void
+lpfc_sli_iocb_ring_abort(struct lpfc_hba *phba, struct lpfc_sli_ring *pring)
+{
+ LIST_HEAD(completions);
+ struct lpfc_iocbq *iocb, *next_iocb;
+
+ if (pring->ringno == LPFC_ELS_RING)
+ lpfc_fabric_abort_hba(phba);
+
+ spin_lock_irq(&phba->hbalock);
+
+ /* Take off all the iocbs on txq for cancelling */
+ list_splice_init(&pring->txq, &completions);
+ pring->txq_cnt = 0;
+
+ /* Next issue ABTS for everything on the txcmplq */
+ list_for_each_entry_safe(iocb, next_iocb, &pring->txcmplq, list)
+ lpfc_sli_abort_iotag_issue(phba, pring, iocb);
+
+ spin_unlock_irq(&phba->hbalock);
+
+ /* Cancel all the IOCBs from the completions list */
+ lpfc_sli_cancel_iocbs(phba, &completions, IOSTAT_LOCAL_REJECT,
+ IOERR_SLI_ABORTED);
+}
+
+/**
+ * lpfc_sli_hba_iocb_abort - Abort all iocbs to an hba.
+ * @phba: pointer to lpfc HBA data structure.
+ *
+ * This routine will abort all pending and outstanding iocbs to an HBA.
+ **/
+void
+lpfc_sli_hba_iocb_abort(struct lpfc_hba *phba)
+{
+ struct lpfc_sli *psli = &phba->sli;
+ struct lpfc_sli_ring *pring;
+ int i;
+
+ for (i = 0; i < psli->num_rings; i++) {
+ pring = &psli->ring[i];
+ lpfc_sli_iocb_ring_abort(phba, pring);
+ }
+}
+
/**
* lpfc_sli_validate_fcp_iocb - find commands associated with a vport or LUN
* @iocbq: Pointer to driver iocb object.
/* Issue the mailbox command asynchronously */
mboxq->vport = phba->pport;
mboxq->mbox_cmpl = lpfc_mbx_cmpl_fcf_scan_read_fcf_rec;
+
+ spin_lock_irq(&phba->hbalock);
+ phba->hba_flag |= FCF_TS_INPROG;
+ spin_unlock_irq(&phba->hbalock);
+
rc = lpfc_sli_issue_mbox(phba, mboxq, MBX_NOWAIT);
if (rc == MBX_NOT_FINISHED)
error = -EIO;
else {
- spin_lock_irq(&phba->hbalock);
- phba->hba_flag |= FCF_DISC_INPROGRESS;
- spin_unlock_irq(&phba->hbalock);
/* Reset eligible FCF count for new scan */
if (fcf_index == LPFC_FCOE_FCF_GET_FIRST)
phba->fcf.eligible_fcf_cnt = 0;
if (error) {
if (mboxq)
lpfc_sli4_mbox_cmd_free(phba, mboxq);
- /* FCF scan failed, clear FCF_DISC_INPROGRESS flag */
+ /* FCF scan failed, clear FCF_TS_INPROG flag */
spin_lock_irq(&phba->hbalock);
- phba->hba_flag &= ~FCF_DISC_INPROGRESS;
+ phba->hba_flag &= ~FCF_TS_INPROG;
spin_unlock_irq(&phba->hbalock);
}
return error;
}
/**
- * lpfc_sli4_fcf_rr_read_fcf_rec - Read hba fcf record for round robin fcf.
+ * lpfc_sli4_fcf_rr_read_fcf_rec - Read hba fcf record for roundrobin fcf.
* @phba: pointer to lpfc hba data structure.
* @fcf_index: FCF table entry offset.
*
* This routine is invoked to read an FCF record indicated by @fcf_index
- * and to use it for FLOGI round robin FCF failover.
+ * and to use it for FLOGI roundrobin FCF failover.
*
* Return 0 if the mailbox command is submitted sucessfully, none 0
* otherwise.
* @fcf_index: FCF table entry offset.
*
* This routine is invoked to read an FCF record indicated by @fcf_index to
- * determine whether it's eligible for FLOGI round robin failover list.
+ * determine whether it's eligible for FLOGI roundrobin failover list.
*
* Return 0 if the mailbox command is submitted sucessfully, none 0
* otherwise.
*
* This routine is to get the next eligible FCF record index in a round
* robin fashion. If the next eligible FCF record index equals to the
- * initial round robin FCF record index, LPFC_FCOE_FCF_NEXT_NONE (0xFFFF)
+ * initial roundrobin FCF record index, LPFC_FCOE_FCF_NEXT_NONE (0xFFFF)
* shall be returned, otherwise, the next eligible FCF record's index
* shall be returned.
**/
return LPFC_FCOE_FCF_NEXT_NONE;
}
- /* Check roundrobin failover index bmask stop condition */
- if (next_fcf_index == phba->fcf.fcf_rr_init_indx) {
- if (!(phba->fcf.fcf_flag & FCF_REDISC_RRU)) {
- lpfc_printf_log(phba, KERN_WARNING, LOG_FIP,
- "2847 Round robin failover FCF index "
- "search hit stop condition:x%x\n",
- next_fcf_index);
- return LPFC_FCOE_FCF_NEXT_NONE;
- }
- /* The roundrobin failover index bmask updated, start over */
- lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
- "2848 Round robin failover FCF index bmask "
- "updated, start over\n");
- spin_lock_irq(&phba->hbalock);
- phba->fcf.fcf_flag &= ~FCF_REDISC_RRU;
- spin_unlock_irq(&phba->hbalock);
- return phba->fcf.fcf_rr_init_indx;
- }
-
lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
- "2845 Get next round robin failover "
- "FCF index x%x\n", next_fcf_index);
+ "2845 Get next roundrobin failover FCF (x%x)\n",
+ next_fcf_index);
+
return next_fcf_index;
}
* @phba: pointer to lpfc hba data structure.
*
* This routine sets the FCF record index in to the eligible bmask for
- * round robin failover search. It checks to make sure that the index
+ * roundrobin failover search. It checks to make sure that the index
* does not go beyond the range of the driver allocated bmask dimension
* before setting the bit.
*
{
if (fcf_index >= LPFC_SLI4_FCF_TBL_INDX_MAX) {
lpfc_printf_log(phba, KERN_ERR, LOG_FIP,
- "2610 HBA FCF index reached driver's "
- "book keeping dimension: fcf_index:%d, "
- "driver_bmask_max:%d\n",
+ "2610 FCF (x%x) reached driver's book "
+ "keeping dimension:x%x\n",
fcf_index, LPFC_SLI4_FCF_TBL_INDX_MAX);
return -EINVAL;
}
/* Set the eligible FCF record index bmask */
set_bit(fcf_index, phba->fcf.fcf_rr_bmask);
- /* Set the roundrobin index bmask updated */
- spin_lock_irq(&phba->hbalock);
- phba->fcf.fcf_flag |= FCF_REDISC_RRU;
- spin_unlock_irq(&phba->hbalock);
-
lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
- "2790 Set FCF index x%x to round robin failover "
+ "2790 Set FCF (x%x) to roundrobin FCF failover "
"bmask\n", fcf_index);
return 0;
* @phba: pointer to lpfc hba data structure.
*
* This routine clears the FCF record index from the eligible bmask for
- * round robin failover search. It checks to make sure that the index
+ * roundrobin failover search. It checks to make sure that the index
* does not go beyond the range of the driver allocated bmask dimension
* before clearing the bit.
**/
{
if (fcf_index >= LPFC_SLI4_FCF_TBL_INDX_MAX) {
lpfc_printf_log(phba, KERN_ERR, LOG_FIP,
- "2762 HBA FCF index goes beyond driver's "
- "book keeping dimension: fcf_index:%d, "
- "driver_bmask_max:%d\n",
+ "2762 FCF (x%x) reached driver's book "
+ "keeping dimension:x%x\n",
fcf_index, LPFC_SLI4_FCF_TBL_INDX_MAX);
return;
}
clear_bit(fcf_index, phba->fcf.fcf_rr_bmask);
lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
- "2791 Clear FCF index x%x from round robin failover "
+ "2791 Clear FCF (x%x) from roundrobin failover "
"bmask\n", fcf_index);
}
}
} else {
lpfc_printf_log(phba, KERN_INFO, LOG_FIP,
- "2775 Start FCF rediscovery quiescent period "
- "wait timer before scaning FCF table\n");
+ "2775 Start FCF rediscover quiescent timer\n");
/*
* Start FCF rediscovery wait timer for pending FCF
* before rescan FCF record table.
*******************************************************************/
#define LPFC_ACTIVE_MBOX_WAIT_CNT 100
+#define LPFC_XRI_EXCH_BUSY_WAIT_TMO 10000
+#define LPFC_XRI_EXCH_BUSY_WAIT_T1 10
+#define LPFC_XRI_EXCH_BUSY_WAIT_T2 30000
#define LPFC_RELEASE_NOTIFICATION_INTERVAL 32
#define LPFC_GET_QE_REL_INT 32
#define LPFC_RPI_LOW_WATER_MARK 10
+#define LPFC_UNREG_FCF 1
+#define LPFC_SKIP_UNREG_FCF 0
+
/* Amount of time in seconds for waiting FCF rediscovery to complete */
#define LPFC_FCF_REDISCOVER_WAIT_TMO 2000 /* msec */
#define FCF_REDISC_PEND 0x80 /* FCF rediscovery pending */
#define FCF_REDISC_EVT 0x100 /* FCF rediscovery event to worker thread */
#define FCF_REDISC_FOV 0x200 /* Post FCF rediscovery fast failover */
-#define FCF_REDISC_RRU 0x400 /* Roundrobin bitmap updated */
+#define FCF_REDISC_PROG (FCF_REDISC_PEND | FCF_REDISC_EVT)
uint32_t addr_mode;
- uint16_t fcf_rr_init_indx;
uint32_t eligible_fcf_cnt;
struct lpfc_fcf_rec current_rec;
struct lpfc_fcf_rec failover_rec;
* included with this package. *
*******************************************************************/
-#define LPFC_DRIVER_VERSION "8.3.17"
+#define LPFC_DRIVER_VERSION "8.3.18"
#define LPFC_DRIVER_NAME "lpfc"
#define LPFC_SP_DRIVER_HANDLER_NAME "lpfc:sp"
#define LPFC_FP_DRIVER_HANDLER_NAME "lpfc:fp"
* 2 of the License, or (at your option) any later version.
*
* FILE : megaraid_sas.c
- * Version : v00.00.04.17.1-rc1
+ * Version : v00.00.04.31-rc1
*
* Authors:
* (email-id : megaraidlinux@lsi.com)
MODULE_PARM_DESC(poll_mode_io,
"Complete cmds from IO path, (default=0)");
+/*
+ * Number of sectors per IO command
+ * Will be set in megasas_init_mfi if user does not provide
+ */
+static unsigned int max_sectors;
+module_param_named(max_sectors, max_sectors, int, 0);
+MODULE_PARM_DESC(max_sectors,
+ "Maximum number of sectors per IO command");
+
MODULE_LICENSE("GPL");
MODULE_VERSION(MEGASAS_VERSION);
MODULE_AUTHOR("megaraidlinux@lsi.com");
static DECLARE_WAIT_QUEUE_HEAD(megasas_poll_wait);
static u32 support_poll_for_event;
static u32 megasas_dbg_lvl;
+static u32 support_device_change;
/* define lock for aen poll */
spinlock_t poll_aen_lock;
megasas_check_reset_gen2(struct megasas_instance *instance,
struct megasas_register_set __iomem *regs)
{
+ if (instance->adprecovery != MEGASAS_HBA_OPERATIONAL) {
+ return 1;
+ }
+
return 0;
}
mfi_sgl->sge_skinny[i].length = sg_dma_len(os_sgl);
mfi_sgl->sge_skinny[i].phys_addr =
sg_dma_address(os_sgl);
+ mfi_sgl->sge_skinny[i].flag = 0;
}
}
return sge_count;
}
}
+static void
+megasas_internal_reset_defer_cmds(struct megasas_instance *instance);
+
+static void
+process_fw_state_change_wq(struct work_struct *work);
+
+void megasas_do_ocr(struct megasas_instance *instance)
+{
+ if ((instance->pdev->device == PCI_DEVICE_ID_LSI_SAS1064R) ||
+ (instance->pdev->device == PCI_DEVICE_ID_DELL_PERC5) ||
+ (instance->pdev->device == PCI_DEVICE_ID_LSI_VERDE_ZCR)) {
+ *instance->consumer = MEGASAS_ADPRESET_INPROG_SIGN;
+ }
+ instance->instancet->disable_intr(instance->reg_set);
+ instance->adprecovery = MEGASAS_ADPRESET_SM_INFAULT;
+ instance->issuepend_done = 0;
+
+ atomic_set(&instance->fw_outstanding, 0);
+ megasas_internal_reset_defer_cmds(instance);
+ process_fw_state_change_wq(&instance->work_init);
+}
+
/**
* megasas_wait_for_outstanding - Wait for all outstanding cmds
* @instance: Adapter soft state
unsigned long flags;
struct list_head clist_local;
struct megasas_cmd *reset_cmd;
+ u32 fw_state;
+ u8 kill_adapter_flag;
spin_lock_irqsave(&instance->hba_lock, flags);
adprecovery = instance->adprecovery;
msleep(1000);
}
- if (atomic_read(&instance->fw_outstanding)) {
+ i = 0;
+ kill_adapter_flag = 0;
+ do {
+ fw_state = instance->instancet->read_fw_status_reg(
+ instance->reg_set) & MFI_STATE_MASK;
+ if ((fw_state == MFI_STATE_FAULT) &&
+ (instance->disableOnlineCtrlReset == 0)) {
+ if (i == 3) {
+ kill_adapter_flag = 2;
+ break;
+ }
+ megasas_do_ocr(instance);
+ kill_adapter_flag = 1;
+
+ /* wait for 1 secs to let FW finish the pending cmds */
+ msleep(1000);
+ }
+ i++;
+ } while (i <= 3);
+
+ if (atomic_read(&instance->fw_outstanding) &&
+ !kill_adapter_flag) {
+ if (instance->disableOnlineCtrlReset == 0) {
+
+ megasas_do_ocr(instance);
+
+ /* wait for 5 secs to let FW finish the pending cmds */
+ for (i = 0; i < wait_time; i++) {
+ int outstanding =
+ atomic_read(&instance->fw_outstanding);
+ if (!outstanding)
+ return SUCCESS;
+ msleep(1000);
+ }
+ }
+ }
+
+ if (atomic_read(&instance->fw_outstanding) ||
+ (kill_adapter_flag == 2)) {
printk(KERN_NOTICE "megaraid_sas: pending cmds after reset\n");
/*
* Send signal to FW to stop processing any pending cmds.
return -ENOMEM;
}
+ memset(cmd->frame, 0, total_sz);
cmd->frame->io.context = cmd->index;
cmd->frame->io.pad_0 = 0;
}
instance->max_fw_cmds - MEGASAS_INT_CMDS;
host->this_id = instance->init_id;
host->sg_tablesize = instance->max_num_sge;
+ /*
+ * Check if the module parameter value for max_sectors can be used
+ */
+ if (max_sectors && max_sectors < instance->max_sectors_per_req)
+ instance->max_sectors_per_req = max_sectors;
+ else {
+ if (max_sectors) {
+ if (((instance->pdev->device ==
+ PCI_DEVICE_ID_LSI_SAS1078GEN2) ||
+ (instance->pdev->device ==
+ PCI_DEVICE_ID_LSI_SAS0079GEN2)) &&
+ (max_sectors <= MEGASAS_MAX_SECTORS)) {
+ instance->max_sectors_per_req = max_sectors;
+ } else {
+ printk(KERN_INFO "megasas: max_sectors should be > 0"
+ "and <= %d (or < 1MB for GEN2 controller)\n",
+ instance->max_sectors_per_req);
+ }
+ }
+ }
+
host->max_sectors = instance->max_sectors_per_req;
host->cmd_per_lun = 128;
host->max_channel = MEGASAS_MAX_CHANNELS - 1;
static DRIVER_ATTR(support_poll_for_event, S_IRUGO,
megasas_sysfs_show_support_poll_for_event, NULL);
+ static ssize_t
+megasas_sysfs_show_support_device_change(struct device_driver *dd, char *buf)
+{
+ return sprintf(buf, "%u\n", support_device_change);
+}
+
+static DRIVER_ATTR(support_device_change, S_IRUGO,
+ megasas_sysfs_show_support_device_change, NULL);
+
static ssize_t
megasas_sysfs_show_dbg_lvl(struct device_driver *dd, char *buf)
{
MEGASAS_EXT_VERSION);
support_poll_for_event = 2;
+ support_device_change = 1;
memset(&megasas_mgmt_info, 0, sizeof(megasas_mgmt_info));
if (rval)
goto err_dcf_poll_mode_io;
+ rval = driver_create_file(&megasas_pci_driver.driver,
+ &driver_attr_support_device_change);
+ if (rval)
+ goto err_dcf_support_device_change;
+
return rval;
+err_dcf_support_device_change:
+ driver_remove_file(&megasas_pci_driver.driver,
+ &driver_attr_poll_mode_io);
+
err_dcf_poll_mode_io:
driver_remove_file(&megasas_pci_driver.driver,
&driver_attr_dbg_lvl);
&driver_attr_poll_mode_io);
driver_remove_file(&megasas_pci_driver.driver,
&driver_attr_dbg_lvl);
+ driver_remove_file(&megasas_pci_driver.driver,
+ &driver_attr_support_poll_for_event);
+ driver_remove_file(&megasas_pci_driver.driver,
+ &driver_attr_support_device_change);
driver_remove_file(&megasas_pci_driver.driver,
&driver_attr_release_date);
driver_remove_file(&megasas_pci_driver.driver, &driver_attr_version);
/*
* MegaRAID SAS Driver meta data
*/
-#define MEGASAS_VERSION "00.00.04.17.1-rc1"
-#define MEGASAS_RELDATE "Oct. 29, 2009"
-#define MEGASAS_EXT_VERSION "Thu. Oct. 29, 11:41:51 PST 2009"
+#define MEGASAS_VERSION "00.00.04.31-rc1"
+#define MEGASAS_RELDATE "May 3, 2010"
+#define MEGASAS_EXT_VERSION "Mon. May 3, 11:41:51 PST 2010"
/*
* Device IDs
#define MEGASAS_MAX_LD_IDS (MEGASAS_MAX_LD_CHANNELS * \
MEGASAS_MAX_DEV_PER_CHANNEL)
+#define MEGASAS_MAX_SECTORS (2*1024)
#define MEGASAS_DBG_LVL 1
#define MEGASAS_FW_BUSY 1
{
struct request *rq = or->request;
- _osd_free_seg(or, &or->set_attr);
- _osd_free_seg(or, &or->enc_get_attr);
- _osd_free_seg(or, &or->get_attr);
-
if (rq) {
if (rq->next_rq) {
_put_request(rq->next_rq);
_put_request(rq);
}
+
+ _osd_free_seg(or, &or->get_attr);
+ _osd_free_seg(or, &or->enc_get_attr);
+ _osd_free_seg(or, &or->set_attr);
+ _osd_free_seg(or, &or->cdb_cont);
+
_osd_request_free(or);
}
EXPORT_SYMBOL(osd_end_request);
return 0;
}
+static int _alloc_cdb_cont(struct osd_request *or, unsigned total_bytes)
+{
+ OSD_DEBUG("total_bytes=%d\n", total_bytes);
+ return _osd_realloc_seg(or, &or->cdb_cont, total_bytes);
+}
+
static int _alloc_set_attr_list(struct osd_request *or,
const struct osd_attr *oa, unsigned nelem, unsigned add_bytes)
{
}
EXPORT_SYMBOL(osd_req_read_kern);
+static int _add_sg_continuation_descriptor(struct osd_request *or,
+ const struct osd_sg_entry *sglist, unsigned numentries, u64 *len)
+{
+ struct osd_sg_continuation_descriptor *oscd;
+ u32 oscd_size;
+ unsigned i;
+ int ret;
+
+ oscd_size = sizeof(*oscd) + numentries * sizeof(oscd->entries[0]);
+
+ if (!or->cdb_cont.total_bytes) {
+ /* First time, jump over the header, we will write to:
+ * cdb_cont.buff + cdb_cont.total_bytes
+ */
+ or->cdb_cont.total_bytes =
+ sizeof(struct osd_continuation_segment_header);
+ }
+
+ ret = _alloc_cdb_cont(or, or->cdb_cont.total_bytes + oscd_size);
+ if (unlikely(ret))
+ return ret;
+
+ oscd = or->cdb_cont.buff + or->cdb_cont.total_bytes;
+ oscd->hdr.type = cpu_to_be16(SCATTER_GATHER_LIST);
+ oscd->hdr.pad_length = 0;
+ oscd->hdr.length = cpu_to_be32(oscd_size - sizeof(*oscd));
+
+ *len = 0;
+ /* copy the sg entries and convert to network byte order */
+ for (i = 0; i < numentries; i++) {
+ oscd->entries[i].offset = cpu_to_be64(sglist[i].offset);
+ oscd->entries[i].len = cpu_to_be64(sglist[i].len);
+ *len += sglist[i].len;
+ }
+
+ or->cdb_cont.total_bytes += oscd_size;
+ OSD_DEBUG("total_bytes=%d oscd_size=%d numentries=%d\n",
+ or->cdb_cont.total_bytes, oscd_size, numentries);
+ return 0;
+}
+
+static int _osd_req_finalize_cdb_cont(struct osd_request *or, const u8 *cap_key)
+{
+ struct request_queue *req_q = osd_request_queue(or->osd_dev);
+ struct bio *bio;
+ struct osd_cdb_head *cdbh = osd_cdb_head(&or->cdb);
+ struct osd_continuation_segment_header *cont_seg_hdr;
+
+ if (!or->cdb_cont.total_bytes)
+ return 0;
+
+ cont_seg_hdr = or->cdb_cont.buff;
+ cont_seg_hdr->format = CDB_CONTINUATION_FORMAT_V2;
+ cont_seg_hdr->service_action = cdbh->varlen_cdb.service_action;
+
+ /* create a bio for continuation segment */
+ bio = bio_map_kern(req_q, or->cdb_cont.buff, or->cdb_cont.total_bytes,
+ GFP_KERNEL);
+ if (unlikely(!bio))
+ return -ENOMEM;
+
+ bio->bi_rw |= REQ_WRITE;
+
+ /* integrity check the continuation before the bio is linked
+ * with the other data segments since the continuation
+ * integrity is separate from the other data segments.
+ */
+ osd_sec_sign_data(cont_seg_hdr->integrity_check, bio, cap_key);
+
+ cdbh->v2.cdb_continuation_length = cpu_to_be32(or->cdb_cont.total_bytes);
+
+ /* we can't use _req_append_segment, because we need to link in the
+ * continuation bio to the head of the bio list - the
+ * continuation segment (if it exists) is always the first segment in
+ * the out data buffer.
+ */
+ bio->bi_next = or->out.bio;
+ or->out.bio = bio;
+ or->out.total_bytes += or->cdb_cont.total_bytes;
+
+ return 0;
+}
+
+/* osd_req_write_sg: Takes a @bio that points to the data out buffer and an
+ * @sglist that has the scatter gather entries. Scatter-gather enables a write
+ * of multiple none-contiguous areas of an object, in a single call. The extents
+ * may overlap and/or be in any order. The only constrain is that:
+ * total_bytes(sglist) >= total_bytes(bio)
+ */
+int osd_req_write_sg(struct osd_request *or,
+ const struct osd_obj_id *obj, struct bio *bio,
+ const struct osd_sg_entry *sglist, unsigned numentries)
+{
+ u64 len;
+ int ret = _add_sg_continuation_descriptor(or, sglist, numentries, &len);
+
+ if (ret)
+ return ret;
+ osd_req_write(or, obj, 0, bio, len);
+
+ return 0;
+}
+EXPORT_SYMBOL(osd_req_write_sg);
+
+/* osd_req_read_sg: Read multiple extents of an object into @bio
+ * See osd_req_write_sg
+ */
+int osd_req_read_sg(struct osd_request *or,
+ const struct osd_obj_id *obj, struct bio *bio,
+ const struct osd_sg_entry *sglist, unsigned numentries)
+{
+ u64 len;
+ int ret = _add_sg_continuation_descriptor(or, sglist, numentries, &len);
+
+ if (ret)
+ return ret;
+ osd_req_read(or, obj, 0, bio, len);
+
+ return 0;
+}
+EXPORT_SYMBOL(osd_req_read_sg);
+
+/* SG-list write/read Kern API
+ *
+ * osd_req_{write,read}_sg_kern takes an array of @buff pointers and an array
+ * of sg_entries. @numentries indicates how many pointers and sg_entries there
+ * are. By requiring an array of buff pointers. This allows a caller to do a
+ * single write/read and scatter into multiple buffers.
+ * NOTE: Each buffer + len should not cross a page boundary.
+ */
+static struct bio *_create_sg_bios(struct osd_request *or,
+ void **buff, const struct osd_sg_entry *sglist, unsigned numentries)
+{
+ struct request_queue *q = osd_request_queue(or->osd_dev);
+ struct bio *bio;
+ unsigned i;
+
+ bio = bio_kmalloc(GFP_KERNEL, numentries);
+ if (unlikely(!bio)) {
+ OSD_DEBUG("Faild to allocate BIO size=%u\n", numentries);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ for (i = 0; i < numentries; i++) {
+ unsigned offset = offset_in_page(buff[i]);
+ struct page *page = virt_to_page(buff[i]);
+ unsigned len = sglist[i].len;
+ unsigned added_len;
+
+ BUG_ON(offset + len > PAGE_SIZE);
+ added_len = bio_add_pc_page(q, bio, page, len, offset);
+ if (unlikely(len != added_len)) {
+ OSD_DEBUG("bio_add_pc_page len(%d) != added_len(%d)\n",
+ len, added_len);
+ bio_put(bio);
+ return ERR_PTR(-ENOMEM);
+ }
+ }
+
+ return bio;
+}
+
+int osd_req_write_sg_kern(struct osd_request *or,
+ const struct osd_obj_id *obj, void **buff,
+ const struct osd_sg_entry *sglist, unsigned numentries)
+{
+ struct bio *bio = _create_sg_bios(or, buff, sglist, numentries);
+ if (IS_ERR(bio))
+ return PTR_ERR(bio);
+
+ bio->bi_rw |= REQ_WRITE;
+ osd_req_write_sg(or, obj, bio, sglist, numentries);
+
+ return 0;
+}
+EXPORT_SYMBOL(osd_req_write_sg_kern);
+
+int osd_req_read_sg_kern(struct osd_request *or,
+ const struct osd_obj_id *obj, void **buff,
+ const struct osd_sg_entry *sglist, unsigned numentries)
+{
+ struct bio *bio = _create_sg_bios(or, buff, sglist, numentries);
+ if (IS_ERR(bio))
+ return PTR_ERR(bio);
+
+ osd_req_read_sg(or, obj, bio, sglist, numentries);
+
+ return 0;
+}
+EXPORT_SYMBOL(osd_req_read_sg_kern);
+
+
+
void osd_req_get_attributes(struct osd_request *or,
const struct osd_obj_id *obj)
{
or->get_attr.buff = attar_page;
or->get_attr.total_bytes = max_page_len;
- or->set_attr.buff = set_one_attr->val_ptr;
- or->set_attr.total_bytes = set_one_attr->len;
-
cdbh->attrs_page.get_attr_page = cpu_to_be32(page_id);
cdbh->attrs_page.get_attr_alloc_length = cpu_to_be32(max_page_len);
- /* ocdb->attrs_page.get_attr_offset; */
+
+ if (!set_one_attr || !set_one_attr->attr_page)
+ return 0; /* The set is optional */
+
+ or->set_attr.buff = set_one_attr->val_ptr;
+ or->set_attr.total_bytes = set_one_attr->len;
cdbh->attrs_page.set_attr_page = cpu_to_be32(set_one_attr->attr_page);
cdbh->attrs_page.set_attr_id = cpu_to_be32(set_one_attr->attr_id);
cdbh->attrs_page.set_attr_length = cpu_to_be32(set_one_attr->len);
- /* ocdb->attrs_page.set_attr_offset; */
return 0;
}
EXPORT_SYMBOL(osd_req_add_get_attr_page);
if (ret)
return ret;
+ if (or->set_attr.total_bytes == 0)
+ return 0;
+
/* set one value */
cdbh->attrs_page.set_attr_offset =
osd_req_encode_offset(or, or->out.total_bytes, &out_padding);
- ret = _req_append_segment(or, out_padding, &or->enc_get_attr, NULL,
+ ret = _req_append_segment(or, out_padding, &or->set_attr, NULL,
&or->out);
return ret;
}
}
static int _osd_req_finalize_data_integrity(struct osd_request *or,
- bool has_in, bool has_out, u64 out_data_bytes, const u8 *cap_key)
+ bool has_in, bool has_out, struct bio *out_data_bio, u64 out_data_bytes,
+ const u8 *cap_key)
{
struct osd_security_parameters *sec_parms = _osd_req_sec_params(or);
int ret;
or->out.last_seg = NULL;
/* they are now all chained to request sign them all together */
- osd_sec_sign_data(&or->out_data_integ, or->out.req->bio,
+ osd_sec_sign_data(&or->out_data_integ, out_data_bio,
cap_key);
}
{
struct osd_cdb_head *cdbh = osd_cdb_head(&or->cdb);
bool has_in, has_out;
+ /* Save for data_integrity without the cdb_continuation */
+ struct bio *out_data_bio = or->out.bio;
u64 out_data_bytes = or->out.total_bytes;
int ret;
osd_set_caps(&or->cdb, cap);
has_in = or->in.bio || or->get_attr.total_bytes;
- has_out = or->out.bio || or->set_attr.total_bytes ||
- or->enc_get_attr.total_bytes;
+ has_out = or->out.bio || or->cdb_cont.total_bytes ||
+ or->set_attr.total_bytes || or->enc_get_attr.total_bytes;
+ ret = _osd_req_finalize_cdb_cont(or, cap_key);
+ if (ret) {
+ OSD_DEBUG("_osd_req_finalize_cdb_cont failed\n");
+ return ret;
+ }
ret = _init_blk_request(or, has_in, has_out);
if (ret) {
OSD_DEBUG("_init_blk_request failed\n");
}
ret = _osd_req_finalize_data_integrity(or, has_in, has_out,
- out_data_bytes, cap_key);
+ out_data_bio, out_data_bytes,
+ cap_key);
if (ret)
return ret;
cfg_entry = &ccn_hcam->cfg_entry;
fw_version = be16_to_cpu(pinstance->inq_data->fw_version);
- pmcraid_info
- ("CCN(%x): %x type: %x lost: %x flags: %x res: %x:%x:%x:%x\n",
+ pmcraid_info("CCN(%x): %x timestamp: %llx type: %x lost: %x flags: %x \
+ res: %x:%x:%x:%x\n",
pinstance->ccn.hcam->ilid,
pinstance->ccn.hcam->op_code,
+ ((pinstance->ccn.hcam->timestamp1) |
+ ((pinstance->ccn.hcam->timestamp2 & 0xffffffffLL) << 32)),
pinstance->ccn.hcam->notification_type,
pinstance->ccn.hcam->notification_lost,
pinstance->ccn.hcam->flags,
* none
*/
static void pmcraid_initiate_reset(struct pmcraid_instance *);
+static void pmcraid_set_timestamp(struct pmcraid_cmd *cmd);
static void pmcraid_process_ldn(struct pmcraid_cmd *cmd)
{
lock_flags);
return;
}
+ if (fd_ioasc == PMCRAID_IOASC_TIME_STAMP_OUT_OF_SYNC) {
+ pinstance->timestamp_error = 1;
+ pmcraid_set_timestamp(cmd);
+ }
} else {
dev_info(&pinstance->pdev->dev,
"Host RCB(LDN) failed with IOASC: 0x%08X\n", ioasc);
sg_size = buflen;
for (i = 0; i < num_elem; i++) {
- page = alloc_pages(GFP_KERNEL|GFP_DMA, order);
+ page = alloc_pages(GFP_KERNEL|GFP_DMA|__GFP_ZERO, order);
if (!page) {
for (j = i - 1; j >= 0; j--)
__free_pages(sg_page(&scatterlist[j]), order);
unsigned long request_buffer;
unsigned long request_offset;
unsigned long lock_flags;
+ void *ioasa;
u32 ioasc;
int request_size;
int buffer_size;
rc = __copy_from_user(buffer,
(struct pmcraid_passthrough_ioctl_buffer *) arg,
sizeof(struct pmcraid_passthrough_ioctl_buffer));
+
+ ioasa =
+ (void *)(arg +
+ offsetof(struct pmcraid_passthrough_ioctl_buffer, ioasa));
+
if (rc) {
pmcraid_err("ioctl: can't copy passthrough buffer\n");
rc = -EFAULT;
}
out_handle_response:
- /* If the command failed for any reason, copy entire IOASA buffer and
- * return IOCTL success. If copying IOASA to user-buffer fails, return
+ /* copy entire IOASA buffer and return IOCTL success.
+ * If copying IOASA to user-buffer fails, return
* EFAULT
*/
- if (PMCRAID_IOASC_SENSE_KEY(le32_to_cpu(cmd->ioa_cb->ioasa.ioasc))) {
- void *ioasa =
- (void *)(arg +
- offsetof(struct pmcraid_passthrough_ioctl_buffer, ioasa));
-
- pmcraid_info("command failed with %x\n",
- le32_to_cpu(cmd->ioa_cb->ioasa.ioasc));
- if (copy_to_user(ioasa, &cmd->ioa_cb->ioasa,
- sizeof(struct pmcraid_ioasa))) {
- pmcraid_err("failed to copy ioasa buffer to user\n");
- rc = -EFAULT;
- }
+ if (copy_to_user(ioasa, &cmd->ioa_cb->ioasa,
+ sizeof(struct pmcraid_ioasa))) {
+ pmcraid_err("failed to copy ioasa buffer to user\n");
+ rc = -EFAULT;
}
/* If the data transfer was from device, copy the data onto user
pinstance->inq_data = NULL;
pinstance->inq_data_baddr = 0;
}
+
+ if (pinstance->timestamp_data != NULL) {
+ pci_free_consistent(pinstance->pdev,
+ sizeof(struct pmcraid_timestamp_data),
+ pinstance->timestamp_data,
+ pinstance->timestamp_data_baddr);
+
+ pinstance->timestamp_data = NULL;
+ pinstance->timestamp_data_baddr = 0;
+ }
}
/**
return -ENOMEM;
}
+ /* allocate DMAable memory for set timestamp data buffer */
+ pinstance->timestamp_data = pci_alloc_consistent(
+ pinstance->pdev,
+ sizeof(struct pmcraid_timestamp_data),
+ &pinstance->timestamp_data_baddr);
+
+ if (pinstance->timestamp_data == NULL) {
+ pmcraid_err("couldn't allocate DMA memory for \
+ set time_stamp \n");
+ pmcraid_release_buffers(pinstance);
+ return -ENOMEM;
+ }
+
+
/* Initialize all the command blocks and add them to free pool. No
* need to lock (free_pool_lock) as this is done in initialization
* itself
return;
}
+/**
+ * pmcraid_set_timestamp - set the timestamp to IOAFP
+ *
+ * @cmd: pointer to pmcraid_cmd structure
+ *
+ * Return Value
+ * 0 for success or non-zero for failure cases
+ */
+static void pmcraid_set_timestamp(struct pmcraid_cmd *cmd)
+{
+ struct pmcraid_instance *pinstance = cmd->drv_inst;
+ struct pmcraid_ioarcb *ioarcb = &cmd->ioa_cb->ioarcb;
+ __be32 time_stamp_len = cpu_to_be32(PMCRAID_TIMESTAMP_LEN);
+ struct pmcraid_ioadl_desc *ioadl = ioarcb->add_data.u.ioadl;
+
+ struct timeval tv;
+ __le64 timestamp;
+
+ do_gettimeofday(&tv);
+ timestamp = tv.tv_sec * 1000;
+
+ pinstance->timestamp_data->timestamp[0] = (__u8)(timestamp);
+ pinstance->timestamp_data->timestamp[1] = (__u8)((timestamp) >> 8);
+ pinstance->timestamp_data->timestamp[2] = (__u8)((timestamp) >> 16);
+ pinstance->timestamp_data->timestamp[3] = (__u8)((timestamp) >> 24);
+ pinstance->timestamp_data->timestamp[4] = (__u8)((timestamp) >> 32);
+ pinstance->timestamp_data->timestamp[5] = (__u8)((timestamp) >> 40);
+
+ pmcraid_reinit_cmdblk(cmd);
+ ioarcb->request_type = REQ_TYPE_SCSI;
+ ioarcb->resource_handle = cpu_to_le32(PMCRAID_IOA_RES_HANDLE);
+ ioarcb->cdb[0] = PMCRAID_SCSI_SET_TIMESTAMP;
+ ioarcb->cdb[1] = PMCRAID_SCSI_SERVICE_ACTION;
+ memcpy(&(ioarcb->cdb[6]), &time_stamp_len, sizeof(time_stamp_len));
+
+ ioarcb->ioadl_bus_addr = cpu_to_le64((cmd->ioa_cb_bus_addr) +
+ offsetof(struct pmcraid_ioarcb,
+ add_data.u.ioadl[0]));
+ ioarcb->ioadl_length = cpu_to_le32(sizeof(struct pmcraid_ioadl_desc));
+ ioarcb->ioarcb_bus_addr &= ~(0x1FULL);
+
+ ioarcb->request_flags0 |= NO_LINK_DESCS;
+ ioarcb->request_flags0 |= TRANSFER_DIR_WRITE;
+ ioarcb->data_transfer_length =
+ cpu_to_le32(sizeof(struct pmcraid_timestamp_data));
+ ioadl = &(ioarcb->add_data.u.ioadl[0]);
+ ioadl->flags = IOADL_FLAGS_LAST_DESC;
+ ioadl->address = cpu_to_le64(pinstance->timestamp_data_baddr);
+ ioadl->data_len = cpu_to_le32(sizeof(struct pmcraid_timestamp_data));
+
+ if (!pinstance->timestamp_error) {
+ pinstance->timestamp_error = 0;
+ pmcraid_send_cmd(cmd, pmcraid_set_supported_devs,
+ PMCRAID_INTERNAL_TIMEOUT, pmcraid_timeout_handler);
+ } else {
+ pmcraid_send_cmd(cmd, pmcraid_return_cmd,
+ PMCRAID_INTERNAL_TIMEOUT, pmcraid_timeout_handler);
+ return;
+ }
+}
+
+
/**
* pmcraid_init_res_table - Initialize the resource table
* @cmd: pointer to pmcraid command struct
/* release the resource list lock */
spin_unlock_irqrestore(&pinstance->resource_lock, lock_flags);
- pmcraid_set_supported_devs(cmd);
+ pmcraid_set_timestamp(cmd);
}
/**
static void __exit pmcraid_exit(void)
{
pmcraid_netlink_release();
- class_destroy(pmcraid_class);
unregister_chrdev_region(MKDEV(pmcraid_major, 0),
PMCRAID_MAX_ADAPTERS);
pci_unregister_driver(&pmcraid_driver);
+ class_destroy(pmcraid_class);
}
module_init(pmcraid_init);
*/
#define PMCRAID_DRIVER_NAME "PMC MaxRAID"
#define PMCRAID_DEVFILE "pmcsas"
-#define PMCRAID_DRIVER_VERSION "2.0.2"
+#define PMCRAID_DRIVER_VERSION "2.0.3"
#define PMCRAID_DRIVER_DATE __DATE__
#define PMCRAID_FW_VERSION_1 0x002
#define PMCRAID_IOASC_IR_INVALID_RESOURCE_HANDLE 0x05250000
#define PMCRAID_IOASC_AC_TERMINATED_BY_HOST 0x0B5A0000
#define PMCRAID_IOASC_UA_BUS_WAS_RESET 0x06290000
+#define PMCRAID_IOASC_TIME_STAMP_OUT_OF_SYNC 0x06908B00
#define PMCRAID_IOASC_UA_BUS_WAS_RESET_BY_OTHER 0x06298000
/* Driver defined IOASCs */
__u8 reserved3[16];
};
+#define PMCRAID_TIMESTAMP_LEN 12
+#define PMCRAID_REQ_TM_STR_LEN 6
+#define PMCRAID_SCSI_SET_TIMESTAMP 0xA4
+#define PMCRAID_SCSI_SERVICE_ACTION 0x0F
+
+struct pmcraid_timestamp_data {
+ __u8 reserved1[4];
+ __u8 timestamp[PMCRAID_REQ_TM_STR_LEN]; /* current time value */
+ __u8 reserved2[2];
+};
+
/* pmcraid_cmd - LLD representation of SCSI command */
struct pmcraid_cmd {
struct pmcraid_control_block *ioa_cb;
dma_addr_t ioa_cb_bus_addr;
dma_addr_t dma_handle;
- u8 *sense_buffer;
/* pointer to mid layer structure of SCSI commands */
struct scsi_cmnd *scsi_cmd;
struct pmcraid_inquiry_data *inq_data;
dma_addr_t inq_data_baddr;
+ struct pmcraid_timestamp_data *timestamp_data;
+ dma_addr_t timestamp_data_baddr;
+
/* size of configuration table entry, varies based on the firmware */
u32 config_table_entry_size;
#define SHUTDOWN_NONE 0x0
#define SHUTDOWN_NORMAL 0x1
#define SHUTDOWN_ABBREV 0x2
+ u32 timestamp_error:1; /* indicate set timestamp for out of sync */
};
#define PMCRAID_PASSTHROUGH_IOCTL 'F'
#define DRV_IOCTL(n, size) \
- _IOC(_IOC_READ|_IOC_WRITE, PMCRAID_DRIVER_IOCTL, (n), (size))
+ _IOC(_IOC_READ|_IOC_WRITE, PMCRAID_DRIVER_IOCTL, (n), (size))
#define FMW_IOCTL(n, size) \
- _IOC(_IOC_READ|_IOC_WRITE, PMCRAID_PASSTHROUGH_IOCTL, (n), (size))
+ _IOC(_IOC_READ|_IOC_WRITE, PMCRAID_PASSTHROUGH_IOCTL, (n), (size))
/*
* _ARGSIZE: macro that gives size of the argument type passed to an IOCTL cmd.
if (!fcport)
return;
+ /* Now that the rport has been deleted, set the fcport state to
+ FCS_DEVICE_DEAD */
+ atomic_set(&fcport->state, FCS_DEVICE_DEAD);
+
/*
* Transport has effectively 'deleted' the rport, clear
* all local references.
return rval;
}
+static int
+qla2x00_optrom_setup(struct fc_bsg_job *bsg_job, struct qla_hw_data *ha,
+ uint8_t is_update)
+{
+ uint32_t start = 0;
+ int valid = 0;
+
+ bsg_job->reply->reply_payload_rcv_len = 0;
+
+ if (unlikely(pci_channel_offline(ha->pdev)))
+ return -EINVAL;
+
+ start = bsg_job->request->rqst_data.h_vendor.vendor_cmd[1];
+ if (start > ha->optrom_size)
+ return -EINVAL;
+
+ if (ha->optrom_state != QLA_SWAITING)
+ return -EBUSY;
+
+ ha->optrom_region_start = start;
+
+ if (is_update) {
+ if (ha->optrom_size == OPTROM_SIZE_2300 && start == 0)
+ valid = 1;
+ else if (start == (ha->flt_region_boot * 4) ||
+ start == (ha->flt_region_fw * 4))
+ valid = 1;
+ else if (IS_QLA24XX_TYPE(ha) || IS_QLA25XX(ha) ||
+ IS_QLA8XXX_TYPE(ha))
+ valid = 1;
+ if (!valid) {
+ qla_printk(KERN_WARNING, ha,
+ "Invalid start region 0x%x/0x%x.\n",
+ start, bsg_job->request_payload.payload_len);
+ return -EINVAL;
+ }
+
+ ha->optrom_region_size = start +
+ bsg_job->request_payload.payload_len > ha->optrom_size ?
+ ha->optrom_size - start :
+ bsg_job->request_payload.payload_len;
+ ha->optrom_state = QLA_SWRITING;
+ } else {
+ ha->optrom_region_size = start +
+ bsg_job->reply_payload.payload_len > ha->optrom_size ?
+ ha->optrom_size - start :
+ bsg_job->reply_payload.payload_len;
+ ha->optrom_state = QLA_SREADING;
+ }
+
+ ha->optrom_buffer = vmalloc(ha->optrom_region_size);
+ if (!ha->optrom_buffer) {
+ qla_printk(KERN_WARNING, ha,
+ "Read: Unable to allocate memory for optrom retrieval "
+ "(%x).\n", ha->optrom_region_size);
+
+ ha->optrom_state = QLA_SWAITING;
+ return -ENOMEM;
+ }
+
+ memset(ha->optrom_buffer, 0, ha->optrom_region_size);
+ return 0;
+}
+
+static int
+qla2x00_read_optrom(struct fc_bsg_job *bsg_job)
+{
+ struct Scsi_Host *host = bsg_job->shost;
+ scsi_qla_host_t *vha = shost_priv(host);
+ struct qla_hw_data *ha = vha->hw;
+ int rval = 0;
+
+ rval = qla2x00_optrom_setup(bsg_job, ha, 0);
+ if (rval)
+ return rval;
+
+ ha->isp_ops->read_optrom(vha, ha->optrom_buffer,
+ ha->optrom_region_start, ha->optrom_region_size);
+
+ sg_copy_from_buffer(bsg_job->reply_payload.sg_list,
+ bsg_job->reply_payload.sg_cnt, ha->optrom_buffer,
+ ha->optrom_region_size);
+
+ bsg_job->reply->reply_payload_rcv_len = ha->optrom_region_size;
+ bsg_job->reply->result = DID_OK;
+ vfree(ha->optrom_buffer);
+ ha->optrom_buffer = NULL;
+ ha->optrom_state = QLA_SWAITING;
+ bsg_job->job_done(bsg_job);
+ return rval;
+}
+
+static int
+qla2x00_update_optrom(struct fc_bsg_job *bsg_job)
+{
+ struct Scsi_Host *host = bsg_job->shost;
+ scsi_qla_host_t *vha = shost_priv(host);
+ struct qla_hw_data *ha = vha->hw;
+ int rval = 0;
+
+ rval = qla2x00_optrom_setup(bsg_job, ha, 1);
+ if (rval)
+ return rval;
+
+ sg_copy_to_buffer(bsg_job->request_payload.sg_list,
+ bsg_job->request_payload.sg_cnt, ha->optrom_buffer,
+ ha->optrom_region_size);
+
+ ha->isp_ops->write_optrom(vha, ha->optrom_buffer,
+ ha->optrom_region_start, ha->optrom_region_size);
+
+ bsg_job->reply->result = DID_OK;
+ vfree(ha->optrom_buffer);
+ ha->optrom_buffer = NULL;
+ ha->optrom_state = QLA_SWAITING;
+ bsg_job->job_done(bsg_job);
+ return rval;
+}
+
static int
qla2x00_process_vendor_specific(struct fc_bsg_job *bsg_job)
{
case QL_VND_FCP_PRIO_CFG_CMD:
return qla24xx_proc_fcp_prio_cfg_cmd(bsg_job);
+ case QL_VND_READ_FLASH:
+ return qla2x00_read_optrom(bsg_job);
+
+ case QL_VND_UPDATE_FLASH:
+ return qla2x00_update_optrom(bsg_job);
+
default:
bsg_job->reply->result = (DID_ERROR << 16);
bsg_job->job_done(bsg_job);
#define QL_VND_A84_MGMT_CMD 0x04
#define QL_VND_IIDMA 0x05
#define QL_VND_FCP_PRIO_CFG_CMD 0x06
+#define QL_VND_READ_FLASH 0x07
+#define QL_VND_UPDATE_FLASH 0x08
/* BSG definations for interpreting CommandSent field */
#define INT_DEF_LB_LOOPBACK_CMD 0
atomic_t state;
uint32_t flags;
- int port_login_retry_count;
int login_retry;
- atomic_t port_down_timer;
struct fc_rport *rport, *drport;
u32 supported_classes;
extern int ql2xdbwr;
extern int ql2xdontresethba;
extern int ql2xasynctmfenable;
+extern int ql2xgffidenable;
extern int ql2xenabledif;
extern int ql2xenablehba_err_chk;
extern int ql2xtargetreset;
struct srb_iocb *iocb = ctx->u.iocb_cmd;
struct scsi_qla_host *vha = sp->fcport->vha;
- del_timer_sync(&iocb->timer);
+ del_timer(&iocb->timer);
kfree(iocb);
kfree(ctx);
mempool_free(sp, sp->fcport->vha->hw->srb_mempool);
qla_printk(KERN_WARNING, ha, "Unable to allocate (%d KB) for "
"firmware dump!!!\n", dump_size / 1024);
+ if (ha->fce) {
+ dma_free_coherent(&ha->pdev->dev, FCE_SIZE, ha->fce,
+ ha->fce_dma);
+ ha->fce = NULL;
+ ha->fce_dma = 0;
+ }
+
if (ha->eft) {
dma_free_coherent(&ha->pdev->dev, eft_size, ha->eft,
ha->eft_dma);
qla2x00_init_response_q_entries(rsp);
}
- spin_lock_irqsave(&ha->vport_slock, flags);
+ spin_lock(&ha->vport_slock);
/* Clear RSCN queue. */
list_for_each_entry(vp, &ha->vp_list, list) {
vp->rscn_in_ptr = 0;
vp->rscn_out_ptr = 0;
}
- spin_unlock_irqrestore(&ha->vport_slock, flags);
+ spin_unlock(&ha->vport_slock);
ha->isp_ops->config_rings(vha);
void
qla2x00_update_fcport(scsi_qla_host_t *vha, fc_port_t *fcport)
{
- struct qla_hw_data *ha = vha->hw;
-
fcport->vha = vha;
fcport->login_retry = 0;
- fcport->port_login_retry_count = ha->port_down_retry_count *
- PORT_RETRY_TIME;
- atomic_set(&fcport->port_down_timer, ha->port_down_retry_count *
- PORT_RETRY_TIME);
fcport->flags &= ~(FCF_LOGIN_NEEDED | FCF_ASYNC_SENT);
qla2x00_iidma_fcport(vha, fcport);
-
- atomic_set(&fcport->state, FCS_ONLINE);
-
qla2x00_reg_remote_port(vha, fcport);
+ atomic_set(&fcport->state, FCS_ONLINE);
}
/*
continue;
/* Bypass ports whose FCP-4 type is not FCP_SCSI */
- if (new_fcport->fc4_type != FC4_TYPE_FCP_SCSI &&
- new_fcport->fc4_type != FC4_TYPE_UNKNOWN)
+ if (ql2xgffidenable &&
+ (new_fcport->fc4_type != FC4_TYPE_FCP_SCSI &&
+ new_fcport->fc4_type != FC4_TYPE_UNKNOWN))
continue;
/* Locate matching device in database. */
ha = vha->hw;
DEBUG18(printk(KERN_DEBUG
- "%s(%ld): Executing cmd sp %p, pid=%ld, prot_op=%u.\n", __func__,
- vha->host_no, sp, cmd->serial_number, scsi_get_prot_op(sp->cmd)));
+ "%s(%ld): Executing cmd sp %p, prot_op=%u.\n", __func__,
+ vha->host_no, sp, scsi_get_prot_op(sp->cmd)));
cmd_pkt->vp_index = sp->fcport->vp_idx;
case LSC_SCODE_NPORT_USED:
data[0] = MBS_LOOP_ID_USED;
break;
- case LSC_SCODE_CMD_FAILED:
- if ((iop[1] & 0xff) == 0x05) {
- data[0] = MBS_NOT_LOGGED_IN;
- break;
- }
- /* Fall through. */
default:
data[0] = MBS_COMMAND_ERROR;
break;
rsp->status_srb = sp;
DEBUG5(printk("%s(): Check condition Sense data, scsi(%ld:%d:%d:%d) "
- "cmd=%p pid=%ld\n", __func__, sp->fcport->vha->host_no,
- cp->device->channel, cp->device->id, cp->device->lun, cp,
- cp->serial_number));
+ "cmd=%p\n", __func__, sp->fcport->vha->host_no,
+ cp->device->channel, cp->device->id, cp->device->lun, cp));
if (sense_len)
DEBUG5(qla2x00_dump_buffer(cp->sense_buffer, sense_len));
}
case CS_INCOMPLETE:
case CS_PORT_UNAVAILABLE:
case CS_TIMEOUT:
+ case CS_RESET:
+
/*
* We are going to have the fc class block the rport
* while we try to recover so instruct the mid layer
qla2x00_mark_device_lost(fcport->vha, fcport, 1, 1);
break;
- case CS_RESET:
- cp->result = DID_TRANSPORT_DISRUPTED << 16;
- break;
-
case CS_ABORTED:
cp->result = DID_RESET << 16;
break;
if (logit)
DEBUG2(qla_printk(KERN_INFO, ha,
"scsi(%ld:%d:%d) FCP command status: 0x%x-0x%x (0x%x) "
- "oxid=0x%x ser=0x%lx cdb=%02x%02x%02x len=0x%x "
+ "oxid=0x%x cdb=%02x%02x%02x len=0x%x "
"rsp_info=0x%x resid=0x%x fw_resid=0x%x\n", vha->host_no,
cp->device->id, cp->device->lun, comp_status, scsi_status,
- cp->result, ox_id, cp->serial_number, cp->cmnd[0],
+ cp->result, ox_id, cp->cmnd[0],
cp->cmnd[1], cp->cmnd[2], scsi_bufflen(cp), rsp_info_len,
resid_len, fw_resid_len));
"Enable target reset."
"Default is 1 - use hw defaults.");
+int ql2xgffidenable;
+module_param(ql2xgffidenable, int, S_IRUGO|S_IRUSR);
+MODULE_PARM_DESC(ql2xgffidenable,
+ "Enables GFF_ID checks of port type. "
+ "Default is 0 - Do not use GFF_ID information.");
int ql2xasynctmfenable;
module_param(ql2xasynctmfenable, int, S_IRUGO|S_IRUSR);
static int qla2x00_mem_alloc(struct qla_hw_data *, uint16_t, uint16_t,
struct req_que **, struct rsp_que **);
+static void qla2x00_free_fw_dump(struct qla_hw_data *);
static void qla2x00_mem_free(struct qla_hw_data *);
static void qla2x00_sp_free_dma(srb_t *);
srb_t *sp;
int rval;
+ spin_unlock_irq(vha->host->host_lock);
if (ha->flags.eeh_busy) {
if (ha->flags.pci_channel_io_perm_failure)
cmd->result = DID_NO_CONNECT << 16;
goto qc24_fail_command;
}
- /* Close window on fcport/rport state-transitioning. */
- if (fcport->drport)
- goto qc24_target_busy;
-
if (!vha->flags.difdix_supported &&
scsi_get_prot_op(cmd) != SCSI_PROT_NORMAL) {
DEBUG2(qla_printk(KERN_ERR, ha,
}
if (atomic_read(&fcport->state) != FCS_ONLINE) {
if (atomic_read(&fcport->state) == FCS_DEVICE_DEAD ||
- atomic_read(&base_vha->loop_state) == LOOP_DEAD) {
+ atomic_read(&fcport->state) == FCS_DEVICE_LOST ||
+ atomic_read(&base_vha->loop_state) == LOOP_DEAD) {
cmd->result = DID_NO_CONNECT << 16;
goto qc24_fail_command;
}
goto qc24_target_busy;
}
- spin_unlock_irq(vha->host->host_lock);
-
sp = qla2x00_get_new_sp(base_vha, fcport, cmd, done);
if (!sp)
goto qc24_host_busy_lock;
return SCSI_MLQUEUE_HOST_BUSY;
qc24_target_busy:
+ spin_lock_irq(vha->host->host_lock);
return SCSI_MLQUEUE_TARGET_BUSY;
qc24_fail_command:
+ spin_lock_irq(vha->host->host_lock);
done(cmd);
return 0;
{
scsi_qla_host_t *vha = shost_priv(cmd->device->host);
srb_t *sp;
- int ret, i;
+ int ret;
unsigned int id, lun;
- unsigned long serial;
unsigned long flags;
int wait = 0;
struct qla_hw_data *ha = vha->hw;
- struct req_que *req = vha->req;
- srb_t *spt;
- int got_ref = 0;
fc_block_scsi_eh(cmd);
if (!CMD_SP(cmd))
return SUCCESS;
- ret = SUCCESS;
-
id = cmd->device->id;
lun = cmd->device->lun;
- serial = cmd->serial_number;
- spt = (srb_t *) CMD_SP(cmd);
- if (!spt)
- return SUCCESS;
- /* Check active list for command command. */
spin_lock_irqsave(&ha->hardware_lock, flags);
- for (i = 1; i < MAX_OUTSTANDING_COMMANDS; i++) {
- sp = req->outstanding_cmds[i];
-
- if (sp == NULL)
- continue;
- if ((sp->ctx) && !(sp->flags & SRB_FCP_CMND_DMA_VALID) &&
- !IS_PROT_IO(sp))
- continue;
- if (sp->cmd != cmd)
- continue;
+ sp = (srb_t *) CMD_SP(cmd);
+ if (!sp) {
+ spin_unlock_irqrestore(&ha->hardware_lock, flags);
+ return SUCCESS;
+ }
- DEBUG2(printk("%s(%ld): aborting sp %p from RISC."
- " pid=%ld.\n", __func__, vha->host_no, sp, serial));
+ DEBUG2(printk("%s(%ld): aborting sp %p from RISC.",
+ __func__, vha->host_no, sp));
- /* Get a reference to the sp and drop the lock.*/
- sp_get(sp);
- got_ref++;
+ /* Get a reference to the sp and drop the lock.*/
+ sp_get(sp);
- spin_unlock_irqrestore(&ha->hardware_lock, flags);
- if (ha->isp_ops->abort_command(sp)) {
- DEBUG2(printk("%s(%ld): abort_command "
- "mbx failed.\n", __func__, vha->host_no));
- ret = FAILED;
- } else {
- DEBUG3(printk("%s(%ld): abort_command "
- "mbx success.\n", __func__, vha->host_no));
- wait = 1;
- }
- spin_lock_irqsave(&ha->hardware_lock, flags);
- break;
- }
spin_unlock_irqrestore(&ha->hardware_lock, flags);
+ if (ha->isp_ops->abort_command(sp)) {
+ DEBUG2(printk("%s(%ld): abort_command "
+ "mbx failed.\n", __func__, vha->host_no));
+ ret = FAILED;
+ } else {
+ DEBUG3(printk("%s(%ld): abort_command "
+ "mbx success.\n", __func__, vha->host_no));
+ wait = 1;
+ }
+ qla2x00_sp_compl(ha, sp);
/* Wait for the command to be returned. */
if (wait) {
if (qla2x00_eh_wait_on_command(cmd) != QLA_SUCCESS) {
qla_printk(KERN_ERR, ha,
- "scsi(%ld:%d:%d): Abort handler timed out -- %lx "
- "%x.\n", vha->host_no, id, lun, serial, ret);
+ "scsi(%ld:%d:%d): Abort handler timed out -- %x.\n",
+ vha->host_no, id, lun, ret);
ret = FAILED;
}
}
- if (got_ref)
- qla2x00_sp_compl(ha, sp);
-
qla_printk(KERN_INFO, ha,
- "scsi(%ld:%d:%d): Abort command issued -- %d %lx %x.\n",
- vha->host_no, id, lun, wait, serial, ret);
+ "scsi(%ld:%d:%d): Abort command issued -- %d %x.\n",
+ vha->host_no, id, lun, wait, ret);
return ret;
}
fc_port_t *fcport = (struct fc_port *) cmd->device->hostdata;
int ret = FAILED;
unsigned int id, lun;
- unsigned long serial;
fc_block_scsi_eh(cmd);
id = cmd->device->id;
lun = cmd->device->lun;
- serial = cmd->serial_number;
if (!fcport)
return ret;
struct qla_hw_data *ha = vha->hw;
int ret = FAILED;
unsigned int id, lun;
- unsigned long serial;
scsi_qla_host_t *base_vha = pci_get_drvdata(ha->pdev);
fc_block_scsi_eh(cmd);
id = cmd->device->id;
lun = cmd->device->lun;
- serial = cmd->serial_number;
if (!fcport)
return ret;
ha->bars = bars;
ha->mem_only = mem_only;
spin_lock_init(&ha->hardware_lock);
+ spin_lock_init(&ha->vport_slock);
/* Set ISP-type information. */
qla2x00_set_isp_flags(ha);
return ret;
}
+static void
+qla2x00_shutdown(struct pci_dev *pdev)
+{
+ scsi_qla_host_t *vha;
+ struct qla_hw_data *ha;
+
+ vha = pci_get_drvdata(pdev);
+ ha = vha->hw;
+
+ /* Turn-off FCE trace */
+ if (ha->flags.fce_enabled) {
+ qla2x00_disable_fce_trace(vha, NULL, NULL);
+ ha->flags.fce_enabled = 0;
+ }
+
+ /* Turn-off EFT trace */
+ if (ha->eft)
+ qla2x00_disable_eft_trace(vha);
+
+ /* Stop currently executing firmware. */
+ qla2x00_try_to_stop_firmware(vha);
+
+ /* Turn adapter off line */
+ vha->flags.online = 0;
+
+ /* turn-off interrupts on the card */
+ if (ha->interrupts_on) {
+ vha->flags.init_done = 0;
+ ha->isp_ops->disable_intrs(ha);
+ }
+
+ qla2x00_free_irqs(vha);
+
+ qla2x00_free_fw_dump(ha);
+}
+
static void
qla2x00_remove_one(struct pci_dev *pdev)
{
if (atomic_read(&fcport->state) == FCS_DEVICE_DEAD)
continue;
if (atomic_read(&fcport->state) == FCS_ONLINE) {
+ atomic_set(&fcport->state, FCS_DEVICE_LOST);
if (defer)
qla2x00_schedule_rport_del(vha, fcport, defer);
else if (vha->vp_idx == fcport->vp_idx)
qla2x00_schedule_rport_del(vha, fcport, defer);
}
- atomic_set(&fcport->state, FCS_DEVICE_LOST);
}
}
}
/*
-* qla2x00_mem_free
-* Frees all adapter allocated memory.
+* qla2x00_free_fw_dump
+* Frees fw dump stuff.
*
* Input:
-* ha = adapter block pointer.
+* ha = adapter block pointer.
*/
static void
-qla2x00_mem_free(struct qla_hw_data *ha)
+qla2x00_free_fw_dump(struct qla_hw_data *ha)
{
- if (ha->srb_mempool)
- mempool_destroy(ha->srb_mempool);
-
if (ha->fce)
dma_free_coherent(&ha->pdev->dev, FCE_SIZE, ha->fce,
- ha->fce_dma);
+ ha->fce_dma);
if (ha->fw_dump) {
if (ha->eft)
dma_free_coherent(&ha->pdev->dev,
- ntohl(ha->fw_dump->eft_size), ha->eft, ha->eft_dma);
+ ntohl(ha->fw_dump->eft_size), ha->eft, ha->eft_dma);
vfree(ha->fw_dump);
}
+ ha->fce = NULL;
+ ha->fce_dma = 0;
+ ha->eft = NULL;
+ ha->eft_dma = 0;
+ ha->fw_dump = NULL;
+ ha->fw_dumped = 0;
+ ha->fw_dump_reading = 0;
+}
+
+/*
+* qla2x00_mem_free
+* Frees all adapter allocated memory.
+*
+* Input:
+* ha = adapter block pointer.
+*/
+static void
+qla2x00_mem_free(struct qla_hw_data *ha)
+{
+ qla2x00_free_fw_dump(ha);
+
+ if (ha->srb_mempool)
+ mempool_destroy(ha->srb_mempool);
if (ha->dcbx_tlv)
dma_free_coherent(&ha->pdev->dev, DCBX_TLV_DATA_SIZE,
ha->srb_mempool = NULL;
ha->ctx_mempool = NULL;
- ha->eft = NULL;
- ha->eft_dma = 0;
ha->sns_cmd = NULL;
ha->sns_cmd_dma = 0;
ha->ct_sns = NULL;
ha->gid_list = NULL;
ha->gid_list_dma = 0;
-
- ha->fw_dump = NULL;
- ha->fw_dumped = 0;
- ha->fw_dump_reading = 0;
}
struct scsi_qla_host *qla2x00_create_host(struct scsi_host_template *sht,
qla2x00_timer(scsi_qla_host_t *vha)
{
unsigned long cpu_flags = 0;
- fc_port_t *fcport;
int start_dpc = 0;
int index;
srb_t *sp;
- int t;
uint16_t w;
struct qla_hw_data *ha = vha->hw;
struct req_que *req;
/* Hardware read to raise pending EEH errors during mailbox waits. */
if (!pci_channel_offline(ha->pdev))
pci_read_config_word(ha->pdev, PCI_VENDOR_ID, &w);
- /*
- * Ports - Port down timer.
- *
- * Whenever, a port is in the LOST state we start decrementing its port
- * down timer every second until it reaches zero. Once it reaches zero
- * the port it marked DEAD.
- */
- t = 0;
- list_for_each_entry(fcport, &vha->vp_fcports, list) {
- if (fcport->port_type != FCT_TARGET)
- continue;
-
- if (atomic_read(&fcport->state) == FCS_DEVICE_LOST) {
-
- if (atomic_read(&fcport->port_down_timer) == 0)
- continue;
-
- if (atomic_dec_and_test(&fcport->port_down_timer) != 0)
- atomic_set(&fcport->state, FCS_DEVICE_DEAD);
-
- DEBUG(printk("scsi(%ld): fcport-%d - port retry count: "
- "%d remaining\n",
- vha->host_no,
- t, atomic_read(&fcport->port_down_timer)));
- }
- t++;
- } /* End of for fcport */
-
/* Loop down handler. */
if (atomic_read(&vha->loop_down_timer) > 0 &&
.id_table = qla2xxx_pci_tbl,
.probe = qla2x00_probe_one,
.remove = qla2x00_remove_one,
+ .shutdown = qla2x00_shutdown,
.err_handler = &qla2xxx_err_handler,
};
printk(KERN_INFO "\n");
}
+void qla4xxx_dump_registers(struct scsi_qla_host *ha)
+{
+ uint8_t i;
+
+ if (is_qla8022(ha)) {
+ for (i = 1; i < MBOX_REG_COUNT; i++)
+ printk(KERN_INFO "mailbox[%d] = 0x%08X\n",
+ i, readl(&ha->qla4_8xxx_reg->mailbox_in[i]));
+ return;
+ }
+
+ for (i = 0; i < MBOX_REG_COUNT; i++) {
+ printk(KERN_INFO "0x%02X mailbox[%d] = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, mailbox[i]), i,
+ readw(&ha->reg->mailbox[i]));
+ }
+
+ printk(KERN_INFO "0x%02X flash_address = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, flash_address),
+ readw(&ha->reg->flash_address));
+ printk(KERN_INFO "0x%02X flash_data = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, flash_data),
+ readw(&ha->reg->flash_data));
+ printk(KERN_INFO "0x%02X ctrl_status = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, ctrl_status),
+ readw(&ha->reg->ctrl_status));
+
+ if (is_qla4010(ha)) {
+ printk(KERN_INFO "0x%02X nvram = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u1.isp4010.nvram),
+ readw(&ha->reg->u1.isp4010.nvram));
+ } else if (is_qla4022(ha) | is_qla4032(ha)) {
+ printk(KERN_INFO "0x%02X intr_mask = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u1.isp4022.intr_mask),
+ readw(&ha->reg->u1.isp4022.intr_mask));
+ printk(KERN_INFO "0x%02X nvram = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u1.isp4022.nvram),
+ readw(&ha->reg->u1.isp4022.nvram));
+ printk(KERN_INFO "0x%02X semaphore = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u1.isp4022.semaphore),
+ readw(&ha->reg->u1.isp4022.semaphore));
+ }
+ printk(KERN_INFO "0x%02X req_q_in = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, req_q_in),
+ readw(&ha->reg->req_q_in));
+ printk(KERN_INFO "0x%02X rsp_q_out = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, rsp_q_out),
+ readw(&ha->reg->rsp_q_out));
+
+ if (is_qla4010(ha)) {
+ printk(KERN_INFO "0x%02X ext_hw_conf = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u2.isp4010.ext_hw_conf),
+ readw(&ha->reg->u2.isp4010.ext_hw_conf));
+ printk(KERN_INFO "0x%02X port_ctrl = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u2.isp4010.port_ctrl),
+ readw(&ha->reg->u2.isp4010.port_ctrl));
+ printk(KERN_INFO "0x%02X port_status = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u2.isp4010.port_status),
+ readw(&ha->reg->u2.isp4010.port_status));
+ printk(KERN_INFO "0x%02X req_q_out = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u2.isp4010.req_q_out),
+ readw(&ha->reg->u2.isp4010.req_q_out));
+ printk(KERN_INFO "0x%02X gp_out = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u2.isp4010.gp_out),
+ readw(&ha->reg->u2.isp4010.gp_out));
+ printk(KERN_INFO "0x%02X gp_in = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u2.isp4010.gp_in),
+ readw(&ha->reg->u2.isp4010.gp_in));
+ printk(KERN_INFO "0x%02X port_err_status = 0x%08X\n", (uint8_t)
+ offsetof(struct isp_reg, u2.isp4010.port_err_status),
+ readw(&ha->reg->u2.isp4010.port_err_status));
+ } else if (is_qla4022(ha) | is_qla4032(ha)) {
+ printk(KERN_INFO "Page 0 Registers:\n");
+ printk(KERN_INFO "0x%02X ext_hw_conf = 0x%08X\n", (uint8_t)
+ offsetof(struct isp_reg, u2.isp4022.p0.ext_hw_conf),
+ readw(&ha->reg->u2.isp4022.p0.ext_hw_conf));
+ printk(KERN_INFO "0x%02X port_ctrl = 0x%08X\n", (uint8_t)
+ offsetof(struct isp_reg, u2.isp4022.p0.port_ctrl),
+ readw(&ha->reg->u2.isp4022.p0.port_ctrl));
+ printk(KERN_INFO "0x%02X port_status = 0x%08X\n", (uint8_t)
+ offsetof(struct isp_reg, u2.isp4022.p0.port_status),
+ readw(&ha->reg->u2.isp4022.p0.port_status));
+ printk(KERN_INFO "0x%02X gp_out = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u2.isp4022.p0.gp_out),
+ readw(&ha->reg->u2.isp4022.p0.gp_out));
+ printk(KERN_INFO "0x%02X gp_in = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u2.isp4022.p0.gp_in),
+ readw(&ha->reg->u2.isp4022.p0.gp_in));
+ printk(KERN_INFO "0x%02X port_err_status = 0x%08X\n", (uint8_t)
+ offsetof(struct isp_reg, u2.isp4022.p0.port_err_status),
+ readw(&ha->reg->u2.isp4022.p0.port_err_status));
+ printk(KERN_INFO "Page 1 Registers:\n");
+ writel(HOST_MEM_CFG_PAGE & set_rmask(CSR_SCSI_PAGE_SELECT),
+ &ha->reg->ctrl_status);
+ printk(KERN_INFO "0x%02X req_q_out = 0x%08X\n",
+ (uint8_t) offsetof(struct isp_reg, u2.isp4022.p1.req_q_out),
+ readw(&ha->reg->u2.isp4022.p1.req_q_out));
+ writel(PORT_CTRL_STAT_PAGE & set_rmask(CSR_SCSI_PAGE_SELECT),
+ &ha->reg->ctrl_status);
+ }
+}
#include <linux/delay.h>
#include <linux/interrupt.h>
#include <linux/mutex.h>
+#include <linux/aer.h>
#include <net/tcp.h>
#include <scsi/scsi.h>
#include "ql4_dbg.h"
#include "ql4_nx.h"
-#if defined(CONFIG_PCIEAER)
-#include <linux/aer.h>
-#else
-/* AER releated */
-static inline int pci_enable_pcie_error_reporting(struct pci_dev *dev)
-{
- return -EINVAL;
-}
-static inline int pci_disable_pcie_error_reporting(struct pci_dev *dev)
-{
- return -EINVAL;
-}
-static inline int pci_cleanup_aer_uncorrect_error_status(struct pci_dev *dev)
-{
- return -EINVAL;
-}
-#endif
-
#ifndef PCI_DEVICE_ID_QLOGIC_ISP4010
#define PCI_DEVICE_ID_QLOGIC_ISP4010 0x4010
#endif
#define IOCB_TOV_MARGIN 10
#define RELOGIN_TOV 18
#define ISNS_DEREG_TOV 5
+#define HBA_ONLINE_TOV 30
#define MAX_RESET_HA_RETRIES 2
#define MBOX_ASTS_IPV6_ND_PREFIX_IGNORED 0x802C
#define MBOX_ASTS_IPV6_LCL_PREFIX_IGNORED 0x802D
#define MBOX_ASTS_ICMPV6_ERROR_MSG_RCVD 0x802E
+#define MBOX_ASTS_TXSCVR_INSERTED 0x8130
+#define MBOX_ASTS_TXSCVR_REMOVED 0x8131
#define ISNS_EVENT_DATA_RECEIVED 0x0000
#define ISNS_EVENT_CONNECTION_OPENED 0x0001
#define FWOPT_SESSION_MODE 0x0040
#define FWOPT_INITIATOR_MODE 0x0020
#define FWOPT_TARGET_MODE 0x0010
+#define FWOPT_ENABLE_CRBDB 0x8000
uint16_t exec_throttle; /* 04-05 */
uint8_t zio_count; /* 06 */
void qla4xxx_wake_dpc(struct scsi_qla_host *ha);
void qla4xxx_get_conn_event_log(struct scsi_qla_host *ha);
void qla4xxx_mailbox_premature_completion(struct scsi_qla_host *ha);
+void qla4xxx_dump_registers(struct scsi_qla_host *ha);
void qla4_8xxx_pci_config(struct scsi_qla_host *);
int qla4_8xxx_iospace_config(struct scsi_qla_host *ha);
break;
DEBUG2(printk(KERN_INFO "scsi%ld: %s: Waiting for boot "
- "firmware to complete... ctrl_sts=0x%x\n",
- ha->host_no, __func__, ctrl_status));
+ "firmware to complete... ctrl_sts=0x%x, remaining=%ld\n",
+ ha->host_no, __func__, ctrl_status, max_wait_time));
msleep_interruptible(250);
} while (!time_after_eq(jiffies, max_wait_time));
exit_init_online:
set_bit(AF_ONLINE, &ha->flags);
exit_init_hba:
+ if (is_qla8022(ha) && (status == QLA_ERROR)) {
+ /* Since interrupts are registered in start_firmware for
+ * 82xx, release them here if initialize_adapter fails */
+ qla4xxx_free_irqs(ha);
+ }
+
DEBUG2(printk("scsi%ld: initialize adapter: %s\n", ha->host_no,
status == QLA_ERROR ? "FAILED" : "SUCCEDED"));
return status;
void qla4_8xxx_queue_iocb(struct scsi_qla_host *ha)
{
uint32_t dbval = 0;
- unsigned long wtime;
dbval = 0x14 | (ha->func_num << 5);
dbval = dbval | (0 << 8) | (ha->request_in << 16);
- writel(dbval, (unsigned long __iomem *)ha->nx_db_wr_ptr);
- wmb();
- wtime = jiffies + (2 * HZ);
- while (readl((void __iomem *)ha->nx_db_rd_ptr) != dbval &&
- !time_after_eq(jiffies, wtime)) {
- writel(dbval, (unsigned long __iomem *)ha->nx_db_wr_ptr);
- wmb();
- }
+ qla4_8xxx_wr_32(ha, ha->nx_db_wr_ptr, ha->request_in);
}
/**
{
struct srb *srb = ha->status_srb;
struct scsi_cmnd *cmd;
- uint8_t sense_len;
+ uint16_t sense_len;
if (srb == NULL)
return;
case MBOX_ASTS_SYSTEM_ERROR:
/* Log Mailbox registers */
ql4_printk(KERN_INFO, ha, "%s: System Err\n", __func__);
+ qla4xxx_dump_registers(ha);
+
if (ql4xdontresethba) {
DEBUG2(printk("scsi%ld: %s:Don't Reset HBA\n",
ha->host_no, __func__));
}
break;
+ case MBOX_ASTS_TXSCVR_INSERTED:
+ DEBUG2(printk(KERN_WARNING
+ "scsi%ld: AEN %04x Transceiver"
+ " inserted\n", ha->host_no, mbox_sts[0]));
+ break;
+
+ case MBOX_ASTS_TXSCVR_REMOVED:
+ DEBUG2(printk(KERN_WARNING
+ "scsi%ld: AEN %04x Transceiver"
+ " removed\n", ha->host_no, mbox_sts[0]));
+ break;
+
default:
DEBUG2(printk(KERN_WARNING
"scsi%ld: AEN %04x UNKNOWN\n",
{
memset(mbox_cmd, 0, sizeof(mbox_cmd[0]) * MBOX_REG_COUNT);
memset(mbox_sts, 0, sizeof(mbox_sts[0]) * MBOX_REG_COUNT);
+
+ if (is_qla8022(ha))
+ qla4_8xxx_wr_32(ha, ha->nx_db_wr_ptr, 0);
+
mbox_cmd[0] = MBOX_CMD_INITIALIZE_FIRMWARE;
mbox_cmd[1] = 0;
mbox_cmd[2] = LSDW(init_fw_cb_dma);
init_fw_cb->fw_options |=
__constant_cpu_to_le16(FWOPT_SESSION_MODE |
FWOPT_INITIATOR_MODE);
+
+ if (is_qla8022(ha))
+ init_fw_cb->fw_options |=
+ __constant_cpu_to_le16(FWOPT_ENABLE_CRBDB);
+
init_fw_cb->fw_options &= __constant_cpu_to_le16(~FWOPT_TARGET_MODE);
if (qla4xxx_set_ifcb(ha, &mbox_cmd[0], &mbox_sts[0], init_fw_cb_dma)
}
ql4_printk(KERN_INFO, ha, "%ld firmare IOCBs available (%d).\n",
- ha->host_no, mbox_cmd[2]);
+ ha->host_no, mbox_sts[2]);
return QLA_SUCCESS;
}
done = qla4_8xxx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM2_LOCK));
if (done == 1)
break;
- if (timeout >= qla4_8xxx_rom_lock_timeout)
+ if (timeout >= qla4_8xxx_rom_lock_timeout) {
+ ql4_printk(KERN_WARNING, ha,
+ "%s: Failed to acquire rom lock", __func__);
return -1;
+ }
timeout++;
return 0;
}
-static int qla4_8xxx_check_for_bad_spd(struct scsi_qla_host *ha)
-{
- u32 val = 0;
- val = qla4_8xxx_rd_32(ha, BOOT_LOADER_DIMM_STATUS) ;
- val &= QLA82XX_BOOT_LOADER_MN_ISSUE;
- if (val & QLA82XX_PEG_TUNE_MN_SPD_ZEROED) {
- printk("Memory DIMM SPD not programmed. Assumed valid.\n");
- return 1;
- } else if (val) {
- printk("Memory DIMM type incorrect. Info:%08X.\n", val);
- return 2;
- }
- return 0;
-}
-
static int
qla4_8xxx_load_from_flash(struct scsi_qla_host *ha, uint32_t image_start)
{
} while (--retries);
- qla4_8xxx_check_for_bad_spd(ha);
-
if (!retries) {
pegtune_val = qla4_8xxx_rd_32(ha,
QLA82XX_ROMUSB_GLB_PEGTUNE_DONE);
ql4_printk(KERN_INFO, ha,
"FW: Attempting to load firmware from flash...\n");
rval = qla4_8xxx_start_firmware(ha, ha->hw.flt_region_fw);
- if (rval == QLA_SUCCESS)
- return rval;
- ql4_printk(KERN_ERR, ha, "FW: Load firmware from flash FAILED...\n");
+ if (rval != QLA_SUCCESS) {
+ ql4_printk(KERN_ERR, ha, "FW: Load firmware from flash"
+ " FAILED...\n");
+ return rval;
+ }
return rval;
}
+static void qla4_8xxx_rom_lock_recovery(struct scsi_qla_host *ha)
+{
+ if (qla4_8xxx_rom_lock(ha)) {
+ /* Someone else is holding the lock. */
+ dev_info(&ha->pdev->dev, "Resetting rom_lock\n");
+ }
+
+ /*
+ * Either we got the lock, or someone
+ * else died while holding it.
+ * In either case, unlock.
+ */
+ qla4_8xxx_rom_unlock(ha);
+}
+
/**
* qla4_8xxx_device_bootstrap - Initialize device, set DEV_READY, start fw
* @ha: pointer to adapter structure
static int
qla4_8xxx_device_bootstrap(struct scsi_qla_host *ha)
{
- int rval, i, timeout;
+ int rval = QLA_ERROR;
+ int i, timeout;
uint32_t old_count, count;
+ int need_reset = 0, peg_stuck = 1;
- if (qla4_8xxx_need_reset(ha))
- goto dev_initialize;
+ need_reset = qla4_8xxx_need_reset(ha);
old_count = qla4_8xxx_rd_32(ha, QLA82XX_PEG_ALIVE_COUNTER);
if (timeout) {
qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE,
QLA82XX_DEV_FAILED);
- return QLA_ERROR;
+ return rval;
}
count = qla4_8xxx_rd_32(ha, QLA82XX_PEG_ALIVE_COUNTER);
if (count != old_count)
+ peg_stuck = 0;
+ }
+
+ if (need_reset) {
+ /* We are trying to perform a recovery here. */
+ if (peg_stuck)
+ qla4_8xxx_rom_lock_recovery(ha);
+ goto dev_initialize;
+ } else {
+ /* Start of day for this ha context. */
+ if (peg_stuck) {
+ /* Either we are the first or recovery in progress. */
+ qla4_8xxx_rom_lock_recovery(ha);
+ goto dev_initialize;
+ } else {
+ /* Firmware already running. */
+ rval = QLA_SUCCESS;
goto dev_ready;
+ }
}
dev_initialize:
ql4_printk(KERN_INFO, ha, "HW State: READY\n");
qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA82XX_DEV_READY);
- return QLA_SUCCESS;
+ return rval;
}
/**
int retval;
retval = qla4_8xxx_device_state_handler(ha);
- if (retval == QLA_SUCCESS &&
- !test_bit(AF_INIT_DONE, &ha->flags)) {
+ if (retval == QLA_SUCCESS && !test_bit(AF_INIT_DONE, &ha->flags))
retval = qla4xxx_request_irqs(ha);
- if (retval != QLA_SUCCESS) {
- ql4_printk(KERN_WARNING, ha,
- "Failed to reserve interrupt %d already in use.\n",
- ha->pdev->irq);
- } else {
- set_bit(AF_IRQ_ATTACHED, &ha->flags);
- ha->host->irq = ha->pdev->irq;
- ql4_printk(KERN_INFO, ha, "%s: irq %d attached\n",
- __func__, ha->pdev->irq);
- }
- }
+
return retval;
}
#define CRB_CMDPEG_STATE QLA82XX_REG(0x50)
#define CRB_RCVPEG_STATE QLA82XX_REG(0x13c)
-#define BOOT_LOADER_DIMM_STATUS QLA82XX_REG(0x54)
#define CRB_DMA_SHIFT QLA82XX_REG(0xcc)
#define QLA82XX_HW_H0_CH_HUB_ADR 0x05
# define QLA82XX_CAM_RAM_BASE (QLA82XX_CRB_CAM + 0x02000)
# define QLA82XX_CAM_RAM(reg) (QLA82XX_CAM_RAM_BASE + (reg))
-#define QLA82XX_PEG_TUNE_MN_SPD_ZEROED 0x80000000
-#define QLA82XX_BOOT_LOADER_MN_ISSUE 0xff00ffff
#define QLA82XX_PORT_MODE_ADDR (QLA82XX_CAM_RAM(0x24))
#define QLA82XX_PEG_HALT_STATUS1 (QLA82XX_CAM_RAM(0xa8))
#define QLA82XX_PEG_HALT_STATUS2 (QLA82XX_CAM_RAM(0xac))
#define QLA82XX_PEG_ALIVE_COUNTER (QLA82XX_CAM_RAM(0xb0))
+#define QLA82XX_CAM_RAM_DB1 (QLA82XX_CAM_RAM(0x1b0))
+#define QLA82XX_CAM_RAM_DB2 (QLA82XX_CAM_RAM(0x1b4))
#define HALT_STATUS_UNRECOVERABLE 0x80000000
#define HALT_STATUS_RECOVERABLE 0x40000000
"of (%d) secs exhausted, marking device DEAD.\n",
ha->host_no, __func__, ddb_entry->fw_ddb_index,
QL4_SESS_RECOVERY_TMO));
-
- qla4xxx_wake_dpc(ha);
}
}
if (ha->nx_pcibase)
iounmap(
(struct device_reg_82xx __iomem *)ha->nx_pcibase);
-
- if (ha->nx_db_wr_ptr)
- iounmap(
- (struct device_reg_82xx __iomem *)ha->nx_db_wr_ptr);
} else if (ha->reg)
iounmap(ha->reg);
pci_release_regions(ha->pdev);
qla4xxx_wake_dpc(ha);
qla4xxx_mailbox_premature_completion(ha);
}
- }
+ } else
+ ha->seconds_since_last_heartbeat = 0;
+
ha->fw_heartbeat_counter = fw_heartbeat_counter;
}
/* Find a command that hasn't completed. */
for (index = 0; index < ha->host->can_queue; index++) {
cmd = scsi_host_find_tag(ha->host, index);
- if (cmd != NULL)
+ /*
+ * We cannot just check if the index is valid,
+ * becase if we are run from the scsi eh, then
+ * the scsi/block layer is going to prevent
+ * the tag from being released.
+ */
+ if (cmd != NULL && CMD_SP(cmd))
break;
}
spin_unlock_irqrestore(&ha->hardware_lock, flags);
{
uint32_t max_wait_time;
unsigned long flags = 0;
- int status = QLA_ERROR;
+ int status;
uint32_t ctrl_status;
- qla4xxx_hw_reset(ha);
+ status = qla4xxx_hw_reset(ha);
+ if (status != QLA_SUCCESS)
+ return status;
+ status = QLA_ERROR;
/* Wait until the Network Reset Intr bit is cleared */
max_wait_time = RESET_INTR_TOV;
do {
ha->host_no, __func__));
status = ha->isp_ops->reset_firmware(ha);
if (status == QLA_SUCCESS) {
- qla4xxx_cmd_wait(ha);
+ if (!test_bit(AF_FW_RECOVERY, &ha->flags))
+ qla4xxx_cmd_wait(ha);
ha->isp_ops->disable_intrs(ha);
qla4xxx_process_aen(ha, FLUSH_DDB_CHANGED_AENS);
qla4xxx_abort_active_cmds(ha, DID_RESET << 16);
* or if stop_firmware fails for ISP-82xx.
* This is the default case for ISP-4xxx */
if (!is_qla8022(ha) || reset_chip) {
- qla4xxx_cmd_wait(ha);
+ if (!test_bit(AF_FW_RECOVERY, &ha->flags))
+ qla4xxx_cmd_wait(ha);
qla4xxx_process_aen(ha, FLUSH_DDB_CHANGED_AENS);
qla4xxx_abort_active_cmds(ha, DID_RESET << 16);
DEBUG2(ql4_printk(KERN_INFO, ha,
db_base = pci_resource_start(pdev, 4); /* doorbell is on bar 4 */
db_len = pci_resource_len(pdev, 4);
- /* mapping of doorbell write pointer */
- ha->nx_db_wr_ptr = (unsigned long)ioremap(db_base +
- (ha->pdev->devfn << 12), 4);
- if (!ha->nx_db_wr_ptr) {
- printk(KERN_ERR
- "cannot remap MMIO doorbell-write (%s), aborting\n",
- pci_name(pdev));
- goto iospace_error_exit;
- }
- /* mapping of doorbell read pointer */
- ha->nx_db_rd_ptr = (uint8_t *) ha->nx_pcibase + (512 * 1024) +
- (ha->pdev->devfn * 8);
- if (!ha->nx_db_rd_ptr)
- printk(KERN_ERR
- "cannot remap MMIO doorbell-read (%s), aborting\n",
- pci_name(pdev));
- return 0;
+ ha->nx_db_wr_ptr = (ha->pdev->devfn == 4 ? QLA82XX_CAM_RAM_DB1 :
+ QLA82XX_CAM_RAM_DB2);
+ return 0;
iospace_error_exit:
return -ENOMEM;
}
{
unsigned long wait_online;
- wait_online = jiffies + (30 * HZ);
+ wait_online = jiffies + (HBA_ONLINE_TOV * HZ);
while (time_before(jiffies, wait_online)) {
if (adapter_up(ha))
return QLA_SUCCESS;
- else if (ha->retry_reset_ha_cnt == 0)
- return QLA_ERROR;
msleep(2000);
}
unsigned int id = cmd->device->id;
unsigned int lun = cmd->device->lun;
unsigned long serial = cmd->serial_number;
+ unsigned long flags;
struct srb *srb = NULL;
int ret = SUCCESS;
int wait = 0;
"scsi%ld:%d:%d: Abort command issued cmd=%p, pid=%ld\n",
ha->host_no, id, lun, cmd, serial);
+ spin_lock_irqsave(&ha->hardware_lock, flags);
srb = (struct srb *) CMD_SP(cmd);
-
- if (!srb)
+ if (!srb) {
+ spin_unlock_irqrestore(&ha->hardware_lock, flags);
return SUCCESS;
-
+ }
kref_get(&srb->srb_ref);
+ spin_unlock_irqrestore(&ha->hardware_lock, flags);
if (qla4xxx_abort_task(ha, srb) != QLA_SUCCESS) {
DEBUG3(printk("scsi%ld:%d:%d: Abort_task mbx failed.\n",
qla4xxx_mailbox_premature_completion(ha);
qla4xxx_free_irqs(ha);
pci_disable_device(pdev);
+ /* Return back all IOs */
+ qla4xxx_abort_active_cmds(ha, DID_RESET << 16);
return PCI_ERS_RESULT_NEED_RESET;
case pci_channel_io_perm_failure:
set_bit(AF_EEH_BUSY, &ha->flags);
if (!is_aer_supported(ha))
return PCI_ERS_RESULT_NONE;
- if (test_bit(AF_FW_RECOVERY, &ha->flags)) {
- ql4_printk(KERN_WARNING, ha, "scsi%ld: %s: firmware hang -- "
- "mmio_enabled\n", ha->host_no, __func__);
- return PCI_ERS_RESULT_NEED_RESET;
- } else
- return PCI_ERS_RESULT_RECOVERED;
+ return PCI_ERS_RESULT_RECOVERED;
}
-uint32_t qla4_8xxx_error_recovery(struct scsi_qla_host *ha)
+static uint32_t qla4_8xxx_error_recovery(struct scsi_qla_host *ha)
{
uint32_t rval = QLA_ERROR;
+ uint32_t ret = 0;
int fn;
struct pci_dev *other_pdev = NULL;
clear_bit(AF_ONLINE, &ha->flags);
qla4xxx_mark_all_devices_missing(ha);
qla4xxx_process_aen(ha, FLUSH_DDB_CHANGED_AENS);
- qla4xxx_abort_active_cmds(ha, DID_RESET << 16);
}
fn = PCI_FUNC(ha->pdev->devfn);
/* Clear driver state register */
qla4_8xxx_wr_32(ha, QLA82XX_CRB_DRV_STATE, 0);
qla4_8xxx_set_drv_active(ha);
- ha->isp_ops->enable_intrs(ha);
+ ret = qla4xxx_request_irqs(ha);
+ if (ret) {
+ ql4_printk(KERN_WARNING, ha, "Failed to "
+ "reserve interrupt %d already in use.\n",
+ ha->pdev->irq);
+ rval = QLA_ERROR;
+ } else {
+ ha->isp_ops->enable_intrs(ha);
+ rval = QLA_SUCCESS;
+ }
}
qla4_8xxx_idc_unlock(ha);
} else {
clear_bit(AF_FW_RECOVERY, &ha->flags);
rval = qla4xxx_initialize_adapter(ha,
PRESERVE_DDB_LIST);
- if (rval == QLA_SUCCESS)
- ha->isp_ops->enable_intrs(ha);
+ if (rval == QLA_SUCCESS) {
+ ret = qla4xxx_request_irqs(ha);
+ if (ret) {
+ ql4_printk(KERN_WARNING, ha, "Failed to"
+ " reserve interrupt %d already in"
+ " use.\n", ha->pdev->irq);
+ rval = QLA_ERROR;
+ } else {
+ ha->isp_ops->enable_intrs(ha);
+ rval = QLA_SUCCESS;
+ }
+ }
qla4_8xxx_idc_lock(ha);
qla4_8xxx_set_drv_active(ha);
qla4_8xxx_idc_unlock(ha);
goto exit_slot_reset;
}
- ret = qla4xxx_request_irqs(ha);
- if (ret) {
- ql4_printk(KERN_WARNING, ha, "Failed to reserve interrupt %d"
- " already in use.\n", pdev->irq);
- goto exit_slot_reset;
- }
+ ha->isp_ops->disable_intrs(ha);
if (is_qla8022(ha)) {
if (qla4_8xxx_error_recovery(ha) == QLA_SUCCESS) {
* See LICENSE.qla4xxx for copyright and licensing details.
*/
-#define QLA4XXX_DRIVER_VERSION "5.02.00-k3"
+#define QLA4XXX_DRIVER_VERSION "5.02.00-k4"
sdev->sdev_state = SDEV_RUNNING;
else if (sdev->sdev_state == SDEV_CREATED_BLOCK)
sdev->sdev_state = SDEV_CREATED;
- else
+ else if (sdev->sdev_state != SDEV_CANCEL &&
+ sdev->sdev_state != SDEV_OFFLINE)
return -EINVAL;
spin_lock_irqsave(q->queue_lock, flags);
list_for_each_entry(sdev, &shost->__devices, siblings) {
if (sdev->channel != starget->channel ||
sdev->id != starget->id ||
- sdev->sdev_state == SDEV_DEL)
+ scsi_device_get(sdev))
continue;
spin_unlock_irqrestore(shost->host_lock, flags);
scsi_remove_device(sdev);
+ scsi_device_put(sdev);
spin_lock_irqsave(shost->host_lock, flags);
goto restart;
}
return snprintf(buf, 20, "%u\n", sdkp->protection_type);
}
+static ssize_t
+sd_show_protection_mode(struct device *dev, struct device_attribute *attr,
+ char *buf)
+{
+ struct scsi_disk *sdkp = to_scsi_disk(dev);
+ struct scsi_device *sdp = sdkp->device;
+ unsigned int dif, dix;
+
+ dif = scsi_host_dif_capable(sdp->host, sdkp->protection_type);
+ dix = scsi_host_dix_capable(sdp->host, sdkp->protection_type);
+
+ if (!dix && scsi_host_dix_capable(sdp->host, SD_DIF_TYPE0_PROTECTION)) {
+ dif = 0;
+ dix = 1;
+ }
+
+ if (!dif && !dix)
+ return snprintf(buf, 20, "none\n");
+
+ return snprintf(buf, 20, "%s%u\n", dix ? "dix" : "dif", dif);
+}
+
static ssize_t
sd_show_app_tag_own(struct device *dev, struct device_attribute *attr,
char *buf)
__ATTR(manage_start_stop, S_IRUGO|S_IWUSR, sd_show_manage_start_stop,
sd_store_manage_start_stop),
__ATTR(protection_type, S_IRUGO, sd_show_protection_type, NULL),
+ __ATTR(protection_mode, S_IRUGO, sd_show_protection_mode, NULL),
__ATTR(app_tag_own, S_IRUGO, sd_show_app_tag_own, NULL),
__ATTR(thin_provisioning, S_IRUGO, sd_show_thin_provisioning, NULL),
__ATTR_NULL,
return CDS_NO_DISC;
}
+ /*
+ * SK/ASC/ASCQ of 2/4/2 means "initialization required"
+ * Using CD_TRAY_OPEN results in an START_STOP_UNIT to close
+ * the tray, which resolves the initialization requirement.
+ */
+ if (scsi_sense_valid(&sshdr) && sshdr.sense_key == NOT_READY
+ && sshdr.asc == 0x04 && sshdr.ascq == 0x02)
+ return CDS_TRAY_OPEN;
+
/*
* 0x04 is format in progress .. but there must be a disc present!
*/
CONFIG_ETRAX_RS485_LTC1387_RXEN_PORT_G_BIT, 1);
#endif
- info->rs485.flags = r->flags;
- if (r->delay_rts_before_send >= 1000)
+ info->rs485 = *r;
+
+ /* Maximum delay before RTS equal to 1000 */
+ if (info->rs485.delay_rts_before_send >= 1000)
info->rs485.delay_rts_before_send = 1000;
- else
- info->rs485.delay_rts_before_send = r->delay_rts_before_send;
+
/* printk("rts: on send = %i, after = %i, enabled = %i",
info->rs485.rts_on_send,
info->rs485.rts_after_sent,
e100_disable_rx(info);
e100_enable_rx_irq(info);
#endif
-
- if (info->rs485.delay_rts_before_send > 0)
- msleep(info->rs485.delay_rts_before_send);
+ if ((info->rs485.flags & SER_RS485_RTS_BEFORE_SEND) &&
+ (info->rs485.delay_rts_before_send > 0))
+ msleep(info->rs485.delay_rts_before_send);
}
#endif /* CONFIG_ETRAX_RS485 */
rs485data.delay_rts_before_send = rs485ctrl.delay_rts_before_send;
rs485data.flags = 0;
+ if (rs485data.delay_rts_before_send != 0)
+ rs485data.flags |= SER_RS485_RTS_BEFORE_SEND;
+ else
+ rs485data.flags &= ~(SER_RS485_RTS_BEFORE_SEND);
+
if (rs485ctrl.enabled)
rs485data.flags |= SER_RS485_ENABLED;
else
/* This is the ioctl to get RS485 data from user-space */
if (copy_to_user((struct serial_rs485 *) arg,
rs485data,
- sizeof(serial_rs485)))
+ sizeof(struct serial_rs485)))
return -EFAULT;
break;
}
/* Set sane defaults */
info->rs485.flags &= ~(SER_RS485_RTS_ON_SEND);
info->rs485.flags |= SER_RS485_RTS_AFTER_SEND;
+ info->rs485.flags &= ~(SER_RS485_RTS_BEFORE_SEND);
info->rs485.delay_rts_before_send = 0;
info->rs485.flags &= ~(SER_RS485_ENABLED);
#endif
static long clk_rate_round_helper(struct clk_rate_round_data *rounder)
{
unsigned long rate_error, rate_error_prev = ~0UL;
- unsigned long rate_best_fit = rounder->rate;
unsigned long highest, lowest, freq;
+ long rate_best_fit = -ENOENT;
int i;
highest = 0;
};
if (clk->nr_freqs < 1)
- return 0;
+ return -ENOSYS;
return clk_rate_round_helper(&table_round);
}
}
EXPORT_SYMBOL_GPL(clk_round_rate);
+long clk_round_parent(struct clk *clk, unsigned long target,
+ unsigned long *best_freq, unsigned long *parent_freq,
+ unsigned int div_min, unsigned int div_max)
+{
+ struct cpufreq_frequency_table *freq, *best = NULL;
+ unsigned long error = ULONG_MAX, freq_high, freq_low, div;
+ struct clk *parent = clk_get_parent(clk);
+
+ if (!parent) {
+ *parent_freq = 0;
+ *best_freq = clk_round_rate(clk, target);
+ return abs(target - *best_freq);
+ }
+
+ for (freq = parent->freq_table; freq->frequency != CPUFREQ_TABLE_END;
+ freq++) {
+ if (freq->frequency == CPUFREQ_ENTRY_INVALID)
+ continue;
+
+ if (unlikely(freq->frequency / target <= div_min - 1)) {
+ unsigned long freq_max;
+
+ freq_max = (freq->frequency + div_min / 2) / div_min;
+ if (error > target - freq_max) {
+ error = target - freq_max;
+ best = freq;
+ if (best_freq)
+ *best_freq = freq_max;
+ }
+
+ pr_debug("too low freq %lu, error %lu\n", freq->frequency,
+ target - freq_max);
+
+ if (!error)
+ break;
+
+ continue;
+ }
+
+ if (unlikely(freq->frequency / target >= div_max)) {
+ unsigned long freq_min;
+
+ freq_min = (freq->frequency + div_max / 2) / div_max;
+ if (error > freq_min - target) {
+ error = freq_min - target;
+ best = freq;
+ if (best_freq)
+ *best_freq = freq_min;
+ }
+
+ pr_debug("too high freq %lu, error %lu\n", freq->frequency,
+ freq_min - target);
+
+ if (!error)
+ break;
+
+ continue;
+ }
+
+ div = freq->frequency / target;
+ freq_high = freq->frequency / div;
+ freq_low = freq->frequency / (div + 1);
+
+ if (freq_high - target < error) {
+ error = freq_high - target;
+ best = freq;
+ if (best_freq)
+ *best_freq = freq_high;
+ }
+
+ if (target - freq_low < error) {
+ error = target - freq_low;
+ best = freq;
+ if (best_freq)
+ *best_freq = freq_low;
+ }
+
+ pr_debug("%u / %lu = %lu, / %lu = %lu, best %lu, parent %u\n",
+ freq->frequency, div, freq_high, div + 1, freq_low,
+ *best_freq, best->frequency);
+
+ if (!error)
+ break;
+ }
+
+ if (parent_freq)
+ *parent_freq = best->frequency;
+
+ return error;
+}
+EXPORT_SYMBOL_GPL(clk_round_parent);
+
#ifdef CONFIG_PM
static int clks_sysdev_suspend(struct sys_device *dev, pm_message_t state)
{
* Register the IRQ position with the global IRQ map, then insert
* it in to the radix tree.
*/
- irq_reserve_irqs(irq, 1);
+ irq_reserve_irq(irq);
raw_spin_lock_irqsave(&intc_big_lock, flags);
radix_tree_insert(&d->tree, enum_id, intc_irq_xlate_get(irq));
int i;
for (i = 0; i < nr_vecs; i++)
- irq_reserve_irqs(evt2irq(vectors[i].vect), 1);
+ irq_reserve_irq(evt2irq(vectors[i].vect));
}
#include <linux/spi/spi.h>
#include <linux/of_spi.h>
-
-/* SPI bustype and spi_master class are registered after board init code
- * provides the SPI device tables, ensuring that both are present by the
- * time controller driver registration causes spi_devices to "enumerate".
- */
static void spidev_release(struct device *dev)
{
struct spi_device *spi = to_spi_device(dev);
struct boardinfo {
struct list_head list;
- unsigned n_board_info;
- struct spi_board_info board_info[0];
+ struct spi_board_info board_info;
};
static LIST_HEAD(board_list);
+static LIST_HEAD(spi_master_list);
+
+/*
+ * Used to protect add/del opertion for board_info list and
+ * spi_master list, and their matching process
+ */
static DEFINE_MUTEX(board_lock);
/**
*/
status = spi_setup(spi);
if (status < 0) {
- dev_err(dev, "can't %s %s, status %d\n",
- "setup", dev_name(&spi->dev), status);
+ dev_err(dev, "can't setup %s, status %d\n",
+ dev_name(&spi->dev), status);
goto done;
}
/* Device may be bound to an active driver when this returns */
status = device_add(&spi->dev);
if (status < 0)
- dev_err(dev, "can't %s %s, status %d\n",
- "add", dev_name(&spi->dev), status);
+ dev_err(dev, "can't add %s, status %d\n",
+ dev_name(&spi->dev), status);
else
dev_dbg(dev, "registered child %s\n", dev_name(&spi->dev));
}
EXPORT_SYMBOL_GPL(spi_new_device);
+static void spi_match_master_to_boardinfo(struct spi_master *master,
+ struct spi_board_info *bi)
+{
+ struct spi_device *dev;
+
+ if (master->bus_num != bi->bus_num)
+ return;
+
+ dev = spi_new_device(master, bi);
+ if (!dev)
+ dev_err(master->dev.parent, "can't create new device for %s\n",
+ bi->modalias);
+}
+
/**
* spi_register_board_info - register SPI devices for a given board
* @info: array of chip descriptors
int __init
spi_register_board_info(struct spi_board_info const *info, unsigned n)
{
- struct boardinfo *bi;
+ struct boardinfo *bi;
+ int i;
- bi = kmalloc(sizeof(*bi) + n * sizeof *info, GFP_KERNEL);
+ bi = kzalloc(n * sizeof(*bi), GFP_KERNEL);
if (!bi)
return -ENOMEM;
- bi->n_board_info = n;
- memcpy(bi->board_info, info, n * sizeof *info);
- mutex_lock(&board_lock);
- list_add_tail(&bi->list, &board_list);
- mutex_unlock(&board_lock);
- return 0;
-}
+ for (i = 0; i < n; i++, bi++, info++) {
+ struct spi_master *master;
-/* FIXME someone should add support for a __setup("spi", ...) that
- * creates board info from kernel command lines
- */
-
-static void scan_boardinfo(struct spi_master *master)
-{
- struct boardinfo *bi;
-
- mutex_lock(&board_lock);
- list_for_each_entry(bi, &board_list, list) {
- struct spi_board_info *chip = bi->board_info;
- unsigned n;
-
- for (n = bi->n_board_info; n > 0; n--, chip++) {
- if (chip->bus_num != master->bus_num)
- continue;
- /* NOTE: this relies on spi_new_device to
- * issue diagnostics when given bogus inputs
- */
- (void) spi_new_device(master, chip);
- }
+ memcpy(&bi->board_info, info, sizeof(*info));
+ mutex_lock(&board_lock);
+ list_add_tail(&bi->list, &board_list);
+ list_for_each_entry(master, &spi_master_list, list)
+ spi_match_master_to_boardinfo(master, &bi->board_info);
+ mutex_unlock(&board_lock);
}
- mutex_unlock(&board_lock);
+
+ return 0;
}
/*-------------------------------------------------------------------------*/
{
static atomic_t dyn_bus_id = ATOMIC_INIT((1<<15) - 1);
struct device *dev = master->dev.parent;
+ struct boardinfo *bi;
int status = -ENODEV;
int dynamic = 0;
dev_dbg(dev, "registered master %s%s\n", dev_name(&master->dev),
dynamic ? " (dynamic)" : "");
- /* populate children from any spi device tables */
- scan_boardinfo(master);
+ mutex_lock(&board_lock);
+ list_add_tail(&master->list, &spi_master_list);
+ list_for_each_entry(bi, &board_list, list)
+ spi_match_master_to_boardinfo(master, &bi->board_info);
+ mutex_unlock(&board_lock);
+
status = 0;
/* Register devices from the device tree */
{
int dummy;
- dummy = device_for_each_child(&master->dev, NULL, __unregister);
+ mutex_lock(&board_lock);
+ list_del(&master->list);
+ mutex_unlock(&board_lock);
+
+ dummy = device_for_each_child(master->dev.parent, &master->dev,
+ __unregister);
device_unregister(&master->dev);
}
EXPORT_SYMBOL_GPL(spi_unregister_master);
*/
bad_bits = spi->mode & ~spi->master->mode_bits;
if (bad_bits) {
- dev_dbg(&spi->dev, "setup: unsupported mode bits %x\n",
+ dev_err(&spi->dev, "setup: unsupported mode bits %x\n",
bad_bits);
return -EINVAL;
}
"in dma_irq_handler dmastat:0x%x spistat:0x%x\n",
dmastat, spistat);
+ if (drv_data->rx != NULL) {
+ u16 cr = read_CTRL(drv_data);
+ /* discard old RX data and clear RXS */
+ bfin_spi_dummy_read(drv_data);
+ write_CTRL(drv_data, cr & ~BIT_CTL_ENABLE); /* Disable SPI */
+ write_CTRL(drv_data, cr & ~BIT_CTL_TIMOD); /* Restore State */
+ write_STAT(drv_data, BIT_STAT_CLR); /* Clear Status */
+ }
+
clear_dma_irqstat(drv_data->dma_channel);
/*
}
if (chip->chip_select_num >= MAX_CTRL_CS) {
- ret = gpio_request(chip->cs_gpio, spi->modalias);
- if (ret) {
- dev_err(&spi->dev, "gpio_request() error\n");
- goto pin_error;
+ /* Only request on first setup */
+ if (spi_get_ctldata(spi) == NULL) {
+ ret = gpio_request(chip->cs_gpio, spi->modalias);
+ if (ret) {
+ dev_err(&spi->dev, "gpio_request() error\n");
+ goto pin_error;
+ }
+ gpio_direction_output(chip->cs_gpio, 1);
}
- gpio_direction_output(chip->cs_gpio, 1);
}
dev_dbg(&spi->dev, "setup spi chip %s, width is %d, dma is %d\n",
do {
/* check if host supports scatter requests and it meets our requirements */
- if (device->func->card->host->max_hw_segs < MAX_SCATTER_ENTRIES_PER_REQ) {
+ if (device->func->card->host->max_segs < MAX_SCATTER_ENTRIES_PER_REQ) {
AR_DEBUG_PRINTF(ATH_DEBUG_ERR,("HIF-SCATTER : host only supports scatter of : %d entries, need: %d \n",
- device->func->card->host->max_hw_segs, MAX_SCATTER_ENTRIES_PER_REQ));
+ device->func->card->host->max_segs, MAX_SCATTER_ENTRIES_PER_REQ));
status = A_ENOTSUP;
break;
}
for (streamID = HTC_RAW_STREAM_0; streamID < HTC_RAW_STREAM_NUM_MAX; streamID++) {
/* Initialize the data structures */
- init_MUTEX(&arRaw->raw_htc_read_sem[streamID]);
- init_MUTEX(&arRaw->raw_htc_write_sem[streamID]);
+ sema_init(&arRaw->raw_htc_read_sem[streamID], 1);
+ sema_init(&arRaw->raw_htc_write_sem[streamID], 1);
init_waitqueue_head(&arRaw->raw_htc_read_queue[streamID]);
init_waitqueue_head(&arRaw->raw_htc_write_queue[streamID]);
goto fail;
net->netdev_ops = NULL;
- init_MUTEX(&dhd->proto_sem);
+ sema_init(&dhd->proto_sem, 1);
/* Initialize other structure content */
init_waitqueue_head(&dhd->ioctl_resp_wait);
init_waitqueue_head(&dhd->ctrl_wait);
dhd->timer.function = dhd_watchdog;
/* Initialize thread based operation and lock */
- init_MUTEX(&dhd->sdsem);
+ sema_init(&dhd->sdsem, 1);
if ((dhd_watchdog_prio >= 0) && (dhd_dpc_prio >= 0))
dhd->threads_only = true;
else
spin_lock_init(&wl->rpcq_lock);
spin_lock_init(&wl->txq_lock);
- init_MUTEX(&wl->sem);
+ sema_init(&wl->sem, 1);
#else
spin_lock_init(&wl->lock);
spin_lock_init(&wl->isr_lock);
#define DT9812_NUM_SLOTS 16
-static DECLARE_MUTEX(dt9812_mutex);
+static DEFINE_SEMAPHORE(dt9812_mutex);
static const struct usb_device_id dt9812_table[] = {
{USB_DEVICE(0x0867, 0x9812)},
*/
static struct usbduxsub usbduxsub[NUMUSBDUX];
-static DECLARE_MUTEX(start_stop_sem);
+static DEFINE_SEMAPHORE(start_stop_sem);
/*
* Stops the data acquision
dev_dbg(dev, "comedi_: usbdux: "
"usbduxsub[%d] is ready to connect to comedi.\n", index);
- init_MUTEX(&(usbduxsub[index].sem));
+ sema_init(&(usbduxsub[index].sem), 1);
/* save a pointer to the usb device */
usbduxsub[index].usbdev = udev;
*/
static struct usbduxfastsub_s usbduxfastsub[NUMUSBDUXFAST];
-static DECLARE_MUTEX(start_stop_sem);
+static DEFINE_SEMAPHORE(start_stop_sem);
/*
* bulk transfers to usbduxfast
"connect to comedi.\n", index);
#endif
- init_MUTEX(&(usbduxfastsub[index].sem));
+ sema_init(&(usbduxfastsub[index].sem), 1);
/* save a pointer to the usb device */
usbduxfastsub[index].usbdev = udev;
return ret;
}
-DECLARE_MUTEX(msm_fb_pan_sem);
+DEFINE_SEMAPHORE(msm_fb_pan_sem);
static int msm_fb_pan_display(struct fb_var_screeninfo *var,
struct fb_info *info)
#endif
-DECLARE_MUTEX(msm_fb_ioctl_ppp_sem);
+DEFINE_SEMAPHORE(msm_fb_ioctl_ppp_sem);
DEFINE_MUTEX(msm_fb_ioctl_lut_sem);
DEFINE_MUTEX(msm_fb_ioctl_hist_sem);
static inline void _rtl_rwlock_init(struct semaphore *prwlock)
{
- init_MUTEX(prwlock);
+ sema_init(prwlock, 1);
}
static inline void _init_listhead(struct list_head *list)
server->mnt = NULL;
server->sock_file = NULL;
init_waitqueue_head(&server->conn_wq);
- init_MUTEX(&server->sem);
+ sema_init(&server->sem, 1);
INIT_LIST_HEAD(&server->entry);
INIT_LIST_HEAD(&server->xmitq);
INIT_LIST_HEAD(&server->recvq);
&solo_enc->lock,
V4L2_BUF_TYPE_VIDEO_CAPTURE,
V4L2_FIELD_INTERLACED,
- sizeof(struct videobuf_buffer), fh);
+ sizeof(struct videobuf_buffer), fh, NULL);
spin_unlock(&solo_enc->lock);
&solo_dev->pdev->dev, &fh->slock,
V4L2_BUF_TYPE_VIDEO_CAPTURE,
SOLO_DISP_PIX_FIELD,
- sizeof(struct videobuf_buffer), fh);
+ sizeof(struct videobuf_buffer), fh, NULL);
return 0;
}
static struct i2c_adapter tm6000_adap_template = {
.owner = THIS_MODULE,
- .class = I2C_CLASS_TV_ANALOG | I2C_CLASS_TV_DIGITAL,
.name = "tm6000",
.algo = &tm6000_algo,
};
/* pointer to west bridge block data device superstructure */
static struct cyasblkdev_blk_data *gl_bd;
-static DECLARE_MUTEX(open_lock);
+static DEFINE_SEMAPHORE(open_lock);
/* local forwardd declarationss */
static cy_as_device_handle *cyas_dev_handle;
init_completion(&bq->thread_complete);
init_waitqueue_head(&bq->thread_wq);
- init_MUTEX(&bq->thread_sem);
+ sema_init(&bq->thread_sem, 1);
ret = kernel_thread(cyasblkdev_queue_thread, bq, CLONE_KERNEL);
if (ret >= 0) {
--- /dev/null
+obj-y += tty_io.o n_tty.o tty_ioctl.o tty_ldisc.o \
+ tty_buffer.o tty_port.o tty_mutex.o
+obj-$(CONFIG_LEGACY_PTYS) += pty.o
+obj-$(CONFIG_UNIX98_PTYS) += pty.o
+obj-$(CONFIG_AUDIT) += tty_audit.o
+obj-$(CONFIG_MAGIC_SYSRQ) += sysrq.o
+obj-$(CONFIG_N_HDLC) += n_hdlc.o
+obj-$(CONFIG_N_GSM) += n_gsm.o
+obj-$(CONFIG_R3964) += n_r3964.o
+
+obj-y += vt/
--- /dev/null
+#
+# This file contains the font map for the default (hardware) font
+#
+FONTMAPFILE = cp437.uni
+
+obj-$(CONFIG_VT) += vt_ioctl.o vc_screen.o \
+ selection.o keyboard.o
+obj-$(CONFIG_CONSOLE_TRANSLATIONS) += consolemap.o consolemap_deftbl.o
+obj-$(CONFIG_HW_CONSOLE) += vt.o defkeymap.o
+
+# Files generated that shall be removed upon make clean
+clean-files := consolemap_deftbl.c defkeymap.c
+
+quiet_cmd_conmk = CONMK $@
+ cmd_conmk = scripts/conmakehash $< > $@
+
+$(obj)/consolemap_deftbl.c: $(src)/$(FONTMAPFILE)
+ $(call cmd,conmk)
+
+$(obj)/defkeymap.o: $(obj)/defkeymap.c
+
+# Uncomment if you're changing the keymap and have an appropriate
+# loadkeys version for the map. By default, we'll use the shipped
+# versions.
+# GENERATE_KEYMAP := 1
+
+ifdef GENERATE_KEYMAP
+
+$(obj)/defkeymap.c: $(obj)/%.c: $(src)/%.map
+ loadkeys --mktable $< > $@.tmp
+ sed -e 's/^static *//' $@.tmp > $@
+ rm $@.tmp
+
+endif
INFO(dev, "MAC %pM\n", net->dev_addr);
INFO(dev, "HOST MAC %pM\n", dev->host_mac);
- netif_stop_queue(net);
the_dev = dev;
}
struct device_attribute *attr, const char *buf, size_t count)
{
struct adp8860_bl *data = dev_get_drvdata(dev);
+ int ret = strict_strtoul(buf, 10, &data->cached_daylight_max);
+ if (ret)
+ return ret;
- strict_strtoul(buf, 10, &data->cached_daylight_max);
return adp8860_store(dev, buf, count, ADP8860_BLMX1);
}
static DEVICE_ATTR(l1_daylight_max, 0664, adp8860_bl_l1_daylight_max_show,
if (val == 0) {
/* Enable automatic ambient light sensing */
adp8860_set_bits(data->client, ADP8860_MDCR, CMP_AUTOEN);
- } else if ((val > 0) && (val < 6)) {
+ } else if ((val > 0) && (val <= 3)) {
/* Disable automatic ambient light sensing */
adp8860_clr_bits(data->client, ADP8860_MDCR, CMP_AUTOEN);
mutex_lock(&data->lock);
adp8860_read(data->client, ADP8860_CFGR, ®_val);
reg_val &= ~(CFGR_BLV_MASK << CFGR_BLV_SHIFT);
- reg_val |= val << CFGR_BLV_SHIFT;
+ reg_val |= (val - 1) << CFGR_BLV_SHIFT;
adp8860_write(data->client, ADP8860_CFGR, reg_val);
mutex_unlock(&data->lock);
}
const u16 slpin = 0x10;
const u16 disoff = 0x28;
- if (power) {
+ if (power <= FB_BLANK_NORMAL) {
if (priv->lcd_on)
return 0;
struct spi_device *spi = st->spi;
struct lms283gf05_pdata *pdata = spi->dev.platform_data;
- if (power) {
+ if (power <= FB_BLANK_NORMAL) {
if (pdata)
lms283gf05_reset(pdata->reset_gpio,
pdata->reset_inverted);
},
.driver_data = (void *)&nvidia_chipset_data,
},
+ {
+ .callback = mbp_dmi_match,
+ .ident = "MacBookAir 3,1",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "MacBookAir3,1"),
+ },
+ .driver_data = (void *)&nvidia_chipset_data,
+ },
+ {
+ .callback = mbp_dmi_match,
+ .ident = "MacBookAir 3,2",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "MacBookAir3,2"),
+ },
+ .driver_data = (void *)&nvidia_chipset_data,
+ },
{ }
};
struct pwm_device *pwm;
struct device *dev;
unsigned int period;
+ unsigned int lth_brightness;
int (*notify)(struct device *,
int brightness);
};
pwm_config(pb->pwm, 0, pb->period);
pwm_disable(pb->pwm);
} else {
- pwm_config(pb->pwm, brightness * pb->period / max, pb->period);
+ brightness = pb->lth_brightness +
+ (brightness * (pb->period - pb->lth_brightness) / max);
+ pwm_config(pb->pwm, brightness, pb->period);
pwm_enable(pb->pwm);
}
return 0;
pb->period = data->pwm_period_ns;
pb->notify = data->notify;
+ pb->lth_brightness = data->lth_brightness *
+ (data->pwm_period_ns / data->max_brightness);
pb->dev = &pdev->dev;
pb->pwm = pwm_request(data->pwm_id, "backlight");
return strlen(buf);
}
-static DEVICE_ATTR(gamma_table, 0644,
+static DEVICE_ATTR(gamma_table, 0444,
s6e63m0_sysfs_show_gamma_table, NULL);
-static int __init s6e63m0_probe(struct spi_device *spi)
+static int __devinit s6e63m0_probe(struct spi_device *spi)
{
int ret = 0;
struct s6e63m0 *lcd = NULL;
struct s6e63m0 *lcd = dev_get_drvdata(&spi->dev);
s6e63m0_power(lcd, FB_BLANK_POWERDOWN);
+ device_remove_file(&spi->dev, &dev_attr_gamma_table);
+ device_remove_file(&spi->dev, &dev_attr_gamma_mode);
+ backlight_device_unregister(lcd->bd);
lcd_device_unregister(lcd->ld);
kfree(lcd);
v) mount check for unmatched uids
-w) Add support for new vfs entry points for setlease and fallocate
+w) Add support for new vfs entry point for fallocate
x) Fix Samba 3 server to handle Linux kernel aio so dbench with lots of
processes can proceed better in parallel (on the server)
* the GNU Lesser General Public License for more details.
*
*/
-#include <linux/radix-tree.h>
+#include <linux/rbtree.h>
#ifndef _CIFS_FS_SB_H
#define _CIFS_FS_SB_H
#define CIFS_MOUNT_MULTIUSER 0x20000 /* multiuser mount */
struct cifs_sb_info {
- struct radix_tree_root tlink_tree;
-#define CIFS_TLINK_MASTER_TAG 0 /* is "master" (mount) tcon */
+ struct rb_root tlink_tree;
spinlock_t tlink_tree_lock;
+ struct tcon_link *master_tlink;
struct nls_table *local_nls;
unsigned int rsize;
unsigned int wsize;
return -ENOMEM;
spin_lock_init(&cifs_sb->tlink_tree_lock);
- INIT_RADIX_TREE(&cifs_sb->tlink_tree, GFP_KERNEL);
+ cifs_sb->tlink_tree = RB_ROOT;
rc = bdi_setup_and_register(&cifs_sb->bdi, "cifs", BDI_CAP_MAP_COPY);
if (rc) {
/* Until the file is open and we have gotten oplock
info back from the server, can not assume caching of
file data or metadata */
- cifs_inode->clientCanCacheRead = false;
- cifs_inode->clientCanCacheAll = false;
+ cifs_set_oplock_level(cifs_inode, 0);
cifs_inode->delete_pending = false;
cifs_inode->invalid_mapping = false;
cifs_inode->vfs_inode.i_blkbits = 14; /* 2**14 = CIFS_MAX_MSGSIZE */
knows that the file won't be changed on the server
by anyone else */
return generic_setlease(file, arg, lease);
- else {
- if (arg != F_UNLCK)
- locks_free_lock(*lease);
+ else
return -EAGAIN;
- }
}
struct file_system_type cifs_fs_type = {
* "get" on the container.
*/
struct tcon_link {
- unsigned long tl_index;
+ struct rb_node tl_rbnode;
+ uid_t tl_uid;
unsigned long tl_flags;
#define TCON_LINK_MASTER 0
#define TCON_LINK_PENDING 1
extern u64 cifs_UnixTimeToNT(struct timespec);
extern struct timespec cnvrtDosUnixTm(__le16 le_date, __le16 le_time,
int offset);
+extern void cifs_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock);
extern struct cifsFileInfo *cifs_new_fileinfo(__u16 fileHandle,
struct file *file, struct tcon_link *tlink,
static int ipv4_connect(struct TCP_Server_Info *server);
static int ipv6_connect(struct TCP_Server_Info *server);
+static void tlink_rb_insert(struct rb_root *root, struct tcon_link *new_tlink);
static void cifs_prune_tlinks(struct work_struct *work);
/*
goto mount_fail_check;
}
- tlink->tl_index = pSesInfo->linux_uid;
+ tlink->tl_uid = pSesInfo->linux_uid;
tlink->tl_tcon = tcon;
tlink->tl_time = jiffies;
set_bit(TCON_LINK_MASTER, &tlink->tl_flags);
set_bit(TCON_LINK_IN_TREE, &tlink->tl_flags);
- rc = radix_tree_preload(GFP_KERNEL);
- if (rc == -ENOMEM) {
- kfree(tlink);
- goto mount_fail_check;
- }
-
+ cifs_sb->master_tlink = tlink;
spin_lock(&cifs_sb->tlink_tree_lock);
- radix_tree_insert(&cifs_sb->tlink_tree, pSesInfo->linux_uid, tlink);
- radix_tree_tag_set(&cifs_sb->tlink_tree, pSesInfo->linux_uid,
- CIFS_TLINK_MASTER_TAG);
+ tlink_rb_insert(&cifs_sb->tlink_tree, tlink);
spin_unlock(&cifs_sb->tlink_tree_lock);
- radix_tree_preload_end();
queue_delayed_work(system_nrt_wq, &cifs_sb->prune_tlinks,
TLINK_IDLE_EXPIRE);
int
cifs_umount(struct super_block *sb, struct cifs_sb_info *cifs_sb)
{
- int i, ret;
+ struct rb_root *root = &cifs_sb->tlink_tree;
+ struct rb_node *node;
+ struct tcon_link *tlink;
char *tmp;
- struct tcon_link *tlink[8];
- unsigned long index = 0;
cancel_delayed_work_sync(&cifs_sb->prune_tlinks);
- do {
- spin_lock(&cifs_sb->tlink_tree_lock);
- ret = radix_tree_gang_lookup(&cifs_sb->tlink_tree,
- (void **)tlink, index,
- ARRAY_SIZE(tlink));
- /* increment index for next pass */
- if (ret > 0)
- index = tlink[ret - 1]->tl_index + 1;
- for (i = 0; i < ret; i++) {
- cifs_get_tlink(tlink[i]);
- clear_bit(TCON_LINK_IN_TREE, &tlink[i]->tl_flags);
- radix_tree_delete(&cifs_sb->tlink_tree,
- tlink[i]->tl_index);
- }
- spin_unlock(&cifs_sb->tlink_tree_lock);
+ spin_lock(&cifs_sb->tlink_tree_lock);
+ while ((node = rb_first(root))) {
+ tlink = rb_entry(node, struct tcon_link, tl_rbnode);
+ cifs_get_tlink(tlink);
+ clear_bit(TCON_LINK_IN_TREE, &tlink->tl_flags);
+ rb_erase(node, root);
- for (i = 0; i < ret; i++)
- cifs_put_tlink(tlink[i]);
- } while (ret != 0);
+ spin_unlock(&cifs_sb->tlink_tree_lock);
+ cifs_put_tlink(tlink);
+ spin_lock(&cifs_sb->tlink_tree_lock);
+ }
+ spin_unlock(&cifs_sb->tlink_tree_lock);
tmp = cifs_sb->prepath;
cifs_sb->prepathlen = 0;
return tcon;
}
-static struct tcon_link *
+static inline struct tcon_link *
cifs_sb_master_tlink(struct cifs_sb_info *cifs_sb)
{
- struct tcon_link *tlink;
- unsigned int ret;
-
- spin_lock(&cifs_sb->tlink_tree_lock);
- ret = radix_tree_gang_lookup_tag(&cifs_sb->tlink_tree, (void **)&tlink,
- 0, 1, CIFS_TLINK_MASTER_TAG);
- spin_unlock(&cifs_sb->tlink_tree_lock);
-
- /* the master tcon should always be present */
- if (ret == 0)
- BUG();
-
- return tlink;
+ return cifs_sb->master_tlink;
}
struct cifsTconInfo *
return signal_pending(current) ? -ERESTARTSYS : 0;
}
+/* find and return a tlink with given uid */
+static struct tcon_link *
+tlink_rb_search(struct rb_root *root, uid_t uid)
+{
+ struct rb_node *node = root->rb_node;
+ struct tcon_link *tlink;
+
+ while (node) {
+ tlink = rb_entry(node, struct tcon_link, tl_rbnode);
+
+ if (tlink->tl_uid > uid)
+ node = node->rb_left;
+ else if (tlink->tl_uid < uid)
+ node = node->rb_right;
+ else
+ return tlink;
+ }
+ return NULL;
+}
+
+/* insert a tcon_link into the tree */
+static void
+tlink_rb_insert(struct rb_root *root, struct tcon_link *new_tlink)
+{
+ struct rb_node **new = &(root->rb_node), *parent = NULL;
+ struct tcon_link *tlink;
+
+ while (*new) {
+ tlink = rb_entry(*new, struct tcon_link, tl_rbnode);
+ parent = *new;
+
+ if (tlink->tl_uid > new_tlink->tl_uid)
+ new = &((*new)->rb_left);
+ else
+ new = &((*new)->rb_right);
+ }
+
+ rb_link_node(&new_tlink->tl_rbnode, parent, new);
+ rb_insert_color(&new_tlink->tl_rbnode, root);
+}
+
/*
* Find or construct an appropriate tcon given a cifs_sb and the fsuid of the
* current task.
* If the superblock doesn't refer to a multiuser mount, then just return
* the master tcon for the mount.
*
- * First, search the radix tree for an existing tcon for this fsuid. If one
+ * First, search the rbtree for an existing tcon for this fsuid. If one
* exists, then check to see if it's pending construction. If it is then wait
* for construction to complete. Once it's no longer pending, check to see if
* it failed and either return an error or retry construction, depending on
cifs_sb_tlink(struct cifs_sb_info *cifs_sb)
{
int ret;
- unsigned long fsuid = (unsigned long) current_fsuid();
+ uid_t fsuid = current_fsuid();
struct tcon_link *tlink, *newtlink;
if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER))
return cifs_get_tlink(cifs_sb_master_tlink(cifs_sb));
spin_lock(&cifs_sb->tlink_tree_lock);
- tlink = radix_tree_lookup(&cifs_sb->tlink_tree, fsuid);
+ tlink = tlink_rb_search(&cifs_sb->tlink_tree, fsuid);
if (tlink)
cifs_get_tlink(tlink);
spin_unlock(&cifs_sb->tlink_tree_lock);
newtlink = kzalloc(sizeof(*tlink), GFP_KERNEL);
if (newtlink == NULL)
return ERR_PTR(-ENOMEM);
- newtlink->tl_index = fsuid;
+ newtlink->tl_uid = fsuid;
newtlink->tl_tcon = ERR_PTR(-EACCES);
set_bit(TCON_LINK_PENDING, &newtlink->tl_flags);
set_bit(TCON_LINK_IN_TREE, &newtlink->tl_flags);
cifs_get_tlink(newtlink);
- ret = radix_tree_preload(GFP_KERNEL);
- if (ret != 0) {
- kfree(newtlink);
- return ERR_PTR(ret);
- }
-
spin_lock(&cifs_sb->tlink_tree_lock);
/* was one inserted after previous search? */
- tlink = radix_tree_lookup(&cifs_sb->tlink_tree, fsuid);
+ tlink = tlink_rb_search(&cifs_sb->tlink_tree, fsuid);
if (tlink) {
cifs_get_tlink(tlink);
spin_unlock(&cifs_sb->tlink_tree_lock);
- radix_tree_preload_end();
kfree(newtlink);
goto wait_for_construction;
}
- ret = radix_tree_insert(&cifs_sb->tlink_tree, fsuid, newtlink);
- spin_unlock(&cifs_sb->tlink_tree_lock);
- radix_tree_preload_end();
- if (ret) {
- kfree(newtlink);
- return ERR_PTR(ret);
- }
tlink = newtlink;
+ tlink_rb_insert(&cifs_sb->tlink_tree, tlink);
+ spin_unlock(&cifs_sb->tlink_tree_lock);
} else {
wait_for_construction:
ret = wait_on_bit(&tlink->tl_flags, TCON_LINK_PENDING,
{
struct cifs_sb_info *cifs_sb = container_of(work, struct cifs_sb_info,
prune_tlinks.work);
- struct tcon_link *tlink[8];
- unsigned long now = jiffies;
- unsigned long index = 0;
- int i, ret;
+ struct rb_root *root = &cifs_sb->tlink_tree;
+ struct rb_node *node = rb_first(root);
+ struct rb_node *tmp;
+ struct tcon_link *tlink;
- do {
- spin_lock(&cifs_sb->tlink_tree_lock);
- ret = radix_tree_gang_lookup(&cifs_sb->tlink_tree,
- (void **)tlink, index,
- ARRAY_SIZE(tlink));
- /* increment index for next pass */
- if (ret > 0)
- index = tlink[ret - 1]->tl_index + 1;
- for (i = 0; i < ret; i++) {
- if (test_bit(TCON_LINK_MASTER, &tlink[i]->tl_flags) ||
- atomic_read(&tlink[i]->tl_count) != 0 ||
- time_after(tlink[i]->tl_time + TLINK_IDLE_EXPIRE,
- now)) {
- tlink[i] = NULL;
- continue;
- }
- cifs_get_tlink(tlink[i]);
- clear_bit(TCON_LINK_IN_TREE, &tlink[i]->tl_flags);
- radix_tree_delete(&cifs_sb->tlink_tree,
- tlink[i]->tl_index);
- }
- spin_unlock(&cifs_sb->tlink_tree_lock);
+ /*
+ * Because we drop the spinlock in the loop in order to put the tlink
+ * it's not guarded against removal of links from the tree. The only
+ * places that remove entries from the tree are this function and
+ * umounts. Because this function is non-reentrant and is canceled
+ * before umount can proceed, this is safe.
+ */
+ spin_lock(&cifs_sb->tlink_tree_lock);
+ node = rb_first(root);
+ while (node != NULL) {
+ tmp = node;
+ node = rb_next(tmp);
+ tlink = rb_entry(tmp, struct tcon_link, tl_rbnode);
+
+ if (test_bit(TCON_LINK_MASTER, &tlink->tl_flags) ||
+ atomic_read(&tlink->tl_count) != 0 ||
+ time_after(tlink->tl_time + TLINK_IDLE_EXPIRE, jiffies))
+ continue;
- for (i = 0; i < ret; i++) {
- if (tlink[i] != NULL)
- cifs_put_tlink(tlink[i]);
- }
- } while (ret != 0);
+ cifs_get_tlink(tlink);
+ clear_bit(TCON_LINK_IN_TREE, &tlink->tl_flags);
+ rb_erase(tmp, root);
+
+ spin_unlock(&cifs_sb->tlink_tree_lock);
+ cifs_put_tlink(tlink);
+ spin_lock(&cifs_sb->tlink_tree_lock);
+ }
+ spin_unlock(&cifs_sb->tlink_tree_lock);
queue_delayed_work(system_nrt_wq, &cifs_sb->prune_tlinks,
TLINK_IDLE_EXPIRE);
rc = cifs_get_inode_info(&inode, full_path, buf, inode->i_sb,
xid, NULL);
- if ((oplock & 0xF) == OPLOCK_EXCLUSIVE) {
- pCifsInode->clientCanCacheAll = true;
- pCifsInode->clientCanCacheRead = true;
- cFYI(1, "Exclusive Oplock granted on inode %p", inode);
- } else if ((oplock & 0xF) == OPLOCK_READ)
- pCifsInode->clientCanCacheRead = true;
+ cifs_set_oplock_level(pCifsInode, oplock);
return rc;
}
list_add_tail(&pCifsFile->flist, &pCifsInode->openFileList);
spin_unlock(&cifs_file_list_lock);
- if ((oplock & 0xF) == OPLOCK_EXCLUSIVE) {
- pCifsInode->clientCanCacheAll = true;
- pCifsInode->clientCanCacheRead = true;
- cFYI(1, "Exclusive Oplock inode %p", inode);
- } else if ((oplock & 0xF) == OPLOCK_READ)
- pCifsInode->clientCanCacheRead = true;
+ cifs_set_oplock_level(pCifsInode, oplock);
file->private_data = pCifsFile;
return pCifsFile;
*/
void cifsFileInfo_put(struct cifsFileInfo *cifs_file)
{
+ struct inode *inode = cifs_file->dentry->d_inode;
struct cifsTconInfo *tcon = tlink_tcon(cifs_file->tlink);
- struct cifsInodeInfo *cifsi = CIFS_I(cifs_file->dentry->d_inode);
+ struct cifsInodeInfo *cifsi = CIFS_I(inode);
struct cifsLockInfo *li, *tmp;
spin_lock(&cifs_file_list_lock);
if (list_empty(&cifsi->openFileList)) {
cFYI(1, "closing last open instance for inode %p",
cifs_file->dentry->d_inode);
- cifsi->clientCanCacheRead = false;
- cifsi->clientCanCacheAll = false;
+ cifs_set_oplock_level(cifsi, 0);
}
spin_unlock(&cifs_file_list_lock);
rc = filemap_write_and_wait(inode->i_mapping);
mapping_set_error(inode->i_mapping, rc);
- pCifsInode->clientCanCacheAll = false;
- pCifsInode->clientCanCacheRead = false;
if (tcon->unix_ext)
rc = cifs_get_inode_info_unix(&inode,
full_path, inode->i_sb, xid);
invalidate the current end of file on the server
we can not go to the server to get the new inod
info */
- if ((oplock & 0xF) == OPLOCK_EXCLUSIVE) {
- pCifsInode->clientCanCacheAll = true;
- pCifsInode->clientCanCacheRead = true;
- cFYI(1, "Exclusive Oplock granted on inode %p",
- pCifsFile->dentry->d_inode);
- } else if ((oplock & 0xF) == OPLOCK_READ) {
- pCifsInode->clientCanCacheRead = true;
- pCifsInode->clientCanCacheAll = false;
- } else {
- pCifsInode->clientCanCacheRead = false;
- pCifsInode->clientCanCacheAll = false;
- }
+
+ cifs_set_oplock_level(pCifsInode, oplock);
+
cifs_relock_file(pCifsFile);
reopen_error_exit:
cifs_sb = CIFS_SB(file->f_path.dentry->d_sb);
tcon = tlink_tcon(((struct cifsFileInfo *)file->private_data)->tlink);
-
- if (file->private_data == NULL) {
- rc = -EBADF;
- FreeXid(xid);
- return rc;
- }
netfid = ((struct cifsFileInfo *)file->private_data)->netfid;
if ((tcon->ses->capabilities & CAP_UNIX) &&
ssize_t cifs_user_write(struct file *file, const char __user *write_data,
size_t write_size, loff_t *poffset)
{
+ struct inode *inode = file->f_path.dentry->d_inode;
int rc = 0;
unsigned int bytes_written = 0;
unsigned int total_written;
struct cifsTconInfo *pTcon;
int xid, long_op;
struct cifsFileInfo *open_file;
- struct cifsInodeInfo *cifsi = CIFS_I(file->f_path.dentry->d_inode);
+ struct cifsInodeInfo *cifsi = CIFS_I(inode);
cifs_sb = CIFS_SB(file->f_path.dentry->d_sb);
cifs_stats_bytes_written(pTcon, total_written);
- /* since the write may have blocked check these pointers again */
- if ((file->f_path.dentry) && (file->f_path.dentry->d_inode)) {
- struct inode *inode = file->f_path.dentry->d_inode;
/* Do not update local mtime - server will set its actual value on write
- * inode->i_ctime = inode->i_mtime =
- * current_fs_time(inode->i_sb);*/
- if (total_written > 0) {
- spin_lock(&inode->i_lock);
- if (*poffset > file->f_path.dentry->d_inode->i_size)
- i_size_write(file->f_path.dentry->d_inode,
- *poffset);
- spin_unlock(&inode->i_lock);
- }
- mark_inode_dirty_sync(file->f_path.dentry->d_inode);
+ * inode->i_ctime = inode->i_mtime =
+ * current_fs_time(inode->i_sb);*/
+ if (total_written > 0) {
+ spin_lock(&inode->i_lock);
+ if (*poffset > inode->i_size)
+ i_size_write(inode, *poffset);
+ spin_unlock(&inode->i_lock);
}
+ mark_inode_dirty_sync(inode);
+
FreeXid(xid);
return total_written;
}
bool fsuid_only)
{
struct cifsFileInfo *open_file;
- struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_inode->vfs_inode.i_sb);
+ struct cifs_sb_info *cifs_sb;
bool any_available = false;
int rc;
return NULL;
}
+ cifs_sb = CIFS_SB(cifs_inode->vfs_inode.i_sb);
+
/* only filter by fsuid on multiuser mounts */
if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER))
fsuid_only = false;
setattr_copy(inode, attrs);
mark_inode_dirty(inode);
- return 0;
cifs_setattr_exit:
kfree(full_path);
struct cifs_sb_info *cifs_sb;
#ifdef CONFIG_CIFS_POSIX
struct cifsFileInfo *pSMBFile = filep->private_data;
- struct cifsTconInfo *tcon = tlink_tcon(pSMBFile->tlink);
+ struct cifsTconInfo *tcon;
__u64 ExtAttrBits = 0;
__u64 ExtAttrMask = 0;
- __u64 caps = le64_to_cpu(tcon->fsUnixInfo.Capability);
+ __u64 caps;
#endif /* CONFIG_CIFS_POSIX */
xid = GetXid();
break;
#ifdef CONFIG_CIFS_POSIX
case FS_IOC_GETFLAGS:
+ if (pSMBFile == NULL)
+ break;
+ tcon = tlink_tcon(pSMBFile->tlink);
+ caps = le64_to_cpu(tcon->fsUnixInfo.Capability);
if (CIFS_UNIX_EXTATTR_CAP & caps) {
- if (pSMBFile == NULL)
- break;
rc = CIFSGetExtAttr(xid, tcon, pSMBFile->netfid,
&ExtAttrBits, &ExtAttrMask);
if (rc == 0)
break;
case FS_IOC_SETFLAGS:
+ if (pSMBFile == NULL)
+ break;
+ tcon = tlink_tcon(pSMBFile->tlink);
+ caps = le64_to_cpu(tcon->fsUnixInfo.Capability);
if (CIFS_UNIX_EXTATTR_CAP & caps) {
if (get_user(ExtAttrBits, (int __user *)arg)) {
rc = -EFAULT;
break;
}
- if (pSMBFile == NULL)
- break;
/* rc= CIFSGetExtAttr(xid,tcon,pSMBFile->netfid,
extAttrBits, &ExtAttrMask);*/
}
cFYI(1, "file id match, oplock break");
pCifsInode = CIFS_I(netfile->dentry->d_inode);
- pCifsInode->clientCanCacheAll = false;
- if (pSMB->OplockLevel == 0)
- pCifsInode->clientCanCacheRead = false;
+ cifs_set_oplock_level(pCifsInode,
+ pSMB->OplockLevel);
/*
* cifs_oplock_break_put() can't be called
* from here. Get reference after queueing
cifs_sb_master_tcon(cifs_sb)->treeName);
}
}
+
+void cifs_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock)
+{
+ oplock &= 0xF;
+
+ if (oplock == OPLOCK_EXCLUSIVE) {
+ cinode->clientCanCacheAll = true;
+ cinode->clientCanCacheRead = true;
+ cFYI(1, "Exclusive Oplock granted on inode %p",
+ &cinode->vfs_inode);
+ } else if (oplock == OPLOCK_READ) {
+ cinode->clientCanCacheAll = false;
+ cinode->clientCanCacheRead = true;
+ cFYI(1, "Level II Oplock granted on inode %p",
+ &cinode->vfs_inode);
+ } else {
+ cinode->clientCanCacheAll = false;
+ cinode->clientCanCacheRead = false;
+ }
+}
struct ext4_io_page {
struct page *p_page;
- int p_count;
+ atomic_t p_count;
};
#define MAX_IO_PAGES 128
spinlock_t i_completed_io_lock;
/* current io_end structure for async DIO write*/
ext4_io_end_t *cur_aio_dio;
+ atomic_t i_ioend_count; /* Number of outstanding io_end structs */
/*
* Transactions that contain inode's metadata needed to complete
/* page-io.c */
extern int __init ext4_init_pageio(void);
extern void ext4_exit_pageio(void);
+extern void ext4_ioend_wait(struct inode *);
extern void ext4_free_io_end(ext4_io_end_t *io);
extern ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags);
extern int ext4_end_io_nolock(ext4_io_end_t *io);
static inline int ext4_begin_ordered_truncate(struct inode *inode,
loff_t new_size)
{
+ trace_ext4_begin_ordered_truncate(inode, new_size);
return jbd2_journal_begin_ordered_truncate(
EXT4_SB(inode->i_sb)->s_journal,
&EXT4_I(inode)->jinode,
handle_t *handle;
int err;
+ trace_ext4_evict_inode(inode);
if (inode->i_nlink) {
truncate_inode_pages(&inode->i_data, 0);
goto no_delete;
* will return the blocks that include the delayed allocation
* blocks for this file.
*/
- spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
delalloc_blocks = EXT4_I(inode)->i_reserved_data_blocks;
- spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
stat->blocks += (delalloc_blocks << inode->i_sb->s_blocksize_bits)>>9;
return 0;
int err, ret;
might_sleep();
+ trace_ext4_mark_inode_dirty(inode, _RET_IP_);
err = ext4_reserve_inode_write(handle, inode, &iloc);
if (ext4_handle_valid(handle) &&
EXT4_I(inode)->i_extra_isize < sbi->s_want_extra_isize &&
* with group lock held. generate_buddy look at
* them with group lock_held
*/
- if (test_opt(sb, DISCARD))
- ext4_issue_discard(sb, block_group, bit, count);
ext4_lock_group(sb, block_group);
mb_clear_bits(bitmap_bh->b_data, bit, count);
mb_free_blocks(inode, &e4b, bit, count);
static struct kmem_cache *io_page_cachep, *io_end_cachep;
+#define WQ_HASH_SZ 37
+#define to_ioend_wq(v) (&ioend_wq[((unsigned long)v) % WQ_HASH_SZ])
+static wait_queue_head_t ioend_wq[WQ_HASH_SZ];
+
int __init ext4_init_pageio(void)
{
+ int i;
+
io_page_cachep = KMEM_CACHE(ext4_io_page, SLAB_RECLAIM_ACCOUNT);
if (io_page_cachep == NULL)
return -ENOMEM;
kmem_cache_destroy(io_page_cachep);
return -ENOMEM;
}
+ for (i = 0; i < WQ_HASH_SZ; i++)
+ init_waitqueue_head(&ioend_wq[i]);
return 0;
}
kmem_cache_destroy(io_page_cachep);
}
+void ext4_ioend_wait(struct inode *inode)
+{
+ wait_queue_head_t *wq = to_ioend_wq(inode);
+
+ wait_event(*wq, (atomic_read(&EXT4_I(inode)->i_ioend_count) == 0));
+}
+
+static void put_io_page(struct ext4_io_page *io_page)
+{
+ if (atomic_dec_and_test(&io_page->p_count)) {
+ end_page_writeback(io_page->p_page);
+ put_page(io_page->p_page);
+ kmem_cache_free(io_page_cachep, io_page);
+ }
+}
+
void ext4_free_io_end(ext4_io_end_t *io)
{
int i;
+ wait_queue_head_t *wq;
BUG_ON(!io);
if (io->page)
put_page(io->page);
- for (i = 0; i < io->num_io_pages; i++) {
- if (--io->pages[i]->p_count == 0) {
- struct page *page = io->pages[i]->p_page;
-
- end_page_writeback(page);
- put_page(page);
- kmem_cache_free(io_page_cachep, io->pages[i]);
- }
- }
+ for (i = 0; i < io->num_io_pages; i++)
+ put_io_page(io->pages[i]);
io->num_io_pages = 0;
- iput(io->inode);
+ wq = to_ioend_wq(io->inode);
+ if (atomic_dec_and_test(&EXT4_I(io->inode)->i_ioend_count) &&
+ waitqueue_active(wq))
+ wake_up_all(wq);
kmem_cache_free(io_end_cachep, io);
}
io = kmem_cache_alloc(io_end_cachep, flags);
if (io) {
memset(io, 0, sizeof(*io));
- io->inode = igrab(inode);
- BUG_ON(!io->inode);
+ atomic_inc(&EXT4_I(inode)->i_ioend_count);
+ io->inode = inode;
INIT_WORK(&io->work, ext4_end_io_work);
INIT_LIST_HEAD(&io->list);
}
struct workqueue_struct *wq;
struct inode *inode;
unsigned long flags;
- ext4_fsblk_t err_block;
int i;
BUG_ON(!io_end);
- inode = io_end->inode;
bio->bi_private = NULL;
bio->bi_end_io = NULL;
if (test_bit(BIO_UPTODATE, &bio->bi_flags))
error = 0;
- err_block = bio->bi_sector >> (inode->i_blkbits - 9);
bio_put(bio);
- if (!(inode->i_sb->s_flags & MS_ACTIVE)) {
- pr_err("sb umounted, discard end_io request for inode %lu\n",
- io_end->inode->i_ino);
- ext4_free_io_end(io_end);
- return;
- }
-
- if (error) {
- io_end->flag |= EXT4_IO_END_ERROR;
- ext4_warning(inode->i_sb, "I/O error writing to inode %lu "
- "(offset %llu size %ld starting block %llu)",
- inode->i_ino,
- (unsigned long long) io_end->offset,
- (long) io_end->size,
- (unsigned long long) err_block);
- }
-
for (i = 0; i < io_end->num_io_pages; i++) {
struct page *page = io_end->pages[i]->p_page;
struct buffer_head *bh, *head;
} while (bh != head);
}
- if (--io_end->pages[i]->p_count == 0) {
- struct page *page = io_end->pages[i]->p_page;
-
- end_page_writeback(page);
- put_page(page);
- kmem_cache_free(io_page_cachep, io_end->pages[i]);
- }
+ put_io_page(io_end->pages[i]);
/*
* If this is a partial write which happened to make
if (!partial_write)
SetPageUptodate(page);
}
-
io_end->num_io_pages = 0;
+ inode = io_end->inode;
+
+ if (error) {
+ io_end->flag |= EXT4_IO_END_ERROR;
+ ext4_warning(inode->i_sb, "I/O error writing to inode %lu "
+ "(offset %llu size %ld starting block %llu)",
+ inode->i_ino,
+ (unsigned long long) io_end->offset,
+ (long) io_end->size,
+ (unsigned long long)
+ bio->bi_sector >> (inode->i_blkbits - 9));
+ }
/* Add the io_end to per-inode completed io list*/
spin_lock_irqsave(&EXT4_I(inode)->i_completed_io_lock, flags);
bio->bi_private = io->io_end = io_end;
bio->bi_end_io = ext4_end_bio;
- io_end->inode = inode;
io_end->offset = (page->index << PAGE_CACHE_SHIFT) + bh_offset(bh);
io->io_bio = bio;
if ((io_end->num_io_pages == 0) ||
(io_end->pages[io_end->num_io_pages-1] != io_page)) {
io_end->pages[io_end->num_io_pages++] = io_page;
- io_page->p_count++;
+ atomic_inc(&io_page->p_count);
}
return 0;
}
return -ENOMEM;
}
io_page->p_page = page;
- io_page->p_count = 0;
+ atomic_set(&io_page->p_count, 1);
get_page(page);
for (bh = head = page_buffers(page), block_start = 0;
* PageWriteback bit from the page to prevent the system from
* wedging later on.
*/
- if (io_page->p_count == 0) {
- put_page(page);
- end_page_writeback(page);
- kmem_cache_free(io_page_cachep, io_page);
- }
+ put_io_page(io_page);
return ret;
}
ei->cur_aio_dio = NULL;
ei->i_sync_tid = 0;
ei->i_datasync_tid = 0;
+ atomic_set(&ei->i_ioend_count, 0);
return &ei->vfs_inode;
}
+static int ext4_drop_inode(struct inode *inode)
+{
+ int drop = generic_drop_inode(inode);
+
+ trace_ext4_drop_inode(inode, drop);
+ return drop;
+}
+
static void ext4_destroy_inode(struct inode *inode)
{
+ ext4_ioend_wait(inode);
if (!list_empty(&(EXT4_I(inode)->i_orphan))) {
ext4_msg(inode->i_sb, KERN_ERR,
"Inode %lu (%p): orphan list check failed!",
.destroy_inode = ext4_destroy_inode,
.write_inode = ext4_write_inode,
.dirty_inode = ext4_dirty_inode,
+ .drop_inode = ext4_drop_inode,
.evict_inode = ext4_evict_inode,
.put_super = ext4_put_super,
.sync_fs = ext4_sync_fs,
.destroy_inode = ext4_destroy_inode,
.write_inode = ext4_write_inode,
.dirty_inode = ext4_dirty_inode,
+ .drop_inode = ext4_drop_inode,
.evict_inode = ext4_evict_inode,
.write_super = ext4_write_super,
.put_super = ext4_put_super,
struct ext4_li_request *elr;
unsigned long next_wakeup;
DEFINE_WAIT(wait);
- int ret;
BUG_ON(NULL == eli);
elr = list_entry(pos, struct ext4_li_request,
lr_request);
- if (time_after_eq(jiffies, elr->lr_next_sched))
- ret = ext4_run_li_request(elr);
-
- if (ret) {
- ret = 0;
- ext4_remove_li_request(elr);
- continue;
+ if (time_after_eq(jiffies, elr->lr_next_sched)) {
+ if (ext4_run_li_request(elr) != 0) {
+ /* error, remove the lazy_init job */
+ ext4_remove_li_request(elr);
+ continue;
+ }
}
if (time_before(elr->lr_next_sched, next_wakeup))
if (freezing(current))
refrigerator();
- if (time_after_eq(jiffies, next_wakeup)) {
+ if ((time_after_eq(jiffies, next_wakeup)) ||
+ (MAX_JIFFY_OFFSET == next_wakeup)) {
cond_resched();
continue;
}
get_random_bytes(&sbi->s_next_generation, sizeof(u32));
spin_lock_init(&sbi->s_next_gen_lock);
+ err = percpu_counter_init(&sbi->s_freeblocks_counter,
+ ext4_count_free_blocks(sb));
+ if (!err) {
+ err = percpu_counter_init(&sbi->s_freeinodes_counter,
+ ext4_count_free_inodes(sb));
+ }
+ if (!err) {
+ err = percpu_counter_init(&sbi->s_dirs_counter,
+ ext4_count_dirs(sb));
+ }
+ if (!err) {
+ err = percpu_counter_init(&sbi->s_dirtyblocks_counter, 0);
+ }
+ if (err) {
+ ext4_msg(sb, KERN_ERR, "insufficient memory");
+ goto failed_mount3;
+ }
+
sbi->s_stripe = ext4_get_stripe_size(sbi);
sbi->s_max_writeback_mb_bump = 128;
}
set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
-no_journal:
- err = percpu_counter_init(&sbi->s_freeblocks_counter,
- ext4_count_free_blocks(sb));
- if (!err)
- err = percpu_counter_init(&sbi->s_freeinodes_counter,
- ext4_count_free_inodes(sb));
- if (!err)
- err = percpu_counter_init(&sbi->s_dirs_counter,
- ext4_count_dirs(sb));
- if (!err)
- err = percpu_counter_init(&sbi->s_dirtyblocks_counter, 0);
- if (err) {
- ext4_msg(sb, KERN_ERR, "insufficient memory");
- goto failed_mount_wq;
- }
+ /*
+ * The journal may have updated the bg summary counts, so we
+ * need to update the global counters.
+ */
+ percpu_counter_set(&sbi->s_freeblocks_counter,
+ ext4_count_free_blocks(sb));
+ percpu_counter_set(&sbi->s_freeinodes_counter,
+ ext4_count_free_inodes(sb));
+ percpu_counter_set(&sbi->s_dirs_counter,
+ ext4_count_dirs(sb));
+ percpu_counter_set(&sbi->s_dirtyblocks_counter, 0);
+no_journal:
EXT4_SB(sb)->dio_unwritten_wq = create_workqueue("ext4-dio-unwritten");
if (!EXT4_SB(sb)->dio_unwritten_wq) {
printk(KERN_ERR "EXT4-fs: failed to create DIO workqueue\n");
jbd2_journal_destroy(sbi->s_journal);
sbi->s_journal = NULL;
}
- percpu_counter_destroy(&sbi->s_freeblocks_counter);
- percpu_counter_destroy(&sbi->s_freeinodes_counter);
- percpu_counter_destroy(&sbi->s_dirs_counter);
- percpu_counter_destroy(&sbi->s_dirtyblocks_counter);
failed_mount3:
if (sbi->s_flex_groups) {
if (is_vmalloc_addr(sbi->s_flex_groups))
else
kfree(sbi->s_flex_groups);
}
+ percpu_counter_destroy(&sbi->s_freeblocks_counter);
+ percpu_counter_destroy(&sbi->s_freeinodes_counter);
+ percpu_counter_destroy(&sbi->s_dirs_counter);
+ percpu_counter_destroy(&sbi->s_dirtyblocks_counter);
failed_mount2:
for (i = 0; i < db_count; i++)
brelse(sbi->s_group_desc[i]);
else
es->s_kbytes_written =
cpu_to_le64(EXT4_SB(sb)->s_kbytes_written);
- if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeblocks_counter))
- ext4_free_blocks_count_set(es, percpu_counter_sum_positive(
- &EXT4_SB(sb)->s_freeblocks_counter));
- if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeinodes_counter))
- es->s_free_inodes_count =
- cpu_to_le32(percpu_counter_sum_positive(
- &EXT4_SB(sb)->s_freeinodes_counter));
+ ext4_free_blocks_count_set(es, percpu_counter_sum_positive(
+ &EXT4_SB(sb)->s_freeblocks_counter));
+ es->s_free_inodes_count =
+ cpu_to_le32(percpu_counter_sum_positive(
+ &EXT4_SB(sb)->s_freeinodes_counter));
sb->s_dirt = 0;
BUFFER_TRACE(sbh, "marking dirty");
mark_buffer_dirty(sbh);
static int ext4_quota_off(struct super_block *sb, int type)
{
- /* Force all delayed allocation blocks to be allocated */
- if (test_opt(sb, DELALLOC)) {
- down_read(&sb->s_umount);
+ /* Force all delayed allocation blocks to be allocated.
+ * Caller already holds s_umount sem */
+ if (test_opt(sb, DELALLOC))
sync_filesystem(sb);
- up_read(&sb->s_umount);
- }
return dquot_quota_off(sb, type);
}
static int gfs2_setlease(struct file *file, long arg, struct file_lock **fl)
{
- if (arg != F_UNLCK)
- locks_free_lock(*fl);
return -EINVAL;
}
#ifdef DEBUG_LOCKS
printk("lock creation\n");
#endif
- down(&hpfs_sb(s)->hpfs_creation_de);
+ mutex_lock(&hpfs_sb(s)->hpfs_creation_de);
}
void hpfs_unlock_creation(struct super_block *s)
#ifdef DEBUG_LOCKS
printk("unlock creation\n");
#endif
- up(&hpfs_sb(s)->hpfs_creation_de);
+ mutex_unlock(&hpfs_sb(s)->hpfs_creation_de);
}
/* Map a sector into a buffer and return pointers to it and to the buffer. */
unsigned *sb_bmp_dir; /* main bitmap directory */
unsigned sb_c_bitmap; /* current bitmap */
unsigned sb_max_fwd_alloc; /* max forwad allocation */
- struct semaphore hpfs_creation_de; /* when creating dirents, nobody else
+ struct mutex hpfs_creation_de; /* when creating dirents, nobody else
can alloc blocks */
/*unsigned sb_mounting : 1;*/
int sb_timeshift;
sbi->sb_bmp_dir = NULL;
sbi->sb_cp_table = NULL;
- init_MUTEX(&sbi->hpfs_creation_de);
+ mutex_init(&sbi->hpfs_creation_de);
uid = current_uid();
gid = current_gid();
if (creat_flags == HUGETLB_SHMFS_INODE && !can_do_hugetlb_shm()) {
*user = current_user();
if (user_shm_lock(size, *user)) {
- WARN_ONCE(1,
- "Using mlock ulimits for SHM_HUGETLB deprecated\n");
+ printk_once(KERN_WARNING "Using mlock ulimits for SHM_HUGETLB is deprecated\n");
} else {
*user = NULL;
return ERR_PTR(-EPERM);
*/
#define JBD2_MAX_SLABS 8
static struct kmem_cache *jbd2_slab[JBD2_MAX_SLABS];
-static DECLARE_MUTEX(jbd2_slab_create_sem);
static const char *jbd2_slab_names[JBD2_MAX_SLABS] = {
"jbd2_1k", "jbd2_2k", "jbd2_4k", "jbd2_8k",
static int jbd2_journal_create_slab(size_t size)
{
+ static DEFINE_MUTEX(jbd2_slab_create_mutex);
int i = order_base_2(size) - 10;
size_t slab_size;
if (unlikely(i < 0))
i = 0;
- down(&jbd2_slab_create_sem);
+ mutex_lock(&jbd2_slab_create_mutex);
if (jbd2_slab[i]) {
- up(&jbd2_slab_create_sem);
+ mutex_unlock(&jbd2_slab_create_mutex);
return 0; /* Already created */
}
slab_size = 1 << (i+10);
jbd2_slab[i] = kmem_cache_create(jbd2_slab_names[i], slab_size,
slab_size, 0, NULL);
- up(&jbd2_slab_create_sem);
+ mutex_unlock(&jbd2_slab_create_mutex);
if (!jbd2_slab[i]) {
printk(KERN_EMERG "JBD2: no memory for jbd2_slab cache\n");
return -ENOMEM;
fl->fl_ops->fl_copy_lock(new, fl);
new->fl_ops = fl->fl_ops;
}
- if (fl->fl_lmops) {
- if (fl->fl_lmops->fl_copy_lock)
- fl->fl_lmops->fl_copy_lock(new, fl);
+ if (fl->fl_lmops)
new->fl_lmops = fl->fl_lmops;
- }
}
/*
goto out;
if (my_before != NULL) {
- *flp = *my_before;
error = lease->fl_lmops->fl_change(my_before, arg);
+ if (!error)
+ *flp = *my_before;
goto out;
}
return 0;
out:
- if (arg != F_UNLCK)
- locks_free_lock(lease);
return error;
}
EXPORT_SYMBOL(generic_setlease);
static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg)
{
- struct file_lock *fl;
+ struct file_lock *fl, *ret;
struct fasync_struct *new;
- struct inode *inode = filp->f_path.dentry->d_inode;
int error;
fl = lease_alloc(filp, arg);
locks_free_lock(fl);
return -ENOMEM;
}
+ ret = fl;
lock_flocks();
- error = __vfs_setlease(filp, arg, &fl);
- if (error)
- goto out_unlock;
+ error = __vfs_setlease(filp, arg, &ret);
+ if (error) {
+ unlock_flocks();
+ locks_free_lock(fl);
+ goto out_free_fasync;
+ }
+ if (ret != fl)
+ locks_free_lock(fl);
/*
* fasync_insert_entry() returns the old entry if any.
* inserted it into the fasync list. Clear new so that
* we don't release it here.
*/
- if (!fasync_insert_entry(fd, filp, &fl->fl_fasync, new))
+ if (!fasync_insert_entry(fd, filp, &ret->fl_fasync, new))
new = NULL;
- if (error < 0) {
- /* remove lease just inserted by setlease */
- fl->fl_type = F_UNLCK | F_INPROGRESS;
- fl->fl_break_time = jiffies - 10;
- time_out_leases(inode);
- goto out_unlock;
- }
-
error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0);
-out_unlock:
unlock_flocks();
+
+out_free_fasync:
if (new)
fasync_free(new);
return error;
/* dev_mtd.c */
#ifdef CONFIG_MTD
-int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr)
+int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr);
#else
static inline int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr)
{
dprintk("NFS: setlease(%s/%s, arg=%ld)\n",
file->f_path.dentry->d_parent->d_name.name,
file->f_path.dentry->d_name.name, arg);
- if (arg != F_UNLCK)
- locks_free_lock(*fl);
return -EINVAL;
}
spin_unlock(&clp->cl_lock);
}
-static void nfsd4_register_conn(struct nfsd4_conn *conn)
+static int nfsd4_register_conn(struct nfsd4_conn *conn)
{
conn->cn_xpt_user.callback = nfsd4_conn_lost;
- register_xpt_user(conn->cn_xprt, &conn->cn_xpt_user);
+ return register_xpt_user(conn->cn_xprt, &conn->cn_xpt_user);
}
static __be32 nfsd4_new_conn(struct svc_rqst *rqstp, struct nfsd4_session *ses)
{
struct nfsd4_conn *conn;
u32 flags = NFS4_CDFC4_FORE;
+ int ret;
if (ses->se_flags & SESSION4_BACK_CHAN)
flags |= NFS4_CDFC4_BACK;
if (!conn)
return nfserr_jukebox;
nfsd4_hash_conn(conn, ses);
- nfsd4_register_conn(conn);
+ ret = nfsd4_register_conn(conn);
+ if (ret)
+ /* oops; xprt is already down: */
+ nfsd4_conn_lost(&conn->cn_xpt_user);
return nfs_ok;
}
{
struct nfs4_client *clp = ses->se_client;
struct nfsd4_conn *c;
+ int ret;
spin_lock(&clp->cl_lock);
c = __nfsd4_find_conn(new->cn_xprt, ses);
}
__nfsd4_hash_conn(new, ses);
spin_unlock(&clp->cl_lock);
- nfsd4_register_conn(new);
+ ret = nfsd4_register_conn(new);
+ if (ret)
+ /* oops; xprt is already down: */
+ nfsd4_conn_lost(&new->cn_xpt_user);
return;
}
if ((status = vfs_setlease(fl->fl_file, fl->fl_type, &fl))) {
dprintk("NFSD: setlease failed [%d], no delegation\n", status);
dp->dl_flock = NULL;
+ locks_free_lock(fl);
unhash_delegation(dp);
flag = NFS4_OPEN_DELEGATE_NONE;
goto out;
static struct dentry *openprom_mount(struct file_system_type *fs_type,
int flags, const char *dev_name, void *data)
{
- return mount_single(fs_type, flags, data, openprom_fill_super)
+ return mount_single(fs_type, flags, data, openprom_fill_super);
}
static struct file_system_type openprom_fs_type = {
uptodate = 0;
/*
- * A hole may still be marked uptodate because discard_buffer
- * leaves the flag set.
+ * set_page_dirty dirties all buffers in a page, independent
+ * of their state. The dirty state however is entirely
+ * meaningless for holes (!mapped && uptodate), so skip
+ * buffers covering holes here.
*/
if (!buffer_mapped(bh) && buffer_uptodate(bh)) {
- ASSERT(!buffer_dirty(bh));
imap_valid = 0;
continue;
}
INIT_LIST_HEAD(list);
spin_lock(dwlk);
list_for_each_entry_safe(bp, n, dwq, b_list) {
- trace_xfs_buf_delwri_split(bp, _RET_IP_);
ASSERT(bp->b_flags & XBF_DELWRI);
if (!XFS_BUF_ISPINNED(bp) && !xfs_buf_cond_lock(bp)) {
_XBF_RUN_QUEUES);
bp->b_flags |= XBF_WRITE;
list_move_tail(&bp->b_list, list);
+ trace_xfs_buf_delwri_split(bp, _RET_IP_);
} else
skipped++;
}
if (IS_ERR(dentry))
return PTR_ERR(dentry);
- kbuf = kmalloc(al_hreq.buflen, GFP_KERNEL);
+ kbuf = kzalloc(al_hreq.buflen, GFP_KERNEL);
if (!kbuf)
goto out_dput;
inode->i_state = I_NEW;
inode_sb_list_add(inode);
- insert_inode_hash(inode);
+ /* make the inode look hashed for the writeback code */
+ hlist_add_fake(&inode->i_hash);
inode->i_mode = ip->i_d.di_mode;
inode->i_nlink = ip->i_d.di_nlink;
mp->m_qflags &= ~XFS_OQUOTA_ENFD;
} else if (!strcmp(this_char, MNTOPT_DELAYLOG)) {
mp->m_flags |= XFS_MOUNT_DELAYLOG;
- cmn_err(CE_WARN,
- "Enabling EXPERIMENTAL delayed logging feature "
- "- use at your own risk.\n");
} else if (!strcmp(this_char, MNTOPT_NODELAYLOG)) {
mp->m_flags &= ~XFS_MOUNT_DELAYLOG;
} else if (!strcmp(this_char, "ihashsize")) {
if (trylock) {
if (!mutex_trylock(&pag->pag_ici_reclaim_lock)) {
skipped++;
+ xfs_perag_put(pag);
continue;
}
first_index = pag->pag_ici_reclaim_cursor;
* If the file's parent directory is known, take its iolock in exclusive
* mode to prevent two sibling files from racing each other to migrate
* themselves and their parent to different AGs.
+ *
+ * Note that we lock the parent directory iolock inside the child
+ * iolock here. That's fine as we never hold both parent and child
+ * iolock in any other place. This is different from the ilock,
+ * which requires locking of the child after the parent for namespace
+ * operations.
*/
if (pip)
- xfs_ilock(pip, XFS_IOLOCK_EXCL);
+ xfs_ilock(pip, XFS_IOLOCK_EXCL | XFS_IOLOCK_PARENT);
/*
* A new AG needs to be found for the file. If the file's parent
pag = radix_tree_delete(&mp->m_perag_tree, agno);
spin_unlock(&mp->m_perag_lock);
ASSERT(pag);
+ ASSERT(atomic_read(&pag->pag_ref) == 0);
call_rcu(&pag->rcu_head, __xfs_free_perag);
}
}
#define xfs_trans_mod_dquot_byino(tp, ip, fields, delta)
#define xfs_trans_apply_dquot_deltas(tp)
#define xfs_trans_unreserve_and_mod_dquots(tp)
-#define xfs_trans_reserve_quota_nblks(tp, ip, nblks, ninos, flags) (0)
-#define xfs_trans_reserve_quota_bydquots(tp, mp, u, g, nb, ni, fl) (0)
+static inline int xfs_trans_reserve_quota_nblks(struct xfs_trans *tp,
+ struct xfs_inode *ip, long nblks, long ninos, uint flags)
+{
+ return 0;
+}
+static inline int xfs_trans_reserve_quota_bydquots(struct xfs_trans *tp,
+ struct xfs_mount *mp, struct xfs_dquot *udqp,
+ struct xfs_dquot *gdqp, long nblks, long nions, uint flags)
+{
+ return 0;
+}
#define xfs_qm_vop_create_dqattach(tp, ip, u, g)
#define xfs_qm_vop_rename_dqattach(it) (0)
#define xfs_qm_vop_chown(tp, ip, old, new) (NULL)
#define xfs_qm_dqdetach(ip)
#define xfs_qm_dqrele(d)
#define xfs_qm_statvfs(ip, s)
-#define xfs_qm_sync(mp, fl) (0)
+static inline int xfs_qm_sync(struct xfs_mount *mp, int flags)
+{
+ return 0;
+}
#define xfs_qm_newmount(mp, a, b) (0)
#define xfs_qm_mount_quotas(mp)
#define xfs_qm_unmount(mp)
-#define xfs_qm_unmount_quotas(mp) (0)
+#define xfs_qm_unmount_quotas(mp)
#endif /* CONFIG_XFS_QUOTA */
#define xfs_trans_unreserve_quota_nblks(tp, ip, nblks, ninos, flags) \
int st_blksize; /* Optimal block size for I/O. */
int __pad2;
long st_blocks; /* Number 512-byte blocks allocated. */
- int st_atime; /* Time of last access. */
- unsigned int st_atime_nsec;
- int st_mtime; /* Time of last modification. */
- unsigned int st_mtime_nsec;
- int st_ctime; /* Time of last status change. */
- unsigned int st_ctime_nsec;
+ long st_atime; /* Time of last access. */
+ unsigned long st_atime_nsec;
+ long st_mtime; /* Time of last modification. */
+ unsigned long st_mtime_nsec;
+ long st_ctime; /* Time of last status change. */
+ unsigned long st_ctime_nsec;
unsigned int __unused4;
unsigned int __unused5;
};
-#if __BITS_PER_LONG != 64
/* This matches struct stat64 in glibc2.1. Only used for 32 bit. */
+#if __BITS_PER_LONG != 64 || defined(__ARCH_WANT_STAT64)
struct stat64 {
unsigned long long st_dev; /* Device. */
unsigned long long st_ino; /* File serial number. */
* together with the @destroy function,
* enables driver-specific objects derived from a ttm_buffer_object.
* On successful return, the object kref and list_kref are set to 1.
+ * If a failure occurs, the function will call the @destroy function, or
+ * kfree() if @destroy is NULL. Thus, after a failure, dereferencing @bo is
+ * illegal and will likely cause memory corruption.
+ *
* Returns
* -ENOMEM: Out of memory.
* -EINVAL: Invalid placement flags.
struct ttm_mem_type_manager;
struct ttm_mem_type_manager_func {
+ /**
+ * struct ttm_mem_type_manager member init
+ *
+ * @man: Pointer to a memory type manager.
+ * @p_size: Implementation dependent, but typically the size of the
+ * range to be managed in pages.
+ *
+ * Called to initialize a private range manager. The function is
+ * expected to initialize the man::priv member.
+ * Returns 0 on success, negative error code on failure.
+ */
int (*init)(struct ttm_mem_type_manager *man, unsigned long p_size);
+
+ /**
+ * struct ttm_mem_type_manager member takedown
+ *
+ * @man: Pointer to a memory type manager.
+ *
+ * Called to undo the setup done in init. All allocated resources
+ * should be freed.
+ */
int (*takedown)(struct ttm_mem_type_manager *man);
+
+ /**
+ * struct ttm_mem_type_manager member get_node
+ *
+ * @man: Pointer to a memory type manager.
+ * @bo: Pointer to the buffer object we're allocating space for.
+ * @placement: Placement details.
+ * @mem: Pointer to a struct ttm_mem_reg to be filled in.
+ *
+ * This function should allocate space in the memory type managed
+ * by @man. Placement details if
+ * applicable are given by @placement. If successful,
+ * @mem::mm_node should be set to a non-null value, and
+ * @mem::start should be set to a value identifying the beginning
+ * of the range allocated, and the function should return zero.
+ * If the memory region accomodate the buffer object, @mem::mm_node
+ * should be set to NULL, and the function should return 0.
+ * If a system error occured, preventing the request to be fulfilled,
+ * the function should return a negative error code.
+ *
+ * Note that @mem::mm_node will only be dereferenced by
+ * struct ttm_mem_type_manager functions and optionally by the driver,
+ * which has knowledge of the underlying type.
+ *
+ * This function may not be called from within atomic context, so
+ * an implementation can and must use either a mutex or a spinlock to
+ * protect any data structures managing the space.
+ */
int (*get_node)(struct ttm_mem_type_manager *man,
struct ttm_buffer_object *bo,
struct ttm_placement *placement,
struct ttm_mem_reg *mem);
+
+ /**
+ * struct ttm_mem_type_manager member put_node
+ *
+ * @man: Pointer to a memory type manager.
+ * @mem: Pointer to a struct ttm_mem_reg to be filled in.
+ *
+ * This function frees memory type resources previously allocated
+ * and that are identified by @mem::mm_node and @mem::start. May not
+ * be called from within atomic context.
+ */
void (*put_node)(struct ttm_mem_type_manager *man,
struct ttm_mem_reg *mem);
+
+ /**
+ * struct ttm_mem_type_manager member debug
+ *
+ * @man: Pointer to a memory type manager.
+ * @prefix: Prefix to be used in printout to identify the caller.
+ *
+ * This function is called to print out the state of the memory
+ * type manager to aid debugging of out-of-memory conditions.
+ * It may not be called from within atomic context.
+ */
void (*debug)(struct ttm_mem_type_manager *man, const char *prefix);
};
uint64_t size;
uint32_t available_caching;
uint32_t default_caching;
+ const struct ttm_mem_type_manager_func *func;
+ void *priv;
/*
- * Protected by the bdev->lru_lock.
- * TODO: Consider one lru_lock per ttm_mem_type_manager.
- * Plays ill with list removal, though.
+ * Protected by the global->lru_lock.
*/
- const struct ttm_mem_type_manager_func *func;
- void *priv;
+
struct list_head lru;
};
--- /dev/null
+#ifndef _LINUX_ATOMIC_H
+#define _LINUX_ATOMIC_H
+#include <asm/atomic.h>
+
+/**
+ * atomic_inc_not_zero_hint - increment if not null
+ * @v: pointer of type atomic_t
+ * @hint: probable value of the atomic before the increment
+ *
+ * This version of atomic_inc_not_zero() gives a hint of probable
+ * value of the atomic. This helps processor to not read the memory
+ * before doing the atomic read/modify/write cycle, lowering
+ * number of bus transactions on some arches.
+ *
+ * Returns: 0 if increment was not done, 1 otherwise.
+ */
+#ifndef atomic_inc_not_zero_hint
+static inline int atomic_inc_not_zero_hint(atomic_t *v, int hint)
+{
+ int val, c = hint;
+
+ /* sanity test, should be removed by compiler if hint is a constant */
+ if (!hint)
+ return atomic_inc_not_zero(v);
+
+ do {
+ val = atomic_cmpxchg(v, c, c + 1);
+ if (val == c)
+ return 1;
+ c = val;
+ } while (c);
+
+ return 0;
+}
+#endif
+
+#endif /* _LINUX_ATOMIC_H */
int (*fl_compare_owner)(struct file_lock *, struct file_lock *);
void (*fl_notify)(struct file_lock *); /* unblock callback */
int (*fl_grant)(struct file_lock *, struct file_lock *, int);
- void (*fl_copy_lock)(struct file_lock *, struct file_lock *);
void (*fl_release_private)(struct file_lock *);
void (*fl_break)(struct file_lock *);
int (*fl_mylease)(struct file_lock *, struct file_lock *);
*/
#define in_nmi() (preempt_count() & NMI_MASK)
-#if defined(CONFIG_PREEMPT)
+#if defined(CONFIG_PREEMPT) && defined(CONFIG_BKL)
# define PREEMPT_INATOMIC_BASE kernel_locked()
-# define PREEMPT_CHECK_OFFSET 1
#else
# define PREEMPT_INATOMIC_BASE 0
+#endif
+
+#if defined(CONFIG_PREEMPT)
+# define PREEMPT_CHECK_OFFSET 1
+#else
# define PREEMPT_CHECK_OFFSET 0
#endif
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/uaccess.h>
+#include <linux/hardirq.h>
#include <asm/cacheflush.h>
/* i2c adapter classes (bitmask) */
#define I2C_CLASS_HWMON (1<<0) /* lm_sensors, ... */
-#define I2C_CLASS_TV_ANALOG (1<<1) /* bttv + friends */
-#define I2C_CLASS_TV_DIGITAL (1<<2) /* dvb cards */
#define I2C_CLASS_DDC (1<<3) /* DDC bus on graphics adapters */
#define I2C_CLASS_SPD (1<<7) /* SPD EEPROMs and similar */
irq_free_descs(irq, 1);
}
+static inline int irq_reserve_irq(unsigned int irq)
+{
+ return irq_reserve_irqs(irq, 1);
+}
+
#endif /* CONFIG_GENERIC_HARDIRQS */
#endif /* !CONFIG_S390 */
else
#ifdef CONFIG_SMP
-#define irq_node(irq) (irq_to_desc(irq)->node)
+#define irq_node(irq) (irq_get_irq_data(irq)->node)
#else
#define irq_node(irq) 0
#endif
#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
#define roundup(x, y) ( \
{ \
- typeof(y) __y = y; \
+ const typeof(y) __y = y; \
(((x) + (__y - 1)) / __y) * __y; \
} \
)
unsigned int interval_msec);
extern int printk_delay_msec;
+extern int dmesg_restrict;
/*
* Print a one-time message (analogous to WARN_ONCE() et al):
--- /dev/null
+/*
+ * LP5521 LED chip driver.
+ *
+ * Copyright (C) 2010 Nokia Corporation
+ *
+ * Contact: Samu Onkalo <samu.p.onkalo@nokia.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
+ * 02110-1301 USA
+ */
+
+#ifndef __LINUX_LP5521_H
+#define __LINUX_LP5521_H
+
+/* See Documentation/leds/leds-lp5521.txt */
+
+struct lp5521_led_config {
+ u8 chan_nr;
+ u8 led_current; /* mA x10, 0 if led is not connected */
+ u8 max_current;
+};
+
+#define LP5521_CLOCK_AUTO 0
+#define LP5521_CLOCK_INT 1
+#define LP5521_CLOCK_EXT 2
+
+struct lp5521_platform_data {
+ struct lp5521_led_config *led_config;
+ u8 num_channels;
+ u8 clock_mode;
+ int (*setup_resources)(void);
+ void (*release_resources)(void);
+ void (*enable)(bool state);
+};
+
+#endif /* __LINUX_LP5521_H */
--- /dev/null
+/*
+ * LP5523 LED Driver
+ *
+ * Copyright (C) 2010 Nokia Corporation
+ *
+ * Contact: Samu Onkalo <samu.p.onkalo@nokia.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
+ * 02110-1301 USA
+ */
+
+#ifndef __LINUX_LP5523_H
+#define __LINUX_LP5523_H
+
+/* See Documentation/leds/leds-lp5523.txt */
+
+struct lp5523_led_config {
+ u8 chan_nr;
+ u8 led_current; /* mA x10, 0 if led is not connected */
+ u8 max_current;
+};
+
+#define LP5523_CLOCK_AUTO 0
+#define LP5523_CLOCK_INT 1
+#define LP5523_CLOCK_EXT 2
+
+struct lp5523_platform_data {
+ struct lp5523_led_config *led_config;
+ u8 num_channels;
+ u8 clock_mode;
+ int (*setup_resources)(void);
+ void (*release_resources)(void);
+ void (*enable)(bool state);
+};
+
+#endif /* __LINUX_LP5523_H */
#include <linux/list.h>
#include <linux/spinlock.h>
#include <linux/rwsem.h>
+#include <linux/timer.h>
struct device;
/*
/* Get LED brightness level */
enum led_brightness (*brightness_get)(struct led_classdev *led_cdev);
- /* Activate hardware accelerated blink, delays are in
- * miliseconds and if none is provided then a sensible default
- * should be chosen. The call can adjust the timings if it can't
- * match the values specified exactly. */
+ /*
+ * Activate hardware accelerated blink, delays are in milliseconds
+ * and if both are zero then a sensible default should be chosen.
+ * The call should adjust the timings in that case and if it can't
+ * match the values specified exactly.
+ * Deactivate blinking again when the brightness is set to a fixed
+ * value via the brightness_set() callback.
+ */
int (*blink_set)(struct led_classdev *led_cdev,
unsigned long *delay_on,
unsigned long *delay_off);
struct list_head node; /* LED Device list */
const char *default_trigger; /* Trigger to use */
+ unsigned long blink_delay_on, blink_delay_off;
+ struct timer_list blink_timer;
+ int blink_brightness;
+
#ifdef CONFIG_LEDS_TRIGGERS
/* Protects the trigger data below */
struct rw_semaphore trigger_lock;
extern void led_classdev_suspend(struct led_classdev *led_cdev);
extern void led_classdev_resume(struct led_classdev *led_cdev);
+/**
+ * led_blink_set - set blinking with software fallback
+ * @led_cdev: the LED to start blinking
+ * @delay_on: the time it should be on (in ms)
+ * @delay_off: the time it should ble off (in ms)
+ *
+ * This function makes the LED blink, attempting to use the
+ * hardware acceleration if possible, but falling back to
+ * software blinking if there is no hardware blinking or if
+ * the LED refuses the passed values.
+ *
+ * Note that if software blinking is active, simply calling
+ * led_cdev->brightness_set() will not stop the blinking,
+ * use led_classdev_brightness_set() instead.
+ */
+extern void led_blink_set(struct led_classdev *led_cdev,
+ unsigned long *delay_on,
+ unsigned long *delay_off);
+/**
+ * led_brightness_set - set LED brightness
+ * @led_cdev: the LED to set
+ * @brightness: the brightness to set it to
+ *
+ * Set an LED's brightness, and, if necessary, cancel the
+ * software blink timer that implements blinking when the
+ * hardware doesn't.
+ */
+extern void led_brightness_set(struct led_classdev *led_cdev,
+ enum led_brightness brightness);
+
/*
* LED Triggers
*/
#define MMCIF_CE_HOST_STS2 0x0000004C
#define MMCIF_CE_VERSION 0x0000007C
-extern inline u32 sh_mmcif_readl(void __iomem *addr, int reg)
+static inline u32 sh_mmcif_readl(void __iomem *addr, int reg)
{
return readl(addr + reg);
}
-extern inline void sh_mmcif_writel(void __iomem *addr, int reg, u32 val)
+static inline void sh_mmcif_writel(void __iomem *addr, int reg, u32 val)
{
writel(val, addr + reg);
}
#define SH_MMCIF_BBS 512 /* boot block size */
-extern inline void sh_mmcif_boot_cmd_send(void __iomem *base,
+static inline void sh_mmcif_boot_cmd_send(void __iomem *base,
unsigned long cmd, unsigned long arg)
{
sh_mmcif_writel(base, MMCIF_CE_INT, 0);
sh_mmcif_writel(base, MMCIF_CE_CMD_SET, cmd);
}
-extern inline int sh_mmcif_boot_cmd_poll(void __iomem *base, unsigned long mask)
+static inline int sh_mmcif_boot_cmd_poll(void __iomem *base, unsigned long mask)
{
unsigned long tmp;
int cnt;
return -1;
}
-extern inline int sh_mmcif_boot_cmd(void __iomem *base,
+static inline int sh_mmcif_boot_cmd(void __iomem *base,
unsigned long cmd, unsigned long arg)
{
sh_mmcif_boot_cmd_send(base, cmd, arg);
return sh_mmcif_boot_cmd_poll(base, 0x00010000);
}
-extern inline int sh_mmcif_boot_do_read_single(void __iomem *base,
+static inline int sh_mmcif_boot_do_read_single(void __iomem *base,
unsigned int block_nr,
unsigned long *buf)
{
return 0;
}
-extern inline int sh_mmcif_boot_do_read(void __iomem *base,
+static inline int sh_mmcif_boot_do_read(void __iomem *base,
unsigned long first_block,
unsigned long nr_blocks,
void *buf)
return ret;
}
-extern inline void sh_mmcif_boot_init(void __iomem *base)
+static inline void sh_mmcif_boot_init(void __iomem *base)
{
unsigned long tmp;
sh_mmcif_boot_cmd(base, 0x03400040, 0x00010000);
}
-extern inline void sh_mmcif_boot_slurp(void __iomem *base,
+static inline void sh_mmcif_boot_slurp(void __iomem *base,
unsigned char *buf,
unsigned long no_bytes)
{
#define PCI_DEVICE_ID_INTEL_COUGARPOINT_SMBUS 0x1c22
#define PCI_DEVICE_ID_INTEL_COUGARPOINT_LPC_MIN 0x1c41
#define PCI_DEVICE_ID_INTEL_COUGARPOINT_LPC_MAX 0x1c5f
+#define PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS 0x1d22
#define PCI_DEVICE_ID_INTEL_PATSBURG_LPC 0x1d40
#define PCI_DEVICE_ID_INTEL_82801AA_0 0x2410
#define PCI_DEVICE_ID_INTEL_82801AA_1 0x2411
u64 tstamp_running;
u64 tstamp_stopped;
+ /*
+ * timestamp shadows the actual context timing but it can
+ * be safely used in NMI interrupt context. It reflects the
+ * context time as it was when the event was last scheduled in.
+ *
+ * ctx_time already accounts for ctx->timestamp. Therefore to
+ * compute ctx_time for a sample, simply add perf_clock().
+ */
+ u64 shadow_ctx_time;
+
struct perf_event_attr attr;
struct hw_perf_event hw;
int pwm_id;
unsigned int max_brightness;
unsigned int dft_brightness;
+ unsigned int lth_brightness;
unsigned int pwm_period_ns;
int (*init)(struct device *dev);
int (*notify)(struct device *dev, int brightness);
* needed for RCU lookups (because root->height is unreliable). The only
* time callers need worry about this is when doing a lookup_slot under
* RCU.
+ *
+ * Indirect pointer in fact is also used to tag the last pointer of a node
+ * when it is shrunk, before we rcu free the node. See shrink code for
+ * details.
*/
#define RADIX_TREE_INDIRECT_PTR 1
-#define RADIX_TREE_RETRY ((void *)-1UL)
-
-static inline void *radix_tree_ptr_to_indirect(void *ptr)
-{
- return (void *)((unsigned long)ptr | RADIX_TREE_INDIRECT_PTR);
-}
-static inline void *radix_tree_indirect_to_ptr(void *ptr)
-{
- return (void *)((unsigned long)ptr & ~RADIX_TREE_INDIRECT_PTR);
-}
#define radix_tree_indirect_to_ptr(ptr) \
radix_tree_indirect_to_ptr((void __force *)(ptr))
* removed.
*
* For use with radix_tree_lookup_slot(). Caller must hold tree at least read
- * locked across slot lookup and dereference. More likely, will be used with
- * radix_tree_replace_slot(), as well, so caller will hold tree write locked.
+ * locked across slot lookup and dereference. Not required if write lock is
+ * held (ie. items cannot be concurrently inserted).
+ *
+ * radix_tree_deref_retry must be used to confirm validity of the pointer if
+ * only the read lock is held.
*/
static inline void *radix_tree_deref_slot(void **pslot)
{
- void *ret = rcu_dereference(*pslot);
- if (unlikely(radix_tree_is_indirect_ptr(ret)))
- ret = RADIX_TREE_RETRY;
- return ret;
+ return rcu_dereference(*pslot);
}
+
+/**
+ * radix_tree_deref_retry - check radix_tree_deref_slot
+ * @arg: pointer returned by radix_tree_deref_slot
+ * Returns: 0 if retry is not required, otherwise retry is required
+ *
+ * radix_tree_deref_retry must be used with radix_tree_deref_slot.
+ */
+static inline int radix_tree_deref_retry(void *arg)
+{
+ return unlikely((unsigned long)arg & RADIX_TREE_INDIRECT_PTR);
+}
+
/**
* radix_tree_replace_slot - replace item in a slot
* @pslot: pointer to slot, returned by radix_tree_lookup_slot
#define _LINUX_RESOURCE_H
#include <linux/time.h>
+#include <linux/types.h>
/*
* Resource control/accounting header file for linux
#define DEFINE_SEMAPHORE(name) \
struct semaphore name = __SEMAPHORE_INITIALIZER(name, 1)
-#define DECLARE_MUTEX(name) \
- struct semaphore name = __SEMAPHORE_INITIALIZER(name, 1)
-
static inline void sema_init(struct semaphore *sem, int val)
{
static struct lock_class_key __key;
lockdep_init_map(&sem->lock.dep_map, "semaphore->lock", &__key, 0);
}
-#define init_MUTEX(sem) sema_init(sem, 1)
-#define init_MUTEX_LOCKED(sem) sema_init(sem, 0)
-
extern void down(struct semaphore *sem);
extern int __must_check down_interruptible(struct semaphore *sem);
extern int __must_check down_killable(struct semaphore *sem);
long clk_rate_div_range_round(struct clk *clk, unsigned int div_min,
unsigned int div_max, unsigned long rate);
+long clk_round_parent(struct clk *clk, unsigned long target,
+ unsigned long *best_freq, unsigned long *parent_freq,
+ unsigned int div_min, unsigned int div_max);
+
#define SH_CLK_MSTP32(_parent, _enable_reg, _enable_bit, _flags) \
{ \
.parent = _parent, \
char *name;
long channel_offset;
int timer_bit;
- char *clk;
unsigned long clockevent_rating;
unsigned long clocksource_rating;
};
/**
* struct spi_master - interface to SPI master controller
* @dev: device interface to this driver
+ * @list: link with the global spi_master list
* @bus_num: board-specific (and often SOC-specific) identifier for a
* given SPI controller.
* @num_chipselect: chipselects are used to distinguish individual
struct spi_master {
struct device dev;
+ struct list_head list;
+
/* other than negative (== assign one dynamically), bus_num is fully
* board-specific. usually that simplifies to being SOC-specific.
* example: one SOC has three SPI controllers, numbered 0..2,
struct net *xpt_net;
};
-static inline void register_xpt_user(struct svc_xprt *xpt, struct svc_xpt_user *u)
+static inline void unregister_xpt_user(struct svc_xprt *xpt, struct svc_xpt_user *u)
{
spin_lock(&xpt->xpt_lock);
- list_add(&u->list, &xpt->xpt_users);
+ list_del_init(&u->list);
spin_unlock(&xpt->xpt_lock);
}
-static inline void unregister_xpt_user(struct svc_xprt *xpt, struct svc_xpt_user *u)
+static inline int register_xpt_user(struct svc_xprt *xpt, struct svc_xpt_user *u)
{
spin_lock(&xpt->xpt_lock);
- list_del_init(&u->list);
+ if (test_bit(XPT_CLOSE, &xpt->xpt_flags)) {
+ /*
+ * The connection is about to be deleted soon (or,
+ * worse, may already be deleted--in which case we've
+ * already notified the xpt_users).
+ */
+ spin_unlock(&xpt->xpt_lock);
+ return -ENOTCONN;
+ }
+ list_add(&u->list, &xpt->xpt_users);
spin_unlock(&xpt->xpt_lock);
+ return 0;
}
int svc_reg_xprt_class(struct svc_xprt_class *);
* @sockaddr: Socket address to connect.
* @priority: Priority of the connection.
* @link_selector: Link selector (high bandwidth or low latency)
- * @link_name: Name of the CAIF Link Layer to use.
+ * @ifindex: kernel index of the interface.
* @param: Connect Request parameters (CAIF_SO_REQ_PARAM).
*
* This struct is used when connecting a CAIF channel.
struct sockaddr_caif sockaddr;
enum caif_channel_priority priority;
enum caif_link_selector link_selector;
- char link_name[16];
+ int ifindex;
struct caif_param param;
};
wait_queue_head_t wait;
spinlock_t lock;
bool flow_stop;
+ bool slave;
+ bool slave_talked;
#ifdef CONFIG_DEBUG_FS
enum cfspi_state dbg_state;
u16 pcmd;
enum cfcnfg_phy_preference phy_pref);
/**
- * cfcnfg_get_named() - Get the Physical Identifier of CAIF Link Layer
+ * cfcnfg_get_id_from_ifi() - Get the Physical Identifier of ifindex,
+ * it matches caif physical id with the kernel interface id.
* @cnfg: Configuration object
- * @name: Name of the Physical Layer (Caif Link Layer)
+ * @ifi: ifindex obtained from socket.c bindtodevice.
*/
-int cfcnfg_get_named(struct cfcnfg *cnfg, char *name);
-
+int cfcnfg_get_id_from_ifi(struct cfcnfg *cnfg, int ifi);
#endif /* CFCNFG_H_ */
*
* Returns the first attribute which matches the specified type.
*/
-static inline struct nlattr *nlmsg_find_attr(struct nlmsghdr *nlh,
+static inline struct nlattr *nlmsg_find_attr(const struct nlmsghdr *nlh,
int hdrlen, int attrtype)
{
return nla_find(nlmsg_attrdata(nlh, hdrlen),
* struct fc_disc - Discovery context
* @retry_count: Number of retries
* @pending: 1 if discovery is pending, 0 if not
- * @requesting: 1 if discovery has been requested, 0 if not
+ * @requested: 1 if discovery has been requested, 0 if not
* @seq_count: Number of sequences used for discovery
* @buf_len: Length of the discovery buffer
* @disc_id: Discovery ID
void *buff;
unsigned alloc_size; /* 0 here means: don't call kfree */
unsigned total_bytes;
- } set_attr, enc_get_attr, get_attr;
+ } cdb_cont, set_attr, enc_get_attr, get_attr;
struct _osd_io_info {
struct bio *bio;
int osd_req_read_kern(struct osd_request *or,
const struct osd_obj_id *obj, u64 offset, void *buff, u64 len);
+/* Scatter/Gather write/read commands */
+int osd_req_write_sg(struct osd_request *or,
+ const struct osd_obj_id *obj, struct bio *bio,
+ const struct osd_sg_entry *sglist, unsigned numentries);
+int osd_req_read_sg(struct osd_request *or,
+ const struct osd_obj_id *obj, struct bio *bio,
+ const struct osd_sg_entry *sglist, unsigned numentries);
+int osd_req_write_sg_kern(struct osd_request *or,
+ const struct osd_obj_id *obj, void **buff,
+ const struct osd_sg_entry *sglist, unsigned numentries);
+int osd_req_read_sg_kern(struct osd_request *or,
+ const struct osd_obj_id *obj, void **buff,
+ const struct osd_sg_entry *sglist, unsigned numentries);
+
/*
* Root/Partition/Collection/Object Attributes commands
*/
put_unaligned_le16(bit_mask, &cap->permissions_bit_mask);
}
+/* osd2r05a sec 5.3: CDB continuation segment formats */
+enum osd_continuation_segment_format {
+ CDB_CONTINUATION_FORMAT_V2 = 0x01,
+};
+
+struct osd_continuation_segment_header {
+ u8 format;
+ u8 reserved1;
+ __be16 service_action;
+ __be32 reserved2;
+ u8 integrity_check[OSDv2_CRYPTO_KEYID_SIZE];
+} __packed;
+
+/* osd2r05a sec 5.4.1: CDB continuation descriptors */
+enum osd_continuation_descriptor_type {
+ NO_MORE_DESCRIPTORS = 0x0000,
+ SCATTER_GATHER_LIST = 0x0001,
+ QUERY_LIST = 0x0002,
+ USER_OBJECT = 0x0003,
+ COPY_USER_OBJECT_SOURCE = 0x0101,
+ EXTENSION_CAPABILITIES = 0xFFEE
+};
+
+struct osd_continuation_descriptor_header {
+ __be16 type;
+ u8 reserved;
+ u8 pad_length;
+ __be32 length;
+} __packed;
+
+
+/* osd2r05a sec 5.4.2: Scatter/gather list */
+struct osd_sg_list_entry {
+ __be64 offset;
+ __be64 len;
+};
+
+struct osd_sg_continuation_descriptor {
+ struct osd_continuation_descriptor_header hdr;
+ struct osd_sg_list_entry entries[];
+};
+
#endif /* ndef __OSD_PROTOCOL_H__ */
void *val_ptr; /* in network order */
};
+struct osd_sg_entry {
+ u64 offset;
+ u64 len;
+};
+
#endif /* ndef __OSD_TYPES_H__ */
(unsigned long) __entry->dir, __entry->mode)
);
+TRACE_EVENT(ext4_evict_inode,
+ TP_PROTO(struct inode *inode),
+
+ TP_ARGS(inode),
+
+ TP_STRUCT__entry(
+ __field( int, dev_major )
+ __field( int, dev_minor )
+ __field( ino_t, ino )
+ __field( int, nlink )
+ ),
+
+ TP_fast_assign(
+ __entry->dev_major = MAJOR(inode->i_sb->s_dev);
+ __entry->dev_minor = MINOR(inode->i_sb->s_dev);
+ __entry->ino = inode->i_ino;
+ __entry->nlink = inode->i_nlink;
+ ),
+
+ TP_printk("dev %d,%d ino %lu nlink %d",
+ __entry->dev_major, __entry->dev_minor,
+ (unsigned long) __entry->ino, __entry->nlink)
+);
+
+TRACE_EVENT(ext4_drop_inode,
+ TP_PROTO(struct inode *inode, int drop),
+
+ TP_ARGS(inode, drop),
+
+ TP_STRUCT__entry(
+ __field( int, dev_major )
+ __field( int, dev_minor )
+ __field( ino_t, ino )
+ __field( int, drop )
+ ),
+
+ TP_fast_assign(
+ __entry->dev_major = MAJOR(inode->i_sb->s_dev);
+ __entry->dev_minor = MINOR(inode->i_sb->s_dev);
+ __entry->ino = inode->i_ino;
+ __entry->drop = drop;
+ ),
+
+ TP_printk("dev %d,%d ino %lu drop %d",
+ __entry->dev_major, __entry->dev_minor,
+ (unsigned long) __entry->ino, __entry->drop)
+);
+
+TRACE_EVENT(ext4_mark_inode_dirty,
+ TP_PROTO(struct inode *inode, unsigned long IP),
+
+ TP_ARGS(inode, IP),
+
+ TP_STRUCT__entry(
+ __field( int, dev_major )
+ __field( int, dev_minor )
+ __field( ino_t, ino )
+ __field(unsigned long, ip )
+ ),
+
+ TP_fast_assign(
+ __entry->dev_major = MAJOR(inode->i_sb->s_dev);
+ __entry->dev_minor = MINOR(inode->i_sb->s_dev);
+ __entry->ino = inode->i_ino;
+ __entry->ip = IP;
+ ),
+
+ TP_printk("dev %d,%d ino %lu caller %pF",
+ __entry->dev_major, __entry->dev_minor,
+ (unsigned long) __entry->ino, (void *)__entry->ip)
+);
+
+TRACE_EVENT(ext4_begin_ordered_truncate,
+ TP_PROTO(struct inode *inode, loff_t new_size),
+
+ TP_ARGS(inode, new_size),
+
+ TP_STRUCT__entry(
+ __field( int, dev_major )
+ __field( int, dev_minor )
+ __field( ino_t, ino )
+ __field( loff_t, new_size )
+ ),
+
+ TP_fast_assign(
+ __entry->dev_major = MAJOR(inode->i_sb->s_dev);
+ __entry->dev_minor = MINOR(inode->i_sb->s_dev);
+ __entry->ino = inode->i_ino;
+ __entry->new_size = new_size;
+ ),
+
+ TP_printk("dev %d,%d ino %lu new_size %lld",
+ __entry->dev_major, __entry->dev_minor,
+ (unsigned long) __entry->ino,
+ (long long) __entry->new_size)
+);
+
DECLARE_EVENT_CLASS(ext4__write_begin,
TP_PROTO(struct inode *inode, loff_t pos, unsigned int len,
tty = sig->tty;
sig->tty = NULL;
} else {
+ /*
+ * This can only happen if the caller is de_thread().
+ * FIXME: this is the temporary hack, we should teach
+ * posix-cpu-timers to handle this case correctly.
+ */
+ if (unlikely(has_group_leader_pid(tsk)))
+ posix_cpu_timers_exit_group(tsk);
+
/*
* If there is any task waiting for the group exit
* then notify it:
if (!desc)
return;
+ if (WARN(!desc->irq_data.chip || !desc->irq_data.chip->irq_enable,
+ KERN_ERR "enable_irq before setup/request_irq: irq %u\n", irq))
+ return;
+
chip_bus_lock(desc);
raw_spin_lock_irqsave(&desc->lock, flags);
__enable_irq(desc, irq, false);
account_global_scheduler_latency(tsk, &lat);
- /*
- * short term hack; if we're > 32 we stop; future we recycle:
- */
- tsk->latency_record_count++;
- if (tsk->latency_record_count >= LT_SAVECOUNT)
- goto out_unlock;
-
- for (i = 0; i < LT_SAVECOUNT; i++) {
+ for (i = 0; i < tsk->latency_record_count; i++) {
struct latency_record *mylat;
int same = 1;
}
}
+ /*
+ * short term hack; if we're > 32 we stop; future we recycle:
+ */
+ if (tsk->latency_record_count >= LT_SAVECOUNT)
+ goto out_unlock;
+
/* Allocated a new one: */
- i = tsk->latency_record_count;
+ i = tsk->latency_record_count++;
memcpy(&tsk->latency_record[i], &lat, sizeof(struct latency_record));
out_unlock:
event->tstamp_running += ctx->time - event->tstamp_stopped;
+ event->shadow_ctx_time = ctx->time - ctx->timestamp;
+
if (!is_software_event(event))
cpuctx->active_oncpu++;
ctx->nr_active++;
}
static void perf_output_read_one(struct perf_output_handle *handle,
- struct perf_event *event)
+ struct perf_event *event,
+ u64 enabled, u64 running)
{
u64 read_format = event->attr.read_format;
u64 values[4];
values[n++] = perf_event_count(event);
if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) {
- values[n++] = event->total_time_enabled +
+ values[n++] = enabled +
atomic64_read(&event->child_total_time_enabled);
}
if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) {
- values[n++] = event->total_time_running +
+ values[n++] = running +
atomic64_read(&event->child_total_time_running);
}
if (read_format & PERF_FORMAT_ID)
* XXX PERF_FORMAT_GROUP vs inherited events seems difficult.
*/
static void perf_output_read_group(struct perf_output_handle *handle,
- struct perf_event *event)
+ struct perf_event *event,
+ u64 enabled, u64 running)
{
struct perf_event *leader = event->group_leader, *sub;
u64 read_format = event->attr.read_format;
values[n++] = 1 + leader->nr_siblings;
if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
- values[n++] = leader->total_time_enabled;
+ values[n++] = enabled;
if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
- values[n++] = leader->total_time_running;
+ values[n++] = running;
if (leader != event)
leader->pmu->read(leader);
}
}
+#define PERF_FORMAT_TOTAL_TIMES (PERF_FORMAT_TOTAL_TIME_ENABLED|\
+ PERF_FORMAT_TOTAL_TIME_RUNNING)
+
static void perf_output_read(struct perf_output_handle *handle,
struct perf_event *event)
{
+ u64 enabled = 0, running = 0, now, ctx_time;
+ u64 read_format = event->attr.read_format;
+
+ /*
+ * compute total_time_enabled, total_time_running
+ * based on snapshot values taken when the event
+ * was last scheduled in.
+ *
+ * we cannot simply called update_context_time()
+ * because of locking issue as we are called in
+ * NMI context
+ */
+ if (read_format & PERF_FORMAT_TOTAL_TIMES) {
+ now = perf_clock();
+ ctx_time = event->shadow_ctx_time + now;
+ enabled = ctx_time - event->tstamp_enabled;
+ running = ctx_time - event->tstamp_running;
+ }
+
if (event->attr.read_format & PERF_FORMAT_GROUP)
- perf_output_read_group(handle, event);
+ perf_output_read_group(handle, event, enabled, running);
else
- perf_output_read_one(handle, event);
+ perf_output_read_one(handle, event, enabled, running);
}
void perf_output_sample(struct perf_output_handle *handle,
}
#endif
+#ifdef CONFIG_SECURITY_DMESG_RESTRICT
+int dmesg_restrict = 1;
+#else
+int dmesg_restrict;
+#endif
+
int do_syslog(int type, char __user *buf, int len, bool from_file)
{
unsigned i, j, limit, count;
int clean_sort_range(struct range *range, int az)
{
- int i, j, k = az - 1, nr_range = 0;
+ int i, j, k = az - 1, nr_range = az;
for (i = 0; i < k; i++) {
if (range[i].end)
*/
static struct page **relay_alloc_page_array(unsigned int n_pages)
{
- struct page **array;
- size_t pa_size = n_pages * sizeof(struct page *);
-
- if (pa_size > PAGE_SIZE) {
- array = vmalloc(pa_size);
- if (array)
- memset(array, 0, pa_size);
- } else {
- array = kzalloc(pa_size, GFP_KERNEL);
- }
- return array;
+ const size_t pa_size = n_pages * sizeof(struct page *);
+ if (pa_size > PAGE_SIZE)
+ return vzalloc(pa_size);
+ return kzalloc(pa_size, GFP_KERNEL);
}
/*
.extra2 = &ten_thousand,
},
#endif
+ {
+ .procname = "dmesg_restrict",
+ .data = &dmesg_restrict,
+ .maxlen = sizeof(int),
+ .mode = 0644,
+ .proc_handler = proc_dointvec_minmax,
+ .extra1 = &zero,
+ .extra2 = &one,
+ },
{
.procname = "ngroups_max",
.data = &ngroups_max,
static DEFINE_PER_CPU(struct perf_event *, watchdog_ev);
#endif
-static int __initdata no_watchdog;
+static int no_watchdog;
/* boot commands */
};
static DEFINE_PER_CPU(struct radix_tree_preload, radix_tree_preloads) = { 0, };
+static inline void *ptr_to_indirect(void *ptr)
+{
+ return (void *)((unsigned long)ptr | RADIX_TREE_INDIRECT_PTR);
+}
+
+static inline void *indirect_to_ptr(void *ptr)
+{
+ return (void *)((unsigned long)ptr & ~RADIX_TREE_INDIRECT_PTR);
+}
+
static inline gfp_t root_gfp_mask(struct radix_tree_root *root)
{
return root->gfp_mask & __GFP_BITS_MASK;
return -ENOMEM;
/* Increase the height. */
- node->slots[0] = radix_tree_indirect_to_ptr(root->rnode);
+ node->slots[0] = indirect_to_ptr(root->rnode);
/* Propagate the aggregated tag info into the new root */
for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) {
newheight = root->height+1;
node->height = newheight;
node->count = 1;
- node = radix_tree_ptr_to_indirect(node);
+ node = ptr_to_indirect(node);
rcu_assign_pointer(root->rnode, node);
root->height = newheight;
} while (height > root->height);
return error;
}
- slot = radix_tree_indirect_to_ptr(root->rnode);
+ slot = indirect_to_ptr(root->rnode);
height = root->height;
shift = (height-1) * RADIX_TREE_MAP_SHIFT;
rcu_assign_pointer(node->slots[offset], slot);
node->count++;
} else
- rcu_assign_pointer(root->rnode,
- radix_tree_ptr_to_indirect(slot));
+ rcu_assign_pointer(root->rnode, ptr_to_indirect(slot));
}
/* Go a level down */
return NULL;
return is_slot ? (void *)&root->rnode : node;
}
- node = radix_tree_indirect_to_ptr(node);
+ node = indirect_to_ptr(node);
height = node->height;
if (index > radix_tree_maxindex(height))
height--;
} while (height > 0);
- return is_slot ? (void *)slot:node;
+ return is_slot ? (void *)slot : indirect_to_ptr(node);
}
/**
height = root->height;
BUG_ON(index > radix_tree_maxindex(height));
- slot = radix_tree_indirect_to_ptr(root->rnode);
+ slot = indirect_to_ptr(root->rnode);
shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
while (height > 0) {
shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
pathp->node = NULL;
- slot = radix_tree_indirect_to_ptr(root->rnode);
+ slot = indirect_to_ptr(root->rnode);
while (height > 0) {
int offset;
if (!radix_tree_is_indirect_ptr(node))
return (index == 0);
- node = radix_tree_indirect_to_ptr(node);
+ node = indirect_to_ptr(node);
height = node->height;
if (index > radix_tree_maxindex(height))
}
shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
- slot = radix_tree_indirect_to_ptr(root->rnode);
+ slot = indirect_to_ptr(root->rnode);
/*
* we fill the path from (root->height - 2) to 0, leaving the index at
results[0] = node;
return 1;
}
- node = radix_tree_indirect_to_ptr(node);
+ node = indirect_to_ptr(node);
max_index = radix_tree_maxindex(node->height);
slot = *(((void ***)results)[ret + i]);
if (!slot)
continue;
- results[ret + nr_found] = rcu_dereference_raw(slot);
+ results[ret + nr_found] =
+ indirect_to_ptr(rcu_dereference_raw(slot));
nr_found++;
}
ret += nr_found;
results[0] = (void **)&root->rnode;
return 1;
}
- node = radix_tree_indirect_to_ptr(node);
+ node = indirect_to_ptr(node);
max_index = radix_tree_maxindex(node->height);
results[0] = node;
return 1;
}
- node = radix_tree_indirect_to_ptr(node);
+ node = indirect_to_ptr(node);
max_index = radix_tree_maxindex(node->height);
slot = *(((void ***)results)[ret + i]);
if (!slot)
continue;
- results[ret + nr_found] = rcu_dereference_raw(slot);
+ results[ret + nr_found] =
+ indirect_to_ptr(rcu_dereference_raw(slot));
nr_found++;
}
ret += nr_found;
results[0] = (void **)&root->rnode;
return 1;
}
- node = radix_tree_indirect_to_ptr(node);
+ node = indirect_to_ptr(node);
max_index = radix_tree_maxindex(node->height);
void *newptr;
BUG_ON(!radix_tree_is_indirect_ptr(to_free));
- to_free = radix_tree_indirect_to_ptr(to_free);
+ to_free = indirect_to_ptr(to_free);
/*
* The candidate node has more than one child, or its child
/*
* We don't need rcu_assign_pointer(), since we are simply
- * moving the node from one part of the tree to another. If
- * it was safe to dereference the old pointer to it
+ * moving the node from one part of the tree to another: if it
+ * was safe to dereference the old pointer to it
* (to_free->slots[0]), it will be safe to dereference the new
- * one (root->rnode).
+ * one (root->rnode) as far as dependent read barriers go.
*/
newptr = to_free->slots[0];
if (root->height > 1)
- newptr = radix_tree_ptr_to_indirect(newptr);
+ newptr = ptr_to_indirect(newptr);
root->rnode = newptr;
root->height--;
+
+ /*
+ * We have a dilemma here. The node's slot[0] must not be
+ * NULLed in case there are concurrent lookups expecting to
+ * find the item. However if this was a bottom-level node,
+ * then it may be subject to the slot pointer being visible
+ * to callers dereferencing it. If item corresponding to
+ * slot[0] is subsequently deleted, these callers would expect
+ * their slot to become empty sooner or later.
+ *
+ * For example, lockless pagecache will look up a slot, deref
+ * the page pointer, and if the page is 0 refcount it means it
+ * was concurrently deleted from pagecache so try the deref
+ * again. Fortunately there is already a requirement for logic
+ * to retry the entire slot lookup -- the indirect pointer
+ * problem (replacing direct root node with an indirect pointer
+ * also results in a stale slot). So tag the slot as indirect
+ * to force callers to retry.
+ */
+ if (root->height == 0)
+ *((unsigned long *)&to_free->slots[0]) |=
+ RADIX_TREE_INDIRECT_PTR;
+
radix_tree_node_free(to_free);
}
}
root->rnode = NULL;
goto out;
}
- slot = radix_tree_indirect_to_ptr(slot);
+ slot = indirect_to_ptr(slot);
shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
pathp->node = NULL;
radix_tree_node_free(to_free);
if (pathp->node->count) {
- if (pathp->node ==
- radix_tree_indirect_to_ptr(root->rnode))
+ if (pathp->node == indirect_to_ptr(root->rnode))
radix_tree_shrink(root);
goto out;
}
pagep = radix_tree_lookup_slot(&mapping->page_tree, offset);
if (pagep) {
page = radix_tree_deref_slot(pagep);
- if (unlikely(!page || page == RADIX_TREE_RETRY))
+ if (unlikely(!page))
+ goto out;
+ if (radix_tree_deref_retry(page))
goto repeat;
if (!page_cache_get_speculative(page))
goto repeat;
}
}
+out:
rcu_read_unlock();
return page;
page = radix_tree_deref_slot((void **)pages[i]);
if (unlikely(!page))
continue;
- /*
- * this can only trigger if nr_found == 1, making livelock
- * a non issue.
- */
- if (unlikely(page == RADIX_TREE_RETRY))
+ if (radix_tree_deref_retry(page)) {
+ if (ret)
+ start = pages[ret-1]->index;
goto restart;
+ }
if (!page_cache_get_speculative(page))
goto repeat;
page = radix_tree_deref_slot((void **)pages[i]);
if (unlikely(!page))
continue;
- /*
- * this can only trigger if nr_found == 1, making livelock
- * a non issue.
- */
- if (unlikely(page == RADIX_TREE_RETRY))
+ if (radix_tree_deref_retry(page))
goto restart;
if (page->mapping == NULL || page->index != index)
page = radix_tree_deref_slot((void **)pages[i]);
if (unlikely(!page))
continue;
- /*
- * this can only trigger if nr_found == 1, making livelock
- * a non issue.
- */
- if (unlikely(page == RADIX_TREE_RETRY))
+ if (radix_tree_deref_retry(page))
goto restart;
if (!page_cache_get_speculative(page))
goto page_not_up_to_date;
if (!trylock_page(page))
goto page_not_up_to_date;
+ /* Did it get truncated before we got the lock? */
+ if (!page->mapping)
+ goto page_not_up_to_date_locked;
if (!mapping->a_ops->is_partially_uptodate(page,
desc, offset))
goto page_not_up_to_date_locked;
goto no_cached_page;
}
- if (!lock_page_or_retry(page, vma->vm_mm, vmf->flags))
+ if (!lock_page_or_retry(page, vma->vm_mm, vmf->flags)) {
+ page_cache_release(page);
return ret | VM_FAULT_RETRY;
+ }
/* Did it get truncated? */
if (unlikely(page->mapping != mapping)) {
memset(mem, 0, size);
mem->stat = alloc_percpu(struct mem_cgroup_stat_cpu);
- if (!mem->stat) {
- if (size < PAGE_SIZE)
- kfree(mem);
- else
- vfree(mem);
- mem = NULL;
- }
+ if (!mem->stat)
+ goto out_free;
spin_lock_init(&mem->pcp_counter_lock);
return mem;
+
+out_free:
+ if (size < PAGE_SIZE)
+ kfree(mem);
+ else
+ vfree(mem);
+ return NULL;
}
/*
mmu_notifier_invalidate_range_end(mm, start, end);
vm_stat_account(mm, oldflags, vma->vm_file, -nrpages);
vm_stat_account(mm, newflags, vma->vm_file, nrpages);
+ perf_event_mmap(vma);
return 0;
fail:
error = mprotect_fixup(vma, &prev, nstart, tmp, newflags);
if (error)
goto out;
- perf_event_mmap(vma);
nstart = tmp;
if (nstart < prev->vm_end)
* back off and wait for congestion to clear because further reclaim
* will encounter the same problem
*/
- if (nr_dirty == nr_congested)
+ if (nr_dirty == nr_congested && nr_dirty != 0)
zone_set_flag(zone, ZONE_CONGESTED);
free_page_list(&free_pages);
v[PGPGIN] /= 2; /* sectors -> kbytes */
v[PGPGOUT] /= 2;
#endif
- return m->private + *pos;
+ return (unsigned long *)m->private + *pos;
}
static void *vmstat_next(struct seq_file *m, void *arg, loff_t *pos)
{
struct dev_info *dev_info;
enum cfcnfg_phy_preference pref;
+ int res;
+
memset(l, 0, sizeof(*l));
- l->priority = s->priority;
+ /* In caif protocol low value is high priority */
+ l->priority = CAIF_PRIO_MAX - s->priority + 1;
- if (s->link_name[0] != '\0')
- l->phyid = cfcnfg_get_named(cnfg, s->link_name);
+ if (s->ifindex != 0){
+ res = cfcnfg_get_id_from_ifi(cnfg, s->ifindex);
+ if (res < 0)
+ return res;
+ l->phyid = res;
+ }
else {
switch (s->link_selector) {
case CAIF_LINK_HIGH_BANDW:
case NETDEV_UNREGISTER:
caifd = caif_get(dev);
+ if (caifd == NULL)
+ break;
netdev_info(dev, "unregister\n");
atomic_set(&caifd->state, what);
caif_device_destroy(dev);
{
struct sock *sk = sock->sk;
struct caifsock *cf_sk = container_of(sk, struct caifsock, sk);
- int prio, linksel;
- struct ifreq ifreq;
+ int linksel;
if (cf_sk->sk.sk_socket->state != SS_UNCONNECTED)
return -ENOPROTOOPT;
release_sock(&cf_sk->sk);
return 0;
- case SO_PRIORITY:
- if (lvl != SOL_SOCKET)
- goto bad_sol;
- if (ol < sizeof(int))
- return -EINVAL;
- if (copy_from_user(&prio, ov, sizeof(int)))
- return -EINVAL;
- lock_sock(&(cf_sk->sk));
- cf_sk->conn_req.priority = prio;
- release_sock(&cf_sk->sk);
- return 0;
-
- case SO_BINDTODEVICE:
- if (lvl != SOL_SOCKET)
- goto bad_sol;
- if (ol < sizeof(struct ifreq))
- return -EINVAL;
- if (copy_from_user(&ifreq, ov, sizeof(ifreq)))
- return -EFAULT;
- lock_sock(&(cf_sk->sk));
- strncpy(cf_sk->conn_req.link_name, ifreq.ifr_name,
- sizeof(cf_sk->conn_req.link_name));
- cf_sk->conn_req.link_name
- [sizeof(cf_sk->conn_req.link_name)-1] = 0;
- release_sock(&cf_sk->sk);
- return 0;
-
case CAIFSO_REQ_PARAM:
if (lvl != SOL_CAIF)
goto bad_sol;
sock->state = SS_CONNECTING;
sk->sk_state = CAIF_CONNECTING;
+ /* Check priority value comming from socket */
+ /* if priority value is out of range it will be ajusted */
+ if (cf_sk->sk.sk_priority > CAIF_PRIO_MAX)
+ cf_sk->conn_req.priority = CAIF_PRIO_MAX;
+ else if (cf_sk->sk.sk_priority < CAIF_PRIO_MIN)
+ cf_sk->conn_req.priority = CAIF_PRIO_MIN;
+ else
+ cf_sk->conn_req.priority = cf_sk->sk.sk_priority;
+
+ /*ifindex = id of the interface.*/
+ cf_sk->conn_req.ifindex = cf_sk->sk.sk_bound_dev_if;
+
dbfs_atomic_inc(&cnt.num_connect_req);
cf_sk->layer.receive = caif_sktrecv_cb;
err = caif_connect_client(&cf_sk->conn_req,
cf_sk->maxframe = mtu - (headroom + tailroom);
if (cf_sk->maxframe < 1) {
pr_warn("CAIF Interface MTU too small (%d)\n", dev->mtu);
+ err = -ENODEV;
goto out;
}
set_rx_flow_on(cf_sk);
/* Set default options on configuration */
- cf_sk->conn_req.priority = CAIF_PRIO_NORMAL;
+ cf_sk->sk.sk_priority= CAIF_PRIO_NORMAL;
cf_sk->conn_req.link_selector = CAIF_LINK_LOW_LATENCY;
cf_sk->conn_req.protocol = protocol;
/* Increase the number of sockets created. */
return NULL;
}
-int cfcnfg_get_named(struct cfcnfg *cnfg, char *name)
+
+int cfcnfg_get_id_from_ifi(struct cfcnfg *cnfg, int ifi)
{
int i;
-
- /* Try to match with specified name */
- for (i = 0; i < MAX_PHY_LAYERS; i++) {
- if (cnfg->phy_layers[i].frm_layer != NULL
- && strcmp(cnfg->phy_layers[i].phy_layer->name,
- name) == 0)
- return cnfg->phy_layers[i].frm_layer->id;
- }
- return 0;
+ for (i = 0; i < MAX_PHY_LAYERS; i++)
+ if (cnfg->phy_layers[i].frm_layer != NULL &&
+ cnfg->phy_layers[i].ifindex == ifi)
+ return i;
+ return -ENODEV;
}
int cfcnfg_disconn_adapt_layer(struct cfcnfg *cnfg, struct cflayer *adap_layer)
struct cfctrl_request_info *p, *tmp;
struct cfctrl *ctrl = container_obj(layr);
spin_lock(&ctrl->info_list_lock);
- pr_warn("enter\n");
list_for_each_entry_safe(p, tmp, &ctrl->list, list) {
if (p->client_layer == adap_layer) {
- pr_warn("cancel req :%d\n", p->sequence_no);
+ pr_debug("cancel req :%d\n", p->sequence_no);
list_del(&p->list);
kfree(p);
}
#include <net/caif/cfsrvl.h>
#include <net/caif/cfpkt.h>
+#define container_obj(layr) ((struct cfsrvl *) layr)
+
static int cfdbgl_receive(struct cflayer *layr, struct cfpkt *pkt);
static int cfdbgl_transmit(struct cflayer *layr, struct cfpkt *pkt);
static int cfdbgl_transmit(struct cflayer *layr, struct cfpkt *pkt)
{
+ struct cfsrvl *service = container_obj(layr);
+ struct caif_payload_info *info;
+ int ret;
+
+ if (!cfsrvl_ready(service, &ret))
+ return ret;
+
+ /* Add info for MUX-layer to route the packet out */
+ info = cfpkt_info(pkt);
+ info->channel_id = service->layer.id;
+ info->dev_info = &service->dev_info;
+
return layr->dn->transmit(layr->dn, pkt);
}
static int cfrfml_transmit_segment(struct cfrfml *rfml, struct cfpkt *pkt)
{
- caif_assert(cfpkt_getlen(pkt) >= rfml->fragment_size);
+ caif_assert(cfpkt_getlen(pkt) < rfml->fragment_size);
/* Add info for MUX-layer to route the packet out. */
cfpkt_info(pkt)->channel_id = rfml->serv.layer.id;
} else {
struct sock *sk = skb->sk;
queue_index = sk_tx_queue_get(sk);
- if (queue_index < 0) {
+ if (queue_index < 0 || queue_index >= dev->real_num_tx_queues) {
queue_index = 0;
if (dev->real_num_tx_queues > 1)
static inline void fib_result_assign(struct fib_result *res,
struct fib_info *fi)
{
- if (res->fi != NULL)
- fib_info_put(res->fi);
+ /* we used to play games with refcounts, but we now use RCU */
res->fi = fi;
- if (fi != NULL)
- atomic_inc(&fi->fib_clntref);
}
#endif /* _FIB_LOOKUP_H */
{
struct inet_diag_req *r = NLMSG_DATA(cb->nlh);
- if (cb->nlh->nlmsg_len > 4 + NLMSG_SPACE(sizeof(*r))) {
+ if (nlmsg_attrlen(cb->nlh, sizeof(*r))) {
struct inet_diag_entry entry;
- struct rtattr *bc = (struct rtattr *)(r + 1);
+ const struct nlattr *bc = nlmsg_find_attr(cb->nlh,
+ sizeof(*r),
+ INET_DIAG_REQ_BYTECODE);
struct inet_sock *inet = inet_sk(sk);
entry.family = sk->sk_family;
entry.dport = ntohs(inet->inet_dport);
entry.userlocks = sk->sk_userlocks;
- if (!inet_diag_bc_run(RTA_DATA(bc), RTA_PAYLOAD(bc), &entry))
+ if (!inet_diag_bc_run(nla_data(bc), nla_len(bc), &entry))
return 0;
}
{
struct inet_diag_req *r = NLMSG_DATA(cb->nlh);
- if (cb->nlh->nlmsg_len > 4 + NLMSG_SPACE(sizeof(*r))) {
+ if (nlmsg_attrlen(cb->nlh, sizeof(*r))) {
struct inet_diag_entry entry;
- struct rtattr *bc = (struct rtattr *)(r + 1);
+ const struct nlattr *bc = nlmsg_find_attr(cb->nlh,
+ sizeof(*r),
+ INET_DIAG_REQ_BYTECODE);
entry.family = tw->tw_family;
#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
entry.dport = ntohs(tw->tw_dport);
entry.userlocks = 0;
- if (!inet_diag_bc_run(RTA_DATA(bc), RTA_PAYLOAD(bc), &entry))
+ if (!inet_diag_bc_run(nla_data(bc), nla_len(bc), &entry))
return 0;
}
struct inet_diag_req *r = NLMSG_DATA(cb->nlh);
struct inet_connection_sock *icsk = inet_csk(sk);
struct listen_sock *lopt;
- struct rtattr *bc = NULL;
+ const struct nlattr *bc = NULL;
struct inet_sock *inet = inet_sk(sk);
int j, s_j;
int reqnum, s_reqnum;
if (!lopt || !lopt->qlen)
goto out;
- if (cb->nlh->nlmsg_len > 4 + NLMSG_SPACE(sizeof(*r))) {
- bc = (struct rtattr *)(r + 1);
+ if (nlmsg_attrlen(cb->nlh, sizeof(*r))) {
+ bc = nlmsg_find_attr(cb->nlh, sizeof(*r),
+ INET_DIAG_REQ_BYTECODE);
entry.sport = inet->inet_num;
entry.userlocks = sk->sk_userlocks;
}
&ireq->rmt_addr;
entry.dport = ntohs(ireq->rmt_port);
- if (!inet_diag_bc_run(RTA_DATA(bc),
- RTA_PAYLOAD(bc), &entry))
+ if (!inet_diag_bc_run(nla_data(bc),
+ nla_len(bc), &entry))
continue;
}
private = &tmp;
}
#endif
+ memset(&info, 0, sizeof(info));
info.valid_hooks = t->valid_hooks;
memcpy(info.hook_entry, private->hook_entry,
sizeof(info.hook_entry));
private = &tmp;
}
#endif
+ memset(&info, 0, sizeof(info));
info.valid_hooks = t->valid_hooks;
memcpy(info.hook_entry, private->hook_entry,
sizeof(info.hook_entry));
return rcu_dereference(nf_nat_protos[protonum]);
}
-static const struct nf_nat_protocol *
-nf_nat_proto_find_get(u_int8_t protonum)
-{
- const struct nf_nat_protocol *p;
-
- rcu_read_lock();
- p = __nf_nat_proto_find(protonum);
- if (!try_module_get(p->me))
- p = &nf_nat_unknown_protocol;
- rcu_read_unlock();
-
- return p;
-}
-
-static void
-nf_nat_proto_put(const struct nf_nat_protocol *p)
-{
- module_put(p->me);
-}
-
/* We keep an extra hash for each conntrack, for fast searching. */
static inline unsigned int
hash_by_src(const struct net *net, u16 zone,
#include <linux/netfilter/nfnetlink.h>
#include <linux/netfilter/nfnetlink_conntrack.h>
+static const struct nf_nat_protocol *
+nf_nat_proto_find_get(u_int8_t protonum)
+{
+ const struct nf_nat_protocol *p;
+
+ rcu_read_lock();
+ p = __nf_nat_proto_find(protonum);
+ if (!try_module_get(p->me))
+ p = &nf_nat_unknown_protocol;
+ rcu_read_unlock();
+
+ return p;
+}
+
+static void
+nf_nat_proto_put(const struct nf_nat_protocol *p)
+{
+ module_put(p->me);
+}
+
static const struct nla_policy protonat_nla_policy[CTA_PROTONAT_MAX+1] = {
[CTA_PROTONAT_PORT_MIN] = { .type = NLA_U16 },
[CTA_PROTONAT_PORT_MAX] = { .type = NLA_U16 },
private = &tmp;
}
#endif
+ memset(&info, 0, sizeof(info));
info.valid_hooks = t->valid_hooks;
memcpy(info.hook_entry, private->hook_entry,
sizeof(info.hook_entry));
kfree(net->ipv6.ip6_prohibit_entry);
kfree(net->ipv6.ip6_blk_hole_entry);
#endif
+ dst_entries_destroy(&net->ipv6.ip6_dst_ops);
}
static struct pernet_operations ip6_route_net_ops = {
xfrm6_fini();
fib6_gc_cleanup();
unregister_pernet_subsys(&ip6_route_net_ops);
+ dst_entries_destroy(&ip6_dst_blackhole_ops);
kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep);
}
struct seq_file *seq;
int rc = -ENOMEM;
- pd = kzalloc(GFP_KERNEL, sizeof(*pd));
+ pd = kzalloc(sizeof(*pd), GFP_KERNEL);
if (pd == NULL)
goto out;
if (!hash) {
*vmalloced = 1;
printk(KERN_WARNING "nf_conntrack: falling back to vmalloc.\n");
- hash = __vmalloc(sz, GFP_KERNEL | __GFP_ZERO, PAGE_KERNEL);
+ hash = __vmalloc(sz, GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO,
+ PAGE_KERNEL);
}
if (hash && nulls)
for (i = 0; i < MAX_NF_CT_PROTO; i++)
proto_array[i] = &nf_conntrack_l4proto_generic;
+
+ /* Before making proto_array visible to lockless readers,
+ * we must make sure its content is committed to memory.
+ */
+ smp_wmb();
+
nf_ct_protos[l4proto->l3proto] = proto_array;
} else if (nf_ct_protos[l4proto->l3proto][l4proto->l4proto] !=
&nf_conntrack_l4proto_generic) {
static void rds_loop_conn_free(void *arg)
{
struct rds_loop_connection *lc = arg;
+ unsigned long flags;
+
rdsdebug("lc %p\n", lc);
+ spin_lock_irqsave(&loop_conns_lock, flags);
list_del(&lc->loop_node);
+ spin_unlock_irqrestore(&loop_conns_lock, flags);
kfree(lc);
}
static void rds_tcp_conn_free(void *arg)
{
struct rds_tcp_connection *tc = arg;
+ unsigned long flags;
rdsdebug("freeing tc %p\n", tc);
+
+ spin_lock_irqsave(&rds_tcp_conn_lock, flags);
+ list_del(&tc->t_tcp_node);
+ spin_unlock_irqrestore(&rds_tcp_conn_lock, flags);
+
kmem_cache_free(rds_tcp_conn_slab, tc);
}
.populate = cgrp_populate,
#ifdef CONFIG_NET_CLS_CGROUP
.subsys_id = net_cls_subsys_id,
-#else
-#define net_cls_subsys_id net_cls_subsys.subsys_id
#endif
.module = THIS_MODULE,
};
static void em_text_destroy(struct tcf_proto *tp, struct tcf_ematch *m)
{
- textsearch_destroy(EM_TEXT_PRIV(m)->config);
+ if (EM_TEXT_PRIV(m) && EM_TEXT_PRIV(m)->config)
+ textsearch_destroy(EM_TEXT_PRIV(m)->config);
}
static int em_text_dump(struct sk_buff *skb, struct tcf_ematch *m)
case X25_FAC_CLASS_D:
switch (*p) {
case X25_FAC_CALLING_AE:
- if (p[1] > X25_MAX_DTE_FACIL_LEN)
- break;
+ if (p[1] > X25_MAX_DTE_FACIL_LEN || p[1] <= 1)
+ return 0;
dte_facs->calling_len = p[2];
memcpy(dte_facs->calling_ae, &p[3], p[1] - 1);
*vc_fac_mask |= X25_MASK_CALLING_AE;
break;
case X25_FAC_CALLED_AE:
- if (p[1] > X25_MAX_DTE_FACIL_LEN)
- break;
+ if (p[1] > X25_MAX_DTE_FACIL_LEN || p[1] <= 1)
+ return 0;
dte_facs->called_len = p[2];
memcpy(dte_facs->called_ae, &p[3], p[1] - 1);
*vc_fac_mask |= X25_MASK_CALLED_AE;
&x25->vc_facil_mask);
if (len > 0)
skb_pull(skb, len);
+ else
+ return -1;
/*
* Copy any Call User Data.
*/
WARN("__func__ should be used instead of gcc specific __FUNCTION__\n" . $herecurr);
}
-# check for semaphores used as mutexes
- if ($line =~ /^.\s*(DECLARE_MUTEX|init_MUTEX)\s*\(/) {
- WARN("mutexes are preferred for single holder semaphores\n" . $herecurr);
- }
-# check for semaphores used as mutexes
- if ($line =~ /^.\s*init_MUTEX_LOCKED\s*\(/) {
+# check for semaphores initialized locked
+ if ($line =~ /^.\s*sema_init.+,\W?0\W?\)/) {
WARN("consider using a completion\n" . $herecurr);
}
symval = sym_get_string_value(sym);
}
- newlen = strlen(res) + strlen(symval) + strlen(src);
+ newlen = strlen(res) + strlen(symval) + strlen(src) + 1;
if (newlen > reslen) {
reslen = newlen;
res = realloc(res, reslen);
If you are unsure as to whether this is required, answer N.
+config SECURITY_DMESG_RESTRICT
+ bool "Restrict unprivileged access to the kernel syslog"
+ default n
+ help
+ This enforces restrictions on unprivileged users reading the kernel
+ syslog via dmesg(8).
+
+ If this option is not selected, no restrictions will be enforced
+ unless the dmesg_restrict sysctl is explicitly set to (1).
+
+ If you are unsure how to answer this question, answer N.
+
config SECURITY
bool "Enable different security models"
depends on SYSFS
error = register_security(&apparmor_ops);
if (error) {
AA_ERROR("Unable to register AppArmor\n");
- goto register_security_out;
+ goto set_init_cxt_out;
}
/* Report that AppArmor successfully initialized */
return error;
+set_init_cxt_out:
+ aa_free_task_context(current->real_cred->security);
+
register_security_out:
aa_free_root_ns();
apparmor_enabled = 0;
return error;
-
}
security_initcall(apparmor_init);
return ns;
fail_unconfined:
- kzfree(ns->base.name);
+ kzfree(ns->base.hname);
fail_ns:
kzfree(ns);
return NULL;
{
if (type != SYSLOG_ACTION_OPEN && from_file)
return 0;
+ if (dmesg_restrict && !capable(CAP_SYS_ADMIN))
+ return -EPERM;
if ((type != SYSLOG_ACTION_READ_ALL &&
type != SYSLOG_ACTION_SIZE_BUFFER) && !capable(CAP_SYS_ADMIN))
return -EPERM;
control_cache_size, (struct hpi_control_cache_info *)
&phw->control_cache[0]
);
+ if (!phw->p_cache)
+ pao->has_control_cache = 0;
} else
pao->has_control_cache = 0;
interface->control_cache.size_in_bytes,
(struct hpi_control_cache_info *)
p_control_cache_virtual);
+ if (!phw->p_cache)
+ err = HPI_ERROR_MEMORY_ALLOC;
}
if (!err) {
err = hpios_locked_mem_get_phys_addr(&phw->
{
struct hpi_control_cache *p_cache =
kmalloc(sizeof(*p_cache), GFP_KERNEL);
+ if (!p_cache)
+ return NULL;
+ p_cache->p_info =
+ kmalloc(sizeof(*p_cache->p_info) * number_of_controls,
+ GFP_KERNEL);
+ if (!p_cache->p_info) {
+ kfree(p_cache);
+ return NULL;
+ }
p_cache->cache_size_in_bytes = size_in_bytes;
p_cache->control_count = number_of_controls;
p_cache->p_cache =
(struct hpi_control_cache_single *)pDSP_control_buffer;
p_cache->init = 0;
- p_cache->p_info =
- kmalloc(sizeof(*p_cache->p_info) * p_cache->control_count,
- GFP_KERNEL);
return p_cache;
}
{
struct dsp_spos_instance * ins = kzalloc(sizeof(struct dsp_spos_instance), GFP_KERNEL);
- if (ins == NULL)
+ if (ins == NULL)
return NULL;
/* better to use vmalloc for this big table */
- ins->symbol_table.nsymbols = 0;
ins->symbol_table.symbols = vmalloc(sizeof(struct dsp_symbol_entry) *
DSP_MAX_SYMBOLS);
- ins->symbol_table.highest_frag_index = 0;
-
- if (ins->symbol_table.symbols == NULL) {
+ ins->code.data = kmalloc(DSP_CODE_BYTE_SIZE, GFP_KERNEL);
+ ins->modules = kmalloc(sizeof(struct dsp_module_desc) * DSP_MAX_MODULES, GFP_KERNEL);
+ if (!ins->symbol_table.symbols || !ins->code.data || !ins->modules) {
cs46xx_dsp_spos_destroy(chip);
goto error;
}
-
+ ins->symbol_table.nsymbols = 0;
+ ins->symbol_table.highest_frag_index = 0;
ins->code.offset = 0;
ins->code.size = 0;
- ins->code.data = kmalloc(DSP_CODE_BYTE_SIZE, GFP_KERNEL);
-
- if (ins->code.data == NULL) {
- cs46xx_dsp_spos_destroy(chip);
- goto error;
- }
-
ins->nscb = 0;
ins->ntask = 0;
-
ins->nmodules = 0;
- ins->modules = kmalloc(sizeof(struct dsp_module_desc) * DSP_MAX_MODULES, GFP_KERNEL);
-
- if (ins->modules == NULL) {
- cs46xx_dsp_spos_destroy(chip);
- goto error;
- }
/* default SPDIF input sample rate
to 48000 khz */
/* set left and right validity bits and
default channel status */
- ins->spdif_csuv_default =
- ins->spdif_csuv_stream =
+ ins->spdif_csuv_default =
+ ins->spdif_csuv_stream =
/* byte 0 */ ((unsigned int)_wrap_all_bits( (SNDRV_PCM_DEFAULT_CON_SPDIF & 0xff)) << 24) |
/* byte 1 */ ((unsigned int)_wrap_all_bits( ((SNDRV_PCM_DEFAULT_CON_SPDIF >> 8) & 0xff)) << 16) |
/* byte 3 */ (unsigned int)_wrap_all_bits( (SNDRV_PCM_DEFAULT_CON_SPDIF >> 24) & 0xff) |
return ins;
error:
+ kfree(ins->modules);
+ kfree(ins->code.data);
+ vfree(ins->symbol_table.symbols);
kfree(ins);
return NULL;
}
static struct snd_pci_quirk cs420x_cfg_tbl[] = {
SND_PCI_QUIRK(0x10de, 0x0ac0, "MacBookPro 5,3", CS420X_MBP53),
+ SND_PCI_QUIRK(0x10de, 0x0d94, "MacBookAir 3,1(2)", CS420X_MBP55),
SND_PCI_QUIRK(0x10de, 0xcb79, "MacBookPro 5,5", CS420X_MBP55),
SND_PCI_QUIRK(0x10de, 0xcb89, "MacBookPro 7,1", CS420X_MBP55),
SND_PCI_QUIRK(0x8086, 0x7270, "IMac 27 Inch", CS420X_IMAC27),
static void lx_trigger_start(struct lx6464es *chip, struct lx_stream *lx_stream)
{
struct snd_pcm_substream *substream = lx_stream->stream;
- const int is_capture = lx_stream->is_capture;
+ const unsigned int is_capture = lx_stream->is_capture;
int err;
static void lx_trigger_stop(struct lx6464es *chip, struct lx_stream *lx_stream)
{
- const int is_capture = lx_stream->is_capture;
+ const unsigned int is_capture = lx_stream->is_capture;
int err;
snd_printd(LXP "stopping: stopping stream\n");
snd_pcm_uframes_t frame_pos;
enum lx_stream_status status; /* free, open, running, draining
* pause */
- int is_capture:1;
+ unsigned int is_capture:1;
};
struct lx_stream *lx_stream)
{
struct snd_pcm_substream *substream = lx_stream->stream;
- int is_capture = lx_stream->is_capture;
+ const unsigned int is_capture = lx_stream->is_capture;
int err;
unsigned long flags;
select SND_SOC_CQ0093VC if MFD_DAVINCI_VOICECODEC
select SND_SOC_CS42L51 if I2C
select SND_SOC_CS4270 if I2C
+ select SND_SOC_CX20442
select SND_SOC_DA7210 if I2C
- select SND_SOC_JZ4740 if SOC_JZ4740
+ select SND_SOC_JZ4740_CODEC if SOC_JZ4740
select SND_SOC_MAX98088 if I2C
select SND_SOC_MAX9877 if I2C
select SND_SOC_PCM3008
(1000000000 / ((rate * 1000) / samples))
#define US_TO_SAMPLES(rate, us) \
- (rate / (1000000 / us))
+ (rate / (1000000 / (us < 1000000 ? us : 1000000)))
#define UTHR_FROM_PERIOD_SIZE(samples, playrate, burstrate) \
((samples * 5000) / ((burstrate * 5000) / (burstrate - playrate)))
u8 *value)
{
struct tlv320dac33_priv *dac33 = snd_soc_codec_get_drvdata(codec);
- int val;
+ int val, ret = 0;
*value = reg & 0xff;
if (val < 0) {
dev_err(codec->dev, "Read failed (%d)\n", val);
value[0] = dac33_read_reg_cache(codec, reg);
+ ret = val;
} else {
value[0] = val;
dac33_write_reg_cache(codec, reg, val);
value[0] = dac33_read_reg_cache(codec, reg);
}
- return 0;
+ return ret;
}
static int dac33_write(struct snd_soc_codec *codec, unsigned int reg,
dac33_read_reg_cache(codec, DAC33_LINER_TO_RLO_VOL));
}
-static inline void dac33_read_id(struct snd_soc_codec *codec)
+static inline int dac33_read_id(struct snd_soc_codec *codec)
{
+ int i, ret = 0;
u8 reg;
- dac33_read(codec, DAC33_DEVICE_ID_MSB, ®);
- dac33_read(codec, DAC33_DEVICE_ID_LSB, ®);
- dac33_read(codec, DAC33_DEVICE_REV_ID, ®);
+ for (i = 0; i < 3; i++) {
+ ret = dac33_read(codec, DAC33_DEVICE_ID_MSB + i, ®);
+ if (ret < 0)
+ break;
+ }
+
+ return ret;
}
static inline void dac33_soft_power(struct snd_soc_codec *codec, int power)
/* Number of samples under i2c latency */
dac33->alarm_threshold = US_TO_SAMPLES(rate,
dac33->mode1_latency);
+ nsample_limit = DAC33_BUFFER_SIZE_SAMPLES -
+ dac33->alarm_threshold;
+
if (dac33->auto_fifo_config) {
if (period_size <= dac33->alarm_threshold)
/*
((dac33->alarm_threshold / period_size) +
(dac33->alarm_threshold % period_size ?
1 : 0));
+ else if (period_size > nsample_limit)
+ dac33->nsample = nsample_limit;
else
dac33->nsample = period_size;
} else {
*/
dac33->nsample_max = substream->runtime->buffer_size -
period_size;
- nsample_limit = DAC33_BUFFER_SIZE_SAMPLES -
- dac33->alarm_threshold;
+
if (dac33->nsample_max > nsample_limit)
dac33->nsample_max = nsample_limit;
dev_err(codec->dev, "Failed to power up codec: %d\n", ret);
goto err_power;
}
- dac33_read_id(codec);
+ ret = dac33_read_id(codec);
dac33_hard_power(codec, 0);
+ if (ret < 0) {
+ dev_err(codec->dev, "Failed to read chip ID: %d\n", ret);
+ ret = -ENODEV;
+ goto err_power;
+ }
+
/* Check if the IRQ number is valid and request it */
if (dac33->irq >= 0) {
ret = request_irq(dac33->irq, dac33_interrupt_handler,
{
struct tpa6130a2_data *data;
u8 val;
- int ret;
+ int ret = 0;
BUG_ON(tpa6130a2_client == NULL);
data = i2c_get_clientdata(tpa6130a2_client);
mutex_lock(&data->mutex);
- if (power) {
+ if (power && !data->power_state) {
/* Power on */
if (data->power_gpio >= 0)
gpio_set_value(data->power_gpio, 1);
val = tpa6130a2_read(TPA6130A2_REG_CONTROL);
val &= ~TPA6130A2_SWS;
tpa6130a2_i2c_write(TPA6130A2_REG_CONTROL, val);
- } else {
+ } else if (!power && data->power_state) {
/* set SWS */
val = tpa6130a2_read(TPA6130A2_REG_CONTROL);
val |= TPA6130A2_SWS;
{
switch (reg) {
case WM8900_REG_ID:
- case WM8900_REG_POWER1:
return 1;
default:
return 0;
return -ENODEV;
}
- /* Read back from the chip */
- reg = snd_soc_read(codec, WM8900_REG_POWER1);
- reg = (reg >> 12) & 0xf;
- dev_info(codec->dev, "WM8900 revision %d\n", reg);
-
wm8900_reset(codec);
/* Turn the chip on */
reg_r = reg & WM8993_DCS_DAC_WR_VAL_0_MASK;
break;
default:
- WARN(1, "Unknown DCS readback method");
+ WARN(1, "Unknown DCS readback method\n");
break;
}
static int tosa_startup(struct snd_pcm_substream *substream)
{
struct snd_soc_pcm_runtime *rtd = substream->private_data;
- struct snd_soc_codec *codec = rtd->card->codec;
+ struct snd_soc_codec *codec = rtd->codec;
/* check the jack status at stream startup */
tosa_ext_control(codec);
{
struct snd_soc_pcm_runtime *rtd =
container_of(dev, struct snd_soc_pcm_runtime, dev);
+ int ret;
- strict_strtol(buf, 10, &rtd->pmdown_time);
+ ret = strict_strtol(buf, 10, &rtd->pmdown_time);
+ if (ret)
+ return ret;
return count;
}
{ USB_ID(0x041e, 0x3000), 0, 1, 2, 1, 18, 0x0013 }, /* Extigy */
{ USB_ID(0x041e, 0x3020), 2, 1, 6, 6, 18, 0x0013 }, /* Audigy 2 NX */
{ USB_ID(0x041e, 0x3040), 2, 2, 6, 6, 2, 0x6e91 }, /* Live! 24-bit */
- { USB_ID(0x041e, 0x3042), 0, 1, 1, 1, 1, 0x000d }, /* Usb X-Fi */
+ { USB_ID(0x041e, 0x3042), 0, 1, 1, 1, 1, 0x000d }, /* Usb X-Fi S51 */
{ USB_ID(0x041e, 0x3048), 2, 2, 6, 6, 2, 0x6e91 }, /* Toshiba SB0500 */
};
if (value > 1)
return -EINVAL;
changed = value != mixer->audigy2nx_leds[index];
- err = snd_usb_ctl_msg(mixer->chip->dev,
+ if (mixer->chip->usb_id == USB_ID(0x041e, 0x3042))
+ err = snd_usb_ctl_msg(mixer->chip->dev,
+ usb_sndctrlpipe(mixer->chip->dev, 0), 0x24,
+ USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_OTHER,
+ !value, 0, NULL, 0, 100);
+ else
+ err = snd_usb_ctl_msg(mixer->chip->dev,
usb_sndctrlpipe(mixer->chip->dev, 0), 0x24,
USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_OTHER,
value, index + 2, NULL, 0, 100);
int i, err;
for (i = 0; i < ARRAY_SIZE(snd_audigy2nx_controls); ++i) {
+ /* USB X-Fi S51 doesn't have a CMSS LED */
+ if ((mixer->chip->usb_id == USB_ID(0x041e, 0x3042)) && i == 0)
+ continue;
if (i > 1 && /* Live24ext has 2 LEDs only */
(mixer->chip->usb_id == USB_ID(0x041e, 0x3040) ||
+ mixer->chip->usb_id == USB_ID(0x041e, 0x3042) ||
mixer->chip->usb_id == USB_ID(0x041e, 0x3048)))
break;
err = snd_ctl_add(mixer->chip->card,
if (mixer->chip->usb_id == USB_ID(0x041e, 0x3020) ||
mixer->chip->usb_id == USB_ID(0x041e, 0x3040) ||
+ mixer->chip->usb_id == USB_ID(0x041e, 0x3042) ||
mixer->chip->usb_id == USB_ID(0x041e, 0x3048)) {
if ((err = snd_audigy2nx_controls_create(mixer)) < 0)
return err;
if (!needs_knot)
return 0;
- subs->rate_list.count = count;
subs->rate_list.list = kmalloc(sizeof(int) * count, GFP_KERNEL);
+ if (!subs->rate_list.list)
+ return -ENOMEM;
+ subs->rate_list.count = count;
subs->rate_list.mask = 0;
count = 0;
list_for_each_entry(fp, &subs->fmt_list, list) {
SYNOPSIS
--------
[verse]
-'perf trace' {record <script> | report <script> [args] }
+'perf trace' [<options>]
+'perf trace' [<options>] record <script> [<record-options>] <command>
+'perf trace' [<options>] report <script> [script-args]
+'perf trace' [<options>] <script> <required-script-args> [<record-options>] <command>
+'perf trace' [<options>] <top-script> [script-args]
DESCRIPTION
-----------
available via 'perf trace -l'). The following variants allow you to
record and run those scripts:
- 'perf trace record <script>' to record the events required for 'perf
- trace report'. <script> is the name displayed in the output of
- 'perf trace --list' i.e. the actual script name minus any language
- extension.
+ 'perf trace record <script> <command>' to record the events required
+ for 'perf trace report'. <script> is the name displayed in the
+ output of 'perf trace --list' i.e. the actual script name minus any
+ language extension. If <command> is not specified, the events are
+ recorded using the -a (system-wide) 'perf record' option.
- 'perf trace report <script>' to run and display the results of
- <script>. <script> is the name displayed in the output of 'perf
+ 'perf trace report <script> [args]' to run and display the results
+ of <script>. <script> is the name displayed in the output of 'perf
trace --list' i.e. the actual script name minus any language
extension. The perf.data output from a previous run of 'perf trace
record <script>' is used and should be present for this command to
- succeed.
+ succeed. [args] refers to the (mainly optional) args expected by
+ the script.
+
+ 'perf trace <script> <required-script-args> <command>' to both
+ record the events required for <script> and to run the <script>
+ using 'live-mode' i.e. without writing anything to disk. <script>
+ is the name displayed in the output of 'perf trace --list' i.e. the
+ actual script name minus any language extension. If <command> is
+ not specified, the events are recorded using the -a (system-wide)
+ 'perf record' option. If <script> has any required args, they
+ should be specified before <command>. This mode doesn't allow for
+ optional script args to be specified; if optional script args are
+ desired, they can be specified using separate 'perf trace record'
+ and 'perf trace report' commands, with the stdout of the record step
+ piped to the stdin of the report script, using the '-o -' and '-i -'
+ options of the corresponding commands.
+
+ 'perf trace <top-script>' to both record the events required for
+ <top-script> and to run the <top-script> using 'live-mode'
+ i.e. without writing anything to disk. <top-script> is the name
+ displayed in the output of 'perf trace --list' i.e. the actual
+ script name minus any language extension; a <top-script> is defined
+ as any script name ending with the string 'top'.
+
+ [<record-options>] can be passed to the record steps of 'perf trace
+ record' and 'live-mode' variants; this isn't possible however for
+ <top-script> 'live-mode' or 'perf trace report' variants.
See the 'SEE ALSO' section for links to language-specific
information on how to write and run your own trace scripts.
OPTIONS
-------
+<command>...::
+ Any command you can specify in a shell.
+
-D::
--dump-raw-trace=::
Display verbose dump of the trace data.
Generate perf-trace.[ext] starter script for given language,
using current perf.data.
+-a::
+ Force system-wide collection. Scripts run without a <command>
+ normally use -a by default, while scripts run with a <command>
+ normally don't - this option allows the latter to be run in
+ system-wide mode.
+
+
SEE ALSO
--------
linkperf:perf-record[1], linkperf:perf-trace-perl[1],
static bool force, append_file;
-static const struct option options[] = {
+const struct option record_options[] = {
OPT_CALLBACK('e', "event", NULL, "event",
"event selector. use 'perf list' to list available events",
parse_events),
{
int i, j, err = -ENOMEM;
- argc = parse_options(argc, argv, options, record_usage,
+ argc = parse_options(argc, argv, record_options, record_usage,
PARSE_OPT_STOP_AT_NON_OPTION);
if (!argc && target_pid == -1 && target_tid == -1 &&
!system_wide && !cpu_list)
- usage_with_options(record_usage, options);
+ usage_with_options(record_usage, record_options);
if (force && append_file) {
fprintf(stderr, "Can't overwrite and append at the same time."
" You need to choose between -f and -A");
- usage_with_options(record_usage, options);
+ usage_with_options(record_usage, record_options);
} else if (append_file) {
write_mode = WRITE_APPEND;
} else {
if (thread_num <= 0) {
fprintf(stderr, "Can't find all threads of pid %d\n",
target_pid);
- usage_with_options(record_usage, options);
+ usage_with_options(record_usage, record_options);
}
} else {
all_tids=malloc(sizeof(pid_t));
static pid_t *all_tids = NULL;
static int thread_num = 0;
static bool inherit = false;
-static int profile_cpu = -1;
static int nr_cpus = 0;
static int realtime_prio = 0;
static bool group = false;
else
printf(" (all");
- if (profile_cpu != -1)
- printf(", cpu: %d)\n", profile_cpu);
+ if (cpu_list)
+ printf(", CPU%s: %s)\n", nr_cpus > 1 ? "s" : "", cpu_list);
else {
if (target_tid != -1)
printf(")\n");
else
- printf(", %d CPUs)\n", nr_cpus);
+ printf(", %d CPU%s)\n", nr_cpus, nr_cpus > 1 ? "s" : "");
}
printf("%-*.*s\n", win_width, win_width, graph_dotted_line);
static void start_counter(int i, int counter)
{
struct perf_event_attr *attr;
- int cpu;
+ int cpu = -1;
int thread_index;
- cpu = profile_cpu;
- if (target_tid == -1 && profile_cpu == -1)
+ if (target_tid == -1)
cpu = cpumap[i];
attr = attrs + counter;
#include "util/symbol.h"
#include "util/thread.h"
#include "util/trace-event.h"
+#include "util/parse-options.h"
#include "util/util.h"
static char const *script_name;
static bool debug_mode;
static u64 last_timestamp;
static u64 nr_unordered;
+extern const struct option record_options[];
static int default_start_script(const char *script __unused,
int argc __unused,
{
struct script_desc *s = zalloc(sizeof(*s));
- if (s != NULL)
+ if (s != NULL && name)
s->name = strdup(name);
return s;
static void script_desc__delete(struct script_desc *s)
{
free(s->name);
+ free(s->half_liner);
+ free(s->args);
free(s);
}
return path;
}
+static bool is_top_script(const char *script_path)
+{
+ return ends_with((char *)script_path, "top") == NULL ? false : true;
+}
+
+static int has_required_arg(char *script_path)
+{
+ struct script_desc *desc;
+ int n_args = 0;
+ char *p;
+
+ desc = script_desc__new(NULL);
+
+ if (read_script_info(desc, script_path))
+ goto out;
+
+ if (!desc->args)
+ goto out;
+
+ for (p = desc->args; *p; p++)
+ if (*p == '<')
+ n_args++;
+out:
+ script_desc__delete(desc);
+
+ return n_args;
+}
+
static const char * const trace_usage[] = {
- "perf trace [<options>] <command>",
+ "perf trace [<options>]",
+ "perf trace [<options>] record <script> [<record-options>] <command>",
+ "perf trace [<options>] report <script> [script-args]",
+ "perf trace [<options>] <script> [<record-options>] <command>",
+ "perf trace [<options>] <top-script> [script-args]",
NULL
};
OPT_END()
};
+static bool have_cmd(int argc, const char **argv)
+{
+ char **__argv = malloc(sizeof(const char *) * argc);
+
+ if (!__argv)
+ die("malloc");
+ memcpy(__argv, argv, sizeof(const char *) * argc);
+ argc = parse_options(argc, (const char **)__argv, record_options,
+ NULL, PARSE_OPT_STOP_AT_NON_OPTION);
+ free(__argv);
+
+ return argc != 0;
+}
+
int cmd_trace(int argc, const char **argv, const char *prefix __used)
{
+ char *rec_script_path = NULL;
+ char *rep_script_path = NULL;
struct perf_session *session;
- const char *suffix = NULL;
+ char *script_path = NULL;
const char **__argv;
- char *script_path;
- int i, err;
+ bool system_wide;
+ int i, j, err;
- if (argc >= 2 && strncmp(argv[1], "rec", strlen("rec")) == 0) {
- if (argc < 3) {
- fprintf(stderr,
- "Please specify a record script\n");
- return -1;
- }
- suffix = RECORD_SUFFIX;
+ setup_scripting();
+
+ argc = parse_options(argc, argv, options, trace_usage,
+ PARSE_OPT_STOP_AT_NON_OPTION);
+
+ if (argc > 1 && !strncmp(argv[0], "rec", strlen("rec"))) {
+ rec_script_path = get_script_path(argv[1], RECORD_SUFFIX);
+ if (!rec_script_path)
+ return cmd_record(argc, argv, NULL);
}
- if (argc >= 2 && strncmp(argv[1], "rep", strlen("rep")) == 0) {
- if (argc < 3) {
+ if (argc > 1 && !strncmp(argv[0], "rep", strlen("rep"))) {
+ rep_script_path = get_script_path(argv[1], REPORT_SUFFIX);
+ if (!rep_script_path) {
fprintf(stderr,
- "Please specify a report script\n");
+ "Please specify a valid report script"
+ "(see 'perf trace -l' for listing)\n");
return -1;
}
- suffix = REPORT_SUFFIX;
}
/* make sure PERF_EXEC_PATH is set for scripts */
perf_set_argv_exec_path(perf_exec_path());
- if (!suffix && argc >= 2 && strncmp(argv[1], "-", strlen("-")) != 0) {
- char *record_script_path, *report_script_path;
+ if (argc && !script_name && !rec_script_path && !rep_script_path) {
int live_pipe[2];
+ int rep_args;
pid_t pid;
- record_script_path = get_script_path(argv[1], RECORD_SUFFIX);
- if (!record_script_path) {
- fprintf(stderr, "record script not found\n");
- return -1;
+ rec_script_path = get_script_path(argv[0], RECORD_SUFFIX);
+ rep_script_path = get_script_path(argv[0], REPORT_SUFFIX);
+
+ if (!rec_script_path && !rep_script_path) {
+ fprintf(stderr, " Couldn't find script %s\n\n See perf"
+ " trace -l for available scripts.\n", argv[0]);
+ usage_with_options(trace_usage, options);
}
- report_script_path = get_script_path(argv[1], REPORT_SUFFIX);
- if (!report_script_path) {
- fprintf(stderr, "report script not found\n");
- return -1;
+ if (is_top_script(argv[0])) {
+ rep_args = argc - 1;
+ } else {
+ int rec_args;
+
+ rep_args = has_required_arg(rep_script_path);
+ rec_args = (argc - 1) - rep_args;
+ if (rec_args < 0) {
+ fprintf(stderr, " %s script requires options."
+ "\n\n See perf trace -l for available "
+ "scripts and options.\n", argv[0]);
+ usage_with_options(trace_usage, options);
+ }
}
if (pipe(live_pipe) < 0) {
}
if (!pid) {
+ system_wide = true;
+ j = 0;
+
dup2(live_pipe[1], 1);
close(live_pipe[0]);
- __argv = malloc(6 * sizeof(const char *));
- __argv[0] = "/bin/sh";
- __argv[1] = record_script_path;
- __argv[2] = "-q";
- __argv[3] = "-o";
- __argv[4] = "-";
- __argv[5] = NULL;
+ if (!is_top_script(argv[0]))
+ system_wide = !have_cmd(argc - rep_args,
+ &argv[rep_args]);
+
+ __argv = malloc((argc + 6) * sizeof(const char *));
+ if (!__argv)
+ die("malloc");
+
+ __argv[j++] = "/bin/sh";
+ __argv[j++] = rec_script_path;
+ if (system_wide)
+ __argv[j++] = "-a";
+ __argv[j++] = "-q";
+ __argv[j++] = "-o";
+ __argv[j++] = "-";
+ for (i = rep_args + 1; i < argc; i++)
+ __argv[j++] = argv[i];
+ __argv[j++] = NULL;
execvp("/bin/sh", (char **)__argv);
+ free(__argv);
exit(-1);
}
dup2(live_pipe[0], 0);
close(live_pipe[1]);
- __argv = malloc((argc + 3) * sizeof(const char *));
- __argv[0] = "/bin/sh";
- __argv[1] = report_script_path;
- for (i = 2; i < argc; i++)
- __argv[i] = argv[i];
- __argv[i++] = "-i";
- __argv[i++] = "-";
- __argv[i++] = NULL;
+ __argv = malloc((argc + 4) * sizeof(const char *));
+ if (!__argv)
+ die("malloc");
+ j = 0;
+ __argv[j++] = "/bin/sh";
+ __argv[j++] = rep_script_path;
+ for (i = 1; i < rep_args + 1; i++)
+ __argv[j++] = argv[i];
+ __argv[j++] = "-i";
+ __argv[j++] = "-";
+ __argv[j++] = NULL;
execvp("/bin/sh", (char **)__argv);
+ free(__argv);
exit(-1);
}
- if (suffix) {
- script_path = get_script_path(argv[2], suffix);
- if (!script_path) {
- fprintf(stderr, "script not found\n");
- return -1;
- }
-
- __argv = malloc((argc + 1) * sizeof(const char *));
- __argv[0] = "/bin/sh";
- __argv[1] = script_path;
- for (i = 3; i < argc; i++)
- __argv[i - 1] = argv[i];
- __argv[argc - 1] = NULL;
+ if (rec_script_path)
+ script_path = rec_script_path;
+ if (rep_script_path)
+ script_path = rep_script_path;
+
+ if (script_path) {
+ system_wide = false;
+ j = 0;
+
+ if (rec_script_path)
+ system_wide = !have_cmd(argc - 1, &argv[1]);
+
+ __argv = malloc((argc + 2) * sizeof(const char *));
+ if (!__argv)
+ die("malloc");
+ __argv[j++] = "/bin/sh";
+ __argv[j++] = script_path;
+ if (system_wide)
+ __argv[j++] = "-a";
+ for (i = 2; i < argc; i++)
+ __argv[j++] = argv[i];
+ __argv[j++] = NULL;
execvp("/bin/sh", (char **)__argv);
+ free(__argv);
exit(-1);
}
- setup_scripting();
-
- argc = parse_options(argc, argv, options, trace_usage,
- PARSE_OPT_STOP_AT_NON_OPTION);
-
if (symbol__init() < 0)
return -1;
if (!script_name)
#!/bin/bash
-perf record -a -e raw_syscalls:sys_exit $@
+perf record -e raw_syscalls:sys_exit $@
#!/bin/bash
-perf record -a -e syscalls:sys_enter_read -e syscalls:sys_enter_write $@
+perf record -e syscalls:sys_enter_read -e syscalls:sys_enter_write $@
#!/bin/bash
-perf record -a -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@
+perf record -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@
#!/bin/bash
-perf record -a -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@
+perf record -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@
#!/bin/bash
-perf record -a -e sched:sched_switch -e sched:sched_wakeup $@
+perf record -e sched:sched_switch -e sched:sched_wakeup $@
#!/bin/bash
-perf record -a -e workqueue:workqueue_creation -e workqueue:workqueue_destruction -e workqueue:workqueue_execution -e workqueue:workqueue_insertion $@
+perf record -e workqueue:workqueue_creation -e workqueue:workqueue_destruction -e workqueue:workqueue_execution -e workqueue:workqueue_insertion $@
#!/bin/bash
-perf record -a -e raw_syscalls:sys_exit $@
+perf record -e raw_syscalls:sys_exit $@
#!/bin/bash
-perf record -a -e syscalls:sys_enter_futex -e syscalls:sys_exit_futex $@
+perf record -e syscalls:sys_enter_futex -e syscalls:sys_exit_futex $@
#!/bin/bash
-perf record -a -e net:net_dev_xmit -e net:net_dev_queue \
+perf record -e net:net_dev_xmit -e net:net_dev_queue \
-e net:netif_receive_skb -e net:netif_rx \
-e skb:consume_skb -e skb:kfree_skb \
-e skb:skb_copy_datagram_iovec -e napi:napi_poll \
#!/bin/bash
-perf record -m 16384 -a -e sched:sched_wakeup -e sched:sched_wakeup_new -e sched:sched_switch -e sched:sched_migrate_task $@
+perf record -m 16384 -e sched:sched_wakeup -e sched:sched_wakeup_new -e sched:sched_switch -e sched:sched_migrate_task $@
#!/bin/bash
-perf record -a -e raw_syscalls:sys_enter $@
+perf record -e raw_syscalls:sys_enter $@
#!/bin/bash
-perf record -a -e raw_syscalls:sys_enter $@
+perf record -e raw_syscalls:sys_enter $@
#!/bin/bash
-perf record -a -e raw_syscalls:sys_enter $@
+perf record -e raw_syscalls:sys_enter $@
return rc;
}
+static const char yes[] = "Yes", no[] = "No";
+
bool ui__dialog_yesno(const char *msg)
{
/* newtWinChoice should really be accepting const char pointers... */
- char yes[] = "Yes", no[] = "No";
- return newtWinChoice(NULL, yes, no, (char *)msg) == 1;
+ return newtWinChoice(NULL, (char *)yes, (char *)no, (char *)msg) == 1;
}
.section .init.ramfs.info,"a"
.globl __initramfs_size
__initramfs_size:
-#ifdef CONFIG_32BIT
- .long __irf_end - __irf_start
-#else
+#ifdef CONFIG_64BIT
.quad __irf_end - __irf_start
+#else
+ .long __irf_end - __irf_start
#endif
git://bbs.cooldavid.org / net-next-2.6.git / commitdiff