}
}
ioat->tail += i;
- BUG_ON(!seen_current); /* no active descs have written a completion? */
+ BUG_ON(active && !seen_current); /* no active descs have written a completion? */
chan->last_completion = phys_complete;
- if (ioat->head == ioat->tail) {
+
+ active = ioat2_ring_active(ioat);
+ if (active == 0) {
dev_dbg(to_dev(chan), "%s: cancel completion timeout\n",
__func__);
clear_bit(IOAT_COMPLETION_PENDING, &chan->state);
mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
}
+ /* 5 microsecond delay per pending descriptor */
+ writew(min((5 * active), IOAT_INTRDELAY_MASK),
+ chan->device->reg_base + IOAT_INTRDELAY_OFFSET);
}
-static void ioat3_cleanup(struct ioat2_dma_chan *ioat)
+/* try to cleanup, but yield (via spin_trylock) to incoming submissions
+ * with the expectation that we will immediately poll again shortly
+ */
+static void ioat3_cleanup_poll(struct ioat2_dma_chan *ioat)
{
struct ioat_chan_common *chan = &ioat->base;
unsigned long phys_complete;
spin_unlock_bh(&chan->cleanup_lock);
}
+/* run cleanup now because we already delayed the interrupt via INTRDELAY */
+static void ioat3_cleanup_sync(struct ioat2_dma_chan *ioat)
+{
+ struct ioat_chan_common *chan = &ioat->base;
+ unsigned long phys_complete;
+
+ prefetch(chan->completion);
+
+ spin_lock_bh(&chan->cleanup_lock);
+ if (!ioat_cleanup_preamble(chan, &phys_complete)) {
+ spin_unlock_bh(&chan->cleanup_lock);
+ return;
+ }
+ spin_lock_bh(&ioat->ring_lock);
+
+ __cleanup(ioat, phys_complete);
+
+ spin_unlock_bh(&ioat->ring_lock);
+ spin_unlock_bh(&chan->cleanup_lock);
+}
+
static void ioat3_cleanup_tasklet(unsigned long data)
{
struct ioat2_dma_chan *ioat = (void *) data;
- ioat3_cleanup(ioat);
- writew(IOAT_CHANCTRL_RUN | IOAT3_CHANCTRL_COMPL_DCA_EN,
- ioat->base.reg_base + IOAT_CHANCTRL_OFFSET);
+ ioat3_cleanup_sync(ioat);
+ writew(IOAT_CHANCTRL_RUN, ioat->base.reg_base + IOAT_CHANCTRL_OFFSET);
}
static void ioat3_restart_channel(struct ioat2_dma_chan *ioat)
{
struct ioat_chan_common *chan = &ioat->base;
unsigned long phys_complete;
- u32 status;
-
- status = ioat_chansts(chan);
- if (is_ioat_active(status) || is_ioat_idle(status))
- ioat_suspend(chan);
- while (is_ioat_active(status) || is_ioat_idle(status)) {
- status = ioat_chansts(chan);
- cpu_relax();
- }
+ ioat2_quiesce(chan, 0);
if (ioat_cleanup_preamble(chan, &phys_complete))
__cleanup(ioat, phys_complete);
if (ioat_is_complete(c, cookie, done, used) == DMA_SUCCESS)
return DMA_SUCCESS;
- ioat3_cleanup(ioat);
+ ioat3_cleanup_poll(ioat);
return ioat_is_complete(c, cookie, done, used);
}
num_descs = ioat2_xferlen_to_descs(ioat, len);
/* we need 2x the number of descriptors to cover greater than 3
- * sources
+ * sources (we need 1 extra source in the q-only continuation
+ * case and 3 extra sources in the p+q continuation case.
*/
- if (src_cnt > 3 || flags & DMA_PREP_CONTINUE) {
+ if (src_cnt + dmaf_p_disabled_continue(flags) > 3 ||
+ (dmaf_continue(flags) && !dmaf_p_disabled_continue(flags))) {
with_ext = 1;
num_descs *= 2;
} else
return 0;
}
+static int ioat3_reset_hw(struct ioat_chan_common *chan)
+{
+ /* throw away whatever the channel was doing and get it
+ * initialized, with ioat3 specific workarounds
+ */
+ struct ioatdma_device *device = chan->device;
+ struct pci_dev *pdev = device->pdev;
+ u32 chanerr;
+ u16 dev_id;
+ int err;
+
+ ioat2_quiesce(chan, msecs_to_jiffies(100));
+
+ chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
+ writel(chanerr, chan->reg_base + IOAT_CHANERR_OFFSET);
+
+ /* -= IOAT ver.3 workarounds =- */
+ /* Write CHANERRMSK_INT with 3E07h to mask out the errors
+ * that can cause stability issues for IOAT ver.3, and clear any
+ * pending errors
+ */
+ pci_write_config_dword(pdev, IOAT_PCI_CHANERRMASK_INT_OFFSET, 0x3e07);
+ err = pci_read_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, &chanerr);
+ if (err) {
+ dev_err(&pdev->dev, "channel error register unreachable\n");
+ return err;
+ }
+ pci_write_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, chanerr);
+
+ /* Clear DMAUNCERRSTS Cfg-Reg Parity Error status bit
+ * (workaround for spurious config parity error after restart)
+ */
+ pci_read_config_word(pdev, IOAT_PCI_DEVICE_ID_OFFSET, &dev_id);
+ if (dev_id == PCI_DEVICE_ID_INTEL_IOAT_TBG0)
+ pci_write_config_dword(pdev, IOAT_PCI_DMAUNCERRSTS_OFFSET, 0x10);
+
+ return ioat2_reset_sync(chan, msecs_to_jiffies(200));
+}
+
int __devinit ioat3_dma_probe(struct ioatdma_device *device, int dca)
{
struct pci_dev *pdev = device->pdev;
struct ioat_chan_common *chan;
bool is_raid_device = false;
int err;
- u16 dev_id;
u32 cap;
device->enumerate_channels = ioat2_enumerate_channels;
+ device->reset_hw = ioat3_reset_hw;
device->self_test = ioat3_dma_self_test;
dma = &device->common;
dma->device_prep_dma_memcpy = ioat2_dma_prep_memcpy_lock;
dma->device_prep_dma_xor_val = NULL;
#endif
- /* -= IOAT ver.3 workarounds =- */
- /* Write CHANERRMSK_INT with 3E07h to mask out the errors
- * that can cause stability issues for IOAT ver.3
- */
- pci_write_config_dword(pdev, IOAT_PCI_CHANERRMASK_INT_OFFSET, 0x3e07);
-
- /* Clear DMAUNCERRSTS Cfg-Reg Parity Error status bit
- * (workaround for spurious config parity error after restart)
- */
- pci_read_config_word(pdev, IOAT_PCI_DEVICE_ID_OFFSET, &dev_id);
- if (dev_id == PCI_DEVICE_ID_INTEL_IOAT_TBG0)
- pci_write_config_dword(pdev, IOAT_PCI_DMAUNCERRSTS_OFFSET, 0x10);
-
err = ioat_probe(device);
if (err)
return err;
git://bbs.cooldavid.org / net-next-2.6.git / blobdiff