]> bbs.cooldavid.org Git - net-next-2.6.git/commit
vsprintf: Change struct printf_spec.precision from s8 to s16
authorJoe Perches <joe@perches.com>
Wed, 14 Apr 2010 16:27:40 +0000 (09:27 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Wed, 14 Apr 2010 17:32:35 +0000 (10:32 -0700)
commit4e310fda91cb095915395f811d10b2c900c9589e
tree5e4d6b2bd76abe6f2bb0db12d6d82f697007d1e5
parent2ba3abd8186f24c7fb418927025b4e2120e3a362
vsprintf: Change struct printf_spec.precision from s8 to s16

Commit ef0658f3de484bf9b173639cd47544584e01efa5 changed precision
from int to s8.

There is existing kernel code that uses a larger precision.

An example from the audit code:
vsnprintf(...,..., " msg='%.1024s'", (char *)data);
which overflows precision and truncates to nothing.

Extending precision size fixes the audit system issue.

Other changes:

Change the size of the struct printf_spec.type from u16 to u8 so
sizeof(struct printf_spec) stays as small as possible.
Reorder the struct members so sizeof(struct printf_spec) remains 64 bits
without alignment holes.
Document the struct members a bit more.

Original-patch-by: Eric Paris <eparis@redhat.com>
Signed-off-by: Joe Perches <joe@perches.com>
Tested-by: Justin P. Mattock <justinmattock@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
lib/vsprintf.c