]> git.baikalelectronics.ru Git - kernel.git/commit
x86/asm/entry: Replace this_cpu_sp0() with current_top_of_stack() and fix it on x86_32
authorAndy Lutomirski <luto@amacapital.net>
Sat, 7 Mar 2015 01:50:19 +0000 (17:50 -0800)
committerIngo Molnar <mingo@kernel.org>
Sat, 7 Mar 2015 08:34:03 +0000 (09:34 +0100)
commit694b1640a42c650012e6064f7a0be74b5de04227
tree3de032435c43db84cdfc4f7e46df4d118b86fe82
parentb80192ef96a1e0ed57fef66a9a3f8d59594fec91
x86/asm/entry: Replace this_cpu_sp0() with current_top_of_stack() and fix it on x86_32

I broke 32-bit kernels.  The implementation of sp0 was correct
as far as I can tell, but sp0 was much weirder on x86_32 than I
realized.  It has the following issues:

 - Init's sp0 is inconsistent with everything else's: non-init tasks
   are offset by 8 bytes.  (I have no idea why, and the comment is unhelpful.)

 - vm86 does crazy things to sp0.

Fix it up by replacing this_cpu_sp0() with
current_top_of_stack() and using a new percpu variable to track
the top of the stack on x86_32.

Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: ca8e1e3cc619 ("x86/asm/entry: Switch all C consumers of kernel_stack to this_cpu_sp0()")
Link: http://lkml.kernel.org/r/d09dbe270883433776e0cbee3c7079433349e96d.1425692936.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
arch/x86/include/asm/processor.h
arch/x86/include/asm/thread_info.h
arch/x86/kernel/cpu/common.c
arch/x86/kernel/process_32.c
arch/x86/kernel/smpboot.c
arch/x86/kernel/traps.c