Re: [PATCH] i386: fix stack alignment for signal handlers
From: Markus F.X.J. Oberhumer
Date: Mon Oct 10 2005 - 19:30:50 EST
I've just seen that Linus has merged my second patch, so here is one more
missing piece to fix ia64 in ia32 emulation as well.
p.s. this patch has not been tested due to lack of hardware
Andi Kleen wrote:
On Sunday 09 October 2005 18:54, Markus F.X.J. Oberhumer wrote:
Here is a somewhat simplified version of my previous patch with
Attached is also a new small user-space test program which does not
depend on any special gcc features and should trigger the problem on all
I already have a version of the patch in my queue, but it's not a strict
bugfix so it's only for after 2.6.14.
Markus Oberhumer, <markus@xxxxxxxxxxxxx>, http://www.oberhumer.com/
[PATCH] i386: fix stack alignment for signal handlers (ia64)
This fixes the setup of the alignment of the signal frame, so that all
signal handlers are run with a properly aligned stack frame.
The current code "over-aligns" the stack pointer so that the stack frame
is effectively always mis-aligned by 4 bytes. But what we really want
is that on function entry ((sp + 4) & 15) == 0, which matches what would
happen if the stack were aligned before a "call" instruction.
[ This patch fixes ia64. i386 and x86_64 are already fixed by
git commit d347f372273c2b3d86a66e2e1c94c790c208e166 ]
Signed-off-by: Markus F.X.J. Oberhumer <markus@xxxxxxxxxxxxx>
@@ -810,7 +810,11 @@
/* Legacy stack switching not supported */
- return (void __user *)((esp - frame_size) & -8ul);
+ esp -= frame_size;
+ /* Align the stack pointer according to the i386 ABI,
+ * i.e. so that on function entry ((sp + 4) & 15) == 0. */
+ esp = ((esp + 4) & -16ul) - 4;
+ return (void __user *) esp;