SSE vs. stack alignment vs. pthread

[Daniel Eischen]

On Wed, 24 Nov 2004, Craig Boston wrote:

> First of all, I'd like to apologize for cross-posting to -hackers and
> -threads.  I'm not sure yet if this is an application bug, a gcc
> bug, or a pthreads bug, so here goes...
>
> I'm currently working on the audacity port.  It's up to 1.2.3, but I
> want to get a problem I've observed with 1.2.2 resolved to make sure
> that it doesn't crop up later or affect other software...
>
> Long story short, audacity is a threaded program.  A straight compile of
> 1.2.2 results in a 100% reproducible bus error that happens on multiple
> Pentium-4 machines (5.3-STABLE).  It always happens at this instruction:
>
> 0x081807c4: movaps %xmm0,0xffffff68(%ebp)
>
> Now, at that time ebp is 0xbfadc6c0, so ebp+0xffffff68 (-0x152) is 0xbfadc56e.
> Oops, that's not 16-byte aligned like SSE wants.  The offsets vary sligthly
> depending on the compile flags, etc., but the result is always the same --
> SIGBUS.

Tor Egge reported similar problem to me yesterday.  I haven't had a chance
to test his patch, but this supposedly fixes it.


Index: lib/libc/i386/gen/makecontext.c
===================================================================
RCS file: /home/ncvs/src/lib/libc/i386/gen/makecontext.c,v
retrieving revision 1.4
diff -u -r1.4 makecontext.c
--- lib/libc/i386/gen/makecontext.c     2 Jul 2004 14:19:44 -0000       1.4
+++ lib/libc/i386/gen/makecontext.c     22 Nov 2004 22:51:49 -0000
@@ -118,7 +118,9 @@
 		 * address, _ctx_start, and ucp) and argc arguments.
 		 * We allow the arguments to be pointers also.
 		 */
-		stack_top = stack_top - (sizeof(intptr_t) * (3 + argc));
+		stack_top = stack_top - (sizeof(intptr_t) * (1 + argc));
+		stack_top -= ((long) stack_top & 15); /* 16 bytes alignment */
+		stack_top -= sizeof(intptr_t) * 2;
 		argp = (intptr_t *)stack_top;

 		/*