Hi Willy,

This is a v3 revision.

The ABI mandates that the %esp register must be a multiple of 16 when
executing a 'call' instruction.

Commit 2ab446336b17 ("tools/nolibc: i386: shrink _start with _start_c")
simplified the _start function, but it didn't take care of the %esp
alignment, causing SIGSEGV on SSE and AVX programs that use aligned move
instruction (e.g., movdqa, movaps, and vmovdqa).

  $eax   : 0x56559000  →  0x00003f90
  $ebx   : 0x56559000  →  0x00003f90
  $ecx   : 0x1
  $edx   : 0xf7fcaaa0  →   endbr32 
  $esp   : 0xffffcdbc  →  0x00000001
  $ebp   : 0x0
  $esi   : 0xffffce7c  →  0xffffd096
  $edi   : 0x56556060  →  <_start+0> xor %ebp, %ebp
  $eip   : 0x56556489  →  <sse_pq_add+25> movaps %xmm0, 0x30(%esp)

    <sse_pq_add+11>  pop    %eax
    <sse_pq_add+12>  add    $0x2b85, %eax
    <sse_pq_add+18>  movups -0x1fd0(%eax), %xmm0
  → <sse_pq_add+25>  movaps %xmm0, 0x30(%esp)     <== trapping instruction
    <sse_pq_add+30>  movups -0x1fe0(%eax), %xmm1
    <sse_pq_add+37>  movaps %xmm1, 0x20(%esp)
    <sse_pq_add+42>  movups -0x1ff0(%eax), %xmm2
    <sse_pq_add+49>  movaps %xmm2, 0x10(%esp)
    <sse_pq_add+54>  movups -0x2000(%eax), %xmm3

  [#0] Id 1, Name: "test", stopped 0x56556489 in sse_pq_add (), reason: SIGSEGV

  (gdb) bt
  #0  0x56556489 in sse_pq_add ()

Ensure the %esp is a multiple of 16 when executing the call instruction.

Changes since v2:
  - Avoid over-estimating the stack size (per Willy).
  - Add the link to a test program to validate the alignment (per Zhangjin).

Changes since v1:
  - Change 'sub $12, %esp' to 'sub $(16 - 4), %esp' (per Zhangjin).
  - Fix the reference format (per Thomas).
  - Explain more about the logic behind the fix (per Thomas).
  - Append an Acked-by tag from Thomas.

Signed-off-by: Ammar Faizi <ammarfaizi2@gnuweeb.org>
---

Ammar Faizi (1):
  tools/nolibc: i386: Fix a stack misalign bug on _start

 tools/include/nolibc/arch-i386.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)


base-commit: 6269320850097903b30be8f07a5c61d9f7592393
-- 
Ammar Faizi

On Wed, Aug 30, 2023 at 08:02:22AM +0700, Ammar Faizi wrote:
> Hi Willy,
> 
> This is a v3 revision.
> 
> The ABI mandates that the %esp register must be a multiple of 16 when
> executing a 'call' instruction.
(...)

Thanks Ammar, now queued into the fixes branch.

Willy

On Wed, 30 Aug 2023 05:45:16 +0200, Willy Tarreau wrote:
> Thanks Ammar, now queued into the fixes branch.

Reviewed-by: Alviro Iskandar Setiawan <alviro.iskandar@gnuweeb.org>

-- Viro