[PATCH] Fix AMD64 backtrace

[Andreas Jaeger]

Andreas Jaeger <aj@suse.de> writes:

> Jakub Jelinek <jakub@redhat.com> writes:
> [...]
>>> libc/sysdeps/x86_64/__longjmp.S
>
> What's the best way to handle the destroying of the registers with
> CFI?  Should we just mark the registers with "cfi_undefined"?

I'm appending a proposed patch for __longjmp.S.  Or can anybody
envision a better way doing this?

Andreas

2004-01-10  Andreas Jaeger  <aj@suse.de>

	* sysdeps/x86_64/__longjmp.S: Add cfi directives.

============================================================
Index: sysdeps/x86_64/__longjmp.S
--- sysdeps/x86_64/__longjmp.S	19 Sep 2001 10:12:07 -0000	1.1
+++ sysdeps/x86_64/__longjmp.S	10 Jan 2004 19:47:01 -0000
@@ -1,4 +1,4 @@
-/* Copyright (C) 2001 Free Software Foundation, Inc.
+/* Copyright (C) 2001, 2004 Free Software Foundation, Inc.
    This file is part of the GNU C Library.
 
    The GNU C Library is free software; you can redistribute it and/or
@@ -28,11 +28,17 @@
 ENTRY(__longjmp)
 	/* Restore registers.  */
 	movq (JB_RBX*8)(%rdi),%rbx
+	cfi_undefined(%rbx)
 	movq (JB_RBP*8)(%rdi),%rbp
+	cfi_undefined(%rbp)
 	movq (JB_R12*8)(%rdi),%r12
+	cfi_undefined(%r12)
 	movq (JB_R13*8)(%rdi),%r13
+	cfi_undefined(%r13)
 	movq (JB_R14*8)(%rdi),%r14
+	cfi_undefined(%r14)
 	movq (JB_R15*8)(%rdi),%r15
+	cfi_undefined(%r15)
 	/* Set return value for setjmp.  */
 	test %esi,%esi
 	mov $01,%eax
@@ -40,5 +46,6 @@ ENTRY(__longjmp)
 	mov %esi, %eax
 	movq (JB_PC*8)(%rdi),%rdx
 	movq (JB_RSP*8)(%rdi),%rsp
+	cfi_undefined(%rsp)
 	jmpq *%rdx
 END (BP_SYM (__longjmp))

-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj
  SuSE Linux AG, Maxfeldstr. 5, 90409 Nürnberg, Germany
   GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 188 bytes
Desc: not available
URL: <http://sourceware.org/pipermail/libc-hacker/attachments/20040110/450eb0d0/attachment.sig>