Bug 10255 - no .eh_frame_hdr table will be created.
Summary: no .eh_frame_hdr table will be created.
Status: RESOLVED FIXED
Alias: None
Product: binutils
Classification: Unclassified
Component: gas (show other bugs)
Version: 2.19
: P2 normal
Target Milestone: ---
Assignee: Jakub Jelinek
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2009-06-09 10:51 UTC by Joost VandeVondele
Modified: 2009-06-29 09:55 UTC (History)
3 users (show)

See Also:
Host:
Target: x86_64-unknown-linux-gnu
Build:
Last reconfirmed:

Attachments
z2.s.bz2 (53.97 KB, application/octet-stream)
2009-06-09 11:43 UTC, Jakub Jelinek 		Details
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Joost VandeVondele 2009-06-09 10:51:02 UTC 
as discussed in the gcc bugzilla, linking a recent version of cp2k with trunk
gcc produces an number of unexpected errors:

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40332

this happens with older and more recent (2.19.1) ld. 

http://www.pci.uzh.ch/vandevondele/tmp/PR40332.tgz

contains the relevant object files and libraries, plus a README that provides
the exact command line I use
Comment 1 Jakub Jelinek 2009-06-09 11:43:11 UTC 
Created attachment 3988 [details]
z2.s.bz2

This looks like a gas bug to me.  At least from brief look at .cfi_* directives
in the assembly they look ok, but after assembling we see:
00000044 0000002c 00000048 FDE cie=00000000 pc=0000339f..00016d6a
  DW_CFA_advance_loc: 1 to 000033a0
  DW_CFA_def_cfa_offset: 16
  DW_CFA_advance_loc: 3 to 000033a3
  DW_CFA_offset: r6 (rbp) at cfa-16
  DW_CFA_def_cfa_register: r6 (rbp)
  DW_CFA_advance_loc2: 332 to 000034ef
  DW_CFA_offset: r3 (rbx) at cfa-56
  DW_CFA_offset: r12 (r12) at cfa-48
  DW_CFA_offset: r13 (r13) at cfa-40
  DW_CFA_offset: r14 (r14) at cfa-32
  DW_CFA_offset: r15 (r15) at cfa-24
  DW_CFA_nop
  DW_CFA_advance_loc: 58 to 00003529
  DW_CFA_??? (User defined call frame op: 0x38)
in the second FDE.  I guess gas just doesn't handle the huge location advance
well (from 34ef to 16d69).  Looking into it...
Comment 2 Jakub Jelinek 2009-06-09 11:48:09 UTC 
Small testcase:
.text
.cfi_startproc
.skip 16
.cfi_def_cfa 0, 16
.skip 75031
.cfi_def_cfa 0, 64
.cfi_endproc
Comment 3 Paul Pluzhnikov 2009-06-17 20:38:04 UTC 
I just hit this as well: corrupt .eh_frame in one of my object files with a huge
location advance.

Verified that this change to binutils/gas

2009-06-09  Jakub Jelinek  <jakub@redhat.com>

        PR gas/10255
        * dw2gencfi.c (output_cfi_insn): Initialize fragment before rs_cfa
        to DW_CFA_advance_loc4.

fixes it for me.
Comment 4 H.J. Lu 2009-06-18 13:57:00 UTC 
Fixed.
Comment 5 Peter Bergner 2009-06-18 19:47:42 UTC 
The new test case is failing for me on ppc64 like:

/home/bergner/binutils/build/binutils-pr10255/gas/testsuite/../../binutils/readelf
 -wf dump.o
Executing on host: sh -c
{/home/bergner/binutils/build/binutils-pr10255/gas/testsuite/../../binutils/readelf
 -wf dump.o >dump.out 2>gas.stderr}  /dev/null  (timeout = 300)
regexp_diff match failure
regexp "^00000014 000000(1c|20) 00000018 FDE cie=00000000 pc=.*$"
line   "00000014 00000018 00000018 FDE cie=00000000 pc=00000000..00012527"
regexp_diff match failure
regexp "^  DW_CFA_advance_loc4: 75031 to .*$"
line   "  DW_CFA_advance_loc2: 75028 to 00012524"
FAIL: CFI common 7

Configure options used were:
  --target=powerpc-linux
  --host=powerpc-linux
  --build=powerpc-linux
  --enable-targets=powerpc64-linux
Comment 6 Alan Modra 2009-06-18 23:20:17 UTC 
Testcase fixed.
Comment 7 M R Swami Reddy 2009-06-29 09:55:06 UTC 
Hello, Similar issue Igetting with cr16-elf tools? Is this patch taget 
specific? Thanks for any input.

Thanks
Swami