Re: [Rpm-maint] [rpm-software-management/rpm] debugedit fails to update section headers (#423)
Closed #423. -- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/issues/423#event-2713882777___ Rpm-maint mailing list Rpm-maint@lists.rpm.org http://lists.rpm.org/mailman/listinfo/rpm-maint
Re: [Rpm-maint] [rpm-software-management/rpm] debugedit fails to update section headers (#423)
Okay so this was confirmed fixed on elfutils side, we can close it here. -- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/issues/423#issuecomment-542200334___ Rpm-maint mailing list Rpm-maint@lists.rpm.org http://lists.rpm.org/mailman/listinfo/rpm-maint
Re: [Rpm-maint] [rpm-software-management/rpm] debugedit fails to update section headers (#423)
> But I now believe this is actually an elfutils libelf issue: > https://sourceware.org/ml/elfutils-devel/2019-q2/msg00077.html Could you try > that out and see if it resolves the issue for you? (The actual patch is just > 3 lines, but there is also a larger testcase added to make sure the issue was > really resolved.) Thanks, Mark The patch helps here: applied on top of `elfutils-0.176` and tested against `debugedit-4.14.2`. -- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/issues/423#issuecomment-491705457___ Rpm-maint mailing list Rpm-maint@lists.rpm.org http://lists.rpm.org/mailman/listinfo/rpm-maint
Re: [Rpm-maint] [rpm-software-management/rpm] debugedit fails to update section headers (#423)
On Mon, Apr 29, 2019 at 02:24:33PM +, Michael Schroeder wrote: > Mark, any update on this? We (SUSE) get corrupt debuginfo files from > time to time because of this bug. Apologies for the slow response. I looked at it before, but didn't understand what was going wrong. But I now believe this is actually an elfutils libelf issue: https://sourceware.org/ml/elfutils-devel/2019-q2/msg00077.html Could you try that out and see if it resolves the issue for you? (The actual patch is just 3 lines, but there is also a larger testcase added to make sure the issue was really resolved.) Thanks, Mark ___ Rpm-maint mailing list Rpm-maint@lists.rpm.org http://lists.rpm.org/mailman/listinfo/rpm-maint
Re: [Rpm-maint] [rpm-software-management/rpm] debugedit fails to update section headers (#423)
Mark, any update on this? We (SUSE) get corrupt debuginfo files from time to time because of this bug. -- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/issues/423#issuecomment-487600464___ Rpm-maint mailing list Rpm-maint@lists.rpm.org http://lists.rpm.org/mailman/listinfo/rpm-maint
Re: [Rpm-maint] [rpm-software-management/rpm] debugedit fails to update section headers (#423)
The following workaround allows to maintain original section order:
```diff
--- a/tools/debugedit.c
+++ b/tools/debugedit.c
@@ -2342,21 +2342,21 @@ fdopen_dso (int fd, const char *name)
}
if (elf_getphdrnum (elf, ) != 0)
{
error (0, 0, "Couldn't get number of phdrs: %s", elf_errmsg (-1));
goto error_out;
}
/* If there are phdrs we want to maintain the layout of the
allocated sections in the file. */
- if (phnum != 0)
+ //if (phnum != 0)
elf_flagelf (elf, ELF_C_SET, ELF_F_LAYOUT);
memset (dso, 0, sizeof(DSO));
dso->elf = elf;
dso->phnum = phnum;
dso->ehdr = ehdr;
dso->scn = (Elf_Scn **) >shdr[ehdr.e_shnum + 20];
for (i = 0; i < ehdr.e_shnum; ++i)
{
```
--
You are receiving this because you commented.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/423#issuecomment-424147426___
Rpm-maint mailing list
Rpm-maint@lists.rpm.org
http://lists.rpm.org/mailman/listinfo/rpm-maint
Re: [Rpm-maint] [rpm-software-management/rpm] debugedit fails to update section headers (#423)
I think in my case it happens because `debugedit` was not able to delete any sections (no file size changes) but it reordered sections in the file and did not update section offsets. Attaching original, mangled file and a script:[__t.tar.gz](https://github.com/rpm-software-management/rpm/files/2412961/__t.tar.gz) -- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/issues/423#issuecomment-424133239___ Rpm-maint mailing list Rpm-maint@lists.rpm.org http://lists.rpm.org/mailman/listinfo/rpm-maint
Re: [Rpm-maint] [rpm-software-management/rpm] debugedit fails to update section headers (#423)
debugedit also can break gcc's crtstartup: https://bugs.gentoo.org/666954 . I think it's the same issue. Will try to provide more detail today. -- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/issues/423#issuecomment-424051949___ Rpm-maint mailing list Rpm-maint@lists.rpm.org http://lists.rpm.org/mailman/listinfo/rpm-maint
Re: [Rpm-maint] [rpm-software-management/rpm] debugedit fails to update section headers (#423)
[xt_DHCPMAC.ko.gz](https://github.com/rpm-software-management/rpm/files/1859743/xt_DHCPMAC.ko.gz) -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/issues/423#issuecomment-377198192___ Rpm-maint mailing list Rpm-maint@lists.rpm.org http://lists.rpm.org/mailman/listinfo/rpm-maint
[Rpm-maint] [rpm-software-management/rpm] debugedit fails to update section headers (#423)
debugedit -b /home/abuild/rpmbuild/BUILD/xtables-addons-3.0 -d
/usr/src/debug/xtables-addons-3.0-0.x86_64 test.ko
results in
There are 41 section headers, starting at offset 0x870a8:
Section Headers:
[Nr] Name Type Address Offset
Size EntSize Flags Link Info Align
[ 0] etNULL
0 0 0
[ 1] abNOTE 0040
0024 A 0 0 4
[ 2] ela.text PROGBITS 0070
0340 AX 0 0 16
...
the .shstrtab (40) starts with
Hex dump of section 'ab':
0x 6574 2e73796d 74616200 2e737472 et...symtab..str
0x0010 74616200 2e736873 74727461 62002e6e tab..shstrtab..n
0x0020 6f74652e 676e752e 6275696c 642d6964 ote.gnu.build-id
and suspiciously 37 (.rela.debug_frame) contains (look at the end!)
Hex dump of section 'ck':
0x 455f4944 5f766572 6d616769 63313400 E_ID_vermagic14.
0x0010 736b625f 636f7079 5f626974 73007874 skb_copy_bits.xt
0x0020 5f726567 69737465 725f6d61 74636800 _register_match.
0x0030 5f5f7468 69735f6d 6f64756c 6500736b __this_module.sk
0x0040 625f6d61 6b655f77 72697461 626c6500 b_make_writable.
0x0050 636c6561 6e75705f 6d6f6475 6c65005f cleanup_module._
0x0060 5f66656e 7472795f 5f00696e 69745f6d _fentry__.init_m
0x0070 6f64756c 65007874 5f756e72 65676973 odule.xt_unregis
0x0080 7465725f 6d617463 68005f5f 73746163 ter_match.__stac
0x0090 6b5f6368 6b5f6661 696c0078 745f7265 k_chk_fail.xt_re
0x00a0 67697374 65725f74 61726765 74007874 gister_target.xt
0x00b0 5f756e72 65676973 7465725f 74617267 _unregister_targ
so the real .shstrtab should start at 00086f1c instead of 00086f18.
So there's a off-by-4 error somewhere...
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/423___
Rpm-maint mailing list
Rpm-maint@lists.rpm.org
http://lists.rpm.org/mailman/listinfo/rpm-maint
