[Bug ld/29031] The fix of CVE-2021-20197 may cause some .so files not linked
https://sourceware.org/bugzilla/show_bug.cgi?id=29031 Nick Clifton changed: What|Removed |Added CC||nickc at redhat dot com --- Comment #1 from Nick Clifton --- So - to be clear - the problem exists with the 2.34 sources with extra patches applied by yourselves, but not with the 2.37 sources (or later). Is that right ? When you talk about compiling "concurrently" does this mean that some of the shared objects might not have been fully linked by the time that libcwall.so is linked ? Ie, could the reason for the change in behaviour be down to the fact that the applied patches have slowed down the linker so that some of the processes in your concurrent build are taking longer than before ? I am not really sure what you expect us to do here. If you can reproduce the problem using the current 2.38 release sources, or the current mainline development sources, then we might be willing to investigate. But the 2.34 sources, with a set of extra patches applied ? Sorry but that is just outside of the scope of support that we try to provide. If you are able to reproduce the problem using the latest sources, please could you upload a testcase to this PR, so that we can take this issue further. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29038] Debug info for function in Windows PE binary on wrong instruction
https://sourceware.org/bugzilla/show_bug.cgi?id=29038 Nick Clifton changed: What|Removed |Added CC||nickc at redhat dot com --- Comment #1 from Nick Clifton --- Hi CP, Please could you upload the roca.exe binary so that I might investigate further ? Cheers Nick -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29038] Debug info for function in Windows PE binary on wrong instruction
https://sourceware.org/bugzilla/show_bug.cgi?id=29038 --- Comment #2 from CP --- Created attachment 14054 --> https://sourceware.org/bugzilla/attachment.cgi?id=14054&action=edit roca.exe Attached as requested -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29038] Debug info for function in Windows PE binary on wrong instruction
https://sourceware.org/bugzilla/show_bug.cgi?id=29038 --- Comment #3 from cvs-commit at gcc dot gnu.org --- The master branch has been updated by Nick Clifton : https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=7d41169b6d4dbc3950516d0cdf0d8ca462b52bbb commit 7d41169b6d4dbc3950516d0cdf0d8ca462b52bbb Author: Nick Clifton Date: Fri Apr 8 16:04:22 2022 +0100 Debug info for function in Windows PE binary on wrong instruction PR 29038 * coffgen.c (coff_find_nearest_line_with_names): Fix typo retrieving saved bias. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29038] Debug info for function in Windows PE binary on wrong instruction
https://sourceware.org/bugzilla/show_bug.cgi?id=29038 Nick Clifton changed: What|Removed |Added Resolution|--- |FIXED Status|UNCONFIRMED |RESOLVED --- Comment #4 from Nick Clifton --- Oh this was embarrassing. There was a typo in the code. I had this: if (sec_data != NULL && sec_data->saved_bias) bias = sec_data->saved_bias; whereas it should have been: if (sec_data != NULL && sec_data->saved_bias) bias = sec_data->bias; (saved_bias is a boolean that indicates whether the bias field has been set...) I have applied a patch to fix the bug. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29038] Debug info for function in Windows PE binary on wrong instruction
https://sourceware.org/bugzilla/show_bug.cgi?id=29038 --- Comment #5 from CP --- Wonderful! Confirmed fixed, thanks very much. -- You are receiving this mail because: You are on the CC list for the bug.
