Please use --enable-fortify-source when testing

Carlos O'Donell carlos@redhat.com
Thu Jul 27 02:10:57 GMT 2023


On 7/26/23 08:53, Adhemerval Zanella Netto via Libc-alpha wrote:
> 
> 
> On 26/07/23 06:30, Andreas K. Huettel via Libc-alpha wrote:
>> Dear all, 
>>
>> the 2.38 release date and with it a hard cutoff for fixes is coming closer and
>> closer. Still, if you are doing machine testing, 
>>
>> ** Please use --enable-fortify-source **
>>
>> We need more data on possible compiler/binutils/architecture... combinations there.
> 
> With my patches to fix old gcc build [1][2] and i686 with fortify [3], I checked
> full builds for aarch64, x86_64, i686, s390x, powerpc64, and mips64 for with all 
> supported gcc (from 6 to 13, it vary depending of the architecture) with default
> no fortify and with fortify from 1 to 3 (if compiler supports it).  I have not
> found issues, although I only run check for x86 so far.
> 
> I might try some more architectures, although I don't expect more failures.
> 
> [1] https://patchwork.sourceware.org/project/glibc/list/?series=22751
> [2] https://patchwork.sourceware.org/project/glibc/list/?series=22754
> [3] https://patchwork.sourceware.org/project/glibc/list/?series=22775

Thank you Adhemerval for doing all of this testing! It is really awesome.

I'm testing on x86_64, i686, aarch64, s390x, and ppc64le right now with
gcc 13 (Fedora 39), 11 (RHEL9), and 8 (RHEL8) with a variety of binutils.

I am rechecking again with --enable-fortify-source.

-- 
Cheers,
Carlos.



More information about the Libc-alpha mailing list