Reviewer for a small clang Driver code change needed

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Reviewer for a small clang Driver code change needed

David Blaikie via cfe-dev
Hi!

I have made a small change to the clang Driver code [1] which fixes the testsuite
on Linux/sparc64

Currently, more than 400 tests fail on Linux/sparc64 [2] which drops to around 70
failures with the fix applied.

The reason for that is that without the fix, all 32-bit tests will fail on 64-bit
Linux/sparc64 since clang is unable to find the 32-bit shared libraries on the
64-bit system.

What this fix does is actually changing the Linux/sparc64 Driver code to use the
same Debian MultiArch tests and library paths as they are already used for x86
and PowerPC, hence we're actually not introducing anything new, just make SPARC
behave the same as x86 and PowerPC.

The Debian Multilib test that this change replace also make no sense since there
was never a Debian 8 release for sparc. The last official Debian sparc release
was version 7.10 (Debian Wheezy).

After that, Debian/sparc64 has been the only SPARC port in Debian - although this
change will not break the 32-bit Debian/sparc port as sparc and sparc64 use MultiArch
the same way on Debian as x86 and PowerPC do.

So, could someone review and commit that change for me so we can reduce the number
of testsuite failures on Linux/sparc64?

Thanks,
Adrian

> [1] https://reviews.llvm.org/D90524
> [2] lab.llvm.org:8014/#/workers/109

--
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer - [hidden email]
`. `'   Freie Universitaet Berlin - [hidden email]
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
_______________________________________________
cfe-dev mailing list
[hidden email]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: Reviewer for a small clang Driver code change needed

David Blaikie via cfe-dev
Generally it's best to ping the previous/existing thread - while I
realize it's rather tedious/thankless/feels a bit futile, it's
possibly more likely to get attention because people can then see how
long it's been waiting/how much it's been repeatedly brought
up/requested.

Richard - I know it's a bit unowned right now (Chandler did the last
big rewrite of it many years ago, I think) - I guess your stamp of
approval on the driver change discussed here might be valuable.

On Mon, Nov 23, 2020 at 2:53 PM John Paul Adrian Glaubitz via cfe-dev
<[hidden email]> wrote:

>
> Hi!
>
> I have made a small change to the clang Driver code [1] which fixes the testsuite
> on Linux/sparc64
>
> Currently, more than 400 tests fail on Linux/sparc64 [2] which drops to around 70
> failures with the fix applied.
>
> The reason for that is that without the fix, all 32-bit tests will fail on 64-bit
> Linux/sparc64 since clang is unable to find the 32-bit shared libraries on the
> 64-bit system.
>
> What this fix does is actually changing the Linux/sparc64 Driver code to use the
> same Debian MultiArch tests and library paths as they are already used for x86
> and PowerPC, hence we're actually not introducing anything new, just make SPARC
> behave the same as x86 and PowerPC.
>
> The Debian Multilib test that this change replace also make no sense since there
> was never a Debian 8 release for sparc. The last official Debian sparc release
> was version 7.10 (Debian Wheezy).
>
> After that, Debian/sparc64 has been the only SPARC port in Debian - although this
> change will not break the 32-bit Debian/sparc port as sparc and sparc64 use MultiArch
> the same way on Debian as x86 and PowerPC do.
>
> So, could someone review and commit that change for me so we can reduce the number
> of testsuite failures on Linux/sparc64?
>
> Thanks,
> Adrian
>
> > [1] https://reviews.llvm.org/D90524
> > [2] lab.llvm.org:8014/#/workers/109
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - [hidden email]
> `. `'   Freie Universitaet Berlin - [hidden email]
>   `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
> _______________________________________________
> cfe-dev mailing list
> [hidden email]
> https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
_______________________________________________
cfe-dev mailing list
[hidden email]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev