[llvm] r275027 - Let FuncAttrs infer the 'returned' argument attribute

Hal Finkel via llvm-commits llvm-commits at lists.llvm.org
Mon Jul 11 04:51:45 PDT 2016


Hi Renato,

Yes, it was this commit. I reverted it because it broke the AArch64 bot - and I sent you a note about it (http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20160704/371455.html). The AArch64 bot has now recovered (http://lab.llvm.org:8011/builders/clang-cmake-aarch64-full/builds/2566), and it does not look like the ARM bots have picked up the revert yet.

Thanks again,
Hal

----- Original Message -----
> From: "Renato Golin" <renato.golin at linaro.org>
> To: "Hal Finkel" <hfinkel at anl.gov>
> Cc: "LLVM Commits" <llvm-commits at lists.llvm.org>, "Diana Picus" <diana.picus at linaro.org>
> Sent: Monday, July 11, 2016 4:10:38 AM
> Subject: Re: [llvm] r275027 - Let FuncAttrs infer the 'returned' argument attribute
> 
> On 10 July 2016 at 23:02, Hal Finkel via llvm-commits
> <llvm-commits at lists.llvm.org> wrote:
> > Author: hfinkel
> > Date: Sun Jul 10 17:02:55 2016
> > New Revision: 275027
> >
> > URL: http://llvm.org/viewvc/llvm-project?rev=275027&view=rev
> > Log:
> > Let FuncAttrs infer the 'returned' argument attribute
> >
> > A function can have one argument with the 'returned' attribute,
> > indicating that
> > the associated argument is always the return value of the function.
> > Add
> > FuncAttrs inference logic.
> >
> > Differential Revision: http://reviews.llvm.org/D22202
> 
> Hi Hal,
> 
> I believe one of your two recent patches have broken the bots:
> 
> http://lab.llvm.org:8011/builders/clang-cmake-aarch64-full/builds/2564
> 
> http://lab.llvm.org:8011/builders/clang-cmake-armv7-a15-selfhost/builds/7401
> 
> /test/Analysis/LoopAccessAnalysis/backward-dep-different-types.ll:15:10:
> error: expected string not found in input
> ; CHECK: Report: unsafe dependent memory operations in loop
>          ^
> <stdin>:1:1: note: scanning from here
> Printing analysis 'Loop Access Analysis' for function 'f':
> ^
> <stdin>:27:5: note: possible intended match here
>  Store to invariant address was not found in loop.
>     ^
> 
> Any idea?
> 
> cheers,
> --renato
> 

-- 
Hal Finkel
Assistant Computational Scientist
Leadership Computing Facility
Argonne National Laboratory


More information about the llvm-commits mailing list