[PATCH] D52907: AMDGPU: Don't merge DS opcodes on SI to fix corruption in Hitman

Marek Olšák via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Mon Nov 12 19:37:47 PST 2018


mareko abandoned this revision.
mareko added a comment.

In https://reviews.llvm.org/D52907#1258465, @arsenm wrote:

> In https://reviews.llvm.org/D52907#1258433, @mareko wrote:
>
> > In https://reviews.llvm.org/D52907#1258426, @arsenm wrote:
> >
> > > Why exactly? Is it possible the Mesa lds allocation isn’t aligning properly?
> >
> >
> > Since the issue only occurs on SI, I don't think Mesa is doing anything bad. Unless there is some LDS hw difference on SI...
>
>
> I do remember one bug we have that may be related. We try to use the ds_read2_b32 with 4-byte signed trick on SI, without checking that we can use the offsets if the base address isn't known positive





Repository:
  rL LLVM

https://reviews.llvm.org/D52907





More information about the llvm-commits mailing list