[PATCH] D157680: [X86]Support options -mno-gather -mno-scatter

Wang, Xin via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Sun Aug 13 23:39:02 PDT 2023

XinWang10 marked an inline comment as done.
XinWang10 added inline comments.

Comment at: llvm/lib/Target/X86/X86.td:437
+    : SubtargetFeature<"prefer-no-gather", "PreferGather", "false",
+                       "Indicates if gather prefer to be disabled">;
+def FeaturePreferNoScatter
XinWang10 wrote:
> pengfei wrote:
> > XinWang10 wrote:
> > > skan wrote:
> > > > Does "Prefer no gather instructions" sound better?
> > > > 
> > > > I think these two should be put under "X86 Subtarget Tuning features"?
> > > I think the two options are to mitigate security issues. Could refer to link in summary.
> > It depends on if the micro code was applied. We should assume user care of this option should have applied the micro code. So it's more like a performance turning rather than mitigation. And you cannot disable all gather/scatter instructions with these options.
> Micro code applied? You mean we should keep eye on the byte code generated.
> And what's the essential difference between performance turning and mitigation? Do mitigation for no-gather means we could not emit gather whenever but performance turning could exist some gather considering the performance?
> Second, I think the intention for this option is to mitigate the security issue but not tune the performance.
Talked offline and agreed to move to performance tuning.

  rG LLVM Github Monorepo



More information about the llvm-commits mailing list