[PATCH] D33455: [AMDGPU] Combine and (srl) into shl (bfe)

Matt Arsenault via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Tue May 23 11:51:36 PDT 2017


arsenm added inline comments.


================
Comment at: lib/Target/AMDGPU/AMDGPUISelLowering.cpp:3482
+    if ((OffsetVal + WidthVal) >= 32 &&
+        !(Subtarget->getGeneration() >= AMDGPUSubtarget::VOLCANIC_ISLANDS &&
+          OffsetVal == 16 && WidthVal == 16)) {
----------------
rampitec wrote:
> arsenm wrote:
> > There is an SDWA subtarget feature
> AMDGPUSubtarget does not know anything about SDWA, it is in SISubtarget.
The SDWA predicate could be moved into AMDGPUSUbtarget if needed here, there's nothing really requiring it be part of SISubtarget


Repository:
  rL LLVM

https://reviews.llvm.org/D33455





More information about the llvm-commits mailing list