[llvm] [AMDGPU] Replace dynamic VGPR feature with attribute (PR #133444)
Matt Arsenault via llvm-commits
llvm-commits at lists.llvm.org
Fri Apr 25 04:33:30 PDT 2025
================
@@ -221,11 +226,10 @@ static void testAbsoluteLimits(StringRef CPUName, StringRef FS,
}
TEST(AMDGPU, TestOccupancyAbsoluteLimits) {
- testAbsoluteLimits("gfx1200", "+wavefrontsize32", 1, 16, 256);
- testAbsoluteLimits("gfx1200", "+wavefrontsize32,+dynamic-vgpr", 1, 16, 128);
- testAbsoluteLimits(
- "gfx1200", "+wavefrontsize32,+dynamic-vgpr,+dynamic-vgpr-block-size-32",
- 1, 16, 256);
+ testAbsoluteLimits("gfx1200", "+wavefrontsize32", false, 1, 16, 256);
+ testAbsoluteLimits("gfx1200", "+wavefrontsize32", true, 1, 16, 128);
+ testAbsoluteLimits("gfx1200", "+wavefrontsize32,+dynamic-vgpr-block-size-32",
+ true, 1, 16, 256);
----------------
arsenm wrote:
Is dynamic-vgpr-block-size-32 a fixed hardware property, or just a parameter to this feature? Should this consolidate into making this an attribute with a value, with 0 meaning disable?
https://github.com/llvm/llvm-project/pull/133444
More information about the llvm-commits
mailing list