[SPIRV] Upd SPV_KHR_shader_clock extension #160147
Conversation
Add __spirv_ReadClockKHR support
|
Thank you for submitting a Pull Request (PR) to the LLVM Project! This PR will be automatically labeled and the relevant teams will be notified. If you wish to, you can add reviewers by using the "Reviewers" section on this page. If this is not working for you, it is probably because you do not have write permissions for the repository. In which case you can instead tag reviewers by name in a comment by using If you have received no comments on your PR for a week, you can request a review by "ping"ing the PR by adding a comment “Ping”. The common courtesy "ping" rate is once a week. Please remember that you are asking for valuable time from other developers. If you have further questions, they may be answered by the LLVM GitHub User Guide. You can also ask questions in a comment on this PR, on the LLVM Discord or on the forums. |
|
@llvm/pr-subscribers-backend-spir-v Author: Nikita Kornev (KornevNikita) ChangesAdd __spirv_ReadClockKHR support Full diff: https://github.com/llvm/llvm-project/pull/160147.diff 3 Files Affected:
diff --git a/llvm/lib/Target/SPIRV/SPIRVBuiltins.cpp b/llvm/lib/Target/SPIRV/SPIRVBuiltins.cpp
index 2abd9d36f7606..8bedd8c860ab2 100644
--- a/llvm/lib/Target/SPIRV/SPIRVBuiltins.cpp
+++ b/llvm/lib/Target/SPIRV/SPIRVBuiltins.cpp
@@ -1443,18 +1443,25 @@ static bool generateKernelClockInst(const SPIRV::IncomingCall *Call,
Register ResultReg = Call->ReturnRegister;
- // Deduce the `Scope` operand from the builtin function name.
- SPIRV::Scope::Scope ScopeArg =
- StringSwitch<SPIRV::Scope::Scope>(Builtin->Name)
- .EndsWith("device", SPIRV::Scope::Scope::Device)
- .EndsWith("work_group", SPIRV::Scope::Scope::Workgroup)
- .EndsWith("sub_group", SPIRV::Scope::Scope::Subgroup);
- Register ScopeReg = buildConstantIntReg32(ScopeArg, MIRBuilder, GR);
-
- MIRBuilder.buildInstr(SPIRV::OpReadClockKHR)
- .addDef(ResultReg)
- .addUse(GR->getSPIRVTypeID(Call->ReturnType))
- .addUse(ScopeReg);
+ if (Builtin->Name == "__spirv_ReadClockKHR") {
+ MIRBuilder.buildInstr(SPIRV::OpReadClockKHR)
+ .addDef(ResultReg)
+ .addUse(GR->getSPIRVTypeID(Call->ReturnType))
+ .addUse(Call->Arguments[0]);
+ } else {
+ // Deduce the `Scope` operand from the builtin function name.
+ SPIRV::Scope::Scope ScopeArg =
+ StringSwitch<SPIRV::Scope::Scope>(Builtin->Name)
+ .EndsWith("device", SPIRV::Scope::Scope::Device)
+ .EndsWith("work_group", SPIRV::Scope::Scope::Workgroup)
+ .EndsWith("sub_group", SPIRV::Scope::Scope::Subgroup);
+ Register ScopeReg = buildConstantIntReg32(ScopeArg, MIRBuilder, GR);
+
+ MIRBuilder.buildInstr(SPIRV::OpReadClockKHR)
+ .addDef(ResultReg)
+ .addUse(GR->getSPIRVTypeID(Call->ReturnType))
+ .addUse(ScopeReg);
+ }
return true;
}
diff --git a/llvm/lib/Target/SPIRV/SPIRVBuiltins.td b/llvm/lib/Target/SPIRV/SPIRVBuiltins.td
index d08560bb6565a..02f15bed396a1 100644
--- a/llvm/lib/Target/SPIRV/SPIRVBuiltins.td
+++ b/llvm/lib/Target/SPIRV/SPIRVBuiltins.td
@@ -1145,6 +1145,7 @@ defm : DemangledNativeBuiltin<"clock_read_sub_group", OpenCL_std, KernelClock, 0
defm : DemangledNativeBuiltin<"clock_read_hilo_device", OpenCL_std, KernelClock, 0, 0, OpReadClockKHR>;
defm : DemangledNativeBuiltin<"clock_read_hilo_work_group", OpenCL_std, KernelClock, 0, 0, OpReadClockKHR>;
defm : DemangledNativeBuiltin<"clock_read_hilo_sub_group", OpenCL_std, KernelClock, 0, 0, OpReadClockKHR>;
+defm : DemangledNativeBuiltin<"__spirv_ReadClockKHR", OpenCL_std, KernelClock, 1, 1, OpReadClockKHR>;
//===----------------------------------------------------------------------===//
// Class defining an atomic instruction on floating-point numbers.
diff --git a/llvm/test/CodeGen/SPIRV/extensions/SPV_KHR_shader_clock/shader_clock.ll b/llvm/test/CodeGen/SPIRV/extensions/SPV_KHR_shader_clock/shader_clock.ll
index bd07ba1316ece..eb4cf76c96da6 100644
--- a/llvm/test/CodeGen/SPIRV/extensions/SPV_KHR_shader_clock/shader_clock.ll
+++ b/llvm/test/CodeGen/SPIRV/extensions/SPV_KHR_shader_clock/shader_clock.ll
@@ -20,6 +20,9 @@
; CHECK: OpReadClockKHR [[v2uint]] [[uint_1]]
; CHECK: OpReadClockKHR [[v2uint]] [[uint_2]]
; CHECK: OpReadClockKHR [[v2uint]] [[uint_3]]
+; CHECK: OpReadClockKHR [[ulong]] [[uint_1]]
+; CHECK: OpReadClockKHR [[ulong]] [[uint_2]]
+; CHECK: OpReadClockKHR [[ulong]] [[uint_3]]
define dso_local spir_kernel void @test_clocks(ptr addrspace(1) nocapture noundef writeonly align 8 %out64, ptr addrspace(1) nocapture noundef writeonly align 8 %outv2) {
entry:
@@ -39,6 +42,9 @@ entry:
%call9 = tail call spir_func <2 x i32> @_Z25clock_read_hilo_sub_groupv()
%arrayidx10 = getelementptr inbounds i8, ptr addrspace(1) %outv2, i32 16
store <2 x i32> %call9, ptr addrspace(1) %arrayidx10, align 8
+ %call10 = call spir_func i64 @_Z27__spirv_ReadClockKHR_Rulongi(i32 1)
+ %call11 = call spir_func i64 @_Z27__spirv_ReadClockKHR_Rulongi(i32 2)
+ %call12 = call spir_func i64 @_Z27__spirv_ReadClockKHR_Rulongi(i32 3)
ret void
}
@@ -59,3 +65,6 @@ declare spir_func <2 x i32> @_Z26clock_read_hilo_work_groupv() local_unnamed_add
; Function Attrs: convergent nounwind
declare spir_func <2 x i32> @_Z25clock_read_hilo_sub_groupv() local_unnamed_addr
+
+; Function Attrs: nounwind
+declare spir_func i64 @_Z27__spirv_ReadClockKHR_Rulongi(i32)
|
KornevNikita
changed the title
| if (Builtin->Name == "__spirv_ReadClockKHR") { | ||
| MIRBuilder.buildInstr(SPIRV::OpReadClockKHR) | ||
| .addDef(ResultReg) | ||
| .addUse(GR->getSPIRVTypeID(Call->ReturnType)) | ||
| .addUse(Call->Arguments[0]); | ||
| } else { |
There was a problem hiding this comment.
I believe it's not needed. I'm looking at #96790 and it seems like changing
defm : DemangledNativeBuiltin<"__spirv_ReadClockKHR", OpenCL_std, KernelClock, 1, 1, OpReadClockKHR>;
to
defm : DemangledGroupBuiltinWrapper<"__spirv_ReadClockKHR", OpenCL_std, 1, 1, OpReadClockKHR>;
should be enough
There was a problem hiding this comment.
Tried to apply this, but it doesn't work like that - llc crashes. I believe DemangledGroupBuiltinWrapper is intended for different Group operations. Looking at how other built-ins are processed, I guess the way suggested here should be correct.
|
@MrSidims could you please re-run the pre-commit testing |
|
Ping |
|
@MrSidims thanks, I guess it's ready to merge? |
|
@KornevNikita Congratulations on having your first Pull Request (PR) merged into the LLVM Project! Your changes will be combined with recent changes from other authors, then tested by our build bots. If there is a problem with a build, you may receive a report in an email or a comment on this PR. Please check whether problems have been caused by your change specifically, as the builds can include changes from many authors. It is not uncommon for your change to be included in a build that fails due to someone else's changes, or infrastructure issues. How to do this, and the rest of the post-merge process, is covered in detail here. If your change does cause a problem, it may be reverted, or you can revert it yourself. This is a normal part of LLVM development. You can fix your changes and open a new PR to merge them again. If you don't get any reports, no action is required from you. Your changes are working as expected, well done! |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/162/builds/33835 Here is the relevant piece of the build log for the reference |
4 participants
Add __spirv_ReadClockKHR support