Skip to content

Update cl_ext_float_atomics patch #279

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Sep 15, 2021
Merged

Update cl_ext_float_atomics patch #279

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
54 changes: 33 additions & 21 deletions patches/spirv/0001-Add-support-for-cl_ext_float_atomics-in-SPIRVWriter.patch
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,34 +1,33 @@
From 794c7d32e521060190bbfca7c52f0bb4fa49f2f2 Mon Sep 17 00:00:00 2001
From 89e3be862e9a389ba858faa1270f0f8856b37bcf Mon Sep 17 00:00:00 2001
From: haonanya <haonan.yang@intel.com>
Date: Mon, 19 Jul 2021 10:14:20 +0800
Subject: [PATCH] Add support for cl_ext_float_atomics in SPIRVWriter

Signed-off-by: haonanya <haonan.yang@intel.com>
Signed-off-by: Haonan Yang <haonan.yang@intel.com>
---
lib/SPIRV/OCL20ToSPIRV.cpp | 26 +++++-
lib/SPIRV/OCLUtil.cpp | 19 +++--
lib/SPIRV/OCLUtil.cpp | 15 ++--
lib/SPIRV/SPIRVToOCL.h | 3 +
lib/SPIRV/SPIRVToOCL12.cpp | 21 +++++
lib/SPIRV/SPIRVToOCL20.cpp | 28 ++++++-
lib/SPIRV/libSPIRV/SPIRVNameMapEnum.h | 1 -
lib/SPIRV/libSPIRV/SPIRVOpCode.h | 8 +-
test/AtomicBuiltinsFloat.ll | 79 ++++++++++++++++++
test/AtomicBuiltinsFloat.ll | 94 +++++++++++++++++++++
test/AtomicFAddEXTForOCL.ll | 88 ++++++++++++++++++++
test/AtomicFAddExt.ll | 111 ++++++++-----------------
test/AtomicFMaxEXT.ll | 113 +++++++-------------------
test/AtomicFMaxEXTForOCL.ll | 88 ++++++++++++++++++++
test/AtomicFMinEXT.ll | 113 +++++++-------------------
test/AtomicFMinEXTForOCL.ll | 85 +++++++++++++++++++
test/InvalidAtomicBuiltins.cl | 16 ----
15 files changed, 526 insertions(+), 273 deletions(-)
15 files changed, 539 insertions(+), 271 deletions(-)
create mode 100644 test/AtomicBuiltinsFloat.ll
create mode 100644 test/AtomicFAddEXTForOCL.ll
create mode 100644 test/AtomicFMaxEXTForOCL.ll
create mode 100644 test/AtomicFMinEXTForOCL.ll

diff --git a/lib/SPIRV/OCL20ToSPIRV.cpp b/lib/SPIRV/OCL20ToSPIRV.cpp
index e30aa5be..98d4289e 100644
index e30aa5be..79323de2 100644
--- a/lib/SPIRV/OCL20ToSPIRV.cpp
+++ b/lib/SPIRV/OCL20ToSPIRV.cpp
@@ -407,7 +407,6 @@ void OCL20ToSPIRV::visitCallInst(CallInst &CI) {
Expand All @@ -53,7 +52,7 @@ index e30aa5be..98d4289e 100644
Args.end() - Offset);
}
- return getSPIRVFuncName(OCLSPIRVBuiltinMap::map(Info.UniqName));
+ llvm::Type* AtomicBuiltinsReturnType =
+ llvm::Type* AtomicBuiltinsReturnType =
+ CI->getCalledFunction()->getReturnType();
+ auto IsFPType = [](llvm::Type *ReturnType) {
+ return ReturnType->isHalfTy() || ReturnType->isFloatTy() ||
Expand All @@ -79,30 +78,28 @@ index e30aa5be..98d4289e 100644
&Attrs);
}
diff --git a/lib/SPIRV/OCLUtil.cpp b/lib/SPIRV/OCLUtil.cpp
index 992f173f..539c196c 100644
index 992f173f..8ae882c6 100644
--- a/lib/SPIRV/OCLUtil.cpp
+++ b/lib/SPIRV/OCLUtil.cpp
@@ -120,29 +120,32 @@ size_t getSPIRVAtomicBuiltinNumMemoryOrderArgs(Op OC) {
return 1;
}

+// atomic_fetch_[add, sub, min, max] and atomic_fetch_[add, sub, min,
+// max]_explicit functions are defined on OpenCL headers, they are not
+// translated to function call
+// atomic_fetch_[add, min, max] and atomic_fetch_[add, min, max]_explicit
+// functions declared in clang headers should be translated to corresponding
+// FP-typed Atomic Instructions
bool isComputeAtomicOCLBuiltin(StringRef DemangledName) {
if (!DemangledName.startswith(kOCLBuiltinName::AtomicPrefix) &&
!DemangledName.startswith(kOCLBuiltinName::AtomPrefix))
return false;

return llvm::StringSwitch<bool>(DemangledName)
- .EndsWith("add", true)
- .EndsWith("sub", true)
.EndsWith("sub", true)
+ .EndsWith("atomic_add", true)
+ .EndsWith("atomic_sub", true)
+ .EndsWith("atomic_min", true)
+ .EndsWith("atomic_max", true)
+ .EndsWith("atom_add", true)
+ .EndsWith("atom_sub", true)
+ .EndsWith("atom_min", true)
+ .EndsWith("atom_max", true)
.EndsWith("inc", true)
Expand All @@ -114,7 +111,7 @@ index 992f173f..539c196c 100644
.EndsWith("or", true)
.EndsWith("xor", true)
- .EndsWith("add_explicit", true)
- .EndsWith("sub_explicit", true)
.EndsWith("sub_explicit", true)
.EndsWith("or_explicit", true)
.EndsWith("xor_explicit", true)
.EndsWith("and_explicit", true)
Expand Down Expand Up @@ -184,7 +181,7 @@ index 1a62c6b8..dc0ba9cc 100644
}

diff --git a/lib/SPIRV/SPIRVToOCL20.cpp b/lib/SPIRV/SPIRVToOCL20.cpp
index 8c437858..0033b853 100644
index 8c437858..07612ce6 100644
--- a/lib/SPIRV/SPIRVToOCL20.cpp
+++ b/lib/SPIRV/SPIRVToOCL20.cpp
@@ -82,6 +82,9 @@ public:
Expand Down Expand Up @@ -232,8 +229,8 @@ index 8c437858..0033b853 100644
}
auto Ptr = findFirstPtr(Args);
- auto Name = OCLSPIRVBuiltinMap::rmap(OC);
+ std::string Name;
+ // Map fp atomic instructions to regular OpenCL built-ins.
+ std::string Name;
+ // Map fp atomic instructions to regular OpenCL built-ins.
+ if (isFPAtomicOpCode(OC))
+ Name = mapFPAtomicName(OC);
+ else
Expand Down Expand Up @@ -278,11 +275,13 @@ index feec70f6..8e595e83 100644
return ((unsigned)OpCode >= OpIAdd && (unsigned)OpCode <= OpFMod) ||
diff --git a/test/AtomicBuiltinsFloat.ll b/test/AtomicBuiltinsFloat.ll
new file mode 100644
index 00000000..18ee1c86
index 00000000..b688cb2a
--- /dev/null
+++ b/test/AtomicBuiltinsFloat.ll
@@ -0,0 +1,79 @@
@@ -0,0 +1,94 @@
+; Check that translator generate atomic instructions for atomic builtins
+; FP-typed atomic_fetch_sub and atomic_fetch_sub_explicit should be translated
+; to FunctionCall
+; RUN: llvm-as %s -o %t.bc
+; RUN: llvm-spirv %t.bc -spirv-text -o - | FileCheck %s
+; RUN: llvm-spirv %t.bc -o %t.spv
Expand All @@ -293,6 +292,7 @@ index 00000000..18ee1c86
+; CHECK-COUNT-3: AtomicStore
+; CHECK-COUNT-3: AtomicLoad
+; CHECK-COUNT-3: AtomicExchange
+; CHECK-COUNT-3: FunctionCall
+
+target datalayout = "e-p:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024"
+target triple = "spir-unknown-unknown"
Expand All @@ -311,6 +311,9 @@ index 00000000..18ee1c86
+ %call3 = tail call spir_func float @_Z15atomic_exchangePU3AS4VU7_Atomicff(float addrspace(4)* %0, float 1.000000e+00) #2
+ %call4 = tail call spir_func float @_Z24atomic_exchange_explicitPU3AS4VU7_Atomicff12memory_order(float addrspace(4)* %0, float 1.000000e+00, i32 0) #2
+ %call5 = tail call spir_func float @_Z24atomic_exchange_explicitPU3AS4VU7_Atomicff12memory_order12memory_scope(float addrspace(4)* %0, float 1.000000e+00, i32 0, i32 1) #2
+ %call6 = tail call spir_func float @_Z16atomic_fetch_subPU3AS3VU7_Atomicff(float addrspace(3)* %ff, float 1.000000e+00) #2
+ %call7 = tail call spir_func float @_Z25atomic_fetch_sub_explicitPU3AS3VU7_Atomicff12memory_order(float addrspace(3)* %ff, float 1.000000e+00, i32 0) #2
+ %call8 = tail call spir_func float @_Z25atomic_fetch_sub_explicitPU3AS3VU7_Atomicff12memory_order12memory_scope(float addrspace(3)* %ff, float 1.000000e+00, i32 0, i32 1) #2
+ ret void
+}
+
Expand Down Expand Up @@ -344,6 +347,15 @@ index 00000000..18ee1c86
+; Function Attrs: convergent
+declare spir_func float @_Z24atomic_exchange_explicitPU3AS4VU7_Atomicff12memory_order12memory_scope(float addrspace(4)*, float, i32, i32) local_unnamed_addr #1
+
+; Function Attrs: convergent
+declare spir_func float @_Z16atomic_fetch_subPU3AS3VU7_Atomicff(float addrspace(3)*, float) local_unnamed_addr #1
+
+; Function Attrs: convergent
+declare spir_func float @_Z25atomic_fetch_sub_explicitPU3AS3VU7_Atomicff12memory_order(float addrspace(3)*, float, i32) local_unnamed_addr #1
+
+; Function Attrs: convergent
+declare spir_func float @_Z25atomic_fetch_sub_explicitPU3AS3VU7_Atomicff12memory_order12memory_scope(float addrspace(3)*, float, i32, i32) local_unnamed_addr #1
+
+attributes #0 = { convergent nounwind "correctly-rounded-divide-sqrt-fp-math"="false" "denorms-are-zero"="false" "disable-tail-calls"="false" "frame-pointer"="none" "less-precise-fpmad"="false" "min-legal-vector-width"="0" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "uniform-work-group-size"="false" "unsafe-fp-math"="false" "use-soft-float"="false" }
+attributes #1 = { convergent "correctly-rounded-divide-sqrt-fp-math"="false" "denorms-are-zero"="false" "disable-tail-calls"="false" "frame-pointer"="none" "less-precise-fpmad"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
+attributes #2 = { convergent nounwind }
Expand All @@ -355,7 +367,7 @@ index 00000000..18ee1c86
+
+!0 = !{i32 1, !"wchar_size", i32 4}
+!1 = !{i32 2, i32 0}
+!2 = !{!"clang version 10.0.1 (8560093eba963fba2edd47ca85404cdaff22f174)"}
+!2 = !{!"clang version 10.0.1 (0d3ce3267dd78b77ab8f302347cc547afaf43ede)"}
+!3 = !{i32 3}
+!4 = !{!"none"}
+!5 = !{!"atomic_float*"}
Expand Down