maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/NVPTX/nvvm-reflect.ll

87 lines
2.8 KiB
LLVM
Raw Normal View History

[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; We run nvvm-reflect (and then optimize) this module twice, once with metadata
; that enables FTZ, and again with metadata that disables it.
[NVPTX] Add NVVMReflect pass to allow compile-time selection of specific code paths. This allows us to write code like: if (__nvvm_reflect("FOO")) // Do something else // Do something else and compile into a library, then give "FOO" a value at kernel compile-time so the check becomes a no-op. llvm-svn: 178416
2013-03-30 22:29:25 +08:00
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; RUN: cat %s > %t.noftz
; RUN: echo '!0 = !{i32 4, !"nvvm-reflect-ftz", i32 0}' >> %t.noftz
; RUN: opt %t.noftz -S -nvvm-reflect -O2 \
; RUN: | FileCheck %s --check-prefix=USE_FTZ_0 --check-prefix=CHECK
; RUN: cat %s > %t.ftz
; RUN: echo '!0 = !{i32 4, !"nvvm-reflect-ftz", i32 1}' >> %t.ftz
; RUN: opt %t.ftz -S -nvvm-reflect -O2 \
; RUN: | FileCheck %s --check-prefix=USE_FTZ_1 --check-prefix=CHECK
@str = private unnamed_addr addrspace(4) constant [11 x i8] c"__CUDA_FTZ\00"
[NVPTX] Add NVVMReflect pass to allow compile-time selection of specific code paths. This allows us to write code like: if (__nvvm_reflect("FOO")) // Do something else // Do something else and compile into a library, then give "FOO" a value at kernel compile-time so the check becomes a no-op. llvm-svn: 178416
2013-03-30 22:29:25 +08:00
declare i32 @__nvvm_reflect(i8*)
declare i8* @llvm.nvvm.ptr.constant.to.gen.p0i8.p4i8(i8 addrspace(4)*)
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; CHECK-LABEL: @foo
[NVPTX] Add NVVMReflect pass to allow compile-time selection of specific code paths. This allows us to write code like: if (__nvvm_reflect("FOO")) // Do something else // Do something else and compile into a library, then give "FOO" a value at kernel compile-time so the check becomes a no-op. llvm-svn: 178416
2013-03-30 22:29:25 +08:00
define float @foo(float %a, float %b) {
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; CHECK-NOT: call i32 @__nvvm_reflect
%ptr = tail call i8* @llvm.nvvm.ptr.constant.to.gen.p0i8.p4i8(i8 addrspace(4)* getelementptr inbounds ([11 x i8], [11 x i8] addrspace(4)* @str, i32 0, i32 0))
[NVPTX] Add NVVMReflect pass to allow compile-time selection of specific code paths. This allows us to write code like: if (__nvvm_reflect("FOO")) // Do something else // Do something else and compile into a library, then give "FOO" a value at kernel compile-time so the check becomes a no-op. llvm-svn: 178416
2013-03-30 22:29:25 +08:00
%reflect = tail call i32 @__nvvm_reflect(i8* %ptr)
%cmp = icmp ugt i32 %reflect, 0
br i1 %cmp, label %use_mul, label %use_add
use_mul:
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; USE_FTZ_1: fmul float %a, %b
; USE_FTZ_0-NOT: fadd float %a, %b
[NVPTX] Add NVVMReflect pass to allow compile-time selection of specific code paths. This allows us to write code like: if (__nvvm_reflect("FOO")) // Do something else // Do something else and compile into a library, then give "FOO" a value at kernel compile-time so the check becomes a no-op. llvm-svn: 178416
2013-03-30 22:29:25 +08:00
%ret1 = fmul float %a, %b
br label %exit
use_add:
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; USE_FTZ_0: fadd float %a, %b
; USE_FTZ_1-NOT: fmul float %a, %b
[NVPTX] Add NVVMReflect pass to allow compile-time selection of specific code paths. This allows us to write code like: if (__nvvm_reflect("FOO")) // Do something else // Do something else and compile into a library, then give "FOO" a value at kernel compile-time so the check becomes a no-op. llvm-svn: 178416
2013-03-30 22:29:25 +08:00
%ret2 = fadd float %a, %b
br label %exit
exit:
%ret = phi float [%ret1, %use_mul], [%ret2, %use_add]
ret float %ret
}
[NVPTX] Add reflect intrinsic (better than matching by function name) Also clean up some of the logic in NVVMReflect.cpp while we're messing around in there. llvm-svn: 211948
2014-06-28 02:36:11 +08:00
declare i32 @llvm.nvvm.reflect.p0i8(i8*)
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; CHECK-LABEL: define i32 @intrinsic
[NVPTX] Add reflect intrinsic (better than matching by function name) Also clean up some of the logic in NVVMReflect.cpp while we're messing around in there. llvm-svn: 211948
2014-06-28 02:36:11 +08:00
define i32 @intrinsic() {
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; CHECK-NOT: call i32 @llvm.nvvm.reflect
; USE_FTZ_0: ret i32 0
; USE_FTZ_1: ret i32 1
%ptr = tail call i8* @llvm.nvvm.ptr.constant.to.gen.p0i8.p4i8(i8 addrspace(4)* getelementptr inbounds ([11 x i8], [11 x i8] addrspace(4)* @str, i32 0, i32 0))
[NVPTX] Add reflect intrinsic (better than matching by function name) Also clean up some of the logic in NVVMReflect.cpp while we're messing around in there. llvm-svn: 211948
2014-06-28 02:36:11 +08:00
%reflect = tail call i32 @llvm.nvvm.reflect.p0i8(i8* %ptr)
ret i32 %reflect
}
Add support for __nvvm_reflect changes in libdevice in CUDA-7.0 Summary: CUDA 7.0's libdevice uses slightly different IR to call __nvvm_reflect and that triggers an assertion in nvvm_reflect optimization pass. This change allows nvvm_reflect pass to deal with both old and new ways to pass an argument to __nvvm_reflect. Test Plan: ninja check-all Reviewers: eliben, echristo Subscribers: jholewinski, llvm-commits Differential Revision: http://reviews.llvm.org/D8399 llvm-svn: 232732
2015-03-20 01:05:35 +08:00
; CUDA-7.0 passes __nvvm_reflect argument slightly differently.
; Verify that it works, too
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
@"$str" = private addrspace(1) constant [11 x i8] c"__CUDA_FTZ\00"
Add support for __nvvm_reflect changes in libdevice in CUDA-7.0 Summary: CUDA 7.0's libdevice uses slightly different IR to call __nvvm_reflect and that triggers an assertion in nvvm_reflect optimization pass. This change allows nvvm_reflect pass to deal with both old and new ways to pass an argument to __nvvm_reflect. Test Plan: ninja check-all Reviewers: eliben, echristo Subscribers: jholewinski, llvm-commits Differential Revision: http://reviews.llvm.org/D8399 llvm-svn: 232732
2015-03-20 01:05:35 +08:00
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; CHECK-LABEL: @bar
Add support for __nvvm_reflect changes in libdevice in CUDA-7.0 Summary: CUDA 7.0's libdevice uses slightly different IR to call __nvvm_reflect and that triggers an assertion in nvvm_reflect optimization pass. This change allows nvvm_reflect pass to deal with both old and new ways to pass an argument to __nvvm_reflect. Test Plan: ninja check-all Reviewers: eliben, echristo Subscribers: jholewinski, llvm-commits Differential Revision: http://reviews.llvm.org/D8399 llvm-svn: 232732
2015-03-20 01:05:35 +08:00
define float @bar(float %a, float %b) {
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; CHECK-NOT: call i32 @__nvvm_reflect
%reflect = call i32 @__nvvm_reflect(i8* addrspacecast (i8 addrspace(1)* getelementptr inbounds ([11 x i8], [11 x i8] addrspace(1)* @"$str", i32 0, i32 0) to i8*))
Add support for __nvvm_reflect changes in libdevice in CUDA-7.0 Summary: CUDA 7.0's libdevice uses slightly different IR to call __nvvm_reflect and that triggers an assertion in nvvm_reflect optimization pass. This change allows nvvm_reflect pass to deal with both old and new ways to pass an argument to __nvvm_reflect. Test Plan: ninja check-all Reviewers: eliben, echristo Subscribers: jholewinski, llvm-commits Differential Revision: http://reviews.llvm.org/D8399 llvm-svn: 232732
2015-03-20 01:05:35 +08:00
%cmp = icmp ne i32 %reflect, 0
br i1 %cmp, label %use_mul, label %use_add
use_mul:
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; USE_FTZ_1: fmul float %a, %b
; USE_FTZ_0-NOT: fadd float %a, %b
Add support for __nvvm_reflect changes in libdevice in CUDA-7.0 Summary: CUDA 7.0's libdevice uses slightly different IR to call __nvvm_reflect and that triggers an assertion in nvvm_reflect optimization pass. This change allows nvvm_reflect pass to deal with both old and new ways to pass an argument to __nvvm_reflect. Test Plan: ninja check-all Reviewers: eliben, echristo Subscribers: jholewinski, llvm-commits Differential Revision: http://reviews.llvm.org/D8399 llvm-svn: 232732
2015-03-20 01:05:35 +08:00
%ret1 = fmul float %a, %b
br label %exit
use_add:
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
; USE_FTZ_0: fadd float %a, %b
; USE_FTZ_1-NOT: fmul float %a, %b
Add support for __nvvm_reflect changes in libdevice in CUDA-7.0 Summary: CUDA 7.0's libdevice uses slightly different IR to call __nvvm_reflect and that triggers an assertion in nvvm_reflect optimization pass. This change allows nvvm_reflect pass to deal with both old and new ways to pass an argument to __nvvm_reflect. Test Plan: ninja check-all Reviewers: eliben, echristo Subscribers: jholewinski, llvm-commits Differential Revision: http://reviews.llvm.org/D8399 llvm-svn: 232732
2015-03-20 01:05:35 +08:00
%ret2 = fadd float %a, %b
br label %exit
exit:
%ret = phi float [%ret1, %use_mul], [%ret2, %use_add]
ret float %ret
}
[NVPTX] Let there be One True Way to set NVVMReflect params. Summary: Previously there were three ways to inform the NVVMReflect pass whether you wanted to flush denormals to zero: * An LLVM command-line option * Parameters to the NVVMReflect constructor * Metadata on the module itself. This change removes the first two, leaving only the third. The motivation for this change, aside from simplifying things, is that we want LLVM to be aware of whether it's operating in FTZ mode, so other passes can use this information. Ideally we'd have a target-generic piece of metadata on the module. This change moves us in that direction. Reviewers: tra Subscribers: jholewinski, llvm-commits Differential Revision: https://reviews.llvm.org/D28700 llvm-svn: 292068
2017-01-16 00:54:35 +08:00
!llvm.module.flags = !{!0}
; A module flag is added to the end of this file by the RUN lines at the top.