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/AMDGPU/ds-sub-offset.ll

164 lines
6.4 KiB
LLVM
Raw Normal View History

AMDGPU: Use gfx9 carry-less add/sub instructions llvm-svn: 319491
2017-12-01 06:51:26 +08:00
; RUN: llc -march=amdgcn -mcpu=bonaire -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,CI %s
; RUN: llc -march=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GFX9 %s
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
AMDGPU: Remove some old intrinsic uses from tests llvm-svn: 260493
2016-02-11 14:02:01 +08:00
declare i32 @llvm.amdgcn.workitem.id.x() #0
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
@lds.obj = addrspace(3) global [256 x i32] undef, align 4
; GCN-LABEL: {{^}}write_ds_sub0_offset0_global:
; GCN: v_lshlrev_b32_e32 [[SHL:v[0-9]+]], 2, v0
AMDGPU: Write LDS objects out as global symbols in code generation Summary: The symbols use the processor-specific SHN_AMDGPU_LDS section index introduced with a previous change. The linker is then expected to resolve relocations, which are also emitted. Initially disabled for HSA and PAL environments until they have caught up in terms of linker and runtime loader. Some notes: - The llvm.amdgcn.groupstaticsize intrinsics can no longer be lowered to a constant at compile times, which means some tests can no longer be applied. The current "solution" is a terrible hack, but the intrinsic isn't used by Mesa, so we can keep it for now. - We no longer know the full LDS size per kernel at compile time, which means that we can no longer generate a relevant error message at compile time. It would be possible to add a check for the size of individual variables, but ultimately the linker will have to perform the final check. Change-Id: If66dbf33fccfbf3609aefefa2558ac0850d42275 Reviewers: arsenm, rampitec, t-tye, b-sumner, jsjodin Subscribers: qcolombet, kzhuravl, jvesely, wdng, yaxunl, dstuttard, tpr, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D61494 llvm-svn: 364297
2019-06-25 19:52:30 +08:00
; GCN: v_sub_{{[iu]}}32_e32 [[BASEPTR:v[0-9]+]], {{(vcc, )?}}lds.obj@abs32@lo, [[SHL]]
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
; GCN: v_mov_b32_e32 [[VAL:v[0-9]+]], 0x7b
; GCN: ds_write_b32 [[BASEPTR]], [[VAL]] offset:12
AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @write_ds_sub0_offset0_global() #0 {
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
entry:
AMDGPU: Remove some old intrinsic uses from tests llvm-svn: 260493
2016-02-11 14:02:01 +08:00
%x.i = call i32 @llvm.amdgcn.workitem.id.x() #1
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
%sub1 = sub i32 0, %x.i
%tmp0 = getelementptr [256 x i32], [256 x i32] addrspace(3)* @lds.obj, i32 0, i32 %sub1
%arrayidx = getelementptr inbounds i32, i32 addrspace(3)* %tmp0, i32 3
store i32 123, i32 addrspace(3)* %arrayidx
ret void
}
[AMDGPU] Add more test cases of D59608. Summary: - Add more test cases. Reviewers: arsenm Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D60071 llvm-svn: 357442
2019-04-02 08:36:37 +08:00
; GFX9-LABEL: {{^}}write_ds_sub0_offset0_global_clamp_bit:
; GFX9: v_sub_u32
; GFX9: s_endpgm
define amdgpu_kernel void @write_ds_sub0_offset0_global_clamp_bit(float %dummy.val) #0 {
entry:
%x.i = call i32 @llvm.amdgcn.workitem.id.x() #1
%sub1 = sub i32 0, %x.i
%tmp0 = getelementptr [256 x i32], [256 x i32] addrspace(3)* @lds.obj, i32 0, i32 %sub1
%arrayidx = getelementptr inbounds i32, i32 addrspace(3)* %tmp0, i32 3
store i32 123, i32 addrspace(3)* %arrayidx
%fmas = call float @llvm.amdgcn.div.fmas.f32(float %dummy.val, float %dummy.val, float %dummy.val, i1 false)
store volatile float %fmas, float addrspace(1)* null
ret void
}
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
; GCN-LABEL: {{^}}add_x_shl_neg_to_sub_max_offset:
; GCN-DAG: v_lshlrev_b32_e32 [[SCALED:v[0-9]+]], 2, v0
AMDGPU: Use gfx9 carry-less add/sub instructions llvm-svn: 319491
2017-12-01 06:51:26 +08:00
; CI-DAG: v_sub_i32_e32 [[NEG:v[0-9]+]], vcc, 0, [[SCALED]]
; GFX9-DAG: v_sub_u32_e32 [[NEG:v[0-9]+]], 0, [[SCALED]]
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
; GCN-DAG: v_mov_b32_e32 [[K:v[0-9]+]], 13
; GCN: ds_write_b8 [[NEG]], [[K]] offset:65535
AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @add_x_shl_neg_to_sub_max_offset() #1 {
AMDGPU: Remove some old intrinsic uses from tests llvm-svn: 260493
2016-02-11 14:02:01 +08:00
%x.i = call i32 @llvm.amdgcn.workitem.id.x() #0
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
%neg = sub i32 0, %x.i
%shl = shl i32 %neg, 2
%add = add i32 65535, %shl
%ptr = inttoptr i32 %add to i8 addrspace(3)*
store i8 13, i8 addrspace(3)* %ptr
ret void
}
; GCN-LABEL: {{^}}add_x_shl_neg_to_sub_max_offset_p1:
; GCN-DAG: v_lshlrev_b32_e32 [[SCALED:v[0-9]+]], 2, v0
AMDGPU: Use gfx9 carry-less add/sub instructions llvm-svn: 319491
2017-12-01 06:51:26 +08:00
; CI-DAG: v_sub_i32_e32 [[NEG:v[0-9]+]], vcc, 0x10000, [[SCALED]]
; GFX9-DAG: v_sub_u32_e32 [[NEG:v[0-9]+]], 0x10000, [[SCALED]]
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
; GCN-DAG: v_mov_b32_e32 [[K:v[0-9]+]], 13
; GCN: ds_write_b8 [[NEG]], [[K]]{{$}}
AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @add_x_shl_neg_to_sub_max_offset_p1() #1 {
AMDGPU: Remove some old intrinsic uses from tests llvm-svn: 260493
2016-02-11 14:02:01 +08:00
%x.i = call i32 @llvm.amdgcn.workitem.id.x() #0
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
%neg = sub i32 0, %x.i
%shl = shl i32 %neg, 2
%add = add i32 65536, %shl
%ptr = inttoptr i32 %add to i8 addrspace(3)*
store i8 13, i8 addrspace(3)* %ptr
ret void
}
; GCN-LABEL: {{^}}add_x_shl_neg_to_sub_multi_use:
; GCN-DAG: v_lshlrev_b32_e32 [[SCALED:v[0-9]+]], 2, v0
AMDGPU: Use gfx9 carry-less add/sub instructions llvm-svn: 319491
2017-12-01 06:51:26 +08:00
; CI-DAG: v_sub_i32_e32 [[NEG:v[0-9]+]], vcc, 0, [[SCALED]]
; GFX9-DAG: v_sub_u32_e32 [[NEG:v[0-9]+]], 0, [[SCALED]]
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
; GCN-DAG: v_mov_b32_e32 [[K:v[0-9]+]], 13
; GCN-NOT: v_sub
; GCN: ds_write_b32 [[NEG]], [[K]] offset:123{{$}}
; GCN-NOT: v_sub
; GCN: ds_write_b32 [[NEG]], [[K]] offset:456{{$}}
; GCN: s_endpgm
AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @add_x_shl_neg_to_sub_multi_use() #1 {
AMDGPU: Remove some old intrinsic uses from tests llvm-svn: 260493
2016-02-11 14:02:01 +08:00
%x.i = call i32 @llvm.amdgcn.workitem.id.x() #0
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
%neg = sub i32 0, %x.i
%shl = shl i32 %neg, 2
%add0 = add i32 123, %shl
%add1 = add i32 456, %shl
%ptr0 = inttoptr i32 %add0 to i32 addrspace(3)*
store volatile i32 13, i32 addrspace(3)* %ptr0
%ptr1 = inttoptr i32 %add1 to i32 addrspace(3)*
store volatile i32 13, i32 addrspace(3)* %ptr1
ret void
}
; GCN-LABEL: {{^}}add_x_shl_neg_to_sub_multi_use_same_offset:
; GCN-DAG: v_lshlrev_b32_e32 [[SCALED:v[0-9]+]], 2, v0
AMDGPU: Use gfx9 carry-less add/sub instructions llvm-svn: 319491
2017-12-01 06:51:26 +08:00
; CI-DAG: v_sub_i32_e32 [[NEG:v[0-9]+]], vcc, 0, [[SCALED]]
; GFX9-DAG: v_sub_u32_e32 [[NEG:v[0-9]+]], 0, [[SCALED]]
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
; GCN-DAG: v_mov_b32_e32 [[K:v[0-9]+]], 13
; GCN-NOT: v_sub
; GCN: ds_write_b32 [[NEG]], [[K]] offset:123{{$}}
; GCN-NOT: v_sub
; GCN: ds_write_b32 [[NEG]], [[K]] offset:123{{$}}
; GCN: s_endpgm
AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @add_x_shl_neg_to_sub_multi_use_same_offset() #1 {
AMDGPU: Remove some old intrinsic uses from tests llvm-svn: 260493
2016-02-11 14:02:01 +08:00
%x.i = call i32 @llvm.amdgcn.workitem.id.x() #0
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
%neg = sub i32 0, %x.i
%shl = shl i32 %neg, 2
%add = add i32 123, %shl
%ptr = inttoptr i32 %add to i32 addrspace(3)*
store volatile i32 13, i32 addrspace(3)* %ptr
store volatile i32 13, i32 addrspace(3)* %ptr
ret void
}
; GCN-LABEL: {{^}}add_x_shl_neg_to_sub_misaligned_i64_max_offset:
; GCN-DAG: v_lshlrev_b32_e32 [[SCALED:v[0-9]+]], 2, v0
AMDGPU: Use gfx9 carry-less add/sub instructions llvm-svn: 319491
2017-12-01 06:51:26 +08:00
; CI-DAG: v_sub_i32_e32 [[NEG:v[0-9]+]], vcc, 0, [[SCALED]]
; GFX9-DAG: v_sub_u32_e32 [[NEG:v[0-9]+]], 0, [[SCALED]]
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
; GCN: ds_write2_b32 [[NEG]], {{v[0-9]+}}, {{v[0-9]+}} offset0:254 offset1:255
AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @add_x_shl_neg_to_sub_misaligned_i64_max_offset() #1 {
AMDGPU: Remove some old intrinsic uses from tests llvm-svn: 260493
2016-02-11 14:02:01 +08:00
%x.i = call i32 @llvm.amdgcn.workitem.id.x() #0
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
%neg = sub i32 0, %x.i
%shl = shl i32 %neg, 2
%add = add i32 1019, %shl
%ptr = inttoptr i32 %add to i64 addrspace(3)*
store i64 123, i64 addrspace(3)* %ptr, align 4
ret void
}
[AMDGPU] Add more test cases of D59608. Summary: - Add more test cases. Reviewers: arsenm Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D60071 llvm-svn: 357442
2019-04-02 08:36:37 +08:00
; GFX9-LABEL: {{^}}add_x_shl_neg_to_sub_misaligned_i64_max_offset_clamp_bit:
; GFX9: v_sub_u32
; GFX9: s_endpgm
define amdgpu_kernel void @add_x_shl_neg_to_sub_misaligned_i64_max_offset_clamp_bit(float %dummy.val) #1 {
%x.i = call i32 @llvm.amdgcn.workitem.id.x() #0
%neg = sub i32 0, %x.i
%shl = shl i32 %neg, 2
%add = add i32 1019, %shl
%ptr = inttoptr i32 %add to i64 addrspace(3)*
store i64 123, i64 addrspace(3)* %ptr, align 4
%fmas = call float @llvm.amdgcn.div.fmas.f32(float %dummy.val, float %dummy.val, float %dummy.val, i1 false)
store volatile float %fmas, float addrspace(1)* null
ret void
}
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
; GCN-LABEL: {{^}}add_x_shl_neg_to_sub_misaligned_i64_max_offset_p1:
; GCN-DAG: v_lshlrev_b32_e32 [[SCALED:v[0-9]+]], 2, v0
AMDGPU: Use gfx9 carry-less add/sub instructions llvm-svn: 319491
2017-12-01 06:51:26 +08:00
; CI-DAG: v_sub_i32_e32 [[NEG:v[0-9]+]], vcc, 0x3fc, [[SCALED]]
; GFX9-DAG: v_sub_u32_e32 [[NEG:v[0-9]+]], 0x3fc, [[SCALED]]
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
; GCN: ds_write2_b32 [[NEG]], {{v[0-9]+}}, {{v[0-9]+}} offset1:1{{$}}
AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @add_x_shl_neg_to_sub_misaligned_i64_max_offset_p1() #1 {
AMDGPU: Remove some old intrinsic uses from tests llvm-svn: 260493
2016-02-11 14:02:01 +08:00
%x.i = call i32 @llvm.amdgcn.workitem.id.x() #0
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
%neg = sub i32 0, %x.i
%shl = shl i32 %neg, 2
%add = add i32 1020, %shl
%ptr = inttoptr i32 %add to i64 addrspace(3)*
store i64 123, i64 addrspace(3)* %ptr, align 4
ret void
}
[AMDGPU] Add more test cases of D59608. Summary: - Add more test cases. Reviewers: arsenm Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D60071 llvm-svn: 357442
2019-04-02 08:36:37 +08:00
declare float @llvm.amdgcn.div.fmas.f32(float, float, float, i1)
AMDGPU: Handle sub of constant for DS offset folding sub C, x - > add (sub 0, x), C for DS offsets. This is mostly to fix regressions that show up when SeparateConstOffsetFromGEP is enabled. llvm-svn: 247054
2015-09-09 03:34:22 +08:00
attributes #0 = { nounwind readnone }
attributes #1 = { nounwind }
AMDGPU: Switch barrier intrinsics to using convergent noduplicate prevents unrolling of small loops that happen to have barriers in them. If a loop has a barrier in it, it is OK to duplicate it for the unroll. llvm-svn: 256075
2015-12-19 09:46:41 +08:00
attributes #2 = { nounwind convergent }