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/llvm.amdgcn.image.atomic.di...

218 lines
9.7 KiB
LLVM
Raw Normal View History

AMDGPU: Convert test cases to the dimension-aware intrinsics Summary: Also explicitly port over some tests in llvm.amdgcn.image.* that were missing. Some tests are removed because they no longer apply (i.e. explicitly testing building an address vector via insertelement). This is in preparation for the eventual removal of the old-style intrinsics. Some additional notes: - constant-address-space-32bit.ll: change some GCN-NEXT to GCN because the instruction schedule was subtly altered - insert_vector_elt.ll: the old test didn't actually test anything, because %tmp1 was not used; remove the load, because it doesn't work (Because of the amdgpu_ps calling convention? In any case, it's orthogonal to what the test claims to be testing.) Change-Id: Idfa99b6512ad139e755e82b8b89548ab08f0afcf Reviewers: arsenm, rampitec Subscribers: MatzeB, qcolombet, kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye, javed.absar, llvm-commits Differential Revision: https://reviews.llvm.org/D48018 llvm-svn: 335229
2018-06-21 21:37:19 +08:00
; RUN: llc -march=amdgcn -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
AMDGPU: Dimension-aware image intrinsics Summary: These new image intrinsics contain the texture type as part of their name and have each component of the address/coordinate as individual parameters. This is a preparatory step for implementing the A16 feature, where coordinates are passed as half-floats or -ints, but the Z compare value and texel offsets are still full dwords, making it difficult or impossible to distinguish between A16 on or off in the old-style intrinsics. Additionally, these intrinsics pass the 'texfailpolicy' and 'cachectrl' as i32 bit fields to reduce operand clutter and allow for future extensibility. v2: - gather4 supports 2darray images - fix a bug with 1D images on SI Change-Id: I099f309e0a394082a5901ea196c3967afb867f04 Reviewers: arsenm, rampitec, b-sumner Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, llvm-commits, t-tye Differential Revision: https://reviews.llvm.org/D44939 llvm-svn: 329166
2018-04-04 18:58:54 +08:00
; RUN: llc -march=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
; GCN-LABEL: {{^}}atomic_swap_1d:
; GCN: image_atomic_swap v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_swap_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_add_1d:
; GCN: image_atomic_add v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_add_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.add.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_sub_1d:
; GCN: image_atomic_sub v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_sub_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.sub.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_smin_1d:
; GCN: image_atomic_smin v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_smin_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.smin.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_umin_1d:
; GCN: image_atomic_umin v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_umin_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.umin.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_smax_1d:
; GCN: image_atomic_smax v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_smax_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.smax.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_umax_1d:
; GCN: image_atomic_umax v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_umax_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.umax.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_and_1d:
; GCN: image_atomic_and v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_and_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.and.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_or_1d:
; GCN: image_atomic_or v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_or_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.or.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_xor_1d:
; GCN: image_atomic_xor v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_xor_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.xor.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_inc_1d:
; GCN: image_atomic_inc v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_inc_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.inc.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_dec_1d:
; GCN: image_atomic_dec v0, v1, s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_dec_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.dec.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_cmpswap_1d:
; GCN: image_atomic_cmpswap v[0:1], v2, s[0:7] dmask:0x3 unorm glc{{$}}
define amdgpu_ps float @atomic_cmpswap_1d(<8 x i32> inreg %rsrc, i32 %cmp, i32 %swap, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_add_2d:
; GCN: image_atomic_add v0, v[1:2], s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_add_2d(<8 x i32> inreg %rsrc, i32 %data, i32 %s, i32 %t) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.add.2d.i32.i32(i32 %data, i32 %s, i32 %t, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_add_3d:
; GCN: image_atomic_add v0, v[1:4], s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_add_3d(<8 x i32> inreg %rsrc, i32 %data, i32 %s, i32 %t, i32 %r) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.add.3d.i32.i32(i32 %data, i32 %s, i32 %t, i32 %r, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_add_cube:
; GCN: image_atomic_add v0, v[1:4], s[0:7] dmask:0x1 unorm glc da{{$}}
define amdgpu_ps float @atomic_add_cube(<8 x i32> inreg %rsrc, i32 %data, i32 %s, i32 %t, i32 %face) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.add.cube.i32.i32(i32 %data, i32 %s, i32 %t, i32 %face, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_add_1darray:
; GCN: image_atomic_add v0, v[1:2], s[0:7] dmask:0x1 unorm glc da{{$}}
define amdgpu_ps float @atomic_add_1darray(<8 x i32> inreg %rsrc, i32 %data, i32 %s, i32 %slice) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.add.1darray.i32.i32(i32 %data, i32 %s, i32 %slice, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_add_2darray:
; GCN: image_atomic_add v0, v[1:4], s[0:7] dmask:0x1 unorm glc da{{$}}
define amdgpu_ps float @atomic_add_2darray(<8 x i32> inreg %rsrc, i32 %data, i32 %s, i32 %t, i32 %slice) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.add.2darray.i32.i32(i32 %data, i32 %s, i32 %t, i32 %slice, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_add_2dmsaa:
; GCN: image_atomic_add v0, v[1:4], s[0:7] dmask:0x1 unorm glc{{$}}
define amdgpu_ps float @atomic_add_2dmsaa(<8 x i32> inreg %rsrc, i32 %data, i32 %s, i32 %t, i32 %fragid) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.add.2dmsaa.i32.i32(i32 %data, i32 %s, i32 %t, i32 %fragid, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_add_2darraymsaa:
; GCN: image_atomic_add v0, v[1:4], s[0:7] dmask:0x1 unorm glc da{{$}}
define amdgpu_ps float @atomic_add_2darraymsaa(<8 x i32> inreg %rsrc, i32 %data, i32 %s, i32 %t, i32 %slice, i32 %fragid) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.add.2darraymsaa.i32.i32(i32 %data, i32 %s, i32 %t, i32 %slice, i32 %fragid, <8 x i32> %rsrc, i32 0, i32 0)
%out = bitcast i32 %v to float
ret float %out
}
; GCN-LABEL: {{^}}atomic_add_1d_slc:
; GCN: image_atomic_add v0, v1, s[0:7] dmask:0x1 unorm glc slc{{$}}
define amdgpu_ps float @atomic_add_1d_slc(<8 x i32> inreg %rsrc, i32 %data, i32 %s) {
main_body:
%v = call i32 @llvm.amdgcn.image.atomic.add.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 2)
%out = bitcast i32 %v to float
ret float %out
}
declare i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.add.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.sub.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.smin.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.umin.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.smax.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.umax.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.and.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.or.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.xor.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.inc.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.dec.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32, i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.add.2d.i32.i32(i32, i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.add.3d.i32.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.add.cube.i32.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.add.1darray.i32.i32(i32, i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.add.2darray.i32.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.add.2dmsaa.i32.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
declare i32 @llvm.amdgcn.image.atomic.add.2darraymsaa.i32.i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
attributes #0 = { nounwind }
attributes #1 = { nounwind readonly }
attributes #2 = { nounwind readnone }