caio/clang-p2996
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
4794bbffb228894de3ca0dad50fc027eb94f743e
clang-p2996/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll
Jon Chesterfield 3a20597776 [amdgpu] Implement lds kernel id intrinsic 
Implement an intrinsic for use lowering LDS variables to different
addresses from different kernels. This will allow kernels that cannot
reach an LDS variable to avoid wasting space for it.

There are a number of implicit arguments accessed by intrinsic already
so this implementation closely follows the existing handling. It is slightly
novel in that this SGPR is written by the kernel prologue.

It is necessary in the general case to put variables at different addresses
such that they can be compactly allocated and thus necessary for an
indirect function call to have some means of determining where a
given variable was allocated. Claiming an arbitrary SGPR into which
an integer can be written by the kernel, in this implementation based
on metadata associated with that kernel, which is then passed on to
indirect call sites is sufficient to determine the variable address.

The intent is to emit a __const array of LDS addresses and index into it.

Reviewed By: arsenm

Differential Revision: https://reviews.llvm.org/D125060
2022-07-19 17:46:19 +01:00

236 lines
16 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --function-signature --check-globals
; RUN: opt -mtriple=amdgcn-unknown-amdhsa -S -amdgpu-annotate-kernel-features < %s | FileCheck -check-prefixes=HSA,AKF_HSA %s
; RUN: opt -mtriple=amdgcn-unknown-amdhsa -S -amdgpu-attributor < %s | FileCheck -check-prefixes=HSA,ATTRIBUTOR_HSA %s
declare void @llvm.memcpy.p1i32.p4i32.i32(i32 addrspace(1)* nocapture, i32 addrspace(4)* nocapture, i32, i1) #0
@lds.i32 = unnamed_addr addrspace(3) global i32 undef, align 4
@lds.arr = unnamed_addr addrspace(3) global [256 x i32] undef, align 4
@global.i32 = unnamed_addr addrspace(1) global i32 undef, align 4
@global.arr = unnamed_addr addrspace(1) global [256 x i32] undef, align 4
;.
; HSA: @[[LDS_I32:[a-zA-Z0-9_$"\\.-]+]] = unnamed_addr addrspace(3) global i32 undef, align 4
; HSA: @[[LDS_ARR:[a-zA-Z0-9_$"\\.-]+]] = unnamed_addr addrspace(3) global [256 x i32] undef, align 4
; HSA: @[[GLOBAL_I32:[a-zA-Z0-9_$"\\.-]+]] = unnamed_addr addrspace(1) global i32 undef, align 4
; HSA: @[[GLOBAL_ARR:[a-zA-Z0-9_$"\\.-]+]] = unnamed_addr addrspace(1) global [256 x i32] undef, align 4
;.
define amdgpu_kernel void @store_cast_0_flat_to_group_addrspacecast() #1 {
; HSA-LABEL: define {{[^@]+}}@store_cast_0_flat_to_group_addrspacecast
; HSA-SAME: () #[[ATTR1:[0-9]+]] {
; HSA-NEXT: store i32 7, i32 addrspace(3)* addrspacecast (i32 addrspace(4)* null to i32 addrspace(3)*), align 4
; HSA-NEXT: ret void
;
store i32 7, i32 addrspace(3)* addrspacecast (i32 addrspace(4)* null to i32 addrspace(3)*)
ret void
}
define amdgpu_kernel void @store_cast_0_group_to_flat_addrspacecast() #1 {
; AKF_HSA-LABEL: define {{[^@]+}}@store_cast_0_group_to_flat_addrspacecast
; AKF_HSA-SAME: () #[[ATTR1]] {
; AKF_HSA-NEXT: store i32 7, i32 addrspace(4)* addrspacecast (i32 addrspace(3)* null to i32 addrspace(4)*), align 4
; AKF_HSA-NEXT: ret void
;
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@store_cast_0_group_to_flat_addrspacecast
; ATTRIBUTOR_HSA-SAME: () #[[ATTR2:[0-9]+]] {
; ATTRIBUTOR_HSA-NEXT: store i32 7, i32 addrspace(4)* addrspacecast (i32 addrspace(3)* null to i32 addrspace(4)*), align 4
; ATTRIBUTOR_HSA-NEXT: ret void
;
store i32 7, i32 addrspace(4)* addrspacecast (i32 addrspace(3)* null to i32 addrspace(4)*)
ret void
}
define amdgpu_kernel void @store_constant_cast_group_gv_to_flat() #1 {
; AKF_HSA-LABEL: define {{[^@]+}}@store_constant_cast_group_gv_to_flat
; AKF_HSA-SAME: () #[[ATTR1]] {
; AKF_HSA-NEXT: store i32 7, i32 addrspace(4)* addrspacecast (i32 addrspace(3)* @lds.i32 to i32 addrspace(4)*), align 4
; AKF_HSA-NEXT: ret void
;
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@store_constant_cast_group_gv_to_flat
; ATTRIBUTOR_HSA-SAME: () #[[ATTR2]] {
; ATTRIBUTOR_HSA-NEXT: store i32 7, i32 addrspace(4)* addrspacecast (i32 addrspace(3)* @lds.i32 to i32 addrspace(4)*), align 4
; ATTRIBUTOR_HSA-NEXT: ret void
;
store i32 7, i32 addrspace(4)* addrspacecast (i32 addrspace(3)* @lds.i32 to i32 addrspace(4)*)
ret void
}
define amdgpu_kernel void @store_constant_cast_group_gv_gep_to_flat() #1 {
; AKF_HSA-LABEL: define {{[^@]+}}@store_constant_cast_group_gv_gep_to_flat
; AKF_HSA-SAME: () #[[ATTR1]] {
; AKF_HSA-NEXT: store i32 7, i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), align 4
; AKF_HSA-NEXT: ret void
;
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@store_constant_cast_group_gv_gep_to_flat
; ATTRIBUTOR_HSA-SAME: () #[[ATTR2]] {
; ATTRIBUTOR_HSA-NEXT: store i32 7, i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), align 4
; ATTRIBUTOR_HSA-NEXT: ret void
;
store i32 7, i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8)
ret void
}
define amdgpu_kernel void @store_constant_cast_global_gv_to_flat() #1 {
; HSA-LABEL: define {{[^@]+}}@store_constant_cast_global_gv_to_flat
; HSA-SAME: () #[[ATTR1]] {
; HSA-NEXT: store i32 7, i32 addrspace(4)* addrspacecast (i32 addrspace(1)* @global.i32 to i32 addrspace(4)*), align 4
; HSA-NEXT: ret void
;
store i32 7, i32 addrspace(4)* addrspacecast (i32 addrspace(1)* @global.i32 to i32 addrspace(4)*)
ret void
}
define amdgpu_kernel void @store_constant_cast_global_gv_gep_to_flat() #1 {
; HSA-LABEL: define {{[^@]+}}@store_constant_cast_global_gv_gep_to_flat
; HSA-SAME: () #[[ATTR1]] {
; HSA-NEXT: store i32 7, i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(1)* @global.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), align 4
; HSA-NEXT: ret void
;
store i32 7, i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(1)* @global.arr to [256 x i32] addrspace(4)*), i64 0, i64 8)
ret void
}
define amdgpu_kernel void @load_constant_cast_group_gv_gep_to_flat(i32 addrspace(1)* %out) #1 {
; AKF_HSA-LABEL: define {{[^@]+}}@load_constant_cast_group_gv_gep_to_flat
; AKF_HSA-SAME: (i32 addrspace(1)* [[OUT:%.*]]) #[[ATTR1]] {
; AKF_HSA-NEXT: [[VAL:%.*]] = load i32, i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), align 4
; AKF_HSA-NEXT: store i32 [[VAL]], i32 addrspace(1)* [[OUT]], align 4
; AKF_HSA-NEXT: ret void
;
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@load_constant_cast_group_gv_gep_to_flat
; ATTRIBUTOR_HSA-SAME: (i32 addrspace(1)* [[OUT:%.*]]) #[[ATTR2]] {
; ATTRIBUTOR_HSA-NEXT: [[VAL:%.*]] = load i32, i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), align 4
; ATTRIBUTOR_HSA-NEXT: store i32 [[VAL]], i32 addrspace(1)* [[OUT]], align 4
; ATTRIBUTOR_HSA-NEXT: ret void
;
%val = load i32, i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8)
store i32 %val, i32 addrspace(1)* %out
ret void
}
define amdgpu_kernel void @atomicrmw_constant_cast_group_gv_gep_to_flat(i32 addrspace(1)* %out) #1 {
; AKF_HSA-LABEL: define {{[^@]+}}@atomicrmw_constant_cast_group_gv_gep_to_flat
; AKF_HSA-SAME: (i32 addrspace(1)* [[OUT:%.*]]) #[[ATTR1]] {
; AKF_HSA-NEXT: [[VAL:%.*]] = atomicrmw add i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 1 seq_cst, align 4
; AKF_HSA-NEXT: store i32 [[VAL]], i32 addrspace(1)* [[OUT]], align 4
; AKF_HSA-NEXT: ret void
;
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@atomicrmw_constant_cast_group_gv_gep_to_flat
; ATTRIBUTOR_HSA-SAME: (i32 addrspace(1)* [[OUT:%.*]]) #[[ATTR2]] {
; ATTRIBUTOR_HSA-NEXT: [[VAL:%.*]] = atomicrmw add i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 1 seq_cst, align 4
; ATTRIBUTOR_HSA-NEXT: store i32 [[VAL]], i32 addrspace(1)* [[OUT]], align 4
; ATTRIBUTOR_HSA-NEXT: ret void
;
%val = atomicrmw add i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 1 seq_cst
store i32 %val, i32 addrspace(1)* %out
ret void
}
define amdgpu_kernel void @cmpxchg_constant_cast_group_gv_gep_to_flat(i32 addrspace(1)* %out) #1 {
; AKF_HSA-LABEL: define {{[^@]+}}@cmpxchg_constant_cast_group_gv_gep_to_flat
; AKF_HSA-SAME: (i32 addrspace(1)* [[OUT:%.*]]) #[[ATTR1]] {
; AKF_HSA-NEXT: [[VAL:%.*]] = cmpxchg i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 0, i32 1 seq_cst seq_cst, align 4
; AKF_HSA-NEXT: [[VAL0:%.*]] = extractvalue { i32, i1 } [[VAL]], 0
; AKF_HSA-NEXT: store i32 [[VAL0]], i32 addrspace(1)* [[OUT]], align 4
; AKF_HSA-NEXT: ret void
;
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@cmpxchg_constant_cast_group_gv_gep_to_flat
; ATTRIBUTOR_HSA-SAME: (i32 addrspace(1)* [[OUT:%.*]]) #[[ATTR2]] {
; ATTRIBUTOR_HSA-NEXT: [[VAL:%.*]] = cmpxchg i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 0, i32 1 seq_cst seq_cst, align 4
; ATTRIBUTOR_HSA-NEXT: [[VAL0:%.*]] = extractvalue { i32, i1 } [[VAL]], 0
; ATTRIBUTOR_HSA-NEXT: store i32 [[VAL0]], i32 addrspace(1)* [[OUT]], align 4
; ATTRIBUTOR_HSA-NEXT: ret void
;
%val = cmpxchg i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 0, i32 1 seq_cst seq_cst
%val0 = extractvalue { i32, i1 } %val, 0
store i32 %val0, i32 addrspace(1)* %out
ret void
}
define amdgpu_kernel void @memcpy_constant_cast_group_gv_gep_to_flat(i32 addrspace(1)* %out) #1 {
; AKF_HSA-LABEL: define {{[^@]+}}@memcpy_constant_cast_group_gv_gep_to_flat
; AKF_HSA-SAME: (i32 addrspace(1)* [[OUT:%.*]]) #[[ATTR1]] {
; AKF_HSA-NEXT: call void @llvm.memcpy.p1i32.p4i32.i32(i32 addrspace(1)* align 4 [[OUT]], i32 addrspace(4)* align 4 getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 32, i1 false)
; AKF_HSA-NEXT: ret void
;
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@memcpy_constant_cast_group_gv_gep_to_flat
; ATTRIBUTOR_HSA-SAME: (i32 addrspace(1)* [[OUT:%.*]]) #[[ATTR2]] {
; ATTRIBUTOR_HSA-NEXT: call void @llvm.memcpy.p1i32.p4i32.i32(i32 addrspace(1)* align 4 [[OUT]], i32 addrspace(4)* align 4 getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 32, i1 false)
; ATTRIBUTOR_HSA-NEXT: ret void
;
call void @llvm.memcpy.p1i32.p4i32.i32(i32 addrspace(1)* align 4 %out, i32 addrspace(4)* align 4 getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 32, i1 false)
ret void
}
; Can't just search the pointer value
define amdgpu_kernel void @store_value_constant_cast_lds_gv_gep_to_flat(i32 addrspace(4)* addrspace(1)* %out) #1 {
; AKF_HSA-LABEL: define {{[^@]+}}@store_value_constant_cast_lds_gv_gep_to_flat
; AKF_HSA-SAME: (i32 addrspace(4)* addrspace(1)* [[OUT:%.*]]) #[[ATTR1]] {
; AKF_HSA-NEXT: store i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 addrspace(4)* addrspace(1)* [[OUT]], align 8
; AKF_HSA-NEXT: ret void
;
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@store_value_constant_cast_lds_gv_gep_to_flat
; ATTRIBUTOR_HSA-SAME: (i32 addrspace(4)* addrspace(1)* [[OUT:%.*]]) #[[ATTR2]] {
; ATTRIBUTOR_HSA-NEXT: store i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 addrspace(4)* addrspace(1)* [[OUT]], align 8
; ATTRIBUTOR_HSA-NEXT: ret void
;
store i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8), i32 addrspace(4)* addrspace(1)* %out
ret void
}
; Can't just search pointer types
define amdgpu_kernel void @store_ptrtoint_value_constant_cast_lds_gv_gep_to_flat(i64 addrspace(1)* %out) #1 {
; AKF_HSA-LABEL: define {{[^@]+}}@store_ptrtoint_value_constant_cast_lds_gv_gep_to_flat
; AKF_HSA-SAME: (i64 addrspace(1)* [[OUT:%.*]]) #[[ATTR1]] {
; AKF_HSA-NEXT: store i64 ptrtoint (i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8) to i64), i64 addrspace(1)* [[OUT]], align 4
; AKF_HSA-NEXT: ret void
;
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@store_ptrtoint_value_constant_cast_lds_gv_gep_to_flat
; ATTRIBUTOR_HSA-SAME: (i64 addrspace(1)* [[OUT:%.*]]) #[[ATTR2]] {
; ATTRIBUTOR_HSA-NEXT: store i64 ptrtoint (i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8) to i64), i64 addrspace(1)* [[OUT]], align 4
; ATTRIBUTOR_HSA-NEXT: ret void
;
store i64 ptrtoint (i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8) to i64), i64 addrspace(1)* %out
ret void
}
; Cast group to flat, do GEP, cast back to group
define amdgpu_kernel void @store_constant_cast_group_gv_gep_to_flat_to_group() #1 {
; AKF_HSA-LABEL: define {{[^@]+}}@store_constant_cast_group_gv_gep_to_flat_to_group
; AKF_HSA-SAME: () #[[ATTR1]] {
; AKF_HSA-NEXT: store i32 7, i32 addrspace(3)* addrspacecast (i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8) to i32 addrspace(3)*), align 4
; AKF_HSA-NEXT: ret void
;
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@store_constant_cast_group_gv_gep_to_flat_to_group
; ATTRIBUTOR_HSA-SAME: () #[[ATTR2]] {
; ATTRIBUTOR_HSA-NEXT: store i32 7, i32 addrspace(3)* addrspacecast (i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8) to i32 addrspace(3)*), align 4
; ATTRIBUTOR_HSA-NEXT: ret void
;
store i32 7, i32 addrspace(3)* addrspacecast (i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8) to i32 addrspace(3)*)
ret void
}
define i32 addrspace(3)* @ret_constant_cast_group_gv_gep_to_flat_to_group() #1 {
; AKF_HSA-LABEL: define {{[^@]+}}@ret_constant_cast_group_gv_gep_to_flat_to_group
; AKF_HSA-SAME: () #[[ATTR1]] {
; AKF_HSA-NEXT: ret i32 addrspace(3)* addrspacecast (i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8) to i32 addrspace(3)*)
;
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@ret_constant_cast_group_gv_gep_to_flat_to_group
; ATTRIBUTOR_HSA-SAME: () #[[ATTR2]] {
; ATTRIBUTOR_HSA-NEXT: ret i32 addrspace(3)* addrspacecast (i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8) to i32 addrspace(3)*)
;
ret i32 addrspace(3)* addrspacecast (i32 addrspace(4)* getelementptr ([256 x i32], [256 x i32] addrspace(4)* addrspacecast ([256 x i32] addrspace(3)* @lds.arr to [256 x i32] addrspace(4)*), i64 0, i64 8) to i32 addrspace(3)*)
}
attributes #0 = { argmemonly nounwind }
attributes #1 = { nounwind }
;.
; AKF_HSA: attributes #[[ATTR0:[0-9]+]] = { argmemonly nofree nounwind willreturn }
; AKF_HSA: attributes #[[ATTR1]] = { nounwind }
;.
; ATTRIBUTOR_HSA: attributes #[[ATTR0:[0-9]+]] = { argmemonly nofree nounwind willreturn }
; ATTRIBUTOR_HSA: attributes #[[ATTR1]] = { nounwind "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "amdgpu-no-queue-ptr" "amdgpu-no-workgroup-id-x" "amdgpu-no-workgroup-id-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
; ATTRIBUTOR_HSA: attributes #[[ATTR2]] = { nounwind "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "amdgpu-no-workgroup-id-x" "amdgpu-no-workgroup-id-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
;.
Reference in New Issue View Git Blame Copy Permalink