7dbd6cd294
If a function has `amdgpu-flat-work-group-size`, honor it in `initialize` by taking its value directly; otherwise, it uses the default range as a starting point. We will no longer manipulate the known range, which can cause issues because the known range is a "throttle" to the assumed range such that the assumed range can't get widened properly in `updateImpl` if the known range is not set properly for whatever reasons. Another benefit of not touching the known range is, if we indicate pessimistic state, it also invalidates the AA such that `manifest` will not be called. Since we honor the attribute, we don't want and will not add any half-baked attribute added to a function.
74 lines
2.8 KiB
LLVM
74 lines
2.8 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --function-signature --check-globals
|
|
; RUN: opt -S -mtriple=amdgcn-amd-amdhsa -passes=amdgpu-attributor %s -o - | FileCheck %s
|
|
|
|
@g_fn = addrspace(1) global ptr null
|
|
|
|
;.
|
|
; CHECK: @g_fn = addrspace(1) global ptr null
|
|
;.
|
|
define void @set_fn(ptr %fn) {
|
|
; CHECK-LABEL: define {{[^@]+}}@set_fn
|
|
; CHECK-SAME: (ptr [[FN:%.*]]) #[[ATTR0:[0-9]+]] {
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: store ptr [[FN]], ptr addrspace(1) @g_fn, align 8
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
store ptr %fn, ptr addrspace(1) @g_fn
|
|
ret void
|
|
}
|
|
|
|
define void @get_fn(ptr %fn) {
|
|
; CHECK-LABEL: define {{[^@]+}}@get_fn
|
|
; CHECK-SAME: (ptr [[FN:%.*]]) #[[ATTR0]] {
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: [[LOAD:%.*]] = load ptr, ptr addrspace(1) @g_fn, align 8
|
|
; CHECK-NEXT: store ptr [[LOAD]], ptr [[FN]], align 8
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
%load = load ptr, ptr addrspace(1) @g_fn
|
|
store ptr %load, ptr %fn
|
|
ret void
|
|
}
|
|
|
|
define void @foo() {
|
|
; CHECK-LABEL: define {{[^@]+}}@foo
|
|
; CHECK-SAME: () #[[ATTR1:[0-9]+]] {
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: [[FN:%.*]] = alloca ptr, align 8, addrspace(5)
|
|
; CHECK-NEXT: store ptr null, ptr addrspace(5) [[FN]], align 8
|
|
; CHECK-NEXT: [[FN_CAST:%.*]] = addrspacecast ptr addrspace(5) [[FN]] to ptr
|
|
; CHECK-NEXT: call void @get_fn(ptr [[FN_CAST]])
|
|
; CHECK-NEXT: [[LOAD:%.*]] = load ptr, ptr addrspace(5) [[FN]], align 8
|
|
; CHECK-NEXT: [[TOBOOL:%.*]] = icmp ne ptr [[LOAD]], null
|
|
; CHECK-NEXT: br i1 [[TOBOOL]], label [[IF_THEN:%.*]], label [[IF_END:%.*]]
|
|
; CHECK: if.then:
|
|
; CHECK-NEXT: [[LOAD_1:%.*]] = load ptr, ptr addrspace(5) [[FN]], align 8
|
|
; CHECK-NEXT: call void [[LOAD_1]]()
|
|
; CHECK-NEXT: br label [[IF_END]]
|
|
; CHECK: if.end:
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
%fn = alloca ptr, addrspace(5)
|
|
store ptr null, ptr addrspace(5) %fn
|
|
%fn.cast = addrspacecast ptr addrspace(5) %fn to ptr
|
|
call void @get_fn(ptr %fn.cast)
|
|
%load = load ptr, ptr addrspace(5) %fn
|
|
%tobool = icmp ne ptr %load, null
|
|
br i1 %tobool, label %if.then, label %if.end
|
|
|
|
if.then:
|
|
%load.1 = load ptr, ptr addrspace(5) %fn
|
|
call void %load.1()
|
|
br label %if.end
|
|
|
|
if.end:
|
|
ret void
|
|
}
|
|
;.
|
|
; CHECK: attributes #[[ATTR0]] = { "amdgpu-no-agpr" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
|
|
; CHECK: attributes #[[ATTR1]] = { "uniform-work-group-size"="false" }
|
|
;.
|