
In most cases, the type information attached to load and store instructions is meaningless and inconsistently applied. We can usually use ".b" loads and avoid the complexity of trying to assign the correct type. The one expectation is sign-extending load, which will continue to use ".s" to ensure the sign extension into a larger register is done correctly.
132 lines
6.3 KiB
LLVM
132 lines
6.3 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
|
|
; RUN: llc < %s -march=nvptx64 -mcpu=sm_100a -mattr=+ptx86 | FileCheck --check-prefixes=CHECK_PTX64 %s
|
|
; RUN: llc < %s -march=nvptx64 -mcpu=sm_100a -mattr=+ptx86 --nvptx-short-ptr | FileCheck --check-prefixes=CHECK_PTX64_SHARED32 %s
|
|
; RUN: %if ptxas-12.8 %{ llc < %s -march=nvptx64 -mcpu=sm_100a -mattr=+ptx86 | %ptxas-verify -arch=sm_100a %}
|
|
; RUN: %if ptxas-12.8 %{ llc < %s -march=nvptx64 -mcpu=sm_100a -mattr=+ptx86 --nvptx-short-ptr | %ptxas-verify -arch=sm_100a %}
|
|
|
|
declare void @llvm.nvvm.tcgen05.alloc.cg1(ptr %addr, i32 %ncols)
|
|
declare void @llvm.nvvm.tcgen05.alloc.cg2(ptr %addr, i32 %ncols)
|
|
declare void @llvm.nvvm.tcgen05.alloc.shared.cg1(ptr addrspace(3) %addr, i32 %ncols)
|
|
declare void @llvm.nvvm.tcgen05.alloc.shared.cg2(ptr addrspace(3) %addr, i32 %ncols)
|
|
|
|
; CHECK-LABEL: test_tcgen05_alloc
|
|
define void @test_tcgen05_alloc(ptr %addr, i32 %ncols) {
|
|
; CHECK_PTX64-LABEL: test_tcgen05_alloc(
|
|
; CHECK_PTX64: {
|
|
; CHECK_PTX64-NEXT: .reg .b32 %r<2>;
|
|
; CHECK_PTX64-NEXT: .reg .b64 %rd<2>;
|
|
; CHECK_PTX64-EMPTY:
|
|
; CHECK_PTX64-NEXT: // %bb.0:
|
|
; CHECK_PTX64-NEXT: ld.param.b64 %rd1, [test_tcgen05_alloc_param_0];
|
|
; CHECK_PTX64-NEXT: ld.param.b32 %r1, [test_tcgen05_alloc_param_1];
|
|
; CHECK_PTX64-NEXT: tcgen05.alloc.cta_group::1.sync.aligned.b32 [%rd1], %r1;
|
|
; CHECK_PTX64-NEXT: tcgen05.alloc.cta_group::2.sync.aligned.b32 [%rd1], %r1;
|
|
; CHECK_PTX64-NEXT: ret;
|
|
;
|
|
; CHECK_PTX64_SHARED32-LABEL: test_tcgen05_alloc(
|
|
; CHECK_PTX64_SHARED32: {
|
|
; CHECK_PTX64_SHARED32-NEXT: .reg .b32 %r<2>;
|
|
; CHECK_PTX64_SHARED32-NEXT: .reg .b64 %rd<2>;
|
|
; CHECK_PTX64_SHARED32-EMPTY:
|
|
; CHECK_PTX64_SHARED32-NEXT: // %bb.0:
|
|
; CHECK_PTX64_SHARED32-NEXT: ld.param.b64 %rd1, [test_tcgen05_alloc_param_0];
|
|
; CHECK_PTX64_SHARED32-NEXT: ld.param.b32 %r1, [test_tcgen05_alloc_param_1];
|
|
; CHECK_PTX64_SHARED32-NEXT: tcgen05.alloc.cta_group::1.sync.aligned.b32 [%rd1], %r1;
|
|
; CHECK_PTX64_SHARED32-NEXT: tcgen05.alloc.cta_group::2.sync.aligned.b32 [%rd1], %r1;
|
|
; CHECK_PTX64_SHARED32-NEXT: ret;
|
|
call void @llvm.nvvm.tcgen05.alloc.cg1(ptr %addr, i32 %ncols)
|
|
call void @llvm.nvvm.tcgen05.alloc.cg2(ptr %addr, i32 %ncols)
|
|
|
|
ret void
|
|
}
|
|
|
|
; CHECK-LABEL: test_tcgen05_alloc_shared
|
|
define void @test_tcgen05_alloc_shared(ptr addrspace(3) %addr, i32 %ncols) {
|
|
; CHECK_PTX64-LABEL: test_tcgen05_alloc_shared(
|
|
; CHECK_PTX64: {
|
|
; CHECK_PTX64-NEXT: .reg .b32 %r<2>;
|
|
; CHECK_PTX64-NEXT: .reg .b64 %rd<2>;
|
|
; CHECK_PTX64-EMPTY:
|
|
; CHECK_PTX64-NEXT: // %bb.0:
|
|
; CHECK_PTX64-NEXT: ld.param.b64 %rd1, [test_tcgen05_alloc_shared_param_0];
|
|
; CHECK_PTX64-NEXT: ld.param.b32 %r1, [test_tcgen05_alloc_shared_param_1];
|
|
; CHECK_PTX64-NEXT: tcgen05.alloc.cta_group::1.sync.aligned.shared::cta.b32 [%rd1], %r1;
|
|
; CHECK_PTX64-NEXT: tcgen05.alloc.cta_group::2.sync.aligned.shared::cta.b32 [%rd1], %r1;
|
|
; CHECK_PTX64-NEXT: ret;
|
|
;
|
|
; CHECK_PTX64_SHARED32-LABEL: test_tcgen05_alloc_shared(
|
|
; CHECK_PTX64_SHARED32: {
|
|
; CHECK_PTX64_SHARED32-NEXT: .reg .b32 %r<3>;
|
|
; CHECK_PTX64_SHARED32-EMPTY:
|
|
; CHECK_PTX64_SHARED32-NEXT: // %bb.0:
|
|
; CHECK_PTX64_SHARED32-NEXT: ld.param.b32 %r1, [test_tcgen05_alloc_shared_param_0];
|
|
; CHECK_PTX64_SHARED32-NEXT: ld.param.b32 %r2, [test_tcgen05_alloc_shared_param_1];
|
|
; CHECK_PTX64_SHARED32-NEXT: tcgen05.alloc.cta_group::1.sync.aligned.shared::cta.b32 [%r1], %r2;
|
|
; CHECK_PTX64_SHARED32-NEXT: tcgen05.alloc.cta_group::2.sync.aligned.shared::cta.b32 [%r1], %r2;
|
|
; CHECK_PTX64_SHARED32-NEXT: ret;
|
|
call void @llvm.nvvm.tcgen05.alloc.shared.cg1(ptr addrspace(3) %addr, i32 %ncols)
|
|
|
|
call void @llvm.nvvm.tcgen05.alloc.shared.cg2(ptr addrspace(3) %addr, i32 %ncols)
|
|
ret void
|
|
}
|
|
|
|
declare void @llvm.nvvm.tcgen05.dealloc.cg1(ptr addrspace(6) %tmem_addr, i32 %ncols)
|
|
declare void @llvm.nvvm.tcgen05.dealloc.cg2(ptr addrspace(6) %tmem_addr, i32 %ncols)
|
|
|
|
; CHECK-LABEL: test_tcgen05_dealloc
|
|
define void @test_tcgen05_dealloc(ptr addrspace(6) %tmem_addr, i32 %ncols) {
|
|
; CHECK_PTX64-LABEL: test_tcgen05_dealloc(
|
|
; CHECK_PTX64: {
|
|
; CHECK_PTX64-NEXT: .reg .b32 %r<3>;
|
|
; CHECK_PTX64-EMPTY:
|
|
; CHECK_PTX64-NEXT: // %bb.0:
|
|
; CHECK_PTX64-NEXT: ld.param.b32 %r1, [test_tcgen05_dealloc_param_0];
|
|
; CHECK_PTX64-NEXT: ld.param.b32 %r2, [test_tcgen05_dealloc_param_1];
|
|
; CHECK_PTX64-NEXT: tcgen05.dealloc.cta_group::1.sync.aligned.b32 %r1, %r2;
|
|
; CHECK_PTX64-NEXT: tcgen05.dealloc.cta_group::2.sync.aligned.b32 %r1, %r2;
|
|
; CHECK_PTX64-NEXT: ret;
|
|
;
|
|
; CHECK_PTX64_SHARED32-LABEL: test_tcgen05_dealloc(
|
|
; CHECK_PTX64_SHARED32: {
|
|
; CHECK_PTX64_SHARED32-NEXT: .reg .b32 %r<3>;
|
|
; CHECK_PTX64_SHARED32-EMPTY:
|
|
; CHECK_PTX64_SHARED32-NEXT: // %bb.0:
|
|
; CHECK_PTX64_SHARED32-NEXT: ld.param.b32 %r1, [test_tcgen05_dealloc_param_0];
|
|
; CHECK_PTX64_SHARED32-NEXT: ld.param.b32 %r2, [test_tcgen05_dealloc_param_1];
|
|
; CHECK_PTX64_SHARED32-NEXT: tcgen05.dealloc.cta_group::1.sync.aligned.b32 %r1, %r2;
|
|
; CHECK_PTX64_SHARED32-NEXT: tcgen05.dealloc.cta_group::2.sync.aligned.b32 %r1, %r2;
|
|
; CHECK_PTX64_SHARED32-NEXT: ret;
|
|
call void @llvm.nvvm.tcgen05.dealloc.cg1(ptr addrspace(6) %tmem_addr, i32 %ncols)
|
|
|
|
call void @llvm.nvvm.tcgen05.dealloc.cg2(ptr addrspace(6) %tmem_addr, i32 %ncols)
|
|
ret void
|
|
}
|
|
|
|
declare void @llvm.nvvm.tcgen05.relinq.alloc.permit.cg1()
|
|
declare void @llvm.nvvm.tcgen05.relinq.alloc.permit.cg2()
|
|
|
|
; CHECK-LABEL: test_tcgen05_relinquish_alloc_permit
|
|
define void @test_tcgen05_relinquish_alloc_permit() {
|
|
; CHECK_PTX64-LABEL: test_tcgen05_relinquish_alloc_permit(
|
|
; CHECK_PTX64: {
|
|
; CHECK_PTX64-EMPTY:
|
|
; CHECK_PTX64-EMPTY:
|
|
; CHECK_PTX64-NEXT: // %bb.0:
|
|
; CHECK_PTX64-NEXT: tcgen05.relinquish_alloc_permit.cta_group::1.sync.aligned;
|
|
; CHECK_PTX64-NEXT: tcgen05.relinquish_alloc_permit.cta_group::2.sync.aligned;
|
|
; CHECK_PTX64-NEXT: ret;
|
|
;
|
|
; CHECK_PTX64_SHARED32-LABEL: test_tcgen05_relinquish_alloc_permit(
|
|
; CHECK_PTX64_SHARED32: {
|
|
; CHECK_PTX64_SHARED32-EMPTY:
|
|
; CHECK_PTX64_SHARED32-EMPTY:
|
|
; CHECK_PTX64_SHARED32-NEXT: // %bb.0:
|
|
; CHECK_PTX64_SHARED32-NEXT: tcgen05.relinquish_alloc_permit.cta_group::1.sync.aligned;
|
|
; CHECK_PTX64_SHARED32-NEXT: tcgen05.relinquish_alloc_permit.cta_group::2.sync.aligned;
|
|
; CHECK_PTX64_SHARED32-NEXT: ret;
|
|
call void @llvm.nvvm.tcgen05.relinq.alloc.permit.cg1()
|
|
|
|
call void @llvm.nvvm.tcgen05.relinq.alloc.permit.cg2()
|
|
ret void
|
|
}
|