maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/clang/test/CodeGenOpenCL/addr-space-struct-arg.cl

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

178 lines
7.6 KiB
Common Lisp
Raw Normal View History

[FileCheck] Support comment directives Sometimes you want to disable a FileCheck directive without removing it entirely, or you want to write comments that mention a directive by name. The `COM:` directive makes it easy to do this. For example, you might have: ``` ; X32: pinsrd_1: ; X32: pinsrd $1, 4(%esp), %xmm0 ; COM: FIXME: X64 isn't working correctly yet for this part of codegen, but ; COM: X64 will have something similar to X32: ; COM: ; COM: X64: pinsrd_1: ; COM: X64: pinsrd $1, %edi, %xmm0 ``` Without this patch, you need to use some combination of rewording and directive syntax mangling to prevent FileCheck from recognizing the commented occurrences of `X32:` and `X64:` above as directives. Moreover, FileCheck diagnostics have been proposed that might complain about the occurrences of `X64` that don't have the trailing `:` because they look like directive typos: <http://lists.llvm.org/pipermail/llvm-dev/2020-April/140610.html> I think dodging all these problems can prove tedious for test authors, and directive syntax mangling already makes the purpose of existing test code unclear. `COM:` can avoid all these problems. This patch also updates the small set of existing tests that define `COM` as a check prefix: - clang/test/CodeGen/default-address-space.c - clang/test/CodeGenOpenCL/addr-space-struct-arg.cl - clang/test/Driver/hip-device-libs.hip - llvm/test/Assembler/drop-debug-info-nonzero-alloca.ll I think lit should support `COM:` as well. Perhaps `clang -verify` should too. Reviewed By: jhenderson, thopre Differential Revision: https://reviews.llvm.org/D79276
2020-05-05 06:05:55 +08:00
// RUN: %clang_cc1 %s -emit-llvm -o - -O0 -ffake-address-space-map -triple i686-pc-darwin | FileCheck -enable-var-scope -check-prefixes=ALL,X86 %s
// RUN: %clang_cc1 %s -emit-llvm -o - -O0 -triple amdgcn | FileCheck -enable-var-scope -check-prefixes=ALL,AMDGCN %s
// RUN: %clang_cc1 %s -emit-llvm -o - -cl-std=CL2.0 -O0 -triple amdgcn | FileCheck -enable-var-scope -check-prefixes=ALL,AMDGCN,AMDGCN20 %s
[OpenCL][CodeGen] Fix replacing memcpy with addrspacecast Summary: If a function argument is byval and RV is located in default or alloca address space an optimization of creating addrspacecast instead of memcpy is performed. That is not correct for OpenCL, where that can lead to a situation of address space casting from __private * to __global *. See an example below: ``` typedef struct { int x; } MyStruct; void foo(MyStruct val) {} kernel void KernelOneMember(__global MyStruct* x) { foo (*x); } ``` for this code clang generated following IR: ... %0 = load %struct.MyStruct addrspace(1)*, %struct.MyStruct addrspace(1)** %x.addr, align 4 %1 = addrspacecast %struct.MyStruct addrspace(1)* %0 to %struct.MyStruct* ... So the optimization was disallowed for OpenCL if RV is located in an address space different than that of the argument (0). Reviewers: yaxunl, Anastasia Reviewed By: Anastasia Subscribers: cfe-commits, asavonic Differential Revision: https://reviews.llvm.org/D54947 llvm-svn: 348752
2018-12-10 20:03:00 +08:00
// RUN: %clang_cc1 %s -emit-llvm -o - -cl-std=CL1.2 -O0 -triple spir-unknown-unknown-unknown | FileCheck -enable-var-scope -check-prefixes=SPIR %s
typedef int int2 __attribute__((ext_vector_type(2)));
Prune CRLF. llvm-svn: 220678
2014-10-27 20:37:26 +08:00
typedef struct {
int cells[9];
} Mat3X3;
typedef struct {
int cells[16];
} Mat4X4;
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
typedef struct {
int cells[1024];
} Mat32X32;
typedef struct {
int cells[4096];
} Mat64X64;
CodeGen: Fix invalid bitcast for coerced function argument Clang assumes coerced function argument is in address space 0, which is not always true and results in invalid bitcasts. This patch fixes failure in OpenCL conformance test api/get_kernel_arg_info with amdgcn---amdgizcl triple, where non-zero alloca address space is used. Differential Revision: https://reviews.llvm.org/D34777 llvm-svn: 306721
2017-06-30 02:47:45 +08:00
struct StructOneMember {
int2 x;
};
struct StructTwoMember {
int2 x;
int2 y;
};
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
struct LargeStructOneMember {
int2 x[100];
};
struct LargeStructTwoMember {
int2 x[40];
int2 y[20];
};
Recommit r326946 after reducing CallArgList memory footprint llvm-svn: 327634
2018-03-15 23:25:19 +08:00
#if __OPENCL_C_VERSION__ >= 200
struct LargeStructOneMember g_s;
#endif
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
[test] Add {{.*}} to make tests immune to dso_local/dso_preemptable/(none) differences For a definition (of most linkage types), dso_local is set for ELF -fno-pic/-fpie and COFF, but not for Mach-O. This nuance causes unneeded binary format differences. This patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} ` if there is an explicit linkage. * Clang will set dso_local for Mach-O, which is currently implied by TargetMachine.cpp. This will make COFF/Mach-O and executable ELF similar. * Eventually I hope we can make dso_local the textual LLVM IR default (write explicit "dso_preemptable" when applicable) and -fpic ELF will be similar to everything else. This patch helps move toward that goal.
2020-12-31 12:45:56 +08:00
// X86-LABEL: define{{.*}} void @foo(%struct.Mat4X4* noalias sret(%struct.Mat4X4) align 4 %agg.result, %struct.Mat3X3* byval(%struct.Mat3X3) align 4 %in)
// AMDGCN-LABEL: define{{.*}} %struct.Mat4X4 @foo([9 x i32] %in.coerce)
Prune CRLF. llvm-svn: 220678
2014-10-27 20:37:26 +08:00
Mat4X4 __attribute__((noinline)) foo(Mat3X3 in) {
Mat4X4 out;
return out;
}
[FileCheck] Support comment directives Sometimes you want to disable a FileCheck directive without removing it entirely, or you want to write comments that mention a directive by name. The `COM:` directive makes it easy to do this. For example, you might have: ``` ; X32: pinsrd_1: ; X32: pinsrd $1, 4(%esp), %xmm0 ; COM: FIXME: X64 isn't working correctly yet for this part of codegen, but ; COM: X64 will have something similar to X32: ; COM: ; COM: X64: pinsrd_1: ; COM: X64: pinsrd $1, %edi, %xmm0 ``` Without this patch, you need to use some combination of rewording and directive syntax mangling to prevent FileCheck from recognizing the commented occurrences of `X32:` and `X64:` above as directives. Moreover, FileCheck diagnostics have been proposed that might complain about the occurrences of `X64` that don't have the trailing `:` because they look like directive typos: <http://lists.llvm.org/pipermail/llvm-dev/2020-April/140610.html> I think dodging all these problems can prove tedious for test authors, and directive syntax mangling already makes the purpose of existing test code unclear. `COM:` can avoid all these problems. This patch also updates the small set of existing tests that define `COM` as a check prefix: - clang/test/CodeGen/default-address-space.c - clang/test/CodeGenOpenCL/addr-space-struct-arg.cl - clang/test/Driver/hip-device-libs.hip - llvm/test/Assembler/drop-debug-info-nonzero-alloca.ll I think lit should support `COM:` as well. Perhaps `clang -verify` should too. Reviewed By: jhenderson, thopre Differential Revision: https://reviews.llvm.org/D79276
2020-05-05 06:05:55 +08:00
// ALL-LABEL: define {{.*}} void @ker
CodeGen: Fix invalid bitcast for coerced function argument Clang assumes coerced function argument is in address space 0, which is not always true and results in invalid bitcasts. This patch fixes failure in OpenCL conformance test api/get_kernel_arg_info with amdgcn---amdgizcl triple, where non-zero alloca address space is used. Differential Revision: https://reviews.llvm.org/D34777 llvm-svn: 306721
2017-06-30 02:47:45 +08:00
// Expect two mem copies: one for the argument "in", and one for
// the return value.
// X86: call void @llvm.memcpy.p0i8.p1i8.i32(i8*
// X86: call void @llvm.memcpy.p1i8.p0i8.i32(i8 addrspace(1)*
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
Clean up AMDGCN tests Differential Revision: https://reviews.llvm.org/D43340 llvm-svn: 325279
2018-02-16 03:12:41 +08:00
// AMDGCN: load [9 x i32], [9 x i32] addrspace(1)*
// AMDGCN: call %struct.Mat4X4 @foo([9 x i32]
// AMDGCN: call void @llvm.memcpy.p1i8.p5i8.i64(i8 addrspace(1)*
Prune CRLF. llvm-svn: 220678
2014-10-27 20:37:26 +08:00
kernel void ker(global Mat3X3 *in, global Mat4X4 *out) {
out[0] = foo(in[1]);
}
[test] Add {{.*}} to make tests immune to dso_local/dso_preemptable/(none) differences For a definition (of most linkage types), dso_local is set for ELF -fno-pic/-fpie and COFF, but not for Mach-O. This nuance causes unneeded binary format differences. This patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} ` if there is an explicit linkage. * Clang will set dso_local for Mach-O, which is currently implied by TargetMachine.cpp. This will make COFF/Mach-O and executable ELF similar. * Eventually I hope we can make dso_local the textual LLVM IR default (write explicit "dso_preemptable" when applicable) and -fpic ELF will be similar to everything else. This patch helps move toward that goal.
2020-12-31 12:45:56 +08:00
// X86-LABEL: define{{.*}} void @foo_large(%struct.Mat64X64* noalias sret(%struct.Mat64X64) align 4 %agg.result, %struct.Mat32X32* byval(%struct.Mat32X32) align 4 %in)
// AMDGCN-LABEL: define{{.*}} void @foo_large(%struct.Mat64X64 addrspace(5)* noalias sret(%struct.Mat64X64) align 4 %agg.result, %struct.Mat32X32 addrspace(5)* byval(%struct.Mat32X32) align 4 %in)
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
Mat64X64 __attribute__((noinline)) foo_large(Mat32X32 in) {
Mat64X64 out;
return out;
}
[FileCheck] Support comment directives Sometimes you want to disable a FileCheck directive without removing it entirely, or you want to write comments that mention a directive by name. The `COM:` directive makes it easy to do this. For example, you might have: ``` ; X32: pinsrd_1: ; X32: pinsrd $1, 4(%esp), %xmm0 ; COM: FIXME: X64 isn't working correctly yet for this part of codegen, but ; COM: X64 will have something similar to X32: ; COM: ; COM: X64: pinsrd_1: ; COM: X64: pinsrd $1, %edi, %xmm0 ``` Without this patch, you need to use some combination of rewording and directive syntax mangling to prevent FileCheck from recognizing the commented occurrences of `X32:` and `X64:` above as directives. Moreover, FileCheck diagnostics have been proposed that might complain about the occurrences of `X64` that don't have the trailing `:` because they look like directive typos: <http://lists.llvm.org/pipermail/llvm-dev/2020-April/140610.html> I think dodging all these problems can prove tedious for test authors, and directive syntax mangling already makes the purpose of existing test code unclear. `COM:` can avoid all these problems. This patch also updates the small set of existing tests that define `COM` as a check prefix: - clang/test/CodeGen/default-address-space.c - clang/test/CodeGenOpenCL/addr-space-struct-arg.cl - clang/test/Driver/hip-device-libs.hip - llvm/test/Assembler/drop-debug-info-nonzero-alloca.ll I think lit should support `COM:` as well. Perhaps `clang -verify` should too. Reviewed By: jhenderson, thopre Differential Revision: https://reviews.llvm.org/D79276
2020-05-05 06:05:55 +08:00
// ALL-LABEL: define {{.*}} void @ker_large
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
// Expect two mem copies: one for the argument "in", and one for
// the return value.
// X86: call void @llvm.memcpy.p0i8.p1i8.i32(i8*
// X86: call void @llvm.memcpy.p1i8.p0i8.i32(i8 addrspace(1)*
Clean up AMDGCN tests Differential Revision: https://reviews.llvm.org/D43340 llvm-svn: 325279
2018-02-16 03:12:41 +08:00
// AMDGCN: call void @llvm.memcpy.p5i8.p1i8.i64(i8 addrspace(5)*
// AMDGCN: call void @llvm.memcpy.p1i8.p5i8.i64(i8 addrspace(1)*
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
kernel void ker_large(global Mat32X32 *in, global Mat64X64 *out) {
out[0] = foo_large(in[1]);
}
[test] Add {{.*}} to make tests immune to dso_local/dso_preemptable/(none) differences For a definition (of most linkage types), dso_local is set for ELF -fno-pic/-fpie and COFF, but not for Mach-O. This nuance causes unneeded binary format differences. This patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} ` if there is an explicit linkage. * Clang will set dso_local for Mach-O, which is currently implied by TargetMachine.cpp. This will make COFF/Mach-O and executable ELF similar. * Eventually I hope we can make dso_local the textual LLVM IR default (write explicit "dso_preemptable" when applicable) and -fpic ELF will be similar to everything else. This patch helps move toward that goal.
2020-12-31 12:45:56 +08:00
// AMDGCN-LABEL: define{{.*}} void @FuncOneMember(<2 x i32> %u.coerce)
CodeGen: Fix invalid bitcast for coerced function argument Clang assumes coerced function argument is in address space 0, which is not always true and results in invalid bitcasts. This patch fixes failure in OpenCL conformance test api/get_kernel_arg_info with amdgcn---amdgizcl triple, where non-zero alloca address space is used. Differential Revision: https://reviews.llvm.org/D34777 llvm-svn: 306721
2017-06-30 02:47:45 +08:00
void FuncOneMember(struct StructOneMember u) {
u.x = (int2)(0, 0);
}
[test] Add {{.*}} to make tests immune to dso_local/dso_preemptable/(none) differences For a definition (of most linkage types), dso_local is set for ELF -fno-pic/-fpie and COFF, but not for Mach-O. This nuance causes unneeded binary format differences. This patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} ` if there is an explicit linkage. * Clang will set dso_local for Mach-O, which is currently implied by TargetMachine.cpp. This will make COFF/Mach-O and executable ELF similar. * Eventually I hope we can make dso_local the textual LLVM IR default (write explicit "dso_preemptable" when applicable) and -fpic ELF will be similar to everything else. This patch helps move toward that goal.
2020-12-31 12:45:56 +08:00
// AMDGCN-LABEL: define{{.*}} void @FuncOneLargeMember(%struct.LargeStructOneMember addrspace(5)* byval(%struct.LargeStructOneMember) align 8 %u)
Recommit r326946 after reducing CallArgList memory footprint llvm-svn: 327634
2018-03-15 23:25:19 +08:00
// AMDGCN-NOT: addrspacecast
// AMDGCN: store <2 x i32> %{{.*}}, <2 x i32> addrspace(5)*
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
void FuncOneLargeMember(struct LargeStructOneMember u) {
u.x[0] = (int2)(0, 0);
}
[test] Add {{.*}} to make tests immune to dso_local/dso_preemptable/(none) differences For a definition (of most linkage types), dso_local is set for ELF -fno-pic/-fpie and COFF, but not for Mach-O. This nuance causes unneeded binary format differences. This patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} ` if there is an explicit linkage. * Clang will set dso_local for Mach-O, which is currently implied by TargetMachine.cpp. This will make COFF/Mach-O and executable ELF similar. * Eventually I hope we can make dso_local the textual LLVM IR default (write explicit "dso_preemptable" when applicable) and -fpic ELF will be similar to everything else. This patch helps move toward that goal.
2020-12-31 12:45:56 +08:00
// AMDGCN20-LABEL: define{{.*}} void @test_indirect_arg_globl()
Recommit r326946 after reducing CallArgList memory footprint llvm-svn: 327634
2018-03-15 23:25:19 +08:00
// AMDGCN20: %[[byval_temp:.*]] = alloca %struct.LargeStructOneMember, align 8, addrspace(5)
// AMDGCN20: %[[r0:.*]] = bitcast %struct.LargeStructOneMember addrspace(5)* %[[byval_temp]] to i8 addrspace(5)*
// AMDGCN20: call void @llvm.memcpy.p5i8.p1i8.i64(i8 addrspace(5)* align 8 %[[r0]], i8 addrspace(1)* align 8 bitcast (%struct.LargeStructOneMember addrspace(1)* @g_s to i8 addrspace(1)*), i64 800, i1 false)
LLVM IR: Generate new-style byval-with-Type from Clang LLVM IR recently added a Type parameter to the byval Attribute, so that when pointers become opaque and no longer have an element type the information will still be present in IR. For now the Type parameter is optional (which is why Clang didn't need this change at the time), but it will become mandatory soon. llvm-svn: 362652
2019-06-06 05:12:14 +08:00
// AMDGCN20: call void @FuncOneLargeMember(%struct.LargeStructOneMember addrspace(5)* byval(%struct.LargeStructOneMember) align 8 %[[byval_temp]])
Recommit r326946 after reducing CallArgList memory footprint llvm-svn: 327634
2018-03-15 23:25:19 +08:00
#if __OPENCL_C_VERSION__ >= 200
void test_indirect_arg_globl(void) {
FuncOneLargeMember(g_s);
}
#endif
[test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences For a default visibility external linkage definition, dso_local is set for ELF -fno-pic/-fpie and COFF and Mach-O. Since default clang -cc1 for ELF is similar to -fpic ("PIC Level" is not set), this nuance causes unneeded binary format differences. To make emitted IR similar, ELF -cc1 -fpic will default to -fno-semantic-interposition, which sets dso_local for default visibility external linkage definitions. To make this flip smooth and enable future (dso_local as definition default), this patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} `.
2020-12-31 16:27:11 +08:00
// AMDGCN-LABEL: define{{.*}} amdgpu_kernel void @test_indirect_arg_local()
Recommit r326946 after reducing CallArgList memory footprint llvm-svn: 327634
2018-03-15 23:25:19 +08:00
// AMDGCN: %[[byval_temp:.*]] = alloca %struct.LargeStructOneMember, align 8, addrspace(5)
// AMDGCN: %[[r0:.*]] = bitcast %struct.LargeStructOneMember addrspace(5)* %[[byval_temp]] to i8 addrspace(5)*
// AMDGCN: call void @llvm.memcpy.p5i8.p3i8.i64(i8 addrspace(5)* align 8 %[[r0]], i8 addrspace(3)* align 8 bitcast (%struct.LargeStructOneMember addrspace(3)* @test_indirect_arg_local.l_s to i8 addrspace(3)*), i64 800, i1 false)
LLVM IR: Generate new-style byval-with-Type from Clang LLVM IR recently added a Type parameter to the byval Attribute, so that when pointers become opaque and no longer have an element type the information will still be present in IR. For now the Type parameter is optional (which is why Clang didn't need this change at the time), but it will become mandatory soon. llvm-svn: 362652
2019-06-06 05:12:14 +08:00
// AMDGCN: call void @FuncOneLargeMember(%struct.LargeStructOneMember addrspace(5)* byval(%struct.LargeStructOneMember) align 8 %[[byval_temp]])
Recommit r326946 after reducing CallArgList memory footprint llvm-svn: 327634
2018-03-15 23:25:19 +08:00
kernel void test_indirect_arg_local(void) {
local struct LargeStructOneMember l_s;
FuncOneLargeMember(l_s);
}
[test] Add {{.*}} to make tests immune to dso_local/dso_preemptable/(none) differences For a definition (of most linkage types), dso_local is set for ELF -fno-pic/-fpie and COFF, but not for Mach-O. This nuance causes unneeded binary format differences. This patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} ` if there is an explicit linkage. * Clang will set dso_local for Mach-O, which is currently implied by TargetMachine.cpp. This will make COFF/Mach-O and executable ELF similar. * Eventually I hope we can make dso_local the textual LLVM IR default (write explicit "dso_preemptable" when applicable) and -fpic ELF will be similar to everything else. This patch helps move toward that goal.
2020-12-31 12:45:56 +08:00
// AMDGCN-LABEL: define{{.*}} void @test_indirect_arg_private()
Recommit r326946 after reducing CallArgList memory footprint llvm-svn: 327634
2018-03-15 23:25:19 +08:00
// AMDGCN: %[[p_s:.*]] = alloca %struct.LargeStructOneMember, align 8, addrspace(5)
// AMDGCN-NOT: @llvm.memcpy
LLVM IR: Generate new-style byval-with-Type from Clang LLVM IR recently added a Type parameter to the byval Attribute, so that when pointers become opaque and no longer have an element type the information will still be present in IR. For now the Type parameter is optional (which is why Clang didn't need this change at the time), but it will become mandatory soon. llvm-svn: 362652
2019-06-06 05:12:14 +08:00
// AMDGCN-NEXT: call void @FuncOneLargeMember(%struct.LargeStructOneMember addrspace(5)* byval(%struct.LargeStructOneMember) align 8 %[[p_s]])
Recommit r326946 after reducing CallArgList memory footprint llvm-svn: 327634
2018-03-15 23:25:19 +08:00
void test_indirect_arg_private(void) {
struct LargeStructOneMember p_s;
FuncOneLargeMember(p_s);
}
[test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences For a default visibility external linkage definition, dso_local is set for ELF -fno-pic/-fpie and COFF and Mach-O. Since default clang -cc1 for ELF is similar to -fpic ("PIC Level" is not set), this nuance causes unneeded binary format differences. To make emitted IR similar, ELF -cc1 -fpic will default to -fno-semantic-interposition, which sets dso_local for default visibility external linkage definitions. To make this flip smooth and enable future (dso_local as definition default), this patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} `.
2020-12-31 16:27:11 +08:00
// AMDGCN-LABEL: define{{.*}} amdgpu_kernel void @KernelOneMember
Clean up AMDGCN tests Differential Revision: https://reviews.llvm.org/D43340 llvm-svn: 325279
2018-02-16 03:12:41 +08:00
// AMDGCN-SAME: (<2 x i32> %[[u_coerce:.*]])
// AMDGCN: %[[u:.*]] = alloca %struct.StructOneMember, align 8, addrspace(5)
// AMDGCN: %[[coerce_dive:.*]] = getelementptr inbounds %struct.StructOneMember, %struct.StructOneMember addrspace(5)* %[[u]], i32 0, i32 0
// AMDGCN: store <2 x i32> %[[u_coerce]], <2 x i32> addrspace(5)* %[[coerce_dive]]
// AMDGCN: call void @FuncOneMember(<2 x i32>
CodeGen: Fix invalid bitcast for coerced function argument Clang assumes coerced function argument is in address space 0, which is not always true and results in invalid bitcasts. This patch fixes failure in OpenCL conformance test api/get_kernel_arg_info with amdgcn---amdgizcl triple, where non-zero alloca address space is used. Differential Revision: https://reviews.llvm.org/D34777 llvm-svn: 306721
2017-06-30 02:47:45 +08:00
kernel void KernelOneMember(struct StructOneMember u) {
FuncOneMember(u);
}
[OpenCL][CodeGen] Fix replacing memcpy with addrspacecast Summary: If a function argument is byval and RV is located in default or alloca address space an optimization of creating addrspacecast instead of memcpy is performed. That is not correct for OpenCL, where that can lead to a situation of address space casting from __private * to __global *. See an example below: ``` typedef struct { int x; } MyStruct; void foo(MyStruct val) {} kernel void KernelOneMember(__global MyStruct* x) { foo (*x); } ``` for this code clang generated following IR: ... %0 = load %struct.MyStruct addrspace(1)*, %struct.MyStruct addrspace(1)** %x.addr, align 4 %1 = addrspacecast %struct.MyStruct addrspace(1)* %0 to %struct.MyStruct* ... So the optimization was disallowed for OpenCL if RV is located in an address space different than that of the argument (0). Reviewers: yaxunl, Anastasia Reviewed By: Anastasia Subscribers: cfe-commits, asavonic Differential Revision: https://reviews.llvm.org/D54947 llvm-svn: 348752
2018-12-10 20:03:00 +08:00
// SPIR: call void @llvm.memcpy.p0i8.p1i8.i32
// SPIR-NOT: addrspacecast
kernel void KernelOneMemberSpir(global struct StructOneMember* u) {
FuncOneMember(*u);
}
[test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences For a default visibility external linkage definition, dso_local is set for ELF -fno-pic/-fpie and COFF and Mach-O. Since default clang -cc1 for ELF is similar to -fpic ("PIC Level" is not set), this nuance causes unneeded binary format differences. To make emitted IR similar, ELF -cc1 -fpic will default to -fno-semantic-interposition, which sets dso_local for default visibility external linkage definitions. To make this flip smooth and enable future (dso_local as definition default), this patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} `.
2020-12-31 16:27:11 +08:00
// AMDGCN-LABEL: define{{.*}} amdgpu_kernel void @KernelLargeOneMember(
Clean up AMDGCN tests Differential Revision: https://reviews.llvm.org/D43340 llvm-svn: 325279
2018-02-16 03:12:41 +08:00
// AMDGCN: %[[U:.*]] = alloca %struct.LargeStructOneMember, align 8, addrspace(5)
// AMDGCN: store %struct.LargeStructOneMember %u.coerce, %struct.LargeStructOneMember addrspace(5)* %[[U]], align 8
LLVM IR: Generate new-style byval-with-Type from Clang LLVM IR recently added a Type parameter to the byval Attribute, so that when pointers become opaque and no longer have an element type the information will still be present in IR. For now the Type parameter is optional (which is why Clang didn't need this change at the time), but it will become mandatory soon. llvm-svn: 362652
2019-06-06 05:12:14 +08:00
// AMDGCN: call void @FuncOneLargeMember(%struct.LargeStructOneMember addrspace(5)* byval(%struct.LargeStructOneMember) align 8 %[[U]])
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
kernel void KernelLargeOneMember(struct LargeStructOneMember u) {
FuncOneLargeMember(u);
}
[test] Add {{.*}} to make tests immune to dso_local/dso_preemptable/(none) differences For a definition (of most linkage types), dso_local is set for ELF -fno-pic/-fpie and COFF, but not for Mach-O. This nuance causes unneeded binary format differences. This patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} ` if there is an explicit linkage. * Clang will set dso_local for Mach-O, which is currently implied by TargetMachine.cpp. This will make COFF/Mach-O and executable ELF similar. * Eventually I hope we can make dso_local the textual LLVM IR default (write explicit "dso_preemptable" when applicable) and -fpic ELF will be similar to everything else. This patch helps move toward that goal.
2020-12-31 12:45:56 +08:00
// AMDGCN-LABEL: define{{.*}} void @FuncTwoMember(<2 x i32> %u.coerce0, <2 x i32> %u.coerce1)
CodeGen: Fix invalid bitcast for coerced function argument Clang assumes coerced function argument is in address space 0, which is not always true and results in invalid bitcasts. This patch fixes failure in OpenCL conformance test api/get_kernel_arg_info with amdgcn---amdgizcl triple, where non-zero alloca address space is used. Differential Revision: https://reviews.llvm.org/D34777 llvm-svn: 306721
2017-06-30 02:47:45 +08:00
void FuncTwoMember(struct StructTwoMember u) {
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
u.y = (int2)(0, 0);
}
[test] Add {{.*}} to make tests immune to dso_local/dso_preemptable/(none) differences For a definition (of most linkage types), dso_local is set for ELF -fno-pic/-fpie and COFF, but not for Mach-O. This nuance causes unneeded binary format differences. This patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} ` if there is an explicit linkage. * Clang will set dso_local for Mach-O, which is currently implied by TargetMachine.cpp. This will make COFF/Mach-O and executable ELF similar. * Eventually I hope we can make dso_local the textual LLVM IR default (write explicit "dso_preemptable" when applicable) and -fpic ELF will be similar to everything else. This patch helps move toward that goal.
2020-12-31 12:45:56 +08:00
// AMDGCN-LABEL: define{{.*}} void @FuncLargeTwoMember(%struct.LargeStructTwoMember addrspace(5)* byval(%struct.LargeStructTwoMember) align 8 %u)
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
void FuncLargeTwoMember(struct LargeStructTwoMember u) {
u.y[0] = (int2)(0, 0);
CodeGen: Fix invalid bitcast for coerced function argument Clang assumes coerced function argument is in address space 0, which is not always true and results in invalid bitcasts. This patch fixes failure in OpenCL conformance test api/get_kernel_arg_info with amdgcn---amdgizcl triple, where non-zero alloca address space is used. Differential Revision: https://reviews.llvm.org/D34777 llvm-svn: 306721
2017-06-30 02:47:45 +08:00
}
[test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences For a default visibility external linkage definition, dso_local is set for ELF -fno-pic/-fpie and COFF and Mach-O. Since default clang -cc1 for ELF is similar to -fpic ("PIC Level" is not set), this nuance causes unneeded binary format differences. To make emitted IR similar, ELF -cc1 -fpic will default to -fno-semantic-interposition, which sets dso_local for default visibility external linkage definitions. To make this flip smooth and enable future (dso_local as definition default), this patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} `.
2020-12-31 16:27:11 +08:00
// AMDGCN-LABEL: define{{.*}} amdgpu_kernel void @KernelTwoMember
Clean up AMDGCN tests Differential Revision: https://reviews.llvm.org/D43340 llvm-svn: 325279
2018-02-16 03:12:41 +08:00
// AMDGCN-SAME: (%struct.StructTwoMember %[[u_coerce:.*]])
// AMDGCN: %[[u:.*]] = alloca %struct.StructTwoMember, align 8, addrspace(5)
// AMDGCN: %[[LD0:.*]] = load <2 x i32>, <2 x i32> addrspace(5)*
// AMDGCN: %[[LD1:.*]] = load <2 x i32>, <2 x i32> addrspace(5)*
// AMDGCN: call void @FuncTwoMember(<2 x i32> %[[LD0]], <2 x i32> %[[LD1]])
CodeGen: Fix invalid bitcast for coerced function argument Clang assumes coerced function argument is in address space 0, which is not always true and results in invalid bitcasts. This patch fixes failure in OpenCL conformance test api/get_kernel_arg_info with amdgcn---amdgizcl triple, where non-zero alloca address space is used. Differential Revision: https://reviews.llvm.org/D34777 llvm-svn: 306721
2017-06-30 02:47:45 +08:00
kernel void KernelTwoMember(struct StructTwoMember u) {
FuncTwoMember(u);
}
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
[test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences For a default visibility external linkage definition, dso_local is set for ELF -fno-pic/-fpie and COFF and Mach-O. Since default clang -cc1 for ELF is similar to -fpic ("PIC Level" is not set), this nuance causes unneeded binary format differences. To make emitted IR similar, ELF -cc1 -fpic will default to -fno-semantic-interposition, which sets dso_local for default visibility external linkage definitions. To make this flip smooth and enable future (dso_local as definition default), this patch replaces (function) `define ` with `define{{.*}} `, (variable/constant/alias) `= ` with `={{.*}} `, or inserts appropriate `{{.*}} `.
2020-12-31 16:27:11 +08:00
// AMDGCN-LABEL: define{{.*}} amdgpu_kernel void @KernelLargeTwoMember
Clean up AMDGCN tests Differential Revision: https://reviews.llvm.org/D43340 llvm-svn: 325279
2018-02-16 03:12:41 +08:00
// AMDGCN-SAME: (%struct.LargeStructTwoMember %[[u_coerce:.*]])
// AMDGCN: %[[u:.*]] = alloca %struct.LargeStructTwoMember, align 8, addrspace(5)
// AMDGCN: store %struct.LargeStructTwoMember %[[u_coerce]], %struct.LargeStructTwoMember addrspace(5)* %[[u]]
LLVM IR: Generate new-style byval-with-Type from Clang LLVM IR recently added a Type parameter to the byval Attribute, so that when pointers become opaque and no longer have an element type the information will still be present in IR. For now the Type parameter is optional (which is why Clang didn't need this change at the time), but it will become mandatory soon. llvm-svn: 362652
2019-06-06 05:12:14 +08:00
// AMDGCN: call void @FuncLargeTwoMember(%struct.LargeStructTwoMember addrspace(5)* byval(%struct.LargeStructTwoMember) align 8 %[[u]])
AMDGPU: Use direct struct returns and arguments This is an improvement over always using byval for structs. This will use registers until ~16 are used, and then switch back to byval. This needs more work, since I'm not sure it ever really makes sense to use byval. If the register limit is exceeded, the arguments still end up passed on the stack, but with a different ABI. It also may make sense to base this on number of registers used for non-struct arguments, rather than just arguments that appear first in the argument list. llvm-svn: 310527
2017-08-10 05:44:58 +08:00
kernel void KernelLargeTwoMember(struct LargeStructTwoMember u) {
FuncLargeTwoMember(u);
}