encounter
/
dawn-cmake
mirror of https://github.com/encounter/dawn-cmake.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
dawn-cmake/test/intrinsics/gen/frexp/b87f4e.wgsl.expected.msl

61 lines
3.9 KiB
Plaintext
Raw Normal View History

test: Generate more permutations for pointers Instead of just generating pointers to functions, generate pointers to all permuted storage types and accesses. Change-Id: I930b20150d823877eaf72dd7cac850078fe22f35 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/54656 Kokoro: Kokoro <noreply+kokoro@google.com> Reviewed-by: James Price <jrprice@google.com> Reviewed-by: David Neto <dneto@google.com>
2021-06-18 18:56:13 +00:00
SKIP: FAILED
[writer/msl]: Implement modf and frexp And remove the u32 overload of frexp (it's not in the spec). Brings the number of failing tint end to end tests for MSL down to 19/1098. The WG still haven't found consensus on reworking these two intrinsics. It's very likely that their signature will change so that they return a structure instead of returning a value and outputing another as a pointer. Until the WG makes a decision, let's implement these according to the current spec. Some overloads are still failing due to MSL missing overloads of the pointer parameter being in the `threadgroup` address space. I'm holding off fixing these until we know what's happening with these intrinsics. See also: https://github.com/gpuweb/gpuweb/issues/1480 https://github.com/gpuweb/gpuweb/issues/1846 Change-Id: Ib6764e6659d840db41bc65fed2b8b283d1056c3d Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/57421 Commit-Queue: Ben Clayton <bclayton@chromium.org> Commit-Queue: Ben Clayton <bclayton@google.com> Kokoro: Ben Clayton <bclayton@google.com> Reviewed-by: James Price <jrprice@google.com>
2021-07-08 21:21:27 +00:00
#include <metal_stdlib>
test: Generate more permutations for pointers Instead of just generating pointers to functions, generate pointers to all permuted storage types and accesses. Change-Id: I930b20150d823877eaf72dd7cac850078fe22f35 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/54656 Kokoro: Kokoro <noreply+kokoro@google.com> Reviewed-by: James Price <jrprice@google.com> Reviewed-by: David Neto <dneto@google.com>
2021-06-18 18:56:13 +00:00
[writer/msl]: Implement modf and frexp And remove the u32 overload of frexp (it's not in the spec). Brings the number of failing tint end to end tests for MSL down to 19/1098. The WG still haven't found consensus on reworking these two intrinsics. It's very likely that their signature will change so that they return a structure instead of returning a value and outputing another as a pointer. Until the WG makes a decision, let's implement these according to the current spec. Some overloads are still failing due to MSL missing overloads of the pointer parameter being in the `threadgroup` address space. I'm holding off fixing these until we know what's happening with these intrinsics. See also: https://github.com/gpuweb/gpuweb/issues/1480 https://github.com/gpuweb/gpuweb/issues/1846 Change-Id: Ib6764e6659d840db41bc65fed2b8b283d1056c3d Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/57421 Commit-Queue: Ben Clayton <bclayton@chromium.org> Commit-Queue: Ben Clayton <bclayton@google.com> Kokoro: Ben Clayton <bclayton@google.com> Reviewed-by: James Price <jrprice@google.com>
2021-07-08 21:21:27 +00:00
using namespace metal;
void frexp_b87f4e(threadgroup int4* const tint_symbol_1) {
float4 res = frexp(float4(), *(&(*(tint_symbol_1))));
test: Generate more permutations for pointers Instead of just generating pointers to functions, generate pointers to all permuted storage types and accesses. Change-Id: I930b20150d823877eaf72dd7cac850078fe22f35 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/54656 Kokoro: Kokoro <noreply+kokoro@google.com> Reviewed-by: James Price <jrprice@google.com> Reviewed-by: David Neto <dneto@google.com>
2021-06-18 18:56:13 +00:00
}
[writer/msl]: Implement modf and frexp And remove the u32 overload of frexp (it's not in the spec). Brings the number of failing tint end to end tests for MSL down to 19/1098. The WG still haven't found consensus on reworking these two intrinsics. It's very likely that their signature will change so that they return a structure instead of returning a value and outputing another as a pointer. Until the WG makes a decision, let's implement these according to the current spec. Some overloads are still failing due to MSL missing overloads of the pointer parameter being in the `threadgroup` address space. I'm holding off fixing these until we know what's happening with these intrinsics. See also: https://github.com/gpuweb/gpuweb/issues/1480 https://github.com/gpuweb/gpuweb/issues/1846 Change-Id: Ib6764e6659d840db41bc65fed2b8b283d1056c3d Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/57421 Commit-Queue: Ben Clayton <bclayton@chromium.org> Commit-Queue: Ben Clayton <bclayton@google.com> Kokoro: Ben Clayton <bclayton@google.com> Reviewed-by: James Price <jrprice@google.com>
2021-07-08 21:21:27 +00:00
kernel void compute_main(uint local_invocation_index [[thread_index_in_threadgroup]]) {
threadgroup int4 tint_symbol_2;
if ((local_invocation_index == 0u)) {
tint_symbol_2 = int4();
}
threadgroup_barrier(mem_flags::mem_threadgroup);
frexp_b87f4e(&(tint_symbol_2));
return;
test: Generate more permutations for pointers Instead of just generating pointers to functions, generate pointers to all permuted storage types and accesses. Change-Id: I930b20150d823877eaf72dd7cac850078fe22f35 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/54656 Kokoro: Kokoro <noreply+kokoro@google.com> Reviewed-by: James Price <jrprice@google.com> Reviewed-by: David Neto <dneto@google.com>
2021-06-18 18:56:13 +00:00
}
[writer/msl]: Implement modf and frexp And remove the u32 overload of frexp (it's not in the spec). Brings the number of failing tint end to end tests for MSL down to 19/1098. The WG still haven't found consensus on reworking these two intrinsics. It's very likely that their signature will change so that they return a structure instead of returning a value and outputing another as a pointer. Until the WG makes a decision, let's implement these according to the current spec. Some overloads are still failing due to MSL missing overloads of the pointer parameter being in the `threadgroup` address space. I'm holding off fixing these until we know what's happening with these intrinsics. See also: https://github.com/gpuweb/gpuweb/issues/1480 https://github.com/gpuweb/gpuweb/issues/1846 Change-Id: Ib6764e6659d840db41bc65fed2b8b283d1056c3d Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/57421 Commit-Queue: Ben Clayton <bclayton@chromium.org> Commit-Queue: Ben Clayton <bclayton@google.com> Kokoro: Ben Clayton <bclayton@google.com> Reviewed-by: James Price <jrprice@google.com>
2021-07-08 21:21:27 +00:00
Compilation failed:
program_source:5:16: error: no matching function for call to 'frexp'
float4 res = frexp(float4(), *(&(*(tint_symbol_1))));
^~~~~
/System/Library/PrivateFrameworks/GPUCompiler.framework/Versions/31001/Libraries/lib/clang/31001.189/include/metal/metal_math:5292:19: note: candidate function not viable: address space mismatch in 2nd argument ('threadgroup int4' (vector of 4 'int' values)), parameter type must be 'metal::int4 &' (aka 'int4 &')
METAL_FUNC float4 frexp(float4 x, thread int4 &exp)
^
/System/Library/PrivateFrameworks/GPUCompiler.framework/Versions/31001/Libraries/lib/clang/31001.189/include/metal/metal_math:3092:17: note: candidate function not viable: no known conversion from 'float4' (vector of 4 'float' values) to 'half' for 1st argument
METAL_FUNC half frexp(half x, thread int &exp)
^
/System/Library/PrivateFrameworks/GPUCompiler.framework/Versions/31001/Libraries/lib/clang/31001.189/include/metal/metal_math:3336:18: note: candidate function not viable: no known conversion from 'float4' (vector of 4 'float' values) to 'metal::half2' (aka 'half2') for 1st argument
METAL_FUNC half2 frexp(half2 x, thread int2 &exp)
^
/System/Library/PrivateFrameworks/GPUCompiler.framework/Versions/31001/Libraries/lib/clang/31001.189/include/metal/metal_math:3580:18: note: candidate function not viable: no known conversion from 'float4' (vector of 4 'float' values) to 'metal::half3' (aka 'half3') for 1st argument
METAL_FUNC half3 frexp(half3 x, thread int3 &exp)
^
/System/Library/PrivateFrameworks/GPUCompiler.framework/Versions/31001/Libraries/lib/clang/31001.189/include/metal/metal_math:3824:18: note: candidate function not viable: no known conversion from 'float4' (vector of 4 'float' values) to 'metal::half4' (aka 'half4') for 1st argument
METAL_FUNC half4 frexp(half4 x, thread int4 &exp)
^
/System/Library/PrivateFrameworks/GPUCompiler.framework/Versions/31001/Libraries/lib/clang/31001.189/include/metal/metal_math:4560:18: note: candidate function not viable: no known conversion from 'float4' (vector of 4 'float' values) to 'float' for 1st argument
METAL_FUNC float frexp(float x, thread int &exp)
^
/System/Library/PrivateFrameworks/GPUCompiler.framework/Versions/31001/Libraries/lib/clang/31001.189/include/metal/metal_math:4804:19: note: candidate function not viable: no known conversion from 'float4' (vector of 4 'float' values) to 'metal::float2' (aka 'float2') for 1st argument
METAL_FUNC float2 frexp(float2 x, thread int2 &exp)
^
/System/Library/PrivateFrameworks/GPUCompiler.framework/Versions/31001/Libraries/lib/clang/31001.189/include/metal/metal_math:5048:19: note: candidate function not viable: no known conversion from 'float4' (vector of 4 'float' values) to 'metal::float3' (aka 'float3') for 1st argument
METAL_FUNC float3 frexp(float3 x, thread int3 &exp)
^
program_source:10:31: warning: equality comparison with extraneous parentheses
if ((local_invocation_index == 0u)) {
~~~~~~~~~~~~~~~~~~~~~~~^~~~~
program_source:10:31: note: remove extraneous parentheses around the comparison to silence this warning
if ((local_invocation_index == 0u)) {
~ ^ ~
program_source:10:31: note: use '=' to turn this equality comparison into an assignment
if ((local_invocation_index == 0u)) {
^~
=
test: Generate more permutations for pointers Instead of just generating pointers to functions, generate pointers to all permuted storage types and accesses. Change-Id: I930b20150d823877eaf72dd7cac850078fe22f35 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/54656 Kokoro: Kokoro <noreply+kokoro@google.com> Reviewed-by: James Price <jrprice@google.com> Reviewed-by: David Neto <dneto@google.com>
2021-06-18 18:56:13 +00:00