Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add WaveGetLane* support for Metal and WGSL #6371

Merged
merged 26 commits into from
Feb 28, 2025
Merged

Add WaveGetLane* support for Metal and WGSL #6371

Show file tree
Hide file tree
Changes from 19 commits
Commits
Show all changes
26 commits
Select commit Hold shift + click to select a range
dd7e4f9
support WaveGetLane* for WGSL and Metal
fairywreath Feb 16, 2025
07bd6b7
update test and glsl support
fairywreath Feb 16, 2025
3f334df
address review comments and fix metal test
fairywreath Feb 16, 2025
fd4d4e3
add missing pragma guard
fairywreath Feb 16, 2025
87410a9
Merge branch 'master' into wgsl-global-param-builtin
fairywreath Feb 17, 2025
f2b97e9
update test
fairywreath Feb 20, 2025
8221661
Merge branch 'master' into wgsl-global-param-builtin
fairywreath Feb 20, 2025
260cccd
Revert "update test"
fairywreath Feb 20, 2025
70e3eb1
update failing glsl metal test and added new test
fairywreath Feb 20, 2025
b670683
Merge branch 'master' into wgsl-global-param-builtin
fairywreath Feb 20, 2025
51a565a
make hlsl and glsl outputs similar
fairywreath Feb 20, 2025
1db6f43
update test
fairywreath Feb 21, 2025
2824396
Merge branch 'master' into wgsl-global-param-builtin
fairywreath Feb 21, 2025
1c42aa1
disable tests for Metal and cleanup
fairywreath Feb 25, 2025
1fcc307
Merge branch 'master' into wgsl-global-param-builtin
fairywreath Feb 25, 2025
94eadbd
comment fix
fairywreath Feb 25, 2025
96a2e7b
add expected failures
fairywreath Feb 25, 2025
397865f
Merge branch 'master' into wgsl-global-param-builtin
fairywreath Feb 25, 2025
4dc0742
Merge branch 'master' into wgsl-global-param-builtin
fairywreath Feb 25, 2025
a8088ea
correct expected failures list
fairywreath Feb 26, 2025
bf11684
Merge branch 'master' into wgsl-global-param-builtin
fairywreath Feb 26, 2025
bb606e3
remove expected failure
fairywreath Feb 27, 2025
51256fa
Merge branch 'master' into wgsl-global-param-builtin
fairywreath Feb 27, 2025
9d6dd3f
add tests to expected failure
fairywreath Feb 27, 2025
a881dbe
Merge branch 'master' into wgsl-global-param-builtin
csyonghe Feb 28, 2025
0ae2841
Merge branch 'master' into wgsl-global-param-builtin
csyonghe Feb 28, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions source/slang/core.meta.slang
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -440,8 +440,8 @@ attribute_syntax [Differentiable(order:int = 0)] : BackwardDifferentiableAttribu
__intrinsic_op($(kIROp_RequirePrelude))
void __requirePrelude(constexpr String preludeText);

__intrinsic_op($(kIROp_RequireGLSLExtension))
void __requireGLSLExtension(constexpr String preludeText);
__intrinsic_op($(kIROp_RequireTargetExtension))
void __requireTargetExtension(constexpr String preludeText);

/// @experimetal
/// Perform a compile-time condition check and emit a compile-time error if the condition is false.
Expand Down
56 changes: 29 additions & 27 deletions source/slang/glsl.meta.slang
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4276,7 +4276,7 @@ __generic<T : __BuiltinType>
case glsl:
{
if (__type_equals<T, float>())
__requireGLSLExtension("GL_EXT_shader_atomic_float");
__requireTargetExtension("GL_EXT_shader_atomic_float");
}
case spirv:
if (__type_equals<T, float>())
Expand All @@ -4298,7 +4298,7 @@ __generic<T : __BuiltinType>
case glsl:
{
if (__type_equals<T, float>())
__requireGLSLExtension("GL_EXT_shader_atomic_float2");
__requireTargetExtension("GL_EXT_shader_atomic_float2");
}
case spirv:
if (__type_equals<T, float>())
Expand Down Expand Up @@ -4738,7 +4738,7 @@ void requireGLSLExtForRayTracingBuiltin()
__target_switch
{
case glsl:
__requireGLSLExtension("GL_EXT_ray_tracing");
__requireTargetExtension("GL_EXT_ray_tracing");
__intrinsic_asm "";
default:
return;
Expand Down Expand Up @@ -6284,30 +6284,30 @@ public void traceRayMotionNV(
__generic<T : __BuiltinType>
[ForceInline]
void typeRequireChecks_shader_subgroup_GLSL() {
// the following is a seperate function call, since else the `__requireGLSLExtension` and associated __intrinsic_asm is ignored if the calling function also calls an __intrinsic_asm
// the following is a seperate function call, since else the `__requireTargetExtension` and associated __intrinsic_asm is ignored if the calling function also calls an __intrinsic_asm
__target_switch
{
case glsl:
if (__type_equals<T, half>()
|| __type_equals<T, float16_t>()
) __requireGLSLExtension("GL_EXT_shader_subgroup_extended_types_float16");
) __requireTargetExtension("GL_EXT_shader_subgroup_extended_types_float16");
else if (__type_equals<T, uint8_t>()
|| __type_equals<T, int8_t>()
) __requireGLSLExtension("GL_EXT_shader_subgroup_extended_types_int8");
) __requireTargetExtension("GL_EXT_shader_subgroup_extended_types_int8");
else if (__type_equals<T, uint16_t>()
|| __type_equals<T, int16_t>()
) __requireGLSLExtension("GL_EXT_shader_subgroup_extended_types_int16");
) __requireTargetExtension("GL_EXT_shader_subgroup_extended_types_int16");
else if (__type_equals<T, uint64_t>()
|| __type_equals<T, int64_t>()
) __requireGLSLExtension("GL_EXT_shader_subgroup_extended_types_int64");
) __requireTargetExtension("GL_EXT_shader_subgroup_extended_types_int64");

__intrinsic_asm "";
}
}

__generic<T : __BuiltinType>
void shader_subgroup_preamble() {
// checks needed for shader_subgroup functions; __requireGLSLExtension does not work
// checks needed for shader_subgroup functions; __requireTargetExtension does not work
// (does not add the ext specified correctly to the compile output; using extended type
// will result in error for using the type)
__target_switch
Expand All @@ -6327,14 +6327,14 @@ void requireGLSLExtForSubgroupBasicBuiltin() {
__target_switch
{
case glsl:
__requireGLSLExtension("GL_KHR_shader_subgroup_basic");
__requireTargetExtension("GL_KHR_shader_subgroup_basic");
__intrinsic_asm "";
default:
return;
}
}

[require(cpp_cuda_glsl_hlsl_spirv_wgsl, subgroup_basic)]
[require(cpp_cuda_glsl_hlsl_metal_spirv_wgsl, subgroup_basic)]
void setupExtForSubgroupBasicBuiltIn() {
__target_switch
{
Expand All @@ -6351,7 +6351,7 @@ void requireGLSLExtForSubgroupBallotBuiltin() {
__target_switch
{
case glsl:
__requireGLSLExtension("GL_KHR_shader_subgroup_ballot");
__requireTargetExtension("GL_KHR_shader_subgroup_ballot");
__intrinsic_asm "";
default:
return;
Expand Down Expand Up @@ -6409,7 +6409,8 @@ public property uint gl_SubgroupID

public property uint gl_SubgroupSize
{
[require(cpp_cuda_glsl_hlsl_spirv_wgsl, subgroup_basic)]
[ForceInline]
[require(cpp_cuda_glsl_hlsl_metal_spirv_wgsl, subgroup_basic)]
get {
setupExtForSubgroupBasicBuiltIn();
return WaveGetLaneCount();
Expand All @@ -6418,7 +6419,8 @@ public property uint gl_SubgroupSize

public property uint gl_SubgroupInvocationID
{
[require(cpp_cuda_glsl_hlsl_spirv_wgsl, subgroup_basic)]
[ForceInline]
[require(cpp_cuda_glsl_hlsl_metal_spirv_wgsl, subgroup_basic)]
get {
setupExtForSubgroupBasicBuiltIn();
return WaveGetLaneIndex();
Expand Down Expand Up @@ -8368,8 +8370,8 @@ void typeRequireChecks_atomic_using_float0_tier()
{
case glsl:
{
if (__type_equals<T, uint64_t>() || __type_equals<T, int64_t>())
__requireGLSLExtension("GL_EXT_shader_atomic_int64");
if (__type_equals<T, uint64_t>() || __type_equals<T, int64_t>())
__requireTargetExtension("GL_EXT_shader_atomic_int64");
}
case spirv:
return;
Expand All @@ -8385,16 +8387,16 @@ void typeRequireChecks_atomic_using_float1_tier()
case glsl:
{
if (__type_equals<T, float>())
__requireGLSLExtension("GL_EXT_shader_atomic_float");
__requireTargetExtension("GL_EXT_shader_atomic_float");
else if (__type_equals<T, half>() || __type_equals<T, float16_t>())
{
__requireGLSLExtension("GL_EXT_shader_atomic_float2");
__requireGLSLExtension("GL_EXT_shader_explicit_arithmetic_types");
__requireTargetExtension("GL_EXT_shader_atomic_float2");
__requireTargetExtension("GL_EXT_shader_explicit_arithmetic_types");
}
else if (__type_equals<T, double>())
__requireGLSLExtension("GL_EXT_shader_atomic_float");
__requireTargetExtension("GL_EXT_shader_atomic_float");
else if (__type_equals<T, uint64_t>() || __type_equals<T, int64_t>())
__requireGLSLExtension("GL_EXT_shader_atomic_int64");
__requireTargetExtension("GL_EXT_shader_atomic_int64");
}
case spirv:
return;
Expand All @@ -8410,16 +8412,16 @@ void typeRequireChecks_atomic_using_float2_tier()
case glsl:
{
if (__type_equals<T, float>())
__requireGLSLExtension("GL_EXT_shader_atomic_float2");
__requireTargetExtension("GL_EXT_shader_atomic_float2");
else if (__type_equals<T, half>() || __type_equals<T, float16_t>())
{
__requireGLSLExtension("GL_EXT_shader_atomic_float2");
__requireGLSLExtension("GL_EXT_shader_explicit_arithmetic_types");
__requireTargetExtension("GL_EXT_shader_atomic_float2");
__requireTargetExtension("GL_EXT_shader_explicit_arithmetic_types");
}
else if (__type_equals<T, double>())
__requireGLSLExtension("GL_EXT_shader_atomic_float2");
else if (__type_equals<T, uint64_t>() || __type_equals<T, int64_t>())
__requireGLSLExtension("GL_EXT_shader_atomic_int64");
__requireTargetExtension("GL_EXT_shader_atomic_float2");
else if (__type_equals<T, uint64_t>() || __type_equals<T, int64_t>())
__requireTargetExtension("GL_EXT_shader_atomic_int64");
}
case spirv:
return;
Expand Down
Loading
Loading