AuroraMiddleware/glslang
1
0
Fork 0
mirror of https://github.com/KhronosGroup/glslang synced 2024-11-08 19:40:06 +00:00
Code Issues Packages Projects Releases Wiki Activity
2b5ea9f851
glslang/Test/baseResults/hlsl.array.flatten.frag.out

593 lines
30 KiB
Plaintext
Raw Normal View History

Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
hlsl.array.flatten.frag
HLSL: Remove workarounds for assigning to opaques. This assumes it will be combined with optimizing transforms that eliminate assignments to opaques.
2017-09-07 10:33:11 +00:00
WARNING: AST will form illegal SPIR-V; need to transform to legalize
HLSL: Emit the OpSource HLSL instruction for HLSL, using new headers.
2017-04-07 21:33:08 +00:00
Shader version: 500
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
gl_FragCoord origin is upper left
0:? Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:17 Function Definition: TestFn1( ( temp 4-component vector of float)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:17 Function Parameters:
0:? Sequence
0:18 Branch: Return with expression
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:18 texture ( temp 4-component vector of float)
0:18 Construct combined texture-sampler ( temp sampler1D)
0:? 'g_tex[1]' ( uniform texture1D)
0:? 'g_samp[1]' ( uniform sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:18 Constant:
0:18 0.200000
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:22 Function Definition: TestFn2(t11[3];p1[3]; ( temp 4-component vector of float)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:22 Function Parameters:
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:22 'l_tex' ( in 3-element array of texture1D)
0:22 'l_samp' ( in 3-element array of sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:? Sequence
0:23 Branch: Return with expression
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:23 texture ( temp 4-component vector of float)
0:23 Construct combined texture-sampler ( temp sampler1D)
0:23 direct index ( temp texture1D)
0:23 'l_tex' ( in 3-element array of texture1D)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:23 Constant:
0:23 2 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:23 direct index ( temp sampler)
0:23 'l_samp' ( in 3-element array of sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:23 Constant:
0:23 2 (const int)
0:23 Constant:
0:23 0.200000
0:26 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:26 move second child to first child ( temp 5-element array of int)
0:26 'not_flattened_a' ( global 5-element array of int)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:26 Constant:
0:26 1 (const int)
0:26 2 (const int)
0:26 3 (const int)
0:26 4 (const int)
0:26 5 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:31 Function Definition: @main(struct-PS_OUTPUT-vf41; ( temp void)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:31 Function Parameters:
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:31 'ps_output' ( out structure{ temp 4-component vector of float color})
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:? Sequence
0:33 Sequence
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:33 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:33 move second child to first child ( temp sampler)
0:33 direct index ( temp sampler)
0:33 'local_sampler_array' ( temp 3-element array of sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:33 Constant:
0:33 0 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[0]' ( uniform sampler)
0:33 move second child to first child ( temp sampler)
0:33 direct index ( temp sampler)
0:33 'local_sampler_array' ( temp 3-element array of sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:33 Constant:
0:33 1 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[1]' ( uniform sampler)
0:33 move second child to first child ( temp sampler)
0:33 direct index ( temp sampler)
0:33 'local_sampler_array' ( temp 3-element array of sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:33 Constant:
0:33 2 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[2]' ( uniform sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:34 Sequence
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:34 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:34 move second child to first child ( temp texture1D)
0:34 direct index ( temp texture1D)
0:34 'local_texture_array' ( temp 3-element array of texture1D)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:34 Constant:
0:34 0 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[0]' ( uniform texture1D)
0:34 move second child to first child ( temp texture1D)
0:34 direct index ( temp texture1D)
0:34 'local_texture_array' ( temp 3-element array of texture1D)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:34 Constant:
0:34 1 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[1]' ( uniform texture1D)
0:34 move second child to first child ( temp texture1D)
0:34 direct index ( temp texture1D)
0:34 'local_texture_array' ( temp 3-element array of texture1D)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:34 Constant:
0:34 2 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[2]' ( uniform texture1D)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:35 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:35 move second child to first child ( temp 4-element array of float)
0:35 'local_float_array' ( temp 4-element array of float)
HLSL: Allow use of $Global members in between function calls. This allows global initializers to use $Global members.
2017-03-23 05:21:34 +00:00
0:35 g_floats: direct index for structure ( uniform 4-element array of float)
0:35 'anon@0' (layout( row_major std140) uniform block{ uniform 4-element array of 3X3 matrix of float g_mats, layout( binding=10) uniform 4-element array of 3X3 matrix of float g_mats_explicit, uniform 4-element array of float g_floats})
HLSL Tests: Tests for previous commit, to make it easier to see what's changing.
2016-09-28 05:09:32 +00:00
0:35 Constant:
0:35 2 (const uint)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:37 move second child to first child ( temp 4-component vector of float)
0:37 color: direct index for structure ( temp 4-component vector of float)
0:37 'ps_output' ( out structure{ temp 4-component vector of float color})
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
0:37 Constant:
0:37 0 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:37 add ( temp 4-component vector of float)
0:37 Function Call: TestFn1( ( temp 4-component vector of float)
0:37 Function Call: TestFn2(t11[3];p1[3]; ( temp 4-component vector of float)
0:37 Comma ( temp 3-element array of texture1D)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:37 move second child to first child ( temp texture1D)
0:37 direct index ( temp texture1D)
0:37 'aggShadow' ( temp 3-element array of texture1D)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 0 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[0]' ( uniform texture1D)
0:37 move second child to first child ( temp texture1D)
0:37 direct index ( temp texture1D)
0:37 'aggShadow' ( temp 3-element array of texture1D)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 1 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[1]' ( uniform texture1D)
0:37 move second child to first child ( temp texture1D)
0:37 direct index ( temp texture1D)
0:37 'aggShadow' ( temp 3-element array of texture1D)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 2 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[2]' ( uniform texture1D)
0:37 'aggShadow' ( temp 3-element array of texture1D)
0:37 Comma ( temp 3-element array of sampler)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:37 move second child to first child ( temp sampler)
0:37 direct index ( temp sampler)
0:37 'aggShadow' ( temp 3-element array of sampler)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 0 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[0]' ( uniform sampler)
0:37 move second child to first child ( temp sampler)
0:37 direct index ( temp sampler)
0:37 'aggShadow' ( temp 3-element array of sampler)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 1 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[1]' ( uniform sampler)
0:37 move second child to first child ( temp sampler)
0:37 direct index ( temp sampler)
0:37 'aggShadow' ( temp 3-element array of sampler)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 2 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[2]' ( uniform sampler)
0:37 'aggShadow' ( temp 3-element array of sampler)
0:31 Function Definition: main( ( temp void)
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
0:31 Function Parameters:
0:? Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:31 Function Call: @main(struct-PS_OUTPUT-vf41; ( temp void)
0:? 'ps_output' ( temp structure{ temp 4-component vector of float color})
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
0:31 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:31 move second child to first child ( temp 4-component vector of float)
HLSL: Move debug naming to a simpler, more consistent, scheme. This will help in expanding flattening and reducing splitting.
2017-08-09 20:29:29 +00:00
0:? 'ps_output.color' (layout( location=0) out 4-component vector of float)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:31 color: direct index for structure ( temp 4-component vector of float)
0:? 'ps_output' ( temp structure{ temp 4-component vector of float color})
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
0:31 Constant:
0:31 0 (const int)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:? Linker Objects
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[0]' ( uniform texture1D)
0:? 'g_tex[1]' ( uniform texture1D)
0:? 'g_tex[2]' ( uniform texture1D)
0:? 'g_tex_explicit[0]' (layout( binding=1) uniform texture1D)
0:? 'g_tex_explicit[1]' (layout( binding=2) uniform texture1D)
0:? 'g_tex_explicit[2]' (layout( binding=3) uniform texture1D)
0:? 'g_samp[0]' ( uniform sampler)
0:? 'g_samp[1]' ( uniform sampler)
0:? 'g_samp[2]' ( uniform sampler)
0:? 'g_samp_explicit[0]' (layout( binding=5) uniform sampler)
0:? 'g_samp_explicit[1]' (layout( binding=6) uniform sampler)
0:? 'g_samp_explicit[2]' (layout( binding=7) uniform sampler)
HLSL: Allow use of $Global members in between function calls. This allows global initializers to use $Global members.
2017-03-23 05:21:34 +00:00
0:? 'anon@0' (layout( row_major std140) uniform block{ uniform 4-element array of 3X3 matrix of float g_mats, layout( binding=10) uniform 4-element array of 3X3 matrix of float g_mats_explicit, uniform 4-element array of float g_floats})
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'not_flattened_a' ( global 5-element array of int)
HLSL: Move debug naming to a simpler, more consistent, scheme. This will help in expanding flattening and reducing splitting.
2017-08-09 20:29:29 +00:00
0:? 'ps_output.color' (layout( location=0) out 4-component vector of float)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
Linked fragment stage:
HLSL: Emit the OpSource HLSL instruction for HLSL, using new headers.
2017-04-07 21:33:08 +00:00
Shader version: 500
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
gl_FragCoord origin is upper left
0:? Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:17 Function Definition: TestFn1( ( temp 4-component vector of float)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:17 Function Parameters:
0:? Sequence
0:18 Branch: Return with expression
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:18 texture ( temp 4-component vector of float)
0:18 Construct combined texture-sampler ( temp sampler1D)
0:? 'g_tex[1]' ( uniform texture1D)
0:? 'g_samp[1]' ( uniform sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:18 Constant:
0:18 0.200000
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:22 Function Definition: TestFn2(t11[3];p1[3]; ( temp 4-component vector of float)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:22 Function Parameters:
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:22 'l_tex' ( in 3-element array of texture1D)
0:22 'l_samp' ( in 3-element array of sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:? Sequence
0:23 Branch: Return with expression
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:23 texture ( temp 4-component vector of float)
0:23 Construct combined texture-sampler ( temp sampler1D)
0:23 direct index ( temp texture1D)
0:23 'l_tex' ( in 3-element array of texture1D)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:23 Constant:
0:23 2 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:23 direct index ( temp sampler)
0:23 'l_samp' ( in 3-element array of sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:23 Constant:
0:23 2 (const int)
0:23 Constant:
0:23 0.200000
0:26 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:26 move second child to first child ( temp 5-element array of int)
0:26 'not_flattened_a' ( global 5-element array of int)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:26 Constant:
0:26 1 (const int)
0:26 2 (const int)
0:26 3 (const int)
0:26 4 (const int)
0:26 5 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:31 Function Definition: @main(struct-PS_OUTPUT-vf41; ( temp void)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:31 Function Parameters:
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:31 'ps_output' ( out structure{ temp 4-component vector of float color})
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:? Sequence
0:33 Sequence
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:33 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:33 move second child to first child ( temp sampler)
0:33 direct index ( temp sampler)
0:33 'local_sampler_array' ( temp 3-element array of sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:33 Constant:
0:33 0 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[0]' ( uniform sampler)
0:33 move second child to first child ( temp sampler)
0:33 direct index ( temp sampler)
0:33 'local_sampler_array' ( temp 3-element array of sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:33 Constant:
0:33 1 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[1]' ( uniform sampler)
0:33 move second child to first child ( temp sampler)
0:33 direct index ( temp sampler)
0:33 'local_sampler_array' ( temp 3-element array of sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:33 Constant:
0:33 2 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[2]' ( uniform sampler)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:34 Sequence
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:34 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:34 move second child to first child ( temp texture1D)
0:34 direct index ( temp texture1D)
0:34 'local_texture_array' ( temp 3-element array of texture1D)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:34 Constant:
0:34 0 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[0]' ( uniform texture1D)
0:34 move second child to first child ( temp texture1D)
0:34 direct index ( temp texture1D)
0:34 'local_texture_array' ( temp 3-element array of texture1D)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:34 Constant:
0:34 1 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[1]' ( uniform texture1D)
0:34 move second child to first child ( temp texture1D)
0:34 direct index ( temp texture1D)
0:34 'local_texture_array' ( temp 3-element array of texture1D)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:34 Constant:
0:34 2 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[2]' ( uniform texture1D)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:35 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:35 move second child to first child ( temp 4-element array of float)
0:35 'local_float_array' ( temp 4-element array of float)
HLSL: Allow use of $Global members in between function calls. This allows global initializers to use $Global members.
2017-03-23 05:21:34 +00:00
0:35 g_floats: direct index for structure ( uniform 4-element array of float)
0:35 'anon@0' (layout( row_major std140) uniform block{ uniform 4-element array of 3X3 matrix of float g_mats, layout( binding=10) uniform 4-element array of 3X3 matrix of float g_mats_explicit, uniform 4-element array of float g_floats})
HLSL Tests: Tests for previous commit, to make it easier to see what's changing.
2016-09-28 05:09:32 +00:00
0:35 Constant:
0:35 2 (const uint)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:37 move second child to first child ( temp 4-component vector of float)
0:37 color: direct index for structure ( temp 4-component vector of float)
0:37 'ps_output' ( out structure{ temp 4-component vector of float color})
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
0:37 Constant:
0:37 0 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:37 add ( temp 4-component vector of float)
0:37 Function Call: TestFn1( ( temp 4-component vector of float)
0:37 Function Call: TestFn2(t11[3];p1[3]; ( temp 4-component vector of float)
0:37 Comma ( temp 3-element array of texture1D)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:37 move second child to first child ( temp texture1D)
0:37 direct index ( temp texture1D)
0:37 'aggShadow' ( temp 3-element array of texture1D)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 0 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[0]' ( uniform texture1D)
0:37 move second child to first child ( temp texture1D)
0:37 direct index ( temp texture1D)
0:37 'aggShadow' ( temp 3-element array of texture1D)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 1 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[1]' ( uniform texture1D)
0:37 move second child to first child ( temp texture1D)
0:37 direct index ( temp texture1D)
0:37 'aggShadow' ( temp 3-element array of texture1D)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 2 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[2]' ( uniform texture1D)
0:37 'aggShadow' ( temp 3-element array of texture1D)
0:37 Comma ( temp 3-element array of sampler)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:37 move second child to first child ( temp sampler)
0:37 direct index ( temp sampler)
0:37 'aggShadow' ( temp 3-element array of sampler)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 0 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[0]' ( uniform sampler)
0:37 move second child to first child ( temp sampler)
0:37 direct index ( temp sampler)
0:37 'aggShadow' ( temp 3-element array of sampler)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 1 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[1]' ( uniform sampler)
0:37 move second child to first child ( temp sampler)
0:37 direct index ( temp sampler)
0:37 'aggShadow' ( temp 3-element array of sampler)
HLSL: Line numbers only: Set locations (line numbers) on synthesized flattening code.
2016-10-01 23:17:55 +00:00
0:37 Constant:
0:37 2 (const int)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_samp[2]' ( uniform sampler)
0:37 'aggShadow' ( temp 3-element array of sampler)
0:31 Function Definition: main( ( temp void)
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
0:31 Function Parameters:
0:? Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:31 Function Call: @main(struct-PS_OUTPUT-vf41; ( temp void)
0:? 'ps_output' ( temp structure{ temp 4-component vector of float color})
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
0:31 Sequence
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:31 move second child to first child ( temp 4-component vector of float)
HLSL: Move debug naming to a simpler, more consistent, scheme. This will help in expanding flattening and reducing splitting.
2017-08-09 20:29:29 +00:00
0:? 'ps_output.color' (layout( location=0) out 4-component vector of float)
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:31 color: direct index for structure ( temp 4-component vector of float)
0:? 'ps_output' ( temp structure{ temp 4-component vector of float color})
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
0:31 Constant:
0:31 0 (const int)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
0:? Linker Objects
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'g_tex[0]' ( uniform texture1D)
0:? 'g_tex[1]' ( uniform texture1D)
0:? 'g_tex[2]' ( uniform texture1D)
0:? 'g_tex_explicit[0]' (layout( binding=1) uniform texture1D)
0:? 'g_tex_explicit[1]' (layout( binding=2) uniform texture1D)
0:? 'g_tex_explicit[2]' (layout( binding=3) uniform texture1D)
0:? 'g_samp[0]' ( uniform sampler)
0:? 'g_samp[1]' ( uniform sampler)
0:? 'g_samp[2]' ( uniform sampler)
0:? 'g_samp_explicit[0]' (layout( binding=5) uniform sampler)
0:? 'g_samp_explicit[1]' (layout( binding=6) uniform sampler)
0:? 'g_samp_explicit[2]' (layout( binding=7) uniform sampler)
HLSL: Allow use of $Global members in between function calls. This allows global initializers to use $Global members.
2017-03-23 05:21:34 +00:00
0:? 'anon@0' (layout( row_major std140) uniform block{ uniform 4-element array of 3X3 matrix of float g_mats, layout( binding=10) uniform 4-element array of 3X3 matrix of float g_mats_explicit, uniform 4-element array of float g_floats})
GLSL output: Removed fixed-size buffer; fixes #769. Makes some white-space differences in most output, plus a few cases where more could have been put out but was cut short by the previous fix-sized buffer.
2017-03-15 01:43:13 +00:00
0:? 'not_flattened_a' ( global 5-element array of int)
HLSL: Move debug naming to a simpler, more consistent, scheme. This will help in expanding flattening and reducing splitting.
2017-08-09 20:29:29 +00:00
0:? 'ps_output.color' (layout( location=0) out 4-component vector of float)
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
SPV Version: Emit the requested SPIR-V version, not the header version. Fixes #1236.
2018-02-01 01:35:56 +00:00
// Module Version 10000
SPV: Bump the generator number to account for barrier changes.
2017-12-11 15:24:06 +00:00
// Generated by (magic number): 80003
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
// Id's are bound by 143
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
Capability Shader
Capability Sampled1D
1: ExtInstImport "GLSL.std.450"
MemoryModel Logical GLSL450
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
EntryPoint Fragment 4 "main" 134
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
ExecutionMode 4 OriginUpperLeft
HLSL: Emit the OpSource HLSL instruction for HLSL, using new headers.
2017-04-07 21:33:08 +00:00
Source HLSL 500
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
Name 4 "main"
Name 9 "TestFn1("
Name 22 "TestFn2(t11[3];p1[3];"
Name 20 "l_tex"
Name 21 "l_samp"
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Name 24 "PS_OUTPUT"
MemberName 24(PS_OUTPUT) 0 "color"
Name 28 "@main(struct-PS_OUTPUT-vf41;"
Name 27 "ps_output"
Name 34 "not_flattened_a"
Name 42 "g_tex[1]"
Name 45 "g_samp[1]"
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
Name 63 "local_sampler_array"
Name 65 "g_samp[0]"
Name 70 "g_samp[2]"
Name 73 "local_texture_array"
Name 74 "g_tex[0]"
Name 79 "g_tex[2]"
Name 85 "local_float_array"
Name 91 "$Global"
MemberName 91($Global) 0 "g_mats"
MemberName 91($Global) 1 "g_mats_explicit"
MemberName 91($Global) 2 "g_floats"
Name 93 ""
Name 107 "aggShadow"
Name 114 "aggShadow"
Name 121 "param"
Name 123 "param"
Name 129 "ps_output"
Name 130 "param"
Name 134 "ps_output.color"
Name 137 "g_tex_explicit[0]"
Name 138 "g_tex_explicit[1]"
Name 139 "g_tex_explicit[2]"
Name 140 "g_samp_explicit[0]"
Name 141 "g_samp_explicit[1]"
Name 142 "g_samp_explicit[2]"
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Decorate 42(g_tex[1]) DescriptorSet 0
Decorate 45(g_samp[1]) DescriptorSet 0
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
Decorate 65(g_samp[0]) DescriptorSet 0
Decorate 70(g_samp[2]) DescriptorSet 0
Decorate 74(g_tex[0]) DescriptorSet 0
Decorate 79(g_tex[2]) DescriptorSet 0
Decorate 88 ArrayStride 48
Decorate 89 ArrayStride 48
Decorate 90 ArrayStride 16
MemberDecorate 91($Global) 0 RowMajor
MemberDecorate 91($Global) 0 Offset 0
MemberDecorate 91($Global) 0 MatrixStride 16
MemberDecorate 91($Global) 1 RowMajor
MemberDecorate 91($Global) 1 Offset 192
MemberDecorate 91($Global) 1 MatrixStride 16
MemberDecorate 91($Global) 2 Offset 384
Decorate 91($Global) Block
Decorate 93 DescriptorSet 0
Decorate 134(ps_output.color) Location 0
Decorate 137(g_tex_explicit[0]) DescriptorSet 0
Decorate 137(g_tex_explicit[0]) Binding 1
Decorate 138(g_tex_explicit[1]) DescriptorSet 0
Decorate 138(g_tex_explicit[1]) Binding 2
Decorate 139(g_tex_explicit[2]) DescriptorSet 0
Decorate 139(g_tex_explicit[2]) Binding 3
Decorate 140(g_samp_explicit[0]) DescriptorSet 0
Decorate 140(g_samp_explicit[0]) Binding 5
Decorate 141(g_samp_explicit[1]) DescriptorSet 0
Decorate 141(g_samp_explicit[1]) Binding 6
Decorate 142(g_samp_explicit[2]) DescriptorSet 0
Decorate 142(g_samp_explicit[2]) Binding 7
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
2: TypeVoid
3: TypeFunction 2
6: TypeFloat 32
7: TypeVector 6(float) 4
8: TypeFunction 7(fvec4)
11: TypeImage 6(float) 1D sampled format:Unknown
12: TypeInt 32 0
13: 12(int) Constant 3
14: TypeArray 11 13
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
15: TypePointer Function 14
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
16: TypeSampler
17: TypeArray 16 13
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
18: TypePointer Function 17
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
19: TypeFunction 7(fvec4) 15(ptr) 18(ptr)
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
24(PS_OUTPUT): TypeStruct 7(fvec4)
25: TypePointer Function 24(PS_OUTPUT)
26: TypeFunction 2 25(ptr)
30: TypeInt 32 1
31: 12(int) Constant 5
32: TypeArray 30(int) 31
33: TypePointer Private 32
34(not_flattened_a): 33(ptr) Variable Private
35: 30(int) Constant 1
36: 30(int) Constant 2
37: 30(int) Constant 3
38: 30(int) Constant 4
39: 30(int) Constant 5
40: 32 ConstantComposite 35 36 37 38 39
41: TypePointer UniformConstant 11
42(g_tex[1]): 41(ptr) Variable UniformConstant
44: TypePointer UniformConstant 16
45(g_samp[1]): 44(ptr) Variable UniformConstant
47: TypeSampledImage 11
49: 6(float) Constant 1045220557
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
53: TypePointer Function 11
56: TypePointer Function 16
64: 30(int) Constant 0
65(g_samp[0]): 44(ptr) Variable UniformConstant
70(g_samp[2]): 44(ptr) Variable UniformConstant
74(g_tex[0]): 41(ptr) Variable UniformConstant
79(g_tex[2]): 41(ptr) Variable UniformConstant
82: 12(int) Constant 4
83: TypeArray 6(float) 82
84: TypePointer Function 83
86: TypeVector 6(float) 3
87: TypeMatrix 86(fvec3) 3
88: TypeArray 87 82
89: TypeArray 87 82
90: TypeArray 6(float) 82
91($Global): TypeStruct 88 89 90
92: TypePointer Uniform 91($Global)
93: 92(ptr) Variable Uniform
94: TypePointer Uniform 90
98: TypePointer Function 6(float)
127: TypePointer Function 7(fvec4)
133: TypePointer Output 7(fvec4)
134(ps_output.color): 133(ptr) Variable Output
137(g_tex_explicit[0]): 41(ptr) Variable UniformConstant
138(g_tex_explicit[1]): 41(ptr) Variable UniformConstant
139(g_tex_explicit[2]): 41(ptr) Variable UniformConstant
140(g_samp_explicit[0]): 44(ptr) Variable UniformConstant
141(g_samp_explicit[1]): 44(ptr) Variable UniformConstant
142(g_samp_explicit[2]): 44(ptr) Variable UniformConstant
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
4(main): 2 Function None 3
5: Label
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
129(ps_output): 25(ptr) Variable Function
130(param): 25(ptr) Variable Function
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Store 34(not_flattened_a) 40
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
131: 2 FunctionCall 28(@main(struct-PS_OUTPUT-vf41;) 130(param)
132:24(PS_OUTPUT) Load 130(param)
Store 129(ps_output) 132
135: 127(ptr) AccessChain 129(ps_output) 64
136: 7(fvec4) Load 135
Store 134(ps_output.color) 136
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
Return
FunctionEnd
9(TestFn1(): 7(fvec4) Function None 8
10: Label
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
43: 11 Load 42(g_tex[1])
46: 16 Load 45(g_samp[1])
48: 47 SampledImage 43 46
50: 7(fvec4) ImageSampleImplicitLod 48 49
ReturnValue 50
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
FunctionEnd
22(TestFn2(t11[3];p1[3];): 7(fvec4) Function None 19
20(l_tex): 15(ptr) FunctionParameter
21(l_samp): 18(ptr) FunctionParameter
23: Label
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
54: 53(ptr) AccessChain 20(l_tex) 36
55: 11 Load 54
57: 56(ptr) AccessChain 21(l_samp) 36
58: 16 Load 57
59: 47 SampledImage 55 58
60: 7(fvec4) ImageSampleImplicitLod 59 49
ReturnValue 60
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
FunctionEnd
28(@main(struct-PS_OUTPUT-vf41;): 2 Function None 26
27(ps_output): 25(ptr) FunctionParameter
29: Label
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
63(local_sampler_array): 18(ptr) Variable Function
73(local_texture_array): 15(ptr) Variable Function
85(local_float_array): 84(ptr) Variable Function
107(aggShadow): 15(ptr) Variable Function
114(aggShadow): 18(ptr) Variable Function
121(param): 15(ptr) Variable Function
123(param): 18(ptr) Variable Function
66: 16 Load 65(g_samp[0])
67: 56(ptr) AccessChain 63(local_sampler_array) 64
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Store 67 66
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
68: 16 Load 45(g_samp[1])
69: 56(ptr) AccessChain 63(local_sampler_array) 35
Store 69 68
71: 16 Load 70(g_samp[2])
72: 56(ptr) AccessChain 63(local_sampler_array) 36
Store 72 71
75: 11 Load 74(g_tex[0])
76: 53(ptr) AccessChain 73(local_texture_array) 64
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Store 76 75
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
77: 11 Load 42(g_tex[1])
78: 53(ptr) AccessChain 73(local_texture_array) 35
Store 78 77
80: 11 Load 79(g_tex[2])
81: 53(ptr) AccessChain 73(local_texture_array) 36
Store 81 80
95: 94(ptr) AccessChain 93 36
96: 90 Load 95
97: 6(float) CompositeExtract 96 0
99: 98(ptr) AccessChain 85(local_float_array) 64
Store 99 97
100: 6(float) CompositeExtract 96 1
101: 98(ptr) AccessChain 85(local_float_array) 35
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Store 101 100
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
102: 6(float) CompositeExtract 96 2
103: 98(ptr) AccessChain 85(local_float_array) 36
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Store 103 102
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
104: 6(float) CompositeExtract 96 3
105: 98(ptr) AccessChain 85(local_float_array) 37
Store 105 104
106: 7(fvec4) FunctionCall 9(TestFn1()
108: 11 Load 74(g_tex[0])
109: 53(ptr) AccessChain 107(aggShadow) 64
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Store 109 108
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
110: 11 Load 42(g_tex[1])
111: 53(ptr) AccessChain 107(aggShadow) 35
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Store 111 110
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
112: 11 Load 79(g_tex[2])
113: 53(ptr) AccessChain 107(aggShadow) 36
Store 113 112
115: 16 Load 65(g_samp[0])
116: 56(ptr) AccessChain 114(aggShadow) 64
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Store 116 115
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
117: 16 Load 45(g_samp[1])
118: 56(ptr) AccessChain 114(aggShadow) 35
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Store 118 117
HLSL: Pass opaques by local copy, instead of by interface original. Also, remove assumption that if something is opaque that it must be in the UniformConstant storage class. This allows function declarations to know all parameters will be in the Function storage class.
2017-09-08 08:38:07 +00:00
119: 16 Load 70(g_samp[2])
120: 56(ptr) AccessChain 114(aggShadow) 36
Store 120 119
122: 14 Load 107(aggShadow)
Store 121(param) 122
124: 17 Load 114(aggShadow)
Store 123(param) 124
125: 7(fvec4) FunctionCall 22(TestFn2(t11[3];p1[3];) 121(param) 123(param)
126: 7(fvec4) FAdd 106 125
128: 127(ptr) AccessChain 27(ps_output) 64
Store 128 126
HLSL: Wrap the entry-point; need to write 'in' args, and support 'inout' args. This needs some render testing, but is destined to be part of master. This also leads to a variety of other simplifications. - IO are global symbols, so only need one list of linkage nodes (deferred) - no longer need parse-context-wide 'inEntryPoint' state, entry-point is localized - several parts of splitting/flattening are now localized
2017-01-19 22:41:47 +00:00
Return
Flatten uniform arrays This checkin adds a --flatten-uniform-arrays option which can break uniform arrays of samplers, textures, or UBOs up into individual scalars named (e.g) myarray[0], myarray[1], etc. These appear as individual linkage objects. Code notes: - shouldFlatten internally calls shouldFlattenIO, and shouldFlattenUniform, but is the only flattening query directly called. - flattenVariable will handle structs or arrays (but not yet arrayed structs; this is tested an an error is generated). - There's some error checking around unhandled situations. E.g, flattening uniform arrays with initializer lists is not implemented. - This piggybacks on as much of the existing mechanism for struct flattening as it can. E.g, it uses the same flattenMap, and the same flattenAccess() method. - handleAssign() has been generalized to cope with either structs or arrays. - Extended test infrastructure to test flattening ability.
2016-09-16 19:26:37 +00:00
FunctionEnd
Reference in New Issue Copy Permalink