summaryrefslogtreecommitdiffstats
path: root/src/compiler
Commit message (Collapse)AuthorAgeFilesLines
* nir: Share destination rewriting and replacement code in IO lowering.Kenneth Graunke2016-07-151-25/+19
| | | | | | | | Both loads and atomics had identical code to rewrite destinations, and all cases had the same two lines to replace instructions. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* nir: Share get_io_offset handling in nir_lower_io.Kenneth Graunke2016-07-151-24/+9
| | | | | | | | | | | | The load/store/atomic cases all duplicated the get_io_offset code, with a few tiny differences: stores didn't bother checking for per-vertex inputs, because they can't be stored to, and atomics didn't check at all, since shared variables aren't per-vertex. However, it's harmless to check, and allows us to share more code. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* nir: Make a 'var' temporary in nir_lower_io.Kenneth Graunke2016-07-151-16/+12
| | | | | | | Less typing and word wrapping issues than intrin->variables[0]->var. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* glsl/types: Use _mesa_hash_data for hashing function typesJason Ekstrand2016-07-141-14/+2
| | | | | | | | | | This is way better than the stupid string approach especially since you could overflow the string. Again, I thought I had something better at one point but it obviously got lost. Signed-off-by: Jason Ekstrand <[email protected]> Reviewed-by: Iago Toral Quiroga <[email protected]> Cc: "12.0" <[email protected]>
* glsl/types: Fix function type comparison functionJason Ekstrand2016-07-141-1/+1
| | | | | | | | | | It was returning true if the function types have different lengths rather than false. This was new with the SPIR-V to NIR pass and I thought I'd fixed it a while ago but it may have gotten lost in rebasing somewhere. Signed-off-by: Jason Ekstrand <[email protected]> Reviewed-by: Iago Toral Quiroga <[email protected]> Cc: "12.0" <[email protected]>
* nir/spirv: Don't multiply the push constant block size by 4Jason Ekstrand2016-07-131-1/+1
| | | | | | | | | I have no idea why we were multiplying by 4 before. The offsets we get from SPIR-V are in bytes and so is nir->num_uniforms so there's no need to do any adjustment whatsoever. Signed-off-by: Jason Ekstrand <[email protected]> Cc: "12.0" <[email protected]>
* nir: Add optimization for (a || True == True)Eric Anholt2016-07-121-0/+1
| | | | | | | | | | | | This was appearing in vc4 VS/CS in mupen64, due to vertex attrib lowering producing some constants that were getting compared. total instructions in shared programs: 112276 -> 112198 (-0.07%) instructions in affected programs: 2239 -> 2161 (-3.48%) total estimated cycles in shared programs: 283102 -> 283038 (-0.02%) estimated cycles in affected programs: 2365 -> 2301 (-2.71%) Reviewed-by: Jason Ekstrand <[email protected]>
* nir: add glsl_dvec_type() helperTimothy Arceri2016-07-062-0/+7
| | | | Reviewed-by: Kenneth Graunke <[email protected]>
* nir: add glsl_double_type() helperTimothy Arceri2016-07-072-0/+7
| | | | Reviewed-by: Kenneth Graunke <[email protected]>
* nir: use the same driver location for packed varyingsTimothy Arceri2016-07-072-4/+28
| | | | Reviewed-by: Kenneth Graunke <[email protected]>
* nir: add new intrinsic field for storing component offsetTimothy Arceri2016-07-074-6/+23
| | | | | | This offset is used for packing. Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: don't try to lower non-gl builtins as if they were gl_FragDataIlia Mirkin2016-07-051-1/+2
| | | | | | | | | | | | | | If a shader has an output array, it will get treated as though it were gl_FragData and rewritten into gl_out_FragData instances. We only want this to happen on the actual gl_FragData and not everything else. This is a small part of the problem pointed out by the below bug. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96765 Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Marek Olšák <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]> Cc: "11.2 12.0" <[email protected]>
* glsl: Document and enforce restriction on type valuesIan Romanick2016-07-052-0/+10
| | | | | Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* glsl: Pack integer and double varyings as flat even if interpolation mode is ↵Ian Romanick2016-07-053-6/+15
| | | | | | | | | | | | | | none v2: Also update varying_matches::compute_packing_class(). Suggested by Timothy Arceri. Signed-off-by: Ian Romanick <[email protected]> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96358 Reviewed-by: Kenneth Graunke <[email protected]> Cc: "12.0" <[email protected]> Cc: Gregory Hainaut <[email protected]> Cc: Ilia Mirkin <[email protected]>
* glsl: stop allocating memory for UBOs during linkingTimothy Arceri2016-07-051-5/+8
| | | | | | | | | This just stops counting and assigning a storage location for these uniforms, the count is only used to create the uniform storage. These uniform types don't use this storage. Reviewed-by: Samuel Iglesias Gonsálvez <[email protected]>
* glsl: mark link_uniform_blocks_are_compatible() as staticTimothy Arceri2016-07-052-5/+1
| | | | | | Missed this when doing 6d1a59d15b. Reviewed-by: Samuel Iglesias Gonsálvez <[email protected]>
* glsl: add driconf to zero-init unintialized varsRob Clark2016-07-023-0/+10
| | | | | | | | | | | | | Some games are sloppy.. perhaps because it is defined behavior for DX or perhaps because nv blob driver defaults things to zero. So add driconf param to force uninitialized variables to default to zero. This issue was observed with rust, from steam store. But has surfaced elsewhere in the past. Signed-off-by: Rob Clark <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl/standalone: initialize MaxUserAssignableUniformLocationsRob Clark2016-07-021-0/+4
| | | | Signed-off-by: Rob Clark <[email protected]>
* glsl: atomic counters are different than their uniformsAndres Gomez2016-06-301-37/+37
| | | | | | | | | | The linker deals with atomic counters in terms of uniforms but the data structure are called after the atomic counters. Renamed the data structures used in the linker for disambiguation. Reviewed-by: Timothy Arceri <[email protected]> Signed-off-by: Andres Gomez <[email protected]>
* glsl: count atomic counters correctlyAndres Gomez2016-06-301-4/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently the linker uses the uniform count for the total number of atomic counters. However uniforms don't include the innermost array dimension in their count, but atomic counters are expected to include them. Although the spec doesn't directly state this, it's clear how offsets will be assigned for arrays. From OpenGL 4.2 (Core Profile), page 98: " * Arrays of type atomic_uint are stored in memory by element order, with array element member zero at the lowest offset. The difference in offsets between each pair of elements in the array in basic machine units is referred to as the array stride, and is constant across the entire array. The stride can be queried by calling GetIntegerv with a pname of ATOMIC_COUNTER_- ARRAY_STRIDE after a program is linked." From that it is clear how arrays of atomic counters will interact with GL_MAX_ATOMIC_COUNTER_BUFFER_SIZE. For other kinds of uniforms it's also clear that each entry in an array counts against the relevant limits. Hence, although inferred, this is the expected behavior. Fixes GL44-CTS.arrays_of_arrays_gl.AtomicDeclaration Reviewed-by: Timothy Arceri <[email protected]> Signed-off-by: Andres Gomez <[email protected]>
* glsl/mesa: move duplicate shader fields into new struct gl_shader_infoTimothy Arceri2016-06-304-147/+171
| | | | Reviewed-by: Iago Toral Quiroga <[email protected]>
* glsl/main: remove unused params and make function staticTimothy Arceri2016-06-303-5/+3
| | | | Reviewed-by: Iago Toral Quiroga <[email protected]>
* glsl: simplify link_uniform_blocks()Timothy Arceri2016-06-303-10/+5
| | | | | | There is only ever one shader so simplify the input params. Reviewed-by: Iago Toral Quiroga <[email protected]>
* glsl/mesa: split gl_shader in twoTimothy Arceri2016-06-3025-121/+161
| | | | | | | | | | | | | | | | | There are two distinctly different uses of this struct. The first is to store GL shader objects. The second is to store information about a shader stage thats been linked. The two uses actually share few fields and there is clearly confusion about their use. For example the linked shaders map one to one with a program so can simply be destroyed along with the program. However previously we were calling reference counting on the linked shaders. We were also creating linked shaders with a name even though it is always 0 and called the driver version of the _mesa_new_shader() function unnecessarily for GL shader objects. Acked-by: Iago Toral Quiroga <[email protected]>
* glsl: pass symbols to find_matching_signature() rather than shaderTimothy Arceri2016-06-301-25/+22
| | | | | | This will allow us to later split gl_shader into two structs. Reviewed-by: Iago Toral Quiroga <[email protected]>
* glsl: pass symbols rather than shader to _mesa_get_main_function_signature()Timothy Arceri2016-06-305-7/+7
| | | | | | | This will allow us to split gl_shader into two different structs, one for shader objects and one for linked shaders. Reviewed-by: Iago Toral Quiroga <[email protected]>
* glsl: make cross_validate_globals() more genericTimothy Arceri2016-06-301-206/+207
| | | | | | | | | Rather than passing in gl_shader we now pass in the IR. This will allow us to later split gl_shader into two structs. One for use as a linked per stage shader struct and one for use as a GL shader object. Reviewed-by: Iago Toral Quiroga <[email protected]>
* glsl: Ignore ir_texture in lower_const_arrays_to_uniforms.Kenneth Graunke2016-06-271-0/+7
| | | | | | | | | | | | | The only part of an ir_texture which can be an array is the offsets array in textureGatherOffsets() calls. We don't want to lower those, because they're required to remain constants. Fixes textureGatherOffsets with Gallium drivers such as llvmpipe, which commit ef78df8d3b0cf540e5f08c8c2f6caa338b64a6c7 regressed. Cc: [email protected] Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* nir: Fix copy_prop_src when src is an indirect access on a reg.Eric Anholt2016-06-261-1/+1
| | | | | | | | | | | | The intent was to continue down the indirect chain, not to call ourselves with unchanged input arguments. Found by code inspection, and comparison to copy_prop_alu_src(). We haven't hit this because callers of NIR's copy prop are doing so in SSA, before indirect variable dereferences have been lowered to registers. Reviewed-by: Rob Clark <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* nir: Add a NIR_VALIDATE environment variableJason Ekstrand2016-06-251-0/+6
| | | | | | | | | | | It defaults to true so default behavior doesn't change but it allows you to do NIR_VALIDATE=false if you don't want validation. Disabling validation can substantially speed up shader compiles so you frequently want to turn it off if compiler invariants aren't in question. Reviewed-by: Matt Turner <[email protected]> Reviewed-by: Matt Turner <[email protected]> Signed-off-by: Rob Clark <[email protected]>
* Remove wrongly repeated words in commentsGiuseppe Bilotta2016-06-239-10/+10
| | | | | | | | | | | | | | | | | Clean up misrepetitions ('if if', 'the the' etc) found throughout the comments. This has been done manually, after grepping case-insensitively for duplicate if, is, the, then, do, for, an, plus a few other typos corrected in fly-by v2: * proper commit message and non-joke title; * replace two 'as is' followed by 'is' to 'as-is'. v3: * 'a integer' => 'an integer' and similar (originally spotted by Jason Ekstrand, I fixed a few other similar ones while at it) Signed-off-by: Giuseppe Bilotta <[email protected]> Reviewed-by: Chad Versace <[email protected]>
* glsl: Don't constant propagate arrays.Kenneth Graunke2016-06-231-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Constant propagation on arrays doesn't make a lot of sense. If the array is only accessed with constant indexes, then opt_array_splitting would split it up. Otherwise, we have variable indexing. If there's multiple accesses, then constant propagation would end up replicating the data. The lower_const_arrays_to_uniforms pass creates uniforms for each ir_constant with array type that it encounters. This means that it creates redundant uniforms for each copy of the constant, which means uploading too much data. It can even mean exceeding the maximum number of uniform components, causing link failures. We could try and teach the pass to de-duplicate the data by hashing constants, but it makes more sense to avoid duplicating it in the first place. We should promote constant arrays to uniforms, then propagate the uniform access. Fixes the TressFX shaders from Tomb Raider, which exceeded the maximum number of uniform components by a huge margin and failed to link. On Broadwell: total instructions in shared programs: 9067702 -> 9068202 (0.01%) instructions in affected programs: 10335 -> 10835 (4.84%) helped: 10 (Hoard, Shadow of Mordor, Amnesia: The Dark Descent) HURT: 20 (Natural Selection 2) loops in affected programs: 4 -> 0 The hurt programs appear to no longer have a constarray uniform, as all constants were successfully propagated. Apparently before this patch, we successfully unrolled a loop containing array access, but only after promoting constant arrays to uniforms. With this patch, we unroll it first, so all array access is direct, and the array is split up, and individual constants are propagated. This seems better. Cc: [email protected] Reported-by: Karol Herbst <[email protected]> Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* glsl: Make lower_const_arrays_to_uniforms work directly on constants.Kenneth Graunke2016-06-231-8/+3
| | | | | | | | | | | | | There's really no point in looking at ir_dereference_array of a constant. It also misses cases like: (assign () (var_ref tmp) (constant (array ...) ...)) No changes in shader-db, but keeps it working after the next commit. Cc: [email protected] Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* glsl: Propagate invariant/precise after lowering const arrays.Kenneth Graunke2016-06-231-0/+1
| | | | | | | | | | The new uniform may need precise as well. Fixes copy propagation of constant array uniforms in Tomb Raider shaders. Cc: [email protected] Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* glsl: Split arrays even in the presence of whole-array copies.Kenneth Graunke2016-06-231-0/+55
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously, we failed to split constant arrays. Code such as int[2] numbers = int[](1, 2); would generates a whole-array assignment: (assign () (var_ref numbers) (constant (array int 4) (constant int 1) (constant int 2))) opt_array_splitting generally tried to visit ir_dereference_array nodes, and avoid recursing into the inner ir_dereference_variable. So if it ever saw a ir_dereference_variable, it assumed this was a whole-array read and bailed. However, in the above case, there's no array deref, and we can totally handle it - we just have to "unroll" the assignment, creating assignments for each element. This was mitigated by the fact that we constant propagate whole arrays, so a dereference of a single component would usually get the desired single value anyway. However, I plan to stop doing that shortly; early experiments with disabling constant propagation of arrays revealed this shortcoming. This patch causes some arrays in Gl32GSCloth's geometry shaders to be split, which allows other optimizations to eliminate unused GS inputs. The VS then doesn't have to write them, which eliminates the entire VS (5 -> 2 instructions). It still renders correctly. No other change in shader-db. v2: Drop !AOA check and improve a comment (feedback from Tim Arceri). Cc: [email protected] Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* glsl: Make constant propagation's folder not propagate into an LHS.Kenneth Graunke2016-06-231-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | opt_constant_propagation.cpp contains constant folding code which can actually do constant propagation in some cases. It was happily propagating constants into the left-hand-side of assignments. For example, (assign () (var_ref temp) (constant ...)) would brilliantly be turned into: (assign () (constant ...) (constant ....)) This is a bigger hammer than necessary - it prevents propagation into the left-hand-side altogether. We could certainly do better someday. Notably, the constant propagation pass itself already takes this approach - it's just the constant propagation pass's built-in constant folding code (which actually propagates, too) that was broken. No change in shader-db, but prevents regressions after future commits. It seems plausible that this could be hit today, but I haven't seen it happen. Cc: [email protected] Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* glsl/mesa: stop duplicating geom and tcs layout valuesTimothy Arceri2016-06-232-12/+8
| | | | | | | | | | | | We already store these in gl_shader and gl_program here we remove it from gl_shader_program and just use the values from gl_shader. This will allow us to keep the shader cache restore code as simple as it can be while making it somewhat clearer where these values originate from. Reviewed-by: Iago Toral Quiroga <[email protected]>
* glsl/mesa: stop duplicating tes layout valuesTimothy Arceri2016-06-231-4/+0
| | | | | | | | | | | | | | | We already store this in gl_shader and gl_program here we remove it from gl_shader_program and just use the values from gl_shader. This will allow us to keep the shader cache restore code as simple as it can be while making it somewhat clearer where these values originate from. V2: remove unnecessary NULL check Reviewed-by: Marek Olšák <[email protected]> Reviewed-by: Iago Toral <[email protected]>
* glsl: only match gl_FragData and not gl_SecondaryFragDataEXTIlia Mirkin2016-06-211-2/+5
| | | | | | | | | | | | | There's special logic around finding gl_FragData. It latches onto any array with FRAG_RESULT_DATA0. However gl_SecondaryFragDataEXT[], added by GL_EXT_blend_func_extended, fits those parameters as well. The real frag data array should have index 0 though, so we can use that to distinguish them. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96617 Signed-off-by: Ilia Mirkin <[email protected]> Cc: "11.1 11.2 12.0" <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* spirv: Use the system value version of gl_FrontFaceJason Ekstrand2016-06-211-2/+2
| | | | | | | | | | SPIR-V treats it as an input but NIR wants the system value. This shouldn't have been too much of a surprise given that we have to do the same conversion in the GLSL IR to NIR pass. Signed-off-by: Jason Ekstrand <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]> Cc: "12.0" <[email protected]>
* nir/alu_to_scalar: Respect the exact ALU operation qualifierJason Ekstrand2016-06-201-0/+2
| | | | | | | | | | Just setting builder->exact isn't sufficient because that only applies to instructions that are built with the builder but instructions created manually and only inserted using the builder are left alone. Signed-off-by: Jason Ekstrand <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]> Cc: "12.0" <[email protected]>
* nir: Add a pass for propagating invariant decorationsJason Ekstrand2016-06-203-0/+199
| | | | | | | | | | This pass is similar to propagate_invariance in the GLSL compiler. The real "output" of this pass is that any algebraic operations which are eventually consumed by an invariant variable get marked as "exact". Signed-off-by: Jason Ekstrand <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]> Cc: "12.0" <[email protected]>
* nir/algebraic: Remove imprecise flog2 optimizationsJason Ekstrand2016-06-201-2/+0
| | | | | | | | | | | | | | | | | | | | | | | | While mathematically correct, these two optimizations result in an expression with substantially lower precision than the original. For any positive finite floating-point value, log2(x) is well-defined and finite. More precisely, it is in the range [-150, 150] so any sum of logarithms log2(a) + log2(b) is also well-defined and finite as long as a and b are both positive and finite. However, if a and b are either very small or very large, their product may get flushed to infinity or zero causing log2(a * b) to be nowhere close to log2(a) + log2(b). This imprecision was causing incorrect rendering in Talos Principal because part of its HDR rendering process involves doing 8 texture operations, clamping the result to [0, 65000], taking a dot-product with a constant, and then taking the log2. This is done 6 or 8 times and summed to produce the final result which is written to a red texture. In cases where you have a region of the screen that is very dark, it can end up getting a result value of -inf which is not what is intended. Signed-off-by: Jason Ekstrand <[email protected]> Reviewed-by: Matt Turner <[email protected]> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96425 Cc: "11.1 11.2 12.0" <[email protected]>
* mesa: Fix incorrect "see also" commentsIan Romanick2016-06-201-1/+1
| | | | | Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* glsl: Always strip arrayness in precision_qualifier_allowedIan Romanick2016-06-161-11/+6
| | | | | | | | | | | | | | | | Previously some callers of precision_qualifier_allowed would strip the arrayness from the type and some would not. As a result, some places would not notice that float[6], for example, needed a precision qualifier. Fixes the new piglit test no-default-float-array-precision.frag. Signed-off-by: Ian Romanick <[email protected]> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96358 Cc: "12.0" <[email protected]> Cc: Gregory Hainaut <[email protected]> Cc: Ilia Mirkin <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* mesa/glsl: stop using GL shader type internallyTimothy Arceri2016-06-166-9/+8
| | | | | | | | | | | | Instead use the internal gl_shader_stage enum everywhere. This makes things more consistent and gets rid of unnecessary conversions. Ideally it would be nice to remove the Type field from gl_shader altogether but currently it is used to differentiate between gl_shader and gl_shader_program in the ShaderObjects hash table. Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Optionally lower TCS gl_PatchVerticesIn to a uniform.Kenneth Graunke2016-06-151-1/+12
| | | | | | | | | i965 has no special hardware for this, so the best way to implement this is to pass it in via a uniform. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Alejandro Piñeiro <[email protected]> Cc: [email protected]
* glsl: Optionally lower TES gl_PatchVerticesIn to a uniform.Kenneth Graunke2016-06-151-3/+17
| | | | | | | | | | i965 has no special hardware for this, so we need to pass this value in as a uniform (unless the TES is linked against a TCS, in which case the linker can just replace this with a constant). Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Alejandro Piñeiro <[email protected]> Cc: [email protected]
* glsl/builtin_variables: Populate MaxCombinedShaderStorageBlocks on GLSL 4.40Eduardo Lima Mitev2016-06-141-1/+1
| | | | | | | | | | | | | | | Built-in variable "MaxCombinedShaderStorageBlocks" was added to GLSL 4.40 revision 9. Section "1.2.1 Changes since revision 8 of GLSL version 4.40", page 3 of the PDF states: "Bug 11734: Add gl_MaxCombinedShaderOutputResources and mark gl_MaxCombinedImageUnitsAndFragmentOutputs as deprecated." Fixes: GL44-CTS.shader_image_load_store.basic-glsl-const Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: make sure UBO arrays are sized in EScros-mesa-12.1.0-r7-vanillacros-mesa-12.1.0-r5-vanillacros-mesa-12.1.0-r3-vanillachadv/cros-mesa-12.1.0-r7-vanillachadv/cros-mesa-12.1.0-r5-vanillachadv/cros-mesa-12.1.0-r3-vanillaTimothy Arceri2016-06-141-0/+14
| | | | | | | | This check was removed in 5b2675093e86 add it back in. Reviewed-by: Dave Airlie <[email protected]> Cc: "12.0" <[email protected]> https://bugs.freedesktop.org/show_bug.cgi?id=96349
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-24 19:03:51 +0000