summaryrefslogtreecommitdiffstats
path: root/src/compiler
Commit message (Collapse)AuthorAgeFilesLines
* glsl/linker: attempt to match anonymous structures at linkDave Airlie2016-05-191-9/+15
| | | | | | | | | | | This is my attempt at fixing at least one of the UE4 bugs with GL4.3. If we are doing intrastage matching and hit anonymous structs, then we should do a record comparison instead of using the names. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95005 Reviewed-by: Ian Romanick <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* spirv: add switch case for nir_texop_txf_ms_mcs in vtn_handle_texture()Brian Paul2016-05-181-0/+2
| | | | | | | | | | | Mark it as unreachable. Silences a compiler warning: spirv/spirv_to_nir.c:1397:4: warning: enumeration value 'nir_texop_txf_ms_mcs' not handled in switch [-Wswitch] switch (instr->op) { ^ Reviewed-by: Jason Ekstrand <[email protected]>
* glsl: Check that layout is non-null before dereferencing.Matt Turner2016-05-181-1/+1
| | | | | | | layout should only be null for structs, but it's checked everywhere else and confuses Coverity (CID 1358495). Reviewed-by: Timothy Arceri <[email protected]>
* spirv: Properly size the src[] array.Matt Turner2016-05-181-1/+1
| | | | | | | Operations like nir_op_bitfield_insert have four arguments, and Coverity isn't privy to the fact that 4-argument operations aren't possible here, so it thinks this can lead to memory corruption. Just increase the size of the array to quell any fears.
* glsl/linker: Ensure the first stage of an SSO pipeline has input locs assignedIan Romanick2016-05-181-3/+6
| | | | | | | | | | | | | | | | | | | | | Previously an SSO pipeline containing only a tessellation control shader and a tessellation evaluation shader would not get locations assigned for the TCS inputs. This would lead to assertion failures in some piglit tests, such as arb_program_interface_query-resource-query. That piglit test still fails on some tessellation related subtests. Specifically, these subtests fail: 'GL_PROGRAM_INPUT(tcs) active resources' expected 2 but got 3 'GL_PROGRAM_INPUT(tcs) max length name' expected 12 but got 16 'GL_PROGRAM_INPUT(tcs,tes) active resources' expected 2 but got 3 'GL_PROGRAM_INPUT(tcs,tes) max length name' expected 12 but got 16 'GL_PROGRAM_OUTPUT(tcs) active resources' expected 15 but got 3 'GL_PROGRAM_OUTPUT(tcs) max length name' expected 23 but got 12 Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Timothy Arceri <[email protected]> Cc: [email protected]
* glsl/linker: Don't include interface name for built-in blocksIan Romanick2016-05-181-1/+2
| | | | | | | | | | | | | | Commit 11096ec introduced a regression in some piglit tests (e.g., arb_program_interface_query-resource-query). I did not notice this regression because other (unrelated) problems caused failed assertions in those same tests on my system... so they crashed before getting to the new failure. v2: Use is_gl_identifier. Suggested by Tim. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Timothy Arceri <[email protected]> Cc: [email protected]
* glsl: Assert that inputs have a location assignedIan Romanick2016-05-181-0/+2
| | | | | | | This catches a problem previously undetected until deep in the backend. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* glsl/linker: Fix trivial typos in commentsIan Romanick2016-05-182-8/+8
| | | | | Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* glsl/linker: Fix some formatting to match current coding conventionsIan Romanick2016-05-181-8/+6
| | | | | Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* glsl/linker: Silence unused parameter warningIan Romanick2016-05-181-5/+3
| | | | | | | | | | | The use of the parameter was removed in d6b92028. glsl/link_varyings.cpp:1390:39: warning: unused parameter ‘separate_shader’ [-Wunused-parameter] bool separate_shader) ^ Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* glsl/linker: Silence unused parameter warningIan Romanick2016-05-181-3/+2
| | | | | | | | | | | | The parameter appears to have been unused since the function was added in commit 12ba6cfb. Remove it. glsl/linker.cpp:2886:60: warning: unused parameter ‘prog’ [-Wunused-parameter] match_explicit_outputs_to_inputs(struct gl_shader_program *prog, ^ Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* nir/validate: dump annotated shader with error msgsRob Clark2016-05-171-1/+62
| | | | | | | | | Log all the errors, and at the end dump the shader w/ error annotations to make it easier to see where the problems are. Signed-off-by: Rob Clark <[email protected]> Reviewed-by: Eduardo Lima Mitev <[email protected]> Reviewed-by: Connor Abbott <[email protected]>
* nir/validate: assert() -> validate_assert()Rob Clark2016-05-171-126/+128
| | | | | | | Prep work for next patch. Signed-off-by: Rob Clark <[email protected]> Reviewed-by: Connor Abbott <[email protected]>
* nir/print: add support for print annotationsRob Clark2016-05-172-1/+35
| | | | | | | | | | | | | | | Caller can pass a hashtable mapping NIR object (currently instr or var, but I guess others could be added as needed) to annotation msg to print inline with the shader dump. As the annotation msg is printed, it is removed from the hashtable to give the caller a way to know about any unassociated msgs. This is used in the next patch, for nir_validate to try to associate error msgs to nir_print dump. Signed-off-by: Rob Clark <[email protected]> Reviewed-by: Eduardo Lima Mitev <[email protected]> Reviewed-by: Connor Abbott <[email protected]>
* nir: add double input bitmapJuan A. Suarez Romero2016-05-172-0/+3
| | | | | | This bitmap tracks which input attributes are double-precision. Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: make reserved_varying_slot() staticTimothy Arceri2016-05-171-1/+1
| | | | Reviewed-by: Dave Airlie <[email protected]>
* glsl: include per-patch varyings when generating reserved slot bitfieldTimothy Arceri2016-05-171-3/+6
| | | | Reviewed-by: Dave Airlie <[email protected]>
* glsl: don't incorrectly eliminate patches with explicit locationsTimothy Arceri2016-05-171-1/+2
| | | | | | | These varying have a separate location domain from per-vertex varyings and need to be handled separately. Reviewed-by: Dave Airlie <[email protected]>
* glsl: remove remainings tabs in link_varyings.cppTimothy Arceri2016-05-171-2/+2
| | | | Reviewed-by: Dave Airlie <[email protected]>
* glsl: fix location and component packing validation on patchesTimothy Arceri2016-05-172-1/+3
| | | | | | | These varyings have a separate location domain from per-vertex varyings and need to be handled separately. Reviewed-by: Dave Airlie <[email protected]>
* glsl/linker: Include the interface name for input and output blocksIan Romanick2016-05-161-1/+16
| | | | | | | | | On my oes_shader_io_blocks branch, this fixes 71 dEQP-GLES31.functional.program_interface_query.* tests. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]> Cc: [email protected]
* glsl/linker: Use canonical format for ARB_program_interface_query spec quotesIan Romanick2016-05-161-49/+51
| | | | | Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* nir: Mark nir_start_block()/nir_impl_last_block() with returns_nonnull.Matt Turner2016-05-161-4/+5
| | | | Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Don't do constant propagation in opt_constant_folding.Kenneth Graunke2016-05-152-0/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | opt_constant_folding is supposed to fold trees of constants into a single constant. Surprisingly, it was also propagating constant values from variables into expression trees - even when the result couldn't be folded together. This is opt_constant_propagation's job. The ir_dereference_variable::constant_expression_value() method returns a clone of var->constant_value. So we would replace the dereference with a constant, propagating it into the tree. Skip over ir_dereference_variable to avoid this surprising behavior. However, add code to explicitly continue doing it in the constant propagation pass, as it's useful to do so. shader-db statistics on Broadwell: total instructions in shared programs: 8905349 -> 8905126 (-0.00%) instructions in affected programs: 30100 -> 29877 (-0.74%) helped: 93 HURT: 20 total cycles in shared programs: 71017030 -> 71015944 (-0.00%) cycles in affected programs: 132456 -> 131370 (-0.82%) helped: 54 HURT: 45 The only hurt programs are by a single instruction, while the helped ones are helped by 1-4 instructions. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* glsl: Avoid excess tree walking when folding ir_dereference_arrays.Kenneth Graunke2016-05-151-0/+6
| | | | | | | | | | | | | | | | | | | If an ir_dereference_array has non-constant components, there's no point in trying to evaluate its value (which involves walking down the tree and possibly allocating memory for portions of the subtree which are constant). This also removes convoluted tree walking in opt_constant_folding(), which tries to fold constants while walking up the tree. No need to walk down, then up, then down again. We did this for swizzles and expressions already, but I was lazy back in the day and didn't do this for ir_dereference_array. No change in shader-db. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* glsl: Consolidate duplicate copies of constant folding.Kenneth Graunke2016-05-153-32/+19
| | | | | | | | | | We could probably clean this up more (maybe make it a method), but at least there's only one copy of this code now, and that's a start. No change in shader-db. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* glsl: Remove bonus tree walking in opt_constant_folding().Kenneth Graunke2016-05-151-2/+0
| | | | | | | | | | | | | | It looks like this was missed when converting opt_constant_folding() from a hierarchical visitor to an rvalue visitor in 6606fde3. ir_rvalue_visitor already processes values on the way back up the tree, so we will have already visited every child node. There's no point in doing it again. No change in shader-db. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* glsl: Make opt_constant_variable() bail in useless cases.Kenneth Graunke2016-05-151-0/+7
| | | | | | | | | | | | | | | The pass ultimately skips over any entries with assignment_count != 1, so there's no need to do further work once we've determined that there are multiple assignments. The constant value could be a large array (i.e. uvec4[327]), at which point skipping the constant_expression_value() call (and the clone() call within) can save us piles of memory. No change in shader-db. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* nir: Add a nir->info.uses_interp_var_at_offset flag.Kenneth Graunke2016-05-153-0/+10
| | | | | | | | | I've added this to nir_gather_info(), but also to glsl_to_nir() as a temporary measure, since the i965 GL driver today doesn't use nir_gather_info() yet. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* glsl: Drop bad ASSERT_TRUE in gl_CullDistance link_varyings test.Kenneth Graunke2016-05-151-6/+6
| | | | | | | | | | | | I don't know what the intention was here, but this function returns void. We can't assert anything about its return value. Fixes "make check" failures. v2: Also fix prototype for the function (caught by Jordan). Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* nir: forward-declare 'struct gl_shader_program'Rob Clark2016-05-151-1/+2
| | | | | | | | Drop extra #include which is otherwise unneeded (and makes this header difficult to include from outside of src/mesa). Signed-off-by: Rob Clark <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* nir: return progress from lower_idivRob Clark2016-05-152-7/+16
| | | | | | | | | | With algebraic-opt support for lowering div to shift, the driver would like to be able to run this pass *after* the main opt-loop, and then conditionally re-run the opt-loop if this pass actually lowered some- thing. Signed-off-by: Rob Clark <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* nir: fix comment typo about f2d/d2fRob Clark2016-05-151-2/+2
| | | | | Signed-off-by: Rob Clark <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* nir: Add texture opcodes and source types for multisample compressionJason Ekstrand2016-05-142-0/+12
| | | | | | | | | | | Intel hardware does a form of multisample compression that involves an auxilary surface called the MCS. When an MCS is in use, you have to first sample from the MCS with a special opcode and then pass the result of that operation into the next sample instrucion. Normally, we just do this ourselves in the back-end, but we want to expose that functionality to NIR so that we can use MCS values directly in NIR-based blorp. Reviewed-by: Kenneth Graunke <[email protected]>
* nir/builder: Add a helper for grabbing multiple channels from an ssa defJason Ekstrand2016-05-141-0/+14
| | | | | | | This is similar to nir_channel except that it lets you grab more than one channel by providing a mask. Reviewed-by: Kenneth Graunke <[email protected]>
* nir/builder: Generate the alu helpers directly in pythonJason Ekstrand2016-05-142-31/+13
| | | | | | | | | There's no reason for having a macro *and* a python generator. We can easily just do the whole thing in python. This has the advantage that we are no longer definining ALU# macros which conflict with the ones in brw_fs_builder.h. Reviewed-by: Kenneth Graunke <[email protected]>
* nir: Add an info bit for uses_sample_qualifierJason Ekstrand2016-05-143-1/+13
| | | | Reviewed-by: Kenneth Graunke <[email protected]>
* Revert "glsl: Extend lowering pass for gl_ClipDistance to support other ↵Dave Airlie2016-05-143-160/+95
| | | | | | | | arrays (v4)" This reverts commit ad355652c20b245f5f2faa8622e71461e3121a7f. This broke a bunch of clip tests.
* glsl: make sure that textureProj(bias) variants are only exposed in fsIlia Mirkin2016-05-131-37/+37
| | | | | | | | | Many were already marked as fs_only, but not all. This fixes the remaining ir_txb entries. Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* glsl: be more strict when validating shader inputsIlia Mirkin2016-05-131-8/+18
| | | | | | | | | | | | | | | | | | interpolateAt* can only take input variables or an element of an input variable array. No structs. Further, GLSL 4.40 relaxes the requirement to allow swizzles, so enable that as well. This fixes the following dEQP tests: dEQP-GLES31.functional.shaders.multisample_interpolation.interpolate_at_sample.negative.interpolate_struct_member dEQP-GLES31.functional.shaders.multisample_interpolation.interpolate_at_centroid.negative.interpolate_struct_member dEQP-GLES31.functional.shaders.multisample_interpolation.interpolate_at_offset.negative.interpolate_struct_member Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Chris Forbes <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* glsl: make sure that interpolateAt arguments are variablesIlia Mirkin2016-05-131-1/+1
| | | | | | | | | | | | | | | In the case of a constant, it might have been propagated through and variable_referenced() returns NULL. Error out in that case. Fixes 3 dEQP tests: dEQP-GLES31.functional.shaders.multisample_interpolation.interpolate_at_sample.negative.interpolate_constant dEQP-GLES31.functional.shaders.multisample_interpolation.interpolate_at_centroid.negative.interpolate_constant dEQP-GLES31.functional.shaders.multisample_interpolation.interpolate_at_offset.negative.interpolate_constant Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Eduardo Lima Mitev <[email protected]> Reviewed-by: Chris Forbes <[email protected]>
* glsl: Add arb_cull_distance support (v3)Tobias Klausmann2016-05-1411-39/+181
| | | | | | | | | v2: make too large array a compile error v3: squash mesa/prog patch to avoid static compiler errors in bisect Signed-off-by: Tobias Klausmann <[email protected]> Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Kristian Høgsberg <[email protected]>
* glsl: Extend lowering pass for gl_ClipDistance to support other arrays (v4)Tobias Klausmann2016-05-143-95/+160
| | | | | | | | | | | | | | | | | | | This will come in handy when we want to lower gl_CullDistance into gl_CullDistanceMESA. [airlied: drop separate APIs for clip/cull - just use single API to call both passes.] v3: reexamine my sanity, this was pretty broken, the new code creates one copy of gl_ClipDistanceMESA, as the clip distance varying and lowers everything into that in two passes, one for clips one for culls. v4: rework using the passes in clip/cull sizes, instead of the array sizes. Signed-off-by: Tobias Klausmann <[email protected]> Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Kristian Høgsberg <[email protected]>
* glsl: rename lower_clip_distance to lower_distance.Dave Airlie2016-05-142-55/+55
| | | | | | | | | | This just renames the file in anticipation of adding cull lowering, and renames the internals. Signed-off-by: Tobias Klausmann <[email protected]> Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Kristian Høgsberg <[email protected]>
* mesa/main: Add support for GL_ARB_cull_distance (v2)Tobias Klausmann2016-05-143-3/+3
| | | | | | | | | | | | airlied: v2: rename LowerClipDistance to LowerCombinedClipCullDistnace. I don't think we want any other behaviour with any current hw. Signed-off-by: Tobias Klausmann <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Ian Romanick <[email protected]> Reviewed-by: Kristian Høgsberg <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* nir: Remove empty visit_call_src and visit_load_const_src functionsIan Romanick2016-05-121-17/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The guts were removed in dfb3abba. It has been almost exactly a year, so I dont think we're going to "decide we want [predication] back." Silences several "unused parameter" warnings: nir/nir.c: In function ‘visit_call_src’: nir/nir.c:1052:32: warning: unused parameter ‘instr’ [-Wunused-parameter] visit_call_src(nir_call_instr *instr, nir_foreach_src_cb cb, void *state) ^ nir/nir.c:1052:58: warning: unused parameter ‘cb’ [-Wunused-parameter] visit_call_src(nir_call_instr *instr, nir_foreach_src_cb cb, void *state) ^ nir/nir.c:1052:68: warning: unused parameter ‘state’ [-Wunused-parameter] visit_call_src(nir_call_instr *instr, nir_foreach_src_cb cb, void *state) ^ nir/nir.c: In function ‘visit_load_const_src’: nir/nir.c:1058:44: warning: unused parameter ‘instr’ [-Wunused-parameter] visit_load_const_src(nir_load_const_instr *instr, nir_foreach_src_cb cb, ^ nir/nir.c:1058:70: warning: unused parameter ‘cb’ [-Wunused-parameter] visit_load_const_src(nir_load_const_instr *instr, nir_foreach_src_cb cb, ^ nir/nir.c:1059:28: warning: unused parameter ‘state’ [-Wunused-parameter] void *state) ^ v2: Add some comments in nir_foreach_src suggested by Jason. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]> Cc: Connor Abbott <[email protected]>
* nir: Silence unused parameter warningsIan Romanick2016-05-124-6/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | These cases had the parameter removed: nir/nir_lower_vec_to_movs.c: In function ‘try_coalesce’: nir/nir_lower_vec_to_movs.c:124:66: warning: unused parameter ‘shader’ [-Wunused-parameter] try_coalesce(nir_alu_instr *vec, unsigned start_idx, nir_shader *shader) ^ nir/nir_lower_io.c: In function ‘load_op’: nir/nir_lower_io.c:147:32: warning: unused parameter ‘state’ [-Wunused-parameter] load_op(struct lower_io_state *state, ^ These cases had the parameter (void) silenced because the parameter was necessary for an interface: nir/glsl_to_nir.cpp:1900:32: warning: unused parameter 'ir' [-Wunused-parameter] nir_visitor::visit(ir_barrier *ir) ^ nir/nir.c: In function ‘remove_use_cb’: nir/nir.c:802:35: warning: unused parameter ‘state’ [-Wunused-parameter] remove_use_cb(nir_src *src, void *state) ^ nir/nir.c: In function ‘remove_def_cb’: nir/nir.c:811:37: warning: unused parameter ‘state’ [-Wunused-parameter] remove_def_cb(nir_dest *dest, void *state) ^ Number of total warnings in my build reduced from 2543 to 2538 (reduction of 5). Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* nir: glsl_get_bit_size() should take glsl_typeRob Clark2016-05-129-21/+20
| | | | | | | | It's what all the call-sites once, so gets rid of a bunch of inlined glsl_get_base_type() at the call-sites. Signed-off-by: Rob Clark <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* glsl: fixed uninitialized pointerJakob Sinclair2016-05-121-0/+5
| | | | | | | | | | | | | Class "ir_constant" had a bunch of constructors where the pointer member "array_elements" had not been initialized. This could have lead to unsafe code if something had tried to write anything to it. This patch fixes this issue by initializing the pointer to NULL in all the constructors. This issue was discovered by Coverity. CID: 401603, 401604, 401605, 401610 Signed-off-by: Jakob Sinclair <[email protected]> Reviewed-by: Alejandro Piñeiro <[email protected]>
* nir/algebraic: Separate ffma lowering from fusingJason Ekstrand2016-05-112-1/+2
| | | | | | | | The i965 driver has its own pass for fusing mul+add combinations that's much smarter than what nir_opt_algebraic can do so we don't want to get the nir_opt_algebraic one just because we didn't set lower_ffma. Reviewed-by: Kenneth Graunke <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-22 11:36:02 +0000