aboutsummaryrefslogtreecommitdiffstats
path: root/src/compiler/glsl/shader_cache.cpp
Commit message (Collapse)AuthorAgeFilesLines
* glsl/shader_cache: handle SPIR-V shadersAlejandro Piñeiro2019-07-121-5/+5
| | | | | | | | | Right now we don't have cache support for SPIR-V shaders (from ARB_gl_spirv). Right now they are properly skipped because they fall on the ff shader code path (no key, no name), but it would be better to update current comments, and add some guards. Reviewed-by: Caio Marcelo de Oliveira Filho <[email protected]>
* glsl: Fix out of bounds read in shader_cache_read_program_metadataKenneth Graunke2019-06-171-3/+2
| | | | | | | | | | | | | | | | | | The VaryingNames array has NumVaryings entries. But BufferStride is a small array of MAX_FEEDBACK_BUFFERS (4) entries. Programs with more than 4 varyings would read out of bounds. Also, BufferStride is set based on the shader itself, which means that it's inherently already included in the hash, and doesn't need to be included again. At the point when shader_cache_read_program_metadata is called, the linker hasn't even set those fields yet. So, just drop it entirely. Fixes valgrind errors in KHR-GL45.transform_feedback.linking_errors_test. Fixes: 6d830940f78 glsl/shader_cache: Allow shader cache usage with transform feedback Reviewed-by: Timothy Arceri <[email protected]>
* glsl: be much more aggressive when skipping shader compilationTimothy Arceri2019-01-191-6/+1
| | | | | | | | | | | | | | | | | | | | Currently we only add a cache key for a shader once it is linked. However games like Team Fortress 2 compile a whole bunch of shaders which are never actually linked. These compiled shaders can take up a bunch of memory. This patch changes things so that we add the key for the shader to the cache as soon as it is compiled. This means on a warm cache we can avoid the wasted memory from these shaders. Worst case scenario is we need to compile the shaders at link time but this can happen anyway if the shader has been evicted from the cache. Reduces memory use in Team Fortress 2 from 1.3GB -> 770MB on a warm cache from start up to the game menu. V2: only add key to cache when compilation is successful. Acked-by: Marek Olšák <[email protected]>
* Revert "glsl: be much more aggressive when skipping shader compilation"Timothy Arceri2019-01-191-1/+6
| | | | | | This reverts commit 64b8c86d37ebb1e1d286c69d642d52b7bcf051d3. Reverting for now as it was causing some segfaults.
* glsl: be much more aggressive when skipping shader compilationTimothy Arceri2019-01-191-6/+1
| | | | | | | | | | | | | | | | | | Currently we only add a cache key for a shader once it is linked. However games like Team Fortress 2 compile a whole bunch of shaders which are never actually linked. These compiled shaders can take up a bunch of memory. This patch changes things so that we add the key for the shader to the cache as soon as it is compiled. This means on a warm cache we can avoid the wasted memory from these shaders. Worst case scenario is we need to compile the shaders at link time but this can happen anyway if the shader has been evicted from the cache. Reduces memory use in Team Fortress 2 from 1.3GB -> 770MB on a warm cache from start up to the game menu. Acked-by: Marek Olšák <[email protected]>
* glsl: don't skip GLSL IR opts on first-time compilesTimothy Arceri2019-01-191-17/+0
| | | | | | | | | | | | | | This basically reverts c2bc0aa7b188. By running the opts we reduce memory using in Team Fortress 2 from 1.5GB -> 1.3GB from start-up to game menu. This will likely increase Deus Ex start up times as per commit c2bc0aa7b188. However currently 32bit games like Team Fortress 2 can run out of memory on low memory systems, so that seems more important. Reviewed-by: Marek Olšák <[email protected]>
* mesa: Add disk shader cache driver blob callbackJordan Justen2018-07-091-0/+8
| | | | | Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* mesa: include mtypes.h lessMarek Olšák2018-04-121-1/+1
| | | | | | | | | | - remove mtypes.h from most header files - add main/menums.h for often used definitions - remove main/core.h v2: fix radv build Reviewed-by: Brian Paul <[email protected]>
* glsl/shader_cache: Allow shader cache usage with transform feedbackJordan Justen2018-03-191-0/+6
| | | | | | | | Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105444 Suggested-by: Timothy Arceri <[email protected]> Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Tapani Pälli <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* mesa: change gl_link_status enums to uppercaseBrian Paul2018-01-261-1/+1
| | | | | | follow the convention of other enums. Reviewed-by: Neha Bhende <[email protected]>
* mesa: change gl_compile_status enums to uppercaseBrian Paul2018-01-261-1/+1
| | | | | | To follow the convention of other enums. Reviewed-by: Neha Bhende <[email protected]>
* glsl: Split out shader program serializationJordan Justen2017-12-081-1181/+4
| | | | | | | | This will allow us to use the program serialization to implement ARB_get_program_binary. Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* mesa: rework _mesa_add_parameter() to only add a single paramTimothy Arceri2017-11-291-2/+2
| | | | | | | | | This is more inline with what the functions name suggests it should do, and makes the code much easier to follow. This will also make adding uniform packing support much simpler. Reviewed-by: Nicolai Hähnle <[email protected]>
* st/glsl_to_tgsi: make use of driver_cache_blob with the disk cacheTimothy Arceri2017-11-281-1/+1
| | | | | | | | | | | | driver_cache_blob was introduced with the i965 disk cache, it allows us to simplify the cache a little and possibly offers some minor speed improvements since we load the GLSL metadata and TGSI from disk in one pass. Using driver_cache_blob should also make it straight forward to implement binary support for ARB_get_program_binary in gallium. Reviewed-by: Marek Olšák <[email protected]>
* glsl: use the correct parent when allocating program data membersTimothy Arceri2017-11-091-2/+2
| | | | | | Cc: "17.2 17.3" <[email protected]> Reviewed-by: Tapani Pälli <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: drop cache_fallbackTimothy Arceri2017-11-091-1/+1
| | | | | | | | | | This turned out to be a dead end, it is much easier and less error prone to just cache the IR used by the drivers backend e.g. TGSI or NIR. Cc: "17.2 17.3" <[email protected]> Reviewed-by: Tapani Pälli <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl/shader_cache: Save fs (BlendSupport) metadataJordan Justen2017-10-311-0/+6
| | | | | | | | | | | | | | | | | | Fixes many GL 4.5 CTS blend tests, such as: * GL45-CTS.blend_equation_advanced.extension_directive_enable * GL45-CTS.blend_equation_advanced.extension_directive_warn * GL45-CTS.blend_equation_advanced.blend_all.GL_MULTIPLY_KHR_all_qualifier * GL45-CTS.blend_equation_advanced.blend_specific.GL_COLORBURN_KHR v2: * Directly save the BlendSupport field to avoid potentially including a pointer in the future in the structure is updated. (tarceri) Cc: Timothy Arceri <[email protected]> Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Timothy Arceri <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* glsl/shader_cache: Save and restore serialized nir in gl_programJordan Justen2017-10-311-0/+16
| | | | | | | | | v3: * Rename serialized_nir* to driver_cache_blob*. (Tim) Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Timothy Arceri <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* glsl: move shader_cache type handling to glsl_typesConnor Abbott2017-10-251-171/+0
| | | | | | | | | Not sure if this is the best place to put it, but we're going to need this for NIR too. Reviewed-by: Timothy Arceri <[email protected]> Reviewed-by: Jordan Justen <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* compiler/blob: Switch to init/finish instead of create/destroyJason Ekstrand2017-10-121-19/+20
| | | | | | | | | There's no reason why that tiny bit of memory needs to be on the heap. We always put blob_reader on the stack, so why not do the same with the writable blob. Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* util: move string_to_uint_map to glslEmil Velikov2017-08-291-1/+1
| | | | | | | | | | | | | | | | | | The functionality is used by glsl and mesa. With the latter already depending on the former. With this in place the src/util/ static library libmesautil.la no longer has a C++ dependency. Thus objects which use it (like libEGL) don't need the C++ link. Cc: "17.2" <[email protected]> Fixes: 02cc35937277 ("egl/wayland: Use linux-dmabuf interface for buffers") Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101851 Signed-off-by: Emil Velikov <[email protected]> Suggested-by: Jason Ekstrand <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]> Tested-by: Mike Lothian <[email protected]> Tested-by: James Harvey <[email protected]>
* glsl: fix glsl_struct_field size calculations for shader cacheNicolai Hähnle2017-08-251-7/+4
| | | | | | | | | | | | | | | | | | Found by address sanitizer: ==22621==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x61400000cbd8 at pc 0x7f561610a4ff bp 0x7ffca85f9d50 sp 0x7ffca85f94f8 READ of size 344 at 0x61400000cbd8 thread T0 #0 0x7f561610a4fe (/usr/lib/x86_64-linux-gnu/libasan.so.3+0x5f4fe) #1 0x7f560bb305a5 in memcpy /usr/include/x86_64-linux-gnu/bits/string3.h:53 #2 0x7f560bb305a5 in blob_write_bytes ../../../mesa-src/src/compiler/glsl/blob.c:136 #3 0x7f560be7d7ff in encode_type_to_blob ../../../mesa-src/src/compiler/glsl/shader_cache.cpp:153 #4 0x7f560be81222 in write_program_resource_data ../../../mesa-src/src/compiler/glsl/shader_cache.cpp:950 #5 0x7f560be81222 in write_program_resource_list ../../../mesa-src/src/compiler/glsl/shader_cache.cpp:1118 #6 0x7f560be81222 in shader_cache_write_program_metadata(gl_context*, gl_shader_program*) ../../../mesa-src/src/compiler/glsl/shader_cache.cpp:1407 #7 0x7f560b825fdb in link_program ../../../mesa-src/src/mesa/main/shaderapi.c:1163 Fixes: 073a84ff60db ("glsl: stop adding pointers from glsl_struct_field to the cache") Reviewed-by: Timothy Arceri <[email protected]>
* glsl: pass shader source keys to the disk cacheTimothy Arceri2017-08-251-3/+17
| | | | | | | We don't actually write them to disk here. That will happen in the following commit. Reviewed-by: Nicolai Hähnle <[email protected]>
* glsl: stop adding pointers from bindless structs to the cacheTimothy Arceri2017-08-241-4/+8
| | | | | | | This is so we always create reproducible cache entries. Consistency is required for verification of any third party distributed shaders. Reviewed-by: Samuel Pitoiset <[email protected]>
* glsl: stop adding pointers from shader_info to the cacheTimothy Arceri2017-08-241-6/+25
| | | | | | | This is so we always create reproducible cache entries. Consistency is required for verification of any third party distributed shaders. Reviewed-by: Samuel Pitoiset <[email protected]>
* glsl: always write a name/label string to the cacheTimothy Arceri2017-08-241-4/+7
| | | | | | | | | In the following patch we will stop writing the pointer to cache. Unfortunately adding empty strings to that cache seems to be the only thing we can do here once we no longer have the pointers. Reviewed-by: Samuel Pitoiset <[email protected]>
* glsl: don't write uniform storage offset if there isn't oneTimothy Arceri2017-08-241-3/+10
| | | | | | | This is so we always create reproducible cache entries. Consistency is required for verification of any third party distributed shaders. Reviewed-by: Samuel Pitoiset <[email protected]>
* glsl: add has_uniform_storage() helper to shader cacheTimothy Arceri2017-08-241-6/+13
| | | | Reviewed-by: Samuel Pitoiset <[email protected]>
* glsl: stop adding pointers from glsl_struct_field to the cacheTimothy Arceri2017-08-241-7/+38
| | | | | | | This is so we always create reproducible cache entries. Consistency is required for verification of any third party distributed shaders. Reviewed-by: Samuel Pitoiset <[email protected]>
* glsl: stop adding pointers from gl_shader_variable to the cacheTimothy Arceri2017-08-241-12/+28
| | | | | | | This is so we always create reproducible cache entries. Consistency is required for verification of any third party distributed shaders. Reviewed-by: Samuel Pitoiset <[email protected]>
* glsl: allow NULL to be passed to encode_type_to_blob()Timothy Arceri2017-08-241-0/+10
| | | | | | This will be used by the following commit. Reviewed-by: Samuel Pitoiset <[email protected]>
* glsl: Track whether uniforms are active per stageKenneth Graunke2017-06-221-0/+2
| | | | | | | | | | for finer granularity state flagging v2: Marek - use a bitmask, add shader cache support Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Brian Paul <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* glsl: process uniform images declared bindlessSamuel Pitoiset2017-06-141-0/+20
| | | | | Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* glsl: process uniform samplers declared bindlessSamuel Pitoiset2017-06-141-0/+23
| | | | | Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* mesa: add support for glUniformHandleui64*ARB()Samuel Pitoiset2017-06-141-0/+2
| | | | | | | | | | | | | | Bindless sampler/image handles are represented using 64-bit unsigned integers. The ARB_bindless_texture spec says: "The error INVALID_OPERATION is generated by UniformHandleui64{v}ARB if the sampler or image uniform being updated has the "bound_sampler" or "bound_image" layout qualifier"." Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* st/mesa: don't mark the program as in cache_fallback when there is cache missTimothy Arceri2017-05-201-1/+1
| | | | | | | | | | | | | | | | | | When we fallback currently the gl_program objects are re-allocated. This is likely to change when the i965 cache lands, but for now this fixes a crash when using MESA_GLSL=cache_fb. This env var simulates the fallback path taken when a tgsi cache item doesn't exist due to being evicted previously or some kind of error. Unlike i965 we are always falling back at link time so it's safe to just re-allocate everything. We will be unnecessarily freeing and re-allocate a bunch of things here but it's probably not a huge deal, and can be changed when the i965 code lands. Fixes: 0e9991f957e2 ("glsl: don't reference shader prog data during cache fallback") Reviewed-by: Nicolai Hähnle <[email protected]>
* glsl: make use of glsl_type::is_interface()Samuel Pitoiset2017-04-211-1/+1
| | | | | | Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Samuel Iglesias Gonsálvez <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]>
* glsl: get rid of values_for_type()Samuel Pitoiset2017-04-211-2/+2
| | | | | | | | | | | This function is actually a wrapper for component_slots() and it always returns 1 (or N) for samplers. Since component_slots() now return 1 for samplers, it can go. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Timothy Arceri <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* glsl: don't run the GLSL pre-processor when we are skipping compilationTimothy Arceri2017-04-151-0/+10
| | | | | | | | | | | | | | | | | | | | This moves the hashing of shader source for the cache lookup to before the preprocessor. In our experience, shaders are unlikely to hash the same after preprocessing if they didn't hash the same before, so we can skip preprocessing for cache hits. Improves Deus Ex start-up times with a warm cache from ~30 seconds to ~22 seconds. Also fixes the leaking of state. V2: fix indentation v3: add the value of MESA_EXTENSION_OVERRIDE to the hash of the shader. Tested-by (v2): Grazvydas Ignotas <[email protected]> Tested-by: Dieter Nützel <[email protected]> Reviewed-by: Eric Anholt <[email protected]>
* glsl: delay optimisations on individual shaders when cache is availableTimothy Arceri2017-04-151-1/+1
| | | | | | | | | | | | | | | | | | | | | | Due to a max limit of 65,536 entries on the index table that we use to decide if we can skip compiling individual shaders, it is very likely we will have collisions. To avoid doing too much work when the linked program may be in the cache this patch delays calling the optimisations until link time. Improves cold cache start-up times on Deus Ex by ~20 seconds. When deleting the cache index to simulate a worst case scenario of collisions in the index, warm cache start-up time improves by ~45 seconds. V2: fix indentation, make sure to call optimisations on cache fallback, make sure optimisations get called for XFB. Tested-by: Grazvydas Ignotas <[email protected]> Tested-by: Dieter Nützel <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* glsl: Fix blob memory leakBartosz Tomczyk2017-04-041-1/+1
| | | | Reviewed-by: Timothy Arceri <[email protected]>
* glsl, st/shader_cache: check the whole sha1 for zeroGrazvydas Ignotas2017-03-271-1/+2
| | | | | | | | | | | | The checks were only looking at the first byte, while the intention seems to be to check if the whole sha1 is zero. This prevented all shaders with first byte zero in their sha1 from being saved. This shaves around a second from Deus Ex load time on a hot cache. Signed-off-by: Grazvydas Ignotas <[email protected]> Reviewed-by: Timothy Arceri <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]>
* glsl/shader_cache: restore evicted shader keysGrazvydas Ignotas2017-03-271-0/+17
| | | | | | | | | | | | | | | | | Even though the programs themselves stay in cache and are loaded, the shader keys can be evicted separately. If that happens, unnecessary compiles are caused that waste time, and no matter how many times the program is re-run, performance never recovers to the levels of first hot cache run. To deal with this, we need to refresh the shader keys of shaders that were recompiled. An easy way to currently observe this is running Deux Ex, then piglit and Deux Ex again, or deleting just the cache index. The later is causing over a minute of lost time on all later Deux Ex runs, with this patch it returns to normal after 1 run. Signed-off-by: Grazvydas Ignotas <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* util/disk_cache: use a helper to compute cache keysGrazvydas Ignotas2017-03-211-1/+1
| | | | | | | | | | This will allow to hash additional data into the cache keys or even change the hashing algorithm easily, should we decide to do so. v2: don't try to compute key (and crash) if cache is disabled Signed-off-by: Grazvydas Ignotas <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* util/sha1: drop _mesa_sha1_{update, format} return typeEmil Velikov2017-03-151-7/+8
| | | | | | | | | Unused/unchecked by any of the callers. v2: Fix the glsl cases that have crept in since v1 Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Grazvydas Ignotas <[email protected]>
* glsl: don't use ralloc for blob creationTimothy Arceri2017-03-131-2/+2
| | | | | | There is no need to use ralloc here. Reviewed-by: Marek Olšák <[email protected]>
* st/mesa/glsl: build string of dri options and use as input to building sha ↵Timothy Arceri2017-02-171-0/+6
| | | | | | for shaders Reviewed-by: Nicolai Hähnle <[email protected]>
* glsl: reserve parameter storage on cache restoreTimothy Arceri2017-02-171-0/+1
| | | | | | | Since we know how big the list will be we can allocate the storage upfront. Reviewed-by: Nicolai Hähnle <[email protected]>
* glsl: don't try to load/store buffer object values in the cacheTimothy Arceri2017-02-171-2/+11
| | | | | | Also add an assert to catch buffer overflows. Reviewed-by: Nicolai Hähnle <[email protected]>
* mesa/glsl: add cache_fallback flag to gl_shader_program_dataTimothy Arceri2017-02-171-1/+1
| | | | | | | | | | | This will allow us to skip certain things when falling back to a full recompile on a cache miss such as avoiding reinitialising uniforms. In this change we use it to avoid reading the program metadata from the cache and skipping linking during a fallback. Reviewed-by: Nicolai Hähnle <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-02 05:18:09 +0000