summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* radeon/r600_query.c: Minor style fixEdward O'Callaghan2016-03-271-1/+1
| | | | | | Signed-off-by: Edward O'Callaghan <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]> Signed-off-by: Marek Olšák <[email protected]>
* virgl: drop next shader property for now.Dave Airlie2016-03-261-0/+1
| | | | Signed-off-by: Dave Airlie <[email protected]>
* glsl: reduce buffer block duplicationTimothy Arceri2016-03-266-82/+57
| | | | | | | | | | | | | This reduces some of the craziness required for handling buffer blocks. The problem is each shader stage holds its own information about a block in memory, we were copying that information to a program wide list but the per stage information remained meaning when a binding was updated we needed to update all versions of it. This changes the per stage blocks to instead point to a single version of the block information in the program list. Acked-by: Kenneth Graunke <[email protected]>
* st/xa: emit sampler view declarations in shadersBrian Paul2016-03-251-0/+19
| | | | | | | Fixes recent regressions with the VMware gallium driver. Reviewed-by: Charmaine Lee <[email protected]> Tested-by: Charmaine Lee <[email protected]>
* swr: [rasterizer jitter] Fix MASKLOADD AVX prototype (float -> i32)Tim Rowley2016-03-251-1/+1
|
* swr: [rasterizer core] NUMA optimizations...Tim Rowley2016-03-255-65/+105
| | | | | - Affinitize hot-tile memory to specific NUMA nodes. - Only do BE work for macrotiles assoicated with the numa node
* swr: [rasterizer jitter] Fix logic bug for alpha-to-coverage.Tim Rowley2016-03-251-2/+11
|
* swr: [rasterizer core] Fix Compute workitem retirementTim Rowley2016-03-254-31/+22
|
* swr: [rasterizer core] Cleanup state ring arena after last draw that ↵Tim Rowley2016-03-253-2/+14
| | | | | | references it completes Rather than waiting for the API thread to re-use it.
* swr: [rasterizer jitter] add missing include for llvm jiteventsTim Rowley2016-03-251-0/+4
|
* swr: [rasterizer core] Reduce Arena blocksize to 128KB (from 1MB).Tim Rowley2016-03-251-3/+7
| | | | | With global allocator this doesn't seem to affect performance at all. Overall memory consumption drops by up to 85%.
* swr: [rasterizer core] One last pass at Arena optimizationsTim Rowley2016-03-251-15/+15
|
* swr: [rasterizer core] CachedArena optimizationsTim Rowley2016-03-253-210/+161
| | | | | | Reduce list traversal during Alloc and Free. Add ability to have multiple lists based on alloc size (not used for now)
* swr: [rasterizer jitter] support llvm-svnTim Rowley2016-03-256-12/+37
|
* swr: [rasterizer core] Globally cache allocated arena blocks for fast ↵Tim Rowley2016-03-258-46/+168
| | | | re-allocation.
* swr: [rasterizer] more arena workTim Rowley2016-03-255-10/+110
|
* swr: [rasterizer core] Add clipping against user clip distances in the ↵Tim Rowley2016-03-251-2/+12
| | | | NullPS backend.
* swr: [rasterizer core] Arena optimizations - preparing for global allocator.Tim Rowley2016-03-255-187/+131
|
* swr: [rasterizer core] Reset DrawContext arena at end of draw rather than ↵Tim Rowley2016-03-252-21/+4
| | | | | | | upon reclaim of DC Keeps overall memory consumption lower. Also, remove unused knobs.
* swr: [rasterizer core] Add clipping of user clip planes in clipper.Tim Rowley2016-03-251-0/+86
|
* swr: [rasterizer] Reduce max in-flight draws to 96 (by default)Tim Rowley2016-03-251-1/+1
|
* swr: [rasterizer] Fix run-time check assertsTim Rowley2016-03-252-8/+7
| | | | | One innocuous (uninitialized variable), and one not so innocuous (stack corruption).
* swr: [rasterizer jitter] signed immediate builderTim Rowley2016-03-252-0/+8
|
* swr: [rasterizer common] changes for cygwinTim Rowley2016-03-251-1/+4
|
* swr: [rasterizer] code styling and update copyrightsTim Rowley2016-03-2510-328/+328
|
* swr: [rasterizer core] Guard against enquing work to invalid hot tilesTim Rowley2016-03-251-0/+5
|
* swr: [rasterizer] Stop setting viewport size to larger than hottile arrayTim Rowley2016-03-251-0/+6
| | | | Guard against enquing work to invalid tiles
* swr: [rasterizer] Discard work + misc fixesTim Rowley2016-03-259-28/+119
|
* swr: [rasterizer] remove use of BYTE typeTim Rowley2016-03-2515-74/+65
|
* swr: [rasterizer core] Fix crash that can occur when switching contextsTim Rowley2016-03-253-1/+22
|
* swr: [rasterizer] remove unused knobTim Rowley2016-03-251-8/+0
|
* swr: [rasterizer core] subcontext reworkTim Rowley2016-03-257-52/+61
|
* swr: [rasterizer common] add _simd_s[rl]lv_epi32Tim Rowley2016-03-251-0/+115
|
* swr: [rasterizer core] Alleviate potential stack overflow for 32bit buildsTim Rowley2016-03-253-14/+20
| | | | Move large stack allocations in the GS and clipper into thread local storage.
* swr: [rasterizer] remove use of UCHAR and UINT64 typesTim Rowley2016-03-255-8/+6
|
* swr: [rasterizer] remove use of FLOAT typeTim Rowley2016-03-253-5/+4
|
* swr: [rasterizer] Fix Coverity issues reported by Mesa developers.Tim Rowley2016-03-2510-199/+205
|
* swr: [rasterizer] add debug/perf category to knobsTim Rowley2016-03-251-12/+37
|
* swr: [rasterizer core] don't assume linux is 64-bitTim Rowley2016-03-251-2/+1
|
* swr: [rasterizer common] remove old unused win32 typesTim Rowley2016-03-251-6/+0
|
* swr: [rasterizer jitter] vpermps supportTim Rowley2016-03-254-1/+84
|
* swr: [rasterizer] Add rdtsc buckets support for shadersTim Rowley2016-03-256-3/+75
| | | | | | | | | | | Pass pointer to core buckets mgr back to sim layer. Add support for RDTSC_START/RDTSC_STOP macros in the builder. Each unique shader now has a unique bucket associated with it, enabling more detailed reporting at the shader level. Currently due to some llvm issue with thread local storage, 64bit runs require single threaded mode.
* swr: [rasterizer core] backend reorganizationTim Rowley2016-03-258-256/+276
|
* swr: [rasterizer core] store blend output in temporary instead of PS output.Tim Rowley2016-03-251-5/+10
| | | | Fixes additive blend problem with MSAA
* swr: [rasterizer core] Move InitializeHotTiles and corresponding clear code ↵Tim Rowley2016-03-253-254/+258
| | | | out of threads.cpp.
* swr: [rasterizer jitter] Cleanup use of types inside of Builder.Tim Rowley2016-03-254-63/+75
| | | | | Also, cached the simd width since we don't have to keep querying the JitManager for it.
* swr: [rasterizer jitter] Fix type mismatch on select args for SCATTERPSTim Rowley2016-03-251-2/+4
|
* swr: [rasterizer core] fix rasterizing multisampling with scissor enabledTim Rowley2016-03-251-42/+28
| | | | We were not evaluating the scissor edge equations at sample positions.
* swr: [rasterizer core] RingBuffer class for DC/DSTim Rowley2016-03-256-136/+180
| | | | | | | | | | | | | Use head/tail ring buffer indices for thread synchronization. 1. SwrWaitForIdle loops until ring is empty. (head == tail) 2. GetDrawContext waits until ring is not full. (head - tail) == Ring Size 3. Draw enqueues by incrementing head. 4. Last worker thread to move past a DC dequeues by incrementing tail. Todo: To reduce contention we can cache the tail in the API thread. For example, if you know you have 64 free entries in the ring then you don't need to keep checking the tail until you used those 64 entries.
* swr: [rasterizer] switch assert uses to SWR_ASSERTTim Rowley2016-03-251-5/+5
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-23 20:04:55 +0000