Commit Graph

4549 Commits

Author SHA1 Message Date
DrChat 2a908eb86d [SPIR-V] Handle 32-bit normalized unsigned formats 2018-02-20 16:25:37 -06:00
DrChat b3de604f9e [SPIR-V] Fix offsetof on Travis 2018-02-19 19:09:06 -06:00
DrChat eb6ce26d43 Disable warning C4996 on spirv-tools 2018-02-19 12:58:17 -06:00
DrChat b5127cf924 [SPIR-V] Remove is_signed/is_integer check on float formats (they're ignored) 2018-02-19 12:53:53 -06:00
DrChat a3cf3a7694 [SPIR-V] Don't bother setting up vertex bindings if they aren't used. 2018-02-19 12:45:17 -06:00
DrChat be687f5f27 [SPIR-V] Size the vertex bindings input based on actual number of bindings 2018-02-19 11:28:22 -06:00
DrChat 16cf9883ca [Vulkan] Fix a couple validation errors 2018-02-19 11:19:24 -06:00
DrChat 103ecbab7e [SPIR-V] 32 [u]int formats 2018-02-19 11:01:45 -06:00
DrChat ff6d306728 [SPIR-V] Handle 16/16, 16/16/16/16 (float and norm) and 32f 2018-02-19 08:11:31 -06:00
DrChat 5ca81f4c9f Update spirv-tools 2018-02-19 07:54:37 -06:00
DrChat 6fb176528b [SPIR-V] Don't group together unimplemented fetches 2018-02-19 00:23:15 -06:00
DrChat 9479444b3e Update spirv-tools and spirv-headers submodules 2018-02-19 00:20:30 -06:00
DrChat 081c142c4c [SPIR-V] Convert 8888, 32, 2/10/10/10 2018-02-19 00:14:38 -06:00
DrChat fe9c83c15b [Vulkan] First working version of dynamic vfetch! 2018-02-18 23:22:02 -06:00
DrChat 6bd25940b9 [SPIR-V] Update SPIRV-Tools 2018-02-18 22:20:52 -06:00
DrChat 4c766dcc61 [Vulkan] BufferCache: Fix a few problems 2018-02-18 20:54:29 -06:00
DrChat f3f1a7fe42 [SPIR-V] Update glslang
WIP new vertex fetching
2018-02-18 20:13:38 -06:00
DrChat 4e1a55f585 [Vulkan] Bind the vertex set to DS slot 2 2018-02-18 18:50:53 -06:00
DrChat 9463732709 [Vulkan] BufferCache::PrepareVertexSet 2018-02-18 18:48:23 -06:00
DrChat 4be725c4b4 [Vulkan] Create/Free Vertex Descriptor Pool 2018-02-18 18:26:28 -06:00
DrChat af406df793 [Vulkan] Move the constant descriptor set to its own setup/destroy functions 2018-02-18 18:11:52 -06:00
DrChat 300656f788 Merge branch 'master' into vk_vfetch 2018-02-18 17:02:59 -06:00
DrChat 09d16f4b7b [Base] Copy up to alignment boundary in aligned swap boundaries and then use aligned load/store 2018-02-18 17:02:26 -06:00
DrChat 79308b477f Merge branch 'vtx_cache' 2018-02-18 16:49:32 -06:00
DrChat 71e5e967ec [Vulkan] Disable vertex buffer cache expansion for now. 2018-02-18 16:49:15 -06:00
DrChat 7fa27343c7 [Vulkan] Declare feature samplerAnisotropy 2018-02-18 16:23:29 -06:00
DrChat 05192e3c0f [Vulkan] Prevent negative scissor offsets 2018-02-18 15:14:34 -06:00
DrChat 1a113fb82b [Vulkan] Fix cube texture uploads 2018-02-18 14:59:34 -06:00
DrChat ffef59aaa5 [Base] RingBuffer: Add ReadAndSwap instead of piggybacking off of Read 2018-02-18 12:18:20 -06:00
DrChat 705fd7594a [Docs] Add more notes on the single-page offset 2018-02-17 10:33:34 -06:00
Dr. Chat 3d1c2a6c9a [GPU] Update vizquery documentation 2018-02-16 22:03:33 -06:00
DrChat 766974c30a [Kernel] Document XMP[G/S]etPlaybackController 2018-02-16 18:59:45 -06:00
DrChat 22929182e6 [GPU] Update the docs of a few registers 2018-02-16 18:33:01 -06:00
DrChat 1dba231049 [Vulkan] Handle primitive reset index on copy 2018-02-16 18:31:43 -06:00
DrChat 776b5eff36 [Vulkan] BufferCache: Add VertexBuffer struct 2018-02-16 16:51:32 -06:00
DrChat db53559611 Merge branch 'master' 2018-02-16 16:34:17 -06:00
DrChat 43f9cebb0c [JIT] Rewrite CNTLZ and remove one branch 2018-02-16 12:51:39 -06:00
DrChat e4bc596887 [Base] Faster copy-and-swap routines 2018-02-15 21:14:52 -06:00
DrChat 30e0a258e4 [Vulkan] Buffers typically aren't aligned. 2018-02-15 17:05:54 -06:00
DrChat abd21e5034 [Vulkan] Update the loader to SDK 1.0.68.0 2018-02-15 11:14:47 -06:00
DrChat ec501a32c6 [JIT] Apply the same ZeroExtend/Truncate to rlwimix and rlwnmx 2018-02-14 22:05:54 -06:00
DrChat 0cc2b0fc9f [Core] Protect the last 64kb of memory 2018-02-14 21:58:05 -06:00
DrChat 310606d0f6 Merge branch 'master' into vk_vfetch 2018-02-14 20:31:31 -06:00
DrChat b17d6f5088 [Base] Enable aligned copy and swap routines 2018-02-14 20:28:34 -06:00
DrChat e54c24e150 [JIT] New opcodes: OPCODE_LOAD_OFFSET and OPCODE_STORE_OFFSET
These take full advantage of x86 addressing, and eliminate extra add operations.
2018-02-14 16:26:49 -06:00
DrChat 1de598e4ce [JIT] Comment the offsets in PPCContext 2018-02-14 16:22:14 -06:00
DrChat 7818cdab60 [CPU] Check for RawModule memory allocation success 2018-02-14 16:21:16 -06:00
DrChat 02b5a07bc9 [JIT] rlwinmx: Use Truncate/ZeroExtend instead of And 0xFFFFFFFF 2018-02-14 13:50:57 -06:00
DrChat bbafcc089d [Travis] Fix Travis 2018-02-14 07:44:21 -06:00
DrChat f203380040 [Travis] Update libvulkan version 2018-02-13 21:03:53 -06:00