| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2010-04-08 | util: is_array/mixed/etc is only meaningful for plain formats. | José Fonseca | |
| 2010-04-08 | util: Keep const keyword when unpacking formats. | José Fonseca | |
| 2010-04-08 | util: Reorder the code generated function so that they are grouped by format. | José Fonseca | |
| Makes it easier to copy and paste. | |||
| 2010-04-08 | draw llvm: hook up the generated function into the draw elts path | Zack Rusin | |
| we were only using the jited function in the linear case, now drawelts correctly uses the same path. it results in a significant gain in real world apps (openarena went from 23fps to 29fps) | |||
| 2010-04-08 | draw: Fix vertex buffer indexation. | José Fonseca | |
| 2010-04-08 | util: Don't call util_dl_close(library) when util_format_s3tc_init is ↵ | José Fonseca | |
| successful." Otherwise the library will be unloaded and function pointers become invalid. This reverts commit bc2bc0306e4dd8c56bd66a8aabf2433f6689653d. | |||
| 2010-04-08 | util: util_dl_close(library) before exiting util_format_s3tc_init. | Vinson Lee | |
| 2010-04-08 | draw llvm: Remove unnecessary header. | Vinson Lee | |
| 2010-04-07 | draw llvm: highly reduce the compilation times for draw llvm | Zack Rusin | |
| our code resets pipe_vertex_buffer's with different offsets when rendering vbo, meaning that we kept creating insane number of shaders even for simple apps e.g. geartrain had 54 shaders and it was taking almost 27 seconds just to compile them. this patch passes pipe_vertex_buffer's to the jit function and lets it to the stride/buffer_offset computation at run time. the slowdown at runtime is largely unnoticable but the we go from 54 shaders to 3, and from 27 seconds to less than 1. | |||
| 2010-04-07 | gallivm: Get the format translation logic write. | José Fonseca | |
| 2010-04-07 | util: Support fixed formats conversion. | José Fonseca | |
| 2010-04-07 | util: Add copyright header to u_half.h | José Fonseca | |
| 2010-04-07 | util: Remove the half typedef from p_compiler.h. | José Fonseca | |
| Unnecessary, and doesn't even guarantee size. | |||
| 2010-04-07 | util: Use stubs for the dynamically loaded S3TC functions. | José Fonseca | |
| Loosely based on Luca Barbieri's commit 52e9b990a192a9329006d5f7dd2ac222effea5a5. | |||
| 2010-04-07 | draw: Fix MSVC build (snprintf->util_snprintf). | José Fonseca | |
| 2010-04-07 | util: Remove u_tile.c YCbCr's. | José Fonseca | |
| Superseded by u_format_yuv.c. Also PIPE_FORMAT_YUYV's interpretation was inconsistent: it was being interpreted as VYUY. | |||
| 2010-04-07 | util: Implement YUV and subsampled RGB format conversion. | José Fonseca | |
| 2010-04-07 | draw: init draw->pt.middle.general = NULL just to be safe | Brian Paul | |
| This field should be null from the initial calloc() of the draw context, but let's be safe (and improve understanding of the code). | |||
| 2010-04-07 | llvmpipe: Fix USE_DRAW_LLVM build. Use lp_build_engine. | José Fonseca | |
| 2010-04-07 | draw llvm: Remove unnecessary headers. | Vinson Lee | |
| 2010-04-07 | gallium: Add files to Makefile. | Vinson Lee | |
| This was missed in commit ae69f9fbf0a1aab7186e5b644085a5fe5aea99af. New files from the gallium_draw_llvm merge were added to SCons but not make. | |||
| 2010-04-06 | draw llvm: disable printing of the LLVM ir | Zack Rusin | |
| 2010-04-06 | gallium/draw: add additional DRAW_LLVM check | Brian Paul | |
| Fixes build breakage. | |||
| 2010-04-06 | Merge branch 'gallium_draw_llvm' | Zack Rusin | |
| 2010-04-06 | gallivm: Fallback to calling util_format_description::fetch_float for any ↵ | José Fonseca | |
| format we can't code LLVM IR directly. | |||
| 2010-04-06 | gallivm: Move the global LLVM objects (module, engine, provider, target) ↵ | José Fonseca | |
| into here. | |||
| 2010-04-06 | draw llvm: disable debugging output | Zack Rusin | |
| 2010-04-06 | draw llvm: implement simple pipeline caching using variants | Zack Rusin | |
| 2010-04-06 | draw llvm: fix iteration for larger vertex arrays | Zack Rusin | |
| we were trying to store the outputs starting at the same offset we were using for the input arrays, which was writing beyond the end of the output array. | |||
| 2010-04-06 | draw llvm: iterate with the correct stop over the outputs | Zack Rusin | |
| it's whatever the var step is (4 usually) not an unconditional 1 | |||
| 2010-04-06 | util: Respect destination stride in pipe_get_tile_swizzle(). | Michal Krol | |
| 2010-04-06 | draw llvm: fix draw arrays | Zack Rusin | |
| we don't index within the outputs but only within the inputs | |||
| 2010-04-06 | util: Make half float lookup tables constant. | José Fonseca | |
| 2010-04-06 | util: Remove the hand written SRGB format support code. | José Fonseca | |
| 2010-04-06 | util: Support all SRGB formats. | José Fonseca | |
| Based on code from Brian Paul and Michal Krol. | |||
| 2010-04-06 | draw llvm: fix loop iteration and vertex header offsets | Zack Rusin | |
| the loop was doing a NE comparison which we could have skipped if the prim was triangles (3 verts) and our step was 4 verts. also fix offsets in conversion to aos. | |||
| 2010-04-05 | draw llvm: when generating the vertex_header struct adjust its name | Zack Rusin | |
| change the name to not clash and accuretly represent the number of inputs we store in the data member | |||
| 2010-04-05 | util: added util_dl_error() | Brian Paul | |
| 2010-04-04 | Merge branch '7.8' | Ian Romanick | |
| 2010-04-03 | util: Use GCC atomic bultins on GCC 4.1 and higher only. | Vinson Lee | |
| 2010-04-03 | util: Revert unsolicited, untested, unreviewed, and broken changes to format ↵ | José Fonseca | |
| support. Not all is bad, but I'm afraid I'll have to throw the baby with the water given they are all tied to together. | |||
| 2010-04-02 | draw llvm: fix storing of outputs for the rest of the pipeline | Zack Rusin | |
| there's no good way of aligning the output's, and since the vertex_header is variable sized in the first place we need to extract elements from a vector and store them individually into an array. this gets the basic examples working again | |||
| 2010-04-02 | draw llvm: fix translation of formats with variable components | Zack Rusin | |
| 2010-04-02 | gallivm: Add fallthrough comment at end of case statement. | Vinson Lee | |
| The code is correct. Tell Coverity that the fallthrough case is intentional. | |||
| 2010-04-02 | gallium/util: print \n after DXTn printf | Luca Barbieri | |
| 2010-04-01 | gallium/util: remove extraneous semicolon | Brian Paul | |
| 2010-04-02 | gallium/util: revert util_format_init addition | Luca Barbieri | |
| Putting calls to util_format_init all over the codebase is infeasible. Instead, half float tables are pregenerated, and the s3tc library is loaded on demand. I believe this is a solution that combines performance, cleanliness, flexibility and portability. | |||
| 2010-04-02 | gallium/util: add util_format_is_supported to check for pack/unpack | Luca Barbieri | |
| This improves the code by making it more readable, and removes special knowledge of S3TC and other formats from softpipe. | |||
| 2010-04-02 | gallium/util: load s3tc on demand | Luca Barbieri | |
| This changes the S3TC function pointers to be initialized to stubs that load the S3TC library and then delegate to the real functions. If the S3TC library fails to load, the function pointers are replaced with a "nop" function. The code is also changed to attempt to load the library only one time.c Note that unlike checking for a flag, this method has no performance cost at all. The use of the "nop" functions also allows to avoid most checks, that are only preserved when the function does non-trivial work. | |||
| 2010-04-02 | gallium/util: pregenerate half float tables | Luca Barbieri | |
| This solution avoids the issue of how to run the initializers and also allows those pages (and the parts of them in processor caches) to be shared between multiple processes. The drawback is slightly higher library size. | |||
 |
index : android-x86-mesa.git | |
| Androïd/x86 port of Mesa drivers | Hugues Hiegel |
| summaryrefslogtreecommitdiff |