Age | Commit message (Collapse) | Author | |
---|---|---|---|
2008-02-13 | Remove O(n^2) debugging code from non-debug path of dri_bufmgr_fake. | Eric Anholt | |
2008-02-12 | [intel] Remove cached reloc data buffer now that it's not a BO. | Eric Anholt | |
It's not worth the extra effort to avoid a free/malloc, and we'd rather auto-size the reloc data buffer at some point so we don't need to have max_relocs. | |||
2008-02-12 | [intel] Fix type of some more flags variables for uint64_t flags. | Eric Anholt | |
Harmless since we don't yet have any bits above 31 for flags. | |||
2008-02-12 | [intel] Note when BO map/unmap fail with TTM. | Eric Anholt | |
2008-02-12 | [intel] Fix INTEL_DEBUG=bufmgr after relocation interface fixups. | Eric Anholt | |
2008-02-12 | [965] Remove stale brw_state_cache.c comment and function export. | Eric Anholt | |
2008-02-12 | nouveau: ddx versioning changed | Ben Skeggs | |
2008-02-08 | [directfb] Added RGB444 and RGB555. | Claudio Ciccani | |
Also added color expansion for RGB16, ARGB1555 and ARGB4444. | |||
2008-02-07 | [965] Flush icache on new batch, not just new context. | Eric Anholt | |
This is required since our buffer manager may now move our instruction-containing buffers at any batchbuffer emit. | |||
2008-02-06 | [915] Fix COS function using same plan as SIN. | Eric Anholt | |
The previous COS function failed badly outside of [-pi/2, pi/2]. | |||
2008-02-06 | [915] Use a quartic term to improve the accuracy of SIN results. | Eric Anholt | |
This is described in the link in the comment, and is the same technique that r300 uses. | |||
2008-02-06 | [915] Fix fp SIN function, and use a quadratic approximation instead of Taylor. | Eric Anholt | |
The Taylor series notably fails at producing sin(pi) == 0, which leads to discontinuity every 2*pi. The quadratic gets us sin(pi) == 0 behavior, at the expense of going from 2.4% THD with working Taylor series to 3.8% THD (easily seen on comparative graphs of the two). However, our previous implementation was producing sin(pi) < -1 and worse, so any reasonable approximation is an improvement. This also fixes the repeating behavior, where the previous implementation would repeat sin(x) for x>pi as sin(x % pi) and the opposite for x < -pi. | |||
2008-02-05 | [965] Bug 14314: assertion failure with with !AIGLX and depth=24 visual. | Eric Anholt | |
2008-02-05 | [965] Fix TTM relocation caching overzealousness. | Eric Anholt | |
The failure mode that was a available was: reloc 1 -> target_buf exec: PRESUMED_OFFSET wrong, buffer migrates, r1 entry updated. reloc 2 -> target_buf exec: suppose buffer migrates again. PRESUMED_OFFSET wrong. r2 entry updated. reloc 1 -> target_buf exec: suppose buffer doesn't migrate. PRESUMED_OFFSET right. no relocations performed. r1 has stale pointer at original location. Failures were reported with OGLconform's VBO test and SPECviewperf90, though I haven't confirmed that this fixes it. | |||
2008-02-05 | i965: adjust the byte order of clear color. fix #14165 | Xiang, Haihao | |
2008-02-04 | Replace usage of DRM_BO_FLAG_MEM_TT in intel_regions.c with local/cached. | Eric Anholt | |
In addition to potentially binding when it was about to be mapped anyway, failure to use CACHED_MAPPED means eating a full wbinvd on validate. Thanks to airlied for catching this. | |||
2008-02-04 | [965] Convert brw_draw_upload to managing dri_bos, not gl_buffer_objects. | Eric Anholt | |
This helps us avoid a bunch of mess with gl_client_arrays that we filled with unused data and confused readers. | |||
2008-02-04 | [965] Remove dead structure in brw_draw_upload.c. | Eric Anholt | |
2008-02-04 | [965] Move temporary vbo array storage into the function using it. | Eric Anholt | |
2008-02-04 | [965] Remove dead brw_vertex_element members. | Eric Anholt | |
2008-02-04 | [965] Add a wrapper around interleaved copy_array_to_vbo_array for profiling. | Eric Anholt | |
If compiled with optimization, it shouldn't appear at all, and helps me for now. | |||
2008-02-04 | [965] Avoid overloaded use of the term 'input' for clarity. | Eric Anholt | |
2008-02-04 | [965] Replace VEP/VBP state structures with inline batch emits. | Eric Anholt | |
2008-02-04 | r300: fix isosurf on rs690 | Dave Airlie | |
2008-02-03 | i965: fix potential NULL pointer dereference. The third region | Xiang, Haihao | |
isn't created at all for 965 | |||
2008-02-01 | [965] Fix indentation. | Eric Anholt | |
2008-02-01 | Revert "intel: don't apply the relocation optimization if a target" | Eric Anholt | |
This reverts commit e2cb905bc6e23eaafaeeb2abdc9480e70959ee3f. It was a reversion of an optimization hidden as otherwise. pre_target_buf_handle was always NULL, so the optimization was never enabled, rather than fixing the important optimization (resulting in 25-50% performance loss). | |||
2008-02-01 | [965] Replace XXX comment about constant swizzle with an assert. | Eric Anholt | |
2008-02-01 | [965] Fix some indentation in brw_vs_tnl.c. | Eric Anholt | |
2008-02-01 | [intel] fix for previous fix | Zou Nan hai | |
2008-02-01 | [intel] use _mesa_copy_rect for upload compressed texture, | Zou Nan hai | |
this fix bad texture issue in some games(UT and quake). | |||
2008-02-01 | i965: Don't emit state if fall back to software rendering. fix #14116 | Xiang, Haihao | |
2008-01-31 | [i965] renable regative rhw test | Zou Nan hai | |
2008-01-31 | intel: don't apply the relocation optimization if a target | Xiang, Haihao | |
buffer is used for a relocatee in the former relocation process then another target buffer is used for this relocatee at the same offset in the current relocation process. | |||
2008-01-29 | Add new RV380 pci id | Alex Deucher | |
bug 14289 | |||
2008-01-29 | i965: new integrated graphics chipset support | Xiang, Haihao | |
2008-01-27 | r300: add initial rs690 support to Mesa | Dave Airlie | |
The rs690 has an rs4xx style vertex-shader less 3D engine. It uses the new r500 output engine though. It also needs a new drm with rs690 support, which is just getting cleaned up. | |||
2008-01-25 | i965: valid message length includes message header. | Xiang, Haihao | |
2008-01-25 | i965: re-define the type of reg.loopcount. | Xiang, Haihao | |
avoid some issues such that 1 + (-2) gets a big positive value. | |||
2008-01-24 | Bufmgr cleanup from intel-batchbuffer branch of 2d driver. | Eric Anholt | |
2008-01-24 | Clean up comments/dead code from relocation buffer change. | Eric Anholt | |
2008-01-24 | i915: move to using copy from user for relocations | Dave Airlie | |
2008-01-22 | Fix some issues with glDrawBuffer(GL_NONE), bug 14198 | Brian | |
Set _ColorDrawBuffers[0] = NULL if no renderbuffers enabled. Check that _ColorDrawBuffers[0] is non-null before dereferencing in a few places. | |||
2008-01-22 | [intel] Clean up references to screen buffer metrics. | Kristian Høgsberg | |
The screen wide info such as pitch and cpp are obsoleted by the FBO changes, so clean up the last few references to those, except for setting up the legacy screen regions. | |||
2008-01-19 | [965] Fix WM unit cache keying that broke line stipple and polygon offset. | Eric Anholt | |
2008-01-18 | [intel] Fix memory leak with fake bufmgr. | Eric Anholt | |
2008-01-18 | [965] Do a little bit rotation in state hash to reduce collisions. | Eric Anholt | |
This was around 3% improvement in OA. | |||
2008-01-18 | [intel] Use a static array for the validation list instead of a linked list. | Eric Anholt | |
Around 10% of a CPU was being wasted to create the linked list which we threw out immediately after passing it to the kernel. | |||
2008-01-17 | [intel] Make the no_rast option be standard driconf instead of INTEL_NO_RAST. | Eric Anholt | |
2008-01-17 | [i915] Fix driver from cliprects changes, and clean up state emission. | Eric Anholt | |
The fix for pageflipping with cliprects ended up causing a batch flush at an inopportune time, which is fixed by moving it up. Additionally, the recovery code for handling batch wraps at bad times is replaced by just checking for the space up front, and using a no_batch_wrap assert like on 965 to make sure that we weren't wrong about how much space that was. |