summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* intel: combine short memcpy using a temporary allocated bufferChris Wilson2011-02-213-38/+27
| | | | | | | | Using a temporary buffer for large discontiguous uploads into the common buffer and a single buffered upload is faster than performing the discontiguous copies through a mapping into the GTT. Signed-off-by: Chris Wilson <[email protected]>
* i965: upload normal arrays as interleavedChris Wilson2011-02-211-30/+72
| | | | | | | Upload the non-vbo arrays into a single interleaved buffer object, and so need to just emit a single vertex buffer relocation. Signed-off-by: Chris Wilson <[email protected]>
* i965: interleaved vboChris Wilson2011-02-211-12/+27
| | | | | | | If the user passed in several arrays interleaved in the same vbo, only emit a single vertex buffer and relocation. Signed-off-by: Chris Wilson <[email protected]>
* i965: emit one vb packet per vboChris Wilson2011-02-213-77/+83
| | | | | | | Track reuse of the vertex buffer objects and so minimise the number of vertex buffers used by the hardware (and their relocations). Signed-off-by: Chris Wilson <[email protected]>
* i965: upload transient indices into the same discontiguous bufferChris Wilson2011-02-212-13/+8
| | | | | | | | As we now pack the indices into a common upload buffer, we can reuse a single CMD_INDEX_BUFFER packet and translate each invocation with a start vertex offset. Signed-off-by: Chris Wilson <[email protected]>
* i965: suppress repeat-emission of identical vertex elementsChris Wilson2011-02-211-3/+2
| | | | Signed-off-by: Chris Wilson <[email protected]>
* i965: Move repeat-instruction-suppression to batchbuffer coreChris Wilson2011-02-219-152/+120
| | | | | | | | Move the tracking of the last emitted instructions into the core batchbuffer routines and take advantage of the shadow batch copy to avoid extra memory allocations and copies. Signed-off-by: Chris Wilson <[email protected]>
* intel: use pwrite for batchChris Wilson2011-02-2127-302/+219
| | | | | | | | | | | It's faster. Not only is the memcpy more efficiently performed in the kernel (making up for the system call overhead), but by not using mmap we remove the greater overhead of tracking the vma of every batch. And it means we can read back from the batch buffer without incurring the cost of a uncached read through the GTT. Signed-off-by: Chris Wilson <[email protected]>
* i965: drop state_bo references to batch_boChris Wilson2011-02-219-105/+74
| | | | | | | As we use state relocations and we know that all the state belongs to the same bo, we can drop the multiple references to the same bo. Signed-off-by: Chris Wilson <[email protected]>
* i965: directly write wm state to batchChris Wilson2011-02-211-63/+48
| | | | | | | As we write directly into the batch in system memory, we do not need to write first to the stack (as was to avoid read back through the GTT) Signed-off-by: Chris Wilson <[email protected]>
* i965: write cc straight to batchChris Wilson2011-02-211-48/+46
| | | | | | | As we write directly into the batch in system memory, we do not need to write first to the stack (as was to avoid read back through the GTT) Signed-off-by: Chris Wilson <[email protected]>
* i965: switch gen6 to use its own cc state boChris Wilson2011-02-211-6/+6
| | | | | | | In preparation for a greater change, use the color_calc_state_bo already provisioned for this purpose. Signed-off-by: Chris Wilson <[email protected]>
* intel: Buffered uploadChris Wilson2011-02-215-17/+82
| | | | | | | | | Rather than performing lots of little writes to update the common bo upon each update, write those into a static buffer and flush that when full (or at the end of the batch). Doing so gives a dramatic performance improvement over and above using mmaped access. Signed-off-by: Chris Wilson <[email protected]>
* intel: Replace the bo for a complete updateChris Wilson2011-02-211-6/+13
| | | | | | | Rather than performing a blit to completely overwrite a busy bo, simply discard it and create a new one with the fresh data. Signed-off-by: Chris Wilson <[email protected]>
* i965: Combine vb upload buffer with the general upload bufferChris Wilson2011-02-216-97/+71
| | | | | | | Reuse the new common upload buffer for uploading temporary indices and rebuilt vertex arrays. Signed-off-by: Chris Wilson <[email protected]>
* intel: Pack dynamic draws togetherChris Wilson2011-02-217-20/+88
| | | | | | | | Dynamic arrays have the tendency to be small and so allocating a bo for each one is overkill and we can exploit many efficiency gains by packing them together. Signed-off-by: Chris Wilson <[email protected]>
* intel: Use system memory for DYNAMIC_DRAW source objectsChris Wilson2011-02-211-20/+31
| | | | | | | | | | | | Dynamic draw buffers are used by clients for temporary arrays and for uploading normal vertex arrays. By keeping the data in memory, we can avoid reusing active buffer objects and reallocate them as they are changed. This is important for Sandybridge which can not issue blits within a batch and so ends up flushing the batch upon every update, that is each batch only contains a single draw operation (if using dynamic arrays or regular arrays from system memory). Signed-off-by: Chris Wilson <[email protected]>
* i965: Trim the trailing NOOP from 3DSTATE_INDEX_BUFFERChris Wilson2011-02-211-23/+12
| | | | Signed-off-by: Chris Wilson <[email protected]>
* i965: Fallback on encountering a NULL render bufferChris Wilson2011-02-211-0/+5
| | | | | | | | Following a GPU hang, or other error, the render target is not likely to have an allocated BO and so we must fallback to avoid using it. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=32534 Signed-off-by: Chris Wilson <[email protected]>
* i915g: s/__func__/__FUNCTION__/Vinson Lee2011-02-201-1/+1
|
* i915g: kill remnants of mmapped batchbuffer supportDaniel Vetter2011-02-211-12/+0
| | | | | | We're using bo_subdata. Signed-off-by: Daniel Vetter <[email protected]>
* i915g: Add winsys/i915/sw to scons buildJakob Bornecrantz2011-02-211-0/+4
|
* i915g: Fix void ptr arithJakob Bornecrantz2011-02-211-1/+1
|
* i915g: Add dummy flush_frontbufferJakob Bornecrantz2011-02-211-0/+15
|
* egl_dri2: Return NULL when xcb_get_geometry_reply fails.Chia-I Wu2011-02-201-0/+1
| | | | This should fix bug #33946.
* scons: Add aliases for the llvmpipe unit tests.José Fonseca2011-02-191-2/+7
| | | | | | Now one can simply do scons lp_test_format
* gallivm: Use simple scaling plus casting in more unorm->float cases.José Fonseca2011-02-191-25/+34
|
* glsl: Remove $(PWD) from Makefile in favor of .Kenneth Graunke2011-02-191-1/+1
| | | | Hopefully should fix bug #34468.
* r300g: fix a possible race when counting contextsMarek Olšák2011-02-193-2/+8
| | | | Atomics aren't sufficient here.
* r300g: fix invalid dereference in winsysMarek Olšák2011-02-191-1/+1
| | | | radeon_bo_unref may destroy the buffer, so call it after p_atomic_dec, not before.
* svga: Fix NULL dereference.José Fonseca2011-02-183-7/+9
| | | | Probably introduced with the surface view move from screen to context.
* vbo: add debug code to verify that buffers are unmapped before drawingBrian Paul2011-02-181-0/+25
|
* mesa: MESA_VERBOSE logging for glRead/Draw/CopyPixels, glBlitFramebufferBrian Paul2011-02-183-0/+38
|
* st/mesa: set renderbuffer _BaseFormat in a few placesBrian Paul2011-02-181-0/+2
| | | | NOTE: This is a candidate for the 7.9 and 7.10 branches
* st/mesa: check buffer orientation in blit_copy_pixels()Brian Paul2011-02-181-3/+4
| | | | | Can't invert the region if copying between surfaces with different orientations.
* svga: Ensure pending drawing commands other surface operations are emitted ↵José Fonseca2011-02-188-17/+40
| | | | | | | | | before DMAs. This behavior was last when moving the transfers to the contexts. This fixes several piglit failures, which were reading the color renderbuffer before the draw operations were emitted.
* svga: Cannot use negate or abs on source to dsx/dsy instructions.José Fonseca2011-02-181-1/+19
|
* svga: Ensure SWTNL is created after HWTNL.José Fonseca2011-02-181-5/+4
| | | | | Matches the internal driver layering, and prevents null svga->hwtnl dereferencing from inside the swtnl.
* svga: Ensure LRP's restrictions are observed in all uses.José Fonseca2011-02-181-38/+51
| | | | The dst reg must be a temporary, and not be the same as src0 or src2.
* svga: Preserve src swizzles in submit_op2/3/4.José Fonseca2011-02-181-25/+45
| | | | | Several opcodes require scalar swizzle, and this requirement was being was not being observed when creating temporaries for other reasons.
* r300g: remove tracking whether vertex buffers need to be validatedMarek Olšák2011-02-187-88/+27
| | | | | This was getting hard to maintain and didn't really bring any real benefits. Instead, validate buffers when the vertex array state is dirty.
* st/mesa: fix geometry corruption by always re-binding vertex arraysMarek Olšák2011-02-181-1/+3
| | | | | | | | | | | This is a temporary workaround. It fixes sauerbrauten with shaders enabled. I guess we might be changing vertex attribs somewhere and not updating the appropriate dirty flags, therefore we can't rely on them for now. Or maybe we need to make this state dependent on some other flags too. More info: https://bugs.freedesktop.org/show_bug.cgi?id=34378
* svga: Disable surface cache for texturesJakob Bornecrantz2011-02-181-1/+2
| | | | Signed-off-by: Jakob Bornecrantz <[email protected]>
* svga: Describe svga_sampler_views for refcnt debuggingJakob Bornecrantz2011-02-182-1/+21
| | | | Signed-off-by: Jakob Bornecrantz <[email protected]>
* svga: Make sure that refcnt debugger gets the correct backtrace for createJakob Bornecrantz2011-02-182-0/+9
| | | | Signed-off-by: Jakob Bornecrantz <[email protected]>
* util: Make refcnt and symbol debuggers work on windowsJakob Bornecrantz2011-02-182-2/+21
| | | | Signed-off-by: Jakob Bornecrantz <[email protected]>
* Point to bugs.freedesktop.org rather than bugzilla.freedesktop.orgCyril Brulebois2011-02-181-1/+1
| | | | | | Suggested by a freedesktop.org admin. Signed-off-by: Cyril Brulebois <[email protected]>
* u_vbuf_mgr: initialize flag indicating that buffers have been updatedMarek Olšák2011-02-181-1/+1
| | | | | This fixes r300g errors: r300: Cannot get a relocation in radeon_drm_cs_write_reloc.
* gallium/svga: Fix unnecessary swtnl fallbacksThomas Hellstrom2011-02-181-0/+1
| | | | | | | | When we drop the in_swtnl_draw flag, we must force a rerun of update_need_swtnl to reset the need_swtnl flag to its correct value outside of a swtnl vbo draw. Signed-off-by: Thomas Hellstrom <[email protected]>
* r600g: reorganise rgtc pieces.Dave Airlie2011-02-181-16/+23
| | | | | | | when the cs checker fixes go upstream a lot of this can disappear into a drm version check. Signed-off-by: Dave Airlie <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-29 15:15:00 +0000