summaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* nvc0: correct storage type for 16 bit surface formatsChristoph Bumiller2011-02-091-2/+2
|
* nvc0: make CSE work for ops with multiple resultsChristoph Bumiller2011-02-091-13/+72
|
* nvc0: replace branching with predicated insns where feasibleChristoph Bumiller2011-02-095-46/+123
|
* nvc0: implement local memory load and store opsChristoph Bumiller2011-02-093-12/+38
|
* nvc0: make sure phi-ops really have one source per in-blockChristoph Bumiller2011-02-092-19/+40
|
* nv50,nvc0: do not forget to apply sign mode to saved TGSI inputsChristoph Bumiller2011-02-092-11/+3
|
* nvc0: do not generate a backwards jump if a loop ends with BRKChristoph Bumiller2011-02-091-2/+4
|
* nvc0: store only one value per basic block for TGSI regsChristoph Bumiller2011-02-091-1/+7
|
* nv50,nvc0: fix condition code change when commuting SET sourcesChristoph Bumiller2011-02-092-3/+3
|
* nvc0: set basic block on manual instruction insertionChristoph Bumiller2011-02-091-0/+2
|
* nvc0: try to fix register conflicts for vector instructionsChristoph Bumiller2011-02-094-32/+43
| | | | | | | Vector here means using multiple 32 bit regs which are forced to be consecutive in the register file. This still isn't quite nice.
* nvc0: reset texture base address after read transferChristoph Bumiller2011-02-091-2/+4
|
* nvc0: don't combine memory loads across block boundariesChristoph Bumiller2011-02-091-0/+5
|
* nvc0: detect no-op MIN/MAX, do CSE earlier to succeed more oftenChristoph Bumiller2011-02-091-48/+79
|
* gallium/docs: fix typoMarek Olšák2011-02-091-1/+1
|
* st/egl: Fix platform selectionBenjamin Franzke2011-02-081-0/+1
| | | | | A break for case _EGL_PLATFORM_X11 is missing. introduced by: 381ea0d67a6d84a34d23571c49bbf4339ffda364
* r600g: fixup assertionMarek Olšák2011-02-081-1/+1
|
* r600g: add a faster implementation of transfer_inline_writeMarek Olšák2011-02-081-1/+25
| | | | u_default_transfer_inline_write uses util_copy_rect, which is kinda slow.
* r600g: slab-allocate buffer and transfer structuresMarek Olšák2011-02-083-10/+85
|
* r300g: use the same upload buffer for vertices and indicesMarek Olšák2011-02-084-18/+7
|
* r600g: use the same upload buffer for vertices, indices, and constantsMarek Olšák2011-02-084-28/+10
| | | | This should reduce memory consumption.
* u_vbuf_mgr: add a way to specify the BIND flag for the upload bufferMarek Olšák2011-02-084-1/+5
|
* u_vbuf_mgr: remove tabsMarek Olšák2011-02-081-18/+18
|
* u_vbuf_mgr: make the uploader publicMarek Olšák2011-02-083-18/+16
|
* Revert "r600g: do not flush the uploader" (with comments)Marek Olšák2011-02-081-0/+7
| | | | This reverts commit 1c2a4f0820ff2272f993e6da28dcf8bcbbc3252a.
* r600g: silence a few valgrind warningsBrian Paterni2011-02-083-4/+4
|
* r600g: Add support for relative addressing on constant buffers.Henri Verbeet2011-02-076-22/+152
| | | | | | | | | | Relative addressing of constant buffers can't work properly through the kcache, since you can only address within the currently locked kcache window. Instead, this patch binds the constant buffer as a shader resource, and then explicitly fetches the constant using a vertex fetch with fetch type VTX_FETCH_NO_INDEX_OFFSET from the shader. There's probably still some room for improvement, doing the fetch right before the instruction that needs the value may not be quite optimal for example.
* r600g: Set the fetch type in r600_bc_vtx_build().Henri Verbeet2011-02-071-0/+1
|
* r600g: Handle the ADD_INT instruction in r600_bc_get_num_operands().Henri Verbeet2011-02-071-0/+2
|
* r600g: Generalize the pipe_add_vertex_attrib() functions.Henri Verbeet2011-02-074-28/+20
| | | | This allows them to be used for VS or PS buffer resources as well.
* r600g: Remove vs_resource and ps_resource from the pipe context.Henri Verbeet2011-02-073-22/+6
| | | | | These are practically unused, only the vs_resource array is being abused for fetch shader resources.
* r600g: Split constants in r600_shader_from_tgsi().Henri Verbeet2011-02-071-140/+80
|
* r600g: Do the tgsi_full_src_register to r600_shader_src conversion in ↵Henri Verbeet2011-02-071-176/+129
| | | | r600_shader_from_tgsi().
* r600g: Split r600_bc_alu_src.Henri Verbeet2011-02-073-157/+147
| | | | | | The r600_bc_alu_src structure is used in two different ways, as a vector and for the individual channels of that same vector. This is somewhat fragile, and probably confusing.
* r600g: Store literal values in the r600_bc_alu_src structure.Henri Verbeet2011-02-073-16/+13
| | | | | This is much easier to work with, and allows use to get rid of some of the literal handling hacks.
* r600g: tgsi_dst() can't fail.Henri Verbeet2011-02-071-75/+26
|
* r600g: tgsi_src() can't fail.Henri Verbeet2011-02-071-80/+28
|
* st/egl: Add wayland platformBenjamin Franzke2011-02-076-0/+718
|
* st/egl: drm_image: Check for MESA_drm_imageBenjamin Franzke2011-02-071-2/+2
| | | | | MESA_drm_image isnt limited to drm platform, others can enable the extension too.
* st/egl: native_helper: Add resource_surface_import_resourceBenjamin Franzke2011-02-072-0/+13
|
* st/dri: Implement dri2::{Allocate,Release}BufferBenjamin Franzke2011-02-072-0/+111
|
* r600g: correctly report supported vertex formatsMarek Olšák2011-02-073-10/+38
|
* r300g: use the new vertex buffer managerMarek Olšák2011-02-079-560/+144
|
* r600g: use the new vertex buffer managerMarek Olšák2011-02-079-419/+101
|
* util: import a new vertex buffer managerMarek Olšák2011-02-074-1/+722
| | | | | | | This code has originally matured in r300g and was ported to r600g several times. It was obvious it's a code duplication. See also comments in the header file.
* r600g: do not flush the uploaderMarek Olšák2011-02-061-3/+0
|
* r300g: do not flush the uploaderMarek Olšák2011-02-061-3/+0
| | | | | We don't have to unmap and recreate the upload buffer when a flush occurs. This should also prevent buffer allocations from failing.
* r300g: RS400 doesn't have ZMASKMarek Olšák2011-02-061-1/+0
|
* r600g: use surface format not underlying texture formatDave Airlie2011-02-062-6/+6
| | | | | | | This uses the surface format to set the CB up not the underlying texture format, since these can and do differ. Fixes piglit fbo-srgb.
* r600g: don't set tile_type on evergreen.Dave Airlie2011-02-041-1/+1
| | | | | Since we never bind the actual DB to the CB/texture only the flushed one we don't need to track the tile type at the moment.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-12 14:02:21 +0000