summaryrefslogtreecommitdiff
path: root/src/gallium/winsys
AgeCommit message (Collapse)Author
2010-10-04r600g: allow r600_bo to be a sub allocation of a big boJerome Glisse
Add bo offset everywhere needed if r600_bo is ever a sub bo of a bigger bo. Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-10-04r600g: rename radeon_ws_bo to r600_boJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-10-04r600g: TODO domain managementDave Airlie
no wonder it was slow, the code is deliberately forcing stuff into GTT, we used to have domain management but it seems to have disappeared.
2010-10-04r600g: fix wwarning in bo_map functionDave Airlie
2010-10-04r600g: break out of search for reloc bo after finding it.Dave Airlie
this function was taking quite a lot of pointless CPU.
2010-10-01r600g: setup basic loop consts on r600 + evergreen.Dave Airlie
this sets up a single loop constant like r600c does.
2010-10-01r600g: flush SH cache on constant change on evergreenDave Airlie
2010-10-01r600g: fix evergreen draw-buffersDave Airlie
just a typo in the register headers.
2010-10-01r600g: add cb flushing for extra buffers + depth buffer on r600/evergreenDave Airlie
2010-10-01r600g: sync vertex/texture cache on resources on evergreenDave Airlie
this gets rid of lots of the instability on evergreen, which isn't surprising since it really broken not to flush caches.
2010-10-01r600g: add reloc for evergreen color attribDave Airlie
we'll need this for color tiling on evergreen.
2010-10-01r600g: add winsys support for CTL constants.Dave Airlie
These need to be emitted, we also need them to do proper vtx start, instead of abusing index offset.
2010-10-01r600g: use Elements macro instead of manual sizeofsDave Airlie
2010-09-30r600g: keep a mapping around for each boJohn Doe
Save a lot of call into the kernel and thus improve performances. Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-30r600g: don't double count dirty blockJohn Doe
This avoid to overcount the number of dwords we need and thus avoid maximazation of cs buffer use. Signed-off-by: Jerome Glisse <jglisse@redhat.com
2010-09-30evergreeng: avoid overlapping border color btw VS & PSJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-30r600g: use constant buffer instead of register for constantJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-30gallium/winsys: remove duplicated includeNicolas Kaiser
Remove duplicated include. Signed-off-by: Brian Paul <brianp@vmware.com>
2010-09-30r600g: clean up some code from move to new paths.Dave Airlie
mainly remove 2 suffix from function names
2010-09-29r600g: Fix SCons build.Vinson Lee
2010-09-29r600g: more cleanupJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-29r600g: Update SConscript.Vinson Lee
Fixes SCons build.
2010-09-29r600g: cleanupJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-29r600g: delete old pathJerome Glisse
Lot of clean can now happen. Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-29r600g: use a hash table instead of groupJerome Glisse
Instead of creating group of register use a hash table to lookup into which block each register belongs. This simplify code a bit. Signed-off-by: Jerome Glisse <jglisse@redhat.com
2010-09-29r600g: move radeon.h members around to add back map flushing.Dave Airlie
2010-09-28r600g: suspend/resume occlusion query around clear/copyJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-27r600g: fix occlusion query after change to block structureJerome Glisse
block->reg point to register value not block->pm4 which point to packet. Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-27r600g: fix pointsprite & resource unbindingJerome Glisse
When asking to bind NULL resource assume it's unbinding so free resource and unreference assoicated buffer. Also fix pointsprite parameter. Fix glsl-fs-pointcoord & fp-fragment-position Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-27r600g: build packet header onceJerome Glisse
Build packet header once and allow to add fake register support so we can handle things like indexed set of register (evergreen sampler border registers for instance. Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-27r600g: add evergreen texture resource properly.Dave Airlie
adding sampler border looks impossible with current design, another day, another corner case not worked out.
2010-09-26radeong: fix leaksJoakim Sindholt
2010-09-26r600g: disable early cull optimization when occlusion query runningJerome Glisse
When occlusion query are running we want to have accurate fragment count thus disable any early culling optimization GPU has. Based on work from Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-26r600g: Include p_compiler.h instead of malloc.h.Vinson Lee
2010-09-26r600g: Remove unused variables.Vinson Lee
Fixes these GCC warnings. radeon.c: In function 'radeon_new': radeon.c:59: warning: unused variable 'k' radeon.c:59: warning: unused variable 'j' radeon.c:59: warning: unused variable 'id' radeon.c:59: warning: unused variable 'i'
2010-09-26r600g: Don't return a value in function returning void.Vinson Lee
Fixes this GCC warning. radeon_state.c: In function 'radeon_state_fini': radeon_state.c:140: warning: 'return' with a value, in function returning void
2010-09-25r600g: Remove unused variable.Vinson Lee
Fixes this GCC warning. radeon_bo_pb.c: In function 'radeon_bo_pb_create_buffer': radeon_bo_pb.c:178: warning: unused variable 'domain'
2010-09-25r600g: add eg db count control register.Dave Airlie
2010-09-24r600g: bring over fix from old path to new pathJerome Glisse
Up to 2010-09-19: r600g: fix tiling support for ddx supplied buffers 9b146eae2521d8e5f6d3cbefa4f6f7737666313a user buffer seems to be broken... new to fix that. Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-24r600g: fix evergreen new pathJerome Glisse
glxgears seems to work, had somelockup but now they seems to have vanish. Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-24r600g: fix evergreen new pathJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-24r600g: evergreen fix for new designJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-24r600g: move use_mem_constants flags for new designs structure alignmentJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-24r600g: fix typo in evergreen define (resource are in [0x30000;0x34000] range)Jerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-24r300g: make accessing map_list and buffer_handles thread-safeMarek Olšák
NOTE: This is a candidate for the 7.9 branch.
2010-09-24r300g: fixup long-lived BO maps being incorrectly unmapped when flushingMarek Olšák
Based on commit 3ddc714b20ac4e28b80c6f88d1993445fff2262c by Dave Airlie. NOTE: This is a candidate for the 7.9 branch.
2010-09-23r600g: initial evergreen support in new pathJerome Glisse
This doesn't work yet. Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-23r600g: fix typo in evergreen register listDave Airlie
pointed out by glisse on irc.
2010-09-22r600g: disable shader rebuild optimization & account cb flush packetJerome Glisse
Shader rebuild should be more clever, we should store along each shader all the value that change shader program rather than using flags in context (ie change sequence like : change vs buffer, draw, change vs buffer, switch shader will trigger useless shader rebuild). Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-22r600g: flush color buffer after draw commandJerome Glisse
Signed-off-by: Jerome Glisse <jglisse@redhat.com>