aboutsummaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers/nv04
Commit message (Collapse)AuthorAgeFilesLines
* nouveau: Fix glTexSubImage on swizzled surfaces on <=NV40Luca Barbieri2009-12-302-15/+18
| | | | | | | | | | | | | | | | | | Currently in nvXX_transfer_new a temporary as large as the surface is created. If the subrectangle is not the whole texture we would need to read back the whole texture, but we aren't. Thus, everything but the subrectangle specified is loaded as garbage. This can be seen in progs/demos/ray. This patch fixes the problem by creating a temporary that covers only the desired subrectangle. That makes us hit an alignment assert in nv04_surface_2d.c. Fix it using the point registers instead of manipulating the swizzled surface offset to account for the destination coordinates (which do not seem to have a 1024 limit). Signed-off-by: Francisco Jerez <[email protected]>
* nouveau: Unreference state/buffer objects on context/screen destruction.Younes Manton2009-12-281-0/+2
| | | | | | | | - unreference state objects so that buffer objects are unreferenced and eventually destroyed - free channel at screen's destruction Based on Krzysztof Smiechowicz's patch.
* nouveau: Fix swizzling for copies to rectangular texturesLuca Barbieri2009-12-281-3/+12
| | | | | | | | | | | | | nVidia hardware seems to swizzle rectangular texture (with width != height) coordinates by swizzling the lower bits and then adding the higher bits from the larger dimension. However, nv04_swizzle_bits ignores width and height and just interleaves everything. This causes problems with rectangular POT textures with height or width 2048 or 4096 (but not 2048x1024 where it works by chance) since the driver swizzles them in 1024x1024 chunks and gets the start position for the non-first chunks wrong. The following patch seems to fix those problems.
* nouveau: Fix nv20-40 swizzled miptree RTsLuca Barbieri2009-12-282-0/+50
| | | | | | | | | | | | | I just coded a patch that does this and seems to work fine. It must be fixed since it breaks OpenGL (or the state tracker can be changed, but it seems better to do it in the driver). The patch also fixes NV20 and NV30 in the same way. They compile but are untested. I would guess that using the 3D engine is faster for the larger levels, but the 2D engine is faster for the smaller ones (and lacks this issue).
* nv04: Fix build after the latest nouveau_class.h changes.Francisco Jerez2009-12-286-62/+62
|
* gallium: add geometry shader support to galliumZack Rusin2009-12-251-1/+1
|
* Merge branch 'gallium-edgeflags'Roland Scheidegger2009-12-221-6/+0
|\ | | | | | | | | Conflicts: src/mesa/state_tracker/st_draw.c
| * gallium: fix up drivers for edgeflag changesRoland Scheidegger2009-12-191-6/+0
| | | | | | | | | | several drivers which chose to ignore edgeflags might require some more work, while edgeflags never worked there they might now crash.
* | Move the remaining format pf_get_* functions to u_format.h.Michal Krol2009-12-172-7/+9
| | | | | | | | | | Previously they depended on format blocks, but after removing those they started depending on format encoding.
* | nouveau: avoid running out of relocsMaarten Maathuis2009-12-141-3/+6
|/ | | | | | | - Added flush notify functions for NV30 and NV40. - Flushing mid frame will call flush notify, which will resubmit all relocs. - We don't try to recover from reloc failure yet.
* nouveau: use boolean instead of boolRoland Scheidegger2009-12-081-1/+1
|
* nouveau: Work around nv04-nv40 miptrees not matching nouveau_miptree.Younes Manton2009-12-062-1/+3
| | | | | | Thanks to Bob Gleitsmann for the patch. I'll clean this up in a better way later if noone else beats me to it.
* Merge branch 'gallium-noblocks'Roland Scheidegger2009-12-043-24/+10
|\ | | | | | | | | Conflicts: src/gallium/state_trackers/xorg/xorg_exa.c
| * gallium: adapt nv drivers to interface cleanupsRoland Scheidegger2009-12-023-24/+10
| |
* | nv: Update for renamed sampler/texture state setters.Michal Krol2009-12-011-2/+2
|/
* gallium: fix more statetrackers/drivers for not using texture ↵Roland Scheidegger2009-11-263-15/+15
| | | | width/height/depth arrays
* nouveau: Support X8R8G8B8 textures on nv30, nv40 and RTs on nv10-nv40.Younes Manton2009-10-311-0/+1
|
* nouveau: implement is_{texture,buffer}_referenced properlyBen Skeggs2009-10-192-26/+3
|
* Add support for more 8 and 16 bits formatsPatrice Mandin2009-10-121-1/+10
|
* nv04-nv40: Fix swizzle transfers for NPOT sizes.Younes Manton2009-10-061-113/+34
| | | | | | Workarounds not necessary, SIFM can handle NPOT, we just weren't setting dst dimensions properly. SIFM can't handle odd w,h though, that still needs fixing.
* gallium: Preparations for adding more PIPE_TRANSFER_* usage flags.Michel Dänzer2009-10-021-21/+5
| | | | | | Always test for PIPE_TRANSFER_READ/WRITE using the bit-wise and operator, and add a pipe_transfer_buffer_flags() helper for getting the buffer usage flags corresponding to them.
* gallium: Deprecate PIPE_CAP_S3TC.José Fonseca2009-09-161-2/+0
| | | | | No longer used. S3TC support is queried via pipe_screen::is_format_supported.
* nv04: Handle more Z formats.Younes Manton2009-08-261-0/+4
| | | | Based on the patch from Luo Jinghua.
* nouveau: swizzle a single row or column, doing it one pixel at a timePatrice Mandin2009-07-271-28/+80
|
* nouveau: only swizzle square textures for copyPatrice Mandin2009-07-261-4/+11
|
* nouveau: Take into account destination position for copy_swizzle, need to ↵Patrice Mandin2009-07-261-2/+12
| | | | split copy a bit more
* nouveau: Recursively swizzle an NPOT sized copyPatrice Mandin2009-07-261-5/+21
|
* nouveau: use nv04_surface_copy_swizzle only for POT sizesPatrice Mandin2009-07-241-1/+4
|
* nouveau: Take into account sx,sy parameters to read from source surfacePatrice Mandin2009-07-221-2/+2
|
* gallium: Define PIPE_CAP_BLEND_EQUATION_SEPARATE, remove extension from ↵Patrice Mandin2009-07-141-0/+2
| | | | default extension list
* nouveau: return some supported zeta formatsBen Skeggs2009-07-011-0/+7
|
* nouveau: deal with PIPE_CAP_TGSI_CONT_SUPPORTEDBen Skeggs2009-07-011-0/+2
|
* nouveau: Turn off debug cannon.Younes Manton2009-06-291-2/+1
|
* nouveau: remove unneeded code from ws, use pipe_buffer_ instead of ws->Ben Skeggs2009-06-052-10/+13
|
* nouveau: move channel creation into pipe driversBen Skeggs2009-06-053-15/+10
|
* nouveau: call notifier/grobj etc funcs directlyBen Skeggs2009-06-052-11/+22
| | | | | libdrm_nouveau is linked with the winsys, there's no good reason to do all this through yet another layer.
* nv04: remove u_simple_screen usageBen Skeggs2009-06-053-47/+28
|
* nv04: small fix againStephane Marchesin2009-05-051-1/+2
|
* Fix an nv04 bug (thanks Mhopf).Stephane Marchesin2009-05-051-1/+1
|
* gallium: Make sure we flush before some texture / buffer operations.Thomas Hellstrom2009-04-151-0/+27
| | | | | | | Also implement context member functions to optimize away those flushes whenever possible. Signed-off-by: Thomas Hellstrom <thellstrom-at-vmware-dot-com>
* gallium: remove pipe_texture::compressed fieldBrian Paul2009-04-101-1/+0
| | | | | The format field encodes compressed vs. uncompressed already. We can easily check if a texture is compressed with pf_is_compressed(texture->format).
* gallium: Remove some little-used fields from struct pipe_surface.Michel Dänzer2009-03-261-1/+0
|
* gallium: Remove do_flip argument from surface_copyJakob Bornecrantz2009-03-131-10/+1
| | | | | | | I should have gotten most uses and implementation correctly fixed, but things might break. Feel free to blame me.
* nouveau: oops, forgot to keep the NV10_SIFM objectPatrice Mandin2009-03-041-0/+4
|
* nouveau: Use proper SIFM object on NV30Patrice Mandin2009-03-041-6/+10
|
* gallium: Unify reference counting.Michel Dänzer2009-03-042-31/+13
| | | | | | | | | | | | | | The core reference counting code is centralized in p_refcnt.h. This has some consequences related to struct pipe_buffer: * The screen member of struct pipe_buffer must be initialized, or pipe_buffer_reference() will crash trying to destroy a buffer with reference count 0. u_simple_screen takes care of this, but I may have missed some of the drivers not using it. * Except for rare exceptions deep in winsys code, buffers must always be allocated via pipe_buffer_create() or via screen->*buffer_create() rather than via winsys->*buffer_create().
* nouveau: Grab correct surface from temp texture for transfers.Younes Manton2009-03-021-1/+1
|
* nouveau: nv04-nv40 texture transfer.Younes Manton2009-02-259-80/+266
|
* gallium: Improve makefiles for librariesJakob Bornecrantz2009-02-201-10/+1
| | | | | | | | The template makefile that most libraries in gallium included was based on dri and had a bunch unrelevant junk in it. Update it and improve the depending makefiles.
* Merge commit 'origin/draw-vbuf-interface'Keith Whitwell2009-02-181-5/+17
|\
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-01 01:48:25 +0000