Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | nvc0: prevent overlap between load address and destination regs | Christoph Bumiller | 2011-05-13 | 1 | -15/+29 |
| | | | | | | | | For example, an indirect load like "ld b128 $r0q c0[$r0]" seems to overwrite the address register before finishing the load, but only if there are a lot of threads running. Visible as displaced geoemtry in Unigine Heaven. | ||||
* | nv50/nvc0: make transfers aware of PIPE_TRANSFER_MAP_DIRECTLY | Marcin Slusarz | 2011-05-09 | 1 | -0/+3 |
| | | | | | | | | | | If state tracker asked us to map resource directly and we can't do it (because of tiling), return NULL instead of doing full transfer - state tracker should handle it and fallback to some other method or repeat transfer without PIPE_TRANSFER_MAP_DIRECTLY. It greatly improves performance of xorg state tracker on nv50+, because its fallback (DFS/UTS) is much faster than full transfer. | ||||
* | nv50,nvc0: activate seamless cube map filtering | Christoph Bumiller | 2011-05-06 | 1 | -0/+5 |
| | |||||
* | nv50,nvc0: advertise GL_FIXED vertex buffers as supported | Christoph Bumiller | 2011-05-03 | 1 | -0/+18 |
| | | | | | We'll handle them like f64 vertex buffers, by falling back to copying vertex data to the command buffer through translate. | ||||
* | gallium: add fallback for copying buffers to all drivers | Marek Olšák | 2011-04-21 | 1 | -0/+8 |
| | | | | | | Just to keep drivers working. Reviewed-by: Jakob Bornecrantz <[email protected]> | ||||
* | gallium: include u_format.h for util_format_is_supported | Marcin Slusarz | 2011-04-17 | 1 | -0/+1 |
| | | | | | | | | | | | | Without it gcc complains: nv50_screen.c: In function ‘nv50_screen_is_format_supported’: nv50_screen.c:48: warning: implicit declaration of function ‘util_format_is_supported’ and handles it wrongly - util_format_is_supported returns boolean, which is typedef'ed to uchar, but function without prototype is assumed to return int. For me nv50_screen_is_format_supported was returning true for float formats without --enable-texture-float... | ||||
* | nvc0: work around what looks like a code prefetch bug | Christoph Bumiller | 2011-04-17 | 1 | -1/+4 |
| | | | | | | | Sounds very unlikely, but I don't have a better explanation at the moment. The GPU throws page faults at the first page after the code buffer quite frequently on startup, and traces don't show us overflowing. | ||||
* | nv50,nvc0: add new texture and render target formats | Christoph Bumiller | 2011-04-17 | 1 | -5/+61 |
| | |||||
* | nv50,nvc0: fix normalized RG16 and RGBA32 texture format entries | Christoph Bumiller | 2011-04-17 | 1 | -2/+2 |
| | |||||
* | nvc0: fix viewport rectangle calculation | Christoph Bumiller | 2011-04-17 | 1 | -5/+7 |
| | |||||
* | gallium: add and use generic function for querying patented format support (v2) | Marek Olšák | 2011-04-15 | 1 | -11/+2 |
| | | | | v2: Unsigned floats are allowed regardless of the configure switch. | ||||
* | nv50,nvc0: silence shader debug output | Christoph Bumiller | 2011-04-10 | 8 | -60/+33 |
| | |||||
* | nvc0: make the card permit separate colour outputs | Christoph Bumiller | 2011-04-10 | 2 | -14/+4 |
| | |||||
* | nv50-nvc0: make use of COLOR_MASK,BLEND_ENABLE_COMMON | Christoph Bumiller | 2011-04-10 | 5 | -20/+14 |
| | |||||
* | nvc0: shrink CSOs a little | Christoph Bumiller | 2011-04-10 | 3 | -26/+22 |
| | |||||
* | nv50,nvc0: identify some line rasterization state | Christoph Bumiller | 2011-04-10 | 3 | -5/+17 |
| | | | | Fixes width of non-smooth (aliased) lines on nvc0. | ||||
* | nvc0: force perspective correction hint to nicest | Christoph Bumiller | 2011-04-10 | 2 | -0/+7 |
| | | | | | | | | | Fixes issues in e.g. nexuiz (desertfactory) or supertuxkart that look like lighting bugs. They're not visible with the software rasterizers because their notion of linear interpolation seems to be different from that of nv50/nvc0. | ||||
* | nv50,nvc0: don't insert instructions after the block terminator | Christoph Bumiller | 2011-04-10 | 2 | -3/+10 |
| | |||||
* | gallium: set PIPE_CAP_MIXED_COLORBUFFER_FORMATS in some drivers | Marek Olšák | 2011-04-01 | 1 | -0/+1 |
| | |||||
* | nv50,nvc0: implement colour clamping controls | Christoph Bumiller | 2011-03-29 | 4 | -5/+7 |
| | |||||
* | nv50,nvc0: implement texture barrier | Christoph Bumiller | 2011-03-19 | 1 | -0/+10 |
| | |||||
* | nv50,nvc0: don't assert on cso with 0 vertex elements | Christoph Bumiller | 2011-03-13 | 1 | -2/+0 |
| | |||||
* | nvc0: support edge flags | Christoph Bumiller | 2011-03-13 | 4 | -14/+88 |
| | |||||
* | nvc0: fix POLYGON_MODE_BACK macro copy/paste error | Christoph Bumiller | 2011-03-13 | 1 | -1/+1 |
| | |||||
* | nv50,nvc0: fix pipe context switch | Christoph Bumiller | 2011-03-13 | 1 | -5/+32 |
| | |||||
* | nv50,nvc0: clean up flushes | Christoph Bumiller | 2011-03-13 | 2 | -20/+11 |
| | |||||
* | nv50,nvc0: add some missing resource referencing | Christoph Bumiller | 2011-03-13 | 2 | -4/+34 |
| | |||||
* | nvc0: mask out centroid bit for writing FP header | Christoph Bumiller | 2011-03-13 | 1 | -1/+1 |
| | | | | It's only 2 bit per input, centroid is set in the instruction. | ||||
* | nvc0: identify VERTEX_QUARANTINE | Christoph Bumiller | 2011-03-13 | 3 | -8/+17 |
| | | | | | | | | Well, not sure what exactly it is, but it certainly doesn't contain the control flow stack, but vertex data. Not sure about size, I've only seen the first few KiB written, but the binary driver seems to allocate more. | ||||
* | nvc0: don't enable early-z if alpha test is enabled | Christoph Bumiller | 2011-03-13 | 4 | -12/+20 |
| | | | | | | Depth values are also written before the shader is executed, so if early tests are enabled, fragments that failed the alpha test were modifying the depth buffer, but they shouldn't. | ||||
* | nvc0: move sprite coord replace state into cso | Christoph Bumiller | 2011-03-13 | 2 | -27/+8 |
| | | | | It's not dependent on any other state anymore now. | ||||
* | nvc0: s/nblocksx/nblocksy for height in resource_copy_region | Christoph Bumiller | 2011-03-13 | 1 | -1/+1 |
| | |||||
* | nvc0: fix unitialized variable in TGSI sysval decl processing | Christoph Bumiller | 2011-03-13 | 1 | -0/+1 |
| | |||||
* | nvc0: update/fix supported instruction src modifiers | Christoph Bumiller | 2011-03-13 | 5 | -20/+31 |
| | |||||
* | gallium: remove flags from the flush function | Marek Olšák | 2011-03-11 | 1 | -7/+8 |
| | | | | | | | | | | The drivers have been changed so that they behave as if all of the flags were set. This is already implicit in most hardware drivers and required for multiple contexts. Some state trackers were also abusing the PIPE_FLUSH_RENDER_CACHE flag to decide whether flush_frontbuffer should be called. New flag ST_FLUSH_FRONT has been added to st_api.h as a replacement. | ||||
* | gallium: remove the geom_flags param from is_format_supported | Marek Olšák | 2011-03-11 | 1 | -1/+1 |
| | |||||
* | gallium: kill is_resource_referenced | Marek Olšák | 2011-03-11 | 2 | -21/+0 |
| | | | | Only st/xorg used it and even incorrectly with regards to pipelined transfers. | ||||
* | gallium: split CAP_INSTANCE_DRAWING into INSTANCEID and INSTANCE_DIVISOR | Marek Olšák | 2011-03-05 | 1 | -1/+2 |
| | | | | | | | | ARB_instanced_arrays is a subset of D3D9. ARB_draw_instanced is a subset of D3D10. The point of this change is to allow D3D9-level drivers to enable ARB_instanced_arrays without ARB_draw_instanced. | ||||
* | nvc0: Update SConscript. | Vinson Lee | 2011-03-04 | 1 | -3/+0 |
| | |||||
* | nv50,nvc0: get format desc for TIC entry from sampler view format | Christoph Bumiller | 2011-03-05 | 1 | -1/+1 |
| | | | | Fixes piglit/tex-srgb. | ||||
* | nv50,nvc0: share sampler state creation | Christoph Bumiller | 2011-03-05 | 6 | -126/+22 |
| | |||||
* | nv50,nvc0: update the format tables | Christoph Bumiller | 2011-03-05 | 7 | -524/+191 |
| | | | | | | | | Removed sampler view support for USCALED/SSCALED, the texture unit refuses to convert to non-normalized float. The enums are treated like UNORM. Removed duplicate format related headers. | ||||
* | nvc0: use m2mf for resource_copy_region if formats are equal | Christoph Bumiller | 2011-03-05 | 3 | -4/+75 |
| | | | | | Which is always the case, but we'll keep the 2D engine blitter nonetheless. | ||||
* | nv50,nvc0: fix texture layer issues | Christoph Bumiller | 2011-03-05 | 9 | -62/+78 |
| | |||||
* | nouveau: allow pipe driver to define which buffers should start in sysmem | Ben Skeggs | 2011-03-03 | 1 | -3/+2 |
| | | | | | | | PIPE_BIND_CONSTANT_BUFFER alone was OK for nv50/nvc0, but nv30 will need to be able to set others on certain chipsets. Signed-off-by: Ben Skeggs <[email protected]> | ||||
* | nvc0: fix a crash on context destruction | Ben Skeggs | 2011-03-01 | 1 | -2/+6 |
| | | | | Signed-off-by: Ben Skeggs <[email protected]> | ||||
* | nouveau: ensure vbo_dirty is set when buffer write transfer complete | Ben Skeggs | 2011-03-01 | 11 | -102/+111 |
| | | | | | | This introduces a shared nouveau_context struct to track such things. Signed-off-by: Ben Skeggs <[email protected]> | ||||
* | nvc0: port to common fence/mm/buffer code | Ben Skeggs | 2011-03-01 | 21 | -1347/+128 |
| | | | | Signed-off-by: Ben Skeggs <[email protected]> | ||||
* | nv50-nvc0: set cur_ctx during init if none currently bound | Ben Skeggs | 2011-03-01 | 1 | -0/+2 |
| | | | | Signed-off-by: Ben Skeggs <[email protected]> | ||||
* | nvc0: fix PointCoord enable in FP header | Christoph Bumiller | 2011-02-24 | 1 | -2/+5 |
| |