summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* radeonsi: Remove incorrect (and dead) assignment in tex_fetch_args().Michel Dänzer2013-02-071-2/+0
| | | | The proper return type is assigned at the end of the function.
* radeonsi: Use unique names for referring to texture sampling intrinsics.Michel Dänzer2013-02-071-3/+18
| | | | | | | | | | Append the overloaded vector type used for passing in the addressing parameters. Without this, LLVM uses the same function signature for all those types, which cannot work. Fixes problems e.g. with FlightGear and Red Eclipse.
* r300g: put textures with usage=staging in GTT and make them linearMarek Olšák2013-02-072-1/+6
|
* r600g: fix slice tile max for compressed texture and async dmaJerome Glisse2013-02-072-4/+8
| | | | | | | Was using the pixel size instead of the number of block for the slice tile max computation which resulted in dma writing at wrong address. Signed-off-by: Jerome Glisse <[email protected]>
* radeonsi: use new RGBX formatsMarek Olšák2013-02-071-0/+26
|
* r300g: fix blending and alpha-test with RGBX16F and enable MSAA for itMarek Olšák2013-02-076-14/+45
|
* r300g: use new RGBX formatsMarek Olšák2013-02-072-3/+22
|
* r600g: use new RGBX formatsMarek Olšák2013-02-072-0/+52
|
* st/mesa: use new RGBX formatsMarek Olšák2013-02-071-17/+111
| | | | Reviewed-by: Brian Paul <[email protected]>
* mesa: add RGBX formats for existing GL RGB texture formatsMarek Olšák2013-02-076-47/+959
| | | | v2: fix compilation of swrast
* gallium: add RGBX formats for existing GL RGB texture formatsMarek Olšák2013-02-073-0/+36
| | | | Reviewed-by: Brian Paul <[email protected]>
* i965/blorp: Support blits between ARGB and XRGB formats.Kenneth Graunke2013-02-061-14/+12
| | | | | | | | | | | | | | | Now that we have support for overriding alpha to 1.0, we can handle blitting between these formats in either direction. For now, we only support two XRGB formats: MESA_FORMAT_XRGB8888 and MESA_FORMAT_RGBX8888_REV. Most places only appear to worry about the former, so ignore the latter for now. We can always add it later. NOTE: This is a candidate for the 9.1 branch. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Ian Romanick <[email protected]> Tested-by: Martin Steigerwald <[email protected]>
* i965/blorp: Support overriding destination alpha to 1.0.Kenneth Graunke2013-02-061-0/+19
| | | | | | | | | | | | | | | | | | | | Currently, Blorp requires the source and destination formats to be equal. However, we'd really like to be able to blit between XRGB and ARGB formats; our BLT engine paths have supported this for a long time. For ARGB -> XRGB, nothing needs to occur: the missing alpha is already interpreted as 1.0. For XRGB -> ARGB, we need to smash the alpha channel to 1.0 when writing the destination colors. This is fairly straightforward with blending. For now, this code is never used, as the source and destination formats still must be equal. The next patch will relax that restriction. NOTE: This is a candidate for the 9.1 branch. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Ian Romanick <[email protected]> Tested-by: Martin Steigerwald <[email protected]>
* i965: Implement CopyTexSubImage2D via BLORP (and use it by default).Kenneth Graunke2013-02-065-8/+154
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The BLT engine has many limitations. Currently, it can only blit X-tiled buffers (since we don't have a kernel API to whack the BLT tiling mode register), which means all depth/stencil operations get punted to meta code, which can be very CPU-intensive. Even if we used the BLT engine, it can't blit between buffers with different tiling modes, such as an X-tiled non-MSAA ARGB8888 texture and a Y-tiled CMS ARGB8888 renderbuffer. This is a fundamental limitation, and the only way around that is to use BLORP. Previously, BLORP only handled BlitFramebuffer. This patch adds an additional frontend for doing CopyTexSubImage. It also makes it the default. This is partly to increase testing and avoid hiding bugs, and partly because the BLORP path can already handle more cases. With trivial extensions, it should be able to handle everything the BLT can. This helps PlaneShift massively, which tries to CopyTexSubImage2D between depth buffers whenever a player casts a spell. Since these are Y-tiled, we hit meta and software ReadPixels paths, eating 99% CPU while delivering ~1 FPS. This is particularly bad in an MMO setting because people cast spells all the time. It also helps Xonotic in 4X MSAA mode. At default power management settings, I measured a 6.35138% +/- 0.672548% performance boost (n=5). (This data is from v1 of the patch.) No Piglit regressions on Ivybridge (v3) or Sandybridge (v2). v2: Create a fake intel_renderbuffer to wrap the destination texture image and then reuse do_blorp_blit rather than reimplementing most of it. Remove unnecessary clipping code and conditional rendering check. v3: Reuse formats_match() to centralize checks; delete temporary renderbuffers. Reorganize the code. v4: Actually copy stencil when dealing with separate stencil buffers but packed depth/stencil formats. Tested by a new Piglit test. NOTE: This is a candidate for the 9.1 branch. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Paul Berry <[email protected]> [v4] Reviewed-by: Ian Romanick <[email protected]> [v3] Reviewed-and-tested-by: Carl Worth <[email protected]> [v2] Tested-by: Martin Steigerwald <[email protected]> [v3]
* mesa: Put extern "C" guards in renderbuffer.h.Kenneth Graunke2013-02-061-1/+7
| | | | | | | | | I need to use this from C++ code. NOTE: This is a candidate for the 9.1 branch. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* llvmpipe: remove extraneous const qualifierBrian Paul2013-02-062-2/+2
|
* gallium/util: remove duplicated function util_format_is_rgb_no_alphaMarek Olšák2013-02-063-33/+6
| | | | | | It only checks if alpha is present, so it's the same as util_format_has_alpha. Reviewed-by: Brian Paul <[email protected]>
* st/mesa: get rid of GET_CURRENT_CONTEXT in st_choose_formatMarek Olšák2013-02-066-22/+21
| | | | Reviewed-by: Brian Paul <[email protected]>
* st/mesa: adjust texture format selection to try the closest base format firstMarek Olšák2013-02-061-18/+13
| | | | Reviewed-by: Brian Paul <[email protected]>
* st/mesa: put RGBX8 and RGBA8 in the default format listsMarek Olšák2013-02-061-0/+2
| | | | Reviewed-by: Brian Paul <[email protected]>
* st/mesa: add the rest of RGB8 format/type combos to exact_format_mapping tablesMarek Olšák2013-02-061-2/+0
| | | | | | | | These formats were added a few months after these tables were committed. No idea why we have the table though. AFAIK, texstore always takes the slow path for GL_RGBn. Reviewed-by: Brian Paul <[email protected]>
* mesa: fixup inconsistent naming of RG16 formatsMarek Olšák2013-02-0612-37/+37
| | | | Reviewed-by: Brian Paul <[email protected]>
* r600g: report correct control flow depthMarek Olšák2013-02-061-2/+1
|
* glsl: fix incorrect comment about do_common_optimizationMarek Olšák2013-02-061-2/+2
|
* st/mesa: emit saturates in the vertex shader if Shader Model 3.0 is supportedMarek Olšák2013-02-063-3/+6
| | | | v2: change the requirement from GLSL 1.30 to SM 3.0 (R500 can do this)
* st/mesa: advertise ARB_shading_language_packing for GLSL >= 1.30Marek Olšák2013-02-062-0/+16
| | | | Reviewed-by: Brian Paul <[email protected]>
* st/mesa: do most of GLSL lowering outside of the optimization do-while loopMarek Olšák2013-02-061-36/+29
| | | | | | based on the intel driver Reviewed-by: Brian Paul <[email protected]>
* st/mesa: remove dead code depending on EmitCondCodesMarek Olšák2013-02-061-22/+2
| | | | | | EmitCondCodes is always false. Reviewed-by: Brian Paul <[email protected]>
* r300g: try to use color varyings for texcoords if max texcoord limit is exceededMarek Olšák2013-02-066-5/+71
| | | | +35 piglits
* r300/compiler: copy-propagate saturate mode when possibleMarek Olšák2013-02-061-2/+19
| | | | Reviewed-by: Tom Stellard <[email protected]>
* r300/compiler: add support for saturate output modifier in r500 vertex shadersMarek Olšák2013-02-062-10/+19
| | | | | | | The GLSL compiler can simplify clamp(v,0,1) to saturate. The state tracker doesn't use it yet, but it will. Reviewed-by: Tom Stellard <[email protected]>
* r300g: fix blending with RGBX formatsMarek Olšák2013-02-063-122/+202
| | | | Change DST_ALPHA to ONE.
* r300g: fix blending with blend color and RGBA formatsMarek Olšák2013-02-061-0/+11
| | | | NOTE: This is a candidate for the stable branches.
* egl/dri: Don't invoke dri2_dpy->flush if it's NULL.José Fonseca2013-02-061-1/+2
| | | | | | | I'd like to test Mesa OpenGL ES along side with NVIDIA libGL drivers. But without this change, I get a NULL pointer dereference. Reviewed-by: Brian Paul <[email protected]>
* glsl: Initialize ast_parameter_declarator member variables.Vinson Lee2013-02-051-4/+8
| | | | | | | Fixes uninitialized pointer field defect reported by Coverity. Signed-off-by: Vinson Lee <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* svga: fix sRGB renderingBrian Paul2013-02-051-0/+11
| | | | | | | | | | | We weren't emitting the SVGA_RS_OUTPUTGAMMA state so sRGB rendering didn't work properly. Fixes piglit's framebuffer-srgb test. Note: This is a candidate for the stable branches. Reviewed-by: Jose Fonseca <[email protected]>
* r600g/compute: Fix segfault caused by new shader disassemblerTom Stellard2013-02-052-2/+4
|
* i965: Disable write masking when setting up texturing m0.Eric Anholt2013-02-042-1/+4
| | | | | | | | | | v2/Kayden: Also disable write masking in the vec4 backend. Fixes 78 oglconform glsl-bif-tex-* subcases. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Paul Berry <[email protected]> [v1] Reviewed-by: Eric Anholt <[email protected]> [v2]
* intel: Fix regression in intel_create_image_from_name stride handlingTapani Pälli2013-02-041-1/+1
| | | | | | | | | Strangely, the DRIimage interface we have passes the pitch in pixels instead of bytes, which anholt missed in the change to using bytes for region pitch. Signed-off-by: Tapani Pälli <[email protected]> Reviewed-by: Eric Anholt <[email protected]>
* i965: Fix segfaults from 45a28a927ab7f29ff325b9326d386a39ba538c18Eric Anholt2013-02-042-2/+4
| | | | | | If you look up a level that isn't in the miptree, you crash. Reviewed-by: Chad Versace <[email protected]>
* radeonsi: default PA_SC_RASTER_CONFIG to 0Alex Deucher2013-02-041-1/+3
| | | | | | | | That should work in all cases. Signed-off-by: Alex Deucher <[email protected]> Note: this is a candidate for the 9.1 branch.
* radeonsi: add support for Oland chipsAlex Deucher2013-02-044-0/+6
| | | | | | Signed-off-by: Alex Deucher <[email protected]> Note: this is a candidate for the 9.1 branch
* glsl: Support transform feedback of varying structs.Paul Berry2013-02-042-67/+184
| | | | | | | | | | | | | | | | | | | Since transform feedback needs to be able to access individual fields of varying structs, we can no longer match up the arguments to glTransformFeedbackVaryings() with variables in the vertex shader. Instead, we build up a hashtable which records information about each possible name that is a candidate for transform feedback, and then match up the arguments to glTransformFeedbackVaryings() with the contents of that hashtable. Populating the hashtable uses the program_resource_visitor infrastructure, so the logic is shared with how we handle uniforms. NOTE: This is a candidate for the 9.1 branch. Reviewed-by: Jordan Justen <[email protected]> Reviewed-by: Matt Turner <[email protected]>
* glsl: Use parse_program_resource_name to parse transform feedback varyings.Paul Berry2013-02-042-19/+16
| | | | | | | | | | | | | | | | | | | Previously, transform feedback varyings were parsed in an ad-hoc fashion that wasn't compatible with structs (or array of structs). This patch makes it use parse_program_resource_name(), which correctly handles both. Note that parse_program_resource_name()'s technique for handling mal-formed input strings is to simply let them through and rely on the fact that a future name lookup will fail. Because of this, tfeedback_decl::init() no longer needs to return a boolean error code--it always succeeds, and if the input was mal-formed the error will be detected later. NOTE: This is a candidate for the 9.1 branch. Reviewed-by: Jordan Justen <[email protected]> Reviewed-by: Matt Turner <[email protected]>
* glsl: Rename uniform_field_visitor to program_resource_visitor.Paul Berry2013-02-044-33/+34
| | | | | | | | | | | | | | | | There's actually nothing uniform-specific in uniform_field_visitor. It is potentially useful for all kinds of program resources (in particular, future patches will use it for transform feedback varyings). This patch renames it to program_resource_visitor, and clarifies several comments, to reflect the fact that it is useful for more than just uniforms. NOTE: This is a candidate for the 9.1 branch. Reviewed-by: Jordan Justen <[email protected]> Reviewed-by: Matt Turner <[email protected]>
* mesa/glsl: Separate parsing logic from _mesa_get_uniform_location.Paul Berry2013-02-044-61/+91
| | | | | | | | | | | | | | | | The parsing logic is moved to a new function in the GLSL module, parse_program_resource_name(). This name was chosen because it should eventually be useful for handling everything that OpenGL 4.3 calls "program resources" (e.g. uniforms, vertex inputs, fragment outputs, and transform feedback varyings). Future patches will make use of this function for linking transform feedback varyings. NOTE: This is a candidate for the 9.1 branch. Reviewed-by: Jordan Justen <[email protected]> Reviewed-by: Matt Turner <[email protected]>
* gallium/egl: Fix include dirs for VPATH buildQuentin Glidic2013-02-041-0/+1
| | | | | | NOTE: This is a candidate for the 9.1 branch. Reviewed-by: Matt Turner <[email protected]> Signed-off-by: Quentin Glidic <[email protected]>
* intel: make sure to setup image dimension in image_from_planar setupAbdiel Janulgue2013-02-041-0/+1
| | | | | | | | Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=60212 Tested-by: Scott Moreau <[email protected]> Tested-by: Tiago Vignatti <[email protected]> Reviewed-by: Chad Versace <[email protected]> Signed-off-by: Abdiel Janulgue <[email protected]>
* builtin_compiler/build: Don't use *_FOR_BUILD when not cross compilingMatt Turner2013-02-041-11/+32
| | | | | | | | | | | | | | | | | | | | | | | | | Previously we were relying on CFLAGS_FOR_BUILD to be the same as CFLAGS when not cross compiling, but this assumption didn't take into consideration 32-bit builds on 64-bit systems. More generally, not honoring CFLAGS is bad. Automake is evidently too stupid to accept if CROSS_COMPILING CC = @CC_FOR_BUILD@ ... else CC = @CC@ endif without warning that CC has been already defined. The warnings are harmless, but I'd prefer to avoid future reports about them, so define proxy variables, which are assigned inside the conditional and then unconditionally assigned to CC et al. NOTE: This is a candidate for the 9.1 branch. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=59737 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=60038
* st/mesa: emit SQRT opcode when driver supports itBrian Paul2013-02-041-7/+21
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-27 10:14:04 +0000