summaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* gallivm: JIT symbol resolution with linux perf.José Fonseca2013-04-175-59/+101
| | | | | | | Details on docs/llvmpipe.html Reviewed-by: Brian Paul <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* draw: Silence uninitialized var warnings.José Fonseca2013-04-172-3/+7
| | | | Trivial.
* r600g/llvm: Use gprcount from llvmVincent Lejeune2013-04-174-2/+5
|
* gallium: Disambiguate TGSI_OPCODE_IF.José Fonseca2013-04-1715-24/+120
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | TGSI_OPCODE_IF condition had two possible interpretations: - src.x != 0.0f - Mesa statetracker when PIPE_SHADER_CAP_INTEGERS was false either for vertex and fragment shaders - gallivm/llvmpipe - postprocess - vl state tracker - vega state tracker - most old drivers - old internal state trackers - many graw examples - src.x != 0U - Mesa statetracker when PIPE_SHADER_CAP_INTEGERS was true for both vertex and fragment shaders - tgsi_exec/softpipe - r600 - radeonsi - nv50 And drivers that use draw module also were a mess (because Mesa would emit float IFs, but draw module supports native integers so it would interpret IF arg as integers...) This sort of works if the source argument is limited to float +0.0f or +1.0f, integer 0, but would fail if source is float -0.0f, or integer in the float NaN range. It could also fail if source is integer 1, and hardware flushes denormalized numbers to zero. But with this change there are now two opcodes, IF and UIF, with clear meaning. Drivers that do not support native integers do not need to worry about UIF. However, for backwards compatibility with old state trackers and examples, it is advisable that native integer capable drivers also support the float IF opcode. I tried to implement this for r600 and radeonsi based on the surrounding code. I couldn't do this for nouveau, so I just shunted IF/UIF together, which matches the current behavior. Reviewed-by: Roland Scheidegger <[email protected]> Reviewed-by: Marek Olšák <[email protected]> v2: - Incorporate Roland's feedback. - Fix r600_shader.c merge conflict. - Fix typo in radeon, spotted by Michel Dänzer. - Incorporte Christoph Bumiller's patch to handle TGSI_OPCODE_IF(float) properly in nv50/ir.
* gallium: Eliminate TGSI_OPCODE_IFC.José Fonseca2013-04-178-14/+8
| | | | | | Never used or implemented. Reviewed-by: Roland Scheidegger <[email protected]>
* r600/uvd: cleanup disabling tiling on pre EG asicsChristian König2013-04-161-5/+6
| | | | | | Set transfer flag instead of fiddling with the tilling params directly. Signed-off-by: Christian König <[email protected]>
* r600g: Workaround for a harware bug with nested loops on CaymanMartin Andersson2013-04-161-3/+15
| | | | | | | | | | | | | | | | There is a hardware bug on Cayman where a BREAK/CONTINUE followed by LOOP_STARTxxx for nested loops may put the branch stack into a state such that ALU_PUSH_BEFORE doesn't work as expected. Workaround this by replacing the ALU_PUSH_BEFORE with a PUSH + ALU Fixes piglit tests EXT_transform_feedback/order* v2: Use existing loop count and improve comment v3: [Vadim Girlin] Set jump address for PUSH instructions NOTE: This is a candidate for the 9.1 branch Signed-off-by: Vadim Girlin <[email protected]>
* gallium/hud: fix FPS computation for framerate > 4.2kMarek Olšák2013-04-161-1/+2
|
* gallium/hud: increase vertex buffer size for background black rectanglesMarek Olšák2013-04-161-1/+1
| | | | Reviewed-by: Brian Paul <[email protected]>
* gallium/hud: update the contents of GALLIUM_HUD=helpMarek Olšák2013-04-161-2/+17
| | | | Reviewed-by: Brian Paul <[email protected]>
* gallium/hud: remove pipeline-statistics- prefix in query namesMarek Olšák2013-04-161-21/+22
| | | | | | | | for the env var string not to be awfully long v2: fix bug in indexing of "name" Reviewed-by: Brian Paul <[email protected]>
* r600g: implement pipeline statistics queryMarek Olšák2013-04-164-1/+100
|
* winsys/radeon: use query_value for timestamp, remove query_timestampMarek Olšák2013-04-163-26/+13
|
* r600g: add a debug flag for printing virtual addresses of resourcesMarek Olšák2013-04-164-0/+17
|
* r600g: add a query returning the amount of time spent during bo_map sync.Marek Olšák2013-04-167-2/+22
|
* build: Get rid of GALLIUM_WINSYS_DIRSMatt Turner2013-04-152-38/+65
| | | | | | | | configure still uses it to print the enabled winsys. Tested-by: Emil Velikov <[email protected]> Reviewed-and-Tested-by: Andreas Boll <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* build: Get rid of GALLIUM_TARGET_DIRSMatt Turner2013-04-151-1/+151
| | | | | | | | configure still uses it to print the enabled targets. Tested-by: Emil Velikov <[email protected]> Reviewed-and-Tested-by: Andreas Boll <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* build: Build pipe-loader before gallium testsMatt Turner2013-04-152-7/+0
| | | | | | | | | And don't build it from other Makefiles. That's awful, and breaks distclean. Tested-by: Emil Velikov <[email protected]> Reviewed-and-Tested-by: Andreas Boll <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* build: Get rid of GALLIUM_MAKE_DIRSMatt Turner2013-04-152-19/+71
| | | | | | Tested-by: Emil Velikov <[email protected]> Reviewed-and-Tested-by: Andreas Boll <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* build: Stop using GALLIUM_STATE_TRACKERS_DIRS for SUBDIRSMatt Turner2013-04-151-11/+54
| | | | | | | | configure still uses it to print the enabled state trackers. Tested-by: Emil Velikov <[email protected]> Reviewed-and-Tested-by: Andreas Boll <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* build: Remove GALLIUM_DIRSMatt Turner2013-04-151-22/+0
| | | | | | | | It's always constant anyway. Tested-by: Emil Velikov <[email protected]> Reviewed-and-Tested-by: Andreas Boll <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* build: Move src/mapi/mapi/* to src/mapi/Matt Turner2013-04-153-4/+4
| | | | | | Tested-by: Emil Velikov <[email protected]> Reviewed-and-Tested-by: Andreas Boll <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* radeonsi: Read config values from the .AMDGPU.config ELF sectionTom Stellard2013-04-152-8/+28
| | | | | | | | | Instead of emitting configuration values (e.g. number of gprs used) in a predefined order, the LLVM backend now emits these values in register/value pairs. The first dword contains the register address and the second dword contians the value to write. Reviewed-by: Michel Dänzer <[email protected]>
* radeon/llvm: Handle ELF formatted binary output from the LLVM backendTom Stellard2013-04-153-6/+37
|
* radeon/llvm: Use a struct for storing compiled codeTom Stellard2013-04-154-28/+38
|
* gallivm: fix small but severe bug in handling multiple lod level stridesRoland Scheidegger2013-04-151-1/+1
| | | | | | | | | | | | | | | | | Inserting the value for the second quad in the wrong place for the following shuffle. This meant the row or image stride was undefined which is quite catastrophic, can lead to bogus texels fetched or just segfault. This code is only hit for SoA path currently, still surprising it didn't crash more or caused more visible issues (I think llvm used a broadcast shuffle for the undefined parts of the vector, hence the undefined value for the second quad was just the same as that from the first quad, so as long as both quads hit the same mip level everything was fine, and since lower mips always have the same large stride it made it less likely to hit out-of-bound memory in case of differing lods). Note: this is a candidate for stable branches. Reviewed-by: Jose Fonseca <[email protected]>
* clover: Fix usage of incorrect object as destination in ↵Francisco Jerez2013-04-131-1/+1
| | | | | | clEnqueueCopyBufferToImage. Signed-off-by: Francisco Jerez <[email protected]>
* clover: Define platform class and merge with device_registry.Francisco Jerez2013-04-138-68/+155
| | | | | | | | | | | | | Null platform IDs are OK according to the spec, but some applications have been reported to get paranoid and assume that our NULL platform is unusable. As it doesn't hurt to have device enumeration separate from the rest of the device code (quite the opposite, it makes the code cleaner), make the API use an actual platform object that keeps track of the available devices instead of the former NULL pointer. Reported-and-reviewed-by: Tom Stellard <[email protected]> Signed-off-by: Francisco Jerez <[email protected]>
* clover: Add missing fields to the module serializer.Francisco Jerez2013-04-131-0/+2
| | | | Signed-off-by: Francisco Jerez <[email protected]>
* r300g: Fix bug in OMOD optimizationTom Stellard2013-04-121-1/+3
| | | | | | https://bugs.freedesktop.org/show_bug.cgi?id=60503 NOTE: This is a candidate for the stable branches.
* nvc0: set ret variable if launch desc allocation failedEmil Velikov2013-04-121-1/+3
| | | | | | | | | | | | | | | Pointed out by gcc nve4_compute.c: In function 'nve4_launch_grid': nve4_compute.c:511:7: warning: 'ret' may be used uninitialized in this function [-Wmaybe-uninitialized] if (ret) ^ Signed-off-by: Emil Velikov <[email protected]> Edit by Christoph Bumiller: Set it to -1 to indicate failure and only when it's actually required.
* nvc0: bail out early during nve4_compute_setup()Emil Velikov2013-04-121-1/+1
| | | | | | | Exit gracefully rather than trying to create a random object, whenever the chipset is unknown Signed-off-by: Emil Velikov <[email protected]>
* nvc0: compile nve4_cache_split_name() only in debug buildEmil Velikov2013-04-121-1/+1
| | | | | | | | | | As otherwise it is unused - pointed out by gcc nve4_compute.c:586:20: warning: 'nve4_cache_split_name' defined but not used [-Wunused-function] static const char *nve4_cache_split_name(unsigned value) ^ Signed-off-by: Emil Velikov <[email protected]>
* nv50/codegen: do not emitATOM() if the subOp is unknownEmil Velikov2013-04-121-1/+1
| | | | | | | | | | | | For debug build we'll hit the assert, for release we are going to emit random data as subOp is used uninitilised. Spotted by gcc codegen/nv50_ir_emit_nv50.cpp: In member function 'void nv50_ir::CodeEmitterNV50::emitATOM(const nv50_ir::Instruction*)': codegen/nv50_ir_emit_nv50.cpp:1554:12: warning: 'subOp' may be used uninitialized in this function [-Wmaybe-uninitialized] uint8_t subOp; ^ Signed-off-by: Emil Velikov <[email protected]>
* nvc0: implement multisample texturesChristoph Bumiller2013-04-1210-25/+97
|
* nvc0: patch up TEX cases with 5 or 6 sources on nve4Christoph Bumiller2013-04-121-1/+19
| | | | Hackishly fixes alignment requirement of 2nd tuple for now.
* nvc0: fix 2D engine MS2 resolveChristoph Bumiller2013-04-121-2/+2
|
* nv50,nvc0: add RGBX16/32_FLOAT formatsChristoph Bumiller2013-04-121-0/+4
|
* r600g: add get_sample_position support (v3)Dave Airlie2013-04-112-122/+240
| | | | | | | v2: I rewrote this to use the sample positions properly. v3: rewrite properly to use bitfield to cast back to signed ints Signed-off-by: Dave Airlie <[email protected]>
* gallium: add get_sample_position interfaceDave Airlie2013-04-111-0/+13
| | | | | | | This is to be used to implement glGet GL_SAMPLE_POSITION. Reviewed-by: Marek Olšák <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* r600g: fix two issues in compressed msaa reading codeDave Airlie2013-04-111-2/+2
| | | | | | | | | I've no idea when sample_chan would ever be 4 here, but 4 is most definitely wrong, array textures have it as 3 as well. Also the cayman code though unused is obviously wrong. Signed-off-by: Dave Airlie <[email protected]>
* radeon/uvd: add UVD implementation v5Christian König2013-04-1115-21/+1924
| | | | | | | | | | | | | | Just everything you need for UVD with r600g and radeonsi. v2: move UVD code to radeon subdir, clean up build system additions, remove an unused SI function, disable tiling on SI for now. v3: some minor indentation fix and rebased v4: dpb size calculation fixed v5: implement proper fall-back in case the kernel doesn't support UVD, based on patches from Andreas Boll but cleaned up a bit more. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]>
* radeon/winsys: add uvd ring support to winsys v3Christian König2013-04-113-0/+31
| | | | | | | | | | | | | Separated from UVD patch for clarity. v2: sync with next tree for 3.10 v3: as pointed out by Andreas Bool check for drm minor >= 32 http://cgit.freedesktop.org/~agd5f/linux/log/?h=drm-next-3.10-wip Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Andreas Boll <[email protected]>
* r600g: Add support for GL_ARB_texture_buffer_rangeFredrik Höglund2013-04-113-5/+11
| | | | Reviewed-by: Marek Olšák <[email protected]>
* r600g: fix valgrind warning on CaymanMarek Olšák2013-04-101-1/+1
| | | | Warning: "Conditional jump or move depends on uninitialised value(s)".
* gallivm/tgsi: handle untyped movesZack Rusin2013-04-102-0/+10
| | | | | | | | | | | both mov and ucmp can be used to move variables of any type. correctly note that about ucmp in the tgsi_info and make sure gallivm can handle that by correctly casting the untyped moves. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: Jose Fonseca <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* gallivm: fix loops and conditionals within GSZack Rusin2013-04-102-19/+105
| | | | | | | | | | | | | We were using simple temporaries, without using alloca or phi nodes which meant that on every iteration of the loop our temporaries, which were holding the number of vertices and primitives which were emitted, were being reset to zero. Now we're using alloca to allocate those variables to preserve them across conditionals. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: Jose Fonseca <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* llvmpipe: implement PIPE_QUERY_SO_STATISTICSZack Rusin2013-04-102-0/+21
| | | | | | | | | | We were missing the implementation of PIPE_QUERY_SO_STATISTICS query, this change implements it on top of the existing facilities. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: Jose Fonseca <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* gallivm: fix unsigned divide and remainder opcodesZack Rusin2013-04-101-4/+33
| | | | | | | | | | We want to both make sure we never divide by zero to not generate sigfpe and that divide by zero is guaranteed to return 0xffffffff. Based on José idea. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: Jose Fonseca <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* gallivm: fix breakcZack Rusin2013-04-101-12/+14
| | | | | | | | | we break when the mask values are 0 not, 1, plus it's bit comparison not a floating point comparison. This fixes both. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: Jose Fonseca <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>