summaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers
Commit message (Collapse)AuthorAgeFilesLines
* r600g: Implement ARB_texture_viewGlenn Kennard2015-11-112-7/+18
| | | | | | Signed-off-by: Glenn Kennard <[email protected]> Reviewed-by: Marek Olšák <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* r600g: Pass conservative depth parameters to hwGlenn Kennard2015-11-116-0/+46
| | | | | | | Supported on R700 and up. Signed-off-by: Glenn Kennard <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* vc4: Avoid loading undefined (newly-allocated) FBO contents.Eric Anholt2015-11-091-0/+17
| | | | | | | Since X has undefined contents in new pixmaps, it will allocate new textures for an FBO and draw to them without an explicit clear. For VC4, it's much faster to emit a clear than the load of the actual undefined memory contents, so just do that instead.
* vc4: Return NULL when we can't make our shadow for a sampler view.Eric Anholt2015-11-091-0/+4
| | | | | | | I'm not sure what the caller does is appropriate (just have a NULL sampler at this slot), but it fixes the immediate crash. Cc: "11.0" <[email protected]>
* vc4: Return GL_OUT_OF_MEMORY when buffer allocation fails.Eric Anholt2015-11-092-19/+32
| | | | | | | I was afraid our callers weren't prepared for this, but it looks like at least for resource creation, mesa/st throws an error appropriately. Cc: "11.0" <[email protected]>
* vc4: Add CL dumping for GL_ARRAY_PRIMITIVE.Eric Anholt2015-11-091-1/+16
|
* vc4: Fix a compiler warning.Eric Anholt2015-11-091-1/+1
|
* nvc0: enable compute support on FermiSamuel Pitoiset2015-11-081-2/+2
| | | | | | | | | | | Altough the compute support is still not complete because textures and surfaces need to be implemented, it allows to launch very simple compute kernel like one which reads reading MP performance counters. This turns on PIPE_CAP_COMPUTE and PIPE_SHADER_COMPUTE. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* nv50/ir: fix emission of s[] args in certain situationsIlia Mirkin2015-11-071-2/+2
| | | | | | | | | | There might only be a single arg (e.g. cvt), so use mode rather than looking at the source directly. Also we don't want to rely on the type of the value, which can be unreliable, but instead use the instruction's. This works out well since mkSplit doesn't adjust the type. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50/ir: only take abs value when computing high resultIlia Mirkin2015-11-071-1/+1
| | | | | | | | Not reachable from TGSI since it only has UMUL, no IMUL. However it's surprising that setting argument types to s32 will cause sign to get lost. Signed-off-by: Ilia Mirkin <[email protected]>
* nouveau: avoid queueing too much work onto a single fenceIlia Mirkin2015-11-072-26/+43
| | | | | | | | | | Force the fence to get kicked off, which won't actually wait for its completion, but any additional work will be put onto a fresh list. This fixes crashes in teximage-colors --benchmark with too many active maps. Signed-off-by: Ilia Mirkin <[email protected]>
* llvmpipe: disable front updates for nowDave Airlie2015-11-081-1/+1
| | | | | | | | As pointed out by Emil, this sometimes hangs, appears to be due to threading need to rethink how this stuff works for llvmpipe. Signed-off-by: Dave Airlie <[email protected]>
* radeonsi: add register definitions for StoneyMarek Olšák2015-11-071-0/+322
| | | | | | There are a few non-stoney changes too. Reviewed-by: Alex Deucher <[email protected]>
* radeonsi: add workarounds for CP DMA to stay on the fast pathMarek Olšák2015-11-071-5/+88
| | | | | | v2: set emit_scratch_reloc, add a NULL check Reviewed-by: Michel Dänzer <[email protected]>
* radeonsi: unify CP DMA preparation logicMarek Olšák2015-11-071-37/+34
| | | | Reviewed-by: Michel Dänzer <[email protected]>
* radeonsi: unify CP DMA code determining various flagsMarek Olšák2015-11-071-28/+23
| | | | | | v2: don't call get_flush_flags twice per function Reviewed-by: Michel Dänzer <[email protected]>
* radeonsi: only enable write confirmation on the last CP DMA packetMarek Olšák2015-11-071-2/+4
| | | | | | This should improve performance for big copies that need to be split. Reviewed-by: Michel Dänzer <[email protected]>
* nv50/ir: allow emission of immediates in imul/imad opsIlia Mirkin2015-11-071-2/+8
| | | | | | | Nothing actually uses this yet (due to complications), but the emission logic is right. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50/ir: properly set the type of the constant folding resultIlia Mirkin2015-11-061-4/+4
| | | | | | | This removes the hack used for merge, which only covers a fraction of the cases. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50/ir: add support for const-folding OP_CVT with F64 source/destIlia Mirkin2015-11-063-0/+45
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* nv50/ir: add fp64 opcode emission support for G200 (NVA0)Ilia Mirkin2015-11-061-10/+84
| | | | | | Need to emulate rcp/rsq before providing full fp64 support Signed-off-by: Ilia Mirkin <[email protected]>
* nv50/ir: Add support for 64bit immediates to checkSwapSrc01Hans de Goede2015-11-061-5/+6
| | | | | | | | Now that we support 64 bit immediates in insnCanLoad, we need to swap 64 bit immediate sources too for optimal effect. Signed-off-by: Hans de Goede <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* nvc0/ir: Teach insnCanLoad about double immediatesHans de Goede2015-11-061-6/+19
| | | | | | | | | | | | | | | | Teach insnCanLoad about double immediates, together with the "Add support for merge-s to the ConstantFolding pass" This turns the following (nvc0) code: 1: mov u32 $r2 0x00000000 (8) 2: mov u32 $r3 0x3fe00000 (8) 3: add f64 $r0d $r0d $r2d (8) Into: 1: add f64 $r0d $r0d 0.500000 (8) Signed-off-by: Hans de Goede <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* nv50/ir: Add support for merge-s to the ConstantFolding passHans de Goede2015-11-061-0/+15
| | | | | | | | | | | This allows later passes like LoadPropagation to properly deal with 64 bit immediates. If the new 64 bit load this introduces does not get optimized away then split64BitOpPostRA() will split this into 2 instructions again. Signed-off-by: Hans de Goede <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* nv50/ir: disallow 64-bit immediates on nv50 targetsIlia Mirkin2015-11-061-1/+1
| | | | | | No instructions are able to load short immediates like nvc0 can. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50/ir: allow movs with TYPE_F64 destinations to be splitIlia Mirkin2015-11-061-0/+6
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* gm107/ir: Add support for double immediatesHans de Goede2015-11-061-1/+4
| | | | | | | | Add support for encoding double immediates (up to 20 bits of precision) into the generated gm107 machine-code. Signed-off-by: Hans de Goede <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* nvc0/ir: Add support for double immediatesHans de Goede2015-11-061-0/+8
| | | | | | | | Add support for encoding double immediates (up to 20 bits of precision) into the generated nvc0 machine-code. Signed-off-by: Hans de Goede <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* radeon/uvd: fix VC-1 simple/main profile decode v2Boyuan Zhang2015-11-062-2/+7
| | | | | | | | | | | We just needed to set the extra width/height fields to get this working. v2 (chk): rebased, CC stable added, commit message added, fixed coding style Signed-off-by: Boyuan Zhang <[email protected]> Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Cc: "10.6 11.0" <[email protected]>
* freedreno/a4xx: fix blend colorRob Clark2015-11-061-5/+9
| | | | Signed-off-by: Rob Clark <[email protected]>
* freedreno: update generated headersRob Clark2015-11-066-43/+54
| | | | Signed-off-by: Rob Clark <[email protected]>
* freedreno: add a305 supportGuillaume Charifi2015-11-061-0/+1
| | | | Signed-off-by: Rob Clark <[email protected]>
* freedreno/ir3: Use nir_foreach_variableBoyan Ding2015-11-061-3/+3
| | | | | Signed-off-by: Boyan Ding <[email protected]> Signed-off-by: Rob Clark <[email protected]>
* nvc0: reintroduce BGRA4 format supportIlia Mirkin2015-11-062-3/+1
| | | | | | | | | | | | | | Commit 342e68dc60 (nvc0: remove BGRA4 format support) removed the support to fix a WoW trace. However after further experimentation, I was able to get the blit to work by using a different "fake" format in the 2d engine. The reason why this worked on nv50 is that nv50 falls back to the 3d blit path in case either the src or the dst aren't "faithfully" supported, while nvc0 only does it for the dst format. RG8 is better supported by the nvc0 2d engine than R16. Signed-off-by: Ilia Mirkin <[email protected]>
* llvmpipe: disable texture cacheRoland Scheidegger2015-11-051-1/+1
| | | | There are some weird problems with 8-wide vectors.
* nouveau: send back a debug message when waiting for a fence to completeIlia Mirkin2015-11-0510-16/+30
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* nv50,nvc0: provide debug messages with shader compilation statsIlia Mirkin2015-11-0511-9/+28
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* nouveau: add support for sending debug messages via KHR_debugIlia Mirkin2015-11-055-0/+26
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* nouveau: relax fence emit space assertIlia Mirkin2015-11-043-3/+3
| | | | | | | | | We also have the "reserved for kick" space available. Some of my earlier changes can probably be removed, but this is a quick fix for some of the rarer fallout. Signed-off-by: Ilia Mirkin <[email protected]> Cc: <[email protected]>
* vc4: When the create ioctl fails, free our cache and try again.Eric Anholt2015-11-041-5/+24
| | | | | | | | | This greatly increases the pressure you can put on the driver before create fails. Ultimately we need to let the kernel take control of our cached BOs and just take them from us (and other clients) directly, but this is a very easy patch for the moment. Cc: "11.0" <[email protected]>
* vc4: Print the rounded shader size in debug output.Eric Anholt2015-11-041-1/+1
| | | | | It's surprising to see "0kb" printed for debug on short shaders, while 4kb alignment won't be suprising.
* vc4: Fix dumping the size of BOs allocated/cached.Eric Anholt2015-11-041-2/+2
| | | | 60MB of cached BOs are a lot less scary than 600MB.
* svga: implement 'white_fragments' option for VGPU10 fragment shadersBrian Paul2015-11-041-5/+30
| | | | | | | | | | When we emulate XOR logicop mode with blend-subtract, we need to ensure that the fragment shader always emits white. We had this implemented for VGPU9, but not VGPU10. VMware bug 1545492. Reviewed-by: Charmaine Lee <[email protected]>
* svga: use new enum indices_mode typeBrian Paul2015-11-042-2/+4
| | | | Reviewed-by: Charmaine Lee <[email protected]>
* llvmpipe: add cache for compressed texturesRoland Scheidegger2015-11-047-10/+109
| | | | | | | | | | | | | | | | | | | | | | compressed textures are very slow because decoding is rather complex (and because there's no jit code code to decode them too for non-technical reasons). Thus, add some texture cache which holds a couple of decoded blocks. Right now this handles only s3tc format albeit it could be extended to work with other formats rather trivially as long as the result of decode fits into 32bit per texel (ideally, rgtc actually would decode to more than 8 bits per channel, but even then making it work for it shouldn't be too difficult). This can improve performance noticeably but don't expect wonders (uncompressed is unsurprisingly still faster). It's also possible it might be slower in some cases (using nearest filtering for example or if there's otherwise not many cache hits, the cache is only direct mapped which isn't great). Also, actual decode of a block relies on util code, thus even though always full blocks are decoded it is done texel by texel - this could obviously benefit greatly from simd-optimized code decoding full blocks at once... Note the cache is per (raster) thread, and currently only used for fragment shaders. Reviewed-by: Jose Fonseca <[email protected]>
* llvmpipe: use simple coeffs calc for 128bit vectorsOded Gabbay2015-11-041-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are currently two methods in llvmpipe code to calculate coeffs to be used as inputs for the fragment shader. The two methods use slightly different ways to do the floating point calculations and thus produce slightly different results. The decision which method to use is determined by the size of the vector that is used by the platform. For vectors with size of more than 128bit, a single-step method is used, in which coeffs_init_simple() + attribs_update_simple() are called. For vectors with size of 128bit or less, a two-step method is used, in which coeffs_init() + attribs_update() are called. This causes some piglit tests (clip-distance-bulk-copy, interface-vs-unnamed-to-fs-unnamed) to fail when using platforms with 128bit vectors (such as ppc64le or x86-64 without AVX). This patch makes platforms with 128bit vectors use the single-step method (aka "simple" method) instead of the two-step method. This would make the resulting coeffs identical between more platforms, make sure the piglit tests passes, and make debugging and maintainability a bit easier as the generated LLVM IR will be the same for more platforms. The performance impact is negligible for x86-64 without AVX, and basically non-existent for ppc64le, as it can be seen from the following benchmarking results: - glxspheres, on ppc64le: - original code: 4.892745317 frames/sec 5.460303857 Mpixels/sec - with the patch: 4.932083873 frames/sec 5.504205571 Mpixels/sec - Additional 0.8% performance boost - glxspheres, on x86-64 without AVX: - original code: 20.16418809 frames/sec 22.50323395 Mpixels/sec - with the patch: 20.31328989 frames/sec 22.66963152 Mpixels/sec - Additional 0.74% performance boost - glmark2, on ppc64le: - original code: score of 58 - with my change: score of 57 - glmark2, on x86-64 without AVX: - original code: score of 175 - with the patch: score of 167 - Impact of of -4.5% on performance - OpenArena, on ppc64le: - original code: 3398 frames 1719.0 seconds 2.0 fps 255.0/505.9/2773.0/0.0 ms - with the patch: 3398 frames 1690.4 seconds 2.0 fps 241.0/497.5/2563.0/0.2 ms - 29 seconds faster with the patch, which is about 2% - OpenArena, on x86-64 without AVX: - original code: 3398 frames 239.6 seconds 14.2 fps 38.0/70.5/719.0/14.6 ms - with the patch: 3398 frames 244.4 seconds 13.9 fps 38.0/71.9/697.0/14.3 ms - 0.3 fps slower with the patch (about 2%) Additional details can be found at: http://lists.freedesktop.org/archives/mesa-dev/2015-October/098635.html Signed-off-by: Oded Gabbay <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* gallium/radeon: allow returning SDMA fences from pipe->flushMarek Olšák2015-11-041-8/+56
| | | | | | | pipe->flush never returned SDMA fences. This fixes it. This is only an issue on amdgpu where fences can signal out of order. Reviewed-by: Michel Dänzer <[email protected]>
* gallium/radeon: always return the last SDMA fence on SDMA flush if neededMarek Olšák2015-11-042-4/+8
| | | | Reviewed-by: Michel Dänzer <[email protected]>
* nvc0: add missing compute parameters required by cloverSamuel Pitoiset2015-11-031-1/+10
| | | | | | | This fixes crashes with some piglit OpenCL tests. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* nvc0: handle NULL pointer in nvc0_get_compute_param()Samuel Pitoiset2015-11-031-24/+21
| | | | | | | | | To get the size (in bytes) of a compute parameter, clover first calls get_compute_param() with a NULL data pointer. The RET() macro is based on nv50. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-28 19:38:27 +0000