aboutsummaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers/llvmpipe
Commit message (Collapse)AuthorAgeFilesLines
* llvmpipe: do transpose/untwiddle after conversion for 8bit formatsRoland Scheidegger2017-01-061-7/+143
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Generally we should do tranpose after conversion, if the format has less than 32 bits per channel (if it has 32 bits, conversion is going to be a no-op anyway...). This is obviously because there's less vectors to deal with. Though the advantage for 16 bit formats isn't that big, and in fact with AVX there isn't really any (as the 32bit unpacks can be done with 256bit, but the smaller ones cannot, although that would change again with proper AVX2 support). Only makes sense for 2d and not 1d cases. And to keep things easy, only handle 1,2 and 4 channels (rgbx is just fine). For rgba unorm8 format the backend conversion sums up to these instruction totals (not counting the movs for SSE2 due to 2-op syntax - generally every 2 unpacks need an additional mov). SSE2 AVX transpose: 32 unpack 16 unpack untwiddle: 0 8 (128bit low/high permutes) convert: 16 mul + 16 cvt 8 mul + 8 cvt 32->8bit: 12 pack 8 (128bit extract) + 12 pack When doing transpose/untwiddle afterwards we get: convert: 16 mul + 16 cvt 8 mul + 8 cvt 32->8bit: 12 pack 8 (128bit extract) + 12 pack transpose/untwiddle 12 unpack 12 unpack So for SSE2, this drops 20 unpacks (total instruction count 76->56) whereas for AVX it replaces the 16 256bit unpacks with 8 128bit ones and drops the 8 lo/hi permutes (in total 60->48). (Albeit to be fair, the permutes could be dropped even when doing the transpose first, they are extremely pointless but we'd need to be able to tell lp_build_conv to reorder the vectors, for AVX2 we're going to need to be able to tell lp_build_conv about ordering in any case.) (With different ordering going into conversion, it would be possible to do 4 unpacks + 4 pshufbs instead of 12 unpacks, but that might not be better, and not all cpus can do it. Proper AVX2 support should eliminate the 8 128bit extracts, reduce these 12 packs to 6 and the 12 unpacks to 2 pshufb + 2 permq ideally (+ 2 final 128bit extracts).) Reviewed-by: Jose Fonseca <[email protected]>
* llvmpipe: use alpha from already converted color if possibleRoland Scheidegger2017-01-062-18/+54
| | | | | | | | | | | | | For rgbx formats, there is no point in doing alpha conversion again (and with different tranpose even, so llvm can't eliminate it). Albeit it looks like there's some minimal changes needed in the blend code (found by code inspection, no test seemed to complain) if we do this - the blend factors are already sanitized if we have no destination alpha, however for src_alpha_saturate it looks like it still might make a difference (note that we forced has_alpha to true before for some formats and nothing complained, but this seems safer). Reviewed-by: Jose Fonseca <[email protected]>
* llvmpipe: use scalar load instead of vectors for small vectors in fs backendRoland Scheidegger2017-01-061-6/+50
| | | | | | | | | | | | | | | | | | | | | | | | | | llvm has _huge_ problems trying to load things like <4 x i8> vectors and stitching such loads together to form 128bit vectors. My understanding of the problem is that the type legalizer tries to extend that to really a <4 x i32> vector and not a <16 x i8> vector with the 4 elements first then followed by padding, so the shuffles for then combining things together are more or less impossible - you can in fact see the pmovzxd llvm generates. Pre-4.0 llvm just gives up on it completely and does a 30+ pextrb/pinsrb sequence instead. It looks like current llvm has fixed this behavior (my guess would be due to better shuffle combination and load/shuffle folds), but we can avoid this by just loading as <1 x i32> values, combine that and only cast at the end. (I suspect it might also work if we'd pad the loaded vectors immediately before shuffling them together, instead of directly stitching 2 such vectors together pairwise before combining the pair. But this _might_ lose the ability to load the values directly into their right place in the vector with pinsrd.). But using 32bit values is probably easier for llvm as it will never give it funny ideas how the vector should look like. (This is possibly only a problem for 1x8bit formats, since 2x8bit will end up fetching 64bit hence only two vectors are stitched together, not 4, but we use the same strategy anyway.) Reviewed-by: Jose Fonseca <[email protected]>
* gallivm: implement aos unpack (to unorm8) for small unorm formatsRoland Scheidegger2017-01-051-5/+3
| | | | | | | | | | | | | | | | | | | Using bit replication. This path now resembles something which might make sense. (The logic was mostly copied from llvmpipe fs backend.) I am not convinced though it is actually faster than SoA sampling (actually I'm quite certain it's always a loss with AVX). With SoA it's just shift/mask/cvt/mul for getting the colors, whereas there's still roughly 3 shifts, 3 or/and per channel for AoS (i.e. for SoA it's exactly the same as it would be for a rgba8 format, whereas the extra effort for AoS is significant). The filtering might still be faster (albeit with FMA the instruction count gets down quite a bit there on the SoA float filtering path on new cpus). And those small unorm formats often don't have an alpha channel (which makes things worse relatively for AoS path). (This also fixes a trivial bug in the llvmpipe fs code this was derived from, albeit it was only relevant for 4-bit channels.) Reviewed-by: Jose Fonseca <[email protected]>
* llvmpipe: (trivial) minimally simplify mask constructionRoland Scheidegger2017-01-052-17/+51
| | | | | | | | | | | | | | | | simd instruction sets usually have comparisons for equal, not unequal. So use a different comparison against the mask itself - which also means we don't need a all-zero as well as a all-one (for the pxor) reg. Also add code to avoid scalar expansion of i1 values which we definitely shouldn't do. There's problems with this though with llvm select interaction, so it's disabled (basically using llvm select instead of intrinsics may still produce atrocious code, even in cases where we figured it should not, albeit I think this could probably be fixed with some better selection of optimization passes, but I have zero idea there really). Reviewed-by: Jose Fonseca <[email protected]>
* gallium: add PIPE_CAP_GLSL_OPTIMIZE_CONSERVATIVELYMarek Olšák2017-01-051-0/+1
| | | | | | Drivers with good compilers don't need aggressive optimizations before TGSI. Reviewed-by: Eric Anholt <[email protected]>
* llvmpipe: Link tests with CLOCK_LIB.Vinson Lee2016-12-211-1/+2
| | | | | | | | | | Fix linking error with 'make check'. CXXLD lp_test_format ../../../../src/gallium/auxiliary/.libs/libgallium.a(os_time.o): In function `os_time_get_nano': src/gallium/auxiliary/os/os_time.c:59: undefined reference to `clock_gettime' Signed-off-by: Vinson Lee <[email protected]>
* gallium: support for native fence fd'sRob Clark2016-12-011-0/+1
| | | | | | | This enables gallium support for EGL_ANDROID_native_fence_sync, for drivers which support PIPE_CAP_NATIVE_FENCE_FD. Signed-off-by: Rob Clark <[email protected]>
* gallium: add PIPE_CAP_TGSI_CAN_READ_OUTPUTSNicolai Hähnle2016-11-301-0/+1
| | | | | | | | | | | Drivers that support this benefit by saving one lowering pass in the GLSL-to-TGSI conversion. radeonsi already supports this because all outputs are stored in temporary variables before the export (except for TCS outputs, which have always been readable in TGSI anyway due to their special semantics). Reviewed-by: Marek Olšák <[email protected]>
* llvmpipe: Fix build after removal of deprecated attribute API v2Aaron Watry2016-11-092-3/+2
| | | | | | | | | | | | Applies on top of v3 of Tom's gallivm change. v2: - Tom Stellard: Use enums instread of strings. Reviewed-by: Nicolai Hähnle <[email protected]> Signed-off-by: Aaron Watry <[email protected]> CC: Tom Stellard <[email protected]> CC: Jan Vesely <[email protected]>
* gallium: add PIPE_CAP_STREAM_OUTPUT_INTERLEAVE_BUFFERSIlia Mirkin2016-10-221-0/+1
| | | | | | | | | | | | | | This allows the driver to signal that it can't handle random interleaving of attributes across buffers. This is required for ARB_transform_feedback3, and it's initialized to whatever the previous value of PIPE_CAP_STREAM_OUTPUT_PAUSE_RESUME was except for nv50 where it is disabled. Note that the proprietary drivers never expose ARB_transform_feedback3 on any GT21x's (where nouveau previously did), and after some effort I was unable to get it to work. Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* st/mesa: enable ARB_enhanced_layouts and turn the cap onNicolai Hähnle2016-10-121-1/+1
| | | | | | | v2: mark llvmpipe & softpipe properly as well (Jason Wood) Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* gallium: add PIPE_CAP_TGSI_ARRAY_COMPONENTSNicolai Hähnle2016-10-121-0/+1
| | | | | | | | This is a screen cap because drivers are expected to support it either for all shader types or for none of them. Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* llvmpipe: Cap to 2 GB on 32 bitsAxel Davy2016-10-101-0/+6
| | | | | | | | | | | | | | On 32 bits system, application memory is quite limited. llvmpipe uses application memory. To help prevent memory exhaustion, limit reported memory availability to 2GB. Some gallium nine apps do check reported memory by allocating resources until memory is full. Gallium nine refuses allocations when 80% of the reported memory limit is used. This change helps some apps to start. Signed-off-by: Axel Davy <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* llvmpipe: (trivial) honor render_condition_enabled for clear_rt/clear_dsRoland Scheidegger2016-08-291-2/+2
|
* gallium: Use enum pipe_shader_type in set_sampler_views()Kai Wasserbäch2016-08-291-1/+1
| | | | | Signed-off-by: Kai Wasserbäch <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* gallium: Use enum pipe_shader_type in bind_sampler_states() (v2)Kai Wasserbäch2016-08-291-1/+1
| | | | | | | | | | | v1 → v2: - Fixed indentation (noted by Brian Paul) - Removed second assert from nouveau's switch statements (suggested by Brian Paul) Signed-off-by: Kai Wasserbäch <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* gallium: add a pipe_context parameter to resource_get_handleMarek Olšák2016-08-251-0/+1
| | | | | | | | radeonsi needs to do some operations (DCC decompression) for OpenGL-OpenCL interop and this is the only way to make it coherent with the current context. It can optionally be set to NULL. Reviewed-by: Brian Paul <[email protected]>
* gallium: add a cap to expose whether driver supports mixed color/zs bitsIlia Mirkin2016-08-231-0/+1
| | | | | | | | | | Some hardware can't render to color/depth buffers of mixed bitness. When that happens a fallback has to happen, but this allows the driver to express that this isn't an optimal scenario. The purpose of this is to remove such fbconfigs from the GLX/EGL config list. Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* llvmpipe: fix issues with depth clampRoland Scheidegger2016-08-203-49/+94
| | | | | | | | | | | | | | | | We only did depth clamp when the value was written from the fs. This is very wrong both for d3d10 and GL, and only passed the corresponding piglit test due to pure luck (it no longer does with the enhanced test). Also, interpolation clamped values to 1.0 always, which can legitimately happen if depth clip is disabled, so fix that as well (untested). There is one unresolved issue left, d3d10 always does depth clamping, whereas GL does not (but does [0,1] clamp instead for fs depth outputs) - this information isn't in any gallium state object, leave it as-is for now (though it looks like llvmpipe misses the [0,1] clamp as well). This (with the previous patch) fixes piglit depth-clamp-range test. Reviewed-by: Jose Fonseca <[email protected]>
* llvmpipe: fix depth clamping wrt reversed near/far valuesRoland Scheidegger2016-08-201-9/+3
| | | | | | | | | | | This wasn't handled before (the result was that no matter what value got clamped, it always ended up as the near value in this case) (if clamping actually happened). Fix this by using the util helper for that (the math is otherwise "mostly" the same, mostly because there could actually be differences due to float rounding, but I don't even know which one would be more correct). Reviewed-by: Jose Fonseca <[email protected]>
* gallium: change pipe_sampler_view::first_element/last_element -> offset/sizeMarek Olšák2016-08-172-12/+8
| | | | | | | | | | | This is required by OpenGL. Our hardware supports this. Example: Bind RGBA32F with offset = 4 bytes. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97305 Acked-by: Ilia Mirkin <[email protected]> Acked-by: Nicolai Hähnle <[email protected]>
* gallium: add a pipe_context parameter to fence_finishMarek Olšák2016-08-102-1/+3
| | | | | | | | required by glClientWaitSync (GL 4.5 Core spec) that can optionally flush the context Reviewed-by: Rob Clark <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* gallium: add render_condition_enable param to clear_render_target/depth_stencilMarek Olšák2016-08-101-2/+4
| | | | | Reviewed-by: Roland Scheidegger <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* gallium: split transfer_inline_write into buffer and texture callbacksMarek Olšák2016-07-231-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | to reduce the call indirections with u_resource_vtbl. The worst call tree you could get was: - u_transfer_inline_write_vtbl - u_default_transfer_inline_write - u_transfer_map_vtbl - driver_transfer_map - u_transfer_unmap_vtbl - driver_transfer_unmap That's 6 indirect calls. Some drivers only had 5. The goal is to have 1 indirect call for drivers that care. The resource type can be determined statically at most call sites. The new interface is: pipe_context::buffer_subdata(ctx, resource, usage, offset, size, data) pipe_context::texture_subdata(ctx, resource, level, usage, box, data, stride, layer_stride) v2: fix whitespace, correct ilo's behavior Reviewed-by: Nicolai Hähnle <[email protected]> Acked-by: Roland Scheidegger <[email protected]>
* gallium: add a cap for VIEWPORT_SUBPIXEL_BITS (v2)Józef Kucia2016-07-201-0/+1
| | | | | | | | | | | | This allows Gallium drivers to advertise the subpixel precision for floating point viewports bounds. v2: - Set ViewportSubpixelBits in st_init_limits. Signed-off-by: Józef Kucia <[email protected]> Signed-off-by: Marek Olšák <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* gallium: Add a cap for offset_units_unscaledAxel Davy2016-06-251-0/+1
| | | | | | | | | | | | | | D3D9 has a different behaviour for depth bias. For OGL/D3D1X, the depth bias unit is the minimal resolvable value for the depth buffer, which depends on the format (and has different behaviour for float depth buffers). For D3D9, the depth bias unit is 1.0f. Signed-off-by: Axel Davy <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* Remove wrongly repeated words in commentsGiuseppe Bilotta2016-06-231-1/+1
| | | | | | | | | | | | | | | | | Clean up misrepetitions ('if if', 'the the' etc) found throughout the comments. This has been done manually, after grepping case-insensitively for duplicate if, is, the, then, do, for, an, plus a few other typos corrected in fly-by v2: * proper commit message and non-joke title; * replace two 'as is' followed by 'is' to 'as-is'. v3: * 'a integer' => 'an integer' and similar (originally spotted by Jason Ekstrand, I fixed a few other similar ones while at it) Signed-off-by: Giuseppe Bilotta <[email protected]> Reviewed-by: Chad Versace <[email protected]>
* gallium: make constant_buffer constRob Clark2016-06-201-1/+1
| | | | | Signed-off-by: Rob Clark <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* gallium: add PIPE_CAP_MAX_WINDOW_RECTANGLES to all driversIlia Mirkin2016-06-181-0/+1
| | | | | | | | This says how many window rectangles are supported by the implementation, although it may not exceed PIPE_MAX_WINDOW_RECTANGLES. Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* llvmpipe: hack-fix bugs due to bogus bind flagsRoland Scheidegger2016-06-146-17/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | The gallium contract would be that bind flags must indicate all possible bindings a resource might get used, but fact is the mesa state tracker does not set bind flags correctly, and this is more or less unfixable due to GL. This caused a bug with piglit arb_uniform_buffer_object-rendering-dsa since 6e6fd911da8a1d9cd62fe0a8a4cc0fb7bdccfe02 - the commit is correct, but it caused us to miss updates to fs UBOs completely, since the corresponding buffer didn't have the appropriate bind flag set (thus we wouldn't check if it is indeed currently bound). See the discussion about this starting here: https://lists.freedesktop.org/archives/mesa-dev/2016-June/119829.html So, update the bind flags when we detect such usage. Note we update this value for now only in places which matter for us - that is creating sampler/surface view, or binding constant buffer. There's plenty more places (setting streamout buffers, vertex/index buffers, ...) where things can be set with the wrong bind flags, but the bind flags there never matter. While here also make sure we only set dirty constant bit when it's a fs constant buffer - totally doesn't matter if it's vs/gs. Reviewed-by: Jose Fonseca <[email protected]>
* llvmpipe: turn on pipe cap for GL_ARB_copy_image supportBrian Paul2016-06-101-1/+2
| | | | Reviewed-by: Charmaine Lee <[email protected]>
* llvmpipe: don't use 3-component formats, except 32-bit x 3 formatsBrian Paul2016-06-101-11/+12
| | | | | | | | | | | This basically disallows all 8-bit x 3 and 16-bit x 3 formats for textures and render targets. Some 3-component formats were already disallowed before. This avoids problems with GL_ARB_copy_image. v2: the previous version of this patch disallowed all 3-component formats Reviewed-by: Charmaine Lee <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* gallivm: Use llvm.fmuladd.*.Jose Fonseca2016-06-101-11/+6
| | | | Reviewed-by: Roland Scheidegger <[email protected]>
* gallium: add PIPE_CAP_TGSI_VOTE for when the VOTE ops are allowedIlia Mirkin2016-06-061-0/+1
| | | | | Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* gallium: Add a pipe cap for whether primitive restart works for patches.Kenneth Graunke2016-05-231-0/+1
| | | | | | | | | | | | | | | Some hardware supports primitive restart on patch primitives, and other hardware does not. Modern GL and ES include a query for this feature; adding a capability bit will allow us to answer it. As far as I know, AMD hardware does not support this feature, while NVIDIA and Intel hardware does. However, most Gallium drivers do not appear to support tessellation shaders yet. So, I've enabled it for nvc0 and disabled it everywhere else. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* llvm/softpipe: Enable cull_distance as draw supports it.Tobias Klausmann2016-05-231-1/+2
| | | | | | Reviewed-by: Roland Scheidegger <[email protected]> Signed-off-by: Tobias Klausmann <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* gallium: Add a pipe cap for arb_cull_distanceTobias Klausmann2016-05-141-0/+1
| | | | | | | | | This lets us safely enable or disable the extension as needed Signed-off-by: Tobias Klausmann <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Marek Olšák <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* llvmpipe: s/Elements/ARRAY_SIZE/Brian Paul2016-04-2712-54/+54
| | | | Reviewed-by: Jose Fonseca <[email protected]>
* gallium: Include intrin.h instead of defining ourselves.Jose Fonseca2016-04-261-2/+1
| | | | | | | | | More portable, particularly when building with Clang, which implements all MSVC intrisincs in its own intrin.h, but doesn't actually support `#pragma instrinsic`. Reviewed-by: Roland Scheidegger <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* gallium: fix warnings in release buildGrazvydas Ignotas2016-04-251-1/+1
| | | | | | | | Mark variables MAYBE_UNUSED to avoid unused-but-set-variable warnings in release build. Signed-off-by: Grazvydas Ignotas <[email protected]> Reviewed-by: Emil Velikov <[email protected]>
* gallium: add bool return to pipe_context::end_queryNicolai Hähnle2016-04-211-1/+3
| | | | | | | | | Even when begin_query succeeds, there can still be failures in query handling. For example for radeon, additional buffers may have to be allocated when queries span multiple command buffers. Reviewed-by: Samuel Pitoiset <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* gallium: merge PIPE_SWIZZLE_* and UTIL_FORMAT_SWIZZLE_*Marek Olšák2016-04-225-12/+12
| | | | | | | | Use PIPE_SWIZZLE_* everywhere. Use X/Y/Z/W/0/1 instead of RED, GREEN, BLUE, ALPHA, ZERO, ONE. The new enum is called pipe_swizzle. Acked-by: Jose Fonseca <[email protected]>
* llvmpipe: Call LLVMShutdown before exiting.Jose Fonseca2016-04-191-0/+2
| | | | | | So that LLVM frees its globals. Trivial.
* llvmpipe: Avoid LLVMGetGlobalContext in tests.Jose Fonseca2016-04-195-6/+24
| | | | Trivial.
* llvmpipe: Skip false exp2 failure in lp_test_arit due to buggy MSVCRT.Jose Fonseca2016-04-191-4/+34
| | | | | | | | | 64bits MSVCRT's exp2f(-inf) returns -inf instead of 0. Tested with MSVC 2013's CRT. (I haven't tried 2015 yet.) Also this does not happen with MinGW. Reviewed-by: Roland Scheidegger <[email protected]>
* llvmpipe: Test more vector lengths.Jose Fonseca2016-04-191-13/+30
| | | | | | | | | | All power of two of up native vector length. There is actually a bug in lp_build_round for v2, whereby it doesn't round to nearest. Fixing is left to the future, but the test is now able to expect it to fail. Reviewed-by: Roland Scheidegger <[email protected]>
* gallivm: convert size query to using a set of parameters.Dave Airlie2016-04-191-18/+4
| | | | | | | | | | This isn't currently that easy to expand, so fix it up before expanding it later to include dynamic samplers. [airlied: use some local variables (Roland)] Reviewed-by: Roland Scheidegger <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* llvmpipe: (trivial) initialize src1_alpha var to NULLRoland Scheidegger2016-04-151-1/+1
| | | | | | | | | | | The blend code would do a conditional assignment based on it, causing valgrind to complain. Since that variable was actually unused in this case, this doesn't fix anything but the warning. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94955 Cc: "11.1 11.2" <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* scons: Add a "check" target to run all unit tests.Jose Fonseca2016-04-141-5/+1
| | | | | | | | | Except: - u_cache_test -- too long - translate_test -- unreliable (it's probably testing corner cases that translate module doesn't care about.) Reviewed-by: Roland Scheidegger <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-12 14:01:43 +0000