summaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* clover: Handle NULL devices returned by pipe_loader_probe() v2Tom Stellard2015-12-011-1/+2
| | | | | | | | | | | | | | | | | | | | | When probing for devices, clover will call pipe_loader_probe() twice. The first time to retrieve the number of devices, and then second time to retrieve the device structures. We currently assume that the return value of both calls will be the same, but this will not be the case if a device happens to disappear between the two calls. When a device disappears, the pipe_loader_probe() will add a NULL device to the device list, so we need to handle this. v2: - Keep range for loop Reviewed-by: Francisco Jerez <[email protected]> Acked-by: Emil Velikov <[email protected]> CC: <[email protected]>
* automake: fix some occurrences of hardcoded -ldl and -lpthreadJonathan Gray2015-12-011-1/+1
| | | | | | | | | Correct some occurrences of -ldl and -lpthread to use $(DLOPEN_LIBS) and $(PTHREAD_LIBS) respectively. Signed-off-by: Jonathan Gray <[email protected]> Cc: "11.0 11.1" <[email protected]> Reviewed-by: Emil Velikov <[email protected]>
* st/va: also retrieve reference frames info for h264Julien Isorce2015-12-011-0/+52
| | | | | | | | Other hardwares than AMD require to parse: VAPictureParameterBufferH264.ReferenceFrames[16] Signed-off-by: Julien Isorce <[email protected]> Reviewed-by: Christian König <[email protected]>
* st/va: delay decoder creation until max_references is knownJulien Isorce2015-12-014-34/+68
| | | | | | | | | | | | | | | | | | In general max_references cannot be based on num_render_targets. This patch allows to allocate buffers with an accurate size. I.e. no more than necessary. For other codecs it is a fixed value 2. This is similar behaviour as vaapi/vdpau-driver. For now HEVC case defaults to num_render_targets as before. But it could also benefits this change by setting a more accurate max_references number in handlePictureParameterBuffer. Signed-off-by: Julien Isorce <[email protected]> Reviewed-by: Christian König <[email protected]> Reviewed-by: Emil Velikov <[email protected]>
* r600: workaround empty geom shader.Dave Airlie2015-12-011-0/+5
| | | | | | | | | | We need to emit at least one cut/emit in every geometry shader, the easiest workaround it to stick a single CUT at the top of each geom shader. Reviewed-by: Marek Olšák <[email protected]> Cc: "10.6 11.0 11.1" <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* r600: rv670 use at least 16es/gs threadsDave Airlie2015-12-011-4/+5
| | | | | | | | This is specified in the docs for rv670 to work properly. Reviewed-by: Marek Olšák <[email protected]> Cc: "10.6 11.0 11.1" <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* r600: geometry shader gsvs itemsize workaroundDave Airlie2015-12-011-0/+20
| | | | | | | | | | On some chips the GSVS itemsize needs to be aligned to a cacheline size. This only applies to some of the r600 family chips. Reviewed-by: Marek Olšák <[email protected]> Cc: "10.6 11.0 11.1" <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* r600/shader: split address get out to a function.Dave Airlie2015-12-011-1/+6
| | | | | | | This will be used in the tess shaders. Reviewed-by: Oded Gabbay <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* vl/buffers: fixes vl_video_buffer_formats for RGBXJulien Isorce2015-11-301-1/+1
| | | | | | | Fixes: 42a5e143a8d "vl/buffers: add RGBX and BGRX to the supported formats" Cc: [email protected] Signed-off-by: Julien Isorce <[email protected]> Reviewed-by: Emil Velikov <[email protected]>
* r600: move per-type settings into a switch statementDave Airlie2015-11-301-5/+13
| | | | | | | This will allow adding tess stuff much cleaner later. Reviewed-by: Glenn Kennard <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* r600: split out common alu_writes pattern.Dave Airlie2015-11-301-7/+12
| | | | | | | | This just splits out a common pattern into an inline function to make things cleaner to read. Reviewed-by: Glenn Kennard <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* r600/llvm: fix r600/llvm buildDave Airlie2015-11-301-1/+1
| | | | | | Reported on irc by gryffus Signed-off-by: Dave Airlie <[email protected]>
* r600: fixes for register definitions.Dave Airlie2015-11-301-3/+3
| | | | | | Forgot to add these. Signed-off-by: Dave Airlie <[email protected]>
* r600: add missing register to initial stateDave Airlie2015-11-303-7/+15
| | | | | | | | | | We really should initialise HS/LS_2 and SQ_LDS_ALLOC exists on all evergreen not just cayman, so we should initialise it as well. Reviewed-by: Glenn Kennard <[email protected]> Reviewed-by: Marek Olšák <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* r600: define registers required for tessellationDave Airlie2015-11-302-27/+113
| | | | | | | | | This adds the defines for a bunch of registers and shader values that are required to implement tessellation. Reviewed-by: Marek Olšák <[email protected]> Reviewed-by: Glenn Kennard <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* r600: consolidate clip state updatesDave Airlie2015-11-302-17/+16
| | | | | | | | Move some common code into one place, tess will also need to use this function. Reviewed-by: Marek Olšák <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* nv50/ir: always display the opcode number for unknown instructionsSamuel Pitoiset2015-11-292-2/+2
| | | | | | | This helps in debugging unknown instructions. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* auxiliary/vl/dri: fd management cleanupsEmil Velikov2015-11-291-2/+5
| | | | | | | | | | Analogous to previous commit, minus the extra dup. We are the one opening the device thus we can directly use the fd. Spotted by Coverity (CID 1339867, 1339877) Cc: [email protected] Signed-off-by: Emil Velikov <[email protected]>
* auxiliary/vl/drm: fd management cleanupsEmil Velikov2015-11-291-1/+7
| | | | | | | | | Analogous to previous commit. Spotted by Coverity (CID 1339868) Cc: [email protected] Signed-off-by: Emil Velikov <[email protected]>
* st/xa: fd management cleanupsEmil Velikov2015-11-291-1/+8
| | | | | | | | | Analogous to previous commit. Spotted by Coverity (CID 1339866) Cc: [email protected] Signed-off-by: Emil Velikov <[email protected]>
* st/dri: fd management cleanupsEmil Velikov2015-11-291-2/+14
| | | | | | | | | | | Add some checks if the original/dup'd fd is valid and ensure that we don't leak it on error. The former is implicitly handled within the pipe_loader, although let's make things explicit and check beforehand. Spotted by Coverity (CID 1339865) Cc: [email protected] Signed-off-by: Emil Velikov <[email protected]>
* pipe-loader: check if winsys.name is non-null prior to strcmpEmil Velikov2015-11-291-4/+4
| | | | | | | | | | | | In theory this wouldn't be an issue, as we'll find the correct name and break out of the loop before we hit the sentinel. Let's fix this and avoid issues in the future. Spotted by Coverity (CID 1339869, 1339870, 1339871) Cc: [email protected] Signed-off-by: Emil Velikov <[email protected]>
* st/va: add missing break statementEmil Velikov2015-11-291-0/+1
| | | | | | | | | | | | | | Earlier commit factored out the mpeg4 IQ matrix handling into separate function, although it forgot to add a break in its case statement. Thus the data ended up partially overwritten as the mpeg4 and h265 structs are members of the desc union. Spotted by Coverity (CID 1341052) Fixes: 64761a841db "st/va: move MPEG4 functions into separate file" Cc: Julien Isorce <[email protected]> Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Christian König <[email protected]>
* radeon: only suspend queries on flush if they haven't been suspended yetNicolai Hähnle2015-11-282-2/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | Non-timer queries are suspended during blits. When the blits end, the queries are resumed, but this resume operation itself might run out of CS space and trigger a flush. When this happens, we must prevent a duplicate suspend during preflush suspend, and we must also prevent a duplicate resume when the CS flush returns back to the original resume operation. This fixes a regression that was introduced by: commit 8a125afa6e88a3eeddba8c7fdc1a75c9b99d5489 Author: Nicolai Hähnle <[email protected]> Date: Wed Nov 18 18:40:22 2015 +0100 radeon: ensure that timing/profiling queries are suspended on flush The queries_suspended_for_flush flag is redundant because suspended queries are not removed from their respective linked list. Reviewed-by: Marek Olšák <[email protected]> Reported-by: Axel Davy <[email protected]> Cc: "11.1" <[email protected]> Tested-by: Axel Davy <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* scons: Use LD version script for libgl-xlib.Jose Fonseca2015-11-271-2/+6
| | | | Trivial.
* svga: Don't return value from void function.Jose Fonseca2015-11-271-1/+2
| | | | | | | Addresses MSVC warning C4098: 'svga_destroy_query' : 'void' function returning a value. Reviewed-by: Roland Scheidegger <[email protected]>
* gallium: Make pipe_query_result::batch array length non-zero.Jose Fonseca2015-11-271-2/+2
| | | | | | | | | | | | Zero length arrays are non standard: warning C4200: nonstandard extension used : zero-sized array in struct/union Cannot generate copy-ctor or copy-assignment operator when UDT contains a zero-sized array And all code does `N * sizeof query_result->batch[0]`, so it should work exactly the same. Reviewed-by: Roland Scheidegger <[email protected]>
* freedreno/ir3: assign varying locations laterRob Clark2015-11-264-29/+37
| | | | | | | | | | | | | Rather than assigning inloc up front, when we don't yet know if it will be unused, assign it last thing before the legalize pass. Also, realize when inputs are unused (since for frag shader's we can't rely on them being removed from ir->inputs[]). This doesn't make sense if we don't also dynamically assign the inloc's, since we could end up telling the hw the wrong # of varyings (since we currently assume that the # of varyings and max-inloc are related..) Signed-off-by: Rob Clark <[email protected]>
* freedreno/ir3: use instr flag to mark unused instructionsRob Clark2015-11-264-14/+24
| | | | | | Rather than magic depth value, which won't be available in later stages. Signed-off-by: Rob Clark <[email protected]>
* freedreno/a4xx: rework vinterp/vpsreplRob Clark2015-11-261-12/+36
| | | | | | Same as previous commit, for a4xx. Signed-off-by: Rob Clark <[email protected]>
* freedreno/a3xx: rework vinterp/vpsreplRob Clark2015-11-261-12/+37
| | | | | | | | | | | | Make the interpolation / point-sprite replacement mode setup deal with varying packing. In a later commit, we switch to packing just the varying components that are actually used by the frag shader, so we won't be able to assume everything is vec4's aligned to vec4. Which would highly confuse the previous vinterp/vpsrepl logic. Signed-off-by: Rob Clark <[email protected]>
* clover: fix tgsi compiler crash with invalid srcSerge Martin2015-11-261-0/+5
| | | | Reviewed-by: Francisco Jerez <[email protected]>
* radeon: use PIPE_DRIVER_QUERY_FLAG_DONT_LIST for perfcountersNicolai Hähnle2015-11-261-0/+2
| | | | | | | | Since the query names are not very enlightening, and there are thousands of them, GALLIUM_HUD=help should only show the first and last query name for each hardware block. Reviewed-by: Marek Olšák <[email protected]>
* gallium: add PIPE_DRIVER_QUERY_FLAG_DONT_LISTNicolai Hähnle2015-11-262-1/+12
| | | | | | | This allows the driver to give a hint to the HUD so that GALLIUM_HUD=help is less spammy. Reviewed-by: Marek Olšák <[email protected]>
* radeon: delay the generation of driver query names until first useNicolai Hähnle2015-11-263-104/+113
| | | | | | | This shaves a bit more time off the startup of programs that don't actually use performance counters. Reviewed-by: Marek Olšák <[email protected]>
* st/va: add missing profiles in PipeToProfile's switch.Julien Isorce2015-11-261-0/+7
| | | | | | | Otherwise assert is raised from vlVaQueryConfigProfiles's for loop. Signed-off-by: Julien Isorce <[email protected]> Reviewed-by: Christian König <[email protected]>
* targets/xvmc: use the non-inline sw helpersMichel Dänzer2015-11-261-0/+1
| | | | | | | | | | | | | | | This was missed in commit 59cfb21d ("targets: use the non-inline sw helpers"). Fixes build failure: CXXLD libXvMCgallium.la ../../../../src/gallium/auxiliary/pipe-loader/.libs/libpipe_loader_static.a(libpipe_loader_static_la-pipe_loader_sw.o):(.data.rel.ro+0x0): undefined reference to `sw_screen_create' collect2: error: ld returned 1 exit status Makefile:756: recipe for target 'libXvMCgallium.la' failed make[3]: *** [libXvMCgallium.la] Error 1 Trivial.
* targets/nine: remove freedreno targetEmil Velikov2015-11-251-3/+0
| | | | | | | | Analogous to previous commit. As we no longer have anyone who uses NIR we can drop the link. Signed-off-by: Emil Velikov <[email protected]> Acked-by: Rob Clark <[email protected]>
* targets/nine: remove vc4 targetEmil Velikov2015-11-251-2/+0
| | | | | | | There are no users for it. Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Eric Anholt <[email protected]>
* gallium: remove unused function declarationsEmil Velikov2015-11-252-6/+0
| | | | | | | | Unused as of commit 23fb11455b4 "{st,targets}/dri: use static/dynamic pipe-loader" Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* targets: use the non-inline sw helpersEmil Velikov2015-11-257-1/+8
| | | | | | | | | | | | | | | | | | Previously (with the inline ones) things were embedded into the pipe-loader, which means that we cannot control/select what we want in each target. That also meant that at runtime we ended up with the empty sw_screen_create() as the GALLIUM_SOFTPIPE/LLVMPIPE were not set. v2: Cover all the targets, not just dri. Cc: "11.1" <[email protected]> Cc: Ilia Mirkin <[email protected]> Cc: Edward O'Callaghan <[email protected]> Signed-off-by: Emil Velikov <[email protected]> Tested-by: Oded Gabbay <[email protected]> Tested-by: Nick Sarnie <[email protected]>
* target-hepers: add non inline sw helpersEmil Velikov2015-11-252-0/+83
| | | | | | | | | | | | Feeling rather dirty copying the inline ones, yet we need the inline ones for swrast only targets like libgl-xlib, osmesa. Cc: "11.1" <[email protected]> Cc: Ilia Mirkin <[email protected]> Cc: Edward O'Callaghan <[email protected]> Signed-off-by: Emil Velikov <[email protected]> Tested-by: Oded Gabbay <[email protected]> Tested-by: Nick Sarnie <[email protected]>
* pipe-loader: fix off-by one errorEmil Velikov2015-11-251-1/+1
| | | | | | | | | | | | | | | | With earlier commit we've dropped the manual iteration over the fixed size array and prepemtively set the variable storing the size, that is to be returned. Yet we forgot to adjust the comparison, as before we were comparing the index, now we're comparing the size. Fixes: ff9cd8a67ca "pipe-loader: directly use pipe_loader_sw_probe_null() at probe time" Cc: [email protected] Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93091 Reported-by: Tom Stellard <[email protected]> Signed-off-by: Emil Velikov <[email protected]> Tested-by: Tom Stellard <[email protected]> Tested-by: Dieter Nützel <[email protected]>
* radeonsi/compute: Use the compiler's COMPUTE_PGM_RSRC* register valuesTom Stellard2015-11-252-31/+7
| | | | | | | | | The compiler has more information and is able to optimize the bits it sets in these registers. Reviewed-by: Marek Olšák <[email protected]> CC: <[email protected]>
* radeonsi: Rename si_shader::ls_rsrc{1,2} to si_shader::rsrc{1,2}Tom Stellard2015-11-253-6/+6
| | | | | | In the future, these will be used by other shaders types. Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: implement AMD_performance_monitor for CIK+Nicolai Hähnle2015-11-2510-3/+1486
| | | | | | | | | | | | | | | | | | Expose most of the performance counter groups that are exposed by Catalyst. Ideally, the driver will work with GPUPerfStudio at some point, but we are not quite there yet. In any case, this is the reason for grouping multiple instances of hardware blocks in the way it is implemented. The counters can also be shown using the Gallium HUD. If one is interested to see how work is distributed across multiple shader engines, one can set the environment variable RADEON_PC_SEPARATE_SE=1 to obtain finer-grained performance counter groups. Part of the implementation is in radeon because an implementation for older hardware would largely follow along the same lines, but exposing a different set of blocks which are programmed slightly differently. Reviewed-by: Marek Olšák <[email protected]>
* radeon: scale query buffer size to result sizeNicolai Hähnle2015-11-251-1/+1
| | | | | | | Performance monitor queries can become very big, especially considering that instances of a block in different shader engines are queried separately. Reviewed-by: Marek Olšák <[email protected]>
* radeonsi/sid: add performance counter registersNicolai Hähnle2015-11-251-0/+1013
| | | | Reviewed-by: Marek Olšák <[email protected]>
* radeonsi/sid: add hardware constants for COPY_DATA packetNicolai Hähnle2015-11-251-0/+3
| | | | Reviewed-by: Marek Olšák <[email protected]>
* radeon: extend CIK_UCONFIG_REG_END for performance countersNicolai Hähnle2015-11-252-2/+2
| | | | Reviewed-by: Marek Olšák <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-28 02:23:39 +0000