aboutsummaryrefslogtreecommitdiffstats
path: root/src/amd/vulkan/radv_query.c
Commit message (Collapse)AuthorAgeFilesLines
* radv: only emit cache flushes when the pool size is large enoughSamuel Pitoiset2018-03-011-5/+9
| | | | | | | | This is an optimization which reduces the number of flushes for small pool buffers. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: keep track of the query pool sizeSamuel Pitoiset2018-03-011-5/+4
| | | | | Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: make sure to emit cache flushes before starting a querySamuel Pitoiset2018-03-011-7/+21
| | | | | | | | | | | If the query pool has been previously resetted using the compute shader path. Fixes: a41e2e9cf5 ("radv: allow to use a compute shader for resetting the query pool") Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105292 Cc: "18.0" <[email protected]> Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: fix pipeline statistics end query on compute queueDave Airlie2017-12-281-1/+1
| | | | | | | | | | | | It's legal to a pipeline stat query on a compute queue, but we'd emit the wrong packet here. This should fix it to emit the correct packet. Noticed while inspecting the mpv hang. Fixes: ad61eac250 (radv: factor out eop event writing code. (v2)) Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: do not add the query pool BO to the list in vkCmdEndQuery()Samuel Pitoiset2017-11-201-1/+3
| | | | | | | | | As per the spec, the query identified by queryPool and query must currently be active. Applications have to call vkCmdBeginQuery() before, and thus the query pool BO will already be in the list. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: use vk_error() everywhere an error is returnedSamuel Pitoiset2017-11-131-3/+3
| | | | | | | For consistency and it might help for debugging purposes. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: wrap cs_add_buffer in an inline. (v2)Dave Airlie2017-11-061-5/+5
| | | | | | | | | The next patch will try and avoid calling the indirect function. v2: add a missing conversion. Reviewed-by: Samuel Pitoiset <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: allow to use a compute shader for resetting the query poolSamuel Pitoiset2017-10-271-7/+9
| | | | | | | | | | | | | Serious Sam Fusion 2017 uses a huge number of occlusion queries, and the allocated query pool buffer is greater than 4096 bytes. This slightly improves performance (tested in Ultra) from 117.2 FPS to 119.7 FPS (~+2%) on my RX480. This also improves Talos, from 69 FPS to 72/73 FPS (~+5%). Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: add support for local bos. (v3)Dave Airlie2017-10-261-1/+1
| | | | | | | | | | | | This uses the new kernel interfaces for reduced cs overhead, We only set the local flag for memory allocations that don't have a dedicated allocation and ones that aren't imports. v2: add to all the internal buffer creation paths. v3: missed some command submission paths, handle 0/empty bo lists. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* amd: move r600d_common.h into r600gMarek Olšák2017-10-091-1/+1
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radv: convert all COMPUTE operations to the RADV_META_SAVE_XXX flagsSamuel Pitoiset2017-10-061-3/+6
| | | | | Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: do not need to double zero-init the meta state structuresSamuel Pitoiset2017-10-021-2/+0
| | | | | Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: store the amount of saved constants in the compute stateSamuel Pitoiset2017-09-271-1/+1
| | | | | | | It's safer and more elegant. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: fix saved compute state when doing statistics/occlusion queriesSamuel Pitoiset2017-09-261-2/+2
| | | | | | | | | We are pushing 16-bytes of constants, so we have to save/restore the same amount of data to avoid data corruption. Cc: 17.2 <[email protected]> Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Don't use a virtual function for getting the buffer virtual address.Bas Nieuwenhuizen2017-09-201-6/+6
| | | | | | | | | | | We are really not going to use a winsys which does not need to store the va, so might as well store it in a standard field. Not sure this helps perf much though, as most of the cost is in the cache miss accessing the bo anyway, which we stil need to do. Reviewed-by: Dave Airlie <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]>
* radv: add support for cmd predication.Dave Airlie2017-07-061-2/+5
| | | | | | | | This doesn't get used yet, it just adds support to various PKT3 emissions to enable it later. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: Remove SI num RB override for occlusion queries.Bas Nieuwenhuizen2017-06-061-3/+0
| | | | | | | | radeonsi doesn't have it anymore either. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Fixes: f4e499ec791 "radv: add initial non-conformant radv vulkan driver" Reviewed-by: Dave Airlie <[email protected]>
* radv: fewer than 8 RBs are possibleNicolai Hähnle2017-06-051-2/+0
| | | | | | This fixes the subsequent assertion on Bonaire. Reviewed-by: Dave Airlie <[email protected]>
* radv: factor out eop event writing code. (v2)Dave Airlie2017-06-021-40/+17
| | | | | | | | | | | | | | In prep for GFX9 refactor some of the eop event writing code out. This changes behaviour, but aligns with what radeonsi does, it does double emits on CIK/VI, whereas previously it only did this on CIK. v2: bump the size checks. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: factor out si_emit_wait_fence code.Dave Airlie2017-06-021-14/+2
| | | | | | | This code was in a few places, consolidate into one. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: always free nir shaders from modules on stackGrazvydas Ignotas2017-05-101-4/+2
| | | | | | | | | valgrind reports them as leaked, and I could not find anything making a copy of the nir pointer. Also, radv_device_init_meta_blit_color() is already freeing them unconditionally like this. Signed-off-by: Grazvydas Ignotas <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* nir: Embed the shader_info in the nir_shader againJason Ekstrand2017-05-091-14/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit e1af20f18a86f52a9640faf2d4ff8a71b0a4fa9b changed the shader_info from being embedded into being just a pointer. The idea was that sharing the shader_info between NIR and GLSL would be easier if it were a pointer pointing to the same shader_info struct. This, however, has caused a few problems: 1) There are many things which generate NIR without GLSL. This means we have to support both NIR shaders which come from GLSL and ones that don't and need to have an info elsewhere. 2) The solution to (1) raises all sorts of ownership issues which have to be resolved with ralloc_parent checks. 3) Ever since 00620782c92100d77c660f9783504c6d80fa1d58, we've been using nir_gather_info to fill out the final shader_info. Thanks to cloning and the above ownership issues, the nir_shader::info may not point back to the gl_shader anymore and so we have to do a copy of the shader_info from NIR back to GLSL anyway. All of these issues go away if we just embed the shader_info in the nir_shader. There's a little downside of having to copy it back after calling nir_gather_info but, as explained above, we have to do that anyway. Acked-by: Timothy Arceri <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* radv: set base/ranges for push constant loads.Dave Airlie2017-05-081-0/+2
| | | | | | | | | This isn't necessary yet but I'd like to use the range in some future patches. [airlied: add new resolve pass] Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: Use correct stage for ready bit.Bas Nieuwenhuizen2017-05-021-9/+24
| | | | | | | | Set the bit in the same stage as the timestamp, instead always at top of pipe. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]> Tested-by: Grazvydas Ignotas <[email protected]>
* radv: Add top of pipe timestamp queries.Bas Nieuwenhuizen2017-05-021-12/+26
| | | | | | | Does not fix brokenness with the ready bit. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: use push descriptors in metaFredrik Höglund2017-04-141-42/+32
| | | | | | | Use push descriptors instead of temp descriptor sets. Signed-off-by: Fredrik Höglund <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: Stop shadowing the result in radv_GetQueryPoolResults.Bas Nieuwenhuizen2017-04-121-4/+4
| | | | | | | The outer result was referred to, which meant bugs. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Return VK_NOT_READY if the query results are not available.Bas Nieuwenhuizen2017-04-121-0/+6
| | | | | | Signed-off-by: Bas Nieuwenhuizen <[email protected]> Fixes: 8475a14302e ("radv: Implement pipeline statistics queries.") Reviewed-by: Fredrik Höglund <[email protected]>
* radv: Set query availability bit even if we don't wait.Bas Nieuwenhuizen2017-04-121-3/+4
| | | | | | Signed-off-by: Bas Nieuwenhuizen <[email protected]> Fixes: 8475a14302e ("radv: Implement pipeline statistics queries.") Reviewed-by: Fredrik Höglund <[email protected]>
* radv: Implement pipeline statistics queries.Bas Nieuwenhuizen2017-04-111-26/+391
| | | | | | | | | | | The devil is in the shader again, otherwise this is fairly straightforward. The CTS contains no pipeline statistics copy to buffer testcases, so I did a basic smoketest. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Let count be dynamic in radv_break_on_count.Bas Nieuwenhuizen2017-04-111-3/+3
| | | | | Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Rename query pipeline/set layout.Bas Nieuwenhuizen2017-04-111-11/+11
| | | | | | | For using them with both occlusion and pipeline statistics queries. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Use VK_WHOLE_SIZE for the query buffer bindings.Bas Nieuwenhuizen2017-04-111-2/+2
| | | | | | | | The buffer sizes are specified just a few lines earlier, so don't repeat ourselves. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Use a shader for occlusion CmdCopyQueryPoolResults.Bas Nieuwenhuizen2017-04-111-74/+64
| | | | | | | | | | | | | | | | | | Use the new occlusion query copy shader. We don't use the shader for the waiting as a polling loop ineracts badly with having caching enabled. I noticed on my GPU (Tonga) that the values are written out in order, so I just use a WAIT_REG_MEM on the last value. If it turns out other chips don't do that we may need to look a bit more into this. Having 8 WAIT_REG_MEM packets per query doesn't sound ideal. This also restricts the availability word in the pool to timestamp queries only, as occlusion queries don't use it, and pipeline statistic queries likely won't either. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Add occlusion query shader.Bas Nieuwenhuizen2017-04-111-0/+419
| | | | | | | | | Adds a shader for writing occlusion query results to a buffer, as the CP packet isn't support on SI or secondary buffers, and doesn't handle the availability bit (or partial results) nor truncation to 32-bit. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Use correct size for availability flag.Bas Nieuwenhuizen2017-02-271-2/+5
| | | | | | | | | | Per spec, VK_QUERY_RESULT_64_BIT specifies the integer size and the availability flag is an integer. We apparently handled this correctly already for the copy to buffer case. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]> Cc: 13.0 17.0 <[email protected]>
* radv: Only use PKT3_OCCLUSION_QUERY when it doesn't hang.Bas Nieuwenhuizen2017-02-271-20/+45
| | | | | | | | | | | | | PKT3_OCCLUSION_QUERY hangs when used in a nested IB. This only calls it when in a primary command buffer and we change GetQueryPoolResults to not need it. CmdCopyQueryPoolResults still needs it so we break that behavior for secondary command buffers. However, that would hang already and using an unitialized value is better than a hang. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]> Cc: 13.0 17.0 <[email protected]>
* Revert "radv: detect command buffers that do no work and drop them (v2)"Dave Airlie2017-02-201-4/+0
| | | | | | | | | | | | | | | | This just keeps popping up minor problems and regressions we should revisit in a more sustainable manner later. This also reverts: Revert "radv: query cmds should mark a cmd buffer as having draws." Revert "radv: also fixup event emission to not get culled." This reverts commit d1640e79328af4a63c056e3ccab299a3bffbede7. This reverts commit 8b47b97215af7157bc15676167cab73aa5a61a76. This reverts commit b4b19afebeac84ddce3f1f8b7ffe0da6474fe29a. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: query cmds should mark a cmd buffer as having draws.Dave Airlie2017-02-151-0/+4
| | | | | | | | This fixes a regression with the remove non-draw cmd buffers in queries. Fixes: 8b47b97215a radv: detect command buffers that do no work and drop them (v2) Signed-off-by: Dave Airlie <[email protected]>
* radv: Get physical device from radv_device instead of the instance.Bas Nieuwenhuizen2017-01-161-3/+3
| | | | | Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Use RELEASE_MEM packet for MEC timestamp query.Bas Nieuwenhuizen2016-12-181-9/+20
| | | | | Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: fix release build unused variable warningsGrazvydas Ignotas2016-12-101-3/+3
| | | | | | | Just mark with MAYBE_UNUSED. Signed-off-by: Grazvydas Ignotas <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: move to using shared vk_alloc inlines.Dave Airlie2016-10-191-4/+4
| | | | | | | | This moves to the shared vk_alloc inlines for vulkan memory allocations. Acked-by: Jason Ekstrand <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: add initial non-conformant radv vulkan driverDave Airlie2016-10-071-0/+415
This squashes all the radv development up until now into one for merging. History can be found: https://github.com/airlied/mesa/tree/semi-interesting This requires llvm 3.9 and is in no way considered a conformant vulkan implementation. It can run a number of vulkan applications, and supports all GPUs using the amdgpu kernel driver. Thanks to Intel for providing anv and spirv->nir, and Emil Velikov for reviewing build integration. Parts of this are: Reviewed-by: Nicolai Hähnle <[email protected]> Acked-by: Edward O'Callaghan <[email protected]> Authors: Bas Nieuwenhuizen and Dave Airlie Signed-off-by: Dave Airlie <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-03 08:04:39 +0000