Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | radeonsi: implement the workaround for Rocket League - postponed TGSI kill | Marek Olšák | 2017-06-23 | 2 | -0/+6 |
| | | | | | | | | Do KILL at the end of shaders so as not to break WQM. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100070 Reviewed-by: Nicolai Hähnle <[email protected]> | ||||
* | gallium/radeon: pass create_screen flags to r600_common_screen_init | Marek Olšák | 2017-06-23 | 2 | -2/+2 |
| | | | | Reviewed-by: Nicolai Hähnle <[email protected]> | ||||
* | radeonsi/gfx9: enable DCC fast clear | Marek Olšák | 2017-06-22 | 1 | -4/+0 |
| | | | | | | It seems to work now. Reviewed-by: Nicolai Hähnle <[email protected]> | ||||
* | radeonsi: fix DCC fast clear for luminance and alpha formats | Marek Olšák | 2017-06-22 | 1 | -1/+10 |
| | | | | | | | | | | | I reproduced this bug on Polaris11 and Raven. I can't get this bug on Fiji. The reason might be that Fiji doesn't use 2D tiling for the test due to higher 2D tiling alignment requirements. Fixes piglit: spec@ext_framebuffer_object@fbo-fast-clear Reviewed-by: Nicolai Hähnle <[email protected]> | ||||
* | radeonsi: add separate HUD counters for CB and DB cache flushes | Marek Olšák | 2017-06-22 | 3 | -7/+16 |
| | | | | Reviewed-by: Nicolai Hähnle <[email protected]> | ||||
* | gallium/radeon/gfx9: fix PBO texture uploads to compressed textures | Nicolai Hähnle | 2017-06-19 | 1 | -1/+6 |
| | | | | | | | | | st/mesa creates a surface that reinterprets the compressed blocks as RGBA16UI or RGBA32UI. We have to adjust width0 & height0 accordingly to avoid out-of-bounds memory accesses by CB. Cc: 17.1 <[email protected]> Reviewed-by: Marek Olšák <[email protected]> | ||||
* | gallium/radeon: add a new HUD query for the number of resident handles | Samuel Pitoiset | 2017-06-18 | 3 | -0/+9 |
| | | | | | | | | | Useful for debugging performance issues when ARB_bindless_texture is enabled. This query doesn't make a distinction between texture and image handles. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Marek Olšák <[email protected]> | ||||
* | r600, radeon: move radeon_shader_binary_{init,clean} back to radeon | Emil Velikov | 2017-06-17 | 2 | -0/+26 |
| | | | | | | | | | | | | | Those are used by r600 and radeonsi, so moving them within the former was a bad idea. Fixes: d96a210842b ("r600g,compute: provide local copy of functions from ac_binary.c") Cc: Jan Vesely <[email protected]> Cc: Aaron Watry <[email protected]> Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]> Tested-by: Bas Nieuwenhuizen <[email protected]> | ||||
* | r600g,compute: provide local copy of functions from ac_binary.c | Jan Vesely | 2017-06-16 | 2 | -26/+0 |
| | | | | | | | | | | | | | | This is a verbatim copy of the code. The functions can be cleaned up since r600 does not use all the stuff that gcn does. The symbol names have been changed since we still use ac_binary.h header (for struct definition) v2: Add ifdef guard around r600_binary_clean call (Aaron) Remove stray comment Signed-off-by: Jan Vesely <[email protected]> Tested-By: Aaron Watry <[email protected]> Reviewed-by: Emil Velikov <[email protected]> | ||||
* | gallium/radeon: fix initialization of new resource bindless fields | Samuel Pitoiset | 2017-06-15 | 1 | -0/+2 |
| | | | | | | | r600_resource objects are not calloc'd. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Marek Olšák <[email protected]> | ||||
* | radeonsi: upload new descriptors when resident buffers are invalidated | Samuel Pitoiset | 2017-06-14 | 1 | -0/+4 |
| | | | | | | | | | | | | | When texture buffers are invalidated the addr in the resident descriptor has to be updated but we can't create a new descriptor because the resident handle has to be the same. Instead, use the WRITE_DATA packet which allows to update memory directly but graphics/compute have to be idle in case the GPU is reading the descriptor. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Marek Olšák <[email protected]> | ||||
* | gallium/radeon: Include the family name in the renderer string if it's not ↵ | Henri Verbeet | 2017-06-13 | 1 | -14/+18 |
| | | | | | | | | | | | equal to the marketing name. The "family" name is often more informative than the "marketing" name. More importantly, applications, like for example Wine, may recognise GPUs based on the existing "family" names. Reviewed-by: Marek Olšák <[email protected]> Signed-off-by: Henri Verbeet <[email protected]> | ||||
* | radeonsi: pack si_framebuffer better | Marek Olšák | 2017-06-12 | 2 | -2/+2 |
| | | | | Reviewed-by: Nicolai Hähnle <[email protected]> | ||||
* | gallium/radeon: don't allocate HTILE in a separate buffer | Marek Olšák | 2017-06-08 | 2 | -33/+26 |
| | | | | | Reviewed-by: Samuel Pitoiset <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]> | ||||
* | radeonsi: Use libdrm to get chipset name | Samuel Li | 2017-06-07 | 2 | -1/+12 |
| | | | | | | | | v2: Add a func pointer to radeon_winsys to support radeon later. Change-Id: I614ea71424f9e5c97e4ae68654315d28c89eaa5f Signed-off-by: Samuel Li <[email protected]> Signed-off-by: Marek Olšák <[email protected]> | ||||
* | gallium/radeon: clean up a misleading statement from the old days | Marek Olšák | 2017-06-07 | 1 | -4/+1 |
| | | | | Reviewed-by: Nicolai Hähnle <[email protected]> | ||||
* | radeonsi: don't use 1D tiling for Z/S on VI to get TC-compatible HTILE | Marek Olšák | 2017-06-07 | 1 | -3/+13 |
| | | | | | | It's always good to have fewer decompress blits. Reviewed-by: Nicolai Hähnle <[email protected]> | ||||
* | radeon: remove out of date LLVM_REVISION.txt | Emil Velikov | 2017-06-05 | 2 | -4/+0 |
| | | | | | | | | | | | | The file was introduced to track which LLVM revision was required, yet that has quickly gone out of shape. It has seen no updates since 2013. Cc: Nicolai Hähnle <[email protected]> Cc: Marek Olšák <[email protected]> Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Aaron Watry <[email protected]> | ||||
* | radeon: rename has_uvd info to has_hw_decode | Leo Liu | 2017-05-25 | 1 | -2/+2 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/vcn: add decode message for mpeg4 codec | Leo Liu | 2017-05-25 | 1 | -0/+51 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/vcn: add decode message for mpeg2 codec | Leo Liu | 2017-05-25 | 1 | -0/+66 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/vcn: add decode message for vc1 codec | Leo Liu | 2017-05-25 | 1 | -0/+72 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/vcn: add decode message for hevc codec | Leo Liu | 2017-05-25 | 1 | -0/+225 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/vcn: add decode message decode for avc codec | Leo Liu | 2017-05-25 | 1 | -4/+193 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/vcn: add decode message feedback | Leo Liu | 2017-05-25 | 1 | -1/+5 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/vcn: add decode message destroy | Leo Liu | 2017-05-25 | 1 | -1/+10 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/vcn: add decode message create | Leo Liu | 2017-05-25 | 1 | -1/+21 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/vcn: add common decode part | Leo Liu | 2017-05-25 | 3 | -0/+675 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/winsys: add vcn dec ring type | Leo Liu | 2017-05-25 | 1 | -0/+1 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/winsys: add uvd enc ring type | Leo Liu | 2017-05-25 | 1 | -0/+1 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | radeon/vcn: add vcn decode interface | Leo Liu | 2017-05-25 | 2 | -0/+502 |
| | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> | ||||
* | gallium/radeon: pipe AMDGPU_INFO_NUM_VRAM_CPU_PAGE_FAULTS into gallium HUD | Marek Olšák | 2017-05-23 | 3 | -2/+8 |
| | | | | Reviewed-by: Samuel Pitoiset <[email protected]> | ||||
* | radeon: pass flags that can change shaders to disk_cache_create() | Timothy Arceri | 2017-05-23 | 1 | -1/+2 |
| | | | | | | | | | I wasn't sure if I should filter the flags so that we only use flags that actually change the shader output. To avoid manual updates we just pass in everything for now. Reviewed-by: Eduardo Lima Mitev <[email protected]> Reviewed-by: Marek Olšák <[email protected]> | ||||
* | util/disk_cache: add new driver_flags param to cache keys | Timothy Arceri | 2017-05-23 | 1 | -1/+1 |
| | | | | | | | | | This will be used for things such as adding driver specific environment variables to the key. Allowing us to set environment vars that change the shader and not have the driver ignore them if it finds existing shaders in the cache. Reviewed-by: Eduardo Lima Mitev <[email protected]> | ||||
* | gallium/radeon: add a query for monitoring Gallium thread load | Marek Olšák | 2017-05-22 | 2 | -0/+13 |
| | | | | Reviewed-by: Samuel Pitoiset <[email protected]> | ||||
* | ac: add radeon_info::num_{sdma,compute}_rings | Nicolai Hähnle | 2017-05-18 | 1 | -2/+2 |
| | | | | | | Vulkan needs them. Reviewed-by: Marek Olšák <[email protected]> | ||||
* | ac/radeonsi: move struct radeon_info to ac_gpu_info.h | Nicolai Hähnle | 2017-05-18 | 1 | -61/+1 |
| | | | | Reviewed-by: Marek Olšák <[email protected]> | ||||
* | ac/radeonsi: move surface definitions to new header ac_surface.h | Nicolai Hähnle | 2017-05-18 | 1 | -147/+1 |
| | | | | Reviewed-by: Marek Olšák <[email protected]> | ||||
* | gallium/radeon: use a top-of-pipe timestamp for the start of TIME_ELAPSED | Marek Olšák | 2017-05-17 | 1 | -2/+19 |
| | | | | Reviewed-by: Nicolai Hähnle <[email protected]> | ||||
* | gallium/radeon: add threaded context counter monitoring for HUD | Marek Olšák | 2017-05-15 | 3 | -0/+25 |
| | | | | | | | | | "tc" will be initialized by the next commit. v2: rename stuff according to v2 changes in u_threaded_context Reviewed-by: Nicolai Hähnle <[email protected]> (v1) Tested-by: Dieter Nützel <[email protected]> | ||||
* | radeonsi: implement replace_buffer_storage for the threaded context | Marek Olšák | 2017-05-15 | 2 | -1/+33 |
| | | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Tested-by: Dieter Nützel <[email protected]> | ||||
* | gallium/radeon: subclass and handle threaded_query | Marek Olšák | 2017-05-15 | 2 | -7/+12 |
| | | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Tested-by: Dieter Nützel <[email protected]> | ||||
* | gallium/radeon: subclass threaded_transfer | Marek Olšák | 2017-05-15 | 3 | -25/+25 |
| | | | | | | | v2: use assert on rtransfer->b.staging Reviewed-by: Nicolai Hähnle <[email protected]> Tested-by: Dieter Nützel <[email protected]> | ||||
* | gallium/radeon: subclass threaded_resource | Marek Olšák | 2017-05-15 | 3 | -18/+22 |
| | | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Tested-by: Dieter Nützel <[email protected]> | ||||
* | gallium/radeon: handle other map buffer flags from the threaded context | Marek Olšák | 2017-05-15 | 1 | -2/+4 |
| | | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Tested-by: Dieter Nützel <[email protected]> | ||||
* | gallium/radeon: handle TC_TRANSFER_MAP_THREADED_UNSYNC | Marek Olšák | 2017-05-15 | 3 | -1/+14 |
| | | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Tested-by: Dieter Nützel <[email protected]> | ||||
* | gallium/radeon: unwrap a context if we get a wrapped one | Marek Olšák | 2017-05-15 | 2 | -4/+8 |
| | | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Tested-by: Dieter Nützel <[email protected]> | ||||
* | gallium/radeon: require both WRITE and FLUSH_EXPLICIT in buffer_flush_region | Marek Olšák | 2017-05-15 | 1 | -2/+4 |
| | | | | | | | spotted randomly. Reviewed-by: Nicolai Hähnle <[email protected]> Tested-by: Dieter Nützel <[email protected]> | ||||
* | radeonsi/gfx9: add support for Raven | Marek Olšák | 2017-05-15 | 1 | -0/+2 |
| | | | | | | Cc: 17.1 <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]> | ||||
* | radeon: automake: remove unneeded elf Cflags/Libs | Emil Velikov | 2017-05-11 | 1 | -4/+2 |
| | | | | | | | | | | | | No longer required as of commit d90bf4ef3e1 ("radeon: remove unused radeon_elf_util.{c,h}") v2: Add the required libelf link in src/amd/Makefile.common.am Fixes: d90bf4ef3e1 ("radeon: remove unused radeon_elf_util.{c,h}") Cc: Timothy Arceri <[email protected]> Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Marek Olšák <[email protected]> (v1) |