summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* vk: Move anv_CreateSampler to anv_gen8.cKristian Høgsberg Kristensen2015-08-242-95/+100
| | | | Signed-off-by: Kristian Høgsberg Kristensen <[email protected]>
* vk: Move anv_CreateBufferView to anv_gen8.cKristian Høgsberg Kristensen2015-08-244-76/+109
| | | | Signed-off-by: Kristian Høgsberg Kristensen <[email protected]>
* vk: Add new anv_gen8.c and move CreateDynamicRasterState thereKristian Høgsberg Kristensen2015-08-243-36/+75
| | | | Signed-off-by: Kristian Høgsberg Kristensen <[email protected]>
* vk: Implement multi-gen dispatch mechanismKristian Høgsberg Kristensen2015-08-242-9/+45
|
* vk/meta: Implement depth clearsChad Versace2015-08-201-22/+49
| | | | | Fixes Crucible test func.depthstencil.basic-depth.clear-1.0.op-greater.
* vk: Cache each render pass's number of clear opsChad Versace2015-08-203-15/+26
| | | | | | | | | | | | | During vkCreateRenderPass, count the number of clear ops and store them in new members of anv_render_pass: uint32_t num_color_clear_attachments bool has_depth_clear_attachment bool has_stencil_clear_attachment Cacheing these 8 bytes (including padding) reduces the number of times that anv_cmd_buffer_clear_attachments needs to loop over the pass's attachments.
* vk: Use temp var in vkCreateRenderPass's attachment loopChad Versace2015-08-201-7/+8
| | | | | Store the attachment in a temporary variable and s/pass->attachments[i]/att/ .
* vk: Improve memory locality of anv_render_passChad Versace2015-08-201-8/+7
| | | | | Allocate the pass's array of attachments, anv_render_pass::attachments, in the same allocation as the pass itself.
* vk: Unharcode an argument to sizeofChad Versace2015-08-201-1/+1
| | | | s/struct anv_subpass/pass->subpasses[0])/
* vk/meta: Add Z coord to clear verticesChad Versace2015-08-201-6/+6
| | | | | For now, the Z coordinate is always 0.0. Will later be used for depth clears.
* vk/meta: Restore all saved state in anv_cmd_buffer_restore()Chad Versace2015-08-201-4/+6
| | | | | anv_cmd_buffer_restore() did not restore the old VkDynamicColorBlendState.
* vk/meta: Use consistent names and types in anv_saved_stateChad Versace2015-08-201-4/+5
| | | | | | In struct anv_saved_state, each member's type was a pointer to an Anvil struct and each member's name was prefixed with "old" except cb_state, which was a Vulkan handle whose name lacked "old".
* Add mesa.icd to the .gitignoreNeil Roberts2015-08-191-0/+1
| | | | | | | | Since 4d7e0fa8c731776 this file is generated by the configure script. Reviewed-by: Tapani Palli <[email protected]> Reviewed-by: Ben Widawsky <[email protected]> (cherry picked from commit 885762e18291eb4dc0b449297c3a78f7c036bcde)
* vk/meta: Fix dest format of vkCmdCopyImageChad Versace2015-08-181-1/+2
| | | | | | The source image's format was incorrectly used for both the source view and destination view. For vkCmdCopyImage to correctly translate formats, the destination view's format must be that of the destination image's.
* vk: Assert that swap chain format is a color formatChad Versace2015-08-181-0/+2
|
* vk/image: Don't set anv_surface_view::offset twiceChad Versace2015-08-181-1/+0
| | | | | It was set twice a few lines apart, and the second setting always overrode the first.
* vk/meta: Use anv_format_is_color()Chad Versace2015-08-181-1/+1
| | | | | That is, replace !anv_format_is_depth_or_stencil() with anv_format_is_color(). That conveys the meaning better.
* vk: Add anv_format_is_color()Chad Versace2015-08-181-0/+6
|
* vk: Add anv_format reference to anv_render_pass_attachmentChad Versace2015-08-175-19/+15
| | | | | | | Change type of anv_render_pass_attachment::format from VkFormat to const struct anv_format*. This elimiates the repetitive lookups into the VkFormat -> anv_format table when looping over attachments during anv_cmd_buffer_clear_attachments().
* vk/image: Simplify stencil case for anv_image_create()Chad Versace2015-08-171-14/+12
| | | | | | Stop creating a temporary VkImageCreateInfo with overriden format=VK_FORMAT_S8_UINT. Instead, just pass the format override directly to anv_image_make_surface().
* vk/formats: Add global pointer to anv_format for S8_UINTChad Versace2015-08-172-0/+10
| | | | | | | Stencil formats are often a special case. To reduce the number of lookups into the VkFormat-to-anv_format translation table when working with stencil, expose the table's entry for VK_FORMAT_S8_UINT as global variable anv_format_s8_uint.
* vk: Add anv_format reference t anv_surface_viewChad Versace2015-08-175-19/+10
| | | | | | Change type of anv_surface_view::format from VkFormat to const struct anv_format*. This reduces the number of lookups in the VkFormat -> anv_format table.
* vk: Pass anv_format to anv_fill_buffer_surface_state()Chad Versace2015-08-173-8/+8
| | | | | | | This moves the translation of VkFormat to anv_format from anv_fill_buffer_surface_state() to its caller. A prep commit to reduce more VkFormat -> anv_format translations.
* vk: Add anv_format reference to anv_imageChad Versace2015-08-173-29/+22
| | | | | | Change type of anv_image::format from VkFormat to const struct anv_format*. This reduces the number of lookups in the VkFormat -> anv_format table.
* vk: Store the original VkFormat in anv_formatChad Versace2015-08-172-1/+2
| | | | | | Store the original VkFormat as anv_format::vk_format. This will be used to reduce format indirection, such as lookups into the VkFormat -> anv_format translation table.
* vk: Update .gitignore for the autogenerated spirv changesJason Ekstrand2015-08-171-1/+1
|
* vk: Drop aub dumper and PCI ID override featureKristian Høgsberg Kristensen2015-08-174-377/+19
| | | | | | These are now available in intel_aubdump from intel-gpu-tools. Signed-off-by: Kristian Høgsberg Kristensen <[email protected]>
* vk: Use anv_image_create() for creating dmabuf VkImageKristian Høgsberg Kristensen2015-08-174-14/+30
| | | | | | | We need to make sure we use the VkImage infrastructure for creating dmabuf images. Signed-off-by: Kristian Høgsberg Kristensen <[email protected]>
* vk: Add an _autogen suffix autogenerated spirv file namesJason Ekstrand2015-08-172-3/+3
| | | | This prevents make from stomping on nir_spirv.h
* Merge remote-tracking branch 'mesa-public/master' into vulkanJason Ekstrand2015-08-171323-20097/+65538
|\
| * nvc0: disable tessellation on maxwellIlia Mirkin2015-08-141-2/+5
| | | | | | | | | | | | | | | | The address calculations are all different (e.g. see GP), there appear to be sync's in programs, and probably a bunch of other differences. Just disable it for now. Signed-off-by: Ilia Mirkin <[email protected]>
| * nir: Add support for CSE on textures.Eric Anholt2015-08-141-4/+39
| | | | | | | | | | | | | | | | | | | | | | | | NIR instruction count results on i965: total instructions in shared programs: 1261954 -> 1261937 (-0.00%) instructions in affected programs: 455 -> 438 (-3.74%) One in yofrankie, two in tropics. Apparently i965 had also optimized all of these out anyway. Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
| * nir: Zero out texture instructions when creating them.Eric Anholt2015-08-141-1/+1
| | | | | | | | | | | | | | | | There are so many flags in textures, that the CSE pass would have a hard time referencing the correct set when figuring out if two texture ops are the same. By zeroing, we can avoid that fragility. Reviewed-by: Jason Ekstrand <[email protected]>
| * vc4: Move all of our fixed function fragment color handling to NIR.Eric Anholt2015-08-146-388/+538
| | | | | | | | | | | | | | | | | | | | This massively reduces our dependency on VC4-specific optimization passes. shader-db: total uniforms in shared programs: 32077 -> 32067 (-0.03%) uniforms in affected programs: 149 -> 139 (-6.71%) total instructions in shared programs: 98208 -> 98182 (-0.03%) instructions in affected programs: 2154 -> 2128 (-1.21%)
| * vc4: Add a helper for making driver-specific NIR load_uniform for GL stateEric Anholt2015-08-142-2/+30
| | | | | | | | | | | | | | | | In order to move more of our lowering into NIR, we need the ability to reference various pipeline state (like texture rectangle scaling factors or blend colors), so we just set those up as a load_uniform with a big offset to indicate that it's not within the shader's uniform storage and is one of our state values.
| * nir: Don't try to scalarize unpack ops.Eric Anholt2015-08-141-0/+15
| | | | | | | | | | | | | | | | | | Avoids regressions in vc4 when trying to do our blending in NIR. v2: Add the other unpack ops I meant to when writing the original commit message. Reviewed-by: Matt Turner <[email protected]>
| * nir: Add a nir_opt_undef() to handle csels with undef.Eric Anholt2015-08-144-0/+108
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We may find a cause to do more undef optimization in the future, but for now this fixes up things after if flattening. vc4 was handling this internally most of the time, but a GLB2.7 shader that did a conditional discard and assign gl_FragColor in the else was still emitting some extra code. total instructions in shared programs: 100809 -> 100795 (-0.01%) instructions in affected programs: 37 -> 23 (-37.84%) v2: Use nir_instr_rewrite_src() to update def/use on src[0] (by Thomas Helland). v3: Make sure to flag metadata dirties, and copy the swizzle and abs/neg over to src[0], too (by anholt). Reviewed-by: Thomas Helland <[email protected]> (v2) Tested-by: Thomas Helland <[email protected]> (v2)
| * gm107/ir: indirect handle goes first on maxwell alsoIlia Mirkin2015-08-141-8/+4
| | | | | | | | | | | | | | Fixes fs-simple-texture-size.shader_test Signed-off-by: Ilia Mirkin <[email protected]> Cc: "10.6" <[email protected]>
| * nv30: add depth bounds test support for hw that has itIlia Mirkin2015-08-143-2/+14
| | | | | | | | Signed-off-by: Ilia Mirkin <[email protected]>
| * nv50: add depth bounds test supportIlia Mirkin2015-08-143-2/+12
| | | | | | | | Signed-off-by: Ilia Mirkin <[email protected]>
| * nvc0: add depth bounds test supportIlia Mirkin2015-08-143-2/+9
| | | | | | | | Signed-off-by: Ilia Mirkin <[email protected]>
| * docs/relnotes: document amdgpu, GL 4.1 and other new featuresMarek Olšák2015-08-141-0/+6
| |
| * radeonsi: add all new VI PCI IDs including FijiMarek Olšák2015-08-141-0/+24
| |
| * radeonsi: revert a wrong DB bug workaround for VIMarek Olšák2015-08-141-4/+0
| | | | | | | | | | | | | | The bug was misunderstood. Besides that, the bug affects a DB feature we don't use yet. Reviewed-by: Michel Dänzer <[email protected]>
| * radeon/uvd: implement HEVC supportBoyuan Zhang2015-08-143-17/+298
| | | | | | | | | | | | | | | | | | | | | | | | | | add context buffer to fix H265 uvd decode issue. fix H265 corruption issue caused by incorrect assigned ref_pic_list. v2: disable interlace for HEVC add CZ sps flag workaround fix coding style Signed-off-by: Christian König <[email protected]> Signed-off-by: Boyuan Zhang <[email protected]> Reviewed-by: Leo Liu <[email protected]>
| * radeon/vce: disable VCE dual instance for harvest partLeo Liu2015-08-143-1/+5
| | | | | | | | | | | | Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Christian König <[email protected]>
| * radeon/vce: implement VCE dual instance supportLeo Liu2015-08-143-5/+30
| | | | | | | | | | | | | | | | | | | | VCE dual instances are encoding in parallel, it needs two frames for encoding with their own parameters in one IB. Master instance will check the task info to find another frame, assign it to the slave instance Signed-off-by: Leo Liu <[email protected]> Signed-off-by: Christian König <[email protected]> Acked-by: Alex Deucher <[email protected]>
| * radeon/video: config encode stacked frame number based on HWLeo Liu2015-08-141-0/+2
| | | | | | | | | | | | | | | | since VCE 3.0 with dual instances, we need stack frames for them. Signed-off-by: Leo Liu <[email protected]> Acked-by: Alex Deucher <[email protected]> Reviewed-by: Christian König <[email protected]>
| * radeon/vce: make reloc offset signedChristian König2015-08-144-6/+6
| | | | | | | | | | | | | | We need a negative offset for FW 50. Signed-off-by: Christian König <[email protected]> Acked-by: Alex Deucher <[email protected]>
| * radeon/vce: add config task and put task info into encoder v2Leo Liu2015-08-144-33/+47
| | | | | | | | | | | | | | | | | | | | | | The config task has own task ID, extract the configuration functions into config task. v2 (chk): calculate offset automatically Signed-off-by: Leo Liu <[email protected]> Signed-off-by: Christian König <[email protected]> Acked-by: Alex Deucher <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-28 20:53:15 +0000