aboutsummaryrefslogtreecommitdiffstats
path: root/src/amd
Commit message (Collapse)AuthorAgeFilesLines
* radv: Use the guard band.Bas Nieuwenhuizen2017-03-303-13/+90
| | | | | Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Prepare for not using the guard band for lines & points.Bas Nieuwenhuizen2017-03-303-0/+32
| | | | | | | | | | | Vulkan Clipping is defined in terms of vertices, the scissor based clipping happens on pixels. There is a difference with points and lines, as a vertex can be outside the viewport while some pixels are in. On Vulkan thoise pixels shouldn't be drawn, while they would be with the guardband. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Drop the default viewport when 0 viewports are given.Bas Nieuwenhuizen2017-03-301-17/+2
| | | | | Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Set proper viewport & scissor for meta draws.Bas Nieuwenhuizen2017-03-306-75/+214
| | | | | Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* Partially revert "amd/addrlib: silence warnings" to fix builds with DEBUGMarek Olšák2017-03-303-6/+6
| | | | This partially reverts commit 8a74140a21fe6b0d2e8a60b065b890f797f2db51.
* radeonsi/gfx9: image descriptor changes in immutable fieldsMarek Olšák2017-03-301-0/+6
| | | | | | | The border color swizzle logic was copied from Vulkan. It doesn't make any sense to me, but it passes all piglits except the stencil ones. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi/gfx9: init_config changesMarek Olšák2017-03-301-0/+4
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi/gfx9: CP DMA changesMarek Olšák2017-03-301-0/+1
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi/gfx9: add IB parser supportMarek Olšák2017-03-303-19/+36
| | | | | | | | | Both GFX6 and GFX9 fields are printed next to each other in parsed IBs. The Python script parses both headers like one stream and tries to merge all definitions. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi/gfx9: add GFX9 and VEGA10 enumsMarek Olšák2017-03-301-3/+5
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* amd: GFX9 packet changesMarek Olšák2017-03-303-11/+24
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* amd: define event types for GFX9Marek Olšák2017-03-301-0/+54
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* amd: add texture format definitions for GFX9Marek Olšák2017-03-304-23/+148
| | | | | | | | the DATA_FORMAT and NUM_FORMAT fields are the same, but some of the enums differ, thus add GFX6 and GFX9 suffixes, so that the IB parser can show enums for both. Reviewed-by: Nicolai Hähnle <[email protected]>
* amd: resolve remaining definition conflicts with gfx9d.hMarek Olšák2017-03-303-63/+63
| | | | | | | | Add _GFX6 and _GFX9 suffixes to conflicting definitions. sid.h and gfx9d.h can now be included in the same file. Reviewed-by: Nicolai Hähnle <[email protected]>
* amd: normalize register definition formattingMarek Olšák2017-03-302-77/+129
| | | | | | | This resolves trivial conflicts with gfx9d.h caused by different formatting. Some fields are also renamed. Reviewed-by: Nicolai Hähnle <[email protected]>
* amd: import GFX9 register definitionsMarek Olšák2017-03-302-0/+7287
| | | | Acked-by: Nicolai Hähnle <[email protected]>
* amd/addrlib: silence warningsMarek Olšák2017-03-304-15/+15
|
* amd/addrlib: import gfx9 supportNicolai Hähnle2017-03-3019-3/+22053
|
* amd/addrlib: Not all ETC2 formats are 128bpp... add new ETC2 formats to ↵Kevin Furrow2017-03-303-19/+45
| | | | differentiate between 64 and 128bpp formats.
* amd/addrlib: Fix selection of swizzle modes for 3D compressed images.Kevin Furrow2017-03-301-1/+2
|
* amd/addrlib: Add support for ETC2 and ASTC formats.Kevin Furrow2017-03-303-1/+119
|
* amd/addrlib: Bump version to 6.02Joe Ma2017-03-301-1/+1
|
* amd/addrlib: Adjust slie size after pitch and actual height adjustmentFrans Gu2017-03-301-26/+31
|
* amd/addrlib: Apply input pitch after internal pitch aligningFrans Gu2017-03-301-12/+33
|
* amdgpu/addrlib: Bump version to 6.01Nicolai Hähnle2017-03-301-2/+2
| | | | Signed-off-by: Nicolai Hähnle <[email protected]>
* amdgpu/addrlib: Seperate 2 dcc related workarounds by different flagsNicolai Hähnle2017-03-303-3/+9
| | | | | 1) dccCompatible for padding MSAA surface to support fast clear 2) dccPipeWorkaround for padding surface to support dcc
* amdgpu/addrlib: Fix the issue that tcCompatible HTILE slice size is not ↵Nicolai Hähnle2017-03-302-0/+18
| | | | calculated correctly
* amdgpu/addrlib: Add a new output flag to notify client that the returned ↵Nicolai Hähnle2017-03-302-1/+5
| | | | | | | tile index is for PRT on SI If this flag is set for mip0, client should set prt flag for sub mips, so that address lib can select the correct tile index for sub mips.
* amdgpu/addrlib: add matchStencilTileCfg and tcCompatible fixesXavi Zhang2017-03-304-20/+152
| | | | | | | | | | | | | | | | | | | | | | | | | | | The usage should be client first call AddrComputeSurfaceInfo() on depth surface with flag "matchStencilTilecfg", AddrLib will use 2DThin1 tile index for depth as much as possible and do not down grade unless alignment requirement cannot be met. 1. If there is a matched 2DThin1 tile index for stencil which make sure they will share same tile config parameters, then return the stencil 2DThin1 tile index as well. 2. If using 2DThin1 tile mode cannot make sure such thing happen, and TcCompatible flag was set, then ignore this flag then try 2DThin1 tile mode for depth and stencil again. 3. If 2DThin1 tile mode cannot make sure depth and stencil to have same tile config parameters, then down grade depth surface tile mode to 1DThin1. 4. If depth surface's tile mode was 1DThin1, then return 1DThin1 tile index for stencil. 5. If depth surface's tile mode is PRT, then return invalid tile index to stencil since their tile config parameters will never be met. Client driver then check the returned tile index of stencil -- if it is not invalid tile index, then call AddrComputeSurfaceInfo() on stencil surface with the returned stencil tile index to get full output information. Please note, client needs to set flag "useTileIndex" when AddrLib get created.
* amdgpu/addrlib: Adjust bank equation bit order based on macro tile aspect ↵Frans Gu2017-03-304-91/+282
| | | | | | | | ratio settings By this way, we can have valid equation for 2D_THIN1 tile mode. Add flag "preferEquation" to return equation index without adjusting input tile mode.
* amdgpu/addrlib: do some tile mode conversions to display surfaceFrans Gu2017-03-301-2/+3
|
* amdgpu/addrlib: Check prt flag for PRT_THIN1 extra padding for DCC.Xavi Zhang2017-03-306-92/+56
|
* amdgpu/addrlib: Add new flags minimizePadding and maxBaseAlignFrans Gu2017-03-309-157/+401
| | | | | | | | | 1) minimizePadding - Use 1D tile mode if padded size of 2D is bigger than 1D 2) maxBaseAlign - Force PRT tile mode if macro block size is bigger than requested alignment. Also, related changes to tile mode optimization for needEquation.
* amdgpu/addrlib: Always returns pixelPitch in original pixelsXavi Zhang2017-03-301-14/+10
|
* amdgpu/addrlib: fix crash on allocation failureSabre Shao2017-03-305-36/+31
|
* amdgpu/addrlib: Add flag to report if a surface can have dcc ramFrans Gu2017-03-303-4/+28
|
* amdgpu/addrlib: support non-power2 height alignment (for linear surface)Roy Zhan2017-03-301-1/+10
|
* amdgpu/addrlib: Fix family setting for VI and CZ ASICsFrans Gu2017-03-301-0/+2
|
* amdgpu/addrlib: style cleanupNicolai Hähnle2017-03-302-28/+15
| | | | Signed-off-by: Nicolai Hähnle <[email protected]>
* amdgpu/addrlib: Pad pitch to multiples of 256 for DCC surface on FijiNicolai Hähnle2017-03-308-49/+131
| | | | | The change also modifies function CiLib::HwlPadDimensions to report adjusted pitch alignment.
* amdgpu/addrlib: Fix number of //Xavi Zhang2017-03-308-66/+66
| | | | | | Find ^/{80,99}$ and replace them to 100 "/" Signed-off-by: Nicolai Hähnle <[email protected]>
* amdgpu/addrlib: Cleanup.Nicolai Hähnle2017-03-3013-73/+68
| | | | Signed-off-by: Nicolai Hähnle <[email protected]>
* amdgpu/addrlib: Use namespacesXavi Zhang2017-03-3016-892/+969
| | | | Signed-off-by: Nicolai Hähnle <[email protected]>
* amdgpu/addrlib: Adjust 99 "*" to 100 "*" alignmentKevin Zhao2017-03-3018-895/+895
| | | | Signed-off-by: Nicolai Hähnle <[email protected]>
* amdgpu/addrlib: Add a new tile mode ADDR_TM_UNKNOWNFrans Gu2017-03-308-14/+211
| | | | | This can be used by address lib client to ask address lib to select tile mode.
* amdgpu/addrlib: Stylish cleanup.Xavi Zhang2017-03-305-17/+16
| | | | Signed-off-by: Nicolai Hähnle <[email protected]>
* amdgpu/addrlib: Disable tcComaptible when depth surface is not macro tiledRoy Zhan2017-03-301-17/+20
| | | | Experiment show 1D tiling + TcCompatible cannot work together.
* amdgpu/addrlib: fix pixel index calculation of thick micro tilingXavi Zhang2017-03-301-4/+4
|
* amdgpu/addrlib: Add a flag to skip calculate indicesXavi Zhang2017-03-303-14/+21
| | | | | This is useful for debugging and special cases for stencil surfaces do not require texture fetch compatible.
* amdgpu/addrlib: add equation generationNicolai Hähnle2017-03-3012-118/+1344
| | | | | | | | | | | | | | | | 1. Add new surface flags needEquation for client driver use to force the surface tile setting equation compatible. Override 2D/3D macro tile mode to PRT_* tile mode if this flag is TRUE and num slice > 1. 2. Add numEquations and pEquationTable in ADDR_CREATE_OUTPUT structure to return number of equations and the equation table to client driver 3. Add equationIndex in ADDR_COMPUTE_SURFACE_INFO_OUTPUT structure to return the equation index to client driver Please note the use of address equation has following restrictions: 1) The surface can't be splitable 2) The surface can't have non zero tile swizzle value 3) Surface with > 1 slices must have PRT tile mode, which disable slice rotation
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-26 00:10:23 +0000