summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* iris: Copy anv's MI_MATH helpers for multiplication and divisionJason Ekstrand2019-02-212-0/+152
| | | | | (import done by Ken but with author set to Jason because it's his code that's being imported, so he deserves the credit)
* iris: only get space for one offset in stream output targetsKenneth Graunke2019-02-211-2/+2
| | | | | Target corresponds to a buffer, buffer only records one offset, not multiple.
* iris: Move iris_stream_output_target def to iris_context.hKenneth Graunke2019-02-212-10/+10
| | | | now that it doesn't have genxml
* iris: Don't bother packing 3DSTATE_SO_BUFFER at create timeKenneth Graunke2019-02-211-27/+16
| | | | | We have to do half the packet late anyway, we may as well just do it all at set time. This also lets us move the struct def out of genxml
* iris: Add _MI_ALU helpers that don't pasteKenneth Graunke2019-02-211-4/+9
| | | | This lets you pass arguments as function parameters
* iris: Reorder LRR parameters to have dst first.Kenneth Graunke2019-02-213-13/+13
| | | | LRI and LRM both put dst first, be consistent.
* iris: rewrite set_vertex_buffer and VB handlingKenneth Graunke2019-02-212-53/+65
| | | | | | | | | I was using the Gallium API wrong. set_* functions with start_slot and count parameters are supposed to update a subrange of the items. I had been trashing all bound vertex buffers and starting over. This should hopefully also make it easier to slot in additional VERTEX_BUFFER_STATEs at draw time, say, for shader draw parameters.
* iris: handle PatchVerticesIn as a system value.Kenneth Graunke2019-02-214-2/+32
|
* iris: add Android buildTapani Pälli2019-02-216-0/+245
| | | | | | | | | | | | | | | Note that at least following additional libs/components require changes since they refer to BOARD_GPU_DRIVERS variable which is used to select the driver: - mixins - minigbm - libdrm - drm_gralloc v2: (feedback by Gustaw Smolarczyk) Fix trailing \ in a few cases Signed-off-by: Tapani Pälli <[email protected]>
* iris: override alpha to one src1 blend factorsKenneth Graunke2019-02-211-10/+36
| | | | | No idea why this used to pass and doesn't after updating...seems like we should have been handling it all along...
* iris: Always do rasterizer discard in clipperKenneth Graunke2019-02-211-6/+7
| | | | | | but continue doing it in SOL if possible because it's faster Fixes ./bin/ext_transform_feedback-discard-drawarrays - simpler too
* iris: Fix primitive generated query active flagKenneth Graunke2019-02-211-1/+1
|
* iris: Enable guardband clippingKenneth Graunke2019-02-211-9/+10
|
* iris: Clamp viewport extents to the framebuffer dimensionsKenneth Graunke2019-02-211-38/+45
| | | | Fixes arb_framebuffer_no_attachments-query's resize subtest.
* iris: Fix clear dimensionsKenneth Graunke2019-02-211-2/+3
| | | | Fixes depthstencil-render-miplevels 1024 s=z24_s8
* iris: Drop continues in resolveKenneth Graunke2019-02-211-5/+0
| | | | Now that we u_bit_scan we know it exists
* iris: Replace num_textures etc with a bitmask we can scanKenneth Graunke2019-02-213-14/+21
| | | | More accurate bounds, plus can skip dead ones
* iris: Fix set_sampler_views with start > 0Kenneth Graunke2019-02-211-1/+1
|
* iris: fix set_sampler_views to not unbind, be better about boundsKenneth Graunke2019-02-211-9/+6
|
* iris: fix overhead regression from flushing for storage imagesKenneth Graunke2019-02-211-1/+2
| | | | | st calls us with count = 32 but a NULL pointer...we only really care about the highest non-NULL image...
* iris: Fix NOS mechanismKenneth Graunke2019-02-212-8/+7
| | | | Set bits, not values
* iris: re-pin inherited streamout buffersKenneth Graunke2019-02-211-0/+13
|
* iris: reemit SBE when sprite coord origin changesKenneth Graunke2019-02-211-1/+3
| | | | fixes arb_point_sprite-checkerboard
* iris: omask can killKenneth Graunke2019-02-211-1/+3
|
* iris: reject all clipping when we can't use streamout render disabledKenneth Graunke2019-02-212-4/+6
|
* iris: make clipper statistics dynamicKenneth Graunke2019-02-211-1/+2
|
* iris: CS stall for stream out -> VBKenneth Graunke2019-02-211-6/+13
| | | | | i965 doesn't do this, but I suspect it just stalls a lot and doesn't hit this. Fixes ext_transform_feedback-position render among others.
* iris: fix dma buf import stridesKenneth Graunke2019-02-211-1/+1
|
* iris: fix alpha channel for RGB BC1 formatsKenneth Graunke2019-02-211-8/+13
|
* iris: Allocate buffer resources separatelyJason Ekstrand2019-02-212-58/+87
| | | | | | | (cleaned up by Ken - make sure a bunch of things were more obviously not using res->surf, do allow checking res->surf.tiling == LINEAR, drop format cpp checks that aren't needed, drop memzone handling for images, assume buffers / non-buffers in a few places...)
* iris: Don't bother considering if the underlying surface is a cubeKenneth Graunke2019-02-211-3/+1
| | | | | | | Dave fixed it to consider whether the sampler view is a cube. With that, there's no point (possibly harm) in looking if the original resource was a cube...if it's an array view, we don't want to treat it as a cube anymore...
* iris: move some non-buffer case code in a bitKenneth Graunke2019-02-211-4/+6
|
* iris: Stop leaking iris_uncompiled_shaders like madKenneth Graunke2019-02-211-5/+0
| | | | | Now shader-db actually executes. We still need a plan for culling dead iris_compiled_shaders...
* iris: Destroy the bufmgrKenneth Graunke2019-02-211-0/+1
| | | | Plugs a 12360 byte leak
* iris: Fix IRIS_MEMZONE_COUNT to exclude the border color poolKenneth Graunke2019-02-211-1/+1
| | | | | | This is supposed to exclude single address zones. We were getting too many VMA allocators but failing to set them up, which worked out because we also forgot to destroy them...
* iris: Unref unbound_tex resourceKenneth Graunke2019-02-211-0/+2
| | | | Plugs a 12536 byte leak
* iris: Destroy the border color poolKenneth Graunke2019-02-213-0/+10
| | | | This plugs a 12224 byte leak
* iris: Destroy transfer helper on screen teardownKenneth Graunke2019-02-211-0/+2
| | | | Plugs a 16 byte leak
* iris: Fix failed to compile TCS messageKenneth Graunke2019-02-211-1/+1
|
* iris: Rework tiling/modifiers handlingKenneth Graunke2019-02-211-30/+31
| | | | | | | | We were being very picky about things being Y tiled. But, not everything can be - for example, > 16382 surfaces on SKL GT1-3 have to fall back to linear. Instead, give ISL options and let it pick.
* iris: fix conditional compute, don't stomp predicate for pipelined queriesKenneth Graunke2019-02-216-66/+89
|
* iris: check query firstKenneth Graunke2019-02-211-0/+17
| | | | this lets us avoid the predicate bit in more cases, which is nice
* iris: for BLORP, only use the predicate enable bit when USE_BITKenneth Graunke2019-02-212-3/+18
|
* iris: add conditional render supportDave Airlie2019-02-215-3/+119
|
* iris: drop key_size_for_cacheKenneth Graunke2019-02-211-18/+0
| | | | | | dead since my program cache API rework. we could still use it for one function, but it's so trivial to pass the size, that it's probably not worth the extra code
* iris: iris add load register reg32/64Dave Airlie2019-02-212-0/+30
| | | | These will be needed for broadwell and conditional render
* iris: execute compute related query on compute batch.Dave Airlie2019-02-211-6/+12
| | | | This only happens for the compute invocations query.
* iris: fix cube texture viewDave Airlie2019-02-211-0/+4
|
* iris: fix some SO overflow query bugs and tidy the code a bitKenneth Graunke2019-02-211-25/+33
|
* iris: add initial transform feedback overflow query paths (V3)Dave Airlie2019-02-211-2/+129
| | | | | v2: fix cpu overflow calc v3: use a struct
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-26 17:21:30 +0000