The previous pass was based on misconceptions and
rounded up with bug fixes. The new pass is entirely
rewritten and basically just one-to-one from the paper:
"Revisiting Out-of-SSA Translation for Correctness, CodeQuality, and Efficiency"
by B. Boissinot et al.
It also incorporates the value-equality testing.
The regressions are mainly due to creating parallelcopies for
exec phis at loop headers (mitigated in the next commit).
Totals from 4933 (3.61% of 136546) affected shaders (Raven):
SpillSGPRs: 16249 -> 16527 (+1.71%); split: -0.28%, +1.99%
SpillVGPRs: 1771 -> 1595 (-9.94%)
CodeSize: 57544436 -> 58280304 (+1.28%); split: -0.00%, +1.28%
Scratch: 176128 -> 179200 (+1.74%)
Instrs: 11265783 -> 11445884 (+1.60%); split: -0.00%, +1.60%
Latency: 552596156 -> 555880540 (+0.59%); split: -0.53%, +1.13%
InvThroughput: 271431862 -> 273097423 (+0.61%); split: -0.18%, +0.79%
VClause: 160240 -> 160241 (+0.00%); split: -0.02%, +0.02%
SClause: 386863 -> 386685 (-0.05%); split: -0.07%, +0.02%
Copies: 1180801 -> 1345633 (+13.96%); split: -0.02%, +13.98%
Branches: 379129 -> 393052 (+3.67%); split: -0.01%, +3.69%
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9196>
To decide which variables to spill, we must use the distances at the
beginning of the loop-header, and not the distances at the end of the
loop-preheader. The difference are that the former includes phis which are
viable to be spilled as opposed to the phi operands which would be reloaded
by add_coupling_code(), ending up in potentially too high register pressure
before the loop.
Totals from 206 (0.15% of 136546) affected shaders (Raven):
SpillSGPRs: 5154 -> 5000 (-2.99%)
CodeSize: 3654072 -> 3647184 (-0.19%); split: -0.19%, +0.00%
Instrs: 701482 -> 700526 (-0.14%); split: -0.14%, +0.00%
Latency: 40988780 -> 40872506 (-0.28%); split: -0.29%, +0.00%
InvThroughput: 20364560 -> 20306006 (-0.29%)
SClause: 20192 -> 20198 (+0.03%)
Copies: 77732 -> 77688 (-0.06%); split: -0.08%, +0.03%
Branches: 24204 -> 24050 (-0.64%)
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9196>
The KHR_partial_update spec says:
"
If EGL_EXT_buffer_age is supported, the contents of the buffer inside
the damage region may also be relied upon to contain the same content
as the last time they were defined for the current back buffer.
"
but we currently assume that everything inside the damage region will
be overwritten by new data and that the previous content doesn't need
to be reloaded.
Let's get rid of the damage rect inversion logic for now and reload
everything inside the damage extent. We will optimize things further
down the line, using pre-frame DCDs on Bifrost, and a tile enable map
on Midgard.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10033>
pan_blitter.c is meant to replace the pan_blit.c which currently
provides helpers to preload the tile buffer. Some changes
are worth mentioning:
- we use pre-frame DCDs on Bifrost (Midgard still uses a tiler job
to preload the tile buffer)
- the blit shaders are now stored in a hash table and created lazily
- we now cache blend shader binaries and blit RSDs too
This library will soon be extended to cover regular blits.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10033>
The layout is supposed to encode image miplevels/surfaces layout, not
the state data stored in the buffers. It doesn't matter for a gallium
driver, since resources are expected to hold both a layout and a state,
but Vulkan is a bit different. In Vulkan, the image state is explicitly
passed by the user when starting a render pass (vkCmdBeginRenderPass()),
and might evolve depending on the operation done in this render pass.
This state is not effective until the command buffer is queued and
executed. For these reasons, keeping the image state attached to the
VkImage object is not an option, but we'd still like to re-use the
layout and state objects, and all common helpers acting on those objects.
Let's move the state bits out of the layout to make that possible.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10033>
These decorations allow you to override the signedness of image
instructions. This means that we have to override the type we get from
the sampled image.
Apparently both Intel and AMD get the type from the descriptor rather
than the instruction, but this appears to not be the case with Adreno,
which is why this wasn't noticed until now. So this probably won't fix
any preexisting bugs, but it's required to fix
dEQP-VK.image.extend_operands_spirv1p4.* when exposing VK_KHR_spirv_1_4
on turnip.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7968>
When creating the Vextex Shader bin, only store outputs that will be
read later as inputs by the Geometry Shader bin.
This fixes
KHR-GLES31.core.geometry_shader.rendering.rendering.lines_input_triangle_strip_output_line_loop_drawcall
(exposed by previous commit).
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10129>
Vertex Shader has a store_out lowering pass that converts gallium driver
locations in offsets inside the VPM.
One of the consequences is that these offsets are consecutives; that is,
if the VS stores VARYING_SLOT_VAR0.xyz and VARYING_SLOT_VAR1.xyzw, there
isn't a hole in the VPM offsets for the un-stored VARYING_SLOT_VAR0.w.
Thus we need to change how the VPM offset is computed in the Geometry
Shader when loading the inputs.
This bug is exposed by !9050.
v2 (Iago):
- Include explanatory comment.
- Use assert.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10129>
