This worked because the optimizer didn't consider that the 16-bit
instruction would interpret the inline constant differently. This will
change in the next commit.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5245>
p_create_vector can have sub-dword operands with a v1 definition.
No fossil-db changes.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5245>
v_mad_f32 doesn't support denormals but v_fma_f32 does.
No fossil-db changes.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5245>
This check was added because I thought v_mad_f32 didn't preserve the
signess of zero, but I can't reproduce that and this isn't mentioned
anywhere in LLVM.
No fossil-db changes.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5245>
Use sub-dword definitions so that the RA can use SDWA
No fossil-db changes.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5245>
It's tricky to merge XFB-outputs correctly, because we need there to not
be any overlaps when we get to `nir_gather_xfb_info_with_varyings` later
on. We currently trigger an assert there if we end up merging here.
So let's not even try. This is an optimization, and we can optimize this
in safe cases later if needed. For now, let's play it safe.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5329>
The only reason for this dependency was the fd_bo used for the uploaded
shader. But this isn't used by turnip. Now that we've unified the
cleanup path from gallium, it isn't hard to pull the fd_bo upload/free
parts into ir3_gallium.
This cleanup has the added benefit that the shader disk-cache will not
have to deal with it.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5476>
In particular, to move the fd_bo create/delete (which is unneeded by
turnip) out of the shared ir3 code, it is useful to have a single
delete path.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5476>
glsl builtins that have no analog in spirv are emitted as regular varyings,
which means they take up a slot.
we need to ensure that there's no conflict between these regular varying
slots (from user-defined varyings) and the glsl translated builtins, so
we do that by "reserving" the max number of varying slots that can be used
by a given stage, then remapping all glsl builtins with no spirv builtin
to a packed layout location that can be consistent across stages
sort of addresses mesa/mesa#3113 except now there's 10 fewer varying slots
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5432>
this should be all of them, though the check for vertex shader stage needs
to be changed to !fragment stage at some point
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5432>
This seems valid per the SPIR-V spec to use OpSampledImage with
OpUndef instead of OpTypeImage or OpTypeSampler. When the image
operand is undefined, SPIRV->NIR emits an undef instruction that
can be removed later by the compiler.
This fixes shader compilation crashes with Red Dead Redemption II.
Cc: mesa-stable@lists.freedesktop.org>
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5230>
Instead of requiring an explicit unoptimized move, we can implicitly
colour the blend input intrinsic to r0, where it will be preloaded; this
is a simple task for RA, and does not conflict with anything. If there
are multiple duplicate loads, the latter ones can just be simple moves
which will be copypropped.
We don't need to include a explicit synthetic load, since (scanning
backwards) the read will cause the input to become live at the right
time and the lack of an explicit write will keep it live from the
beginning of the shader. So no need to make it more complicated than it
needs to be.
Saves a cycle in blend shaders.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5449>
ir3_nir_move_varying_inputs is broken when there a load input outside of
the first block which depends on the result of a previous load input.
This simplification/rework avoids the problem, and should also be faster.
Fixes this dEQP-VK test:
dEQP-VK.pipeline.multisample_interpolation.offset_interpolate_at_pixel_center.128_128_1.samples_2
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5465>
Old script created files in the source directory, which is generally
considered bad form.
The rewrite to python instead of duct-taping around in the shell script
goes towards the goal of only having cross-platform python scripts,
which is also harder to make mistakes in than shell scripts.
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5155>
There hasn't been a config.h in a long time (it was an artifact of the
autotool build).
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5350>
There hasn't been a config.h in a long time (it was an artifact of the
autotool build).
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5350>
Python 2 is dead and this script is only run by devs, all of which have
had python3 available for basically forever.
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5362>
CoverityID: 1435701
Fixes: e5a81ac704 ("broadcom/vc5: Don't forget to get the BO offset when opening a dmabuf.")
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5263>
Tiling is expensive, so this patch converts textures that appear to be
used for streaming to a linear layout.
Performance of mpv is significantly improved, with software-decoded
1080p mp4 playback on RK3288 going from 30fps to 50fps when testing
with `--untimed --no-audio`.
To keep things simple, conversion only happens when updating the whole
texture and no mipmapping is used.
v2: Make it clear that the heuristic doesn't rely on a texture being
uninitialized, since layout switching code can get confusing (Alyssa).
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4628>
Gen9 and Cherryview have the ability to mark texture instructions with
the End-of-thread bit under some conditions, which allows the texture
result to be written to the render target directly, rather than
returning to the EU.
In order to handle overlapping primitives correctly, we have to use the
'sendc' instruction which stalls until other threads potentially writing
to the same locations in the render target are retired. Unfortunately,
this stall happens before the texture is sampled (rather than in
parallel with stall), so for some literal edge cases (like the diagonal
edge between two triangles forming a rectangle) there can be a
performance penalty. As a result, it's probably not a good idea to use
this optimization in general.
I had planned to leave it enabled only for BLORP, where we use rectangle
primitives and are typically clearing/blitting an entire render target
without any overlapping primitives, but I noticed that the optimization
wasn't applied in some normal cases anyway. For example, in the piglit
test tests/shaders/glsl-fs-texture2d-bias.shader_test it is applied to
one BLORP-blit shader but not another due to some kind of mishandling of
register types (the destination register type of the texture operation
is UD while the color source of the render target write is F).
Additionally the instruction scheduler assumed that the combined texture
and render target write operation took 0 cycles, leading to cycle
estimates that are wildly inaccurate. Since the optimization was not
implemented for SIMD32 and our decision whether to use the SIMD32
program is made by comparing the estimated performance with that of the
SIMD16 shader, we wrongly threw out a bunch of SIMD32 programs that are
likely profitable.
total cycles in shared programs: 472807891 -> 473784245 (0.21%)
cycles in affected programs: 108277 -> 1084631 (901.72%)
helped: 0
HURT: 1290
total sends in shared programs: 998955 -> 1000245 (0.13%)
sends in affected programs: 1400 -> 2690 (92.14%)
helped: 0
HURT: 1290
LOST: 0
GAINED: 33
This patch shows no performance changes in Intel's Mesa performance CI.
Given the problems, the lack of evidence that the pass improves
performance, and the fact that the hardware feature was removed from
subsequent GPU generations, I think that the pass is not valuable and
should be removed.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5412>
This fixes cases where the 3D path is used with layered rendering.
Fixes dEQP-VK.renderpass.suballocation.multisample_resolve.layers* failures
Note the blob's 3D fallback path behaves differently, and uses the
framebuffer information to clear each layer individually (changing the MRT
state each time). But that's not possible in all cases, and the blob fails
to clear properly in dEQP-VK.geometry.layered.*.secondary_cmd_buffer cases.
So this clear path is not based on the blob's behavior.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5426>
This refactor simplifies things a bit, and will make it easier to share
some logic with tu_clear_blit (see next patches).
This changes the order in which some things are emitted, and emits less
for disabled shader stages. There's also as extra write to SP_GS_PRIM_SIZE
that is removed.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5426>
