Add the ability to deduplicate hoisted expressions, which will be
necessary to avoid repeatedly hoisting the same descriptors and blowing
our budget. The offset calculation may have itself been hoisted into the
preamble, so we also have to be able to hoist a bindless_resource_ir3
referencing a load_preamble and connect it to the source of the
corresponding store_preamble.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29873>
We have to be careful here because r63.x is also used to mean "register
not assigned yet," but dummy destinations for prefetches will use r63.x
and we don't want them to count as GPRs when determining whether to
enable early preamble.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29873>
We currently assume that the instruction is already inserted and we are
optimizing it away, but in the use case I have where we are hoisting
instructions into a preamble and deduplicating as we go along, that
isn't the case. Move this responsibility onto the caller, which also
makes it a bit clearer what's going on and turns this into something
more similar to an actual set.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29873>
This allows use cases where we copy over expression trees and
deduplicate as we go along. We can use the matching instruction to build
up the rest of the expression tree.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29873>
Now that there are more rpi5 in the CI, let's add pre-merge jobs.
This also restrict the nightly job to be executed by devices allocated
to run full runs.
Reviewed-by: Eric Engestrom <eric@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30087>
We are about to make a change that will make compiler options
depend on v3d version, so helpers would usually need additional
parameters to retrieve them. Isntead of doing this, we will make
the callers get the options instead and provide them to the
helpers.
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30086>
This requires that our nir compiler options are different between V3D versions
so we can't use a static global any more.
total instructions in shared programs: 11241106 -> 11047872 (-1.72%)
instructions in affected programs: 4634458 -> 4441224 (-4.17%)
helped: 25119
HURT: 1717
Instructions are helped.
total threads in shared programs: 425238 -> 425036 (-0.05%)
threads in affected programs: 878 -> 676 (-23.01%)
helped: 79
HURT: 180
Inconclusive result (%-change mean confidence interval includes 0).
total loops in shared programs: 1968 -> 1933 (-1.78%)
loops in affected programs: 35 -> 0
helped: 35
HURT: 0
Loops are helped.
total uniforms in shared programs: 3845314 -> 3845219 (<.01%)
uniforms in affected programs: 213615 -> 213520 (-0.04%)
helped: 1338
HURT: 1059
Inconclusive result (value mean confidence interval includes 0).
total max-temps in shared programs: 2224313 -> 2221507 (-0.13%)
max-temps in affected programs: 236054 -> 233248 (-1.19%)
helped: 4863
HURT: 3357
Max-temps are helped.
total spills in shared programs: 4264 -> 4294 (0.70%)
spills in affected programs: 274 -> 304 (10.95%)
helped: 8
HURT: 16
total fills in shared programs: 6638 -> 6497 (-2.12%)
fills in affected programs: 2240 -> 2099 (-6.29%)
helped: 55
HURT: 17
total sfu-stalls in shared programs: 14942 -> 14353 (-3.94%)
sfu-stalls in affected programs: 4863 -> 4274 (-12.11%)
helped: 1287
HURT: 1165
Sfu-stalls are helped.
total inst-and-stalls in shared programs: 11256048 -> 11062225 (-1.72%)
inst-and-stalls in affected programs: 4635701 -> 4441878 (-4.18%)
helped: 25074
HURT: 1728
Inst-and-stalls are helped.
total nops in shared programs: 270482 -> 270621 (0.05%)
nops in affected programs: 27579 -> 27718 (0.50%)
helped: 1583
HURT: 1967
Inconclusive result (value mean confidence interval includes 0).
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30086>
Since .sat, .nsat and .max0 are only supported with FMOV we can't copy
propagate an FMOV with any of these unpack modifiers into a different
opcode.
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30086>
Introduce unary and binary versions of the branch instruction. This will
give more ISA_OPC_BRANCH_XXX opcodes to work with. This helps to get rid
of these 'maybe' bitsets and is needed for the assembler.
Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30030>
In isaspec the displayname of a bitset defines what is shown in
dissassembly. The assembler only sees this representation and
needs to be able to handle it.
Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30030>
Don't multiply by 4 twice. While we're here, fix the in-bounds check on
a6xx, since we need to account for the multiplying by 4. This was done
correctly on a7xx but the commit below didn't correctly port it to a6xx
when adding the multiply on a6xx.
Fixes: 01bac643f6 ("freedreno/ir3: Fix ldg/stg offset")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30046>
si_set_patch_vertices was only called if tcs.current was non-NULL but
this condition is not enough for GFX9+ since vs is used as ls.
Add a check in si_update_tess_io_layout_state instead, and set
sctx->do_update_shaders for case where the ls_current is not yet
available.
This fix crashes on GFX6.
Cc: mesa-stable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29876>
buffer_size is not the full buffer size, but the part of the
buffer that is accessed by the compute shader.
This fixes the assert hit in si_set_shader_buffer.
Fixes: 1a99f50c7f ("radeonsi: use a compute shader to convert unsupported indices format")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29876>
We use the method used by the blob, which sets the USE_CPU_MAP flag,
originally intended for SVM, to allocate from a separate address range
and to control the address by passing a preferred address to mmap().
With this we can capture and replay gfxreconstruct traces on kgsl for
apps that use BDA, and we can replay them on msm with a small hack to
increase the address space size:
echo 274877906944 > /sys/module/msm/parameters/address_space_size
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29251>
We select the feedback tranche according to the image format but not the
actual format that we will use for presentation. This breaks direct
scanout in cases where the application selected a visual with an alpha
channel but using EGL_EXT_present_opaque which previously worked as
expected.
Fix this by selecting the feedback tranche according to the actual
presentation format.
Fixes: 9ea9a963aa ("egl/wayland: Fix EGL_EXT_present_opaque")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28153>
When using the EGL_EXT_present_opaque extension we create wayland
buffers with the opaque variant of the surface format, while the
underlying image is created with the normal surface format. However,
there is no guarantee that the compositor supports that opaque format,
or that we can use that format with all the modifiers of the surface
format.
Since this is completely out of the control of the application, and the
compositor will disconnect the client with an error if the format is not
supported, this commit:
1. Fails EGL surface creation if it determines that the opaque format
cannot be used, either because it's not present at all, or because
it shares no modifier with the non-opaque format.
2. When creating the DRI image ensures that we use a modifier that's
also supported by the opaque format.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28153>
This introduces a new helper that will be used to generate a graphics
pipeline hash from a pCreateInfo struct only. Similar to RT.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30049>
