This would have created a 1.0.xx operand from 0x3c00.xx or 0x3c003c00.xy
for vop3p instructions which have 32-bit operands.
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/16296>
We were using a 16-bit inline constant with a 32-bit instruction and the
test would have created
"v1: %_:v[0] = v_alignbyte_b32 0.5, %_:v[1][16:32], 2" instead.
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/16296>
I don't think the presence of med3 here breaks anything, but it shouldn't
be here anyway.
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/16296>
If a shader ends with a workgroup barrier, it must wait for slot #7 at the end
to finish the barrier. After inserting flow control, we get:
BARRIER
NOP.wait
NOP.end
Currently, the flow control pass assumes that .end implies all other control
flow, and will merge this down to
BARRIER.end
However, this is incorrect. Slot #7 is no longer waited on. In theory, this
cannot affect the correctness of the shader. In practice, the hardware checks
that all barriers are reached. Terminating without waiting on slot #7 first
raises an INSTR_BARRIER_FAULT. We need to weaken the flow control merging
slightly to avoid this incorrect merge, instead emitting:
BARRIER.wait
NOP.end
Of course, all of these cases are inefficient: terminal barriers shouldn't be
emitted in the first place. I wrote out an optimization for this. We can merge
it if we find a workload that it actually helps.
Fixes test_half.vstore_half.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17264>
Mesa as a whole requires C++14 nowadays, so this isn't needed any more.
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17307>
Before, we needed CP post-sched to copy-propagate references to NIR
registers produced by out-of-ssa. Now that we're in SSA, this pass ends
up not doing anything useful, and actually gets in the way by occasionally
creating a cycle in the DAG.
The entire shader-db impact is:
instructions HURT: shaders/closed/steam/tropico-5/78.shader_test FRAG: 238 -> 242 (1.68%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17320>
Turns out this was likely a vkd3d-proton issue because it can no
longer be reproduced since it switched to dynamic rendering by default.
AMDGPU-PRO was also affected by the same issue at that time.
According to Hans-Kristian, some bugs related to that have also been
fixed at the same time.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17296>
In a mixed GFX9/GFX10 setup, this would crash for the GFX9 logical
device. Just print a message intead of aborting.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17292>
For unknown reasons, COPY_DATA can hang on GFX10+ while it doesn't
hang on GFX9. Adding PFP_SYNC_ME before/after the COPY_DATA doesn't
fix the hang either.
Using a LOAD_CONTEXT_REG_INDEX packet shouldn't be needed unless the
driver supports preemption (shadow memory) which RADV doesn't support.
I don't have a real explanation but PFP_SYNC_ME+LOAD_CONTEXT_REG_INDEX
fixes a GPU hang with Space Engineers (game uses a bunch of consecutive
calls to vkCmdDrawIndirectByteCountEXT without anything in-between).
Gitlab: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5838
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17290>
Here we move the nir_get_single_slot_attribs_mask() call that sets the
inputs_read mask after NIR optimisations have finished and after
st_nir_assign_vs_in_locations() has been called.
Besides fixing a bug where the mappings would be missaligned if
further NIR optimisations resulted in less inputs being read, it
also allows us to drop an additional nir gather info call.
Fixes: 0909a57b63 ("radeonsi/nir: Set vs_inputs_dual_locations and let NIR do the remap")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6240
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17326>
Avoid overflowing max pkt4 size by splitting VFD_FETCH[] emit.
Otherwise the maximum size of 32 VBOs would overflow and wrap to
zero.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17317>
While a resource might be shared across different contexts all synchronization
of commands is the resposibility of the user (OpenGL spec Chapter 5 "Shared
Objects and Multiple Contexts").
Currently etnaviv tries to be extremely helpful by flushing foreign contexts
when using a resource that is still pending there. This introduces a lot of
issues, as context flushes can now happen at basically any time and also
introduces a lot of overhead due to the needed tracking and locking. Get rid
of all this cross-context tracking and flushing.
The only real requirement here is that we need to track pending resources
without mutating the state of the etna_resource, as this might be shared
across multiple contexts and thus be used by multiple threads concurrently.
Introduce a hash table to track the current pending resources and their
states in the local context.
A side-effect of this change is that we don't need to keep all used
resources referenced until context flush time, as we don't need to mutate
them anymore to track the status. This allows to free some of them a bit
earlier. Note that this introduces a small possibility of a new resource
matching the key of a already destroyed resource still in the
pending_resources hashtable, but even if we get such a collision the worst
outcome is a not strictly necessary flush of the local context being
performed, which is acceptable if it doesn't happen very often.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14466>
Don't access the status member of etna_resource directly, as this
go away to get rid of shared mutable state. Add a wrapper function
that allows to plug in the new status lookup.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14466>
This lock is used to serialize much more than just the lookup tables. In
fact it's used for all device global state including the bo cache. Rename
it to better reflect its real usage.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14466>
This might be called from multiple threads at the same time. To avoid
taking a global lock just to guard against the fairly low chance of
multiple threads calling this on the same BO at the same time, we allow
for the threads to race. All threads will set up a mapping, but only
the first thread is able to set the map member of the etna_bo, all other
threads just roll back and use the mapping set up by the winning thread.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14466>
The mmap offset is the only information we currently get from
DRM_ETNAVIV_GEM_INFO and there is no point in storing this
offset after the mapping has been established. Reduce the
shared mutable state on the etna_bo by inlining fetching the
offset into etna_bo_map.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14466>
Currently the buffer index hash is only used if the BO is used in
multiple streams and the current index is cached on the BO. This
introduces some shared state on the BO, which necessitates the use
of a lock to keep this state consistent across threads, which
negates some of the benefits of caching the index.
Always use the hash to keep track of the submit BOs, to get rid
of the shared state and simplify the code.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14466>
Etnaviv has no restrictions on buffers being mapped during execution. In
fact most buffers are already always mapped during their lifetime as the
unmap is a no-op. Let the frontend know that it doesn't need to bother
with unmapping buffers.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17343>
As we do not suballocate any buffers, the real map buffer alignment
is determined by the GEM BO map alignment, which is at least 4KB.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17343>
The job will still fail, but it will at least go through until the end.
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16841>
This will be used to detect soft resets without aborting the run, but
still make the job fail.
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16841>
Copied and adjusted from the ACO lists.
v2: Martin Roukala
- add an extra test in the list of timeouts
- add an extra test in the list of flakes
- remove a fail
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Martin Roukala <martin.roukala@mupuf.org>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16841>
The LLVM backend is not officially supported by the RADV developers,
but it has been useful early during bring-up, or later when users are
experiencing what looks like a compiler bug. It is thus beneficial to
keep it working.
However, maintaining the vkcts expectations for every platform requires
more work and machine time than what we would like to commit to. This
is why we agreed that we would only keep LLVM tested on the latest
family of Radeon GPUs.
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16841>
Passes following piglit:
- spec@khr_parallel_shader_compile@basic
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16172>
Most CI tests are currently running on LLVM 11 (released over 2 years
ago), which predates some of the GPUs we have in CI and prevents
testing RADV's LLVM backend.
LLVM 13 is known to work for RADV, released almost 8 months ago, and
is already available in most distributions. Fedora 36 is even already
on LLVM 14.
So this commit updates x86 testing on llvm 13.
v2:
- store the llvm apt repo key locally (Michel Dänzer)
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17188>
It isn't compiled to an output param, so can be safely ignored
from the param assignment.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17244>
Set all driver locations to zero.
These are ignored by ac_nir_lower_ngg anyway.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17244>
It seems like the hangs have been resolved on kabini, so let's allow
developers to run their tests on kabini, if they are *very* patient
(~10h).
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17129>
This avoids using a VGPR and uses two SGPRs
instead since we only need to store 2 bits.
Quake II RTX:
Totals from 7 (0.46% of 1513) affected shaders:
CodeSize: 229364 -> 229148 (-0.09%); split: -0.12%, +0.02%
Instrs: 41937 -> 41879 (-0.14%)
Latency: 977374 -> 976723 (-0.07%)
InvThroughput: 651582 -> 651148 (-0.07%)
Copies: 5064 -> 5033 (-0.61%)
PreSGPRs: 430 -> 433 (+0.70%)
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17293>
Rhys Perry
Alyssa Rosenzweig
M Henning
Samuel Pitoiset
Emma Anholt
David Heidelberg
Timothy Arceri
Rob Clark
Jonathan Marek
Lucas Stach
Martin Roukala (né Peres)
Jesse Natalie
Christian Gmeiner
Timur Kristóf
Yiwei Zhang
Konstantin Seurer