On Valhall, register reads may be marked as "last" [1]. Setting the last flag
promises the hardware that the value of the register is no longer required. This
may enable hardware optimizations. In particular, it may permit the hardware to
avoid register file writes if a write to the marked register is still in the
forwarding buffer. This may improve power efficiency.
In principle, this is trivial: run liveness analysis and mark killed sources,
like we would in an SSA-based register allocator. In practice, there are a few
wrinkles to avoid hazards around staging registers and 64-bit register pairs,
requiring some additional data flow analysis and fix ups. However, nothing here
is particularly "hard", and all the ideas are already in use for the Bifrost
scheduler and the Bifrost/Valhall scoreboard analyses.
[1] In Mesa's compiler, this is called discard for historical reasons.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17091>
This helps "contain the crazy" and avoids special casing BLEND in compiler
passes. The Valhall instruction is roughly the same as its Bifrost counterpart,
as long as we fix up the source order (as we already do for bitwise operations)
everything works out.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17091>
Post-RA liveness relies on the caller updating the live variable with the
results of bi_postra_liveness_ins. It is not automatic, as with regular
liveness. This means ignoring the result of bi_postra_liveness_ins is surely an
error. Mark it as MUST_CHECK to catch that error at compile time.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17091>
Add a unit test for the quirk discovered in the previos commit, because this
will cause flakes (instead of fails) if we get it wrong. Better have a
deterministic fail mode.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17091>
For some unknown reason, waiting for general slots (at least for memory stores)
doesn't work properly on a BARRIER instruction. We need to wait for all general
slots right before issuing the BARRIER in addition to the general wait on the
BARRIER itself. I don't know if this is a hardware bug or some hideous
gate-saving quirk, but I observe the Mali-G78 DDK using the same workaround,
which implies this really is necessary.
Fixes rare flakes in:
dEQP-GLES31.functional.compute.shared_var.work_group_size.float_128_1_1
Note that the flakes from that test are extremely timing dependent. Without this
change, that test is racy but we almost always win the race. Reproducing the
issue reliably requires high system load (e.g. running the CTS in the
background) and simultaneously running that test a large number of times.
Minimal shader-db impact. In particular, no cycle count regressions.
total instructions in shared programs: 2699419 -> 2699458 (<.01%)
instructions in affected programs: 22014 -> 22053 (0.18%)
helped: 2
HURT: 25
helped stats (abs) min: 1.0 max: 1.0 x̄: 1.00 x̃: 1
helped stats (rel) min: 0.12% max: 0.12% x̄: 0.12% x̃: 0.12%
HURT stats (abs) min: 1.0 max: 3.0 x̄: 1.64 x̃: 1
HURT stats (rel) min: 0.07% max: 2.82% x̄: 0.69% x̃: 0.49%
95% mean confidence interval for instructions value: 1.01 1.87
95% mean confidence interval for instructions %-change: 0.38% 0.88%
Instructions are HURT.
total cvt in shared programs: 14468.81 -> 14469.42 (<.01%)
cvt in affected programs: 221.33 -> 221.94 (0.28%)
helped: 2
HURT: 25
helped stats (abs) min: 0.015625 max: 0.015625 x̄: 0.02 x̃: 0
helped stats (rel) min: 0.18% max: 0.18% x̄: 0.18% x̃: 0.18%
HURT stats (abs) min: 0.015625 max: 0.046875 x̄: 0.03 x̃: 0
HURT stats (rel) min: 0.10% max: 4.44% x̄: 1.06% x̃: 0.79%
95% mean confidence interval for cvt value: 0.02 0.03
95% mean confidence interval for cvt %-change: 0.57% 1.36%
Cvt are HURT.
total quadwords in shared programs: 1462496 -> 1462528 (<.01%)
quadwords in affected programs: 4632 -> 4664 (0.69%)
helped: 0
HURT: 4
HURT stats (abs) min: 8.0 max: 8.0 x̄: 8.00 x̃: 8
HURT stats (rel) min: 0.35% max: 7.69% x̄: 4.03% x̃: 4.03%
95% mean confidence interval for quadwords value: 8.00 8.00
95% mean confidence interval for quadwords %-change: -2.71% 10.76%
Inconclusive result (%-change mean confidence interval includes 0).
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17091>
Test cases for insert flow are necessarily the reference test cases with the
NOPs stripped out. That means we don't need to duplicate the test bodies.
Deduplicate.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17091>
This definition is a hardware property. It's not specific to the flow control
insertion pass, so move it to common code where other passes can use it.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17091>
This should help with "marge got stuck for an hour and all I got was this
failed job with no results/" when a system intermittently wedges.
This replaces the BM_POE_TIMEOUT ("did we get something on serial in the
last 3 minutes?") that rpi had, in favor of checking that the whole test
job gets through in 20 minutes.
Acked-by: Juan A. Suarez <jasuarez@igalia.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17096>
My local trogdor has a netboot firmware and I want to be able to use it to
test the timeout code I'm working on.
Acked-by: Juan A. Suarez <jasuarez@igalia.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17096>
If the SerialBuffers can just feed the same line queue, then we don't need
the extra threads reading line queues into a new merged line queue.
Less python threading code is always better. Plus, now we can pass args
to SerialBuffer.lines() for timeout/phase.
Acked-by: Juan A. Suarez <jasuarez@igalia.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17096>
This should avoid the 1-hour timeouts if something goes wrong, and just
restart.
Fixes: #6682
Acked-by: Juan A. Suarez <jasuarez@igalia.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17096>
My editor likes to pep8 as I edit, and I'm tired of carefully not
committing those changes.
Acked-by: Juan A. Suarez <jasuarez@igalia.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17096>
Every few weeks we'd get a flake where the script noticed the devcore
right as we were wrapping up, and then tar would exit because the file it
was tarring changed underneath it. Just kill devcores before we do that
-- even if we kill it while it's working, it's so rare that it probably
won't bother anyone.
Fixes: #5867
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16820>
If you set a name of on a swapchain object, because the base object
struct has not been initialized with a VkDevice,
vk_object_base_finish() will segfault when trying to free the object
name.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: cb1e0db23e ("vulkan/wsi: Make wsi_swapchain inherit from vk_object_base")
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17165>
In the following case :
vkCmdBindPipeline(compute_pipeline);
vkCmdDispatch(...);
vkCmdBindPipeline(graphics_pipeline);
vkCmdBindIndexBuffer(buffer)
vkCmdDraw(...);
We're emitting the 3DSTATE_INDEX_BUFFER instruction while the HW is
still in GPGPU mode, because we're dealing the pipeline selection to
vkCmdDraw().
Found while debugging Age Of Empire 4, HW is hung on
3DSTATE_INDEX_BUFFER instruction.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17153>
Enabling depth clamping ensures that the Vulkan driver respects
the depth range that zink sets on viewport objects in zink_draw.
When depth clipping is required, use VK_EXT_depth_clip_enable to
enable that independently of depth clamping.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16929>
Fix the transform to make sure it doesn't disturb the depth range
of the blitted image. Set the Z coordinates of the vertices
by hand instead of relying on the transform to do it.
This is a pre-requisite to Zink always enabling depth clamping.
Fixes: 26c6640835
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16929>
It seems recent versions of Sphinx has started issuing a warning when
language isn't set to a recognized value. Since we're building with
warnings as errors on CI, this causes issues for people.
We're writing English docs, so we should set this to English. I probably
confused this setting with the highlight_language setting or something
when I set this up initially.
Fixes: 00cd1346bf ("docs: use sphinx")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6711
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17159>
Per Vulkan spec only "Draw" commands should be counted towards
occlusion query.
Apparently RB_SAMPLE_COUNT_CONTROL::UNK0 bool controls whether
sample counting is enabled, so we could use it to disable
sample counting for 3d blits which are sometimes used for
clear/copy/blit/gmem-store/resolve operations.
Fixes GL CTS tests running through Zink:
dEQP-GLES3.functional.occlusion_query.depth_clear
dEQP-GLES3.functional.occlusion_query.depth_clear_stencil_clear
dEQP-GLES3.functional.occlusion_query.scissor_depth_clear_stencil_clear
dEQP-GLES3.functional.occlusion_query.scissor_stencil_clear
dEQP-GLES3.functional.occlusion_query.stencil_clear
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6559
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17138>
The test was misidentified as a "fail", when it actually has a very
high failure rate (>95%). The test happened to pass twice within the
last week, so let's mark this test as a flake.
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17131>
Texture2DMSArray element is not initialized, making 2DMSArray DS
views buggy.
Reported-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17158>
On a single core CPU system we end with num_threads being 0 and we do
_NOT_ create any async compile threads in util_queue_init(..).
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17124>
initial_variants_synchronous(..) allows initial shader variant
compilation asynchronously if debug.debug_message == NULL. If the
debug callback gets changed it seems like a good idea to wait until
all async shader compiles are done.
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17124>
Fix defect reported by Coverity Scan.
Side effect in assertion (ASSERT_SIDE_EFFECT)
assignment_where_comparison_intended: Assignment var->type =
glsl_int_type() has a side effect. This code will work differently in a
non-debug build.
Fixes: afb64e10c1 ("microsoft/compiler: Move d3d12_fix_io_uint_type() to dxil_nir.c")
Suggested-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17097>
Fix defect reported by Coverity Scan.
Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking subpass suggests that it may be null,
but it has already been dereferenced on all paths leading to the check.
Fixes: 779e09639b ("radv: configure DB_Z_INFO.NUM_SAMPLES correctly on GFX11")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17100>
r300 and r400 have strict rules with swizzles, so we
will need to convert swizzle back.
Operating on 0, 1, H in this case unnecessarily makes
rest of r300 overly complicated.
(also it's not currently able to handle this)
helps with:
deqp-gles2@functional@shaders@random@exponential@fragment@24
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17117>
Starting with Valhall, the provoking vertex state is specified per-framebuffer
(batch) instead of per-draw. We use the pan_tristate infrastructure to translate
between desktop OpenGL's per-draw semantics to Valhall's per-framebuffer
semantic. This is notably not required for GLES or Vulkan.
If the provoking vertex is unset when the tiler context is generated, it could
be set (incompatibly) later in the batch, and the tiler context's provoking
vertex field would no longer match the framebuffer's. That would violate a
hardware invariant. To ensure that doesn't happen, we make sure to set provoking
vertexes *before* generating the tiler context so it can't change after.
Fixes arb-provoking-vertex-render on Valhall.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17068>
This is useful to ensure that the tri-state has been set. It could be inlined,
but the tri-state structure is designed to be opaque so this matches better.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17068>
If we need to change batch state (currently just point coord origins), not only
do we need to flush the old batch, but also set the desired state on the new
batch. That second step was missing. Fix that so this mechanism works as
intended.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Fixes: 3641dfe436 ("panfrost: Flip point coords in hardware")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17068>
Even with the fixes in the rest of the series, arb-provoking-vertex-render is
still failing on Valhall for a single subcase (involving QUADS). It seems likely
that QUADS support is broken on Valhall, given it's not used in any of the APIs
for which Arm ships drivers.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17068>
This commit extends the graphics hard coding infrastructure to
allow the independent hard coding of stages, i.e. hard code fragment
stage and vertex stage separately instead of having to hard code
everything.
It also extends the infrastructure to allow per device hard coding.
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Rajnesh Kanwal <rajnesh.kanwal@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17106>
This commit adds 33.15.11.3 in the supported BVNCs and addresses
differences in the descriptor limits between devices so that the
physical device limit is accurate.
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Rajnesh Kanwal <rajnesh.kanwal@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17106>
Alyssa Rosenzweig
Emma Anholt
Danylo Piliaiev
Lionel Landwerlin
Timur Kristóf
Sarah Walker
Erik Faye-Lund
Gert Wollny
Martin Roukala (né Peres)
Boris Brezillon
Samuel Pitoiset
Christian Gmeiner
Vinson Lee
David Heidelberg
Filip Gawin
Marcin Ślusarz
Karmjit Mahil
Frank Binns
Juan A. Suarez Romero