Part-of: <!5950>

Part-of: <!5950>

This should be much better than the previous method that was more
guesswork-based than anything else. It returns a value within 1 of the
blob for every input value I've tested, and it seems like it returns
slightly better (but still legal) answers when it differs.

Part-of: <!5950>

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <!5995>

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <!5995>

For implementing panfrost_flush, it suffices to wait on only a single
syncobj, not an entire array of them. This lets us wait on it directly,
without coercing to/from syncfds in the middle (although some complexity
may be added later to support Android winsys).

Further, we should let the fence own the syncobj, tying together the
lifetimes and thus removing the connection between syncobjs and
batch_fence.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <!5995>

With the current kernel UABI, there is no benefit to explicitly
specifiying dependencies, since the kernel by design adds implicit
dependencies to any referenced BOs. This is something we'd like to
address in the future, but efficient handling with future kernels will
require a tweaked design in userspace as well. So let's do the obvious
thing now, and extend later.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <!5995>

It is always false now.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <!5995>

What is intended is to flush the batches and wait on a particular BO at
a later time. Explicitly forcing a wait immediately is redundant.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <!5995>

Fixes lots of tests under dEQP-VK.spirv_assembly.type.*

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <!5993>

Fixes: random KHR-GL45.sample_variables.mask.* fails
Fixes: 66ed9792

 ("nv50: Clear nv50_ir_prog_info of dead and codegen specific variables")
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <!6001>

Fixes: 66ed9792

 ("nv50: Clear nv50_ir_prog_info of dead and codegen specific variables")
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <!6001>

AMDGPU_GEM_CREATE_CPU_GTT_USWC should be faster when CPU reads
are unexpected (because they aren't cached).

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <!5978>

To disable CPU caching.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <!5978>

Detroit: Become Human uses dFdx/dFdy immediately after a quad-divergent
discard, which can cause the image to become white.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Cc: <mesa-stable@lists.freedesktop.org>
Closes: #3212
Part-of: <!5991>

bounds_ctrl is set to true by default which works around some game bugs,
but that isn't enough on GFX10.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <!5991>

We don't want these files shared between builds (it'll get blown away by
the next rsync), and NFS will just increase our latency for hitting the
cache.

Drops a630 gles31 run from 11-17 minutes to 5.5.  Maximum cache size on a
run I've seen is 153M, which it seems we can easily spare.

Fixes: f97acb4b

 ("freedreno/ir3: disk-cache support")
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <!5998>

They were still depending on arm_build, but the build of kernel and
rootfs has been moved to a separate job.

Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-By: Rohan Garg <rohan.garg@collabora.com>
Part-of: <!5472>

Upload failed images and the results.yml file to MinIO, to facilitate
debugging.

Also, fix version checking when git is installed as Mesa is going to
output a different renderer string if git is installed.

Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-By: Rohan Garg <rohan.garg@collabora.com>
Part-of: <!5472>

Downloading the traces directly from git causes very high egress from
GCE, which is expensive.

So we can expand trace testing further, we are going to keep a cache in
freedesktop.org's MinIO instance. This commit implements downloading
from it.

Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-By: Rohan Garg <rohan.garg@collabora.com>
Part-of: <!5472>

Submit lava jobs to replay traces on Veyron (Mali T760) and Kevin (Mali
T860) boards.

Signed-off-by: Rohan Garg <rohan.garg@collabora.com>
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-By: Rohan Garg <rohan.garg@collabora.com>
Part-of: <!5472>

The intention here was to check "Would the GPU be able to compress
this if we used the PBO-based texture upload path?"  Prior to Gen12,
that meant checking for CCS_E.  On Gen12, there are a lot more types
of compression, and basic CCS_E was replaced by GEN12_CCS_E, making
this check simply not work, so we'd take the CPU path instead.

Instead, check if it has CCS, and isn't the basic "fast clear" CCS_D.

Fixes: 39f06e28

 ("iris: Implement pipe->texture_subdata directly")
Tested-by: Mark Janes <mark.a.janes@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <!6005>

Fixes an issue with Renderdoc's shader debugging with ACO.

If nir_opt_algebraic isn't called in-between nir_lower_explicit_io and
nir_lower_int64, we can end up with 64-bit multiplications.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Fixes: 6320e37d ('nir: add amul instruction')
Part-of: <!5709>

We already have all of the shader code for load/store/exchange.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <!5992>

Fixes: e644ed46

 "intel/fs: Implement nir_intrinsic_global_atomic_*"
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <!5992>

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <!5992>

This pulls in commit 63cb1fc131573fa from KhronosGroup/SPIRV-Headers

Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <!5992>

_EGLDriver was an empty wrapper around _EGLAPI, so let's only keep one
of them. "driver" represents better what's being accessed, so that's the
one we're keeping.

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <!5987>

Otherwise most counters return 0. Should be much more user friendly
than having to totally disable clock-gating on the kernel cmdline.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <!5972>

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <!5972>

We already support a superset of VK_EXT_image_robustness via
VK_EXT_robustness2.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <!5985>

It triggers the disk cache for me, and asserts abount not getting the
build id right.

Fixes: f97acb4b ("freedreno/ir3: disk-cache support")
Part-of: <!5989>

All new dEQP-VK.robustness.image_robustness.* pass.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <!5979>

Closes: #2655


Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <!5661>

According to the "Errors" list of the OpenGL 4.6 spec, section 8.6
"Alternate Texture Image Specification Commands":

An INVALID_OPERATION error is generated by *TextureSubImage* if the
effective target of texture does not match the command, as shown in table 8.15.

Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <!5934>

I had a stp testcase that was getting its offset wrong, and by twiddling
bits and feeding it to qc disasm, I found that the comment was sort of
right: some the cat6a bits implicated in the old comment do get used, as
the high bits of the cat6c offset.  Reallocating those bits also fixes how
we were getting r960.y for r0.y.

Part-of: <!5815>

We didn't need the extra branch and temp, we can move it inside of the dst
handling by just duplicating the print of the dst reg.

Part-of: <!5815>

This started with making note of some ldp/stp instructions from the blob
and how we differ from them.  In the process of fixing it, I accidentally
modified behavior of other opcodes, and the other instructions listed will
keep us from doing that.  I also dropped an old stl test that looks like I
took from after a shader 'end' instruction.

Part-of: <!5815>

Part-of: <!5815>

Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <!5983>