platform/upstream/mesa.git
5 years agozink: use dynamic state for line-width
Erik Faye-Lund [Tue, 29 Oct 2019 11:43:56 +0000 (12:43 +0100)]
zink: use dynamic state for line-width

This will lead to fewer pipelines in the cache, which is assumed to
become our most unavoidable performance bottle-neck down the line.

Reviewed-by: Dave Airlie <airlied@redhat.com>
5 years agozink: Use optimal layout instead of general. Reduces valid layer warnings. Fixes...
Duncan Hopkins [Wed, 14 Aug 2019 10:07:47 +0000 (11:07 +0100)]
zink: Use optimal layout instead of general. Reduces valid layer warnings. Fixes RADV image noise.

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
5 years agogitlab-ci: Disable meson-windows job for the time being
Michel Dänzer [Wed, 30 Oct 2019 08:38:20 +0000 (09:38 +0100)]
gitlab-ci: Disable meson-windows job for the time being

It needs a CI runner carrying the mesa-windows tag, but there's none
available currently.

5 years agoradv: make use of radv_sc_read()
Timothy Arceri [Tue, 29 Oct 2019 06:46:57 +0000 (17:46 +1100)]
radv: make use of radv_sc_read()

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoradv: add radv_sc_read() helper
Timothy Arceri [Tue, 29 Oct 2019 06:43:40 +0000 (17:43 +1100)]
radv: add radv_sc_read() helper

This is a function with timeout support for reading from the pipe
between processes used for secure compile.

Initially we hardcode the timeout to 5 seconds. We can adjust the
timeout limit in future if needed.

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoradv: allow select() calls in secure compile
Timothy Arceri [Tue, 29 Oct 2019 06:41:41 +0000 (17:41 +1100)]
radv: allow select() calls in secure compile

This will be used in the following patch to support timeouts for
reading the pipe between processes.

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agomapi: Improve the x86 tsd stubs performance.
Lepton Wu [Wed, 30 Oct 2019 00:52:21 +0000 (17:52 -0700)]
mapi: Improve the x86 tsd stubs performance.

This skips touching %ebx most times and it shows that glGetString performance
increased from 114M/s to 120M/s on my desktop.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Lepton Wu <lepton@chromium.org>
5 years agomapi: Inline call x86_current_tls.
Lepton Wu [Tue, 22 Oct 2019 03:22:18 +0000 (20:22 -0700)]
mapi: Inline call x86_current_tls.

This saves one return and a simple benchmark which calls glGetString
repeatedly on my desktop shows it improves calls per second from 123M
to 141M.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/1997
Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Lepton Wu <lepton@chromium.org>
5 years agomapi: Clean up entry_patch_public for x86 tls
Lepton Wu [Sat, 26 Oct 2019 00:27:04 +0000 (17:27 -0700)]
mapi: Clean up entry_patch_public for x86 tls

Remove hard coded 16 and use entry_generate_or_patch to patch
public stubs. The generated code actually is sightly tighter
than before since the "nop" instructions before the final "jmp"
get removed.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Lepton Wu <lepton@chromium.org>
5 years agomapi: split entry_generate_or_patch for x86 tls
Lepton Wu [Fri, 25 Oct 2019 23:54:35 +0000 (16:54 -0700)]
mapi: split entry_generate_or_patch for x86 tls

The code works exactly the same with before. Just split this function
out so we can reuse it.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Lepton Wu <lepton@chromium.org>
5 years agomapi: Adapted libglvnd x86 tsd changes
Jonathan Gray [Fri, 13 Sep 2019 17:09:15 +0000 (10:09 -0700)]
mapi: Adapted libglvnd x86 tsd changes

The x86 assembly language stub in src/mapi/entry_x86_tsd.h does not
generate PIC (position-independent code). This causes text relocations
which bring troubles on recent versions of FreeBSD, OpenBSD, Android.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108541
Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Lepton Wu <lepton@chromium.org>
5 years agospirv: Don't fail if multiple ordering semantics bits are set
Caio Marcelo de Oliveira Filho [Tue, 29 Oct 2019 19:09:38 +0000 (12:09 -0700)]
spirv: Don't fail if multiple ordering semantics bits are set

Vulkan requires that only one bit for the ordering is set, but old
versions of GLSLang just set all the bits.  This was fixed as part of
https://github.com/KhronosGroup/glslang/commit/c51287d744fb6e7e9ccc09f6f8451e6c64b1dad6
but we can still find older versions (or shaders compiled with it)
around.

So instead of failing, emit a warning and fallback to the effective
result of any combination of multiple bits: AcquireRelease.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2018
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
5 years agointel/isl: Allow stencil buffer to support compression on Gen12+
Sagar Ghuge [Tue, 15 Oct 2019 21:13:29 +0000 (14:13 -0700)]
intel/isl: Allow stencil buffer to support compression on Gen12+

v2: (Nanley Chery)
- Fix commit title
- Fix comment

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agoiris: Resolve stencil resource prior to copy or used by CPU
Sagar Ghuge [Tue, 17 Sep 2019 20:20:16 +0000 (13:20 -0700)]
iris: Resolve stencil resource prior to copy or used by CPU

v2: Decide aux usage in get_copy_region_aux_settings (Nanley Chery)

v3: Use isl_surf_usage_is_stencil function (Nanley Chery)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agoiris: Prepare resources before stencil blit operation
Sagar Ghuge [Tue, 3 Sep 2019 23:30:14 +0000 (16:30 -0700)]
iris: Prepare resources before stencil blit operation

We have to resolve destination surfaces if we are bliting to and from
the same surface.

v2: Revert unrelated change (Nanley Chery)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agoiris: Prepare depth resource if clear_depth enable
Sagar Ghuge [Wed, 28 Aug 2019 07:21:20 +0000 (00:21 -0700)]
iris: Prepare depth resource if clear_depth enable

Avoid preparing depth resource, if we did fast depth clear before.

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agoiris: Prepare stencil resource before clear depth stencil
Sagar Ghuge [Wed, 14 Aug 2019 20:58:57 +0000 (13:58 -0700)]
iris: Prepare stencil resource before clear depth stencil

Let aux surface state tracker track the stencil buffer's aux state while
clearing depth stencil buffer.

v2: Fix condition check (Nanley Chery)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agoiris: Resolve stencil buffer lossless compression with WM_HZ_OP packet
Sagar Ghuge [Wed, 7 Aug 2019 20:42:39 +0000 (13:42 -0700)]
iris: Resolve stencil buffer lossless compression with WM_HZ_OP packet

Even though stencil buffer compression looks like regular lossless color
compression w/o fast clear support, we have to resolve stencil buffer
with WM_HZ_OP packet.

v2: Check if resource is stencil with helper function (Nanley Chery)

v3: Remove unnecessary included file (Nanley Chery)

v4: (Nanley Chery)
- Avoid stencil buffer aux state transition by improving condition check

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agointel/blorp: Set stencil resolve enable bit
Sagar Ghuge [Tue, 17 Sep 2019 18:04:15 +0000 (11:04 -0700)]
intel/blorp: Set stencil resolve enable bit

When set, the stencil buffer is filled with the true stencil values and
we have to disable stencil buffer clear enable bit.

v2: 1) Refactor code little bit (Nanley Chery)
    2) Fix assertion (Nanley Chery)

v3: 1) Remove unncessary assignment (Nanley Chery)
    2) Fix GEN_GEN check (Nanley Chery)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agointel: Track stencil aux usage on Gen12+
Sagar Ghuge [Wed, 23 Oct 2019 23:24:46 +0000 (16:24 -0700)]
intel: Track stencil aux usage on Gen12+

Enable stencil compression enable and control surface enable bit if
stencil buffer lossless compression is enabled.

v2: Remove unnecessary GEN_GEN check (Nanley Chery)

v3: (Nanley Chery)
- Change commit subject tag from intel/isl to intel
- Keep assignment order correct

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agointel/blorp: Add helper function for stencil buffer resolve
Sagar Ghuge [Tue, 15 Oct 2019 18:15:22 +0000 (11:15 -0700)]
intel/blorp: Add helper function for stencil buffer resolve

On Gen12+, Stencil buffer's lossless compression should be resolved
with WM_HZ_OP packet.

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agointel/blorp: Assign correct view while clearing depth stencil
Sagar Ghuge [Wed, 14 Aug 2019 20:58:33 +0000 (13:58 -0700)]
intel/blorp: Assign correct view while clearing depth stencil

We never saw any failures regarding this typo but it's good to assign
correct stencil view while constructing blorp_params.

Fixes: 0cabf93b80d0 "intel/blorp: Add an entrypoint for clearing depth and stencil"

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agogenxml/gen12: Add Stencil Buffer Resolve Enable bit
Sagar Ghuge [Wed, 23 Oct 2019 23:17:48 +0000 (16:17 -0700)]
genxml/gen12: Add Stencil Buffer Resolve Enable bit

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agoiris: Allocate main and aux surfaces together
Nanley Chery [Thu, 24 Oct 2019 16:14:07 +0000 (09:14 -0700)]
iris: Allocate main and aux surfaces together

On Gen12, the CCS buffer address doesn't have to be referenced in state
packets. In the case of a stencil buffer with CCS, the kernel won't know
the location of the CCS unless an extra call is made to pin its address.
To avoid this extra call, make the CCS part of the main surface.

v2. Update comment above bo_size. (Jordan)

Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoiris: Determine aux offsets within configure_aux
Nanley Chery [Fri, 25 Oct 2019 22:07:42 +0000 (15:07 -0700)]
iris: Determine aux offsets within configure_aux

If a resource has a modifier, the main and aux surfaces will share a BO.

Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoiris: Bail resource creation upon aux creation error
Nanley Chery [Fri, 25 Oct 2019 22:38:18 +0000 (15:38 -0700)]
iris: Bail resource creation upon aux creation error

The functions used during aux buffer configuration and creation only
return false for exceptional errors. Don't proceed with surface creation
in those cases.

Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoiris: Drop iris_resource::aux::extra_aux::bo
Nanley Chery [Fri, 25 Oct 2019 19:05:58 +0000 (12:05 -0700)]
iris: Drop iris_resource::aux::extra_aux::bo

The primary and secondary aux buffers are always allocated in the same
BO.

Suggested-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agozink: pass line width from rast_state to gfx_pipeline_state.
Duncan Hopkins [Tue, 24 Sep 2019 15:03:04 +0000 (16:03 +0100)]
zink: pass line width from rast_state to gfx_pipeline_state.

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
5 years agoanv: Reduce the minimum number of relocations
Jason Ekstrand [Mon, 28 Oct 2019 16:17:06 +0000 (11:17 -0500)]
anv: Reduce the minimum number of relocations

The original value of 256 was under the assumption that you're a batch
buffer which is likely going to have a large number of relocations.
However, pipeline objects on Gen7 will have at most 6 relocations (one
per shader stage and one for the workaround BO) so this is a lot of
per-pipeline wasted space.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
5 years agoanv: Delay allocation of relocation lists
Jason Ekstrand [Mon, 28 Oct 2019 15:22:47 +0000 (10:22 -0500)]
anv: Delay allocation of relocation lists

The old relocation list code always allocated 256 relocations and a hash
set up-front without knowing whether or not we really need them.  In
particular, in the softpin case, this is two fairly large allocations
that we don't need to be making.  Also, for pipeline objects on haswell
where we don't have softpin, we don't need relocations unless scratch is
used so this is extra data per-pipeline.  Instead, we should do it
on-demand.  This shaves 3.5% off of a cpu-limited example running with
the Dawn WebGPU implementation.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
5 years agoanv: Implement new way for setting streamout buffers.
Plamena Manolova [Wed, 23 Oct 2019 22:47:03 +0000 (23:47 +0100)]
anv: Implement new way for setting streamout buffers.

For gen12 we set the streamout buffers using 4 separate
commands instead of 3DSTATE_SO_BUFFER.

Signed-off-by: Plamena Manolova <plamena.manolova@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
5 years agoiris: Implement new way for setting streamout buffers.
Plamena Manolova [Wed, 23 Oct 2019 22:45:58 +0000 (23:45 +0100)]
iris: Implement new way for setting streamout buffers.

For gen12 we set the streamout buffers using 4 separate
commands instead of 3DSTATE_SO_BUFFER.

Signed-off-by: Plamena Manolova <plamena.manolova@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
5 years agogenxml: Add 3DSTATE_SO_BUFFER_INDEX_* instructions
Plamena Manolova [Thu, 17 Oct 2019 20:05:55 +0000 (21:05 +0100)]
genxml: Add 3DSTATE_SO_BUFFER_INDEX_* instructions

For gen12 we set the streamout buffers using 4 separate
commands instead of 3DSTATE_SO_BUFFER.

Signed-off-by: Plamena Manolova <plamena.manolova@intel.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
5 years agofreedreno/a6xx: add a618 support
Rob Clark [Thu, 24 Oct 2019 21:29:39 +0000 (14:29 -0700)]
freedreno/a6xx: add a618 support

Signed-off-by: Rob Clark <robdclark@chromium.org>
5 years agofreedreno/a6xx: cleanup magic registers
Rob Clark [Thu, 24 Oct 2019 21:03:32 +0000 (14:03 -0700)]
freedreno/a6xx: cleanup magic registers

Extract out values for the handful of unknown registers which have
different values across different a6xx models, to simplify adding
support for new a6xx's.

Signed-off-by: Rob Clark <robdclark@chromium.org>
5 years agofreedreno/a6xx: remove some left over dead code
Rob Clark [Thu, 24 Oct 2019 21:22:09 +0000 (14:22 -0700)]
freedreno/a6xx: remove some left over dead code

These registers don't exist, just remnants of initial port from a5xx.

Signed-off-by: Rob Clark <robdclark@chromium.org>
5 years agoanv: Set depthBounds to true in anv_GetPhysicalDeviceFeatures.
Plamena Manolova [Mon, 28 Oct 2019 23:47:39 +0000 (23:47 +0000)]
anv: Set depthBounds to true in anv_GetPhysicalDeviceFeatures.

Add depth bounds testing to the list of supported
physical device features.

Signed-off-by: Plamena Manolova <plamena.manolova@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agogenxml: Change 3DSTATE_DEPTH_BOUNDS bias.
Plamena Manolova [Mon, 28 Oct 2019 23:44:28 +0000 (23:44 +0000)]
genxml: Change 3DSTATE_DEPTH_BOUNDS bias.

The bias for the 3DSTATE_DEPTH_BOUNDS instruction
should be 2 not 1.

Signed-off-by: Plamena Manolova <plamena.manolova@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agogitlab-ci: Only run the pipeline if any files affecting it have changed
Michel Dänzer [Fri, 25 Oct 2019 16:59:56 +0000 (18:59 +0200)]
gitlab-ci: Only run the pipeline if any files affecting it have changed

E.g. documentation-only changes cannot affect the outcome of the
pipeline, so don't waste resources on running it.

The thing we need to be careful about here is that the container stage
jobs must always run if any later stage jobs using the corresponding
docker images run. We're currently using the same .ci-run-policy
template for all jobs, so this is trivially true.

v2:
* Add bin/ and common.py (Eric Engestrom)

Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com> # v1
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
5 years agogallium/swr: Enable GL_ARB_gpu_shader5: multiple streams
Krzysztof Raszkowski [Tue, 29 Oct 2019 14:50:02 +0000 (14:50 +0000)]
gallium/swr: Enable GL_ARB_gpu_shader5: multiple streams

Added support for geometry shader multiple streams (part of
GL_ARB_gpu_shader5 extension).

Reviewed-by: Jan Zielinski <jan.zielinski@intel.com>
5 years agopanfrost: Remove unused definitions in mali-job.h
Alyssa Rosenzweig [Sun, 27 Oct 2019 23:46:50 +0000 (19:46 -0400)]
panfrost: Remove unused definitions in mali-job.h

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Cleanup _shader_upper -> shader
Alyssa Rosenzweig [Sun, 27 Oct 2019 23:46:21 +0000 (19:46 -0400)]
panfrost: Cleanup _shader_upper -> shader

I don't believe this is actually a tagged pointer; warn if it is.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agomeson: define _GNU_SOURCE on FreeBSD
Eric Engestrom [Sat, 26 Oct 2019 21:43:50 +0000 (22:43 +0100)]
meson: define _GNU_SOURCE on FreeBSD

_mesa_strtod() needs this to use strtod_l(), which behaves correctly
wrt `,` vs `.` decimal separator.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2008
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
5 years agointel/perf: update ICL configurations
Lionel Landwerlin [Fri, 20 Sep 2019 18:12:13 +0000 (21:12 +0300)]
intel/perf: update ICL configurations

A few equations/programming changes for ICL.

v2: Fix a couple of issues in naming and floating/integer operations (Ken)

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agogitlab-ci: Update required libdrm version
Alexandros Frantzis [Tue, 29 Oct 2019 09:01:57 +0000 (11:01 +0200)]
gitlab-ci: Update required libdrm version

Commit 9edcce2a32ed bumped the required libdrm-amdgpu version to
2.4.100. Update the version we use in our CI scripts to avoid CI
build failures.

Also bump the debian image name for this change to take effect.
Note that amdgpu is only built with the debian-buster image,
so only this image requires an update.

Fixes: 9edcce2a ("ac: get tcc_harvested from the kernel")
Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
5 years agotravis: fix scons build after deprecation warning
Eric Engestrom [Tue, 29 Oct 2019 09:24:36 +0000 (09:24 +0000)]
travis: fix scons build after deprecation warning

Fixes: 54053bc8d0dad89a38e2 ("scons: Print a deprecation warning about using scons on not windows")
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
5 years agoanv: Fix output of INTEL_DEBUG=bat for chained batches
Caio Marcelo de Oliveira Filho [Mon, 28 Oct 2019 21:46:23 +0000 (14:46 -0700)]
anv: Fix output of INTEL_DEBUG=bat for chained batches

The anv_batch_bo contents are linked one to another, and when printing
we have to start with the first of those.  Since in `u_vector` new
elements are added to the head, to get the first element we need the
vector's tail.

Fixes: 32ffd90002b ("anv: add support for INTEL_DEBUG=bat")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
5 years agowinsys/amdgpu: use the new GPU reset query
Marek Olšák [Wed, 9 Oct 2019 23:32:42 +0000 (19:32 -0400)]
winsys/amdgpu: use the new GPU reset query

5 years agoac: get tcc_harvested from the kernel
Marek Olšák [Tue, 24 Sep 2019 21:55:52 +0000 (17:55 -0400)]
ac: get tcc_harvested from the kernel

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoradeonsi: initialize shader compilers in threads on demand
Marek Olšák [Sat, 26 Oct 2019 00:25:59 +0000 (20:25 -0400)]
radeonsi: initialize shader compilers in threads on demand

It takes a noticable amount of time with piglit.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
5 years agoradeonsi: don't print diagnostic LLVM remarks and notes
Marek Olšák [Thu, 24 Oct 2019 04:22:58 +0000 (00:22 -0400)]
radeonsi: don't print diagnostic LLVM remarks and notes

We don't use them.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
5 years agoaco: Introduce vgpr_limit to keep track of available VGPRs.
Timur Kristóf [Thu, 24 Oct 2019 15:34:37 +0000 (17:34 +0200)]
aco: Introduce vgpr_limit to keep track of available VGPRs.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
5 years agoaco: Implement subgroup shuffle in GFX10 wave64 mode.
Timur Kristóf [Sat, 21 Sep 2019 16:03:56 +0000 (18:03 +0200)]
aco: Implement subgroup shuffle in GFX10 wave64 mode.

Previously subgroup shuffle was implemented using the bpermute
instruction, which only works accross half-waves, so by itself it's
not suitable for implementing subgroup shuffle when the shader is
running in wave64 mode.

This commit adds a trick using shared VGPRs that allows to implement
subgroup shuffle still relatively effectively in this mode.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
5 years agoaco: Remove dead code in reduction lowering.
Rhys Perry [Thu, 12 Sep 2019 19:04:20 +0000 (20:04 +0100)]
aco: Remove dead code in reduction lowering.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
5 years agoaco: Fix reductions on GFX10.
Rhys Perry [Thu, 12 Sep 2019 18:28:52 +0000 (19:28 +0100)]
aco: Fix reductions on GFX10.

Fixes p_reduce (all cluster sizes), p_inclusive_scan and p_exclusive_scan
with all reduction operations.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
5 years agoloader: default to iris for all future PCI IDs
Eric Engestrom [Sat, 5 Oct 2019 21:30:51 +0000 (22:30 +0100)]
loader: default to iris for all future PCI IDs

The existing "fallback" code didn't actually do anything, so this
removes it, and instead we just always fallback to `iris` for future
PCI IDs.

Suggested-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoanv: add a couple printflike() annotations
Eric Engestrom [Thu, 24 Oct 2019 12:29:37 +0000 (13:29 +0100)]
anv: add a couple printflike() annotations

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
5 years agost/mesa: lower global vars to local after lowering clip
Erik Faye-Lund [Mon, 28 Oct 2019 13:02:02 +0000 (14:02 +0100)]
st/mesa: lower global vars to local after lowering clip

When this code was merged, this wasn't necessary because the
state-tracker would do it later anyway. But this recently got changed,
without changing the code that depended on this.

Arguably, this was a mistake in the lowering pass to begin with. Either
way, let's fix it by not assuming that the lowering code gets called
later when it's not needed.

This fixed user-defined clip-planes in Zink.

Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Fixes: eaffdad1082 ("st/mesa: don't lower_global_vars_to_local for VS if there are no dead inputs")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
5 years agoiris: Create resource with aux_usage MCS_CCS
Sagar Ghuge [Wed, 18 Sep 2019 20:14:31 +0000 (13:14 -0700)]
iris: Create resource with aux_usage MCS_CCS

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agointel/isl: Support lossless compression with multisamples
Sagar Ghuge [Wed, 18 Sep 2019 19:37:59 +0000 (12:37 -0700)]
intel/isl: Support lossless compression with multisamples

GEN12 adds the ability to losslessly compress each sample plane in a
multisampled buffer that uses MCS compression.

v2: Remove unnecessary assertion (Nanley Chery)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agoiris: Get correct resource aux usage for copy
Sagar Ghuge [Fri, 20 Sep 2019 21:05:58 +0000 (14:05 -0700)]
iris: Get correct resource aux usage for copy

Add case for MCS_CCS so that we get the correct aux usage while copy
operation.

v2: Fix commit subject (Nanley Chery)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agointel/blorp: Use isl_aux_usage_has_mcs instead of comparing
Sagar Ghuge [Thu, 10 Oct 2019 17:40:17 +0000 (10:40 -0700)]
intel/blorp: Use isl_aux_usage_has_mcs instead of comparing

Depending on MCS_CSS or MCS we can emit blorp blit shaders.

As we support MCS_CSS and MCS, it makes sense to use
isl_aux_usage_has_mcs function.

v2: Fix commit message (Nanley Chery)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agoiris: Define MCS_CCS state transitions and usages
Sagar Ghuge [Wed, 18 Sep 2019 20:15:47 +0000 (13:15 -0700)]
iris: Define MCS_CCS state transitions and usages

v2: 1) Fix assertion check (Nanley Chery)
    2) Correct commit subject (Nanley Chery)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agoiris: Initialize CCS to fast clear while using with MCS
Sagar Ghuge [Thu, 19 Sep 2019 15:13:15 +0000 (08:13 -0700)]
iris: Initialize CCS to fast clear while using with MCS

v2: Explain Bsepc quotes properly (Nanley Chery)

v3: 1) Fix comment format (Nanley Chery)
    2) Fix typo in comment (Nanley Chery)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agointel/isl: Don't reconfigure aux surfaces for MCS
Sagar Ghuge [Thu, 19 Sep 2019 15:20:34 +0000 (08:20 -0700)]
intel/isl: Don't reconfigure aux surfaces for MCS

If aux for MCS is already configured, don't configure again.

v2: Fix missing period in commit message (Nanley Chery)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
5 years agozink: emulate optional depth-formats
Erik Faye-Lund [Wed, 23 Oct 2019 09:59:03 +0000 (11:59 +0200)]
zink: emulate optional depth-formats

The Vulkan spec says that an implementation has to support one of
VK_FORMAT_X8_D24_UNORM_PACK32 and VK_FORMAT_D32_SFLOAT, as well of
one of VK_FORMAT_D24_UNORM_S8_UINT and VK_FORMAT_D32_SFLOAT_S8_UINT.

So let's keep track which one is supported of earch pair, and emulate
one on top of the other one.

This won't give the exact result for comparisons, or when mapping and
unmapping the resources. But it's better than flat out failing to create
the resource, and we can fix the map/unmap issue later if needed.

Tested-by: Duncan Hopkins <duncan@thefoundry.co.uk>
5 years agozink: error if VK_KHR_maintenance1 isn't supported
Erik Faye-Lund [Tue, 22 Oct 2019 13:29:55 +0000 (15:29 +0200)]
zink: error if VK_KHR_maintenance1 isn't supported

While we're at it, remove the VK_-prefix from the extension bool; all
extensions have this so it's kinda superfluous.

5 years agoiris: Disallow incomplete resource creation
Nanley Chery [Fri, 2 Aug 2019 22:38:36 +0000 (15:38 -0700)]
iris: Disallow incomplete resource creation

If a modifier specifies an aux, it must be created.

Fixes: 75a3947af46 ("iris/resource: Fall back to no aux if creation fails")
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoiris: Don't leak the resource for unsupported modifier
Nanley Chery [Wed, 25 Sep 2019 19:48:57 +0000 (12:48 -0700)]
iris: Don't leak the resource for unsupported modifier

Make sure the res struct is free'd before returning.

Fixes: 2dce0e94a3d ("iris: Initial commit of a new 'iris' driver for Intel Gen8+ GPUs.")
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoiris: Enable HIZ_CCS sampling
Nanley Chery [Wed, 21 Aug 2019 22:23:24 +0000 (15:23 -0700)]
iris: Enable HIZ_CCS sampling

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel/blorp: Satisfy clear color rules for HIZ_CCS
Nanley Chery [Wed, 18 Sep 2019 16:44:02 +0000 (09:44 -0700)]
intel/blorp: Satisfy clear color rules for HIZ_CCS

Store the converted depth value into two dwords. Avoids regressing the
piglit test "fbo-depth-array depth-clear", when HIZ_CCS sampling is
enabled in a later commit.

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel: Fix and use HIZ_CCS write through mode
Nanley Chery [Wed, 21 Aug 2019 17:57:29 +0000 (10:57 -0700)]
intel: Fix and use HIZ_CCS write through mode

Write through to the CCS if the surface is used as a texture and can be
sampled by the HW with CCS.

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoiris: Start using blorp_can_hiz_clear_depth()
Nanley Chery [Mon, 7 Oct 2019 22:52:21 +0000 (15:52 -0700)]
iris: Start using blorp_can_hiz_clear_depth()

Check that the alignment requirements for HIZ_CCS are satisfied by using
this function.

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel/blorp: Satisfy HIZ_CCS fast-clear alignments
Nanley Chery [Mon, 7 Oct 2019 22:53:44 +0000 (15:53 -0700)]
intel/blorp: Satisfy HIZ_CCS fast-clear alignments

Prevent the piglit test,
amd_vertex_shader_layer-layered-depth-texture-render, from regressing in
in a future commit.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel: Refactor blorp_can_hiz_clear_depth()
Nanley Chery [Mon, 7 Oct 2019 22:48:33 +0000 (15:48 -0700)]
intel: Refactor blorp_can_hiz_clear_depth()

Prepare this function to be used in iris and to handle new Gen12 behavior.

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoisl: Add isl_surf_supports_hiz_ccs_wt()
Nanley Chery [Thu, 19 Sep 2019 20:10:24 +0000 (13:10 -0700)]
isl: Add isl_surf_supports_hiz_ccs_wt()

Add a helper to determine if an ISL surface supports the write-through
mode of HIZ_CCS.

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoiris: Enable HIZ_CCS in depth buffer instructions
Nanley Chery [Fri, 9 Aug 2019 18:08:26 +0000 (11:08 -0700)]
iris: Enable HIZ_CCS in depth buffer instructions

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoiris: Define initial HIZ_CCS state and transitions
Nanley Chery [Fri, 9 Aug 2019 00:39:47 +0000 (17:39 -0700)]
iris: Define initial HIZ_CCS state and transitions

Make it match those of HIZ.

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoiris: Create an unusable secondary aux surface
Nanley Chery [Wed, 7 Aug 2019 23:02:51 +0000 (16:02 -0700)]
iris: Create an unusable secondary aux surface

The HIZ_CCS and MCS_CCS auxiliary surface modes require that drivers
store information about two aux buffers. We choose to represent this as
HiZ/MCS being the primary aux surface and the CCS as an secondary/extra
aux surface. This representation has the effect of placing most of the
code that will have to choose between the two aux surfaces around the
aux-map entry points.

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoiris: Don't guess the aux_usage
Nanley Chery [Thu, 1 Aug 2019 23:49:57 +0000 (16:49 -0700)]
iris: Don't guess the aux_usage

Instead of guessing an aux_usage, then confirming it if the
isl_surf_get_*_surf functions are successful, just call the ISL
functions up-front. This will help us to more easily determine if a
depth buffer supports HIZ_CCS.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel/blorp: Treat HIZ_CCS like HiZ
Nanley Chery [Fri, 9 Aug 2019 17:02:50 +0000 (10:02 -0700)]
intel/blorp: Treat HIZ_CCS like HiZ

Allow it in depth buffer instructions but disable it for blits.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel/blorp: Assert against HiZ in surface states
Nanley Chery [Wed, 21 Aug 2019 23:43:26 +0000 (16:43 -0700)]
intel/blorp: Assert against HiZ in surface states

Avoid unexpected behavior if the caller happens to pass in a HiZ aux
usage.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel: Support HIZ_CCS in isl_surf_get_ccs_surf
Nanley Chery [Mon, 19 Aug 2019 16:17:26 +0000 (09:17 -0700)]
intel: Support HIZ_CCS in isl_surf_get_ccs_surf

Add an extra aux parameter which will be filled out with CCS if the
first two isl_surf parameters fit the requirements for HiZ_CCS.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoisl: Reduce assertions during aux surf creation
Nanley Chery [Wed, 31 Jul 2019 21:38:29 +0000 (14:38 -0700)]
isl: Reduce assertions during aux surf creation

Return false more often to reduce the burden on the caller.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel: Enable CCS_E for R24_UNORM_X8_TYPELESS on TGL+
Nanley Chery [Thu, 8 Aug 2019 20:40:08 +0000 (13:40 -0700)]
intel: Enable CCS_E for R24_UNORM_X8_TYPELESS on TGL+

While this format isn't listed in BSpec: 53911, other documentation and
empirical evidence suggest that it's fine to remap it to R32_FLOAT. I've
filed a bug for the BSpec page.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel: Use 3DSTATE_DEPTH_BUFFER::ControlSurfaceEnable
Nanley Chery [Sat, 10 Aug 2019 00:18:48 +0000 (17:18 -0700)]
intel: Use 3DSTATE_DEPTH_BUFFER::ControlSurfaceEnable

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel/isl: Support HIZ_CCS in emit_depth_stencil_hiz
Jason Ekstrand [Fri, 4 May 2018 16:43:42 +0000 (09:43 -0700)]
intel/isl: Support HIZ_CCS in emit_depth_stencil_hiz

v2. Remove undocumented CCS_E-only mode for depth. (Nanley)

Co-authored-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel: Use RENDER_SURFACE_STATE::DepthStencilResource
Nanley Chery [Sat, 10 Aug 2019 01:04:58 +0000 (18:04 -0700)]
intel: Use RENDER_SURFACE_STATE::DepthStencilResource

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel: Update alignment restrictions for HiZ surfaces.
Jordan Justen [Wed, 30 May 2018 00:10:47 +0000 (17:10 -0700)]
intel: Update alignment restrictions for HiZ surfaces.

v2 (Nanley):
* Maintain a chronological ordering for HiZ alignments. Suggested by
  Ken.

Co-authored-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoiris: Clear ::has_hiz when disabling aux
Nanley Chery [Fri, 30 Aug 2019 21:58:54 +0000 (14:58 -0700)]
iris: Clear ::has_hiz when disabling aux

Fixes: 2cddc953cd0 ("iris: some initial HiZ bits")
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel/blorp: Disable depth testing for slow depth clears
Nanley Chery [Thu, 15 Aug 2019 17:17:11 +0000 (10:17 -0700)]
intel/blorp: Disable depth testing for slow depth clears

We'll start doing slow depth clears more often on HIZ_CCS buffers in a
future commit. Reduce the performance impact by making them use less
bandwidth.

From the Depth Test section of the BSpec:

   This function is enabled by the Depth Test Enable state variable. If
   enabled, the pixel's ("source") depth value is first computed. After
   computation the pixel's depth value is clamped to the range defined
   by Minimum Depth and Maximum Depth in the selected CC_VIEWPORT state.
   Then the current ("destination") depth buffer value for this pixel is
   read.

and from the Depth Buffer Updates section of the BSpec:

   If depth testing is disabled or the depth test passed, the incoming
   pixel's depth value is written to the Depth Buffer.

Taken together, it's clear that depth testing isn't necessary to perform
a depth buffer clear. Mark Janes and I analyzed this patch with
frameretrace and a depthrange piglit test. I disabled HiZ to ensure we'd
get slow depth clears. We've observed the bandwidth consumption by the
depth buffer access to be cut ~50% on BDW and SKL during depth clears.
On a more graphically intensive workload, the Shadowmapping Sascha
benchmark, I took the average of 3 runs on a BDW with a display
resolution of about 1920x1200 (minus some desktop environment
decorations). I measured a 22.61% FPS improvement when HiZ is disabled.

v2. The BSpec doesn't mandate this behavior, update comment accordingly.
    (Ken)

Fixes: bc4bb5a7e30 ("intel/blorp: Emit more complete DEPTH_STENCIL state")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agointel: Enable CCS_E for some formats on Gen12
Nanley Chery [Mon, 25 Mar 2019 21:15:01 +0000 (14:15 -0700)]
intel: Enable CCS_E for some formats on Gen12

In ISL:
   Update the format table to add CCS_E support for some 8BPP formats,
   some 16BPP formats, and R10G10B10A2_UNORM_SRGB.

   In the helper for determining CCS_E support, we return false for some
   16BPP formats because they aren't properly handled in blorp_copy().

In BLORP:
   Allow the new and non-problematic formats for CCS_E-enabled copies.

v2. Update other fields for A1B5G5R5_UNORM and A4B4G4R4_UNORM in table.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com> (v1)
5 years agoisl: Redefine the CCS layout for Gen12
Nanley Chery [Wed, 20 Mar 2019 01:23:46 +0000 (18:23 -0700)]
isl: Redefine the CCS layout for Gen12

The CCS could be described in a number of ways, but this format was
chosen to minimize churn in the drivers. We may decide on an different
direction in the future.

v2. Increase alignment for display surfaces. (Nanley)

Reviewed-by: Jordan Justen <jordan.l.justen@intel.com> (v1)
Acked-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoisl: Add and use isl_tiling_flag_to_enum()
Nanley Chery [Mon, 14 Jan 2019 19:32:21 +0000 (11:32 -0800)]
isl: Add and use isl_tiling_flag_to_enum()

Use a helper that will automatically handle Gen12's CCS tiling when
creating a CCS isl_surf.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoiris: Allow for non-Y-tiled aux allocation
Nanley Chery [Mon, 12 Aug 2019 22:41:11 +0000 (15:41 -0700)]
iris: Allow for non-Y-tiled aux allocation

The Gen12 CCS is not Y-tiled.

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoisl/drm: Map HiZ and CCS tilings to Y
Nanley Chery [Wed, 27 Mar 2019 21:40:58 +0000 (14:40 -0700)]
isl/drm: Map HiZ and CCS tilings to Y

In the function which translates ISL tilings to i915 tilings, map ISL's
HiZ and CCS tilings to Y instead of NONE (linear). The HW docs describe
HiZ and pre-Gen12 CCS surfaces as being Y-tiled in memory.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agointel/isl: Update surf_fill_state for gen12
Jason Ekstrand [Fri, 4 May 2018 16:44:24 +0000 (09:44 -0700)]
intel/isl: Update surf_fill_state for gen12

v2 (Nanley):
* Avoid driver churn for now.
* Include some media compression changes.

Co-authored-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agointel/isl/fill_state: Separate aux_mode handling from aux_surf
Jason Ekstrand [Fri, 4 May 2018 16:34:52 +0000 (09:34 -0700)]
intel/isl/fill_state: Separate aux_mode handling from aux_surf

v2. Avoid driver churn for now. (Nanley)

Co-authored-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agointel/isl: Add new aux modes available on gen12
Jason Ekstrand [Fri, 4 May 2018 16:43:01 +0000 (09:43 -0700)]
intel/isl: Add new aux modes available on gen12

v2. Add media compression. (Nanley)

Co-authored-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoi965/miptree: Avoid -Wswitch for the Gen12 aux modes
Nanley Chery [Fri, 27 Sep 2019 00:23:33 +0000 (17:23 -0700)]
i965/miptree: Avoid -Wswitch for the Gen12 aux modes

Avoid the compiler warnings for the new enums that will be introduced in
a future commit.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>