platform/upstream/mesa.git
5 years agopanfrost/decode: Decode exception status
Tomeu Vizoso [Tue, 25 Jun 2019 06:22:30 +0000 (08:22 +0200)]
panfrost/decode: Decode exception status

Arm's kernel driver mentions how to decode this field, which makes a bit
clearer what had happened.

Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/decode: Print AFBC struct when appropriate
Tomeu Vizoso [Tue, 25 Jun 2019 06:21:15 +0000 (08:21 +0200)]
panfrost/decode: Print AFBC struct when appropriate

Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agoradv: only export clip/cull distances if PS reads them
Samuel Pitoiset [Wed, 26 Jun 2019 14:35:44 +0000 (16:35 +0200)]
radv: only export clip/cull distances if PS reads them

The only exception is the GS copy shader which emits them
unconditionally.

Totals from affected shaders:
SGPRS: 71320 -> 71008 (-0.44 %)
VGPRS: 54372 -> 54240 (-0.24 %)
Code Size: 2952628 -> 2941368 (-0.38 %) bytes
Max Waves: 9689 -> 9723 (0.35 %)

This helps Dota2, Doom, GTAV and Hitman 2.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoradv: fix FMASK expand if layerCount is VK_REMAINING_ARRAY_LAYERS
Samuel Pitoiset [Wed, 26 Jun 2019 14:24:10 +0000 (16:24 +0200)]
radv: fix FMASK expand if layerCount is VK_REMAINING_ARRAY_LAYERS

This doesn't fix anything known, but it's likely going to
break if layerCount is ~0U.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoiris: Disable loop unrolling in GLSL IR.
Kenneth Graunke [Wed, 26 Jun 2019 02:34:45 +0000 (19:34 -0700)]
iris: Disable loop unrolling in GLSL IR.

Leave it to NIR instead, like i965 does.  Thanks to Tim Arceri for
noticing that I'd left this enabled by accident.

shader-db results on Skylake:

total instructions in shared programs: 15522628 -> 15521642 (<.01%)
instructions in affected programs: 94008 -> 93022 (-1.05%)
helped: 34
HURT: 33
helped stats (abs) min: 12 max: 48 x̄: 33.82 x̃: 42
helped stats (rel) min: 0.06% max: 22.14% x̄: 9.86% x̃: 10.89%
HURT stats (abs)   min: 1 max: 16 x̄: 4.97 x̃: 3t
HURT stats (rel)   min: 0.82% max: 3.77% x̄: 1.73% x̃: 1.53%
95% mean confidence interval for instructions value: -20.08 -9.35
95% mean confidence interval for instructions %-change: -5.95% -2.36%
Instructions are helped.

total cycles in shared programs: 367105221 -> 367074230 (<.01%)
cycles in affected programs: 10017660 -> 9986669 (-0.31%)
helped: 266
HURT: 184
helped stats (abs) min: 1 max: 9556 x̄: 151.35 x̃: 12
helped stats (rel) min: 0.08% max: 59.91% x̄: 4.66% x̃: 1.67%
HURT stats (abs)   min: 1 max: 1716 x̄: 50.37 x̃: 6
HURT stats (rel)   min: <.01% max: 24.40% x̄: 2.42% x̃: 0.85%
95% mean confidence interval for cycles value: -133.90 -3.84
95% mean confidence interval for cycles %-change: -2.44% -1.10%
Cycles are helped.

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
5 years agost/mesa: Set EmitNoIndirectSampler if GLSLVersion < 400.
Kenneth Graunke [Wed, 26 Jun 2019 04:00:46 +0000 (21:00 -0700)]
st/mesa: Set EmitNoIndirectSampler if GLSLVersion < 400.

This patch changes the code which sets EmitNoIndirectSampler to check
the core profile GLSL version, rather than the ARB_gpu_shader5 extension
enable.  st/mesa exposes ARB_gpu_shader5 if GLSLVersion (in core
profiles) or GLSLVersionCompat (in compat profiles) >= 400.

The Intel drivers do not currently expose ARB_gpu_shader5 in compat
profiles.  But the backend can absolutely handle indirect samplers.
Looking at the core profile version number should be a good indication
of what the driver supports.

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
5 years agoiris: Delete dead ice->state.streamout_strides field.
Kenneth Graunke [Thu, 27 Jun 2019 03:16:10 +0000 (20:16 -0700)]
iris: Delete dead ice->state.streamout_strides field.

Nothing uses this, it must be a remnant from an earlier approach.

5 years agonir/algebraic: Add helpers and a rule involving wrapping
Caio Marcelo de Oliveira Filho [Wed, 12 Jun 2019 23:48:21 +0000 (16:48 -0700)]
nir/algebraic: Add helpers and a rule involving wrapping

The helpers are needed so we can use the syntax `instr(cond)` in the
algebraic rules.  Add simple rule for dropping a pair of mul-div of
the same value when wrapping is guaranteed to not happen.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
5 years agospirv: Implement NoSignedWrap and NoUnsignedWrap decorations
Caio Marcelo de Oliveira Filho [Sat, 18 May 2019 05:52:42 +0000 (22:52 -0700)]
spirv: Implement NoSignedWrap and NoUnsignedWrap decorations

When handling the specified ALU operations, check for the decorations
and set nir_alu_instr no_signed_wrap and no_unsigned_wrap flags accordingly.

v2: Add a glsl_base_type_is_unsigned_integer() helper.  (Karol)

v3: Rename helper to glsl_base_type_is_uint().

v4: Use two flags, so we don't need the helper anymore.  (Connor)

v5: Pass alu directly to handle function.  (Jason)

Reviewed-by: Karol Herbst <kherbst@redhat.com> [v3]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
5 years agonir: Add a no wrapping bits to nir_alu_instr
Caio Marcelo de Oliveira Filho [Fri, 17 May 2019 20:46:38 +0000 (13:46 -0700)]
nir: Add a no wrapping bits to nir_alu_instr

They indicate the operation does not cause overflow or underflow.
This is motivated by SPIR-V decorations NoSignedWrap and
NoUnsignedWrap.

Change the storage of `exact` to be a single bit, so they pack
together.

v2: Handle no_wrap in nir_instr_set.  (Karol)

v3: Use two separate flags, since the NIR SSA values and certain
    instructions are typeless, so just no_wrap would be insufficient
    to know which one was referred to.  (Connor)

v4: Don't use nir_instr_set to propagate the flags, unlike `exact`,
    consider the instructions different if the flags have different
    values.  Fix hashing/comparing.  (Jason)

Reviewed-by: Karol Herbst <kherbst@redhat.com> [v1]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
5 years agodocs: add news item and link release notes for 19.0.8
Dylan Baker [Wed, 26 Jun 2019 20:48:06 +0000 (13:48 -0700)]
docs: add news item and link release notes for 19.0.8

This is an emergency release due to a critical bug.

5 years agodocs: Add mesa 19.0.8 sha256 sums
Dylan Baker [Wed, 26 Jun 2019 20:42:45 +0000 (13:42 -0700)]
docs: Add mesa 19.0.8 sha256 sums

5 years agodocs: Add docs for 19.0.8
Dylan Baker [Wed, 26 Jun 2019 20:08:54 +0000 (13:08 -0700)]
docs: Add docs for 19.0.8

5 years agonir: remove fnot/fxor/fand/for opcodes
Jonathan Marek [Fri, 21 Jun 2019 03:22:02 +0000 (23:22 -0400)]
nir: remove fnot/fxor/fand/for opcodes

There doesn't seem to be any reason to keep these opcodes around:
* fnot/fxor are not used at all.
* fand/for are only used in lower_alu_to_scalar, but easily replaced

Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
5 years agonir: opt_vectorize: combine different constant sources
Jonathan Marek [Fri, 21 Jun 2019 03:05:13 +0000 (23:05 -0400)]
nir: opt_vectorize: combine different constant sources

We can vectorize instructions with different constant sources by creating
a new load_const and using that.

Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/midgard: Merge embedded constants
Alyssa Rosenzweig [Thu, 20 Jun 2019 22:11:57 +0000 (15:11 -0700)]
panfrost/midgard: Merge embedded constants

In Midgard, a bundle consists of a few ALU instructions. Within the
bundle, there is room for an optional 128-bit constant; this constant is
shared across all instructions in the bundle.

Unfortunately, many instructions want a 128-bit constant all to
themselves (how selfish!). If we run out of space for constants in a
bundle, the bundle has to be broken up, incurring a performance and
space penalty.

As an optimization, the scheduler now analyzes the constants coming in
per-instruction and attempts to merge shared components, adjusting the
swizzle accessing the bundle's constants appropriately. Concretely,
given the GLSL:

   (a * vec4(1.5, 0.5, 0.5, 1.0)) + vec4(1.0, 2.3, 2.3, 0.5)

instead of compiling to the naive two bundles:

   vmul.fmul [temp], [a], r26
   fconstants 1.5, 0.5, 0.5, 1.0

   vadd.fadd [out], [temp], r26
   fconstants 1.0, 2.3, 2.3, 0.5

The scheduler can now fuse into a single (pipelined!) bundle:

   vmul.fmul [temp], [a], r26.xyyz
   vadd.fadd [out], [temp], r26.zwwy
   fconstants 1.5, 0.5, 1.0, 2.3

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/midgard: Share swizzle compose
Alyssa Rosenzweig [Tue, 18 Jun 2019 22:02:44 +0000 (15:02 -0700)]
panfrost/midgard: Share swizzle compose

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/midgard: Share swizzle/mask code
Alyssa Rosenzweig [Tue, 18 Jun 2019 22:02:19 +0000 (15:02 -0700)]
panfrost/midgard: Share swizzle/mask code

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Fix checksumming typo
Alyssa Rosenzweig [Wed, 26 Jun 2019 16:55:15 +0000 (09:55 -0700)]
panfrost: Fix checksumming typo

Fixes: 3e6c6bb0 ("panfrost: Merge checksum buffer with main BO")
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agoiris: Fix overzealous query object batch flushing.
Kenneth Graunke [Wed, 26 Jun 2019 04:14:11 +0000 (21:14 -0700)]
iris: Fix overzealous query object batch flushing.

In the past, each query object had their own BO.  Checking if the batch
referenced that BO was an easy way to check if commands were still
queued to compute the query value.  If so, we needed to flush.

More recently (c24a574e6c), we started using an u_upload_mgr for query
objects, placing multiple queries in the same BO.  One side-effect is
that iris_batch_references is a no longer a reasonable way to check if
commands are still queued for our query.  Ours might be done, but a
later query that happens to be in the same BO might be queued.  We don't
want to flush in that case.

Instead, check if the current batch's signalling syncpt is the one we
referenced when ending the query.  We know the syncpt can't have been
reused because our query is holding a reference, so a simple pointer
comparison should suffice.

Removes all batch flushing caused by query objects in Shadow of Mordor.

Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
5 years agoiris: Make an iris_batch_get_signal_syncpt() helper.
Kenneth Graunke [Wed, 26 Jun 2019 04:11:52 +0000 (21:11 -0700)]
iris: Make an iris_batch_get_signal_syncpt() helper.

This returns a pointer to the signalling syncpt, without incrementing
the reference count.  This can be useful for comparisons.

Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
5 years agopanfrost: Remove unneeded check in panfrost_scissor_culls_everything()
Boris Brezillon [Wed, 26 Jun 2019 09:16:31 +0000 (11:16 +0200)]
panfrost: Remove unneeded check in panfrost_scissor_culls_everything()

The ss local var is guaranteed to be != NULL. Get rid of this useless
check.

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Update copyright identifiers
Alyssa Rosenzweig [Tue, 25 Jun 2019 20:30:17 +0000 (13:30 -0700)]
panfrost: Update copyright identifiers

"Collabora, Ltd." should be listed in lieu of simply "Collabora"

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Suggested-by: Daniel Stone <daniels@collabora.com>
5 years agopanfrost/midgard: Reorder to permit constant bias
Alyssa Rosenzweig [Tue, 18 Jun 2019 16:02:35 +0000 (09:02 -0700)]
panfrost/midgard: Reorder to permit constant bias

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/midgard: Add helper to encode constant bias
Alyssa Rosenzweig [Tue, 18 Jun 2019 16:02:20 +0000 (09:02 -0700)]
panfrost/midgard: Add helper to encode constant bias

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/midgard: Handle negative immediate bias
Alyssa Rosenzweig [Tue, 18 Jun 2019 16:06:25 +0000 (09:06 -0700)]
panfrost/midgard: Handle negative immediate bias

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agofreedreno: correct batch_depends_on() logic
Rob Clark [Wed, 26 Jun 2019 15:25:55 +0000 (08:25 -0700)]
freedreno: correct batch_depends_on() logic

Signed-off-by: Rob Clark <robdclark@chromium.org>
5 years agofreedreno: drop unused arg from fd_batch_flush()
Rob Clark [Tue, 25 Jun 2019 18:21:28 +0000 (11:21 -0700)]
freedreno: drop unused arg from fd_batch_flush()

The `force` arg has been unused for a while.. but apparently I forgot to
garbage collect it.

Signed-off-by: Rob Clark <robdclark@chromium.org>
5 years agost/glsl: fix silly regression finding gs/tes variants
Timothy Arceri [Wed, 26 Jun 2019 13:02:12 +0000 (23:02 +1000)]
st/glsl: fix silly regression finding gs/tes variants

Fixes: d19fe5e67a39 ("st/glsl: support clamping color outputs in compat for gs/tes")

Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
5 years agost/glsl: support clamping color outputs in compat for gs/tes
Timothy Arceri [Tue, 25 Jun 2019 10:53:15 +0000 (20:53 +1000)]
st/glsl: support clamping color outputs in compat for gs/tes

This support requires the driver to be a NIR driver as we use the
NIR lowering pass to do the clamping.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agonir: add tess support to nir_lower_clamp_color_outputs()
Timothy Arceri [Tue, 25 Jun 2019 06:33:22 +0000 (16:33 +1000)]
nir: add tess support to nir_lower_clamp_color_outputs()

This will be used to add compat profile support for higher GL
versions.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
5 years agoglsl: Fix round64 conversion function
Sagar Ghuge [Mon, 24 Jun 2019 22:10:53 +0000 (15:10 -0700)]
glsl: Fix round64 conversion function

Fix round64 function to handle round to nearest even cases specially
with positive and negative numbers with fraction part 0.5.

v2: 1) Simplify unused bits (Elie Tournier)

Fixes:
   KHR-GL45.gpu_shader_fp64.builtin.round_dvec2
   KHR-GL45.gpu_shader_fp64.builtin.round_dvec3
   KHR-GL45.gpu_shader_fp64.builtin.round_dvec4
   KHR-GL45.gpu_shader_fp64.builtin.roundeven_double
   KHR-GL45.gpu_shader_fp64.builtin.roundeven_dvec2
   KHR-GL45.gpu_shader_fp64.builtin.roundeven_dvec3
   KHR-GL45.gpu_shader_fp64.builtin.roundeven_dvec4

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Elie Tournier <elie.tournier@collabora.com>
Acked-by: Anuj Phogat <anuj.phogat@gmail.com>
5 years agopanfrost/ci: Add RK3288 flipflops I don't want to deal with right now
Alyssa Rosenzweig [Tue, 25 Jun 2019 20:42:58 +0000 (13:42 -0700)]
panfrost/ci: Add RK3288 flipflops I don't want to deal with right now

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/ci: Update failures list
Alyssa Rosenzweig [Mon, 24 Jun 2019 17:16:33 +0000 (10:16 -0700)]
panfrost/ci: Update failures list

A ton of tests were fixed by this series. A few were incorrectly passing
before (QualityError, for instance) and now are explicitly failing. A
few legitimate regressions but overwhelmingly positive.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/ci: Set MESA_GLES_VERSION_OVERRIDE=3.0
Alyssa Rosenzweig [Mon, 24 Jun 2019 20:17:28 +0000 (13:17 -0700)]
panfrost/ci: Set MESA_GLES_VERSION_OVERRIDE=3.0

Fixes cube map tests due to disagreements between Mesa, dEQP, and the
spec...

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Cc: Tomeu Vizoso <tomeu.vizoso@collabora.com>
5 years agopanfrost/ci: Run full set of mipmap tests
Alyssa Rosenzweig [Mon, 24 Jun 2019 22:06:47 +0000 (15:06 -0700)]
panfrost/ci: Run full set of mipmap tests

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Advertise support for other 8-bit UNORM formats
Alyssa Rosenzweig [Tue, 25 Jun 2019 14:01:52 +0000 (07:01 -0700)]
panfrost: Advertise support for other 8-bit UNORM formats

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Use pipe_surface->format directly in blitter
Alyssa Rosenzweig [Tue, 25 Jun 2019 15:57:55 +0000 (08:57 -0700)]
panfrost: Use pipe_surface->format directly in blitter

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Invert swizzle for rendering
Alyssa Rosenzweig [Tue, 25 Jun 2019 15:51:48 +0000 (08:51 -0700)]
panfrost: Invert swizzle for rendering

Fixes rendering to e.g. alpha textures.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Honour first_layer...last_layer when sampling
Alyssa Rosenzweig [Mon, 24 Jun 2019 21:53:54 +0000 (14:53 -0700)]
panfrost: Honour first_layer...last_layer when sampling

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Use the sampler_view target (not the textures)
Alyssa Rosenzweig [Mon, 24 Jun 2019 21:39:37 +0000 (14:39 -0700)]
panfrost: Use the sampler_view target (not the textures)

u_blitter gets "special treatment" and uses this mechanism to cast
cube maps to 2D textures in order to texelFetch them.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/midgard: Assert guard texelFetch against cubemaps
Alyssa Rosenzweig [Mon, 24 Jun 2019 21:39:25 +0000 (14:39 -0700)]
panfrost/midgard: Assert guard texelFetch against cubemaps

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Zero pixels in any axis is zero pixels total
Alyssa Rosenzweig [Mon, 24 Jun 2019 21:13:20 +0000 (14:13 -0700)]
panfrost: Zero pixels in any axis is zero pixels total

Multiplication, not addition, so switch the logic operator.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Respect mip level when wallpapering
Alyssa Rosenzweig [Mon, 24 Jun 2019 20:53:44 +0000 (13:53 -0700)]
panfrost: Respect mip level when wallpapering

Fixes DATA_INVALID_FAULT raised when wallpapering while rendering to a
mipmap.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/midgard: Fixup NIR texture op
Alyssa Rosenzweig [Mon, 24 Jun 2019 17:35:03 +0000 (10:35 -0700)]
panfrost/midgard: Fixup NIR texture op

In a vertex shader, a tex op should map to txl, as there *must* be a LOD
given to the hardware (implicitly or explicitly).

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Support (non-)seamless cube maps
Alyssa Rosenzweig [Mon, 24 Jun 2019 16:16:11 +0000 (09:16 -0700)]
panfrost: Support (non-)seamless cube maps

Identify the seamless cubemap bit and passthrough the Gallium state
rather than setting unconditionally.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Merge checksum buffer with main BO
Alyssa Rosenzweig [Mon, 24 Jun 2019 14:08:52 +0000 (07:08 -0700)]
panfrost: Merge checksum buffer with main BO

This is similar to the AFBC merge; now all (non-imported) buffers use a
common backing buffer. Reenables checksumming, eliminating a performance
regression.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/decode: Limit MRT blend count
Alyssa Rosenzweig [Sun, 23 Jun 2019 18:29:46 +0000 (11:29 -0700)]
panfrost/decode: Limit MRT blend count

I thought I already fixed this. Maybe that was a dream...? Then again, I
might be dreaming now.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Clamp tile coordinates before job submission
Alyssa Rosenzweig [Sun, 23 Jun 2019 18:20:00 +0000 (11:20 -0700)]
panfrost: Clamp tile coordinates before job submission

Fixes TILE_RANGE_FAULT raised on some tests in
dEQP-GLES3.functional.fbo.blit.*

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Use dedicated u_blitter context for wallpapers
Alyssa Rosenzweig [Sun, 23 Jun 2019 18:05:10 +0000 (11:05 -0700)]
panfrost: Use dedicated u_blitter context for wallpapers

The main ctx->blitter instance should be reserved for blits originated
from Gallium (like mipmap generation). Since wallpapering is
conceptually different -- wallpaper blits can be triggered by Gallium
blits -- the blitter pipes must be separate to avoid potential u_blitter
recursion.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Sanity check layer
Alyssa Rosenzweig [Sat, 22 Jun 2019 00:39:02 +0000 (17:39 -0700)]
panfrost: Sanity check layer

It doesn't make sense to try to render to multiple array elements at
once.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Divide array_size by 6 for cubemaps
Alyssa Rosenzweig [Sat, 22 Jun 2019 00:27:05 +0000 (17:27 -0700)]
panfrost: Divide array_size by 6 for cubemaps

Addresses the disparity between Mali and Gallium definitions of
array_size.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Use get_texture_address for framebuffer computations
Alyssa Rosenzweig [Sat, 22 Jun 2019 00:23:49 +0000 (17:23 -0700)]
panfrost: Use get_texture_address for framebuffer computations

Allows for sharing some code as well as theoretically allowing cubemap
rendering.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Merge AFBC slab with BO backing
Alyssa Rosenzweig [Fri, 21 Jun 2019 21:54:44 +0000 (14:54 -0700)]
panfrost: Merge AFBC slab with BO backing

Rather than tracking AFBC memory "specially", just use the same codepath
as linear and tiled. Less things to mess up, I figure. This allows us to
use the standard setup_slices() call with AFBC resources, allowing
mipmapped AFBC resources.

Unfortunately, we do have to disable AFBC (and checksumming) in the
meantime to avoid functional regressions, as we don't know _a priori_ if
we'll need to access a resource from software (which is not yet hooked
up with AFBC) and we don't yet have routines to switch the layout of a
BO at runtime.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Z/S can't be tiled
Alyssa Rosenzweig [Fri, 21 Jun 2019 21:41:14 +0000 (14:41 -0700)]
panfrost: Z/S can't be tiled

As far as we know, Utgard-style tiling only works for color render
targets, not depth/stencil, so ensure we don't try to tile it (rather
than compress or plain old linear) and drive ourselves into a corner.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Enable mipmapping
Alyssa Rosenzweig [Fri, 21 Jun 2019 21:26:19 +0000 (14:26 -0700)]
panfrost: Enable mipmapping

Now the autogeneration of mipmaps is working (via u_blitter), we can
finally enable mipmaps!

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Enable blitting
Alyssa Rosenzweig [Fri, 21 Jun 2019 21:25:59 +0000 (14:25 -0700)]
panfrost: Enable blitting

Now that all the prerequisites breaking u_blitter are fixed, we can
finally hook up panfrost_blit.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Allow texelFetch for wallpaper blits
Alyssa Rosenzweig [Sun, 23 Jun 2019 18:49:49 +0000 (11:49 -0700)]
panfrost: Allow texelFetch for wallpaper blits

We just implemented the routine; we may as well use it.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost/midgard: Implement texelFetch (2D only)
Alyssa Rosenzweig [Fri, 21 Jun 2019 23:17:34 +0000 (16:17 -0700)]
panfrost/midgard: Implement texelFetch (2D only)

txf instructions can result from blits, so handle them rather than
crash. Only works for 2D textures (not even 2D array texture) due to a
register allocation constraint that may not be sorted for a while.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Skip flushes only for wallpapers, not any blit
Alyssa Rosenzweig [Fri, 21 Jun 2019 20:59:51 +0000 (13:59 -0700)]
panfrost: Skip flushes only for wallpapers, not any blit

We need the flush from u_blitter for a normal blit (e.g. for mipmaps);
it's only wallpaper-related blits that are special-cased.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Handle generate_mipmap ourselves
Alyssa Rosenzweig [Fri, 21 Jun 2019 20:57:42 +0000 (13:57 -0700)]
panfrost: Handle generate_mipmap ourselves

To avoid interference with the wallpaper code, we need to do some state
tracking when generating mipmaps. In particular, we need to mark the
generated layers as invalid before generating the mipmap, so we don't
try to backblit them if they already had content.

Likewise, we need to flush both before and after generating a mipmap
since our usual set_framebuffer_state flushing isn't quite there yet.
Ideally better optimizations would save the flush but I digress.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agopanfrost: Disable mipmapping if necessary
Alyssa Rosenzweig [Fri, 21 Jun 2019 23:58:48 +0000 (16:58 -0700)]
panfrost: Disable mipmapping if necessary

If a mipfilter is not set, it's legal to have an incomplete mipmap; we
should handle this accordingly. An "easy way out" is to rig the LOD
clamps.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
5 years agointel/blorp: Disable sampler state prefetching on Gen11
Kenneth Graunke [Mon, 24 Jun 2019 22:09:51 +0000 (15:09 -0700)]
intel/blorp: Disable sampler state prefetching on Gen11

Sampler state prefetching is broken on Gen11, and WA_160668216 says
to disable it.  Apparently sampler state prefetching also has basically
zero impact on performance, so we don't need to worry there.

i965, anv, and iris already handle this correctly, but we missed BLORP.
Ideally the kernel should globally disable this by writing SARCHKMD, at
which point we wouldn't have to worry about it.  But let's be defensive
and handle it ourselves too.

v2: separate out from BTP workaround in case we change that eventually

Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com> [v1]
5 years agoanv/descriptor_set: Only write texture swizzles if we have an image view
Jason Ekstrand [Tue, 25 Jun 2019 19:26:56 +0000 (14:26 -0500)]
anv/descriptor_set: Only write texture swizzles if we have an image view

When immutable samplers are set we call write_image_view with a NULL
image view.  This causes issues on IVB where we have to fake texture
swizzling.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110999
Fixes: d2aa65eb18 "anv: Emulate texture swizzle in the shader when..."

5 years agovirgl: add VIRGL_DEBUG_XFER
Chia-I Wu [Mon, 24 Jun 2019 17:47:59 +0000 (10:47 -0700)]
virgl: add VIRGL_DEBUG_XFER

When set, do as requested and skip any transfer optimization.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-By: Gert Wollny <gert.wollny@collabora.com>
Reviewed-By: Alexandros Frantzis <alexandros.frantzis@collabora.com>
5 years agovirgl: add VIRGL_DEBUG_SYNC
Chia-I Wu [Mon, 24 Jun 2019 17:45:36 +0000 (10:45 -0700)]
virgl: add VIRGL_DEBUG_SYNC

When set, wait after every each flush.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-By: Gert Wollny <gert.wollny@collabora.com>
Reviewed-By: Alexandros Frantzis <alexandros.frantzis@collabora.com>
5 years agovirgl: fix the value of VIRGL_DEBUG_BGRA_DEST_SWIZZLE
Chia-I Wu [Mon, 24 Jun 2019 17:30:52 +0000 (10:30 -0700)]
virgl: fix the value of VIRGL_DEBUG_BGRA_DEST_SWIZZLE

VIRGL_DEBUG_BGRA_DEST_SWIZZLE should use bit 3.  Make some cosmetic
changes as well.

Fixes: a478e56fbd33fa23503b63d41265a1c2f3253ed2
    virgl: Add debug flag to bypass driconf to enable the BGRA tweaks

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-By: Gert Wollny <gert.wollny@collabora.com>
Reviewed-By: Alexandros Frantzis <alexandros.frantzis@collabora.com>
5 years agoradv: rename and re-document cache flush flags
Samuel Pitoiset [Tue, 25 Jun 2019 15:57:45 +0000 (17:57 +0200)]
radv: rename and re-document cache flush flags

SMEM and VMEM caches are L0 on gfx10. Ported from RadeonSI.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoradv: set DISABLE_CONSTANT_ENCODE_REG to 1 for Raven2
Samuel Pitoiset [Tue, 25 Jun 2019 14:17:17 +0000 (16:17 +0200)]
radv: set DISABLE_CONSTANT_ENCODE_REG to 1 for Raven2

Ported from RadeonSI, will be emitted for GFX10 too.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoradv: clear CMASK layers instead of the whole buffer on GFX8
Samuel Pitoiset [Mon, 24 Jun 2019 16:40:29 +0000 (18:40 +0200)]
radv: clear CMASK layers instead of the whole buffer on GFX8

This reduces the size of fill operations needed to clear CMASK
for layered color textures.

GFX9 unsupported for now.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoradv: clear FMASK layers instead of the whole buffer on GFX8
Samuel Pitoiset [Mon, 24 Jun 2019 10:18:01 +0000 (12:18 +0200)]
radv: clear FMASK layers instead of the whole buffer on GFX8

This reduces the size of fill operations needed to clear FMASK
for layered color textures.

GFX9 unsupported for now.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoradv: always initialize levels without DCC as fully expanded
Samuel Pitoiset [Mon, 24 Jun 2019 15:03:28 +0000 (17:03 +0200)]
radv: always initialize levels without DCC as fully expanded

This fixes a rendering issue with RoTR/DXVK.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoi965: leaking of upload-BO with push constants
Sergii Romantsov [Tue, 18 Jun 2019 14:07:21 +0000 (17:07 +0300)]
i965: leaking of upload-BO with push constants

In case of any enabled VS members from: uses_firstvertex,
uses_baseinstance, uses_drawid, uses_is_indexed_draw
leaks may happens.
Call gen6_upload_push_constants allocates
stage_stat->push_const_bo. It than takes pointer from
push_const_bo to draw_params_bo (in the call
brw_prepare_shader_draw_parameters by brw_upload_data)
and do reference which finally haven't got unreferenced.

Fixes leak:
 136 bytes in 1 blocks are definitely lost in loss record 6 of 13
    at 0x4C31B25: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
    by 0xC2B64B7: bo_alloc_internal (brw_bufmgr.c:596)
    by 0xC2B6748: brw_bo_alloc (brw_bufmgr.c:672)
    by 0xC314BB3: brw_upload_space (intel_upload.c:88)
    by 0xC2EBBC5: gen6_upload_push_constants (gen6_constant_state.c:155)
    by 0xC9E4FA6: gen9_upload_vs_push_constants (genX_state_upload.c:3300)
    by 0xC2E0EDA: check_and_emit_atom (brw_state_upload.c:540)
    by 0xC2E0EDA: brw_upload_pipeline_state (brw_state_upload.c:659)
    by 0xC2E0FF1: brw_upload_render_state (brw_state_upload.c:681)
    by 0xC2C5D2D: brw_draw_single_prim (brw_draw.c:1052)
    by 0xC2C62CB: brw_draw_prims (brw_draw.c:1175)
    by 0xC488AD1: vbo_exec_vtx_flush (vbo_exec_draw.c:386)
    by 0xC485270: vbo_exec_FlushVertices_internal (vbo_exec_api.c:652)

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reported-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Signed-off-by: Sergii Romantsov <sergii.romantsov@globallogic.com>
5 years agodocs: update calendar, add news item and link release notes for X.Y.Z
Juan A. Suarez Romero [Tue, 25 Jun 2019 11:02:37 +0000 (13:02 +0200)]
docs: update calendar, add news item and link release notes for X.Y.Z

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
5 years agodocs: fix some typos in 19.0.7 release notes
Juan A. Suarez Romero [Tue, 25 Jun 2019 11:01:56 +0000 (13:01 +0200)]
docs: fix some typos in 19.0.7 release notes

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
5 years agodocs: add sha256 checksums for 19.1.1
Juan A. Suarez Romero [Tue, 25 Jun 2019 10:56:10 +0000 (12:56 +0200)]
docs: add sha256 checksums for 19.1.1

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
(cherry picked from commit d54dc24d6d6ad9ce4fbebd5f9a69a92633504c40)

5 years agodocs: add release notes for 19.1.1
Juan A. Suarez Romero [Tue, 25 Jun 2019 10:43:49 +0000 (12:43 +0200)]
docs: add release notes for 19.1.1

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
(cherry picked from commit 22eddd8b9d4d905237494c9841e65e4c073ab514)

5 years agointel/compiler: silence a warning of using different enum type
Tapani Pälli [Mon, 24 Jun 2019 11:36:28 +0000 (14:36 +0300)]
intel/compiler: silence a warning of using different enum type

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
5 years agoegl: replace dead vfunc with an error
Eric Engestrom [Sat, 22 Jun 2019 14:53:36 +0000 (15:53 +0100)]
egl: replace dead vfunc with an error

st/egl used to support eglCreatePbufferFromClientBuffer, but now that
it's gone, any call to it would segfault.

Let's return a nice error instead.

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
5 years agoegl: delete unused vfuncs
Eric Engestrom [Fri, 14 Jun 2019 15:08:44 +0000 (16:08 +0100)]
egl: delete unused vfuncs

Nobody ever uses these, so let's just hard code them instead.

If an EGL driver ever comes around that needs them they're trivial to
re-add.

Suggested-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
5 years agoegl: drop empty eglfallbacks.c
Eric Engestrom [Sat, 22 Jun 2019 15:09:48 +0000 (16:09 +0100)]
egl: drop empty eglfallbacks.c

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
5 years agoegl: move eglGetSyncAttrib() fallback to eglapi.c
Eric Engestrom [Sat, 22 Jun 2019 21:33:00 +0000 (22:33 +0100)]
egl: move eglGetSyncAttrib() fallback to eglapi.c

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
5 years agoegl: move eglSwapInterval() fallback to eglapi.c
Eric Engestrom [Sat, 22 Jun 2019 21:32:50 +0000 (22:32 +0100)]
egl: move eglSwapInterval() fallback to eglapi.c

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
5 years agoegl: move eglSurfaceAttrib() fallback to eglapi.c
Eric Engestrom [Sat, 22 Jun 2019 21:32:26 +0000 (22:32 +0100)]
egl: move eglSurfaceAttrib() fallback to eglapi.c

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
5 years agoegl: move eglQuerySurface() fallback to eglapi.c
Eric Engestrom [Sat, 22 Jun 2019 21:31:53 +0000 (22:31 +0100)]
egl: move eglQuerySurface() fallback to eglapi.c

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
5 years agoegl: move eglQueryContext() fallback to eglapi.c
Eric Engestrom [Sat, 22 Jun 2019 21:31:26 +0000 (22:31 +0100)]
egl: move eglQueryContext() fallback to eglapi.c

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
5 years agoegl: move eglGetConfigAttrib() fallback to eglapi.c
Eric Engestrom [Sat, 22 Jun 2019 21:31:00 +0000 (22:31 +0100)]
egl: move eglGetConfigAttrib() fallback to eglapi.c

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
5 years agoegl: move eglChooseConfig() fallback to eglapi.c
Eric Engestrom [Sat, 22 Jun 2019 21:30:23 +0000 (22:30 +0100)]
egl: move eglChooseConfig() fallback to eglapi.c

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
5 years agoegl: move eglGetConfigs() fallback to eglapi.c
Eric Engestrom [Sat, 22 Jun 2019 21:29:46 +0000 (22:29 +0100)]
egl: move eglGetConfigs() fallback to eglapi.c

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
5 years agofreedreno/a5xx: fix batch leak in fd5 blitter path
Rob Clark [Mon, 24 Jun 2019 22:06:13 +0000 (15:06 -0700)]
freedreno/a5xx: fix batch leak in fd5 blitter path

Fixes: 3d198926a48 freedreno: use fd_bc_alloc_batch instead of fd_batch_create.
Signed-off-by: Rob Clark <robdclark@chromium.org>
5 years agoradeonsi: don't set spi_ps_input_* for monolithic shaders
Marek Olšák [Wed, 19 Jun 2019 23:12:24 +0000 (19:12 -0400)]
radeonsi: don't set spi_ps_input_* for monolithic shaders

The driver doesn't use these values and ac_rtld has assertions
expecting the value of 0.

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoradeonsi: rename and re-document cache flush flags
Marek Olšák [Tue, 18 Jun 2019 23:06:57 +0000 (19:06 -0400)]
radeonsi: rename and re-document cache flush flags

SMEM and VMEM caches are L0 on gfx10.

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoradeonsi: fix AMD_DEBUG=nofmask
Marek Olšák [Fri, 24 May 2019 23:56:17 +0000 (19:56 -0400)]
radeonsi: fix AMD_DEBUG=nofmask

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoradeonsi: flatten the switch for DPBB tunables
Marek Olšák [Thu, 20 Jun 2019 02:24:51 +0000 (22:24 -0400)]
radeonsi: flatten the switch for DPBB tunables

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
5 years agoradeonsi: set the calling convention for inlined function calls
Marek Olšák [Wed, 19 Jun 2019 23:00:50 +0000 (19:00 -0400)]
radeonsi: set the calling convention for inlined function calls

otherwise the behavior is undefined

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
5 years agoradeonsi: refactor si_update_vgt_shader_config
Nicolai Hähnle [Fri, 21 Sep 2018 13:38:42 +0000 (15:38 +0200)]
radeonsi: refactor si_update_vgt_shader_config

We'll have to extend this at some point, and using a bitfield union in
this way makes it easier to get the right index without excessive
branching.

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoamd/rtld: update the ELF representation of LDS symbols
Nicolai Hähnle [Sun, 16 Jun 2019 23:24:29 +0000 (01:24 +0200)]
amd/rtld: update the ELF representation of LDS symbols

The initial prototype used a processor-specific symbol type, but
feedback suggests that an approach using processor-specific section
name that encodes the alignment analogous to SHN_COMMON symbols is
preferred.

This patch keeps both variants around for now to reduce problems
with LLVM compatibility as we switch branches around.

This also cleans up the error reporting in this function.

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agoac/surface: remove addrlib_family_rev_id
Marek Olšák [Thu, 20 Jun 2019 01:47:46 +0000 (21:47 -0400)]
ac/surface: remove addrlib_family_rev_id

Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
5 years agodocs: update calendar, add news item and link release notes for 19.0.7
Dylan Baker [Mon, 24 Jun 2019 23:24:05 +0000 (16:24 -0700)]
docs: update calendar, add news item and link release notes for 19.0.7

5 years agodocs: Add SHA256 sums for 19.0.7
Dylan Baker [Mon, 24 Jun 2019 23:21:34 +0000 (16:21 -0700)]
docs: Add SHA256 sums for 19.0.7