Alyssa Rosenzweig [Tue, 20 Aug 2019 22:36:00 +0000 (15:36 -0700)]
panfrost: Break up usage2 field
This is another bit field describing layout.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 22:24:45 +0000 (15:24 -0700)]
pan/decode: Pretty-print sRGB format
We can just stick an "s" in if it's sRGB.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 22:24:38 +0000 (15:24 -0700)]
panfrost: Remove ancient TODO
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 22:24:18 +0000 (15:24 -0700)]
panfrost: nr_mipmap_levels -> levels
No need to be so verbose.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 21:58:46 +0000 (14:58 -0700)]
pan/decode: Validate texture dimensionality
Textures of a smaller dimension don't need higher dimensions printed.
This allows us to be more compact, while enforcing verification that
higher dimensions must be zero.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 21:48:55 +0000 (14:48 -0700)]
pan/decode: Break out pandecode_texture function
It's massive and hugely nested indentation -- break it out so it's
legible.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 21:40:26 +0000 (14:40 -0700)]
pan/decode: Guard texture unknowns as zero trips
unknown3A I think I've actually seen on T6xx but.. we'll see what
happens in traces going forward. We don't want the zero noise normally,
and if they show up in the wild, we want to draw attention to them.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 21:34:09 +0000 (14:34 -0700)]
pan/decode: Use GLSL style formats/swizzles
This dramatically reduces visual clutter: now an entire
attribute/varying record looks something like:
rgba32f attribute_0[16].bgra;
which is equivalent to the raw structure:
{
.index = 0,
.format = MALI_FORMAT_RGBA32F,
.swizzle = (MALI_CHANNEL_BLUE << 9) | ....,
.src_offset = 16,
}
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 21:03:18 +0000 (14:03 -0700)]
pan/decode: Don't print the default swizzle
It's just noise.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 20:59:26 +0000 (13:59 -0700)]
pan/decode: Validate swizzles against format
We want to make sure we don't access a component in the swizzle that
doesn't exist in the format, since that is (as far as I know) undefined.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 20:58:55 +0000 (13:58 -0700)]
pan/decode: Treat RESERVED swizzles as errors
We've never seen them, so if they come up in trace, we want to draw
attention to that.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 20:33:39 +0000 (13:33 -0700)]
pan/decode: Handle VARYING_DISCARD
Varying discard is not used by Panfrost, but the blob uses it sometimes
to have some padding in the varyings table, probably to minimize
per-draw overhead. (...We should maybe consider this ourselves!)
Let's check for this and ensure the rest of the record is consistent
with a discarded varying.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 20:21:28 +0000 (13:21 -0700)]
panfrost: Don't trip the prefix magic field
What *is* this?
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 20:19:40 +0000 (13:19 -0700)]
pan/decode: Guard attribute unknowns
One should be zero. The other has always been seen as set, so check
this.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 20:14:56 +0000 (13:14 -0700)]
panfrost: Don't crash on GL_CLAMP
It's a legacy GL thing... we don't really need to handle it *right* now,
but we shouldn't crash..
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 20:03:07 +0000 (13:03 -0700)]
panfrost: Do not expose PIPE_CAP_TEXTURE_MIRROR_CLAMP
This CAP controls a desktop-only extension. If the corresponding support
exists in the hardware, we don't know how to use it.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 19:07:31 +0000 (12:07 -0700)]
panfrost: Fix scoreboarding with dependency on job #0
Subtle issue masked by how we emitted SET_VALUE jobs, but this case can
and does occur, so let's fix it.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 18:25:29 +0000 (11:25 -0700)]
pan/decode: Normalize final instances of XXX
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 18:24:32 +0000 (11:24 -0700)]
pan/decode: Normalize case matching XXX format
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 18:18:46 +0000 (11:18 -0700)]
pan/decode: Mark tripped zeroes with XXX
This normalizes the printed format. It also makes it easier for the
future when we may introduce semantic _warn and _error handlers.
A tripped zero is essentially a hazard to check for.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 18:11:46 +0000 (11:11 -0700)]
pan/decode: Check for MFBD preload chicken bit
If this bit is clear, MFBD preload will be enabled, and you.. don't want
that. (At least, when the bit is clear, the old contents of the
framebuffer will be preserved. I'm assuming this is what "MFBD preload"
refers to in kbase.)
Validate that this bit is always set.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 20 Aug 2019 18:06:07 +0000 (11:06 -0700)]
pan/decode: Validate AFBC fields are zero when AFBC is disabled
There is no "chunknown" structure; that part of the union is an artefact
from falsely believing vertex/tiler MFBDs could have render targets
attached (they can't). These are just plain old AFBC fields, and if
there is no AFBC, it's error to set these field.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 22:16:01 +0000 (15:16 -0700)]
pan/decode: Do not print uniform/buffers explicitly
For our purposes of driver debugging, the contents of uniform buffers
are rarely interesting; we're more concerned about the metadata setting
them up.
We do need to be careful to validate the sizes of both uniforms and
uniform buffers.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 22:14:48 +0000 (15:14 -0700)]
pan/decode: Add static bounds checking utility
Many structures in the command stream have a GPU address and size
determined statically. We should check that the pointers we are passed
are valid and the buffers they point to are big enough for the given
size. If they're not, an MMU fault would be raised.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 21:58:36 +0000 (14:58 -0700)]
pan/decode: Don't print unreferenced attribute memory
This is a source of uninitialized memory leaking into the traces.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 21:47:50 +0000 (14:47 -0700)]
pan/decode: Check for a number of potential issues
Verify sizes / masks / etc against something logical to cull down the
trace space and automatically guard against a number of potential
hazards.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 21:30:53 +0000 (14:30 -0700)]
panfrost: Correct polygon size computations
While the algorithm for computing the header size has been correct for a
while, we used a major hack to conservatively guess the body size. Let's
scrap that and figure out the algorithm we actually need to use to be
bit-identical with what the hardware expects.
We do have to be careful to add the header size to total comptued BO
size.
It's not clear how big the polygon list needs to be in practice -- but
it has to be somewhat bigger than the polygon list itself. This needs
more investigation. If we size the polygon list exactly based on the
polygon_list_size field, we get faults like:
[ 1224.219886] panfrost
ff9a0000.gpu: Unhandled Page fault in AS0 at VA 0x000000001BDE8000
Reason: TODO
raw fault status: 0x660003C3
decoded fault status: SLAVE FAULT
exception type 0xC3: TRANSLATION_FAULT_LEVEL3
access type 0x3: WRITE
source id 0x6600
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 19:44:16 +0000 (12:44 -0700)]
panfrost: Remove DRY_RUN
Nobody uses this anymore anyway.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 19:41:18 +0000 (12:41 -0700)]
pan/decode: Print "just right" count of texture pointers
The other commented lines just add noise/entropy we don't want, and can
in fact crash the trace due to asserts failing.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 18:48:32 +0000 (11:48 -0700)]
pan/decode: Verify and omit polygon size
The polygon sizes are computed from the width/height/flags, so we can
reverse the computation and use our computation to verify the two
computation algorithms are bit-identical. If they are, we can omit the
computed fields.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 18:19:15 +0000 (11:19 -0700)]
panfrost: Move pan_tiler.c outside of Gallium
The routines in this file may be shared with Vulkan.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 17:56:23 +0000 (10:56 -0700)]
pan/decode: Bounds check polygon list and tiler heap
We have the BOs available; ensure that the bounds specified in the
command stream are actually the correct bounds.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 17:55:29 +0000 (10:55 -0700)]
pan/decode: Allow updating mmaps
This allows the caller to call track_mmap multiple times for the same
gpu_va for the purpose of updating the mmap. This is used to trace
invisible BOs with kbase and doesn't apply to native traces.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 19 Aug 2019 17:38:25 +0000 (10:38 -0700)]
pan/decode: Express tiler structures as offsets
This allows us to catch a class of errors (for negative offsets, etc)
automatically.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 23:36:39 +0000 (16:36 -0700)]
pan/decode: Don't print zero exception_status
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 23:31:54 +0000 (16:31 -0700)]
pan/decode: Fix missing NULL terminator
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 23:31:34 +0000 (16:31 -0700)]
pan/decode: Silence workgroups_x_shift_2
Since we're bit-identical we can compare the computed value.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 23:31:00 +0000 (16:31 -0700)]
panfrost: Implement workgroups_x_shift_2 quirk
I'm not sure why this is done this way, but let's follow the blob.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 23:22:38 +0000 (16:22 -0700)]
pan/decode: Don't print canonical workgroup encoding
The on-the-wire representation of workgroups is not 1:1 to the decoded
Gallium-level workgroups (there are multiple valid encodings; see the
previous commit). Nevertheless, since we're now bit-identical in packing
vs the blob, we can check for a canonical form and only print the
verbose trace if we fail the canonical form.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 23:21:45 +0000 (16:21 -0700)]
panfrost: Set workgroups z to 32 for non-instanced graphics
This is a blob quirk; in so much as I know, the hardware doesn't care.
But we're trying to be bit-identical to take as much entropy out of
traces as possible, so let's introduce the quirk.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 20:57:38 +0000 (13:57 -0700)]
panfrost: Move pan_invocation to shared panfrost/
The routines in this file have no dependency on Gallium. Let's share
them so they can be used for a theoretical future Vulkan driver or, more
immediately, consulted when tracing.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 20:30:39 +0000 (13:30 -0700)]
pan/decode: Don't print MALI_DRAW_NONE
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 20:28:06 +0000 (13:28 -0700)]
pan/decode: Eliminate DYN_MEMORY_PROP
It's obvious that it's linked by virtue of us printing the struct it
links against. No need to repeat ourselves.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alejandro Piñeiro [Thu, 17 Jan 2019 12:40:51 +0000 (13:40 +0100)]
i965: Enable OpenGL 4.6 for Gen8+
The last remaining stuff was ARB_gl_spirv and ARB_spirv_extensions.
Note that it is really likely that we can enable it for some Gen7 (as
4.5 was), but it was not tested yet.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Alejandro Piñeiro [Sat, 9 Mar 2019 15:06:07 +0000 (16:06 +0100)]
mesa/version: uncomment SPIR-V extensions
As they are implemented on i965, so we can expose 4.6.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Alejandro Piñeiro [Tue, 5 Dec 2017 14:29:43 +0000 (15:29 +0100)]
i965: enable ARB_gl_spirv extension and ARB_spirv_extensions for gen7+
v2: squashed the two enable patches with the docs one (Jason)
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Tomeu Vizoso [Tue, 20 Aug 2019 08:56:23 +0000 (10:56 +0200)]
panfrost/ci: Print load stats
To help make sure we are running tests in the ideal number of threads,
print load stats to make obvious when there's a problem with
utilization.
This will be specially useful when we run tests on a wider variety of
devices.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tomeu Vizoso [Mon, 19 Aug 2019 15:30:10 +0000 (17:30 +0200)]
panfrost/ci: Install qemu-arm-static into chroot
Some runners may be configured such that the qemu binary might not be
available by the time we need to start running commands within the
chroot.
So make sure that it's there to avoid suprising problems in that case.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tomeu Vizoso [Mon, 19 Aug 2019 14:50:01 +0000 (16:50 +0200)]
panfrost/ci: Build kernel with CONFIG_DETECT_HUNG_TASK
There's lots of locking changes going into the Panfrost kernel driver,
so better be prepared.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tomeu Vizoso [Mon, 19 Aug 2019 14:40:53 +0000 (16:40 +0200)]
panfrost/ci: Print bootstrap log
A number of things can go wrong when building the rootfs from within a
non-native chroot, so make sure to print the bootstrap.log so we can
tell what's going on.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tomeu Vizoso [Thu, 15 Aug 2019 08:41:21 +0000 (10:41 +0200)]
panfrost/ci: Use Volt-based runner for dEQP tests
It's able to run tests in parallel, fully utilizing the HW and
shortening considerable the time it takes.
Needed to disable tests in RK3288 for now because Volt doesn't support
armhf yet, though this should be fixed soon.
Tests are now run with --deqp-gl-config-name=rgba8888d24s8ms0, so we are
hitting a few more failures in tests that previously were being skipped.
The time to run the tests decreases from around 8 minutes to 1:45
minutes, allowing for extending coverage without increasing CI times too
much.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Samuel Pitoiset [Wed, 21 Aug 2019 07:04:46 +0000 (09:04 +0200)]
radv: implement VK_AMD_shader_core_properties2
Trivial extension that matches PAL.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Samuel Pitoiset [Tue, 20 Aug 2019 14:50:57 +0000 (16:50 +0200)]
radv: force enable VK_AMD_shader_ballot for Wolfenstein Youngblood
This gives a nice boost, +20% at this time on my Vega 56. Shader
ballot should be enabled by default at some point but it reduces
performance a bit (-6%) with Wolfeinstein II. Enable it only for
Youngblood at the moment, like what we did for Talos in the past.
As a bonus point, it gets rid of some minor artifacts that only
happens when ballot is disabled for some reasons.
Cc: 19.2 <mesa-stable@lists.freedesktop.org
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Samuel Pitoiset [Tue, 20 Aug 2019 14:50:56 +0000 (16:50 +0200)]
radv: add a new debug option called RADV_DEBUG=noshaderballot
Shader ballot will be enabled by default for Wolfenstein
Youngblood. This follows what we did for sisched.
Cc: 19.2 <mesa-stable@lists.freedesktop.org
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Samuel Pitoiset [Wed, 21 Aug 2019 06:38:24 +0000 (08:38 +0200)]
radv: allow to enable VK_AMD_shader_ballot only on GFX8+
Scans aren't implemented on SI/CIK.
Cc: 19.2 <mesa-stable@lists.freedesktop.org>
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Danylo Piliaiev [Tue, 20 Aug 2019 15:48:33 +0000 (18:48 +0300)]
nir/loop_analyze: Treat do{}while(false) loops as 0 iterations
Loops like:
block block_0:
vec1 32 ssa_2 = load_const (0x00000020)
vec1 32 ssa_3 = load_const (0x00000001)
loop {
vec1 32 ssa_7 = phi block_0: ssa_3, block_4: ssa_9
vec1 1 ssa_8 = ige ssa_2, ssa_7
if ssa_8 {
break
} else {
}
vec1 32 ssa_9 = iadd ssa_7, ssa_1
}
Were treated as having more than 1 iteration and after unrolling
produced wrong results, however such loop will exit during
the first iteration if not unrolled.
So we check if loop will actually loop.
Fixes tests/shaders/glsl-fs-loop-while-false-02.shader_test
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Danylo Piliaiev [Mon, 19 Aug 2019 15:13:12 +0000 (18:13 +0300)]
nir/loop_unroll: Prepare loop for unrolling in wrapper_unroll
Without loop_prepare_for_unroll loops are losing phis.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111411
Fixes: 5db98195 "nir: add loop unroll support for wrapper loops"
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Danylo Piliaiev [Mon, 19 Aug 2019 15:05:39 +0000 (18:05 +0300)]
nir/loop_unroll: Update the comments for loop_prepare_for_unroll
The comments say that we should remove continue if it is the last
intruction in a loop however we remove any kind of jump.
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Bas Nieuwenhuizen [Wed, 21 Aug 2019 08:27:05 +0000 (10:27 +0200)]
radv: Emit VGT_GS_ONCHIP_CNTL for tess on GFX10.
Otherwise hangs are possible. This register was already set for
GS and NGG.
Fixes: 5eaed7ecfce "radv/gfx10: enable support for NAVI10, NAVI12 and NAVI14"
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Tue, 20 Aug 2019 23:50:53 +0000 (01:50 +0200)]
radv: Use correct vgpr_comp_cnt for VS if both prim_id and instance_id are needed.
Should take the max of the 2.
Fixes: ea337c8b7e9 "radv/gfx10: fix VS input VGPRs with the legacy path"
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Daniel Schürmann [Wed, 14 Aug 2019 14:25:26 +0000 (16:25 +0200)]
nir/algebraic: some subtraction optimizations
Changes with RADV/ACO:
Totals from affected shaders:
SGPRS: 444087 -> 455543 (2.58 %)
VGPRS: 436468 -> 436768 (0.07 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size:
13448928 ->
13353520 (-0.71 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 68060 -> 67979 (-0.12 %)
Wait states: 0 -> 0 (0.00 %)
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Lionel Landwerlin [Wed, 7 Aug 2019 15:03:37 +0000 (18:03 +0300)]
radeonsi: take reference glsl types for compile threads
An application quitting before the destroying its GL context and
binding a NULL context might still have a radeonsi compiler thread
running and potentially still accessing the types.
Therefore take a reference for the duration of the threads' lifetime.
v2: Only ref the glsl types, the builtins should be used by the time
shader data gets to a gallium driver.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Lionel Landwerlin [Wed, 31 Jul 2019 09:12:10 +0000 (12:12 +0300)]
mesa/compiler: rework tear down of builtin/types
The issue we're running into when running CTS is that glsl types are
deleted while builtins depending on them are not.
This happens because on one hand we have glsl types ref counted, but
builtins are not. Instead builtins are destroyed when unloading libGL
or explicitly calling glReleaseShaderCompiler().
This change removes almost entirely any dealing with glsl types
ref/unref by letting the builtins deal with it instead. In turn we
introduce a builtin ref count mechanism. Each GL context takes a
reference on the builtins when compiling a shader for the first time.
It releases the reference when the context is destroyed. It can also
explicitly release those when glReleaseShaderCompiler() is called.
Finally we also take a reference on the glsl types when loading libGL
to avoid recreating glsl types too often.
v2: Ensure we take a reference if we don't have one in link step (Lionel)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110796
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Lionel Landwerlin [Wed, 31 Jul 2019 09:00:47 +0000 (12:00 +0300)]
compiler: ensure glsl types are not created without a reference
We want to detect invalid refcounting so assert we have at least one
use before creating types.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Lionel Landwerlin [Wed, 31 Jul 2019 12:40:05 +0000 (15:40 +0300)]
nir/tests: take reference on glsl types
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Lionel Landwerlin [Wed, 31 Jul 2019 10:51:50 +0000 (13:51 +0300)]
glsl/tests: take refs on glsl types
Much like each driver, tests as standalone entities must take
references on the glsl types.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Samuel Pitoiset [Tue, 20 Aug 2019 13:23:12 +0000 (15:23 +0200)]
radv/gfx10: hardcode some depth+stencil formats in the format table
The script doesn't handle them correctly and D16_UNORM_S8_UINT
isn't supported by the hardware, mark it as invalid.
This fixes warning when generating gfx10_format_table.h.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111393
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Samuel Pitoiset [Tue, 20 Aug 2019 13:23:11 +0000 (15:23 +0200)]
radv/gfx10: tidy up gfx10_format_table.py
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Ilia Mirkin [Sat, 17 Aug 2019 16:13:34 +0000 (12:13 -0400)]
gallium/vl: use compute preference for all multimedia, not just blit
The compute paths in vl are a bit AMD-specific. For example, they (on
nouveau), try to use a BGRX8 image format, which is not supported.
Fixing all this is probably possible, but since the compute paths aren't
in any way better, it's difficult to care.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111213
Fixes: 9364d66cb7 (gallium/auxiliary/vl: Add video compositor compute shader render)
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Emil Velikov [Tue, 20 Aug 2019 22:14:53 +0000 (23:14 +0100)]
docs: update calendar for 19.2.x
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Emil Velikov [Tue, 20 Aug 2019 21:38:02 +0000 (22:38 +0100)]
docs: add 19.3.0-devel release notes template
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Emil Velikov [Tue, 20 Aug 2019 21:33:49 +0000 (22:33 +0100)]
mesa: bump version to 19.3.0-devel
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Erico Nunes [Sun, 18 Aug 2019 23:04:57 +0000 (01:04 +0200)]
lima/ppir: use ra_get_best_spill_node to select spill node
ra_get_best_spill_node is what other users of the mesa register
allocator use.
Switching to it now also fixes an infinite loop issue with ppir regalloc
with the ppir control flow patchset, and also provides a small gain over
the previous herusitic on number of spilled nodes testing with
shader-db.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Eric Anholt [Wed, 7 Aug 2019 18:50:07 +0000 (11:50 -0700)]
tgsi: Remove unused tgsi_check_soa_dependencies().
Acked-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-By: Gert Wollny <gert.wollny@collabora.com>
Eric Anholt [Wed, 7 Aug 2019 18:43:47 +0000 (11:43 -0700)]
tgsi: Drop the SSE2 constants setup that's been dead code since 2011.
The SSE2 executor was removed in
4eb3225b38ce ("Remove tgsi_sse2.")
Acked-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-By: Gert Wollny <gert.wollny@collabora.com>
Eric Anholt [Wed, 7 Aug 2019 18:24:02 +0000 (11:24 -0700)]
tgsi: drop a stale comment
This was fixed in
912ed84f8338 ("tgsi: move to using vector for system
values.")
Acked-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-By: Gert Wollny <gert.wollny@collabora.com>
Eric Anholt [Fri, 9 Aug 2019 17:32:40 +0000 (10:32 -0700)]
gitlab-ci: Enable the GLES2/3 CTS on softpipe.
The GLES2 CTS takes about 8 minutes of total runtime (at parallel 4 is
~2 minutes in the test stage if runners are free), while GLES3 takes
about 25. Since the GLES3 run is pretty expensive, just do a cheap
touch test of 1 out of every 10 tests in the test list on MRs, until
we can get the runtime down.
v2: Drop the full run for now until we can bring runtime down or bring
up a dedicated mesa runner.
Reviewed-by: Eric Engestrom <eric@engestrom.ch> (v1)
Reviewed-By: Gert Wollny <gert.wollny@collabora.com> (v1)
Jose Maria Casanova Crespo [Tue, 20 Aug 2019 16:34:09 +0000 (18:34 +0200)]
mesa: reverse no_error on compressed_tex_sub_image for TEX_MODE_CURRENT
This fixes the regression introduced on "mesa: refactor
compressed_tex_sub_image function" that started to crash
KHR-GLES2.texture_3d.compressed_texture.negative_compressed_tex_sub_image
Fixes: 7df233d68dc ("mesa: refactor compressed_tex_sub_image function")
Reviewed-by: Eric Anholt <eric@anholt.net>
Adam Jackson [Wed, 14 Aug 2019 17:06:37 +0000 (13:06 -0400)]
glx: Eliminate glx_config::{rgb,float,colorIndex}Mode
These are redundant with glx_config::renderType, let's just use that
consistently.
Adam Jackson [Wed, 14 Aug 2019 16:51:13 +0000 (12:51 -0400)]
glx: Remove unused glx_config::pixmapMode
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Adam Jackson [Wed, 14 Aug 2019 16:14:06 +0000 (12:14 -0400)]
glx: convert glx_config_create_list to one big calloc
Simpler, less failure prone, less malloc overhead, what's not to like.
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Adam Jackson [Wed, 14 Aug 2019 16:04:42 +0000 (12:04 -0400)]
glx: convert a malloc+memset to calloc
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Adam Jackson [Wed, 14 Aug 2019 16:02:49 +0000 (12:02 -0400)]
glx: Fix parameter documentation of glx_config_create_list
'minimum_size' is not, in fact, an argument to this function.
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Arcady Goldmints-Orlov [Sat, 27 Jul 2019 19:52:41 +0000 (14:52 -0500)]
anv: inline uniforms blocks don't count toward descriptor set limits
In a descriptor set inline uniform blocks don't use up any bindings.
However, the presence of any inline uniform blocks doed require the
use of the descriptor buffer, which takes up one binding.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Daniel Schürmann [Mon, 20 May 2019 12:58:23 +0000 (14:58 +0200)]
nir: add divergence analysis pass.
This pass expects the shader to be in LCSSA form.
The algorithm is based on 'The Simple Divergence Analysis' from
Diogo Sampaio, Rafael De Souza, Sylvain Collange, Fernando Magno Quintão Pereira.
Divergence Analysis. ACM Transactions on Programming Languages and Systems (TOPLAS)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Rhys Perry [Wed, 24 Apr 2019 09:25:25 +0000 (10:25 +0100)]
nir/subgroups: Lower clustered reductions with cluster_size >= subgroup_size into reductions
The behavior for reductions with cluster_size >= subgroup_size is implementation defined.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Rhys Perry [Mon, 5 Aug 2019 14:24:18 +0000 (15:24 +0100)]
nir/lcssa: allow to create LCSSA phis for loop-invariant booleans
ACO depends on LCSSA phis for divergent booleans to work correctly.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Daniel Schürmann [Thu, 28 Mar 2019 11:17:51 +0000 (12:17 +0100)]
nir/lcssa: Skip loop invariant variables when converting to LCSSA.
Co-authored-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Rhys Perry [Thu, 31 Jan 2019 16:33:23 +0000 (16:33 +0000)]
nir: make nir_to_lcssa() a general NIR pass.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Daniel Schürmann [Wed, 14 Aug 2019 16:52:50 +0000 (18:52 +0200)]
nir/lcssa: handle deref instructions properly
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Fixes: 414148cdc124 "nir: Support deref instructions in loop_analyze"
Jose Maria Casanova Crespo [Mon, 19 Aug 2019 13:05:25 +0000 (15:05 +0200)]
tgsi_to_nir: only update TGSI properties of the current shader stage
The implementation introduced in "tgsi_to_nir: be careful about not
losing any TGSI properties silently (v2)" updates all the TGSI properties,
but it didn't take into account that the shader_info structure uses a union
to store the different attributes for each shader stage.
Now we only update the attributes if they affect current shader stage,
avoiding to overwrite members of the union that should be overwritten.
This has created hundreds of regressions in v3d.
For example the TGSI_PROPERTY_VS_BLIT_SGPRS_AMD was overwritting the
same position used by TGSI_PROPERY_CS_FIXED_BLOCK_DEPTH.
Fixes: e3003651978 ("tgsi_to_nir: be careful about not losing any TGSI properties silently (v2)")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Samuel Pitoiset [Mon, 19 Aug 2019 13:01:21 +0000 (15:01 +0200)]
radv/gfx10: do not emit PA_SC_TILE_STEERING_OVERRIDE twice
CLEAR_STATE emits it for us.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Samuel Pitoiset [Mon, 19 Aug 2019 13:01:20 +0000 (15:01 +0200)]
radv: do not emit PKT3_CONTEXT_CONTROL with AMDGPU 3.6.0+
It's emitted by the kernel.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Gert Wollny [Mon, 19 Aug 2019 15:41:46 +0000 (17:41 +0200)]
mesa/program: Take ARB_framebuffers_no_attachments into account in wpos correction
If a drawbuffer is an fbo without an attachment then its 'Height' will be zero,
and we have to take its 'DefaultGeometry.Height' into account.
Fixes on softpipe (with the exception of tests that use multisample):
dEQP-GLES31.functional.fbo.no_attachments.*
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Sagar Ghuge [Tue, 16 Jul 2019 18:29:47 +0000 (11:29 -0700)]
iris: Enable non coherent framebuffer fetch on broadwell
v2: Use GEN_GEN in iris_state (Kenneth Graunke)
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Sagar Ghuge [Wed, 7 Aug 2019 21:00:14 +0000 (14:00 -0700)]
iris: Free resource if failed to allocate surface state
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Sagar Ghuge [Wed, 7 Aug 2019 17:38:17 +0000 (10:38 -0700)]
iris: Pass isl_surf to fill_surface_state
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Suggested-by: Kenneth Graunke <kenneth@whitecape.org>
Sagar Ghuge [Tue, 16 Jul 2019 18:08:28 +0000 (11:08 -0700)]
iris: Add infrastructure to support non coherent framebuffer fetch
Create separate SURFACE_STATE for render target read in order to support
non coherent framebuffer fetch on broadwell.
Also we need to resolve framebuffer in order to support CCS_D.
v2: Add outputs_read check (Kenneth Graunke)
v3: 1) Import Curro's comment from get_isl_surf
2) Rename get_isl_surf method
3) Clean up allocation in case of failure
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Sagar Ghuge [Tue, 16 Jul 2019 17:48:19 +0000 (10:48 -0700)]
iris: Add helper functions to get tile offset
All helper functions are ported from i965 driver.
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Sagar Ghuge [Tue, 16 Jul 2019 17:38:00 +0000 (10:38 -0700)]
iris: Add helper function to get isl dim layout
v2: Add missing space (Caio)
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>