Jason Ekstrand [Sun, 4 Dec 2016 01:15:42 +0000 (17:15 -0800)]
intel/fs: Lower large local arrays to scratch
Shader-db results on Kaby Lake:
total instructions in shared programs:
14929212 ->
14880028 (-0.33%)
instructions in affected programs: 72428 -> 23244 (-67.91%)
helped: 6
HURT: 2
helped stats (abs) min: 2165 max: 15981 x̄: 8590.00 x̃: 7624
helped stats (rel) min: 56.06% max: 74.52% x̄: 67.55% x̃: 72.08%
HURT stats (abs) min: 1178 max: 1178 x̄: 1178.00 x̃: 1178
HURT stats (rel) min: 350.60% max: 361.35% x̄: 355.97% x̃: 355.97%
95% mean confidence interval for instructions value: -11947.03 -348.97
95% mean confidence interval for instructions %-change: -125.72% 202.37%
Inconclusive result (%-change mean confidence interval includes 0).
total cycles in shared programs:
368585300 ->
342557344 (-7.06%)
cycles in affected programs:
28144921 ->
2116965 (-92.48%)
helped: 6
HURT: 2
helped stats (abs) min:
1404978 max:
7766106 x̄:
4353922.00 x̃:
3890682
helped stats (rel) min: 82.01% max: 95.57% x̄: 89.95% x̃: 92.28%
HURT stats (abs) min: 47778 max: 47798 x̄: 47788.00 x̃: 47788
HURT stats (rel) min: 278.20% max: 282.98% x̄: 280.59% x̃: 280.59%
95% mean confidence interval for cycles value: -
5900438.73 -606550.27
95% mean confidence interval for cycles %-change: -140.79% 146.16%
Inconclusive result (%-change mean confidence interval includes 0).
total spills in shared programs: 9243 -> 8901 (-3.70%)
spills in affected programs: 2718 -> 2376 (-12.58%)
helped: 4
HURT: 4
total fills in shared programs: 21831 -> 10141 (-53.55%)
fills in affected programs: 11804 -> 114 (-99.03%)
helped: 6
HURT: 2
total sends in shared programs: 815912 -> 815912 (0.00%)
sends in affected programs: 0 -> 0
helped: 0
HURT: 0
LOST: 1
GAINED: 3
The helped shaders are all compute shaders in Aztec Ruins. There is
also a compute shader in synmark2 OglCSDof that's helped but it doesn't
show up in above shader-db results because it went from SIMD8 to SIMD16.
That shader improves enough to yield an 15-20% performance boost to the
benchmark as a whole on my KBL laptop. The hurt shaders are a couple
shaders in Kerbal Space Program and a couple in Aztec Ruins.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Thu, 28 Feb 2019 14:15:30 +0000 (08:15 -0600)]
intel/fs: Implement the new load/store_scratch intrinsics
This commit fills in a number of different pieces:
1. We add support to brw_nir_lower_mem_access_bit_sizes to handle the
new intrinsics. This involves simple plumbing work as well as a
tiny bit of extra logic to always scalarize scratch intrinsics
2. Add code to brw_fs_nir.cpp to turn nir_load/store_scratch intrinsics
into byte/dword scattered read/write messages which use the A32
stateless model.
3. Add code to lower_surface_logical_send to handle dword scattered
messages and the A32 stateless model.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Thu, 28 Feb 2019 16:02:03 +0000 (10:02 -0600)]
intel/nir: Plumb devinfo through lower_mem_access_bit_sizes
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Thu, 28 Feb 2019 16:26:33 +0000 (10:26 -0600)]
intel/fs: refactor surface header setup
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Wed, 8 Apr 2015 09:41:33 +0000 (02:41 -0700)]
intel/fs: Add DWord scattered read/write opcodes
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Wed, 6 Nov 2019 18:36:28 +0000 (12:36 -0600)]
intel/nir: Use nir_extract_bits in lower_mem_access_bit_sizes
The new helper solves most of the annoying problems with data wrangling
in brw_nir_lower_mem_access_bit_sizes.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Sat, 9 Nov 2019 01:24:05 +0000 (19:24 -0600)]
nir: Add tests for nir_extract_bits
Jason Ekstrand [Wed, 6 Nov 2019 18:09:56 +0000 (12:09 -0600)]
nir/builder: Add a nir_extract_bits helper
This new helper is better than nir_bitcast_vector because it's able to
take a (mostly) arbitrary range from the source vector. The only
requirement is that first_bit has to be aligned to the smaller of the
two bit sizes. It wouldn't be hard to lift that requirement but it's
reasonable for now.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Eric Engestrom [Tue, 10 Sep 2019 16:06:09 +0000 (17:06 +0100)]
egl: fix _EGL_NATIVE_PLATFORM fallback
When the X11 or Haiku platforms were compiled in, they would bypass the
`_EGL_NATIVE_PLATFORM` fallback by always returning themselves instead.
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Ricardo Garcia [Thu, 7 Nov 2019 14:38:45 +0000 (15:38 +0100)]
anv: Unify GetDeviceQueue and GetDeviceQueue2
Avoid duplicating some checks and code by making anv_GetDeviceQueue a
subcase of anv_GetDeviceQueue2, like radv does.
Signed-off-by: Ricardo Garcia <rgarcia@igalia.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Alyssa Rosenzweig [Thu, 7 Nov 2019 02:48:33 +0000 (21:48 -0500)]
panfrost: Select format-specific blending intrinsics
If we have an accelerated path for a particular framebuffer format,
let's use it to save a bunch of instructions in a blend shader.
[Tomeu: Only use the faster intrinsic on >T760]
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Alyssa Rosenzweig [Thu, 7 Nov 2019 13:25:27 +0000 (08:25 -0500)]
pan/midgard: Pack load/store masks
While most load/store operations on 32-bit/vec4 intriniscally, some are
not and have special type-size-dependent semantics for the mask. We need
to convert into this native format.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Alyssa Rosenzweig [Thu, 7 Nov 2019 02:50:32 +0000 (21:50 -0500)]
pan/midgard: Implement nir_intrinsic_load_output_u8_as_fp16_pan
We can use the native Midgard ops for this, depending what chip we're
on.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Alyssa Rosenzweig [Thu, 7 Nov 2019 02:49:35 +0000 (21:49 -0500)]
pan/midgard: Identify ld_color_buffer_u8_as_fp16*
There are two versions of this opcode, depending what version of the ISA
you're using. I'm not sure if there's a semantic difference; I think
there might be some slight subtleties but it's too early to know at this
stage.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Alyssa Rosenzweig [Thu, 7 Nov 2019 02:47:23 +0000 (21:47 -0500)]
nir: Add load_output_u8_as_fp16_pan intrinsic
This is a single opcode, at least on newer Midgard chips. It's easier to
have this represented in NIR rather than trying to optimize out the
conversions, so let's add the intrinsic.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Tomeu Vizoso [Wed, 6 Nov 2019 09:04:36 +0000 (10:04 +0100)]
panfrost: Set depth and stencil for SFBD based on the format
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Erik Faye-Lund [Fri, 8 Nov 2019 11:22:00 +0000 (12:22 +0100)]
zink: correct depth-stencil format
When using packed vulkan-formats on little-endian systems, we need to
swap the components for the gallium formats. And since Zink isn't
big-endian safe yet, little-endian is the only endianess we care about
right now.
This fixes a bunch of piglit tests, amongs others:
- spec@arb_depth_texture@depth-level-clamp
- spec@arb_depth_texture@depthstencil-render-miplevels * d=z24
- spec@arb_depth_texture@fbo-depth-gl_depth_component24-blit
- spec@arb_depth_texture@fbo-depth-gl_depth_component24-copypixels
- spec@arb_depth_texture@fbo-depth-gl_depth_component24-drawpixels
- spec@arb_depth_texture@fbo-depth-gl_depth_component24-readpixels
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Fixes: 8d46e35d16e ("zink: introduce opengl over vulkan")
Erik Faye-Lund [Wed, 6 Nov 2019 14:13:58 +0000 (15:13 +0100)]
zink/spirv: add support for nir_op_flrp
This fixes the following piglit:
spec@ati_fragment_shader@ati_fragment_shader-render-fog
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Chris Wilson [Thu, 31 Oct 2019 07:29:55 +0000 (07:29 +0000)]
egl: Mention if swrast is being forced
The system can be disabling HW acceleration unbeknown to the user,
leading to a long debug session trying to work out which component is
failing. A quick mention that it is the environment override would be
very useful.
v2: Use more generic "CPU renderer" and so try to avoid jargon.
Reviewed-By: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Martin Peres <martin.peres@linux.intel.com>
Jason Ekstrand [Thu, 26 Sep 2019 16:56:48 +0000 (11:56 -0500)]
spirv: Sort out the mess that is sampled image
This commit makes two major changes. First, we add a second case to
OpLoad for sampled images which constructs a vtn_sampled_image and
stashes that rather than stashing a pointer to the combined image
sampler like we do for bare samplers and images. This should be more in
line with how SPIR-V is intended to work and hopefully doesn't cause any
weird problems. The second is a rework of vtn_handle_texture to assume
that everything has an image but not everything has a sampler. We also
add a vtn_fail_if for the case where a texture instructions require a
sampler but none is provided.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Mon, 4 Nov 2019 22:44:30 +0000 (16:44 -0600)]
spirv: Add a vtn_decorate_pointer helper
This helper makes a duplicate copy of the pointer if any new access
flags are set at this stage. This way we don't end up propagating
access flags further than they actual SPIR-V decorations. In several
instances where we create new pointers, we still call the decoration
helper directly because no copy is needed.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Jason Ekstrand [Thu, 26 Sep 2019 16:48:44 +0000 (11:48 -0500)]
spirv: Remove the type from sampled_image
We have types on all vtn_values at this point so there's no reason to
carry the redundant type information.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Rob Clark [Mon, 4 Nov 2019 19:41:55 +0000 (11:41 -0800)]
freedreno/ir3: also track # of nops for shader-db
The instruction count is (mostly) a measure of what optimization passes
can do, while # of nops is more an indication of how effectively the
scheduler is balancing register pressure vs instruction count. So track
these independently.
(There could be opportunities to rematerialize values to reduce register
pressure, swapping some nop's with other alu instructions, so nothing is
truely independent.. but it is still useful to break these stats out.)
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Mon, 4 Nov 2019 19:33:54 +0000 (11:33 -0800)]
freedreno/ir3: sync disasm changes from envytools
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Fri, 25 Oct 2019 20:57:49 +0000 (13:57 -0700)]
freedreno/a4xx: fix SP_FS_MRT_REG.HALF_PRECISION
Set flag based on actual output reg type.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Fri, 25 Oct 2019 20:56:30 +0000 (13:56 -0700)]
freedreno/a3xx: fix SP_FS_MRT_REG.HALF_PRECISION
We should really be setting this based on the actual output register
type.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Thu, 24 Oct 2019 18:24:15 +0000 (11:24 -0700)]
freedreno/ir3: remove obsolete comment
The meta PHI instruction was removed long ago. And fanin/fanout
themselves to not contribute actual instructions (at least not by the
time you get to sched, they may prevent copy-propagating away a mov)
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Wed, 23 Oct 2019 22:04:38 +0000 (15:04 -0700)]
freedreno/ir3/ra: remove ir print after livein/out
The IR hasn't changed at this point, so it isn't really adding any
value.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Wed, 23 Oct 2019 16:50:22 +0000 (09:50 -0700)]
freedreno/ir3/ra: move regs_count==0 check
Fold it in to writes_gpr() (since a register that does not reference any
registers by definition does not write a register). This lets us avoid
having to handle this case in a few other places.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Tue, 22 Oct 2019 18:01:11 +0000 (11:01 -0700)]
freedreno/ir3: ir3_print tweaks
Handle HALF/HIGH flags in all cases, and colorize SSA src notation.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Tue, 22 Oct 2019 17:28:04 +0000 (10:28 -0700)]
freedreno/ir3: use SSA flag on dest register too
We did this in some places before, but not consistantly. But it will be
useful for two-pass RA, to identify which registers have already been
assigned.
While we are cleaning this up, use __ssa_src() and new __ssa_dst()
helper more consistently. (If nothing else, this reduces the # of
callers of ir3_reg_create() to audit that we didn't miss something)
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Tue, 22 Oct 2019 16:22:58 +0000 (09:22 -0700)]
freedreno/ir3: split pre-coloring to it's own function
Signed-off-by: Rob Clark <robdclark@chromium.org>
Caio Marcelo de Oliveira Filho [Fri, 8 Nov 2019 23:58:15 +0000 (15:58 -0800)]
spirv: Don't leak GS initialization to other stages
The stage specific fields of shader_info are in an union. We've
likely been lucky that this value was either overwritten or ignored by
other stages. The recent change in shader_info layout in commit
84a1a2578da ("compiler: pack shader_info from 160 bytes to 96 bytes")
made this issue visible.
Fixes: cf2257069cb ("nir/spirv: Set a default number of invocations for geometry shaders")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Marek Olšák [Wed, 6 Nov 2019 00:10:09 +0000 (19:10 -0500)]
compiler: pack shader_info from 160 bytes to 96 bytes
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Marek Olšák [Wed, 6 Nov 2019 00:20:35 +0000 (19:20 -0500)]
glsl/linker: pass shader_info to analyze_clip_cull_usage directly
This will be needed by the next commit.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Marek Olšák [Fri, 8 Nov 2019 21:31:35 +0000 (16:31 -0500)]
radeonsi/nir: fix compute shader crash due to nir_binary == NULL
This partially reverts
8b30114dda8.
Fixes: 8b30114dda8 "radeonsi/nir: call nir_serialize only once per shader"
Marek Olšák [Thu, 7 Nov 2019 23:43:07 +0000 (18:43 -0500)]
radeonsi/nir: call nir_serialize only once per shader
We were calling it twice.
First serialize it, then use it to compute the cache key.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Marek Olšák [Fri, 8 Nov 2019 00:10:55 +0000 (19:10 -0500)]
util: add blob_finish_get_buffer
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Eric Anholt [Thu, 7 Nov 2019 23:24:05 +0000 (15:24 -0800)]
u_format: Fix swizzle of A1R5G5B5.
Found once I started using the generated unpack code from the Mesa side.
Fixes: 4bbaac3782ad ("gallium: Add some more channel orderings of packed formats.")
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
David Stevens [Wed, 23 Oct 2019 02:06:17 +0000 (11:06 +0900)]
virgl: support emulating planar image sampling
Mesa emulates planar format sampling with per-plane samplers. Virgl now
supports this by allowing the plane index to be passed when creating a
sampler view from a planar image. With this change, mesa now passes that
information to virgl.
Signed-off-by: David Stevens <stevensd@chromium.org>
Reviewed-by: Lepton Wu <lepton@chromium.org>
Krzysztof Raszkowski [Fri, 8 Nov 2019 16:04:47 +0000 (16:04 +0000)]
gallium/swr: Enable some ARB_gpu_shader5 extensions
Enable / add to features.txt:
- Enhanced textureGather.
- Geometry shader instancing.
- Geometry shader multiple streams.
Reviewed-by: Jan Zielinski <jan.zielinski@intel.com>
Krzysztof Raszkowski [Fri, 8 Nov 2019 14:52:16 +0000 (14:52 +0000)]
gallium/swr: Fix GS invocation issues
- Fixed proper setting gl_InvocationID.
- Fixed GS vertices output memory overflow.
Reviewed-by: Jan Zielinski <jan.zielinski@intel.com>
Timur Kristóf [Wed, 6 Nov 2019 12:29:26 +0000 (13:29 +0100)]
ac: Handle invalid GFX10 format correctly in ac_get_tbuffer_format.
It happens that some games try to access a vertex buffer without
a valid format. This case was incorrectly handled by
ac_get_tbuffer_format which made ACO emit an invalid instruction.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Cc: 19.3 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Boris Brezillon [Thu, 7 Nov 2019 08:32:31 +0000 (09:32 +0100)]
panfrost: Try to evict unused BOs from the cache
The panfrost BO cache can only grow since all newly allocated BOs are
returned to the cache (unless they've been exported).
With the MADVISE ioctl that's not a big issue because the kernel can
come and reclaim this memory, but MADVISE will only be available on 5.4
kernels. This means an app can currently allocate a lot memory without
ever releasing it, leading to some situations where the OOM-killer kicks
in and kills the app (or even worse, kills another process consuming
more memory than the GL app) to get some of this memory back.
Let's try to limit the amount of BOs we keep in the cache by evicting
entries that have not been used for more than one second (if the app
stopped allocating BOs of this size, it's likely to not allocate
similar BOs in a near future).
This solution is based on the VC4/V3D implementation.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Boris Brezillon [Thu, 7 Nov 2019 07:42:09 +0000 (08:42 +0100)]
panfrost: Move BO cache related fields to a sub-struct
We will soon introduce an LRU list to evict BOs that have been unused
for more than 1 second. Let's first move all BO cache fields to a
sub-struct to clarify which fields are used by the BO caching logic.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 7 Nov 2019 14:31:29 +0000 (09:31 -0500)]
pan/midgard: Switch base for vertex texturing on T720
There aren't texture pipeline registers anymore; instead, space is
shared with work and ldst registers for output and input respectively.
We need to shift the base registers to represent this correctly.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 7 Nov 2019 14:31:02 +0000 (09:31 -0500)]
pan/midgard: Pass shader stage to disassembler
Vertex texturing behaves differently from fragment texturing on some
GPUs.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 7 Nov 2019 14:20:56 +0000 (09:20 -0500)]
pan/midgard: Disassemble half-steps correctly
The meaning of some bits shifts; we need to account for this to print
swizzles sanely.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 7 Nov 2019 13:56:41 +0000 (08:56 -0500)]
pan/midgard: Fix printing of half-registers in texture ops
We were using old style half-registers; let's update that to be
consistent, preparing us for more disassmbler changes in this area.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Kristian H. Kristensen [Fri, 8 Nov 2019 00:44:33 +0000 (16:44 -0800)]
freedreno/ir3: Use regid() helper when setting up precolor regs
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 02:51:38 +0000 (19:51 -0700)]
freedreno/a6xx: Turn on tessellation shaders
Wow. Very triangle. So shader.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Tue, 29 Oct 2019 23:26:34 +0000 (16:26 -0700)]
freedreno/a6xx: Only use merged regs and four quads for VS+FS
When other geometry stages are present, we chose two quads and no
merged regs.
Acked-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Tue, 22 Oct 2019 23:26:11 +0000 (16:26 -0700)]
freedreno/blitter: Save tessellation state
We have tessellation state now.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Fri, 8 Nov 2019 00:32:24 +0000 (16:32 -0800)]
freedreno/a6xx: Only set emit.hs/ds when we're drawing patches
At least the gallium blitter helper will call us to draw with
tessellation shaders set but a non-patch primitive.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 02:49:47 +0000 (19:49 -0700)]
freedreno: Use bypass rendering for tessellation
It seems like tiling could work in the Adreno architecture, but we've
only ever seen bypass rendering with tessellation. For now, let's do
that too.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 03:03:07 +0000 (20:03 -0700)]
freedreno/a6xx: Program state for tessellation stages
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 02:58:59 +0000 (19:58 -0700)]
freedreno/a6xx: Emit constant parameters for tessellation stages
Assemble the information the stages need and emit the constants.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 03:05:47 +0000 (20:05 -0700)]
freedreno/a6xx: Allocate and program tessellation buffer
Tessellation needs a couple of buffers that should hold the entire
output from a full VS+TCS draw call.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 02:47:50 +0000 (19:47 -0700)]
freedreno/a6xx: Build the right draw command for tessellation
We need to select the right primitive type, set a bit to turn on
tessellation and or in the TES output primitive type.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 02:44:42 +0000 (19:44 -0700)]
freedreno/ir3: Allocate const space for tessellation parameters
The tessellation stages need size and stride or the patch layout as
well as locations of attributes in the patch. The tesselation stages
also use two system memory BOs and need the iovas of those.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 02:39:47 +0000 (19:39 -0700)]
freedreno/ir3: Pre-color TCS header and primitive ID inputs
Similar to GS, the registers are shared and not reinitialized betewen
VS and TCS, so we need to make sure to allocate the same registers for
the system values between stages.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 03:21:13 +0000 (20:21 -0700)]
freedreno/ir3: Don't assume binning shader is always VS
In tessellation mode, the TES is (probably) the binning shader.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 02:37:53 +0000 (19:37 -0700)]
freedreno/ir3: Setup inputs and outputs for tessellation stages
Similar to GS, some inputs are reused when the chsh from VS to TCS or
TES to GS, so we need to make sure we setup the right inputs and make
the shared system values outputs so they don't get clobbered.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 02:30:51 +0000 (19:30 -0700)]
freedreno/ir3: Implement TCS synchronization intrinsics
We add two new IR3 specific nir intrinsics that map to the new condend
and endpatch instructions.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 02:29:07 +0000 (19:29 -0700)]
freedreno/ir3: Implement tess coord intrinsic
Our lowering pass made the z component unused by replacing its uses
by 1 - x - y. The intrinsic implementation then just need to return
the x and y components.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 02:26:30 +0000 (19:26 -0700)]
freedreno/ir3: End TES with chsh when using GS
When we have both TES and GS, the TES needs to chain to the VS with
chmask and chsh GS just like the VS does to either TCS or GS.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 01:19:50 +0000 (18:19 -0700)]
freedreno/ir3: Add new synchronization opcodes
There are two new opcodes in use in tesselation control shaders:
category 0, opcodes 13 and 15. unk13 is a kill type of instruction
that terminates threads where !p0.x and it used to narrow down a patch
wavefront to just thread 0. Then, once thread 0 has written the tess
levels, it issues unk15, which might signal the TE that another patch
has been fully written.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 00:30:48 +0000 (17:30 -0700)]
freedreno/ir3: Extend geometry lowering pass to handle tessellation
VS and TCS pass varyings the same way as VS and GS does. TCS then
writes entire patch to a system memory BO and TES eventually reads
back from the BO once the TE starts generating vertices. TES outputs
vertices the same way as VS and GS, except when there's a GS as well,
in which case TES passes varyings to GS same way the VS would.
In addition, the TCS needs a little bit of control flow massaging so
that it only runs for valid invocations needs a couple of unknown
instructions to synchronize with the TE.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Wed, 23 Oct 2019 00:16:09 +0000 (17:16 -0700)]
freedreno/ir3: Add tessellation field to shader key
Whether we're tessellating and which primitives the TES outputs
affects the entire pipeline so let's add a field to the key to track
that.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Tue, 22 Oct 2019 23:37:35 +0000 (16:37 -0700)]
freedreno/ir3: Use imul24 in offset calculations
With the imul24 opcode in place, we can now use it for computing local
offsets (ie for ldlw/stlw).
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Tue, 22 Oct 2019 23:33:18 +0000 (16:33 -0700)]
freedreno/ir3: Add ir3 intrinsics for tessellation
These provide the iovas for system memory buffers used for
tessellation as well as a new HW specific system value.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Tue, 22 Oct 2019 23:24:26 +0000 (16:24 -0700)]
freedreno: Don't count primitives for patches
The gallium helper doesn't like patches and we can't determine how
many primitives it gets tessellated into anyway. On gens where we
have tessellation, we get the prim count from a HW counter so just
skip counting on the CPU.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Tue, 22 Oct 2019 23:16:35 +0000 (16:16 -0700)]
freedreno/ir3: Add load and store intrinsics for global io
These intrinsics take a ivec2 for the 64 bit base address and a
integer offset.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Tue, 22 Oct 2019 23:07:45 +0000 (16:07 -0700)]
freedreno/ir3: Emit link map as byte or dwords offsets as needed
Stages that load inputs with ldlw (TCS, GS) need byte offsets, stages
that load with ldg (TES) need dwords offsets.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Tue, 22 Oct 2019 23:03:36 +0000 (16:03 -0700)]
freedreno/a6xx: Add register offset for STG/LDG
These instructions take a 64 bit iova as two conescutive registers and
a immediate offset. This patch adds support for the offset to be a
single register, which is added to the 64 bit iova.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Thu, 31 Oct 2019 21:43:58 +0000 (14:43 -0700)]
freedreno/a6x: Rename z/s formats
What we call eRB6_Z24_UNORM_S8_UINT now is actually
RB6_Z24_UNORM_S8_UINT_AS_R8G8B8A8 and RB6_X8Z24_UNORM is actually
RB6_Z24_UNORM_S8_UINT.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Fri, 1 Nov 2019 22:16:44 +0000 (15:16 -0700)]
freedreno/a6xx: Fix layered texture type enum
2D array textures and 3D textures are different enum values after all.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Thu, 31 Oct 2019 21:21:32 +0000 (14:21 -0700)]
freedreno: Add nogmem debug option to force bypass rendering
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Fri, 20 Sep 2019 00:13:34 +0000 (17:13 -0700)]
freedreno/a6xx: Clear sysmem with CP_BLIT
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Thu, 31 Oct 2019 00:18:48 +0000 (17:18 -0700)]
freedreno/a6xx: Fix primitive counters again
We use one mechanism for (REG_A6XX_RBBM_PRIMCTR_8_LO)
PIPE_QUERY_PRIMITIVES_GENERATED, which counts all primitives that exit
the geometry pipeline, whether or not xfb is on. Then for
PIPE_QUERY_PRIMITIVES_EMITTED, we use the CP_EVENT_WRITE subfunction
that writes out per-stream counts for generated and emitted, but only
when xfb is enabled.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Thu, 31 Oct 2019 17:01:00 +0000 (10:01 -0700)]
freedreno/registers: Add comments about primitive counters
Adding comments about best guess at what the counters count.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Thu, 31 Oct 2019 17:02:12 +0000 (10:02 -0700)]
freedreno/registers: Move SP_PRIMITIVE_CNTL and SP_VS_VPC_DST
Move these two to be in order with the other VS regs.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Kristian H. Kristensen [Tue, 29 Oct 2019 19:19:28 +0000 (12:19 -0700)]
freedreno/registers: Fix typo
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Rhys Perry [Tue, 29 Oct 2019 13:59:59 +0000 (13:59 +0000)]
aco: add Instruction::usesModifiers() and add more checks in the optimizer
No pipeline-db changes.
v2: use early-exit for VOP3
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev> (v1)
Rhys Perry [Tue, 15 Oct 2019 19:43:39 +0000 (20:43 +0100)]
radv: adjust loop unrolling heuristics for int64
In particular, increase the cost of 64-bit integer division.
Fixes huge shaders with dEQP-VK.spirv_assembly.type.scalar.i64.mod_geom
, with ACO used for GS this creates shaders requiring a branch with
>32767 dword offset.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Erico Nunes [Mon, 28 Oct 2019 19:02:38 +0000 (20:02 +0100)]
lima: fix bo submit memory leak
Fix memory leak on allocation for lima submit, reported by valgrind.
128 bytes in 1 blocks are definitely lost in loss record 38 of 84
at 0x484A6E8: realloc (in /usr/lib/valgrind/vgpreload_memcheck-arm64-linux.so)
by 0x58689C7: util_dynarray_ensure_cap (u_dynarray.h:91)
by 0x5868BBB: util_dynarray_grow_bytes (u_dynarray.h:139)
by 0x5868BBB: lima_submit_add_bo (lima_submit.c:113)
by 0x585D7D3: lima_ctx_buff_va (lima_context.c:57)
by 0x586378F: lima_pack_plbu_cmd (lima_draw.c:802)
by 0x586378F: lima_draw_vbo (lima_draw.c:1351)
by 0x5406A2F: u_vbuf_draw_vbo (u_vbuf.c:1184)
by 0x55D0A57: st_draw_vbo (st_draw.c:268)
by 0x55576CB: _mesa_draw_arrays (draw.c:374)
by 0x55576CB: _mesa_draw_arrays (draw.c:351)
by 0x43610B: Mesh::render_vbo() (mesh.cpp:583)
by 0x415DBB: SceneBuild::draw() (scene-build.cpp:242)
by 0x41131B: MainLoop::draw() (main-loop.cpp:133)
by 0x411947: MainLoop::step() (main-loop.cpp:108)
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Erico Nunes [Mon, 28 Oct 2019 18:59:54 +0000 (19:59 +0100)]
lima: fix nir shader memory leak
Fix memory leak on allocation for nir shader, reported by valgrind.
3,502 (480 direct, 3,022 indirect) bytes in 1 blocks are definitely lost in loss record 77 of 84
at 0x48483F8: malloc (in /usr/lib/valgrind/vgpreload_memcheck-arm64-linux.so)
by 0x5750817: ralloc_size (ralloc.c:119)
by 0x5750977: rzalloc_size (ralloc.c:151)
by 0x575C173: nir_shader_create (nir.c:45)
by 0x5763ACB: nir_shader_clone (nir_clone.c:728)
by 0x55D5003: st_create_fp_variant (st_program.c:1242)
by 0x55D789F: st_get_fp_variant (st_program.c:1522)
by 0x55D789F: st_get_fp_variant (st_program.c:1507)
by 0x56400C3: st_update_fp (st_atom_shader.c:163)
by 0x563D333: st_validate_state (st_atom.c:261)
by 0x55D07CB: prepare_draw (st_draw.c:132)
by 0x55D08DF: st_draw_vbo (st_draw.c:184)
by 0x55576CB: _mesa_draw_arrays (draw.c:374)
by 0x55576CB: _mesa_draw_arrays (draw.c:351)
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Prodea Alexandru-Liviu [Thu, 7 Nov 2019 22:04:50 +0000 (22:04 +0000)]
Meson: Remove lib prefix from graw and osmesa when building with Mingw.
Also remove version sufix from osmesa swrast on Windows.
v2: Make sure we don't remove lib prefix on *nix platforms.
Signed-off-by: Prodea Alexandru-Liviu <liviuprodea@yahoo.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Cc: "19.3" <mesa-stable@lists.freedesktop.org>
Marek Olšák [Wed, 6 Nov 2019 23:08:22 +0000 (18:08 -0500)]
mesa: expose SPIR-V extensions in the Compatibility profile too
We would like to have GL 4.6 Compatibility too.
The extensions don't support compatibility features, so no other changes
are needed.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Drew DeVault [Thu, 7 Nov 2019 17:47:28 +0000 (12:47 -0500)]
st_get_external_sampler_key: improve error message
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Eric Anholt [Fri, 6 Sep 2019 21:33:18 +0000 (14:33 -0700)]
mesa/st: Make st_pipe_format_to_mesa_format an effective no-op.
All callers other than the unit test just wanted to convert back from
a known-mesa-equivalent format, which is now a no-op.
v2: Fix assertion failure in iris GL startup with BGR565 by continuing
to return MESA_FORMAT_NONE for non-Mesa formats.
Reviewed-by: Marek Olšák <marek.olsak@amd.com> (v1)
Eric Anholt [Fri, 6 Sep 2019 21:29:10 +0000 (14:29 -0700)]
mesa/st: Gut most of st_mesa_format_to_pipe_format().
Now that MESA_FORMAT_x is just a PIPE_FORMAT_x define, we can strip
this function down to just the compression fallbacks.
v2: Restore the SRGB format for ASTC SRGB fallback case.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Eric Anholt [Fri, 6 Sep 2019 20:26:15 +0000 (13:26 -0700)]
mesa: Redefine MESA_FORMAT_* in terms of PIPE_FORMAT_*.
There are various places in Mesa where we would like to be able to
have a shared format enum between Mesa and gallium (NIR compiler's
image formats, for example, or mapping from gallium's formats to
mesa's and vice versa in st_format.c). Rewriting all MESA_FORMAT to
PIPE_FORMAT would be disruptive and possibly more work than it's worth
(And I actually prefer MESA_FORMAT's name scheme), so for now just
make it so that there's one shared set of enum values.
The #defines here were generated by printing out from the
tests/st_format.c round-tripping loop, with the exception of 8888
formats where I hand-edited the #defines to point at the corresponding
gallium packed format define.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Eric Anholt [Fri, 6 Sep 2019 21:09:37 +0000 (14:09 -0700)]
mesa: Prepare for the MESA_FORMAT_* enum to be sparse.
To redefine MESA_FORMAT in terms of PIPE_FORMAT enums, we need to fix
places where we iterated up to MESA_FORMAT_COUNT. I use
_mesa_get_format_name(f) == NULL as the signal that it's not an enum
value with a MESA_FORMAT.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Eric Anholt [Fri, 6 Sep 2019 20:33:56 +0000 (13:33 -0700)]
mesa/st: Test round-tripping of all compressed formats.
We checked round-tripping of formats without fallbacks, but weren't
setting the compression support flags in the mock context and thus
needed to skip testing those. Just set all the flags and assert that
no fallbacks are triggered, so we get full test coverage.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Eric Anholt [Thu, 29 Aug 2019 20:20:45 +0000 (13:20 -0700)]
mesa: Stop defining a full separate format for RGBA_UINT8.
We have packed formats for RGBA and ABGR already, so we can just
pack/unpack code.
v2: Rebase on endianness macro rename
Reviewed-by: Marek Olšák <marek.olsak@amd.com> (v1)
Eric Anholt [Mon, 19 Aug 2019 21:33:51 +0000 (14:33 -0700)]
gallium: Add equivalents of packed MESA_FORMAT_*UINT formats.
These are the last formats that MESA_FORMAT had and PIPE_FORMAT
didn't. The .csv entries channel sizes and swizzles all came from the
corresponding UNORM format.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Eric Anholt [Mon, 19 Aug 2019 21:24:12 +0000 (14:24 -0700)]
gallium: Add an equivalent of MESA_FORMAT_BGR_UNORM8.
This is the last unorm format that MESA_FORMAT had and PIPE_FORMAT
didn't. Note that it's an array format on gallium's side as well,
since it's a NPOT pixel size.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Eric Anholt [Mon, 19 Aug 2019 20:58:23 +0000 (13:58 -0700)]
gallium: Add some more channel orderings of packed formats.
This covers everything that MESA_FORMAT had for packed unorm.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Eric Anholt [Mon, 1 Jul 2019 19:39:55 +0000 (12:39 -0700)]
gallium: Add defines for FXT1 texture compression.
This texture compression is exposed by 830 and 915, and to make
MESA_FORMAT match PIPE_FORMAT defines I need a corresponding
PIPE_FORMAT.
v2: Set is_hand_written so we don't try to generate pack/unpack code.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>