"softpin" mode was introduced in the same kernel as the 'DUMP' flag. So
if we are using the legacy non-softpin path, clear the dump flag. OTOH
the 'DUMP' flag isn't quite so needed on older kernels, since we would
get all cmdstream, even SDS stateobjs, dumped regardless, as they would
have cmd table entries.
Fixes: b2c23b1e48 ("freedreno: Mark all ringbuffer BOs as to be dumped on crash.")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Tested-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5081>
A3xx apparently has higher alignment requirements than later gens for
indirect const uploads. It also has fewer of them. Add compiler
parameters for both settings, and set accordingly for a3xx and a4xx+.
This fixes all the ubo test failures caused by this optimization.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5077>
This causes failures on a3xx resulting in the non-sensical dEQP failures
on packUnorm2x16. The same test uses ldlv on a4xx+, so just disallow
(sat) on bary.f on all generations.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5074>
This fixes a number of dEQP tests:
dEQP-GLES3.functional.fbo.blit.conversion.r8*
dEQP-GLES3.texture.specification.basic_teximage2d.r8*
and others. The reason why this enum showed up in traces for R8 is that
it was an "upgraded" texture to R8G8.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5073>
Device UUID becomes SHA1('freedreno' + gpu_id).
Driver UUID becomes SHA1(mesa-version + git-head-sha1).
v2: Don't use build_id for driver UUID since it generates different
values for vulkan and gl shared objects. (Kristian)
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4847>
The new files are created under a 'common' folder under 'src/freedreno',
where shared functionality between GL and Vulkan drivers (that is not
registers, layout or compiler) will be placed.
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4847>
Whoops. After fixing dual-source blending, dEQP-VK.pipeline.blend.* all
go from skipped to pass, and fixes a bunch of
dEQP-VK.api.info.format_properties.* tests where blending is required.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5039>
The hardware expects that where MRT0 and MRT1 would normally go are the
dual sources for MRT0, whereas GLSL has an extra "index" parameter that
indicates which source it is. Remap it when handling FS outputs.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5039>
For the piglit drawoverhead case, 5/18 of the objects' relocs were
duplicated. We can dedupe them at object create time (since objects are
long-lived) and avoid repeated relocation work at emit time.
nohw drawoverhead program statechange throughput 2.34082% +/- 0.645832%
(n=10).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5020>
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
It seems less brittle to not assume they are in the same order for src
and dst instructions.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Cleans up a couple spots that were still open-coding this.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
These intrinsics are supposed to map to the underlying hardware
instructions, which don't have wrmask. We use them when we lower
store_output in the geometry pipeline and since store_output gets
lowered to temps, we always see full wrmasks there.
It saves addressing math, but may cause multiple loads to be done and
bcseled due to NIR not giving us good address alignment information
currently. I don't have any workloads I know of using non-const-uploaded
UBOs, so I don't have perf numbers for it
This makes us match the GLES blob's behavior, and turnip (other than being
bindful).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4858>
With the upcoming LDC usage in the GL driver, we don't want to be
uploading descriptors for every UBO when they aren't actually in use.
Trimming NIR's num_ubos will avoid that, and cleans up num_ubo handling
elsewhere right now.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4858>
The HW requires a log2 width/height of the level 0 meta_* size in the
descriptors, making it pretty clear that UBWC mipmapping is all
power-of-two sized. Fixes a bunch of failures in the upcoming unit UBWC
layout unit tests.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4931>
Using texturator on a P3A at 1024x1024, RG8 has log2w/h of 6x7 instead of
R16I/UI's 6x8. The other blockw/h I verified other than cpp=1
(R8/R8I/R8UI didn't use UBWC) and 32 (would need a bigger type).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4931>
Also, rewrite the format decision code so that we correctly decide when
the linear fallback is needed, even if UBWC is disabled. As part of
that, I also moved around some of the code to handle compressed formats
to make sure that copying compressed formats with a linear staging blit
works (this is now possible since we started allowing tiled compressed
textures).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5007>
This is simpler than a full-blown memory reuse mechanism, but is good
enough to make sure that repeatedly doing a copy that requires the
linear staging buffer workaround won't use excessive memory or be slowed
down due to repeated allocations.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5007>
Similar to what we do in postsched. It is useful for pre-RA sched to be
a bit aware of things that would cause syncs. In particular for the tex
fetches, since the vecN src/dst tends to limit postsched's ability to
re-order them.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4923>
If an instruction's only use is as an output, and it increases register
pressure, then try to avoid scheduling it until there are no other
options.
A semi-common pattern is `fragcolN.a = 1.0`, this pushes all these
immed loads to the end of the shader.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4923>
Similar to avoidance of `(ss)` syncs, it turns out to be helpful to
avoid `(sy)` syncs as well. This helps us turn an tex, (sy)alu, tex,
(sy)alu sequence into tex, tex, (sy)alu, alu, which is a big win in
gfxbench gl_fill2.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4923>
Once we schedule an instruction that will require an `(ss)` sync flag,
there is no need to delay any further instructions that consume an
SFU result (until the next SFU instruction is scheduled).
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4923>
It seems for short frag shaders, too much prefetch can be detrimental.
I think what we *really* want to do is decide after pre-RA sched, when
we also know about nop's and what the actual ir3 instruction count is.
But that will require re-working how prefetch lowering works. For now
this is a super crude heuristic to attempt to approximate a good
solution.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4923>
We can no longer assume that `state->ranges[0]` is block 0. It *often*
is, but when we encounter a "real" ubo that we lower to `load_uniform`
before a block 0 `load_ubo`, it could end up another entry in the table.
Resulting in the second pass after gathering ubo ranges, not finding a
valid range. Which results in a `load_ubo` for a thing that is not
actually a ubo making it's way into ir3 frontend. Resulting in grabbing
what we think is a ubo address out of some unrelated const register, and
trying to dereference that. Which as you can imagine, fails in amusing
ways.
Fixes: fc850080ee ("ir3: Rewrite UBO push analysis to support bindless")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4954>
This commit fixes a GS regression introduced in !4562 where
ir3's GS lowering pass was moved from common code (ir3_nir) to
freedreno-specific code (ir3_shader). For GS support in turnip, we
need to add the GS lowering pass back in, this time in tu_shader.
As for the nir_gather_info change, the GS lowering pass has always
introduced a discard_if intrinsic into the GS. Previously, we simply
ran nir_shader_gather_info before GS lowering, but now since we lower
the GS before we need to remove the assertion that only a FS can use
the discard_if intrinsic.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4892>
It seems this is actually a "minimum pitch" value. For example
TFETCH6_2_BYTE means a minimum pitch of 128 bytes for mipmap levels.
This fixes breakage with compressed formats. For example this test:
dEQP-VK.pipeline.sampler.view_type.2d.format.eac_r11_snorm_block.mipmap.linear.lod.equal_min_3_max_3
Fixes: a34b3fa198 ("freedreno/fdl: Align after dividing by block size")
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5009>
The nir_analyze_ubo_ranges pass removes all UBO block 0 loads to reverse
what nir_lower_uniforms_to_ubo() had done, and we only upload UBO pointers
to the HW for UBO block 1-N, so let's just fix up the shader state.
Fixes an off by one in const state layout setup, and some really dodgy
register addressing trying to deal with dynamic UBO indices when the UBO
pointers happen to be at the start of the constbuf.
There's no fixes tag, though this fixes a bug from September, because it
would require the num_ubos fix in nir_lower_uniforms_to_ubo.
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4992>
Using non-write flags is pretty dubious -- it means the kernel tracking an
array of read-only consumers of the BO and having exclusive consumers wait
on each reader's fence. It allows multiple readers through dma-bufs to do
work in parallel, but at the cost of kernel CPU time and memory management
of the shared array. Other drivers have dropped this distinction since
dma-buf sharing is usually producer-consumer, not producer-two-consumers,
and the userspace and kernel space tracking is expensive.
For us, this lets us drop the flags passed in for relocs and tracked in
the ringbuffer reloc lists. The end result of the flags reduction work is
drawoverhead uniforms test throughput 2.37195% +/- 0.365579% (n=15)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4967>
It's silly to have all the reloc emitters passing around FD_RELOC_READ
when you have to have it set on all relocs (that don't include WRITE,
which implies read) for the kernel to actually track the fences on the BO.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4967>
Fixes the following building errors:
In file included from external/mesa/src/gallium/drivers/freedreno/a6xx/fd6_blitter.c:40:
external/mesa/src/gallium/drivers/freedreno/a6xx/fd6_pack.h:42:10: fatal error: 'adreno-pm4-pack.xml.h' file not found
^~~~~~~~~~~~~~~~~~~~~~~
1 error generated.
In file included from external/mesa/src/gallium/drivers/freedreno/a6xx/fd6_blend.c:36:
external/mesa/src/gallium/drivers/freedreno/a6xx/fd6_pack.h:42:10: fatal error: 'adreno-pm4-pack.xml.h' file not found
^~~~~~~~~~~~~~~~~~~~~~~
1 error generated.
In file included from external/mesa/src/gallium/drivers/freedreno/a6xx/fd6_const.c:26:
external/mesa/src/gallium/drivers/freedreno/a6xx/fd6_pack.h:42:10: fatal error: 'adreno-pm4-pack.xml.h' file not found
^~~~~~~~~~~~~~~~~~~~~~~
1 error generated.
Fixes: ee293160 "freedreno/a6xx: add OUT_PKT()"
Signed-off-by: Mauro Rossi <issor.oruam@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4973>
The dependency is required to get the necessary generated headers
Fixes the following building error:
In file included from external/mesa/src/freedreno/drm/msm_bo.c:27:
In file included from external/mesa/src/freedreno/drm/msm_priv.h:30:
In file included from external/mesa/src/freedreno/drm/freedreno_priv.h:51:
external/mesa/src/freedreno/drm/freedreno_ringbuffer.h:35:10: fatal error: 'adreno_common.xml.h' file not found
#include "adreno_common.xml.h"
^~~~~~~~~~~~~~~~~~~~~
1 error generated.
Fixes: 6c688ae8 ("freedreno: Deduplicate ringbuffer macros with computerator/fdperf")
Signed-off-by: Mauro Rossi <issor.oruam@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4973>
v1. Replace the existed bool type with new bitfield and edit register
files to take a mask instead of duplicating codes to do masking.
v2. Use fragcoord_compmask != 0 instead of fragcoord_compmask > 0 since
it represents a bitfield.
Tested with
dEQP-VK.glsl.builtin_var.simple.fragcoord_xyz/w
dEQP-GLES2.functional.shaders.builtin_variable.fragcoord_xyz/w
Closes: #2680
Signed-off-by: Hyunjun Ko <zzoon@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4723>
Apparently this is allowed, and the CTS started doing this more often
recently which resulted in frequent hangs running the entire CTS. I
copied the code to create an empty FS from radv.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4928>
For compressed formats, we need to align the number of blocks, not the
logical number of pixels in the texture. Only compressed formats have
block width/height > 1, so we can just unconditionally multiply the
alignment by the block width/height.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4868>
robclark noted that the blob wasn't doing range reduction in the mediump
case, and I confirmed it on
dEQP-GLES3.functional.shaders.operator.angle_and_trigonometry.sin.mediump_float_fragment
vs
dEQP-GLES3.functional.shaders.operator.angle_and_trigonometry.sin.highp_float_fragment.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4893>
These come from the disasm tests, and fix our disasm of blob's
uniform/nonuniform cat6 operands. We also now include human-readable names
for all the modes we know about (though bindless gets distinguished by its
.baseN, like Connor's original disasm).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4857>
With this I also brought in a few new control flow instruction disasm
tests that I'd made back when I wrote the disasm test, but which were too
far from correct to include until now.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4857>
We can remove a bunch of conditional code at key comparison time by
computing a bitmask of used key bits at ir3_shader creation time. This
also gives us a nice place to put additional key simplification to reduce
how many variants we create (like skipping rastflat if we don't read
colors in the FS, or skipping vclamp_color if we don't write colors).
It does mean walking the whole key to AND it, but the key is just 28 bytes
so far so that seems pretty fine.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4562>
We weren't filling in the tess mode of the key, or setting has_gs on GS
shaders, resulting in assertion failures when NIR intrinsics didn't get
lowered.
We have to make a guess at prim mode for TCS, but it should be better to
have some shader-db coverage than none, and it will avoid these failures
happening when we start precompiling shaders.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4562>
Some of the negative API tests make shaders for tess stages that don't do
all the stores they need to. Once we start precompiling (or doing
shader-db of tess), we need to at least not segfault when generating them.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4562>
We were failing to tell the allocator about the restriction that scalar
texture instructions (allocated as scalar regs) couldn't be allocated such
that the start of the full unwritemasked vector started before r0. There
was a patch in select_reg_callback on a6xx that tried to work around that,
but you could still end up backed into a corner you shouldn't be because
we didn't tell the RA what it needed.
Fixes compiler assertion failures on a300-a400's blit_z shader, used for
Z32F gmem blits.
Looks like as a result we get tighter register allocation but more nops:
instructions in affected programs: 757945 -> 760356 (0.32%)
nops in affected programs: 317983 -> 320468 (0.78%)
non-nops in affected programs: 27525 -> 27451 (-0.27%)
mov in affected programs: 3098 -> 3023 (-2.42%)
dwords in affected programs: 109664 -> 110656 (0.90%)
last-baryf in affected programs: 112701 -> 112847 (0.13%)
full in affected programs: 4326 -> 4011 (-7.28%)
sstall in affected programs: 120550 -> 120836 (0.24%)
(ss) in affected programs: 13939 -> 13918 (-0.15%)
(sy) in affected programs: 3006 -> 2786 (-7.32%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4562>
When the GS lowering was working on store_output intrinsics, we had to
clean up the split vars to avoid getting confused. Now that we shadow
the output vars instead, there's no confusion and we can drop this
hack.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4562>
We mostly got away with replacing a store_output with a store_var, but
for complex types like structs, that doesn't work. Once the IO has
been lowered from vars to intrinsic, we've lost the deref chains and
can't properly shadow the outputs.
This commits moves the GS lowering up so we do it before the output
variables get lowered to store_output. This way the pass works much
like nir_lower_io_to_temporaries() and cleanly shadows the outputs.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4562>
This pass lowers per-vertex input intrinsics to load_shared_ir3. This
was open coded in the TCS and GS lowering passes before - this way we
can share it. Furthermore, we'll need to run the rest of the GS
lowering earlier (before lowering IO) so we need to split off this
part that operates on the IO intrinsics first.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4562>
If use NIR's 1-bit bool representation , we get exactly the bool behavior
the hardware provides: CMPS produces true or false, AND/OR/XOR work as
intended without extra absnegs, and we can pass those half values directly
to other CMPS. We emit an absneg for b2b1 ("turn a memory load into a
1-bit NIR boolean"), but we would have done so for the ir3_n2b() on the
use of that value anyway. The most awkward bit is that inot(a@1) is now a
sub(1, a), but we can encode the 1 as an immediate so it's fine.
No significant changes to GL_TIME_ELAPSED on my set of traces (n=21).
instructions in affected programs: 1570638 -> 1548702 (-1.40%)
nops in affected programs: 624053 -> 611381 (-2.03%)
non-nops in affected programs: 959061 -> 949797 (-0.97%)
mov in affected programs: 5258 -> 5252 (-0.11%)
cov in affected programs: 15099 -> 15902 (5.32%)
dwords in affected programs: 469600 -> 452768 (-3.58%)
last-baryf in affected programs: 162211 -> 154726 (-4.61%)
full in affected programs: 4881 -> 4797 (-1.72%)
sstall in affected programs: 173953 -> 174545 (0.34%)
(ss) in affected programs: 10922 -> 10934 (0.11%)
(sy) in affected programs: 728 -> 745 (2.34%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4518>
Similar to OUT_REG(), this has the benefits of:
1. No more messing up pkt size
2. Detects errors of mixing up the order of dwords in the packet
3. Optimizes to more efficient code
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4813>
The existing structure dates back to when this code was part of libdrm,
and we wanted some of this not to be exposed as ABI between libdrm and
mesa. Now that this is no longer a constraint, inline things.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4813>
The new option replaces the two other _split lowering options, since
there's no need for separate options.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4738>
The hardware is capable of automatically filling in certain values in
the VPC without writing them from the last geometry stage, like
gl_PointCoord or gl_PrimitiveID when there is no GS. However, we *do*
have to enable these outputs (i.e. set the VPC_VAR_DISABLE bit to 0) as
VPC_VAR_DISABLE is really about FS inputs rather than VS outputs. To do
this, we move the computation of the enable bits to ir3_link_add(),
which is also a nice refactor anyway. In addition we detect the PrimID
case specifically so that the driver can program the location.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4704>
New blit code doesn't change this value, and different values seem to be
related to the driver version and not the GPU version.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4608>
When we originally wrote spirv_to_nir we didn't have a good scalar value
union to handily use so we rolled our own thing for spec constants. Now
that we have nir_const_value, we can use that and simplify a bunch of
the spec constant logic.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4675>
If you have a struct, the var's base driver location is not the last
driver location that will be accessed in that var. We have a shader
struct member with this number for us, already. Fixes overflows in:
dEQP-GLES31.functional.program_interface_query.program_output.type.interface_blocks.out.named_block_explicit_location.struct.mat3x2
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4670>
It was ignoring the sizes of the output variables and assuming
single-slot, and failing to update num_outputs.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4670>
When I was writing drm-shim, I was focused on the v3d kmsro case -- use my
intel device as the kmsro display device and add on a simulator-based v3d
device that we could render with. But for the noop backends we use for
shader-db, it's a lot more useful to just overwrite the first render node
in the system so that you don't have to pass a -d <how many render nodes I
already have in my system> argument.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4664>
Back in a2xx, HW pitches were in pixels, so storing that was reasonable.
Ever since then, the HW wants pitches in bytes, and we have only one
instance of using pitch in pixels in the code (a3xx sysmem path).
Flip things around so that only a2xx has to worry about the cpp for
looking at pitches.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4558>
In vulkan, offsets are already provided through the api
vkCmdBindTransformFeedbackBuffersEXT, so this is duplicated
calculation.
Fixes : 9ff1959ca5
Signed-off-by: Hyunjun Ko <zzoon@igalia.com>
Reviewed-by: Brian Ho <brian@brkho.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4604>
In case of transform feedback query, it writes two integer values,
which one is for primitives written and another is for primitives
generated.
To handle this, the second member of the struct slot_value is worth
to be presented not as a padding.
In addition, we also need to modify get/copy_result to access both
values.
This patch is the prep work for the transform feedback query support.
Tested with
dEQP-VK.pipeline.timestamp.*
dEQP-VK.query_pool.occlusion_query.*
Signed-off-by: Hyunjun Ko <zzoon@igalia.com>
Reviewed-by: Brian Ho <brian@brkho.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4604>
Passes tests in:
dEQP-VK.pipeline.multisample.sample_locations_ext.*
Note that these tests fail because of gl_PrimitiveID not working correctly:
dEQP-VK.pipeline.multisample.sample_locations_ext.verify_location.*
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4665>
discard is supposed to be a terminator, killing the thread, so that it's
possible to exit main solely by a discard e.g. inside of an infinite
loop. However, it currently isn't treated as a terminator in NIR due to
workarounds turning it into demote (d3d-style kill) and even if that
were fixed, we probably wouldn't want to treat discard_if as a jump
since otherwise the scheduler wouldn't be able to schedule things around
it. So, add this workaround which inserts jump instructions as
necessary to guarantee that the program always terminates.
This fixes a hang in dEQP-VK.graphicsfuzz.while-inside-switch, which
conditionally does a discard inside an infinite loop.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4658>
The first block was being added to the list twice, once here and once in
emit_block(), leading to list corruption and infinite loops when trying
to traverse the list of blocks backwards.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4658>
Since we consume NIR, we get FRAG_RESULT_DEPTH in .x. Something must have
been working out for this code to not be trying to get an undefined value,
but go ahead and drop it now.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4668>
The BLIT scissor must be set correctly for tu_store_gmem_attachment.
Fixes this deqp test:
dEQP-VK.pipeline.multisample_shader_builtin.sample_id.137_191_1.samples
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4666>
I wanted to look at the effect of a core NIR change on a2xx codegen, but I
don't have any of those boards. This could also prove useful for quickly
sanity-checking the compiler by running shader-db on it -- a2xx fails in a
few ways on glmark2, and a3xx-a5xx fails on glmark2 in a debug_assert
(which we don't have enabled in our dEQP runs).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4652>
This appears to be limited by VPC_CNTL_0::NUMNONPOSVAR, which is an
8-bit bitfield with no possibility for expansion. Also, in practice
we'll be limited by the vertex shader output maximum, which includes
gl_Position, of 128, so that users won't be able to use more than 124
components anyways. Lower it to match the GL blob.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4641>
The extra bit needs to be used when using the maximum of 128 varying
components. I confirmed that PC_PRIMITIVE_CNTL_1 and SP_PRIMITIVE_CNTL
are expanded using a trace of the Vulkan blob with the maximum number of
varyings, and changed the others by analogy.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4641>
The intersects() function assumes that inside each instruction values
always die before they are defined, so that if the end of one range is
the same instruction as the beginning of the next then they don't
intersect. However, this isn't the case for values that become live at
the beginning of a basic block, which become live *before* the first
instruction, or instructions that die at the end of a basic block which
die after the last instruction.
For example, imagine that we have two values, A which is defined earlier
in the block and B which is defined in the last instruction of the block
and both die at the end of the basic block (e.g. are used in the next
iteration of a loop). We would compute a range for A of, say, (10, 20)
and for B of (20, 20) since each block's end_ip is the same as the ip of
the last instruction, and RA would consider them to not interfere.
There's a similar problem with values that become live at the beginning.
The fix is to offset the block's start_ip and end_ip by one so that they
don't correspond to any actual instruction. One way to think about this
is that we're adding fake instructions at the beginning and end of a
block where values become live & die. We could invert the order, so that
values consumed by each instruction are considered dead at the end of
the previous instruction, but then values that become dead at the
beginning of the basic block would incorrectly have an empty live range,
with a similar problem at the end of the basic block if we try to say
that values are defined at the beginning of the next instruction. So
the extra padding instructions are unavoidable.
This fixes an accidental infinite loop in the shader for
dEQP-VK.spirv_assembly.type.scalar.u32.switch_vert.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4614>
Even though vkCmdRenderPassBegin() isn't allowed inside a secondary
command buffer, vkCmdDispatch() is, and we emit an IB with compute
dispatches, which means that if the secondary command buffer records a
vkCmdDispatch() then we'll have an IB inside an IB, which is illegal.
Fixes hangs in e.g.
dEQP-VK.api.command_buffers.record_simul_use_secondary_one_primary.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4605>
Instead of exposing various layout functions, move image-related logic
into tu_image.c and have the image_view pre-fill relevant register values.
This changes the clear/blit code to use image_view.
This will make it much easier to deal with aspect masks, in particular for
planar formats and D32_S8.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4581>
Even though we normally use the CP_BLIT path with resolves that aren't
aligned, there's a special case when we're resolving the entire image
and there's enough padding so that we can still use CP_EVENT_WRITE::BLIT
when the render area isn't aligned. The hardware seems to not like
unaligned scissors when not clearing, and sometimes hangs rather than
silently round the scissor. This causes hangs in e.g.
dEQP-VK.glsl.derivate.dfdx.texture.msaa4.float_highp.
There was some concern that the CP_BLIT path might use this scissor
also, but I confirmed that this isn't the case by setting it to 0 before
resolving and then noting that CP_BLIT still works (but CP_EVENT_WRITE
doesn't). Furthermore, this is actually impossible because of how the 2D
engine is set up: it gets its own pair of register banks, which can be
switched independently of the 3D register banks, so that 2D events
(CP_BLIT) normally aren't synchronized relative to 3D events
(CP_EVENT_WRITE, CP_DRAW_*, and CP_EXEC_CS) and therefore they can't
share any registers except for non-pipelined registers like RB_CCU_CNTL
that don't use the register bank mechanism at all.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4585>
I had missed that LDC actually uses vec4 units for its offset. This
means that we have to create a new instruction, and lower it in
ir3_nir_lower_io_offsets, similar to the existing SSBO instructions.
Unfortunately we can't assume that loads are always vec4-aligned, so we
have to use the alignment information that NIR gives us. Unfortunately,
it's currently woefully inadequate, and will have to be fixed to give us
good codegen in the future.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4568>
After some experimentation, I believe that GRAS_LAYER_CNTL is
actually just a count register storing the number of layers in the
render target. While debugging cube_array geometry tests, I noticed
that the blob was setting an unknown 0x8 to LAYER_CNTL, so I checked
the value of LAYER_CNTL for various layer sizes:
1: LAYER_CNTL=0
2: LAYER_CNTL=1
3: LAYER_CNTL=2
4: LAYER_CNTL=3
9: LAYER_CNTL=8
256: LAYER_CNTL=255
2000: LAYER_CNTL=1999
Seems like this register just stores a count of the largest layer
that can be written to via gl_Layer. This commit updates the reg
docs, freedreno's gs implementation, and turnip's gs implementation.
Fixes dEQP-VK.geometry.layered.cube_array.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4541>
Previously we were using layout.layer_size for the layer stride, but
in Vulkan, you can alias a 3D image as an array of 2D images via the
VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT flag. One reason to use
this behavior is so the geometry shader can write to a specific
depth in a 3D framebuffer with gl_Layer.
Since the 3D image is not a *true* layered image, layer_size is 0.
Instead, we can copy what freedreno does and use the slice size.
Fixes dEQP-VK.geometry.layered.3d.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4541>
Fixes (with other patches to allow these tests to run):
dEQP-VK.ycbcr.query.size_lod.vertex.*
Suggested-by: Rob Clark <robclark@gmail.com>
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4557>
I missed that this had a micro-optimization to assume that there was
only ever one source, which is no longer valid for the bindless model
since we now have a bindless handle source. Remove the optimization to
fix assertion failures with turnip.
Fixes e.g.
dEQP-VK.glsl.texture_functions.query.texturesize.sampler2d_fixed_vertex
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4548>
Do a better job of pruning when removing unused instructions, including
cleaning up dangling false-deps.
This introduces a new ssa src pointer iterator, which makes it easy to
clear links without having to think about whether it is a normal ssa
src or a false-dep.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4440>
The src of the SFU instruction could also be array/reg (non-SSA).
Handle this case too.
The postsched cp pass makes this scenario more likely.
Fixes: cc82521de4 ("freedreno/ir3: round-robin RA")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4440>
Realize that certain instructions make a vecN live, and account for
this, in hopes of scheduling the remaining components of the vecN
sooner.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4440>
This replaces the depth-first search scheduler with a more traditional
ready-list scheduler. It primarily tries to reduce register pressure
(number of live values), with the exception of trying to schedule kills
as early as possible. (Earlier iterations of this scheduler had a
tendency to push kills later, and in particular moving texture fetches
which may not be necessary ahead of kills.)
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4440>
If the group pass must insert a mov to resolve conflicts, avoid the mov
appearing *after* the meta:collect whose src it is.
The current pre-RA scheduler doesn't really care about the initial
instruction order, but the new one will in some cases.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4440>
for verbose mode, print also the instruction "cycle" (which takes into
account (rptN) and (nopN)) in addition to instruction offset.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4440>
SEL_B32 (and presumably B16) checks for 0 or nonzero in the condition
(tested by just stuffing a uniform's value into it), so there's no need to
do ir3_b2n() on it, or any preceding ir3_n2b().
instructions in affected programs: 664444 -> 659927 (-0.68%)
nops in affected programs: 267898 -> 266312 (-0.59%)
non-nops in affected programs: 420260 -> 417329 (-0.70%)
dwords in affected programs: 144032 -> 137568 (-4.49%)
last-baryf in affected programs: 10801 -> 10321 (-4.44%)
full in affected programs: 2003 -> 2002 (-0.05%)
sstall in affected programs: 76670 -> 77405 (0.96%)
(ss) in affected programs: 4515 -> 4525 (0.22%)
(sy) in affected programs: 612 -> 604 (-1.31%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4516>
The v->binning variant is never added to shader->variants, so just free
each one as we free the nonbinning variant.
Noticed from drm-shim mode running out of open fds, since each bo ends up
with an fd.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4502>
Under the bindless model, there are 5 "base" registers programmed with a
64-bit address, and sam/ldib/ldc and so on each specify a base register
and an offset, in units of 16 dwords. The base registers correspond to
descriptor sets in Vulkan. We allocate a buffer at descriptor set
creation time, hopefully outside the main rendering loop, and then
switching descriptor sets is just a matter of programming the base
registers differently. Note, however, that some kinds of descriptors
need to be patched at command recording time, in particular dynamic
UBO's and SSBO's, which need to be patched at CmdBindDescriptorSets
time, and input attachments which need to be patched at draw time based
on the the pipeline that's bound. We reserve the fifth base register
(which seems to be unused by the blob driver) for these, creating a
descriptor set on-the-fly and combining all the dynamic descriptors from
all the different descriptor sets. This way, we never have to copy the
rest of the descriptor set at draw time like the blob seems to do. I
mostly chose to do this because the infrastructure was already there in
the form of dynamic_descriptors, and other drivers (at least radv) don't
cheat either when implementing this.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4358>
In Vulkan, descriptors for samplers, SSBO's, etc. are collected into
descriptor sets, and shaders can use multiple descriptor sets. At
command-recording time, users can swap out only some of the descriptor
sets, and the driver is supposed to do the minimum amount necessary to
update any internal binding tables, knowing that only some of the
descriptors have changed.
With the old binding model, focused on GL, where there are separate
tables for each type of resource, we can do somewhat better than now by
preserving descriptors from lower descriptor sets when switching higher
descriptor sets. However we still have to copy around descriptors before
each draw.
At least for a6xx, qualcomm went further, essentially copying the Vulkan
binding model as an alternate way to load resources. There's an array of
registers (actually an array for compute and one for everything else),
where each register holds a pointer to a descriptor set that can contain
various different descriptor types. The descriptors are padded out to 16
dwords, so that every instruction can use an index instead of a dword
offset. It's called "bindless", I think, because it can also be used to
implement the old GL bindless extensions (presumably it allows more
samplers and textures than the old model).
This commit adds the register and cmdstream parts. Next up will be the
instruction encoding.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4358>
Carve out some space at the beginning for push constants, and push them
directly, rather than remapping them to a UBO and then relying on the
UBO pushing code. Remapping to a UBO is easy now, where there's a single
table of UBO's, but with the bindless model it'll be a lot harder. I
haven't removed all the code to move the remaining UBO's over by 1,
though, because it's going to all get rewritten with bindless anyways.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4358>
The shader path is used to implement the following cases:
* stencil aspect mask on D24S8 (for image_to_buffer,buffer_to_image)
* clear/copy msaa destination (2D engine can't have msaa dest)
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3783>
* Take tile_mode as input directly
* tu6_format_gmem to tu6_base_format, use may not be limited to GMEM
* Add new helpers that will return the correct tile_mode as for image level
as part of the format.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3783>
This is needed to be able to know the number of samples during
CmdClearAttachments which can be used while the framebuffer is unknown.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3783>
* Correct bypass value for a618
* Bypass value for blitter
* Don't set RB_CCU_CNTL again unnecessarily in tu6_emit_binning_pass
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3783>
Found from piglit fbo-generatemipmaps failures, then tracked down with the
texturator test. The piece that really revealed things was finding that
1024x1 linear RGBA8 on the older blob drivers would have a pitch of 5120
instead of 4096, and the following levels minified that pitch.
Fixes ~124 piglit tests (~8.5% of piglit failures) on cheza.
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3987>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3987>
Trying to work out texture layout by remembering what things looked like
in texturator is hard. Instead, let's use texture layouts from tracing
the blob as a source of truth to make sure that we pick the same layouts
they do (and don't break known-good ones). More testcases will be added
as I fix layout bugs.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3987>
Geometry shaders support an invocations parameter up to a limit
defined by maxGeometryShaderInvocations. This was set to 127, but
executing with invocations > 32 causes a crash. As it turns out, the
blob only advertises a max of 32 invocations, so we set that in
turnip as well.
Fixes dEQP-VK.geometry.instanced.draw_*_instances_{127, 64}_geometry_invocations
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4436>
One of the features of geometry shaders is the ability to render to
different layers by assigning to the gl_Layer (Layer in SPIR-V)
builtin.
While have already plumbed the layer regid to the geometry shader,
we also need to GRAS_LAYER_CNTL to actually use layered rendering.
In addition, gmem does not support layered rendering, so we need to
force sysmem.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4436>
This commit updates VFD_CONTROL to use the GS header and primitive
ID sysvals if a geometry shader stage is present in the pipeline.
Like in the case of VPC, the code here is adapted from fd6_program.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4436>
This commit updates tu6_emit_vpc to selectively emit GS-specifc
configuration. Most of this is repurposed from fd6_program.c.
This also refactors `link_geometry_stages` to ir3_nir_lower_tess.c
so it can be shared between fd and tu.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4436>
Like with other shader types, we need to emit the geometry shader
object and the consts it uses. In addition, we need to emit
additional geometry-specific consts that link primitive/vertex stride
between the vs and gs. In conjunction with the gsheader, these are
used by the vs to determine where to stlw outputs and used by the gs
to determine where to ldlw those outputs from.
FD emits these consts in the draw call because in GL, you can mix
and match shaders in different programs. In Vulkan, however, we
compile and link the shaders at pipeline creation, so we can emit
these in the pipeline IB instead.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4436>
The ir3 compiler only lowers the VS and GS for geometry shading if
the corresponding has_gs key is set in the shader key. Without it,
GS-specific intrinsics like load_per_vertex_input won't get lowered
and the GS header will be initialized with invalid values.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4436>
We don't really track these as the ir is transformed, but it would be a
useful thing for some passes to have. So add a pass to collect this
information. It uses instr->data (generic per-pass ptr), with the
hashsets hanging under a mem_ctx for easy disposal at the end of the
pass.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4423>
Don't fold conversions into array (incl phi lowered to regs/array).
These aren't SSA. Avoids crashes in particular in frag shaders with
flow control, which would leave a dangling array write disconnect from
the original cov src.
Possibly this could be slightly relaxed, if there is no other consumer
of the src, and it were in the same block. But it would require
updating block->keeps, and taking care of barrier state. Which isn't a
thing the cf pass does currently.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4423>
Immediate offsets are currently collapsed for ldlw, but ldlw does
behave correctly with immediate values. For example,
`ldlw.u32 r0.x, l[4], 1` actually means to use the value of
regid 4 (r1.x) as the offset when we actually want it to use the
imm value of 4 as the offset.
This commit disables copy prop for ldlw offsets so the same
intrinsic gets compiled to:
mov.u32u32 r0.y, 0x00000004
ldlw.u32 r0.x, l[r0.y], 1
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4439>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4439>
Previously, turnip advertised 4-bit subpixel precision when in
practice, a6xx seems to render with 8-bit precision. This caused
dEQP-VK.renderpass2.suballocation.subpass_dependencies.late_fragment_tests.*
to fail because they compare images rendered with turnip against
ones rendered via a software reference implementation parameterized
by turnip's VkPhysicalDeviceLimits.subPixelPrecisionBits value.
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4172>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4172>
It turns out that every *_TS event, i.e. every event which requires a
seqno pointer, also allows generating an interrupt in the kernel, at
least since a3xx. And furthermore these interrupts are named by the kgsl
kernel driver and already in envytools. Therefore it's possible to map
out what the *_TS events are with 100% certainty, given access to the
hardware, by sending a CP_EVENT_WRITE with bit 31 set, unmasking all
interrupts in the kernel, and logging which ones get hit. I've done this
for a6xx, and I've also looked at the a5xx firmware, and the list of TS
interrupts is the same as a6xx, so I have a pretty good idea of what the
a5xx events are. I also fixed a few related things along the way:
- VIZQUERY_END overlaps with WT_DONE_TS, but VIZQUERY_START was also a
mess, with neither VIZQUERY_START nor HLSQ_FLUSH using variants. I added
what seems like reasonable variants, based on the existing comment
and the fact that HLSQ_FLUSH is only used in Mesa with a3xx and a4xx.
- CACHE_FLUSH_AND_INVALIDATE seems to come straight from R600, and I
have no idea if it's actually valid with a2xx, but given that RB_DONE_TS
exists in the interrupt mask since a3xx, I guessed that RB_DONE_TS
hasn't changed position since then and put it down as a3xx+ and limited
CACHE_FLUSH_AND_INVALIDATE to a2xx. Someone with the relevant hardware
should be able to confirm.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4065>
Tell users what the base address of the image needs to be aligned to.
These values are based on experimentation via passing an offset to
vkBindImageMemory with turnip and seeing if tests still pass. Note that
r8g8 is also special in this regard, however it actually has an
increased alignment (in bytes).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4357>
Let's make it clear what includes are being added everywhere, so that
they can be cleaned up.
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4360>
Split out the construction of registers and classes (which is the same
on all gens) from setting up conflicts. Prep to re-work how we setup
conflicts on a6xx+ which merged half/full register file.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4272>
Since we are re-assigning the scalars anyways in the second pass, assign
them to the highest free reg in the first pass (rather than lowest) to
allow packing vecN regs as low as possible.
Note this required some changes specifically for tex instructions with a
single component writemask that is not necessarily .x, as previously
these would get assigned in the first RA pass, and since they are still
scalar, we'd end up w/ some r47.* and other similarly way-to-high
assignments after the 2nd pass.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4272>
Using the output of the first pass isn't ideal, as it can bake in the
losses from fragmentation which the scalar pass is intended to fill in.
This gets worse when we start using "vectorish" instructions, due to
higher use of vecN values.
Instead, we can just use the outputs of the liveness analysis to get a
more accurate # of maximum live values at any point.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4272>
Decouple the messy logic of figuring out vreg names defined/used by an
instruction from the logic of what to do about it by introducing
iterators. There is still *some* array vs ssa special casing in
ra_block_compute_live_ranges(), but less than before. And this will
avoid introducing a second copy of the def/use logic in a following
patch which uses the liveranges to calculate the maximum # of live
values (which is the optimal target for max physical register window
to round-robin within).
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4272>
Account for the # of regs an instruction writes, and fix an off-by-one.
(We are about to replace this with calculating the register target using
the live-ranges, but in debugging that it was useful to assert() if it
chose a higher target.)
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4272>