My benchmarking shows no significant change in cache load times with a
single shader cache file vs the existing cache implementation of many
small files (tested with my spinning rust HDD).
However this new single file cache implementation does reduce the total
size on disk used by the shader cache. We have a problem with the existing
cache where writing tiny files to disk causes more disk space to be used than
is actually needed for the files due to the minimum size required for a file.
In pratice this tends to inflate the size of the cache on disk to over 3x
larger.
There are other advantages of using a single file for shader cache entries
such as allowing better removal of cache entries once we hit the max cache
size limit (although we don't implement any max cache size handling in this
initial implementation).
The primary reason for implementing a single file cache for now is to allow
better performance and handling by third party applications such as steam
that collect and distribute precompiled cache entries.
For this reason we also implement a new environment variable
MESA_DISK_CACHE_READ_ONLY_FOZ_DBS which allows a user to pass in a path
to a number of external read only shader cache dbs. There is an initial
limit of 8 dbs that can be passed to mesa like so:
MESA_DISK_CACHE_READ_ONLY_FOZ_DBS=/full_path/filename1, ... ,/full_path/filename8
Where the filename represents the cache db and its index file e.g.
filename1.foz and filename1_idx.foz
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7725>
This allows us to guarantee the different cache implementations will not
interfere with each other and should make it more clear that the max cache
size limits are applied separately for each cache implementation.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7725>
This code is already OS agnostic and moving it here enables code
sharing in the following patches.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7725>
This will be faster and avoids checking for errors with the
compression implementation which we shouldn't need to do. Instead
we trust the compression library does the correct thing and simply
error check the data loaded from disk.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7725>
Without this check, the INTEL_MEASURE environment variable could be
misused to overwrite arbitrary files.
Fixes: 0f4143ec37 ("intel: Print GPU timing data based on INTEL_MEASURE")
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9166>
There are patterns that will re-write the fmin or fmax part into a form
that other patterns will gradually convert to the same ior or iand. For
example,
fmax(b2f(a), b2f(b)) != 0
b2f(a || b) != 0
a || b
No shader-db or fossil-db changes on any Intel platform.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9122>
fmin(-A, -B) is -fmax(A, B), and fmax(-A, -B) is -fmin(A, B). Therefore
the logic joining A and B should toggle between ior and iand for the
negated versions.
At the very least, a shader from Euro Truck Simulator 2 in shader-db is
affected by this. The KIL instruction in the (ARB assembly) shader ends
up with the wrong logic. This is _probably_ the source of
https://gitlab.freedesktop.org/mesa/mesa/-/issues/1346.
That said, the issue mentions that Mesa 18.0.5 works, but commit
68420d8322 ("nir: Simplify min and max of b2f") was added in 17.3.
Moreover, I was not able to reproduce the error in the ETS2 shader from
shader-db from any Mesa commit near the time the original fd.o bugzilla
was submitted (December 2018). 🤷
In fact, the current error in that shader starts with 9167324a86
("nir/algebraic: Mark some logic-joined comparison reductions as
exact"). That's a bit of a red herring as 9167324a86 just sets off a
chain of replacements that eventually leads to the incorrect min/max of
b2f patterns fixed by this commit.
The other affected shaders in the shader-db results are from Cargo
Commander. These are also ARB assembly shaders.
I think any ARB assembly shader that uses the pattern
SLT r0, ...;
...
KIL -r0;
will suffer from issues related to this.
This change fixes the piglit
tests/spec/arb_fragment_program/kil-of-slt.shader_test test added in
https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/454.
shader-db results:
All Gen6+ platforms had similar result. (Ice Lake shown)
total instructions in shared programs: 20034604 -> 20034486 (<.01%)
instructions in affected programs: 3885 -> 3767 (-3.04%)
helped: 47
HURT: 2
helped stats (abs) min: 2 max: 4 x̄: 2.64 x̃: 2
helped stats (rel) min: 2.33% max: 8.33% x̄: 3.48% x̃: 3.39%
HURT stats (abs) min: 3 max: 3 x̄: 3.00 x̃: 3
HURT stats (rel) min: 13.64% max: 16.67% x̄: 15.15% x̃: 15.15%
95% mean confidence interval for instructions value: -2.83 -1.99
95% mean confidence interval for instructions %-change: -3.84% -1.60%
Instructions are helped.
total cycles in shared programs: 979881379 -> 979879406 (<.01%)
cycles in affected programs: 119873 -> 117900 (-1.65%)
helped: 46
HURT: 3
helped stats (abs) min: 10 max: 756 x̄: 45.41 x̃: 26
helped stats (rel) min: 0.53% max: 19.72% x̄: 1.67% x̃: 1.26%
HURT stats (abs) min: 28 max: 56 x̄: 38.67 x̃: 32
HURT stats (rel) min: 1.44% max: 3.54% x̄: 2.75% x̃: 3.27%
95% mean confidence interval for cycles value: -70.83 -9.70
95% mean confidence interval for cycles %-change: -2.23% -0.57%
Cycles are helped.
Iron Lake and GM45 had similar results. (Iron Lake shown)
total instructions in shared programs: 8115098 -> 8115076 (<.01%)
instructions in affected programs: 2592 -> 2570 (-0.85%)
helped: 32
HURT: 2
helped stats (abs) min: 1 max: 1 x̄: 1.00 x̃: 1
helped stats (rel) min: 0.88% max: 2.70% x̄: 1.35% x̃: 1.31%
HURT stats (abs) min: 5 max: 5 x̄: 5.00 x̃: 5
HURT stats (rel) min: 17.24% max: 18.52% x̄: 17.88% x̃: 17.88%
95% mean confidence interval for instructions value: -1.15 -0.15
95% mean confidence interval for instructions %-change: -1.83% 1.39%
Inconclusive result (%-change mean confidence interval includes 0).
total cycles in shared programs: 238189718 -> 238189802 (<.01%)
cycles in affected programs: 75076 -> 75160 (0.11%)
helped: 3
HURT: 31
helped stats (abs) min: 2 max: 130 x̄: 44.67 x̃: 2
helped stats (rel) min: 0.18% max: 5.70% x̄: 2.02% x̃: 0.19%
HURT stats (abs) min: 2 max: 70 x̄: 7.03 x̃: 4
HURT stats (rel) min: 0.07% max: 6.41% x̄: 0.53% x̃: 0.15%
95% mean confidence interval for cycles value: -7.27 12.21
95% mean confidence interval for cycles %-change: -0.33% 0.94%
Inconclusive result (value mean confidence interval includes 0).
No fossil-db changes on any Intel platform.
Fixes: 68420d8322 ("nir: Simplify min and max of b2f")
Closes: #1346
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9122>
Fixes 'nir_tex_src_coord' param was provided to NIR 'txf' operation as a
vec3 for TEXTURE_1D_ARRAY target, causing an assert.
Only following targets require vec3: TEXTURE_2D_ARRAY, TEXTURE_3D,
TEXTURE_CUBE, TEXTURE_CUBE_ARRAY. The rest must use vec2.
Packing layer value into Y-coordinate the same way it was done in
'create_fs' in commit 2bf6dfac.
Fixes: a01ad311 ("st/mesa: Add NIR versions of the PBO upload/download
shaders. ")
Signed-off-by: Yevhenii Kharchenko <yevhenii.kharchenko@globallogic.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9014>
We call update_last_vue_map after updating the shaders, which compares
the new and old VUE maps. Except...updating the shaders may have
dropped the last reference to the variant that ice->shaders.last_vue_map
belonged to, leading to a classic use-after-free.
Fix this by taking a reference to the variant for the last VUE stage,
so it stays around until we're done with it.
Fixes: 1afed51445 ("iris: Store a list of shader variants in the shader itself")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4311
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9143>
Before, we only used 2k of shared memory.
It was found that 5 lower bits of SP_CS_UNKNOWN_A9B1 do control
the available size of shared memory for compute shaders, with
AVAILABLE_SIZE = (SP_CS_UNKNOWN_A9B1_SHARED_SIZE + 1) * 1k
up to 32k. And SP_CS_UNKNOWN_A9B1_SHARED_SIZE being zero enables
all 32k of shared memory.
Fixes tests:
dEQP-VK.rasterization.line_continuity.line-strip
dEQP-VK.memory_model.message_passing.core11.u32.coherent.fence_fence.atomicwrite.workgroup.payload_local.buffer.guard_nonlocal.workgroup.comp
dEQP-VK.memory_model.message_passing.core11.u32.coherent.fence_fence.atomicwrite.workgroup.payload_nonlocal.workgroup.guard_local.buffer.comp
dEQP-VK.memory_model.write_after_read.core11.u32.coherent.fence_fence.atomicwrite.workgroup.payload_local.image.guard_nonlocal.workgroup.comp
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9157>
The top-level gitlab-ci.yml is big and unwieldy when one wants to work on
CI for a single driver. Move the drivers to separate include files for
ease of finding all your driver's tests, and also to pave the way for work
on a single driver's CI to not retest all other drivers.
Reviewed-by: Andres Gomez <agomez@igalia.com>
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9139>
Since vars_to_explicit_types is now where driver_location/offset is filled out,
we need to make sure that we still have all app-provided kernel arg variables
at that point in time so they all get assigned unique offsets. That means
that we can't have removed dead uniforms yet, which also means we can't have
filled out metadata for inline samplers (since usage of them generates tons
of duplicate uniforms).
Reviewed-by: Bill Kristiansen <billkris@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9160>
Otherwise we may get a tile height exceeding the maximum.
Fixes tests:
dEQP-VK.pipeline.render_to_image.core.2d.huge.height.r8g8b8a8_unorm
dEQP-VK.pipeline.render_to_image.core.2d.huge.height.r8g8b8a8_unorm_d16_unorm
dEQP-VK.pipeline.render_to_image.core.2d.huge.height.r8g8b8a8_unorm_s8_uint
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9159>
Just named the variable wrong. Can't really fault anyone for breaking a
target we don't build in CI...
Fixes: 99e25d183d mesa: optimize set_varying_vp_inputs by precomputing the conditions
Acked-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9132>
This patch will set the gfxip into elf_hdr.e_flags by reading data
from struct rad_info->chip_class instead of hardcoding.
v2: remove unused #define
Signed-off-by: Yogesh Mohan Marimuthu <yogesh.mohanmarimuthu@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8609>
This patch dumps code object, code object loader events, pso correlation
databses to given rgp profile file. The data to dump is taken from
struct rgp_code_object, struct rgp_loader_events and
struct rgp_pso_correlation.
v2: fix sqtt_file structure name to match existing names in file.
fix checking wrong variable for pso correlation record_count.
remove freeing rgp database here.
Signed-off-by: Yogesh Mohan Marimuthu <yogesh.mohanmarimuthu@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8609>
struct thread_trace_data holds struct rgp_code_object, struct
rgp_loader_events, struct rgp_pso_correlation data. This data is required
in function ac_sqtt_dump_data(). This patch makes the code changes
required to pass struct thread_trace_data to function ac_sqtt_dump_data().
Signed-off-by: Yogesh Mohan Marimuthu <yogesh.mohanmarimuthu@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8609>
This patch exposes struct rgp_code_object, struct rgp_loader_events and
struct rgp_pso_correlation into struct ac_thread_trace_data. Other modules
can fill in this data structure. The code object, code object loader events,
pso correlation database is written to rgp profile file using data from
these data structure exposed by this patch.
v2: move the data structures to struct ac_thread_trace_data
Signed-off-by: Yogesh Mohan Marimuthu <yogesh.mohanmarimuthu@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8609>
This patch adds helper function ac_rgp_write_elf_object() which
creates elf object from struct rgp_code_object_record. The elf
object is written to given output rgp profile file.
v2: for newly added files, change copyright year from 2020 to 2021
v3: fix 32bit compilation failure
Signed-off-by: Yogesh Mohan Marimuthu <yogesh.mohanmarimuthu@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8609>
This patch adds data structures used to collect data for code object,
code object loader event and pso correlation databases present in
a rgp profile.
v2: fix code review comments from Pierre-Eric
v3: Make loader_event_type into enum
Signed-off-by: Yogesh Mohan Marimuthu <yogesh.mohanmarimuthu@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8609>
This patch adds functions to create msgpack formatted data.
For msgpack specification refer to
github.com/msgpack/msgpack/blob/master/spec.md
This patch only adds formats from msgpac specification that
are required for rgp profile data.
v2: for newly added files, change copyright year from 2020 to 2021
Signed-off-by: Yogesh Mohan Marimuthu <yogesh.mohanmarimuthu@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8609>
Anything higher than INT_MAX results in overflow although the parameter is declared as size_t.
Worse, with (size_t)-1 it is silently ignored and Woverflow is not emitted.
Closes#4226
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Tested-by: Prodea Alexandru-Liviu <liviuprodea@yahoo.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9134>
Blob doesn't apply multipos optimization starting from 11 views
even on a650, however in practice, with the limit of 16 views,
tests pass on a640/a650 and fail on a630.
Fixes tests:
dEQP-VK.multiview.draw_indexed.max_multi_view_view_count
dEQP-VK.multiview.input_attachments.max_multi_view_view_count
dEQP-VK.multiview.masks.max_multi_view_view_count
dEQP-VK.multiview.multisample.max_multi_view_view_count
dEQP-VK.multiview.queries.max_multi_view_view_count
dEQP-VK.multiview.renderpass2.index.fragment_shader.max_multi_view_view_count
dEQP-VK.multiview.secondary_cmd_buffer.max_multi_view_view_count
Fixes: 8d275778 ("tu: Enable multi-position output")
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9135>
Fix defects reported by Coverity Scan.
uninit_member: Non-static class member serial is not initialized in this constructor nor in any functions that it calls.
uninit_member: Non-static class member sched is not initialized in this constructor nor in any functions that it calls.
uninit_member: Non-static class member bb is not initialized in this constructor nor in any functions that it calls.
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9037>
Reduce noise in a6xx.xml by removing LO/HI versions of address registers.
Also fix type="address" registers in register packing (use bit size instead
of checking for "waddress" to use qword)
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8423>
since we aren't (currently) parallelizing and now have barriers, we don't need to cycle
the batch here, which lets us avoid submitting too-small command buffers
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9148>
if first==last layer, this should be a 2D slice of the cube
else if this isn't all the layers, this should be an array of slices
fixes a bunch of spec@arb_shader_image_size@builtin cases
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9080>
