Until now we have assumed that we could skip emitting these barriers
in the general case based on empirical testing and a few assumptions
detailed in a comment in the driver code, however, recent CTS tests
have showed that we actually need them to produce correct behavior.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Error states coming from actual Vulkan applications tend to have fairly
long command buffers and lots of chained batches. 30 total BOs isn't
nearly enough. This commit bumps it to 256, makes some things use the
actual number of sections instead of the #define, and adds asserts if we
ever go over 256 sections.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Our attempt to restart the loop with the second level batch worked at
one point but got broken at some point. It was too fragile anyway and
we're not likely to have enough secondaries to actually overflow the
stack so we may as well recurse in both cases.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
CACHE_MODE_SS is not listed in gfxspecs table for user mode
non-privileged registers. So, making any changes from Mesa
will do nothing. Kernel is already setting this bit in
CACHE_MODE_SS register which is saved/restored to/from
the HW context image.
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Enables SPV_KHR_8bit_storage and VK_KHR_8bit_storage on gen 8+
using the VK_KHR_get_physical_device_properties2 functionality
to expose if the extension is supported or not.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
When the destination is a BYTE type allow raw movs
even if the stride is not exact multiple of destination
type and exec type, execution type is Word and its size is 2.
This restriction was only allowing stride==2 destinations
for 8-bit types.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Since Gen8+ Intel PRM states that "r127 must not be used for return
address when there is a src and dest overlap in send instruction."
This patch implements this restriction creating new grf127_send_hack_node
at the register allocator. This node has a fixed assignation to grf127.
For vgrf that are used as destination of send messages we create node
interfereces with the grf127_send_hack_node. So the register allocator
will never assign to these vgrf a register that involves grf127.
If dispatch_width > 8 we don't create these interferences to the because
all instructions have node interferences between sources and destination.
That is enough to avoid the r127 restriction.
This fixes CTS tests that raised this issue as they were executed as SIMD8:
dEQP-VK.spirv_assembly.instruction.graphics.8bit_storage.8struct_to_32struct.storage_buffer_*int_geom
Shader-db results on Skylake:
total instructions in shared programs: 7686798 -> 7686797 (<.01%)
instructions in affected programs: 301 -> 300 (-0.33%)
helped: 1
HURT: 0
total cycles in shared programs: 337092322 -> 337091919 (<.01%)
cycles in affected programs: 22420415 -> 22420012 (<.01%)
helped: 712
HURT: 588
Shader-db results on Broadwell:
total instructions in shared programs: 7658574 -> 7658625 (<.01%)
instructions in affected programs: 19610 -> 19661 (0.26%)
helped: 3
HURT: 4
total cycles in shared programs: 340694553 -> 340676378 (<.01%)
cycles in affected programs: 24724915 -> 24706740 (-0.07%)
helped: 998
HURT: 916
total spills in shared programs: 4300 -> 4311 (0.26%)
spills in affected programs: 333 -> 344 (3.30%)
helped: 1
HURT: 3
total fills in shared programs: 5370 -> 5378 (0.15%)
fills in affected programs: 274 -> 282 (2.92%)
helped: 1
HURT: 3
v2: Avoid duplicating register classes without grf127. Let's use a node
with a fixed assignation to grf127 and create interferences to send
message vgrf destinations. (Eric Anholt)
v3: Update reference to CTS VK_KHR_8bit_storage failing tests.
(Jose Maria Casanova)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Cc: 18.1 <mesa-stable@lists.freedesktop.org>
Implement at brw_eu_validate the restriction from Intel Broadwell PRM,
vol 07, section "Instruction Set Reference", subsection "EUISA
Instructions", Send Message (page 990):
"r127 must not be used for return address when there is a src and
dest overlap in send instruction."
v2: Style fixes (Matt Turner)
Reviewed-by: Matt Turner <mattst88@gmail.com>
Cc: 18.1 <mesa-stable@lists.freedesktop.org>
We were not properly writing page tables when the virtual address
range spans multiple subtrees of the tables.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
The implementation of CreateRenderPass2 uses the helpers we broke out in
previous commits. The implementations of the new vkCmd functions just
call the old versions.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
This makes certain checks a bit easier and means that we don't have
the attachment information duplicated in the attachment list and in
depth_stencil_attachment.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
This helps us to compact original instruction:
mul(8) g3<1>D g6<8,8,1>UD 0x00000006UD { align1 1Q };
So now we emit:
mul(8) g3<1>UD g6<8,8,1>UD 0x00000006UD { align1 1Q compacted };
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
At the time of commit 7bc6e455e2 (i965: Add support for saturating
immediates.) we thought mixed type saturates would be impossible. We
were only thinking about type converting moves from D to F, for
example. However, type converting moves w/saturate from F to DF are
definitely possible. This change minimally relaxes the restriction to
allow cases that I have been able trigger via piglit tests.
Fixes new piglit tests:
- arb_gpu_shader_fp64/execution/built-in-functions/fs-sign-sat-neg-abs.shader_test
- arb_gpu_shader_fp64/execution/built-in-functions/vs-sign-sat-neg-abs.shader_test
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
This is achived by copying the sign(abs(x)) optimization from the FS
backend.
On Gen7 an earlier platforms, this fixes new piglit tests:
- glsl-1.10/execution/vs-sign-neg-abs.shader_test
- glsl-1.10/execution/vs-sign-sat-neg-abs.shader_test
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
In Python 2, `print` was a statement, but it became a function in
Python 3.
Using print functions everywhere makes the script compatible with Python
versions >= 2.6, including Python 3.
Signed-off-by: Mathieu Bridon <bochecha@daitauha.fr>
Acked-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Dylan Baker <dylan@pnwbakers.com>
The bug fixed by the previous commit went undetected because extra
stderr messages are not flagged by the CI. Copy the solution from
fs_visitor::nir_emit_instr and mark the default case unreachable.
An alternate solution is to delete the default case so that the compiler
will issue a warning. That may require more work since there are other
(impossible) cases that exist.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Some of the lowering passes, nir_lower_locals_to_regs for example, can
cause some previously live code to be dead. This pass in particular
leaves a bunch of nir_instr_type_deref instructions floating around.
This causes shader-db runs on Gen5 through Haswell to spew tons of
messages like:
VS instruction not yet implemented by NIR->vec4
UnrealEngine4/EffectsCaveDemo/239.shader_test is one shader that
generates these messages. Cleaning up the dead code fixes that.
To verify, I did a shader-db before and after. Even though all the
messages are gone, the results make my brain hurt. :(
Haswell
total cycles in shared programs: 411890163 -> 411891145 (<.01%)
cycles in affected programs: 57016 -> 57998 (1.72%)
helped: 3
HURT: 11
helped stats (abs) min: 2 max: 154 x̄: 96.67 x̃: 134
helped stats (rel) min: 0.08% max: 2.23% x̄: 1.42% x̃: 1.96%
HURT stats (abs) min: 18 max: 686 x̄: 115.64 x̃: 20
HURT stats (rel) min: 0.81% max: 7.12% x̄: 1.87% x̃: 0.93%
95% mean confidence interval for cycles value: -51.39 191.67
95% mean confidence interval for cycles %-change: -0.14% 2.46%
Inconclusive result (value mean confidence interval includes 0).
Ivy Bridge
total cycles in shared programs: 259114802 -> 259115032 (<.01%)
cycles in affected programs: 24034 -> 24264 (0.96%)
helped: 1
HURT: 9
helped stats (abs) min: 2 max: 2 x̄: 2.00 x̃: 2
helped stats (rel) min: 0.08% max: 0.08% x̄: 0.08% x̃: 0.08%
HURT stats (abs) min: 18 max: 48 x̄: 25.78 x̃: 20
HURT stats (rel) min: 0.80% max: 1.94% x̄: 1.08% x̃: 0.80%
95% mean confidence interval for cycles value: 12.42 33.58
95% mean confidence interval for cycles %-change: 0.54% 1.38%
Cycles are HURT.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Fixes: 5a02ffb733 nir: Rework lower_locals_to_regs to use deref instructions
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
On Python 2, the default JSON separators are ', ' for items and ': ' for
dicts.
On Python 3, the default is the same when no indent is specified, but if
one is (and we do specify one) then the default items separator becomes
',' (the dict separator remains unchanged).
This change explicitly specifies the Python 3 default, which helps
ensuring that the output is identical, whether it was generated by
Python 2 or 3.
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
We already embed the headers, no need to redefine defines/structs.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
For gen8+, write out PPGTT tables in aub files so that full 48-bit
addresses can be serialized.
v2: Fix handling of `end` index in map_ppgtt
v3: Correctly mark GGTT entry as present (Rafael)
Signed-off-by: Scott D Phillips <scott.d.phillips@intel.com>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Scott added new stuff in IGT.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
With PPGTT mappings, our aubinator implementation can be quite slow if
we request a buffer that doesn't exist. Instead of doing a PPGTT walk
for invalid addresses (0 lengths), wait until we're sure we want to
decode the data.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
v2: by Lionel
Fix memfd_create compilation issue
Fix pml4 address stored on 32 instead of 64bits
Return no buffer if first ppgtt page is not mapped
v3: Drop additional memfd_create() (Rafael)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
We use memfd to store physical pages as they get read/written to and
the GGTT entries translating virtual address to physical pages.
Based on a commit by Scott Phillips.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Now that we're softpinning the address of our BOs in anv & i965, the
addresses selected start at the top of the addressing space. This is a
problem for the current implementation of aubinator which uses only a
40bit mmapped address space.
This change keeps track of all the memory writes from the aub file and
fetch them on request by the batch decoder. As a result we can get rid
of the 1<<40 mmapped address space and only rely on the mmap aub file
\o/
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
On a follow up commit in this series, we stop copying the data from
the mmap'ed file into our big gtt mmap, and start referencing data in
it directly. So reallocating the read buffer and adding more data from
stdin wouldn't work. For that reason, let's stop supporting stdin
process.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
These memory offsets are stored in the gen_batch_decode_ctx.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
In 6f5abf3146 this code was fixed to calculate the maximum size of
an attribute in a seperate pass and then allocate the registers to
that size. However this wasn’t taking into account ranges that overlap
but don’t have the same starting location. For example:
layout(location = 0, component = 0) out float a[4];
layout(location = 2, component = 1) out float b[4];
Previously, if ‘a’ was processed first then it would allocate a
register of size 4 for location 0 and it wouldn’t allocate another
register for location 2 because it would already be covered by the
range of 0. Then if something tries to write to b[2] it would try to
write past the end of the register allocated for ‘a’ and it would hit
an assert.
This patch changes it to scan for any overlapping ranges that start
within each range to calculate the maximum extent and allocate that
instead.
Fixed Piglit’s arb_enhanced_layouts/execution/component-layout/
vs-fs-array-interleave-range.shader_test
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Fixes: 6f5abf3146 "i965: Fix output register sizes when multiple variables
share a slot."
