Change brw_memory_fence to return the number of messages emitted, and
use that to update the send_count statistic in code generation.
This will fix the book-keeping for IVB since the memory fences will
result in two SEND messages.
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4646>
Push constants in particular can get picked up by the hardware at weird
times that happen *before* 3DPRIMITIVE. Therefore, we need to flush
before we emit all our state to ensure that any data they may pick up is
in memory in time. This fixes an app which does vkCmdCopyBuffers
immediately followed by a vkCmdBeginRenderPass and vkCmdDraw which uses
the destination of the copy as a UBO which we push.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4601>
The versions that take a specific number of operands will do various
fixups depending on the platform and the opcode. However, the version
that takes an array of sources did not. This makes all version operate
similarly.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4582>
This commit fixes performance regressions introduced by e03f965280
in which we started bounds checking our push constants. This added a
LOT of shader code to shaders which use the robustBufferAccess feature
and led to substantial spilling. The checking we just added to the FS
back-end is far more efficient for two reasons:
1. It can be done at a whole register granularity rather than per-
scalar and so we emit one SIMD8 SEL per 32B GRF rather than one
SIMD16 SEL (executed as two SELs) for each component loaded.
2. Because we do it with NoMask instructions, we can do it on whole
pushed GRFs without splatting them out to SIMD8 or SIME16 values.
This means that robust buffer access no longer explodes our register
pressure for no good reason.
As a tiny side-benefit, we're now using can use AND instead of SEL which
means no need for the flag and better scheduling.
Vulkan pipeline database results on ICL:
Instructions in all programs: 293586059 -> 238009118 (-18.9%)
SENDs in all programs: 13568515 -> 13568515 (+0.0%)
Loops in all programs: 149720 -> 149720 (+0.0%)
Cycles in all programs: 88499234498 -> 84348917496 (-4.7%)
Spills in all programs: 1229018 -> 184339 (-85.0%)
Fills in all programs: 1348397 -> 246061 (-81.8%)
This also improves the performance of a few apps:
- Shadow of the Tomb Raider: +4%
- Witcher 3: +3.5%
- UE4 Shooter demo: +2%
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4447>
This fixes two bugs: First, if the same block index showed up twice, we
only pick the first one. Second, we weren't multiplying by 32. This
didn't show up in tests because RBA testing is garbage. Found while
looking at shaders from the UE4 Shooter demo.
Fixes: e03f9652 "anv: Bounds-check pushed UBOs when..."
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4578>
Move the calculation to helper functions -- similar to what GL already
needs to do.
This is a preparation for dropping this field since this value is
expected to be calculated by the drivers now for variable group size
case. And also the field would get in the way of brw_compile_cs
producing multiple SIMD variants (like FS).
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4504>
Add new builtin parameters that are used to keep track of the group
size. This will be used to implement ARB_compute_variable_group_size.
The compiler will use the maximum group size supported to pick a
suitable SIMD variant. A later improvement will be to keep all SIMD
variants (like FS) so the driver can select the best one at dispatch
time.
When variable workgroup size is used, the small workgroup optimization
is disabled as it we can't prove at compile time that the barriers
won't be needed.
Extracted from original i965 patch with additional changes by
Caio Marcelo de Oliveira Filho.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4504>
The push.total field had three values but only one was directly
used (size). Replace it with a helper function that explicitly takes
the cs_prog_data and the number of threads -- and use that in the
drivers.
This is a preparation for ARB_compute_variable_group_size where the
number of threads (hence the total size for push constants) is not
defined at compile time (not cs_prog_data->threads).
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4504>
Identify if view_index is used only for position calculation, and use
Primitive Replication to implement Multiview in Gen12. This feature
allows storing per-view position information in a single execution of
the shader, treating position as an array.
The shader is transformed by adding a for-loop around it, that have an
iteration per active view (in the view_mask). Stores to the position
now store into the position array for the current index in the loop,
and load_view_index() will return the view index corresponding to the
current index in the loop.
The feature is controlled by setting the environment variable
ANV_PRIMITIVE_REPLICATION_MAX_VIEWS, which defaults to 2 if unset.
For pipelines with view counts larger than that, the regular
instancing will be used instead of Primitive Replication. To disable
it completely set the variable to 0.
v2: Don't assume position is set in vertex shader; remove only stores
for position; don't apply optimizations since other passes will
do; clone shader body without extract/reinsert; don't use
last_block (potentially stale). (Jason)
Fix view_index immediate to contain the view index, not its order.
Check for maximum number of views supported.
Add guard for gen12.
v3: Clone the entire shader function and change it before reinsert;
disable optimization when shader has memory writes. (Jason)
Use a single environment variable with _DEBUG on the name.
v4: Change to use new nir_deref_instr.
When removing stores, look for mode nir_var_shader_out instead
of the walking the list of outputs.
Ensure unused derefs are removed in the non-position part of the
shader.
Remove dead control flow when identifying if can use or not
primitive replication.
v5: Consider all the active shaders (including fragment) when deciding
that Primitive Replication can be used.
Change environment variable to ANV_PRIMITIVE_REPLICATION.
Squash the emission of 3DSTATE_PRIMITIVE_REPLICATION into this patch.
Disable Prim Rep in blorp_exec_3d.
v6: Use a loop around the shader, instead of manually unrolling, since
the regular unroll pass will kick in.
Document that we don't expect to see copy_deref or load_deref
involving the position variable.
Recover use_primitive_replication value when loading pipeline from
the cache.
Set VARYING_SLOT_LAYER to 0 in the shader. Earlier versions were
relying on ForceZeroRTAIndexEnable but that might not be
sufficient.
Disable Prim Rep in cmd_buffer_so_memcpy.
v7: Don't use Primitive Replication if position is not set, fallback
to instancing; change environment variable to be
ANV_PRIMITVE_REPLICATION_MAX_VIEWS and default it to 2 based on
experiments.
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2313>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2313>
Change brw_compute_vue_map() to also take the number of pos slots. If
more than one slot is used, the VARYING_SLOT_POS is treated as an
array.
When using Primitive Replication, instead of a single position, the
VUE must contain an array of positions. Padding might be
necessary (after clip distance) to ensure rest of attributes start
aligned.
v2: Add note about array in the commit message and assert that
pos_slots >= 1 to make clear 0 is invalid. (Jason)
Move padding to be after the clip distance.
v3: Apply the correct offset when gathering the sources from outputs.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net> [v2]
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2313>
This commit enables the I/O vectorization pass that was originally
written for ACO for Intel drivers. We enable it for UBOs, SSBOs, global
memory, and SLM. We only enable vectorization for the scalar back-end
because it vec4 makes certain alignment assumptions.
Shader-db results with iris on ICL:
total instructions in shared programs: 16077927 -> 16068236 (-0.06%)
instructions in affected programs: 199839 -> 190148 (-4.85%)
helped: 324
HURT: 0
helped stats (abs) min: 2 max: 458 x̄: 29.91 x̃: 4
helped stats (rel) min: 0.11% max: 38.94% x̄: 4.32% x̃: 1.64%
95% mean confidence interval for instructions value: -37.02 -22.80
95% mean confidence interval for instructions %-change: -5.07% -3.58%
Instructions are helped.
total cycles in shared programs: 336806135 -> 336151501 (-0.19%)
cycles in affected programs: 16009735 -> 15355101 (-4.09%)
helped: 458
HURT: 154
helped stats (abs) min: 1 max: 77812 x̄: 1542.50 x̃: 75
helped stats (rel) min: <.01% max: 34.46% x̄: 5.16% x̃: 2.01%
HURT stats (abs) min: 1 max: 22800 x̄: 336.55 x̃: 20
HURT stats (rel) min: <.01% max: 17.11% x̄: 2.12% x̃: 1.00%
95% mean confidence interval for cycles value: -1596.83 -542.49
95% mean confidence interval for cycles %-change: -3.83% -2.82%
Cycles are helped.
total sends in shared programs: 814177 -> 809049 (-0.63%)
sends in affected programs: 15422 -> 10294 (-33.25%)
helped: 324
HURT: 0
helped stats (abs) min: 1 max: 256 x̄: 15.83 x̃: 2
helped stats (rel) min: 1.33% max: 67.90% x̄: 21.21% x̃: 15.38%
95% mean confidence interval for sends value: -19.67 -11.98
95% mean confidence interval for sends %-change: -23.03% -19.39%
Sends are helped.
LOST: 7
GAINED: 2
Most of the helped shaders were in the following titles:
- Doom
- Deus Ex: Mankind Divided
- Aztec Ruins
- Shadow of Mordor
- DiRT Showdown
- Tomb Raider (Rise, I think)
Five of the lost programs are SIMD16 shaders we lost from dirt showdown.
The other two are compute shaders in Aztec Ruins which switched from
SIMD8 to SIMD16.
Vulkan pipeline-db stats on ICL:
Instructions in all programs: 296780486 -> 293493363 (-1.1%)
Loops in all programs: 149669 -> 149669 (+0.0%)
Cycles in all programs: 90999206722 -> 88513844563 (-2.7%)
Spills in all programs: 1710217 -> 1730691 (+1.2%)
Fills in all programs: 1931235 -> 1958138 (+1.4%)
By far the most help was in the Tomb Raider games. A couple of Batman
games with DXVK were also helped. In Shadow of the Tomb Raider:
Instructions in all programs: 41614336 -> 39408023 (-5.3%)
Loops in all programs: 32200 -> 32200 (+0.0%)
Cycles in all programs: 1875498485 -> 1667034831 (-11.1%)
Spills in all programs: 196307 -> 214945 (+9.5%)
Fills in all programs: 282736 -> 307113 (+8.6%)
Benchmarks of real games I've done on this patch:
- Rise of the Tomb Raider: +3%
- Shadow of the Tomb Raider: +10%
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4367>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4367>
Thanks to the NIR vectorizing pass, we're about to see alignments that
are higher than the bit size. Previously, we could use either and we
just happened to choose alignment (probably the wrong choice) so it's
harmless to switch to detecting based on bit size. This commit changes
things to take both into account which is more accurate to what the
messages we're using do. We also beef up the asserts and make them more
consistent, more accurate, and more complete.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4367>
This change incurs a small amount of hurt now, but it enables a lot of
benefit on vec4 shaders on the next commit. nir_opt_algebraic_late
converts dph, dot3, etc. to dhp_replicated, dot_replicated3, etc. In
the process, it introduces extra moves. If the original NIR contained
vec1 32 ssa_45 = fdot4 ssa_51, ssa_44
vec1 32 ssa_46 = fneg ssa_45
nir_opt_algebraic_late will produce
vec4 32 ssa_18 = fdot_replicated4 ssa_1, ssa_15
vec1 32 ssa_19 = mov ssa_18.x
vec1 32 ssa_17 = fneg ssa_19
The algebraic pass added in the next commit can't see through the move
to know that the fneg applies to a fdot_replicated4.
Haswell, Ivy Bridge, and Sandybridge had similar results. (Haswell shown)
total cycles in shared programs: 187077604 -> 187079858 (<.01%)
cycles in affected programs: 350132 -> 352386 (0.64%)
helped: 174
HURT: 194
helped stats (abs) min: 2 max: 124 x̄: 23.60 x̃: 16
helped stats (rel) min: 0.12% max: 15.88% x̄: 4.98% x̃: 3.86%
HURT stats (abs) min: 2 max: 164 x̄: 32.78 x̃: 16
HURT stats (rel) min: 0.17% max: 22.82% x̄: 6.46% x̃: 0.86%
95% mean confidence interval for cycles value: 2.04 10.21
95% mean confidence interval for cycles %-change: 0.17% 1.93%
Cycles are HURT.
No shader-db changes on any other Intel platform.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/1359>
When we originally wrote a bunch of the allocation data structures, we
re-used the GPU memory for CPU-side data structures. It's a bit more
memory efficient and usually ok. However, this has a couple of
problems:
1. It makes it MUCH more likely that the GPU will accidentlly stomp
CPU-side data structures and cause nearly impossible to debug
crashes.
2. With discrete GPUs, the memory will be mapped somehow and that map
may be across the BAR so it could have horribly slow CPU access.
This is bad for our CPU-side data structures.
In the case of anv_state_stream, it also made the data structure
massively more complex than it needed to be.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4336>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4336>
If we have an allocation that's exactly the block size, we end up
computing a new block size to allocate that's exactly the block size,
add in the header, and then assert fail. When computing the block size,
we need to account for the header.
Fixes: 955127db93 "anv/allocator: Add support for large stream..."
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4336>
Rafael Antognolli