LLVM loves take advantage of the fact that vec3s in OpenCL are 16B
aligned and so it can just read/write them as vec4s. This results in a
LOT of vec4->vec3 casts on loads and stores. One solution to this
problem is to get rid of all vec3 variables.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6871>
This pass attempts to optimize three broad categories of memcpy:
1. Self-copies: These we can discard out-of-hand.
2. Vector copies: It doesn't matter what the vector size is or if the
source and destination have different vector types, it's still easy
enough to emit a load/store pair.
3. Tightly packed copies: In the case where a type is tightly packed
(no padding bits), we can replace the memcpy with a copy_deref
instruction which the optimizer is far better at handling.
This has proven capable of getting rid of many of the memcpy instances
in some rather gnarly OpenCL C kernels I've been looking at, even after
coming out of LLVM's optimizer.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6871>
This adds primarily two passes: One is a lowering pass which turns
these conversion intrinsics into a series of ALU ops. The other is an
optimization pass which attempt to simplify the conversion whenever
possible in the hopes that we can turn it into a "normal" conversion op
which doesn't need special treatment.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6945>
Most of these were originally written by Daniel Stone in the Microsoft
ClOn12 branch, reworked by Jesse Natalie, fixed by Boris Brezillon, and
possibly touched by others along the way. Unfortunately, none of that
is in the commit history thanks to living in the CLOn12 branch.
I ported them to mesa master and further reworked things for better
cosmetics. In particular,
1. They now live in a builder helper rather than in vtn_alu.c.
2. Instead of looping inside each builder helper, we just trust NIR
vector instructions to handle vectors.
3. Lots of re-arranging of the helpers for clarity, better asserting,
and better re-use with the upcoming lowering pass.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6945>
Taken mostly directly from the anv pass. A few anv-specific things that
I could leave in anv aren't included. Specifically on turnip we don't
need to set gl_Layer to 0, and we can handle the case where the FS reads
gl_ViewIndex, so that check is moved into anv.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6514>
In CL 1.2, images are required to be either read-only or write-only. We
can always translate the read-only image ops to texture ops. In CL 2.0
(and an extension), the ability is added to have read-write images but
sampling (with a sampler) is only allowed on read-only images. As long
as we only lower read-only images to texture ops, everything should stay
consistent.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6578>
This is very common for backends -- r600, freedreno, and nir_to_tgsi all
needed versions of it. Make a common intrinsic to use for it with a
shared, slightly-tuned-from-ir3 lowering pass.
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6378>
v2 (Karol):
renamed pathes to paths
use more bool
use _mesa_set_intersects
deduplicated some code
fixed some typos
v3 (Karol):
don't enable structurizer as we do this in vtn now
v4 (Jason):
A few clean-ups due to unstructured NIR changes
v5 (Jason):
Misc whitespace and style cleanups
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Tested-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>
The old pass only handled intrinsic load_constant, while the new
nir_opt_shrink_vectors handles ALU ops, nir load_consts, along with all
the load intrinsics.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6050>
Ideally we'd also handle unused middles of vectors and reswizzle ALU-only
uses of it so we could write fewer channels, but that's future work/
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6050>
Lower them to ACCESS_COHERENT to simplify the backend and
probably give better performance than invalidating or writing back the
entire L0/L1 cache.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4905>
nir_schedule already has a struct for options which makes it more than
just a function declaration. Later patches intend to add more structs to
complement these options. In order to make the code easier to manage,
this moves the nir_scheduler-related parts out of nir.h to their own
header.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5953>
Test cases:
opt_if_simplification - the most trivial test case.
opt_if_simplification_single_source_phi_after_if - tests that
opt_if_simplification correctly handles single-source phis after
the if, found in https://gitlab.freedesktop.org/mesa/mesa/-/issues/3282
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5945>
this is needed for zink and other drivers which can support fragcolor but
not fragdata and want to correctly handle EXT_multiview_draw_buffers
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5687>
a pass which rewrites gl_ClipDistance[n] to an undef if the corresponding
clip plane is disabled in the rasterizer state
this pass is needed for zink to handle api disables of clip planes
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5529>
This uses a meson builtin to handle -fvisibility=hidden. This is nice
because we don't need to track which languages are used, if C++ is
suddenly added meson just does the right thing.
Acked-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4740>
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Copied from anv, replaced state with passing model/range directly.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: D Scott Phillips <d.scott.phillips@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4528>
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>
The pass lowers 1-bit booleans produced by NIR to the native bitsize
of the operations that produce them.
v2: change on lower_load_const_instr after upstream changes. Added
TODO2 to explain it, as it was not properly tested yet (see
already existing TODO) (Neil)
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3885>
SPIR-V generates very granular barriers, however HW and backends might
not necessarily take advantage of those. This pass provides a general
mechanism to combine such barriers.
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3224>
This pass is intended to work around game bugs, only!
It also lowers nir_intrinsic_load_helper_invocation to
nir_intrinsic_is_helper_invocation for consistency.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4047>
To facilitate lowering SSBOs to globals, we need a load_ssbo_address
intrinsic. This intrinsic takes an SSBO index and loads the address in
global memory of the SSBO (likely implemented via a uniform in the
driver). In the future, we'll support bounds checking, but at the moment
this is not supported (this pass should only be used for trusted
contexts at the moment, i.e. contexts without robustness extensions).
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2753>
Using LLVM 8 for ppc64el and 7 for s390x (which hits some coroutine
related issues with LLVM 8).
There are some test failures we need to ignore for now. Also, the
timeout needs to be bumped from the default 30s for some tests, because
they can take longer under emulation.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3643>
Etnaviv also does the same thing, so let's try to avoid repetition here,
and use the same for it code as well.
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Tested-by: Paul Cercueil <paul@crapouillou.net>
This is similar to a scheduler I've written for vc4 and i965, but this
time written at the NIR level so that hopefully it's reusable. A notable
new feature it has is Goodman/Hsu's heuristic of "once we've started
processing the uses of a value, prioritize processing the rest of their
uses", which should help avoid the heuristic otherwise making such
systematically bad choices around getting texture results consumed.
Results for v3d:
total instructions in shared programs: 6497588 -> 6518242 (0.32%)
total threads in shared programs: 154000 -> 152828 (-0.76%)
total uniforms in shared programs: 2119629 -> 2068681 (-2.40%)
total spills in shared programs: 4984 -> 472 (-90.53%)
total fills in shared programs: 6418 -> 1546 (-75.91%)
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com> (v1)
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com> (v2)
v2: Use the DAG datastructure, fold in the scheduling-for-parallelism
patch, include SSA defs in live values so we can switch to bottom-up
if we want.
v3: Squash in improvements from Alejandro Piñeiro for getting V3D to
successfully register allocate on GLES3.1 dEQP. Make sure that
discards don't move after store_output. Comment spelling fix.
v7: run nir_opt_algebraic
v9: rework the callback function
v9: update alignment on all loads/stores, even if they're not vectorized
v10: add tests for 64-bit offsets
v10: add tests for signed offsets
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com> (v9)
This pass combines intersecting, adjacent and identical loads/stores into
potentially larger ones and will be used by ACO to greatly reduce the
number of memory operations.
v2: handle nir_deref_type_ptr_as_array
v3: assume explicitly laid out types for derefs
v4: create less deref casts
v4: fix shared boolean vectorization
v4: fix copy+paste error in resources_different
v4: fix extract_subvector() to pass
nir_load_store_vectorize_test.ssbo_load_intersecting_32_32_64
v4: rebase
v5: subtract from deref/offset instead of scheduling offset calculations
v5: various non-functional changes/cleanups
v5: require less metadata and preserve more
v5: rebase
v6: cleanup and improve dependency handling
v6: emit less deref casts
v6: pass undef to components not set in the write_mask for new stores
v7: fix 8-bit extract_vector() with 64-bit input
v7: cleanup creation of store write data
v7: update align correctly for when the bit size of load/store increases
v7: rename extract_vector to extract_component and update comment
v8: prevent combining of row-major matrix column acceses
v9: rework process_block() to be able to vectorize more
v9: rework the callback function
v9: update alignment on all loads/stores, even if they're not vectorized
v9: remove entry::store_value, since it will not be updated if it's was
from a vectorized load
v9: fix bug in subtract_deref(), causing artifacts in Dishonored 2
v9: handle nir_intrinsic_scoped_memory_barrier
v10: use nir_ssa_scalar
v10: handle non-32-bit offsets
v10: use signed offsets for comparison
v10: improve create_entry_key_from_offset()
v10: support load_shared/store_shared
v10: remove strip_deref_casts()
v10: don't ever pass NULL to memcmp
v10: remove recursion in gcd()
v10: fix outdated comment
v11: use the new nir_extract_bits()
v12: remove use of nir_src_as_const_value in resources_different
v13: make entry key hash function deterministic
v13: simplify mask_sign_extend()
v14: add comment in hash_entry_key() about hashing pointers
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com> (v9)
Lower amul to either imul or imul24, depending on whether 24b is enough
bits to calculate an offset within the thing being dereferenced.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Lowering samplers is needed to produce NIR that can actually be
consumed by some gallium drivers, so it doesn't make sense to
to keep it only in the GLSL code.
This commit introduces nir_lower_samplers to compiler/nir,
while maintains the GL-specific function too.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
This pass expects the shader to be in LCSSA form.
The algorithm is based on 'The Simple Divergence Analysis' from
Diogo Sampaio, Rafael De Souza, Sylvain Collange, Fernando Magno Quintão Pereira.
Divergence Analysis. ACM Transactions on Programming Languages and Systems (TOPLAS)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
The OpenGL and OpenGL ES specs require that implementations clamp the
value of gl_PointSize to an implementation-depedent range. This pass
is useful for any GPU hardware that doesn't do this automatically
for either one or both sides of the range, such as V3D.
v2:
- Turn into a generic NIR pass (Eric).
- Make the pass work before lower I/O so we can use the deref variable
to inspect if we are writing to gl_PointSize (Eric).
- Make the pass take the range to clamp as parameter and allow it
to clamp to both sides of the range or just one side.
- Make the pass report progress.
v3:
- Fix copyright header (Eric)
- use fmin/fmax instead of bcsel to clamp (Eric)
Reviewed-by: Eric Anholt <eric@anholt.net>
v2: add to series
v3: update Makefile.sources
v4: don't remove a comment and break statement
v4: use nir_can_move_instr
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
This is mostly the same as nir_move_load_const() but can also move
undef instructions, comparisons and some intrinsics (being careful with
loops).
v2: actually delete nir_move_load_const.c
v3: fix nir_opt_sink() usage in freedreno
v3: update Makefile.sources
v4: replace get_move_def with nir_can_move_instr and nir_instr_ssa_def
v4: handle if uses
v4: fix handling of nested loops
v5: re-write adjust_block_for_loops
v5: re-write setting of use_block for if uses
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Co-authored-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Eric Anholt <eric@anholt.net>
Most integer operations are omitted because dealing with integer
overflow is hard. There are a few things that could be smarter if there
was a small amount more tracking of ranges of integer types (i.e.,
operands are Boolean, operand values fit in 16 bits, etc.).
The changes to nir_search_helpers.h are included in this patch to
simplify reordering the changes to nir_opt_algebraic.py.
v2: Memoize range analysis results. Without this, some shaders appear
to get stuck in infinite loops.
v3: Rebase on many months of Mesa changes, including 1-bit Boolean
changes.
v4: Rebase on "nir: Drop imov/fmov in favor of one mov instruction".
v5: Use nir_alu_srcs_equal for detecting (a*a). Previously just the SSA
value was compared, and this incorrectly matched (a.x*a.y).
v6: Many code improvements including (but not limited to) better names,
more comments, and better use of helper functions. All suggested by
Caio. Rework the handling of several opcodes to use a table for mapping
source ranges to a result range. This change fixed a bug that caused
fmax(gt_zero, ge_zero) to be incorrectly recognized as ge_zero.
Slightly tighten the range of fmul by recognizing that x*x is gt_zero if
x is gt_zero. Add similar handling for -x*x.
v7: Use _______ in the tables as an alias for unknown. Suggested by
Caio.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
