Gfx9 can only have F, but newer GPUs can have F, HF, *D, or *W. The
source and destination types must still match in size.
v2: Simplify the float vs integer logic. Suggested by Ken.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29095>
This bit from the comment should have been a big red flag:
There are currently zero instances of fsign(double(x))*IMM in
shader-db or any test suite, so it is hard to care at this time.
The implementation of that path was incorrect. The XOR instructions
should be predicated like the OR instruction in the non-multiplication
path. As a result, dsign(zero_value) * x will not produce the correct
result.
Instead of fixing this code that is never exercised by anything, replace
it with the simple lowering in NIR.
Ironically, the vec4 implementation is correct. The odds of encountering
an application that is performace limited by dsign performance in vertex
processing stages on Ivy Bridge or Haswell is infinitesimal.
No shader-db changes on any Intel platform.
v2: Delete 's' in emit_fsign as it is now unused.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> [v1]
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29095>
This bit from the comment should have been a big red flag:
There are currently zero instances of fsign(double(x))*IMM in
shader-db or any test suite, so it is hard to care at this time.
The implementation of that path was incorrect. The XOR instructions
should be predicated like the OR instruction in the non-multiplication
path. As a result, dsign(zero_value) * x will not produce the correct
result.
Instead of fixing this code that is never exercised by anything, replace
it with the simple lowering in NIR.
No shader-db or fossil-db changes on any Intel platform.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29095>
Some commas were being skipped, according to history as an attempt
to elide BAD_FILEs, but we still print them, so be consistent. Also
for instructions without any sources, the trailing comma was always
being printed. Fix that too.
Example of instruction output before the change
halt_target(8) (null):UD,
send(8) (mlen: 1) (EOT) (null):UD, 0u, 0u, g126:UD(null):UD NoMask
and after it
halt_target(8) (null):UD
send(8) (mlen: 1) (EOT) (null):UD, 0u, 0u, g126:UD, (null):UD NoMask
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29114>
It was the default to show register pressure for each instruction,
but it gets in the way of cleaner diffs before/after an optimization pass.
Add INTEL_DEBUG=reg-pressure option to show it again.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29114>
The sequential IP cause noise when diffing before/after a pass that
either add or remove instructions.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29114>
Running
KHR-GL46.sparse_texture_clamp_tests.SparseTextureClampLookupColor test
with Zink on Anv we run into an assert :
assert(inst->mlen <= MAX_SAMPLER_MESSAGE_SIZE * reg_unit(devinfo));
Turns out we've not covered all the cases in the SIMD lowering.
It's a bit of a shame to have both files reproduce the same logic.
Will try to think of a better way to extract the layout of the a send
message but that'll be a much bigger rework.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29118>
For CMP/CMPN, use src0 type if destination is null otherwise get the
src0 type register with destination register size.
This fixes dEQP-VK.glsl.builtin_var.frontfacing.* tests cases on Xe2+.
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28679>
Fixes fs-uint-to-float-of-extract-int8.shader_test and
fs-uint-to-float-of-extract-int16.shader_test added by piglit!883.
v2: Expand the comment explaining the potential problem. Suggested by
Caio.
Fixes: e6022281f2 ("intel/elk: Rename files to use elk prefix")
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27891>
Fixes fs-uint-to-float-of-extract-int8.shader_test and
fs-uint-to-float-of-extract-int16.shader_test added by piglit!883.
No shader-db or fossil-db changes on any Intel platform.
v2: Expand the comment explaining the potential problem. Suggested by
Caio.
Fixes: 29ce110be6 ("i965/fs: Remove extract virtual opcodes.")
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27891>
This writes the whole destination register in a single builder call.
Eventually, VEC will write the whole destination register in one go,
allowing better visibility into how it is defined.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28971>
This writes the whole destination register in a single builder call.
Eventually, VEC will write the whole destination register in one go,
allowing better visibility into how it is defined.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28971>
This writes the whole destination register in a single builder call.
Eventually, VEC will write the whole destination register in one go,
allowing better visibility into how it is defined.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28971>
This writes the whole destination register in a single builder call.
Eventually, VEC will write the whole destination register in one go,
allowing better visibility into how it is defined.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28971>
This writes the whole destination register in a single builder call.
Eventually, VEC will write the whole destination register in one go,
allowing better visibility into how it is defined.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28971>
This writes the whole destination register in a single builder call.
Eventually, VEC will write the whole destination register in one go,
allowing better visibility into how it is defined.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28971>
This gathers a number of sources into a contiguous vector register.
Eventually, the plan is that it will use a MOV for a single source,
or LOAD_PAYLOAD for multiple sources. For now, it emits a series of
MOVs to allow us to rewrite a bunch of existing code to use the new
helper, then change them all over at once later.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28971>
When emitting a sampler message, we allocate a temporary destination
large enough to hold 4 values (or 5 for sparse). This is the maximum
size needed to hold any result. However, we shrink the size written by
the sampler message to skip writing any trailing components that NIR
tells us are never read. So we may not write the entire temporary.
The NIR texture instruction has a destination VGRF which is sized
assuming that all components are present. We issue a LOAD_PAYLOAD
instruction to copy our sampler result temporary to the NIR destination.
When we reduce the response length of the sampler messages, then some of
these temporary components have undefined values. The correct way to
indicate that is by using a BAD_FILE source. Unfortunately, we were
naively reading offsets of the temporary that were never written, but
are still part of a larger VGRF. This complicates things.
For example, sampling and only using RGB (not RGBA) was producing this:
txl_logical(8) (written: 3) vgrf3+0.0:F, ...
undef(8) (written: 4) vgrf4:UD
load_payload(8) (written: 4) vgrf4:F, vgrf3+0.0:F, vgrf3+1.0:F, vgrf3+2.0:F, vgrf3+3.0:F
The last source, vgrf3+3.0:F, is undefined, and should be BAD_FILE.
Doing so allows VGRF splitting and other optimizations to work better.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28971>
This has no changes in shader-db or fossil-db, surprisingly, but at
least CSEL will be useful shortly. Presumably the others may matter
somewhere.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28971>
If there's only a single instruction in a basic block, then removing it
would create an empty block. We seem to have trouble representing those
as there are no instructions with an IP inside the block; several places
mess up connections. While most blocks end in control flow instructions
(which are rarely eliminated), ones preceding a DO instruction may end
in an ordinary instruction. This makes such blocks tricky to merge with
adjacent blocks - they may be between loops. Any optimization pass may
may find such an instruction and want to eliminate it, and most of them
are unprepared to perform such CFG link surgery. Nor do we want to make
every pass aware of this issue.
To work around this, we simply replace an instruction with a NOP when
removing it from a block containing only that instruction, leaving the
block in place.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28971>
With the previous commit, we now have new builder helpers that will
allocate a temporary destination for us. So we can eliminate a lot
of the temporary naming and declarations, and build up expressions.
In a number of cases here, the code was confusingly mixing D-type
addresses with UD-immediates, or expecting a UD destination. But the
underlying values should always be positive anyway. To accomodate the
type inference restriction that the base types much match, we switch
these over to be purely UD calculations. It's cleaner to do so anyway.
Compared to the old code, this may in some cases allocate additional
temporary registers for subexpressions.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28957>
In many cases, we calculate an expression by generating a series of
instructions. We'd either overwrite the same register repeatedly,
or call vgrf(BRW_TYPE_X) repeatedly to allocate temporaries for each
intermediate step. In many cases, we overwrote the same register simply
because allocating and naming temporaries for each step was annoying.
This commit adds new builder helpers that will allocate a temporary
destination for you, using simple type interference: unary operations
use the source type, and binary operations require a matching base type
and return the largest of the two types.
The helpers return the destination register, allowing us to write in an
expression-tree style, chaining together builder operations to produce
whole values. Sort of like nir_builder. We still optionally will write
out the fs_inst pointer in case the caller wants to do things like set
predicates or saturation.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28957>
Some instructions can operate on mixed types. Typically this is
something like a binary operation with UD and UW sources resulting
in a UD destination. In order to make it easier to find the result
type of such operations, let's make a type helper that returns the
larger of the two types (but requires the base type to match).
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28957>
We just want to emit an instruction, but we don't need to do anything
further with it, so we don't need to store the resulting inst pointer
anywhere.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28957>
Always select sample barycentric when persample dispatch is unknown at
compile time and let the payload adjustments feed the expected value
based on dispatch.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27803>
Align16 is only used on Gfx9, while Align1 is used on Gfx11+. We can
decode both kinds of encodings in the same function with a simple
devinfo check. One snag is that the align16 encodings didn't have a
separate exec_type field, but we can just pass 0.
This lets us have a single function named brw_type_decode_for_3src,
which is much less of a mouthful.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28847>
Align16 is only used on Gfx9, while Align1 is used on Gfx11+. We can
handle both encodings in the same function with a simple devinfo check,
and give that function a simple name like brw_type_encode_for_3src.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28847>
Both of these helpers do the same thing. We now have brw_type_size_bits
and brw_type_size_bytes and can use whichever makes sense in that place.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28847>
In ancient days, we directly used the hardware register type encodings
throughout the compiler. As more GPU generations came out, encodings
shifted, and we moved to an abstract enum that we could encode/decode
to a particular GPU's hardware encoding. But there was no particular
meaning behind any particular value.
One downside to this approach is that we end up with switch statements
galore. Want to know a type's size? Switch. Convert a unsigned type
to a signed one? Switch. Get a type with the same base type, but
different bit size? Switch. This is both inefficient and inconvenient.
In contrast, nir_alu_type takes a nicer approach - the type encoding has
certain bits representing the base type, and others encoding the size of
the type. Switching base types or sizes is a simple matter of masking
out the relevant field and substituting a different one.
Tigerlake's encoding adopts a similar approach: two bits represent the
size as a 2-bit unsigned number n, where the bit size is (8 * 2^n).
Two more bits represent the base type. Past encodings were a bit ad hoc
as new data types were added over time, but Gfx12 is organized (mostly).
This patch converts our brw_reg_type enum over to a new system that's
patterned after the Tigerlake style (for easy conversion) while
deviating in a few ways that make our vector immediate type size
handling simpler. Should we add additional base types, we're likely
to continue deviating. Still, converting is much simpler.
Type size calculations (which are performed all the time) are now a
simple mask and shift, instead of a switch.
We also adopt the name BRW_TYPE_* instead of BRW_REGISTER_TYPE_* because
it's much shorter and easier to type. Similarly, we create new helper
functions named brw_type_* for working with these types, with a cleaner
naming convention. Legacy names still exist but will we dropped over
the next few patches as pieces get cleaned up.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28847>
Icelake removed the PLN instruction for interpolating fragment shader
inputs, instead adding a special "Native Float" (NF) data type which
was a 66-bit floating point data type that could only be used with the
accumulator. On Tigerlake, they dropped NF support in favor of just
doing the interpolation with MAD instructions.
We stopped using NF years ago (commit 9ea90aae1e),
instead just using the fs_visitor::lower_linterp() pass to emit MADs.
Since this existed only for a short time, and had very limited utility,
we drop it from the compiler. One downside is that we can no longer
disassemble Icelake shaders containing NF types properly, but I doubt
anyone really minds.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28847>
align1 three-source instructions do not exist on gfx9, and this
compiler does not support gfx10. So the oldest case is gfx11.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28847>
