You usually want to go find the highest pressure and figure out why you
couldn't spill or what pattern led to a bunch of pressure leading to that
point.
Fixes: 58bcebd987 ("spirv: Allow [i/u]mulExtended to use new nir opcode")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Now that we have a loop unrolling cost function and loop unrolling isn't
going to kill us the moment we have a 64-bit op in a loop, we can go
ahead and move 64-bit lowering later. This gives us the opportunity to
do more optimizations and actually let the full optimizer run even on
64-bit ops rather than hoping one round of opt_algebraic will fix
everything. This substantially reduces both fp64 shader compile times
and the resulting code size.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Now that we have a loop unrolling cost function and loop unrolling isn't
going to kill us the moment we have a 64-bit op in a loop, we can go
ahead and move 64-bit lowering later. This gives us the opportunity to
do more optimizations and actually let the full optimizer run even on
64-bit ops rather than hoping one round of opt_algebraic will fix
everything. This substantially reduces both fp64 shader compile times
and the resulting code size. On the vs-isnan-dvec test from piglit:
Before this commit:
1684.63s user 17.29s system 99% cpu 28:28.24 total
101479 instructions. 0 loops. 802452 cycles. 79:369 spills:fills.
Peak memory usage (according to massif): 1.435 GB
After this commit:
179.64s user 7.75s system 99% cpu 3:07.92 total
57316 instructions. 0 loops. 459287 cycles. 0:0 spills:fills.
Peak memory usage (according to massif): 531.0 MB
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Instead of trusting the caller to already have created a softfp64
function shader and added all its functions to our shader, we simply
take the softfp64 shader as an argument and do the function inlining
ouselves. This means that there's no more nasty functions lying around
that the caller needs to worry about cleaning up.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
This pulls the guts of function inlining into a builder helper so that
it can be used elsewhere. The rest of the infrastructure is still
needed for most inlining cases to ensure that everything gets inlined
and only ever once. However, there are use-cases where you just want to
inline one little thing. This new helper also has a neat trick where it
can seamlessly inline a function from one nir_shader into another.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
This doesn't really change anything as the functions will all get
inlined anyway. However it does let us do a bit of the work earlier and
in a common place.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Even though this is technically a step in the function inlining process
as laid out in nir_inline_functions.c, it's not really needed. We
already have constant initializers lowered here and no new ones are
added by appending the softfp64 functions.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Instead of looking the devinfo directly, look at the lowering options we
provided to NIR. This is more accurate as it's now checking for "do we
need full software lowering" rather than a hardware bit.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
The lowering we do for 64-bit instructions can cause a single NIR ALU
instruction to blow up into hundreds or thousands of instructions
potentially with control flow. If loop unrolling isn't aware of this,
it can unroll a loop 20 times which contains a nir_op_fsqrt which we
then lower to a full software implementation based on integer math.
Those 20 invocations suddenly get a lot more expensive than NIR loop
unrolling currently expects. By giving it an approximate estimate
function, we can prevent loop unrolling from going to town when it
shouldn't.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
We already have one internally for int64 but we don't have a similar one
for doubles so we'll have to make one.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
This is set to True only for numeric conversion opcodes.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
In the old code, we would generate the exact same instruction for
extract_u8(some_u64, 0) and extract_u8(some_u64, 1). The mask-a-word
trick only works for even numbered bytes.
This fixes the (new) piglit test
tests/spec/arb_gpu_shader_int64/execution/fs-ushr-and-mask.shader_test.
v2: Use a SHR instead of an AND. This saves an instruction compared to
using two moves. Suggested by Jason.
Fixes: 6ac2d16901 ("i965/fs: Fix extract_i8/u8 to a 64-bit destination")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
The parameter is never used, and it's not part of a common interface
idiom. Remove it.
src/intel/compiler/brw_interpolation_map.c: In function ‘brw_setup_vue_interpolation’:
src/intel/compiler/brw_interpolation_map.c:62:59: warning: unused parameter ‘devinfo’ [-Wunused-parameter]
const struct gen_device_info *devinfo)
^~~~~~~
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
In 61e009d2c4 we changed the number of components in the
vulkan_resource_index intrinsic and forgot the update Radv's code for
it.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 61e009d2c4 ("spirv: Use the same types for resource indices as pointers")
Reviewed-by: Samuel Pitoiset samuel.pitoiset@gmail.com
Not complete, mostly just adding things as I encounter them in CTS. But
not getting far enough yet to hit most of the OpenCL.std instructions.
Anyway, this is better than nothing and covers the most common builtins.
v2: add hadd proof from Jason
move some of the lowering into opt_algebraic and create new nir opcodes
simplify nextafter lowering
fix normalize lowering for inf
rework upsample to use nir_pack_bits
add missing files to build systems
v3: split lines of iadd/sub_sat expressions
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
v2: use formula with fewer operations
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
v2: add assert in else clause
make local group intrinsics 32 bit wide
v3: always use 32 bit constant for local_size
v4: add comment by Jason
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
we define it inside 'include/c99_math.h' so it is safe to use.
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
The idea is that for repeated use of the same uniform, we could avoid
loading it on each consumer. The results look pretty good.
total instructions in shared programs: 6413571 -> 6521464 (1.68%)
total threads in shared programs: 154214 -> 154000 (-0.14%)
total uniforms in shared programs: 2393604 -> 2119629 (-11.45%)
total spills in shared programs: 4960 -> 4984 (0.48%)
total fills in shared programs: 6350 -> 6418 (1.07%)
Once we do scheduling at the NIR level, the register pressure (and thus
also instructions) issues we see here will drop back down.
This feels like the right tradeoff for threads vs uniforms, particularly
given that we often have very short thread segments right now:
total instructions in shared programs: 6411504 -> 6413571 (0.03%)
total threads in shared programs: 153946 -> 154214 (0.17%)
total uniforms in shared programs: 2387665 -> 2393604 (0.25%)
On each iteration of successfully spilling a reg, we'd allocate another
copy of temp_registers, and when decrementing thread conut we'd allocate
another copy of the graph. These all got cleaned up on freeing the
compile.
It is already obvious whether the job is building a container or running
a mesa build, so let's drop that prefix so that we can see more
information on the screen (eg. in the jobs list on a pipeline page).
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Previously, only the driver_location was set for all variables,
but constants need to use the location field instead. This change
is necessary because the nine state tracker can produce non-packed
constants whose location needs to be explicitly set.
Signed-Off-By: Timur Kristóf <timur.kristof@gmail.com>
Tested-by: Andre Heider <a.heider@gmail.com>
Tested-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
This patch extracts the interpolation mode translation
into a separate function called ttn_translate_interp_mode,
adds support for TGSI_INTERPOLATE_COLOR which was missing,
and also sets the proper interpolation mode to output
variables, which were not set previously.
Signed-Off-By: Timur Kristóf <timur.kristof@gmail.com>
Tested-by: Andre Heider <a.heider@gmail.com>
Tested-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
v2: fix is_shadow, is_array and txq
Some drivers (eg. iris) need the presence of sampler variables and derefs
so that they can count them to determine the number of samplers used.
This change also makes the output NIR closer to what glsl_to_nir outputs.
Signed-Off-By: Timur Kristóf <timur.kristof@gmail.com>
Tested-by: Andre Heider <a.heider@gmail.com>
Tested-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Previously, FACE was hard-coded as a sysval, but TTN emulated
it incorrectly. Also, POSITION was not supported when it was
a sysval. This patch fixes these by allowing both of them to
be sysvals or inputs, based on driver capabilities. It also
fixes the TGSI FACE emulation based on the TGSI spec.
Signed-Off-By: Timur Kristóf <timur.kristof@gmail.com>
Tested-by: Andre Heider <a.heider@gmail.com>
Tested-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
We'll need to use the same logic in other places, so it makes sense to
have a separate function for this.
Signed-Off-By: Timur Kristóf <timur.kristof@gmail.com>
Tested-by: Andre Heider <a.heider@gmail.com>
Tested-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Minor cleanup to the way system value loads work in tgsi_to_nir.
Signed-Off-By: Timur Kristóf <timur.kristof@gmail.com>
Tested-by: Andre Heider <a.heider@gmail.com>
Tested-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
With this patch, tgsi_to_nir will output NIR that is tailored to
the given pipe, by reading its capabilities and adjusting the NIR code
to those capabilities similarly to how glsl_to_nir works.
It also adds an optimization loop that brings the output NIR in line
with what glsl_to_nir outputs. This is necessary for the same reason
why glsl_to_nir has its own optimization loop: currently not every
driver does these optimizations yet.
For uses which cannot pass a pipe_screen we also keep a variant
called tgsi_to_nir_noscreen which keeps the old behavior.
Signed-Off-By: Timur Kristóf <timur.kristof@gmail.com>
Tested-by: Andre Heider <a.heider@gmail.com>
Tested-by: Rob Clark <robdclark@gmail.com>
Acked-By: Eric Anholt <eric@anholt.net>
