v2: Add brw_ir_performance.cpp and brw_fs_generator.cpp changes. Fix
overlapping register allocation (via has_source_and_destination_hazard). Fix
incorrect destination register file encoding.
v3: Prevent lower_regioning from trying to "fix" DPAS sources.
v4: Add instruction latency information for scheduling and perf
estimates.
v5: Remove all mention of DPASW. Suggested by Curro and Caio. Update
the comment in fs_inst::has_source_and_destination_hazard. Suggested
by Caio.
v6: Add some comments near the src2 calculation in
fs_inst::size_read. Suggested by Caio.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25994>
Once NIR code is lowered and a few optimization passes have run, there
might be flag register interactions between instructions quite far
away from one another.
In the following case :
f0 = and r0, r1
...
fs_interpolate r2, r3
...
if f0
...
endif
If we lower fs_inteporlate while using the f0 register, we completely
garble the value meant for the if block.
To fix this, emit the predication for fs_interpolate in brw_fs_nir.cpp
when doing the NIR translation to the backend IR. This will guarantee
that the flag register interactions are visible to the optimization
passes, avoiding the problem above.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 68027bd38e ("intel/fs: implement dynamic interpolation mode for dynamic persample shaders")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9757
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26306>
This implements the extended 10-bit encoding of the software
scoreboard information used by Xe2 platforms. The new encoding is
different enough that there are few opportunities for sharing code
during translation to machine code, but the high-level tgl_swsb
representation remains roughly the same.
Among other changes the 10-bit SWSB format provides 5 bits worth of
SBID tokens (though they're only usable in large GRF mode) instead of
4 bits, the extended math pipeline is handled as an in-order (RegDist)
pipeline instead of as an out-of-order one, and the dual-argument
encodings support additional combinations of RegDist and SBID
synchronization modes. A new encoding is introduced for preventing
the accumulator hardware scoreboard from being updated, but this is
currently not needed.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25514>
This is what most logical SEND messages do when they take a variable
number of components. 'inst->mlen' is expected to be zero for logical
SEND opcodes, which are expected to behave like plain arithmetic
operations, so certain automated transformations (like SIMD lowering)
can manipulate them without opcode-specific special-casing.
Guessing the number of components from 'inst->mlen' has other
disadvantages, because it requires duplicating the logic that infers
the message payload size in every use of the instruction -- Instead we
can just do the computation once during logical send lowering. In
addition on LNL platform this causes the 'inst->mlen' field of URB
writes to have units inconsistent with every other SEND instruction,
which is likely to lead to confusion and bugs down the road.
Rework:
* Marcin: update emit_urb_indirect_vec4_write
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25195>
Purely from the backend point of view it's just an additional
parameter to sampler messages.
Signed-off-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/23882>
We can lower FS_OPCODE_UNIFORM_PULL_CONSTANT_LOAD into other more
generic sends and drop this internal opcode.
The idea behind this change is to allow bindless surfaces to be used
for UBO pulls and why it's interesting to be able to reuse
setup_surface_descriptors(). But that will come in a later change.
No shader-db changes on TGL & DG2.
Signed-off-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/20416>
These are handled identically in almost all cases. There is one place
in the legacy surface lowering that was obtaining the bitsize from the
opcode, but the LSC-based lowering uses (type_sz(inst->dst.type) * 8)
for that and works just fine. If we just do that in the legacy lowering
too, then we don't need this plethora of opcodes.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rohan Garg <rohan.garg@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20604>
The only reason for the separate opcode was because of the overlapping
BRW_AOP_* enums, making it impossible to tell whether a particular AOP
was the integer or float operation. Now that we use the lsc_opcode
enums, we can just have the legacy lowering inspect the opcode and
select the right descriptor. No need for a separate opcode.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rohan Garg <rohan.garg@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20604>
In the previous patch we adjusted the scoreboard pass to take into
consideration a new case of unordered operations for TGL. Fix the
decoding as well.
v2: use intel_device_info_is_mtl() (Curro, Jordan)
v3: the part where we export num_sources_from_inst() is now a separate patch
(Curro).
v4: Work around false positive maybe-unitialized warning since Marge
uses -Werror=maybe-uninitialized (Marge).
Reviewed-by: Francisco Jerez <currojerez@riseup.net> (v3)
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20072>
In 4ceaed7839 we made scratch surface state allocations part of the
internal heap (mapped to STATE_BASE_ADDRESS::SurfaceStateBaseAddress)
so that it doesn't uses slots in the application's expected 1M
descriptors (especially with vkd3d-proton).
But all our compiler code relies on BSS
(STATE_BASE_ADDRESS::BindlessSurfaceStateBaseAddress).
The additional issue is that there is only 26bits of surface offset
available in CS instruction (CFE_STATE, 3DSTATE_VS, etc...) for
scratch surfaces. So we need the drivers to put the scratch surfaces
in the first chunk of STATE_BASE_ADDRESS::SurfaceStateBaseAddress
(hence all the driver changes).
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 4ceaed7839 ("anv: split internal surface states from descriptors")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7687
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19727>
The changes to fs_visitor::validate() helped track down a place where I
initially forgot to convert a message to the new sources layout. This
had caused a different validation failure in
dEQP-GLES31.functional.tessellation.tesscoord.triangles_equal_spacing,
but this were not detected until after SENDs were lowered.
Tiger Lake, Ice Lake, and Skylake had similar results. (Ice Lake shown)
total instructions in shared programs: 19951145 -> 19951133 (<.01%)
instructions in affected programs: 2429 -> 2417 (-0.49%)
helped: 8 / HURT: 0
total cycles in shared programs: 858904152 -> 858862331 (<.01%)
cycles in affected programs: 5702652 -> 5660831 (-0.73%)
helped: 2138 / HURT: 1255
Broadwell
total cycles in shared programs: 904869459 -> 904835501 (<.01%)
cycles in affected programs: 7686744 -> 7652786 (-0.44%)
helped: 2861 / HURT: 2050
Tiger Lake, Ice Lake, and Skylake had similar results. (Ice Lake shown)
Instructions in all programs: 141442369 -> 141442032 (-0.0%)
Instructions helped: 337
Cycles in all programs: 9099270231 -> 9099036492 (-0.0%)
Cycles helped: 40661
Cycles hurt: 28606
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17605>
The lowering is currently fake. It just changes the opcode from the
_LOGICAL version to the non-_LOGICAL version.
v2: Remove some rebase cruft. 's/gfx8_//;s/simd8_/' in
brw_instruction_name. Both suggested by Ken.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17379>
An argument could be made that all stage-specific opcodes for vec4
stages should be prefixed with VEC4_ like the stage-agnostic opcodes.
I'll leave those additional sed jobs for another day.
egrep -lr '(VS|GS|TCS)_OPCODE_URB_WRITE' src |\
while read f; do
sed --in-place 's/\(VS\|GS\|TCS\)_OPCODE_URB_WRITE/VEC4_\1_OPCODE_URB_WRITE/g' $f
done
egrep -lr 'T.S_OPCODE[_A-Z]*URB_OFFSETS' src |\
while read f; do
sed --in-place 's/\(T.S_OPCODE[_A-Z]*URB_OFFSETS\)/VEC4_\1/g' $f
done
Suggested-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17379>
We haven't exposed this intrinsic as it doesn't directly correspond to
anything in SPIR-V. However, it's used internally by some NIR passes,
namely nir_opt_uniform_atomics().
We reuse most of the infrastructure in brw_find_live_channel, but with
LZD/ADD instead of FBL. A new SHADER_OPCODE_FIND_LAST_LIVE_CHANNEL is
like SHADER_OPCODE_FIND_LIVE_CHANNEL but from the other side.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15484>
In the future we'll want to reuse this intrinsic to deal with ray
queries. Ray queries will use a different global pointer and
programmatically change the control/level arguments of the trace send
instruction.
v2: Comment on barrier after sync trace instruction (Caio)
Generalize lsc helper (Caio)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13719>
We'll want different types of IDs based on topology. Let's make this
more flexible and also move the bit shifting code a layer above where
it's easier to do bitshifting operations, especially if you need to
stash things into temporary registers.
v2: Keep previous comment.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13719>
This allows, for example, fs_inst::components_read() without passing
devinfo as extra argument.
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11766>
on GFX8 onwards, we have only single bit to determine correct return
format.
v2:
- Define macro and use it instead of hardcoded value. (Lionel)
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11766>
When they re-arranged all the dataport stuff and added the LSC, doing
URB fencing through the dataport no longer makes sense. Instead, there
is now a fence message on the URB shared function.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Tested-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13092>
v2: Very significant rebase on changes to previous commits.
Specifically, brw_fs_nir.cpp changes were pretty much rewritten from
scratch after changing the NIR opcode names and types.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12142>
Xe-HPG comes with a massively reworked dataport. The new thing, called
Load/Store Cache or LSC, has a significantly improved interface.
Instead of bespoke messages for every case, there's basically one or two
messages with different bits to control things like address size, how
much data is read/written, etc. It's way nicer but also means we get to
rewrite all our dataport encoding/decoding code. This patch kicks off
the party with all of the new enums.
v2 (Jason Ekstrand, Mark Janes):
- Rename to LSC
v3 (Jason Ekstrand):
- Add numbers to all enums
Co-authored-by: Mark Janes <mark.a.janes@intel.com>
Co-authored-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11600>
This is the vec4 equivalent of d0d039a4d3, required for proper UBO
pushing in vertex stages for Vulkan on HSW. Sadly, the implementation
requires us to do everything in ALIGN1 mode and the vec4 instruction
scheduler doesn't understand HW_GRF <-> UNIFORM interference so it's
easier to do the whole thing in the generator. We add an instruction
to the top of the program which just means "emit the blob" and all the
magic happens in codegen.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10571>
The way we handle spilling for fp64 in vec4 is to emit a series of MOVs
which swizzles the data around and then a pair of 32-bit spills. This
works great except that the next time we go to pick a spill reg, the
compiler isn't smart enough to figure out that the register has already
been spilled. Normally we do this by looking at the sources of spill
instructions (or destinations of fills) but, because it's separated from
the actual value by a MOV, we can't see it. This commit adds a new
opcode VEC4_OPCODE_MOV_FOR_SCRATCH which is identical to MOV in
semantics except that it lets RA know not to spill again.
Fixes: 82c69426a5 "i965/vec4: support basic spilling of 64-bit registers"
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10571>
The execution units of XeHP platforms have multiple asynchronous ALU
pipelines instead of (as far as software is concerned) the single
in-order pipeline that handled most ALU instructions except for
extended math in the original Xe. It's now the compiler's
responsibility to identify cross-pipeline dependencies and insert
synchronization annotations whenever necessary, which are encoded as
some additional bits of the SWSB instruction field.
This commit represents the cross-pipeline synchronization annotations
as part of the existing tgl_swsb structure used for codegen. The
existing tgl_swsb_*() helpers used by hand-crafted assembly are
extended to default to TGL_PIPE_ALL big-hammer synchronization in
order to ensure backwards compatibility with the existing assembly.
The following commits will extend the software scoreboard lowering
pass in order to keep track of cross-pipeline dependencies across IR
instructions, and insert more specific pipeline annotations in the
SWSB field.
The disassembler is also extended here to print out any existing
pipeline sync annotations.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10000>
