GFX11 has a new message type MSG_DEALLOC_VGPRS which can be used to
release a shader's VGPRs. Sending this at the end of a shader (just
before the s_endpgm) can help overall system performance in cases where
the s_endpgm would have to wait for outstanding VMEM stores to complete
before releasing the VGPRs.
Differential Revision: https://reviews.llvm.org/D128442
Pre gfx1030 null for sdst is different.
c97436f8b6 [AMDGPU] Use null for dead sdst operand - requires a change to make
it not apply to pre gfx1030
Differential Revision: https://reviews.llvm.org/D127869
Compared to permlane16, permlane64 has no BC input because it has no
boundary conditions, no fi input because the instruction acts as if FI
were always enabled, and no OLD input because it always writes to every
active lane.
Also use the new intrinsic in the atomic optimizer pass.
Differential Revision: https://reviews.llvm.org/D127662
Nic Curtis done the experiments to prove it is faster than a
separate mul and add.
Fixes: SWDEV-332806
Differential Revision: https://reviews.llvm.org/D127253
Previously SIFoldOperands::foldInstOperand would only fold a
non-inlinable immediate into a single user, so as not to increase code
size by adding the same 32-bit literal operand to many instructions.
This patch removes that restriction, so that a non-inlinable immediate
will be folded into any number of users. The rationale is:
- It reduces the number of registers used for holding constant values,
which might increase occupancy. (On the other hand, many of these
registers are SGPRs which no longer affect occupancy on GFX10+.)
- It reduces ALU stalls between the instruction that loads a constant
into a register, and the instruction that uses it.
- The above benefits are expected to outweigh any increase in code size.
Differential Revision: https://reviews.llvm.org/D114643
Only fold for uniform values on pre-GFX9 chips. GFX9+ allow us
to keep the calculation entirely on the SALU.
For subtargets where integer multiplication isn't full-rate, avoid
folding if the multiply has too many uses.
Finally, we expand 64x32 and 64x64 multiplies here as well, if they
feed into an addition. This results in better code generation than
the generic expansion for such multiplies because we end up using
the accumulator of the MAD instructions.
Differential Revision: https://reviews.llvm.org/D123835
The WWM register has unmodeled register liveness, For v_set_inactive_*,
clobberring source register is dangerous because it will overwrite the
inactive lanes. When the source vgpr is dead at v_set_inactive_lane,
the inactive lanes may be not really dead. This may make common
optimizations doing wrong.
For example in a simple if-then cfg in Machine IR:
bb.if:
%src =
bb.then:
%src1 = COPY %src
%dst = V_SET_INACTIVE %src1(tied-def 0), %inactive
bb.end
... = PHI [0, %bb.then] [%src, %bb.if]
The register coalescer will think it is safe to optimize "%src1 = COPY %src"
in bb.then. And at the same time, there is no interference for the PHI in
bb.end. The source and destination values of the PHI will be assigned
the same register. The single PHI register will be overwritten by the
v_set_inactive, then we would get wrong value in bb.end.
With this change, we will copy the content of the source register before
setting inactive lanes after register allocation. Yes, this will sacrifice
the WWM code generation a little, but I don't have any better idea to do things
correctly.
Differential Revision: https://reviews.llvm.org/D117482
Select SelectionDAG ops smul_lohi/umul_lohi to
v_mad_i64_i32/v_mad_u64_u32 respectively, with an addend of 0.
v_mul_lo, v_mul_hi and v_mad_i64/u64 are all quarter-rate instructions
so it is better to use one instruction than two.
Further improvements are possible to make better use of the addend
operand, but this is already a strict improvement over what we have
now.
Differential Revision: https://reviews.llvm.org/D113986
The compiler was generating symbols in the final code object for local
branch target labels. This bloats the code object, slows down the loader,
and is only used to simplify disassembly.
Use '--symbolize-operands' with llvm-objdump to improve readability of the
branch target operands in disassembly.
Fixes: SWDEV-312223
Reviewed By: scott.linder
Differential Revision: https://reviews.llvm.org/D114273
Use GCNHazardRecognizer in postra sched.
Updated tests for the new schedules.
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D109536
Change-Id: Ia86ba2ae168f12fb34b4d8efdab491f84d936cde
The requested register class priorities weren't respected
globally. Not sure why this is a target option, and not just the
expected behavior (recently added in
1a6dc92be7). This avoids an allocation
failure when many wide tuple spills are introduced. I think this is a
workaround since I would not expect the allocation priority to be
required, and only a performance hint. The allocator should be smarter
about when only a subregister needs to be spilled and restored.
This does regress a couple of degenerate store stress lit tests which
shouldn't be too important.
the compilation time and there is no case for which we see any improvement in
performance. This patch removes this pass and its associated test cases from
the tree.
Differential Revision: https://reviews.llvm.org/D101313
Change-Id: I0599169a7609c19a887f8d847a71e664030cc141
If the result of an atomic operation is not used then it can be more
efficient to build a reduction across all lanes instead of a scan. Do
this for GFX10, where the permlanex16 instruction makes it viable. For
wave64 this saves a couple of dpp operations. For wave32 it saves one
readlane (which are generally bad for performance) and one dpp
operation.
Differential Revision: https://reviews.llvm.org/D98953
Reuse the existing KnownBits multiplication code to handle the 'extend + multiply + extract high bits' pattern for multiply-high ops.
Noticed while looking at the codegen for D88785 / D98587 - the patch helps division-by-constant expansion code in particular, which suggests that we might have some further KnownBits div/rem cases we could handle - but this was far easier to implement.
Differential Revision: https://reviews.llvm.org/D98857
Add some atomic optimizer tests where there is no use of the result of
the atomic operation, which is a common case in real code. NFC.
Differential Revision: https://reviews.llvm.org/D98952
The hazard where a VMEM reads an SGPR written by a VALU counts as a data
dependency hazard, so no nops are required on GFX10. Tested with Vulkan
CTS on GFX10.1 and GFX10.3.
Differential Revision: https://reviews.llvm.org/D97926
Prefer to keep uniform (non-divergent) multiplies on the scalar ALU when
possible. This significantly improves some game cases by eliminating
v_readfirstlane instructions when the result feeds into a scalar
operation, like the address calculation for a scalar load or store.
Since isDivergent is only an approximation of whether a value is in
SGPRs, it can potentially regress some situations where a uniform value
ends up in a VGPR. These should be rare in real code, although the test
changes do contain a number of examples.
Most of the test changes are just using s_mul instead of v_mul/mad which
is generally better for both register pressure and latency (at least on
GFX10 where sgpr pressure doesn't affect occupancy and vector ALU
instructions have significantly longer latency than scalar ALU). Some
R600 tests now use MULLO_INT instead of MUL_UINT24.
GlobalISel appears to handle more scenarios in the desirable way,
although it can also be thrown off and fails to select the 24-bit
multiplies in some cases.
Alternative solution considered and rejected was to allow selecting
MUL_[UI]24 to S_MUL_I32. I've rejected this because the definition of
those SD operations works is don't-care on the most significant 8 bits,
and this fact is used in some combines via SimplifyDemandedBits.
Based on a patch by Nicolai Hähnle.
Differential Revision: https://reviews.llvm.org/D97063
Enabled "bound_ctrl:1" and disabled "bound_ctrl:-1" syntax.
Corrected printer to output "bound_ctrl:1" instead of "bound_ctrl:0".
See bug 35397 for detailed issue description.
Differential Revision: https://reviews.llvm.org/D97048
The AMD GPU SIMemoryLegalizer was using the ordering address space
rather than the instruction address space when determining the
s_waitcnt to generate to ensure that a read-modify-write atomic has
completed. This resulted in additional unnecessary counters being
waited on.
Differential Revision: https://reviews.llvm.org/D96743
Support for XNACK and SRAMECC is not static on some GPUs. We must be able
to differentiate between different scenarios for these dynamic subtarget
features.
The possible settings are:
- Unsupported: The GPU has no support for XNACK/SRAMECC.
- Any: Preference is unspecified. Use conservative settings that can run anywhere.
- Off: Request support for XNACK/SRAMECC Off
- On: Request support for XNACK/SRAMECC On
GCNSubtarget will track the four options based on the following criteria. If
the subtarget does not support XNACK/SRAMECC we say the setting is
"Unsupported". If no subtarget features for XNACK/SRAMECC are requested we
must support "Any" mode. If the subtarget features XNACK/SRAMECC exist in the
feature string when initializing the subtarget, the settings are "On/Off".
The defaults are updated to be conservatively correct, meaning if no setting
for XNACK or SRAMECC is explicitly requested, defaults will be used which
generate code that can be run anywhere. This corresponds to the "Any" setting.
Differential Revision: https://reviews.llvm.org/D85882
Exec mask manipulation inserted by SIWholeQuadMode barriers to
instruction scheduling. Move the entire pass after the machine
instruction scheduler and make changes so pass is correct for
non-SSA operation. These changes should leave the pass still
usable pre-scheduler, although tests have be updated to reflect
post-scheduler results.
Reviewed By: nhaehnle
Differential Revision: https://reviews.llvm.org/D88081
Passes that are run after the post-RA scheduler may insert instructions like
waitcnt which eliminate the need for certain noops. After this patch the
scheduler is still aware of possible latency from hazards but noops will
not be inserted until the dedicated hazard recognizer pass is run.
Depends on D89753.
Reviewed By: foad
Differential Revision: https://reviews.llvm.org/D89754
This is the slowest operation in the already slow pass.
Instead of sorting just put a stall list into an ordered
map.
Differential Revision: https://reviews.llvm.org/D86253
This was assumed to be a simple move, and interpreting the immediate
modifier operand as a materialized immediate. Apparently the SDWA pass
never produces these, but GlobalISel does emit these for some vector
shuffles.
Enable clausing of memory loads on gfx10 by adding a new pass to insert
the s_clause instructions that mark the start of each hard clause.
Differential Revision: https://reviews.llvm.org/D79792
Add a new llvm.amdgcn.ballot intrinsic modeled on the ballot function
in GLSL and other shader languages. It returns a bitfield containing the
result of its boolean argument in all active lanes, and zero in all
inactive lanes.
This is intended to replace the existing llvm.amdgcn.icmp and
llvm.amdgcn.fcmp intrinsics after a suitable transition period.
Use the new intrinsic in the atomic optimizer pass.
Differential Revision: https://reviews.llvm.org/D65088
Summary:
pickNodeBidirectional tried to compare the best top candidate and the
best bottom candidate by examining TopCand.Reason and BotCand.Reason.
This is unsound because, after calling pickNodeFromQueue, Cand.Reason
does not reflect the most important reason why Cand was chosen. Rather
it reflects the most recent reason why it beat some other potential
candidate, which could have been for some low priority tie breaker
reason.
I have seen this cause problems where TopCand is a good candidate, but
because TopCand.Reason is ORDER (which is very low priority) it is
repeatedly ignored in favour of a mediocre BotCand. This is not how
bidirectional scheduling is supposed to work.
To fix this I changed the code to always compare TopCand and BotCand
directly, like the generic implementation of pickNodeBidirectional does.
This removes some uncommented AMDGPU-specific logic; if this logic turns
out to be important then perhaps it could be moved into an override of
tryCandidate instead.
Graphics shader benchmarking on gfx10 shows a lot more positive than
negative effects from this change.
Reviewers: arsenm, tstellar, rampitec, kzhuravl, vpykhtin, dstuttard, tpr, atrick, MatzeB
Subscribers: jvesely, wdng, nhaehnle, yaxunl, t-tye, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D68338
This reverts commit fe23ed2c68.
It was never really clear this was responsible for the performance
regressions that caused this to be reverted. It's been a long time,
and we need to have scalar patterns for this to get GlobalISel
working.
We are relying on atrificial DAG edges inserted by the
MemOpClusterMutation to keep loads and stores together in the
post-RA scheduler. This does not work all the time since it
allows to schedule a completely independent instruction in the
middle of the cluster.
Removed the DAG mutation and added pass to bundle already
clustered instructions. These bundles are unpacked before the
memory legalizer because it does not work with bundles but also
because it allows to insert waitcounts in the middle of a store
cluster.
Removing artificial edges also allows a more relaxed scheduling.
Differential Revision: https://reviews.llvm.org/D72737
The current implementation of skip insertion (SIInsertSkip) makes it a
mandatory pass required for correctness. Initially, the idea was to
have an optional pass. This patch inserts the s_cbranch_execz upfront
during SILowerControlFlow to skip over the sections of code when no
lanes are active. Later, SIRemoveShortExecBranches removes the skips
for short branches, unless there is a sideeffect and the skip branch is
really necessary.
This new pass will replace the handling of skip insertion in the
existing SIInsertSkip Pass.
Differential revision: https://reviews.llvm.org/D68092
The current implementation of skip insertion (SIInsertSkip) makes it a
mandatory pass required for correctness. Initially, the idea was to
have an optional pass. This patch inserts the s_cbranch_execz upfront
during SILowerControlFlow to skip over the sections of code when no
lanes are active. Later, SIRemoveShortExecBranches removes the skips
for short branches, unless there is a sideeffect and the skip branch is
really necessary.
This new pass will replace the handling of skip insertion in the
existing SIInsertSkip Pass.
Differential revision: https://reviews.llvm.org/D68092
Summary:
- `dead-mi-elimination` assumes MIR in the SSA form and cannot be
arranged after phi elimination or DeSSA. It's enhanced to handle the
dead register definition by skipping use check on it. Once a register
def is `dead`, all its uses, if any, should be `undef`.
- Re-arrange the DIE in RA phase for AMDGPU by placing it directly after
`detect-dead-lanes`.
- Many relevant tests are refined due to different register assignment.
Reviewers: rampitec, qcolombet, sunfish
Subscribers: arsenm, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D72709
This defaults to zero fi operand, but we do not expose it
anyway. Should we expose it later it needs to be added to
the pseudo.
This enables dpp combining on gfx10.
Differential Revision: https://reviews.llvm.org/D68888
llvm-svn: 374604
Jay Foad