[MachineSink][AArch64] Sink instruction copies when they can replace copy into hard register or folded into addressing mode

This patch adds a new code transformation to the `MachineSink` pass,
that

[MachineSink][AArch64] Sink instruction copies when they can replace copy into hard register or folded into addressing mode

This patch adds a new code transformation to the `MachineSink` pass,
that tries to sink copies of an instruction, when the copies can be folded
into the addressing modes of load/store instructions, or
replace another instruction (currently, copies into a hard register).

The criteria for performing the transformation is that:
* the register pressure at the sink destination block must not
exceed the register pressure limits
* the latency and throughput of the load/store or the copy must not deteriorate
* the original instruction must be deleted

Reviewed By: dmgreen

Differential Revision: https://reviews.llvm.org/D152828

show more ...

[AArch64] Convert some tests to opaque pointers (NFC)

Nothing interesting, but required some fixups to MIR.

[AArch64] Name instructions in test (NFC)

And regenerate check lines.

[Aarch64] Correct register class for pseudo instructions

This constrains the Mov* and similar pseudo instruction to take
GPR64common register classes rather than GPR64. GPR64 includs XZR
which is in

[Aarch64] Correct register class for pseudo instructions

This constrains the Mov* and similar pseudo instruction to take
GPR64common register classes rather than GPR64. GPR64 includs XZR
which is invalid here, because this pseudo instructions expands
into an adrp/add pair sharing a destination register. XZR is invalid
on add and attempting to encode it will instead increment the stack
pointer causing crashes (downstream report at [1]). The test case
there reproduces on LLVM11, but I do not have a test case that
reaches this code path on main, since it is being masked by
improved dead code elimination introduced in D91513. Nevertheless,
this seems like a good thing to fix in case there are other cases
that dead code elimination doesn't clean up (e.g. if `optnone` is
used and the optimization is skipped).

I think it would be worth auditing uses of GPR64 in pseudo
instructions to see if there are any similar issues, but I do not
have a high enough view of the backend or knowledge of the
Aarch64 architecture to do this quickly.

[1] https://github.com/JuliaLang/julia/issues/39818

Reviewed By: t.p.northover

Differential Revision: https://reviews.llvm.org/D97435

show more ...

CodeGen: Print/parse LLTs in MachineMemOperands

This will currently accept the old number of bytes syntax, and convert
it to a scalar. This should be removed in the near future (I think I
converted

CodeGen: Print/parse LLTs in MachineMemOperands

This will currently accept the old number of bytes syntax, and convert
it to a scalar. This should be removed in the near future (I think I
converted all of the tests already, but likely missed a few).

Not sure what the exact syntax and policy should be. We can continue
printing the number of bytes for non-generic instructions to avoid
test churn and only allow non-scalar types for generic instructions.

This will currently print the LLT in parentheses, but accept parsing
the existing integers and implicitly converting to scalar. The
parentheses are a bit ugly, but the parser logic seems unable to deal
without either parentheses or some keyword to indicate the start of a
type.

show more ...

[MachineLICM][MachineSink] Move SinkIntoLoop to MachineSink.

This moves SinkIntoLoop from MachineLICM to MachineSink. The motivation for
this work is that hoisting is a canonicalisation transformati

[MachineLICM][MachineSink] Move SinkIntoLoop to MachineSink.

This moves SinkIntoLoop from MachineLICM to MachineSink. The motivation for
this work is that hoisting is a canonicalisation transformation, but we do not
really have a good story to sink instructions back if that is better, e.g. to
reduce live-ranges, register pressure and spilling. This has been discussed a
few times on the list, the latest thread is:

https://lists.llvm.org/pipermail/llvm-dev/2020-December/147184.html

There it was pointed out that we have the LoopSink IR pass, but that works on
IR, lacks register pressure informatiom, and is focused on profile guided
optimisations, and then we have MachineLICM and MachineSink that both perform
sinking. MachineLICM is more about hoisting and CSE'ing of hoisted
instructions. It also contained a very incomplete and disabled-by-default
SinkIntoLoop feature, which we now move to MachineSink.

Getting loop-sinking to do something useful is going to be at least a 3-step
approach:

1) This is just moving the code and is almost a NFC, but contains a bug fix.
This uses helper function `isLoopInvariant` that was factored out in D94082 and
added to MachineLoop.
2) A first functional change to make loop-sink a little bit less restrictive,
which it really is at the moment, is the change in D94308. This lets it do
more (alias) analysis using functions in MachineSink, making it a bit more
powerful. Nothing changes much: still off by default. But it shows that
MachineSink is a better home for this, and it starts using its functionality
like `hasStoreBetween`, and in the next step we can use `isProfitableToSinkTo`.
3) This is the going to be he interesting step: decision making when and how
many instructions to sink. This will be driven by the register pressure, and
deciding if reducing live-ranges and loop sinking will help in better
performance.
4) Once we are happy with 3), this should be enabled by default, that should be
the end goal of this exercise.

Differential Revision: https://reviews.llvm.org/D93694

show more ...