* Convert `scalar_to_vector` to ISLE (AArch64)
Converted the exisiting implementation of `scalar_to_vector` for AArch64 to
ISLE.
Copyright (c) 2022 Arm Limited
* Add support for floats and fix FpuExtend
- Added rules to cover `f32 -> f32x4` and `f64 -> f64x2` for
`scalar_to_vector`
- Added tests for `scalar_to_vector` on floats.
- Corrected an invalid instruction emitted by `FpuExtend` on 64-bit
values.
Copyright (c) 2022 Arm Limited
Introduce a new concept in the IR that allows a producer to create
dynamic vector types. An IR function can now contain global value(s)
that represent a dynamic scaling factor, for a given fixed-width
vector type. A dynamic type is then created by 'multiplying' the
corresponding global value with a fixed-width type. These new types
can be used just like the existing types and the type system has a
set of hard-coded dynamic types, such as I32X4XN, which the user
defined types map onto. The dynamic types are also used explicitly
to create dynamic stack slots, which have no set size like their
existing counterparts. New IR instructions are added to access these
new stack entities.
Currently, during codegen, the dynamic scaling factor has to be
lowered to a constant so the dynamic slots do eventually have a
compile-time known size, as do spill slots.
The current lowering for aarch64 just targets Neon, using a dynamic
scale of 1.
Copyright (c) 2022, Arm Limited.
@yuyang-ok reported via zulip that i128 overflow tests were:
1. different from the interpreter implementation
2. wrong on some of the test cases
This fixes both the tests and the aarch64 implementation and adds the
interpreter to the testsuite.
This defines the full set of 32 128-bit vector registers on s390x.
(Note that the VRs overlap the existing FPRs.) In addition, this
adds support to use all 32 vector registers to implement floating-
point operations, by using vector floating-point instructions with
the 'W' bit set to operate only on the first element.
This part of the vector instruction set mostly matches the old FP
instruction set, with two exceptions:
- There is no vector version of the COPY SIGN instruction. Instead,
now use a VECTOR SELECT with an appropriate bit mask to implement
the fcopysign operation.
- There are no vector version of the float <-> int conversion
instructions where source and target differ in bit size. Use
appropriate multiple conversion steps instead. This also requires
use of explicit checking to implement correct overflow handling.
As a side effect, this version now also implements the i8 / i16
variants of all conversions, which had been missing so far.
For all operations except those two above, we continue to use the
old FP instruction if applicable (i.e. if all operands happen to
have been allocated to the original FP register set), and use the
vector instruction otherwise.
This adds infrastructure to allow implementing call and return
instructions in ISLE, and migrates the s390x back-end.
To implement ABI details, this patch creates public accessors
for `ABISig` and makes them accessible in ISLE. All actual
code generation is then done in ISLE rules, following the
information provided by that signature.
[ Note that the s390x back end never requires multiple slots for
a single argument - the infrastructure to handle this should
already be present, however. ]
To implement loops in ISLE rules, this patch uses regular tail
recursion, employing a `Range` data structure holding a range
of integers to be looped over.
The previous `cls` code was producing wrong results when fed with a -1 i8.
The fix here is to sign extend instead of zero extending since we want
to keep the sign bit as one in order for it to be counted correctly
in the cls instruction
This also merges the interpreter only tests now that aarch64
correctly supports this instruction
* Upgrade to regalloc2 v0.2.3 to get bugfix from bytecodealliance/regalloc2#60.
* Update RELEASES.md.
* Update two compile tests based on slightly shifting regalloc output.
The current lowering helper for `cmpxchg` returns the literal RealReg
`rax` as its result. However, this breaks a number of invariants, and
eventually causes a regalloc panic if used as a blockparam arg (pinned
vregs cannot be used in this way).
In general we have to return regular vregs, not a RealReg, as results of
instructions during lowering. However #4223 added a helper for
`x64_cmpxchg` that returns a literal `rax`.
Fortunately we can do the right thing here by just giving a fresh vreg
to the instruction; the regalloc constraints mean that this vreg is
constrained to `rax` at the instruction (at its def/late point), so the
generator of the instruction need not worry about `rax` here.
If an address expression is given to `to_amode` that is completely
constant (no registers at all), then it will produce an `Amode` that has
the resulting constant as an offset, and `(invalid_reg)` as the base.
This is a side-effect of the way we build up the amode step-by-step --
we're waiting to see a register and plug it into the base field. If we
never get a reg though, we need to generate a constant zero into a
register and use that as the base. This PR adds a `finalize_amode`
helper to do just that.
Fixes#4234.
This resolves an edge-case where mul.i128 with an input that continues
to be live after the instruction could cause an invalid regalloc
constraint (basically, the regalloc did not previously support an
instruction use and def both being constrained to the same physical reg;
and the "mul" variant used for mul.i128 on x64 was the only instance of
such operands in Cranelift).
Causes two extra move instructions in the mul.i128 filetest, but that's
the price to pay for the slightly more general (works in all cases)
handling of the constraints.
RA2 recently removed the need for a dedicated scratch register for
cyclic moves (bytecodealliance/regalloc2#51). This has moderate positive
performance impact on function bodies that were register-constrained, as
it means that one more register is available. In Sightglass, I measured
+5-8% on `blake3-scalar`, at least among current benchmarks.
Previously, the pinned register (enabled by the `enable_pinned_reg`
Cranelift setting and used via the `get_pinned_reg` and `set_pinned_reg`
CLIF ops) was only used when Cranelift was embedded in SpiderMonkey, in
order to support a pinned heap register. SpiderMonkey has its own
calling convention in Cranelift (named after the integration layer,
"Baldrdash").
However, the feature is more general, and should be usable with the
default system calling convention too, e.g. SysV or Windows Fastcall.
This PR fixes the ABI code to properly treat the pinned register as a
globally allocated register -- and hence an implicit input and output to
every function, not saved/restored in the prologue/epilogue -- for SysV
on x86-64 and aarch64, and Fastcall on x86-64.
Fixes#4170.
* Upgrade to regalloc2 0.1.3.
This pulls in bytecodealliance/regalloc2#49, which slightly improves
codegen in some cases where a safepoint (for reference-typed values)
occurs in the same liverange as a register-constrained use. For
example, in bytecodealliance/wasmtime#3785, an extra move instruction
appeared and a callee-save register was used (necessitating a more
expensive prologue) because of suboptimal splitting heuristics, which
this PR fixes. The updated RA2 heuristics appear to have no measured
downsides in existing benchmarks and improve the manually-observed
codegen issue.
* Update filetests where regalloc2 improvement altered behavior with reftypes.
Current codegen had a number of logic errors confusing
NAND with AND WITH COMPLEMENT, and NOR with OR WITH COMPLEMENT.
Add support for the missing z15 instructions and fix logic.
This test was added between the last CI run on #4088 and its merge to
main, and the changes in #4088 (use of constants directly in instruction
via load from constant pool, rather than from a register initialized by
a separate instruction) cause it to fail now.
This PR alters the test to be invariant to regalloc and argument
decisions during lowering, as the test is really checking (per the
comment) that we get two cmoves without an intervening move. As such, it
just matches the instruction opcodes, irrespective of the arguments.
* Allow emitting u64 constants into constant pool.
* Use constant pool for constants on x64 that do not fit in a simm32 and are needed as a RegMem or RegMemImm.
* Fix rip-relative addressing bug in pinsrd emission.
In #3744, we identified that extra `mov` instructions were inserted in
between the `cmov` instructions that CLIF's `select` lowers to. The
switch to regalloc2 resolved this and this test checks that no
intervening `mov`s are inserted. Closes#3744.
The pretty-printing had swapped dst and src2; this was introduced when
we moved to RA2 (sorry about that! IMHO we should do something to
automate the mapping between regalloc arg collection and pretty
printing/emission).
`src2` comes at the end because it has a variable number of register
mentions; this is in line with how many of the other inst formats work.
Actual emitted code was never incorrect, just the pretty-printing.
Updated test golden outputs look correct to me now, including the one
that we saw was incorrect in #3945.
This PR refactors the x64 backend address-mode lowering to use an
incremental-build approach, where it considers each node in a tree of
`iadd`s that feed into a load/store address and, at each step, builds
the best possible `Amode`. It will combine an arbitrary number of
constant offsets (an extension beyond the current rules), and can
capture a left-shifted (scaled) index in any position of the tree
(another extension).
This doesn't have any measurable performance improvement on our Wasm
benchmarks in Sightglass, unfortunately, because the IR lowered from
wasm32 will do address computation in 32 bits and then `uextend` it to
add to the 64-bit heap base. We can't quite lift the 32-bit adds to 64
bits because this loses the wraparound semantics.
(We could label adds as "expected not to overflow", and allow *those* to
be lifted to 64 bit operations; wasm32 heap address computation should
fit this. This is `add nuw` (no unsigned wrap) in LLVM IR terms. That's
likely my next step.)
Nevertheless, (i) this generalizes the cases we can handle, which should
be a good thing, all other things being equal (and in this case, no
compile time impact was measured); and (ii) might benefit non-Wasm
frontends.
Currently, we have partial Spectre mitigation: we protect heap accesses
with dynamic bounds checks. Specifically, we guard against errant
accesses on the misspeculated path beyond the bounds-check conditional
branch by adding a conditional move that is also dependent on the
bounds-check condition. This data dependency on the condition is not
speculated and thus will always pick the "safe" value (in the heap case,
a NULL address) on the misspeculated path, until the pipeline flushes
and recovers onto the correct path.
This PR uses the same technique both for table accesses -- used to
implement Wasm tables -- and for jumptables, used to implement Wasm
`br_table` instructions.
In the case of Wasm tables, the cmove picks the table base address on
the misspeculated path. This is equivalent to reading the first table
entry. This prevents loads of arbitrary data addresses on the
misspeculated path.
In the case of `br_table`, the cmove picks index 0 on the misspeculated
path. This is safer than allowing a branch to an address loaded from an
index under misspeculation (i.e., it preserves control-flow integrity
even under misspeculation).
The table mitigation is controlled by a Cranelift setting, on by
default. The br_table mitigation is always on, because it is part of the
single lowering pseudoinstruction. In both cases, the impact should be
minimal: a single extra cmove in a (relatively) rarely-used operation.
The table mitigation is architecture-independent (happens during
legalization); the br_table mitigation has been implemented for both x64
and aarch64. (I don't know enough about s390x to implement this
confidently there, but would happily review a PR to do the same on that
platform.)
Also fix and extend the current implementation:
- AtomicRMWOp::Clr != AtomicRmwOp::And, as the input needs to be
inverted first.
- Inputs to the cmp for the RMWLoop case are sign-extended when
needed.
- Lower Xchg to Swp.
- Lower Sub to Add with a negated input.
- Added more runtests.
Copyright (c) 2022, Arm Limited.
x64 backend: add lowerings with load-op-store fusion.
These lowerings use the `OP [mem], reg` forms (or in AT&T syntax, `OP
%reg, (mem)`) -- i.e., x86 instructions that load from memory, perform
an ALU operation, and store the result, all in one instruction. Using
these instruction forms, we can merge three CLIF ops together: a load,
an arithmetic operation, and a store.
The recent work in #4061 introduced a notion of "unique uses" for CLIF
values that both simplified the load-op merging rules and allowed
loads to merge in some more places.
Unfortunately there's one factor that PR didn't account for: a unique
use at the CLIF level could become a multiple-use at the VCode level,
when a lowering uses a value multiple times!
Making this less error-prone in general is hard, because we don't know
the lowering in VCode until it's emitted, so we can't ahead-of-time
know that a value will be used multiple times and prevent its
merging. But we *can* know in the lowerings themselves when we're
doing this. At least we get a panic from regalloc when we get this
wrong; no bad code (uninitialized register being read) should ever
come from a backend bug like this.
This is still a bit less than ideal, but for now the fix is: in
`cmp_and_choose` in the x64 backend (which compares values, then
picks one or the other with a cmove), explicitly put values in
registers.
Fixes#4067 (thanks @Mrmaxmeier for the report!).
* Cranelift: fix#3953: rework single/multiple-use logic in lowering.
This PR addresses the longstanding issue with loads trying to merge
into compares on x86-64, and more generally, with the lowering
framework falsely recognizing "single uses" of one op by
another (which would normally allow merging of side-effecting ops like
loads) when there is *indirect* duplication.
To fix this, we replace the direct `value_uses` count with a
transitive notion of uniqueness (not unlike Rust's `&`/`&mut` and how
a `&mut` downgrades to `&` when accessed through another `&`!). A
value is used multiple times transitively if it has multiple direct
uses, or is used by another op that is used multiple times
transitively.
The canonical example of badness is:
```
v1 := load
v2 := ifcmp v1, ...
v3 := selectif v2, ...
v4 := selectif v2, ...
```
both `v3` and `v4` effectively merge the `ifcmp` (`v2`), so even
though the use of `v1` is "unique", it is codegenned twice. This is
why we ~~can't have nice things~~ can't merge loads into
compares (#3953).
There is quite a subtle and interesting design space around this
problem and how we might solve it. See the long doc-comment on
`ValueUseState` in this PR for more justification for the particular
design here. In particular, this design deliberately simplifies a bit
relative to an "optimal" solution: some uses can *become* unique
depending on merging, but we don't design our data structures for such
updates because that would require significant extra costly
tracking (some sort of transitive refcounting). For example, in the
above, if `selectif` somehow did not merge `ifcmp`, then we would only
codegen the `ifcmp` once into its result register (and use that
register twice); then the load *is* uniquely used, and could be
merged. But that requires transitioning from "multiple use" back to
"unique use" with careful tracking as we do pattern-matching, which
I've chosen to make out-of-scope here for now. In practice, I don't
think it will matter too much (and we can always improve later).
With this PR, we can now re-enable load-op merging for compares. A
subsequent commit does this.
* Update x64 backend to allow load-op merging for `cmp`.
* Update filetests.
* Add test for cmp-mem merging on x64.
* Comment fixes.
* Rework ValueUseState analysis for better performance.
* Update s390x filetest: iadd_ifcout cannot merge loads anymore because it has multiple outputs (ValueUseState limitation)
* Address review comments.
This PR switches Cranelift over to the new register allocator, regalloc2.
See [this document](https://gist.github.com/cfallin/08553421a91f150254fe878f67301801)
for a summary of the design changes. This switchover has implications for
core VCode/MachInst types and the lowering pass.
Overall, this change brings improvements to both compile time and speed of
generated code (runtime), as reported in #3942:
```
Benchmark Compilation (wallclock) Execution (wallclock)
blake3-scalar 25% faster 28% faster
blake3-simd no diff no diff
meshoptimizer 19% faster 17% faster
pulldown-cmark 17% faster no diff
bz2 15% faster no diff
SpiderMonkey, 21% faster 2% faster
fib(30)
clang.wasm 42% faster N/A
```
Issue #3963 identified a miscompilation with select in which the second
in the pair of `CMOV`s (one pair per `i128` register) used the wrong
flag. This change fixes the error in the x64 ISLE helper function
emitting these `CMOV` instructions.
This change removes all variants of `load*_complex` and `store*_complex`
from Cranelift; this is a breaking change to the instructions exposed by
CLIF. The complete list of instructions removed is: `load_complex`,
`store_complex`, `uload8_complex`, `sload8_complex`, `istore8_complex`,
`sload8_complex`, `uload16_complex`, `sload16_complex`,
`istore16_complex`, `uload32_complex`, `sload32_complex`,
`istore32_complex`, `uload8x8_complex`, `sload8x8_complex`,
`sload16x4_complex`, `uload16x4_complex`, `uload32x2_complex`,
`sload32x2_complex`.
The rationale for this removal is that the Cranelift backend now has the
ability to pattern-match multiple upstream additions in order to
calculate the address to access. Previously, this was not possible so
the `*_complex` instructions were needed. Over time, these instructions
have fallen out of use in this repository, making the additional
overhead of maintaining them a chore.
Previous changes had ported the difficult "`select` based on an `fcmp`"
patterns to ISLE; this completes porting of `select` by moving over the
final two kinds of patterns:
- `select` based on an `icmp`
- `select` based on a value
* x64: port scalar `fcmp` to ISLE
Implement the CLIF lowering for the `fcmp` to ISLE. This adds a new
type-matcher, `ty_scalar_float`, for detecting uses of `F32` and `F64`.
* isle: rename `vec128` to `ty_vec12`
This refactoring changes the name of the `vec128` matcher function to
follow the `ty_*` convention of the other type matchers. It also makes
the helper an inline function call.
* x64: port vector `fcmp` to ISLE
Fuzz testing identified a lowering case for CLIF's `icmp` in which the
double use of a loaded operand resulted in a register allocation error.
This change manually adds `put_in_xmm` to avoid load-coalescing these
values and includes a CLIF filetest to trigger this issue. Closes#3951.
I opened #3953 to discuss a way in which this kind of mistake (i.e.,
forgetting to add `put_in_*` in certain situations) could be avoided.
Previously, we used the flags of `AND` for `SETcc`. This change uses
`TEST` instead, which discards the AND result but sets the flags needed
for `SETcc`. This reduces register pressure slightly for this sequence.
* x64: port GPR-held `icmp` to ISLE
* x64: port equality `icmp` for i128 type
* x64: port `icmp` for vector types
* x64: rename from_intcc to intcc_to_cc
The `fpcmp` helper in the x64 backend uses `put_in_xmm_mem` for one of
its operands, which allows the compiler to merge a load with the compare
instruction (`ucomiss` or `ucomisd`).
Unfortunately, as we saw in #2576 for the integer-compare case, this
does not work with our lowering algorithm because compares can be
lowered more than once (unlike all other instructions) to reproduce the
flags where needed. Merging a load into an op that executes more than
once is invalid in general (the two loads may observe different values,
which violates the original program semantics because there was only one
load originally).
This does not result in a miscompilation, but instead will cause a panic
at regalloc time because the register that should have been defined by
the separate load is never written (the load is never emitted
separately).
I think this (very subtle, easy to miss) condition was unfortunately not
ported over when we moved the logic in #3682.
The existing fcmp-of-load test in `cmp-mem-bug` (from #2576) does not
seem to trigger it, for a reason I haven't fully deduced. I just added
the verbatim function body (happens to come from `clang.wasm`) that
triggers the bug as a test.
Discovered while bringing up regalloc2 support. It's pretty unlikely to
hit by chance, which is why I think none of our fuzzing has hit it yet.
* x64: port `select` using an FP comparison to ISLE
This change includes quite a few interlocking parts, required mainly by
the current x64 conventions in ISLE:
- it adds a way to emit a `cmove` with multiple OR-ing conditions;
because x64 ISLE cannot currently safely emit a comparison followed
by several jumps, this adds `MachInst::CmoveOr` and
`MachInst::XmmCmoveOr` macro instructions. Unfortunately, these macro
instructions hide the multi-instruction sequence in `lower.isle`
- to properly keep track of what instructions consume and produce
flags, @cfallin added a way to pass around variants of
`ConsumesFlags` and `ProducesFlags`--these changes affect all
backends
- then, to lower the `fcmp + select` CLIF, this change adds several
`cmove*_from_values` helpers that perform all of the awkward
conversions between `Value`, `ValueReg`, `Reg`, and `Gpr/Xmm`; one
upside is that now these lowerings have much-improved documentation
explaining why the various `FloatCC` and `CC` choices are made the
the way they are.
Co-authored-by: Chris Fallin <chris@cfallin.org>
Addresses #3809: when we are asked to create a Cranelift backend with
shared flags that indicate support for SIMD, we should check that the
ISA level needed for our SIMD lowerings is present.
In #3721, we have been discussing what to do about the ARM32 backend in
Cranelift. Currently, this backend supports only 32-bit types, which is
insufficient for full Wasm-MVP; it's missing other critical bits, like
floating-point support; and it has only ever been exercised, AFAIK, via
the filetests for the individual CLIF instructions that are implemented.
We were very very thankful for the original contribution of this
backend, even in its partial state, and we had hoped at the time that we
could eventually mature it in-tree until it supported e.g. Wasm and
other use-cases. But that hasn't yet happened -- to the blame of no-one,
to be clear, we just haven't had a contributor with sufficient time.
Unfortunately, the existence of the backend and lack of active
maintainer now potentially pose a bit of a burden as we hope to make
continuing changes to the backend framework. For example, the ISLE
migration, and the use of regalloc2 that it will allow, would need all
of the existing lowering patterns in the hand-written ARM32 backend to
be rewritten as ISLE rules.
Given that we don't currently have the resources to do this, we think
it's probably best if we, sadly, for now remove this partial backend.
This is not in any way a statement of what we might accept in the
future, though. If, in the future, an ARM32 backend updated to our
latest codebase with an active maintainer were to appear, we'd be happy
to merge it (and likewise for any other architecture!). But for now,
this is probably the best path. Thanks again to the original contributor
@jmkrauz and we hope that this work can eventually be brought back and
reused if someone has the time to do so!
This adds support for all atomic operations that were unimplemented
so far in the s390x back end:
- atomic_rmw operations xchg, nand, smin, smax, umin, umax
- $I8 and $I16 versions of atomic_rmw and atomic_cas
- little endian versions of atomic_rmw and atomic_cas
All of these have to be implemented by a compare-and-swap loop;
and for the $I8 and $I16 versions the actual atomic instruction
needs to operate on the surrounding aligned 32-bit word.
Since we cannot emit new control flow during ISLE instruction
selection, these compare-and-swap loops are emitted as a single
meta-instruction to be expanded at emit time.
However, since there is a large number of different versions of
the loop required to implement all the above operations, I've
implemented a facility to allow specifying the loop bodies
from within ISLE after all, by creating a vector of MInst
structures that will be emitted as part of the meta-instruction.
There are still restrictions, in particular instructions that
are part of the loop body may not modify any virtual register.
But even so, this approach looks preferable to doing everything
in emit.rs.
A few instructions needed in those compare-and-swap loop bodies
were added as well, in particular the RxSBG family of instructions
as well as the LOAD REVERSED in-register byte-swap instructions.
This patch also adds filetest runtests to verify the semantics
of all operations, in particular the subword and little-endian
variants (those are currently only executed on s390x).
