This patch implements, for aarch64, the following wasm SIMD extensions.
v128.load32_zero and v128.load64_zero instructions
https://github.com/WebAssembly/simd/pull/237
The changes are straightforward:
* no new CLIF instructions. They are translated into an existing CLIF scalar
load followed by a CLIF `scalar_to_vector`.
* the comment/specification for CLIF `scalar_to_vector` has been changed to
match the actual intended semantics, per consulation with Andrew Brown.
* translation from `scalar_to_vector` to aarch64 `fmov` instruction. This
has been generalised slightly so as to allow both 32- and 64-bit transfers.
* special-case zero in `lower_constant_f128` in order to avoid a
potentially slow call to `Inst::load_fp_constant128`.
* Once "Allow loads to merge into other operations during instruction
selection in MachInst backends"
(https://github.com/bytecodealliance/wasmtime/issues/2340) lands,
we can use that functionality to pattern match the two-CLIF pair and
emit a single AArch64 instruction.
* A simple filetest has been added.
There is no comprehensive testcase in this commit, because that is a separate
repo. The implementation has been tested, nevertheless.
* Rewrite interpreter generically
This change re-implements the Cranelift interpreter to use generic values; this makes it possible to do abstract interpretation of Cranelift instructions. In doing so, the interpretation state is extracted from the `Interpreter` structure and is accessed via a `State` trait; this makes it possible to not only more clearly observe the interpreter's state but also to interpret using a dummy state (e.g. `ImmutableRegisterState`). This addition made it possible to implement more of the Cranelift instructions (~70%, ignoring the x86-specific instructions).
* Replace macros with closures
The immediate splitting code contained a bug causing both low and high
to be equal for i128. This is the root cause for
bjorn3/rustc_codegen_cranelift#1097 and likely the only bug preventing
cg_clif from bootstrapping rustc.
In particular, introduce initial support for the MOVI and MVNI
instructions, with 8-bit elements. Also, treat vector constants
as 32- or 64-bit floating-point numbers, if their value allows
it, by relying on the architectural zero extension. Finally,
stop generating literal loads for 32-bit constants.
Copyright (c) 2020, Arm Limited.
As discussed in #2251, in order to be very confident that NaN signaling bits are correctly handled by the compiler, this switches `DataValue` to use Cranelift's `Ieee32` and `Ieee64` structures. This makes it a bit more inconvenient to interpreter Cranelift FP operations but this should change to something like `rustc_apfloat` in the future.
This PR updates the AArch64 ABI implementation so that it (i) properly
respects that v8-v15 inclusive have callee-save lower halves, and
caller-save upper halves, by conservatively approximating (to full
registers) in the appropriate directions when generating prologue
caller-saves and when informing the regalloc of clobbered regs across
callsites.
In order to prevent saving all of these vector registers in the prologue
of every non-leaf function due to the above approximation, this also
makes use of a new regalloc.rs feature to exclude call instructions'
writes from the clobber set returned by register allocation. This is
safe whenever the caller and callee have the same ABI (because anything
the callee could clobber, the caller is allowed to clobber as well
without saving it in the prologue).
Fixes#2254.
This approach is not the best but avoids an extra instruction; perhaps at some point, as mentioned in https://github.com/bytecodealliance/wasmtime/pull/2248, we will add the extra instruction or refactor things in such a way that this `Inst` variant is unnecessary.
This also passes `fixed_frame_storage_size` (previously `total_sp_adjust`)
into `gen_clobber_save` so that it can be combined with other stack
adjustments.
Copyright (c) 2020, Arm Limited.
This adds a new feature experimental_x64 for CLIF tests.
A test is run in the new x64 backend iff:
- either the test doesn't have an x86_64 target requirement, signaling
it must be target agnostic or not run on this target.
- or the test does require the x86_64 target, and the test is marked
with the `experimental_x64` feature.
This required one workaround in the parser. The reason is that the
parser will try to use information not provided by the TargetIsa adapter
for the Mach backends, like register names. In particular, parsing test
may fail before the test runner realizes that the test must not be run.
In this case, we early return an almost-empty TestFile from the parser,
under the same conditions as above, so that the caller may filter out
the test properly.
This also copies two tests from the test suite using the new backend,
for demonstration purposes.
This commit adds arm32 code generation for some IR insts.
Floating-point instructions are not supported, because regalloc
does not allow to represent overlapping register classes,
which are needed by VFP/Neon.
There is also no support for big-endianness, I64 and I128 types.
* clif-util: do not convert `anyhow::Error`s into strings into `anyhow::Error`s
* filetests: Use the debug formatting of `anyhow::Error`s
This provides the full error context, not just the source error's message.
Previously, in #2128, we factored out a common "vanilla 64-bit ABI"
implementation from the AArch64 ABI code, with the idea that this should
be largely compatible with x64. This PR alters the new x64 backend to
make use of the shared infrastructure, removing the duplication that
existed previously. The generated code is nearly (not exactly) the same;
the only difference relates to how the clobber-save region is padded in
the prologue.
This also changes some register allocations in the aarch64 code because
call support in the shared ABI infra now passes a temp vreg in, rather
than requiring use of a fixed, non-allocable temp; tests have been
updated, and the runtime behavior is unchanged.
When storing an argument to a stack location for consumption by a
callee, or storing a return value to an on-stack return slot for
consumption by the caller, the ABI implementation was properly extending
the value but was then performing a store with only the original width.
This fixes the issue by always performing a 64-bit store of the extended
value.
Issue reported by @uweigand (thanks!).
In the Baldrdash (SpiderMonkey) embedding, we must take care to
zero-extend all function arguments to callees in integer registers when
the types are narrower than 64 bits. This is because, unlike the native
SysV ABI, the Baldrdash ABI expects high bits to be cleared. Not doing
so leads to difficult-to-trace errors where high bits falsely tag an
int32 as e.g. an object pointer, leading to potential security issues.
Previously, our pattern-matching for generating load/store addresses was
somewhat limited. For example, it could not use a register-extend
address mode to handle the following CLIF:
```
v2760 = uextend.i64 v985
v2761 = load.i64 notrap aligned readonly v1
v1018 = iadd v2761, v2760
store v1017, v1018
```
This PR adds more general support for address expressions made up of
additions and extensions. In particular, it pattern-matches a tree of
64-bit `iadd`s, optionally with `uextend`/`sextend` from 32-bit values
at the leaves, to collect the list of all addends that form the address.
It also collects all offsets at leaves, combining them.
It applies a series of heuristics to make the best use of the
available addressing modes, filling the load/store itself with as many
64-bit registers, zero/sign-extended 32-bit registers, and/or an offset,
then computing the rest with add instructions as necessary. It attempts
to make use of immediate forms (add-immediate or subtract-immediate)
whenever possible, and also uses the built-in extend operators on add
instructions when possible. There are certainly cases where this is not
optimal (i.e., does not generate the strictly shortest sequence of
instructions), but it should be good enough for most code.
Using `perf stat` to measure instruction count (runtime only, on
wasmtime, after populating the cache to avoid measuring compilation),
this impacts `bz2` as follows:
```
pre:
1006.410425 task-clock (msec) # 1.000 CPUs utilized
113 context-switches # 0.112 K/sec
1 cpu-migrations # 0.001 K/sec
5,036 page-faults # 0.005 M/sec
3,221,547,476 cycles # 3.201 GHz
4,000,670,104 instructions # 1.24 insn per cycle
<not supported> branches
27,958,613 branch-misses
1.006071348 seconds time elapsed
post:
963.499525 task-clock (msec) # 0.997 CPUs utilized
117 context-switches # 0.121 K/sec
0 cpu-migrations # 0.000 K/sec
5,081 page-faults # 0.005 M/sec
3,039,687,673 cycles # 3.155 GHz
3,837,761,690 instructions # 1.26 insn per cycle
<not supported> branches
28,254,585 branch-misses
0.966072682 seconds time elapsed
```
In other words, this reduces instruction count by 4.1% on `bz2`.
We often see patterns like:
```
mov w2, #0xffff_ffff // uses ORR with logical immediate form
add w0, w1, w2
```
which is just `w0 := w1 - 1`. It would be much better to recognize when
the inverse of an immediate will fit in a 12-bit immediate field if the
immediate itself does not, and flip add to subtract (and vice versa), so
we can instead generate:
```
sub w0, w1, #1
```
We see this pattern in e.g. `bz2`, where this commit makes the following
difference (counting instructions with `perf stat`, filling in the
wasmtime cache first then running again to get just runtime):
pre:
```
992.762250 task-clock (msec) # 0.998 CPUs utilized
109 context-switches # 0.110 K/sec
0 cpu-migrations # 0.000 K/sec
5,035 page-faults # 0.005 M/sec
3,224,119,134 cycles # 3.248 GHz
4,000,521,171 instructions # 1.24 insn per cycle
<not supported> branches
27,573,755 branch-misses
0.995072322 seconds time elapsed
```
post:
```
993.853850 task-clock (msec) # 0.998 CPUs utilized
123 context-switches # 0.124 K/sec
1 cpu-migrations # 0.001 K/sec
5,072 page-faults # 0.005 M/sec
3,201,278,337 cycles # 3.221 GHz
3,917,061,340 instructions # 1.22 insn per cycle
<not supported> branches
28,410,633 branch-misses
0.996008047 seconds time elapsed
```
In other words, a 2.1% reduction in instruction count on `bz2`.
Previously, we simply compared the input bool to 0, which forced the
value into a register (usually via a cmp and cset), zero-extended it,
etc. This patch performs the same pattern-matching that branches do to
directly perform the cmp and use its flag results with the csel.
On the `bz2` benchmark, the runtime is affected as follows (measuring
with `perf stat`, using wasmtime with its cache enabled, and taking the
second run after the first compiles and populates the cache):
pre:
1117.232000 task-clock (msec) # 1.000 CPUs utilized
133 context-switches # 0.119 K/sec
1 cpu-migrations # 0.001 K/sec
5,041 page-faults # 0.005 M/sec
3,511,615,100 cycles # 3.143 GHz
4,272,427,772 instructions # 1.22 insn per cycle
<not supported> branches
27,980,906 branch-misses
1.117299838 seconds time elapsed
post:
1003.738075 task-clock (msec) # 1.000 CPUs utilized
121 context-switches # 0.121 K/sec
0 cpu-migrations # 0.000 K/sec
5,052 page-faults # 0.005 M/sec
3,224,875,393 cycles # 3.213 GHz
4,000,838,686 instructions # 1.24 insn per cycle
<not supported> branches
27,928,232 branch-misses
1.003440004 seconds time elapsed
In other words, with this change, on `bz2`, we see a 6.3% reduction in
executed instructions.
