DHAT reports that when compiling the Spidermonkey Sightglass benchmark,
there are over 100k of these Vec allocations, averaging less than 4
bytes, and with an average lifetime of only about 500 instructions.
This function is only called from one place, which immediately converts
it into an iterator. So this commit just returns the iterator that was
previously being collected into a Vec. The iterator has to borrow from
the DataFlowGraph, so this would change borrow-check results, but in the
one caller that turns out to be okay.
(That sole caller is in cranelift/codegen/src/machinst/lower.rs, in
Lower::lower().)
According to Sightglass, this is a compile-time improvement of between
2% and 12% on the Spidermonkey benchmark:
instantiation :: nanoseconds :: benchmarks/spidermonkey/benchmark.wasm
Δ = 14628.76 ± 10318.59 (confidence = 99%)
main-0e6ffd024.so is 0.87x to 0.98x faster than no-small-vecs.so!
no-small-vecs.so is 1.02x to 1.14x faster than main-0e6ffd024.so!
[142023 187464.24 301522] main-0e6ffd024.so
[103742 172835.48 263917] no-small-vecs.so
compilation :: nanoseconds :: benchmarks/spidermonkey/benchmark.wasm
Δ = 362392705.93 ± 267070467.06 (confidence = 99%)
main-0e6ffd024.so is 0.89x to 0.98x faster than no-small-vecs.so!
no-small-vecs.so is 1.02x to 1.12x faster than main-0e6ffd024.so!
[3655734131 5522594697.83 6471126699] main-0e6ffd024.so
[3278129811 5160201991.90 5810600015] no-small-vecs.so
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.
* Cranelift: make `ir::Type` a `u16`.
* Cranelift: pack ValueData back into 64 bits.
After extending `Type` to a `u16`, `ValueData` became 12 bytes rather
than 8. This packs it back down to 8 bytes (64 bits) by stealing two
bits from the `Type` for the enum discriminant (leaving 14 bits for the
type itself).
Performance comparison (3-way between original (`ty-u8`), 16-bit `Type`
(`ty-u16`), and this PR (`ty-packed`)):
```
~/work/sightglass% target/release/sightglass-cli benchmark \
-e ~/ty-u8.so -e ~/ty-u16.so -e ~/ty-packed.so \
--iterations-per-process 10 --processes 2 \
benchmarks-next/spidermonkey/benchmark.wasm
compilation
benchmarks-next/spidermonkey/benchmark.wasm
cycles
[20654406874 21749213920.50 22958520306] /home/cfallin/ty-packed.so
[22227738316 22584704883.90 22916433748] /home/cfallin/ty-u16.so
[20659150490 21598675968.60 22588108428] /home/cfallin/ty-u8.so
nanoseconds
[5435333269 5723139427.25 6041072883] /home/cfallin/ty-packed.so
[5848788229 5942729637.85 6030030341] /home/cfallin/ty-u16.so
[5436002390 5683248226.10 5943626225] /home/cfallin/ty-u8.so
```
So, when compiling SpiderMonkey.wasm, making `Type` 16 bits regresses
performance by 4.5% (5.683s -> 5.723s), while this PR gets 14 bits for a 1.0%
cost (5.683s -> 5.723s). That's still not great, and we can likely do better,
but it's a start.
* Fix test failure: entities to/from u32 via `{from,to}_bits`, not `{from,to}_u32`.
The current definition of `ValueSlice` is not usable, since any call to
a constructor returning a `ValueSlice` will extend the mutable borrow
on the context taken by the constructor call, with the result that it
cannot be passed to any other constructor ever.
Re-implement `ValueSlice` as a pair of a `ValueList` identifer plus an
offset into the list. This type can simply be copied without requiring
a borrow on the context.
This also paves the way for unifying TargetIsa and MachBackend, since now they map one to one. In theory the two traits could be merged, which would be nice to limit the number of total concepts. Also they have quite different responsibilities, so it might be fine to keep them separate.
Interestingly, this PR started as removing RegInfo from the TargetIsa trait since the adapter returned a dummy value there. From the fallout, noticed that all Display implementations didn't needed an ISA anymore (since these were only used to render ISA specific registers). Also the whole family of RegInfo / ValueLoc / RegUnit was exclusively used for the old backend, and these could be removed. Notably, some IR instructions needed to be removed, because they were using RegUnit too: this was the oddball of regfill / regmove / regspill / copy_special, which were IR instructions inserted by the old regalloc. Fare thee well!
* Change VMMemoryDefinition::current_length to `usize`
This commit changes the definition of
`VMMemoryDefinition::current_length` to `usize` from its previous
definition of `u32`. This is a pretty impactful change because it also
changes the cranelift semantics of "dynamic" heaps where the bound
global value specifier must now match the pointer type for the platform
rather than the index type for the heap.
The motivation for this change is that the `current_length` field (or
bound for the heap) is intended to reflect the current size of the heap.
This is bound by `usize` on the host platform rather than `u32` or`
u64`. The previous choice of `u32` couldn't represent a 4GB memory
because we couldn't put a number representing 4GB into the
`current_length` field. By using `usize`, which reflects the host's
memory allocation, this should better reflect the size of the heap and
allows Wasmtime to support a full 4GB heap for a wasm program (instead
of 4GB minus one page).
This commit also updates the legalization of the `heap_addr` clif
instruction to appropriately cast the address to the platform's pointer
type, handling bounds checks along the way. The practical impact for
today's targets is that a `uextend` is happening sooner than it happened
before, but otherwise there is no intended impact of this change. In the
future when 64-bit memories are supported there will likely need to be
fancier logic which handles offsets a bit differently (especially in the
case of a 64-bit memory on a 32-bit host).
The clif `filetest` changes should show the differences in codegen, and
the Wasmtime changes are largely removing casts here and there.
Closes#3022
* Add tests for memory.size at maximum memory size
* Add a dfg helper method
This ports all of the identity, no-op, simplification, and canonicalization
related optimizations over from being hand-coded to the `peepmatic` DSL. This
does not handle the branch-to-branch optimizations or most of the
divide-by-constant optimizations.
* Manually rename BasicBlock to BlockPredecessor
BasicBlock is a pair of (Ebb, Inst) that is used to represent the
basic block subcomponent of an Ebb that is a predecessor to an Ebb.
Eventually we will be able to remove this struct, but for now it
makes sense to give it a non-conflicting name so that we can start
to transition Ebb to represent a basic block.
I have not updated any comments that refer to BasicBlock, as
eventually we will remove BlockPredecessor and replace with Block,
which is a basic block, so the comments will become correct.
* Manually rename SSABuilder block types to avoid conflict
SSABuilder has its own Block and BlockData types. These along with
associated identifier will cause conflicts in a later commit, so
they are renamed to be more verbose here.
* Automatically rename 'Ebb' to 'Block' in *.rs
* Automatically rename 'EBB' to 'block' in *.rs
* Automatically rename 'ebb' to 'block' in *.rs
* Automatically rename 'extended basic block' to 'basic block' in *.rs
* Automatically rename 'an basic block' to 'a basic block' in *.rs
* Manually update comment for `Block`
`Block`'s wikipedia article required an update.
* Automatically rename 'an `Block`' to 'a `Block`' in *.rs
* Automatically rename 'extended_basic_block' to 'basic_block' in *.rs
* Automatically rename 'ebb' to 'block' in *.clif
* Manually rename clif constant that contains 'ebb' as substring to avoid conflict
* Automatically rename filecheck uses of 'EBB' to 'BB'
'regex: EBB' -> 'regex: BB'
'$EBB' -> '$BB'
* Automatically rename 'EBB' 'Ebb' to 'block' in *.clif
* Automatically rename 'an block' to 'a block' in *.clif
* Fix broken testcase when function name length increases
Test function names are limited to 16 characters. This causes
the new longer name to be truncated and fail a filecheck test. An
outdated comment was also fixed.
* Bitcast vectors immediately before a return
* Bitcast vectors immediately before a block end
* Use helper function for bitcasting arguments
* Add FuncTranslationState::peekn_mut; allows mutating of peeked values
* Bitcast values in place, avoiding an allocation
Also, retrieves the correct EBB header types for bitcasting on Operator::End.
* Bitcast values of a function with no explicit Wasm return instruction
* Add Signature::return_types method
This eliminates some duplicate code and avoids extra `use`s of `Vec`.
* Add Signature::param_types method; only collect normal parameters in both this and Signature::return_types
* Move normal_args to Signature::num_normal_params method
This matches the organization of the other Signature::num_*_params methods.
* Bitcast values of Operator::Call and Operator::CallIndirect
* Add DataFlowGraph::ebb_param_types
* Bitcast values of Operator::Br and Operator::BrIf
* Bitcast values of Operator::BrTable
* Add x86 encodings for `bint` converting to `i8` and `i16`
* Introduce tests for many multi-value returns
* Support arbitrary numbers of return values
This commit implements support for returning an arbitrary number of return
values from a function. During legalization we transform multi-value signatures
to take a struct return ("sret") return pointer, instead of returning its values
in registers. Callers allocate the sret space in their stack frame and pass a
pointer to it into the caller, and once the caller returns to them, they load
the return values back out of the sret stack slot. The callee's return
operations are legalized to store the return values through the given sret
pointer.
* Keep track of old, pre-legalized signatures
When legalizing a call or return for its new legalized signature, we may need to
look at the old signature in order to figure out how to legalize the call or
return.
* Add test for multi-value returns and `call_indirect`
* Encode bool -> int x86 instructions in a loop
* Rename `Signature::uses_sret` to `Signature::uses_struct_return_param`
* Rename `p` to `param`
* Add a clarifiying comment in `num_registers_required`
* Rename `num_registers_required` to `num_return_registers_required`
* Re-add newline
* Handle already-assigned parameters in `num_return_registers_required`
* Document what some debug assertions are checking for
* Make "illegalizing" closure's control flow simpler
* Add unit tests and comments for our rounding-up-to-the-next-multiple-of-a-power-of-2 function
* Use `append_isnt_arg` instead of doing the same thing manually
* Fix grammar in comment
* Add `Signature::uses_special_{param,return}` helper functions
* Inline the definition of `legalize_type_for_sret_load` for readability
* Move sret legalization debug assertions out into their own function
* Add `round_up_to_multiple_of_type_align` helper for readability
* Add a debug assertion that we aren't removing the wrong return value
* Rename `RetPtr` stack slots to `StructReturnSlot`
* Make `legalize_type_for_sret_store` more symmetrical to `legalized_type_for_sret`
* rustfmt
* Remove unnecessary loop labels
* Do not pre-assign offsets to struct return stack slots
Instead, let the existing frame layout algorithm decide where they should go.
* Expand "sret" into explicit "struct return" in doc comment
* typo: "than" -> "then" in comment
* Fold test's debug message into the assertion itself
* [codegen] add encodings for iadd carry variants
Add encodings for iadd carry variants (iadd_cout, iadd_cin, iadd_carry)
for x86_32, enabling the legalization for iadd.i64 to work.
* [codegen] remove support for iadd carry variants on riscv
Previously, the carry variants of iadd (iadd_cin, iadd_cout and
iadd_carry) were being legalized for isa/riscv since RISC architectures
lack a flags register.
This forced us to return and accept booleans for these operations, which
proved to be problematic and inconvenient, especially for x86.
This commit removes support for said statements and all dependent
statements for isa/riscv so that we can work on a better legalization
strategy in the future.
* [codegen] change operand type from bool to iflag for iadd carry variants
The type of the carry operands for the carry variants of the iadd
instruction (iadd_cin, iadd_cout, iadd_carry) was bool for compatibility
reasons for isa/riscv. Since support for these instructions on RISC
architectures has been temporarily suspended, we can safely change the
type to iflags.
