This commit removes Wasmtime's dependency on the `region` crate. The
motivation for this came about when I was updating dependencies and saw
that `region` had a new major version at 3.0.0 as opposed to our
currently used 2.3 track. In reviewing the use cases of `region` within
Wasmtime I found two trends in particular which motivated this commit:
* Some unix-specific areas of `wasmtime_runtime` use
`rustix::mm::mprotect` instead of `region::protect` already. This
means that the usage of `region::protect` for changing virtual memory
protections was already inconsistent.
* Many uses of `region::protect` were already in unix-specific regions
which could make use of `rustix`.
Overall I opted to remove the dependency on the `region` crate to avoid
chasing its versions over time. Unix-specific changes of protections
were easily changed to `rustix::mm::mprotect`. There were two locations
where a windows/unix split is now required and I subjectively ruled
"that seems ok". Finally removing `region` also meant that the "what is
the current page size" query needed to be inlined into
`wasmtime_runtime`, which I have also subjectively ruled "that seems
fine".
Finally one final refactoring here was that the `unix.rs` and `linux.rs`
split for the pooling allocator was merged. These two files already only
differed in one function so I slapped a `cfg_if!` in there to help
reduce the duplication.
This commit is a small refactoring of `wasmtime_runtime::Trap` and
various internals. The `Trap` structure is now a reason plus backtrace,
and the old `Trap` enum is mostly in `TrapReason` now. Additionally all
`Trap`-returning methods of `wasmtime_runtime` are changed to returning
a `TrapCode` to indicate that they never capture a backtrace. Finally
the `UnwindReason` internally now no longer duplicates the trap reasons,
instead only having two variants of "panic" and "trap".
The motivation for this commit is mostly just cleaning up trap internals
and removing the need for methods like
`wasmtime_runtime::Trap::insert_backtrace` to leave it only happening at
the `wasmtime` layer.
* Add shared memories
This change adds the ability to use shared memories in Wasmtime when the
[threads proposal] is enabled. Shared memories are annotated as `shared`
in the WebAssembly syntax, e.g., `(memory 1 1 shared)`, and are
protected from concurrent access during `memory.size` and `memory.grow`.
[threads proposal]: https://github.com/WebAssembly/threads/blob/master/proposals/threads/Overview.md
In order to implement this in Wasmtime, there are two main cases to
cover:
- a program may simply create a shared memory and possibly export it;
this means that Wasmtime itself must be able to create shared
memories
- a user may create a shared memory externally and pass it in as an
import during instantiation; this is the case when the program
contains code like `(import "env" "memory" (memory 1 1
shared))`--this case is handled by a new Wasmtime API
type--`SharedMemory`
Because of the first case, this change allows any of the current
memory-creation mechanisms to work as-is. Wasmtime can still create
either static or dynamic memories in either on-demand or pooling modes,
and any of these memories can be considered shared. When shared, the
`Memory` runtime container will lock appropriately during `memory.size`
and `memory.grow` operations; since all memories use this container, it
is an ideal place for implementing the locking once and once only.
The second case is covered by the new `SharedMemory` structure. It uses
the same `Mmap` allocation under the hood as non-shared memories, but
allows the user to perform the allocation externally to Wasmtime and
share the memory across threads (via an `Arc`). The pointer address to
the actual memory is carefully wired through and owned by the
`SharedMemory` structure itself. This means that there are differing
views of where to access the pointer (i.e., `VMMemoryDefinition`): for
owned memories (the default), the `VMMemoryDefinition` is stored
directly by the `VMContext`; in the `SharedMemory` case, however, this
`VMContext` must point to this separate structure.
To ensure that the `VMContext` can always point to the correct
`VMMemoryDefinition`, this change alters the `VMContext` structure.
Since a `SharedMemory` owns its own `VMMemoryDefinition`, the
`defined_memories` table in the `VMContext` becomes a sequence of
pointers--in the shared memory case, they point to the
`VMMemoryDefinition` owned by the `SharedMemory` and in the owned memory
case (i.e., not shared) they point to `VMMemoryDefinition`s stored in a
new table, `owned_memories`.
This change adds an additional indirection (through the `*mut
VMMemoryDefinition` pointer) that could add overhead. Using an imported
memory as a proxy, we measured a 1-3% overhead of this approach on the
`pulldown-cmark` benchmark. To avoid this, Cranelift-generated code will
special-case the owned memory access (i.e., load a pointer directly to
the `owned_memories` entry) for `memory.size` so that only
shared memories (and imported memories, as before) incur the indirection
cost.
* review: remove thread feature check
* review: swap wasmtime-types dependency for existing wasmtime-environ use
* review: remove unused VMMemoryUnion
* review: reword cross-engine error message
* review: improve tests
* review: refactor to separate prevent Memory <-> SharedMemory conversion
* review: into_shared_memory -> as_shared_memory
* review: remove commented out code
* review: limit shared min/max to 32 bits
* review: skip imported memories
* review: imported memories are not owned
* review: remove TODO
* review: document unsafe send + sync
* review: add limiter assertion
* review: remove TODO
* review: improve tests
* review: fix doc test
* fix: fixes based on discussion with Alex
This changes several key parts:
- adds memory indexes to imports and exports
- makes `VMMemoryDefinition::current_length` an atomic usize
* review: add `Extern::SharedMemory`
* review: remove TODO
* review: atomically load from VMMemoryDescription in JIT-generated code
* review: add test probing the last available memory slot across threads
* fix: move assertion to new location due to rebase
* fix: doc link
* fix: add TODOs to c-api
* fix: broken doc link
* fix: modify pooling allocator messages in tests
* review: make owned_memory_index panic instead of returning an option
* review: clarify calculation of num_owned_memories
* review: move 'use' to top of file
* review: change '*const [u8]' to '*mut [u8]'
* review: remove TODO
* review: avoid hard-coding memory index
* review: remove 'preallocation' parameter from 'Memory::_new'
* fix: component model memory length
* review: check that shared memory plans are static
* review: ignore growth limits for shared memory
* review: improve atomic store comment
* review: add FIXME for memory growth failure
* review: add comment about absence of bounds-checked 'memory.size'
* review: make 'current_length()' doc comment more precise
* review: more comments related to memory.size non-determinism
* review: make 'vmmemory' unreachable for shared memory
* review: move code around
* review: thread plan through to 'wrap()'
* review: disallow shared memory allocation with the pooling allocator
* Add a `VMComponentContext` type and create it on instantiation
This commit fills out the `wasmtime-runtime` crate's support for
`VMComponentContext` and creates it as part of the instantiation
process. This moves a few maps that were temporarily allocated in an
`InstanceData` into the `VMComponentContext` and additionally reads the
canonical options data from there instead.
This type still won't be used in its "full glory" until the lowering of
host functions is completely implemented, however, which will be coming
in a future commit.
* Remove `DerefMut` implementation
* Rebase conflicts
* Change some `VMContext` pointers to `()` pointers
This commit is motivated by my work on the component model
implementation for imported functions. Currently all context pointers in
wasm are `*mut VMContext` but with the component model my plan is to
make some pointers instead along the lines of `*mut VMComponentContext`.
In doing this though one worry I have is breaking what has otherwise
been a core invariant of Wasmtime for quite some time, subtly
introducing bugs by accident.
To help assuage my worry I've opted here to erase knowledge of
`*mut VMContext` where possible. Instead where applicable a context
pointer is simply known as `*mut ()` and the embedder doesn't actually
know anything about this context beyond the value of the pointer. This
will help prevent Wasmtime from accidentally ever trying to interpret
this context pointer as an actual `VMContext` when it might instead be a
`VMComponentContext`.
Overall this was a pretty smooth transition. The main change here is
that the `VMTrampoline` (now sporting more docs) has its first argument
changed to `*mut ()`. The second argument, the caller context, is still
configured as `*mut VMContext` though because all functions are always
called from wasm still. Eventually for component-to-component calls I
think we'll probably "fake" the second argument as the same as the first
argument, losing track of the original caller, as an intentional way of
isolating components from each other.
Along the way there are a few host locations which do actually assume
that the first argument is indeed a `VMContext`. These are valid
assumptions that are upheld from a correct implementation, but I opted
to add a "magic" field to `VMContext` to assert this in debug mode. This
new "magic" field is inintialized during normal vmcontext initialization
and it's checked whenever a `VMContext` is reinterpreted as an
`Instance` (but only in debug mode). My hope here is to catch any future
accidental mistakes, if ever.
* Use a VMOpaqueContext wrapper
* Fix typos
* Support disabling backtraces at compile time
This commit adds support to Wasmtime to disable, at compile time, the
gathering of backtraces on traps. The `wasmtime` crate now sports a
`wasm-backtrace` feature which, when disabled, will mean that backtraces
are never collected at compile time nor are unwinding tables inserted
into compiled objects.
The motivation for this commit stems from the fact that generating a
backtrace is quite a slow operation. Currently backtrace generation is
done with libunwind and `_Unwind_Backtrace` typically found in glibc or
other system libraries. When thousands of modules are loaded into the
same process though this means that the initial backtrace can take
nearly half a second and all subsequent backtraces can take upwards of
hundreds of milliseconds. Relative to all other operations in Wasmtime
this is extremely expensive at this time. In the future we'd like to
implement a more performant backtrace scheme but such an implementation
would require coordination with Cranelift and is a big chunk of work
that may take some time, so in the meantime if embedders don't need a
backtrace they can still use this option to disable backtraces at
compile time and avoid the performance pitfalls of collecting
backtraces.
In general I tried to originally make this a runtime configuration
option but ended up opting for a compile-time option because `Trap::new`
otherwise has no arguments and always captures a backtrace. By making
this a compile-time option it was possible to configure, statically, the
behavior of `Trap::new`. Additionally I also tried to minimize the
amount of `#[cfg]` necessary by largely only having it at the producer
and consumer sites.
Also a noteworthy restriction of this implementation is that if
backtrace support is disabled at compile time then reference types
support will be unconditionally disabled at runtime. With backtrace
support disabled there's no way to trace the stack of wasm frames which
means that GC can't happen given our current implementation.
* Always enable backtraces for the C API
* Delete historical interruptable support in Wasmtime
This commit removes the `Config::interruptable` configuration along with
the `InterruptHandle` type from the `wasmtime` crate. The original
support for adding interruption to WebAssembly was added pretty early on
in the history of Wasmtime when there was no other method to prevent an
infinite loop from the host. Nowadays, however, there are alternative
methods for interruption such as fuel or epoch-based interruption.
One of the major downsides of `Config::interruptable` is that even when
it's not enabled it forces an atomic swap to happen when entering
WebAssembly code. This technically could be a non-atomic swap if the
configuration option isn't enabled but that produces even more branch-y
code on entry into WebAssembly which is already something we try to
optimize. Calling into WebAssembly is on the order of a dozens of
nanoseconds at this time and an atomic swap, even uncontended, can add
up to 5ns on some platforms.
The main goal of this PR is to remove this atomic swap on entry into
WebAssembly. This is done by removing the `Config::interruptable` field
entirely, moving all existing consumers to epochs instead which are
suitable for the same purposes. This means that the stack overflow check
is no longer entangled with the interruption check and perhaps one day
we could continue to optimize that further as well.
Some consequences of this change are:
* Epochs are now the only method of remote-thread interruption.
* There are no more Wasmtime traps that produces the `Interrupted` trap
code, although we may wish to move future traps to this so I left it
in place.
* The C API support for interrupt handles was also removed and bindings
for epoch methods were added.
* Function-entry checks for interruption are a tiny bit less efficient
since one check is performed for the stack limit and a second is
performed for the epoch as opposed to the `Config::interruptable`
style of bundling the stack limit and the interrupt check in one. It's
expected though that this is likely to not really be measurable.
* The old `VMInterrupts` structure is renamed to `VMRuntimeLimits`.
* Remove the `ModuleLimits` pooling configuration structure
This commit is an attempt to improve the usability of the pooling
allocator by removing the need to configure a `ModuleLimits` structure.
Internally this structure has limits on all forms of wasm constructs but
this largely bottoms out in the size of an allocation for an instance in
the instance pooling allocator. Maintaining this list of limits can be
cumbersome as modules may get tweaked over time and there's otherwise no
real reason to limit the number of globals in a module since the main
goal is to limit the memory consumption of a `VMContext` which can be
done with a memory allocation limit rather than fine-tuned control over
each maximum and minimum.
The new approach taken in this commit is to remove `ModuleLimits`. Some
fields, such as `tables`, `table_elements` , `memories`, and
`memory_pages` are moved to `InstanceLimits` since they're still
enforced at runtime. A new field `size` is added to `InstanceLimits`
which indicates, in bytes, the maximum size of the `VMContext`
allocation. If the size of a `VMContext` for a module exceeds this value
then instantiation will fail.
This involved adding a few more checks to `{Table, Memory}::new_static`
to ensure that the minimum size is able to fit in the allocation, since
previously modules were validated at compile time of the module that
everything fit and that validation no longer happens (it happens at
runtime).
A consequence of this commit is that Wasmtime will have no built-in way
to reject modules at compile time if they'll fail to be instantiated
within a particular pooling allocator configuration. Instead a module
must attempt instantiation see if a failure happens.
* Fix benchmark compiles
* Fix some doc links
* Fix a panic by ensuring modules have limited tables/memories
* Review comments
* Add back validation at `Module` time instantiation is possible
This allows for getting an early signal at compile time that a module
will never be instantiable in an engine with matching settings.
* Provide a better error message when sizes are exceeded
Improve the error message when an instance size exceeds the maximum by
providing a breakdown of where the bytes are all going and why the large
size is being requested.
* Try to fix test in qemu
* Flag new test as 64-bit only
Sizes are all specific to 64-bit right now
* Enable copy-on-write heap initialization by default
This commit enables the `Config::memfd` feature by default now that it's
been fuzzed for a few weeks on oss-fuzz, and will continue to be fuzzed
leading up to the next release of Wasmtime in early March. The
documentation of the `Config` option has been updated as well as adding
a CLI flag to disable the feature.
* Remove ubiquitous "memfd" terminology
Switch instead to forms of "memory image" or "cow" or some combination
thereof.
* Update new option names
* Shrink the size of `FuncData`
Before this commit on a 64-bit system the `FuncData` type had a size of
88 bytes and after this commit it has a size of 32 bytes. A `FuncData`
is required for all host functions in a store, including those inserted
from a `Linker` into a store used during linking. This means that
instantiation ends up creating a nontrivial number of these types and
pushing them into the store. Looking at some profiles there were some
surprisingly expensive movements of `FuncData` from the stack to a
vector for moves-by-value generated by Rust. Shrinking this type enables
more efficient code to be generated and additionally means less storage
is needed in a store's function array.
For instantiating the spidermonkey and rustpython modules this improves
instantiation by 10% since they each import a fair number of host
functions and the speedup here is relative to the number of items
imported.
* Use `ptr::copy_nonoverlapping` during initialization
Prevoiusly `ptr::copy` was used for copying imports into place which
translates to `memmove`, but `ptr::copy_nonoverlapping` can be used here
since it's statically known these areas don't overlap. While this
doesn't end up having a performance difference it's something I kept
noticing while looking at the disassembly of `initialize_vmcontext` so I
figured I'd go ahead and implement.
* Indirect shared signature ids in the VMContext
This commit is a small improvement for the instantiation time of modules
by avoiding copying a list of `VMSharedSignatureIndex` entries into each
`VMContext`, instead building one inside of a module and sharing that
amongst all instances. This involves less lookups at instantiation time
and less movement of data during instantiation. The downside is that
type-checks on `call_indirect` now involve an additionally load, but I'm
assuming that these are somewhat pessimized enough as-is that the
runtime impact won't be much there.
For instantiation performance this is a 5-10% win with
rustpyhon/spidermonky instantiation. This should also reduce the size of
each `VMContext` for an instantiation since signatures are no longer
stored inline but shared amongst all instances with one module.
Note that one subtle change here is that the array of
`VMSharedSignatureIndex` was previously indexed by `TypeIndex`, and now
it's indexed by `SignaturedIndex` which is a deduplicated form of
`TypeIndex`. This is done because we already had a list of those lying
around in `Module`, so it was easier to reuse that than to build a
separate array and store it somewhere.
* Reserve space in `Store<T>` with `InstancePre`
This commit updates the instantiation process to reserve space in a
`Store<T>` for the functions that an `InstancePre<T>`, as part of
instantiation, will insert into it. Using an `InstancePre<T>` to
instantiate allows pre-computing the number of host functions that will
be inserted into a store, and by pre-reserving space we can avoid costly
reallocations during instantiation by ensuring the function vector has
enough space to fit everything during the instantiation process.
Overall this makes instantiation of rustpython/spidermonkey about 8%
faster locally.
* Fix tests
* Use checked arithmetic
* Skip memfd creation with precompiled modules
This commit updates the memfd support internally to not actually use a
memfd if a compiled module originally came from disk via the
`wasmtime::Module::deserialize_file` API. In this situation we already
have a file descriptor open and there's no need to copy a module's heap
image to a new file descriptor.
To facilitate a new source of `mmap` the currently-memfd-specific-logic
of creating a heap image is generalized to a new form of
`MemoryInitialization` which is attempted for all modules at
module-compile-time. This means that the serialized artifact to disk
will have the memory image in its entirety waiting for us. Furthermore
the memory image is ensured to be padded and aligned carefully to the
target system's page size, notably meaning that the data section in the
final object file is page-aligned and the size of the data section is
also page aligned.
This means that when a precompiled module is mapped from disk we can
reuse the underlying `File` to mmap all initial memory images. This
means that the offset-within-the-memory-mapped-file can differ for
memfd-vs-not, but that's just another piece of state to track in the
memfd implementation.
In the limit this waters down the term "memfd" for this technique of
quickly initializing memory because we no longer use memfd
unconditionally (only when the backing file isn't available).
This does however open up an avenue in the future to porting this
support to other OSes because while `memfd_create` is Linux-specific
both macOS and Windows support mapping a file with copy-on-write. This
porting isn't done in this PR and is left for a future refactoring.
Closes#3758
* Enable "memfd" support on all unix systems
Cordon off the Linux-specific bits and enable the memfd support to
compile and run on platforms like macOS which have a Linux-like `mmap`.
This only works if a module is mapped from a precompiled module file on
disk, but that's better than not supporting it at all!
* Fix linux compile
* Use `Arc<File>` instead of `MmapVecFileBacking`
* Use a named struct instead of mysterious tuples
* Comment about unsafety in `Module::deserialize_file`
* Fix tests
* Fix uffd compile
* Always align data segments
No need to have conditional alignment since their sizes are all aligned
anyway
* Update comment in build.rs
* Use rustix, not `region`
* Fix some confusing logic/names around memory indexes
These functions all work with memory indexes, not specifically defined
memory indexes.
During instance initialization, we build two sorts of arrays eagerly:
- We create an "anyfunc" (a `VMCallerCheckedAnyfunc`) for every function
in an instance.
- We initialize every element of a funcref table with an initializer to
a pointer to one of these anyfuncs.
Most instances will not touch (via call_indirect or table.get) all
funcref table elements. And most anyfuncs will never be referenced,
because most functions are never placed in tables or used with
`ref.func`. Thus, both of these initialization tasks are quite wasteful.
Profiling shows that a significant fraction of the remaining
instance-initialization time after our other recent optimizations is
going into these two tasks.
This PR implements two basic ideas:
- The anyfunc array can be lazily initialized as long as we retain the
information needed to do so. For now, in this PR, we just recreate the
anyfunc whenever a pointer is taken to it, because doing so is fast
enough; in the future we could keep some state to know whether the
anyfunc has been written yet and skip this work if redundant.
This technique allows us to leave the anyfunc array as uninitialized
memory, which can be a significant savings. Filling it with
initialized anyfuncs is very expensive, but even zeroing it is
expensive: e.g. in a large module, it can be >500KB.
- A funcref table can be lazily initialized as long as we retain a link
to its corresponding instance and function index for each element. A
zero in a table element means "uninitialized", and a slowpath does the
initialization.
Funcref tables are a little tricky because funcrefs can be null. We need
to distinguish "element was initially non-null, but user stored explicit
null later" from "element never touched" (ie the lazy init should not
blow away an explicitly stored null). We solve this by stealing the LSB
from every funcref (anyfunc pointer): when the LSB is set, the funcref
is initialized and we don't hit the lazy-init slowpath. We insert the
bit on storing to the table and mask it off after loading.
We do have to set up a precomputed array of `FuncIndex`s for the table
in order for this to work. We do this as part of the module compilation.
This PR also refactors the way that the runtime crate gains access to
information computed during module compilation.
Performance effect measured with in-tree benches/instantiation.rs, using
SpiderMonkey built for WASI, and with memfd enabled:
```
BEFORE:
sequential/default/spidermonkey.wasm
time: [68.569 us 68.696 us 68.856 us]
sequential/pooling/spidermonkey.wasm
time: [69.406 us 69.435 us 69.465 us]
parallel/default/spidermonkey.wasm: with 1 background thread
time: [69.444 us 69.470 us 69.497 us]
parallel/default/spidermonkey.wasm: with 16 background threads
time: [183.72 us 184.31 us 184.89 us]
parallel/pooling/spidermonkey.wasm: with 1 background thread
time: [69.018 us 69.070 us 69.136 us]
parallel/pooling/spidermonkey.wasm: with 16 background threads
time: [326.81 us 337.32 us 347.01 us]
WITH THIS PR:
sequential/default/spidermonkey.wasm
time: [6.7821 us 6.8096 us 6.8397 us]
change: [-90.245% -90.193% -90.142%] (p = 0.00 < 0.05)
Performance has improved.
sequential/pooling/spidermonkey.wasm
time: [3.0410 us 3.0558 us 3.0724 us]
change: [-95.566% -95.552% -95.537%] (p = 0.00 < 0.05)
Performance has improved.
parallel/default/spidermonkey.wasm: with 1 background thread
time: [7.2643 us 7.2689 us 7.2735 us]
change: [-89.541% -89.533% -89.525%] (p = 0.00 < 0.05)
Performance has improved.
parallel/default/spidermonkey.wasm: with 16 background threads
time: [147.36 us 148.99 us 150.74 us]
change: [-18.997% -18.081% -17.285%] (p = 0.00 < 0.05)
Performance has improved.
parallel/pooling/spidermonkey.wasm: with 1 background thread
time: [3.1009 us 3.1021 us 3.1033 us]
change: [-95.517% -95.511% -95.506%] (p = 0.00 < 0.05)
Performance has improved.
parallel/pooling/spidermonkey.wasm: with 16 background threads
time: [49.449 us 50.475 us 51.540 us]
change: [-85.423% -84.964% -84.465%] (p = 0.00 < 0.05)
Performance has improved.
```
So an improvement of something like 80-95% for a very large module (7420
functions in its one funcref table, 31928 functions total).
* memfd: Reduce some syscalls in the on-demand case
This tweaks the internal organization of the `MemFdSlot` to avoid some
syscalls in the default case as well as opportunistically in the pooling
case. The two cases added here are:
* A `MemFdSlot` is now created with a specified initial size. For
pooling this is 0 but for the on-demand case this can be non-zero.
* When `instantiate` is called with no prior image and the sizes match
(as will be the case for on-demand allocation) then `mprotect` is
skipped entirely.
* In the `clear_and_remain-ready` case the `mprotect` is skipped if the
heap wasn't grown at all.
This should avoid ever using `mprotect` unnecessarily and makes the
ranges we `mprotect` a bit smaller as well.
* Review comments
* Tweak allow to apply to whole crate
As first suggested by Jan on the Zulip here [1], a cheap and effective
way to obtain copy-on-write semantics of a "backing image" for a Wasm
memory is to mmap a file with `MAP_PRIVATE`. The `memfd` mechanism
provided by the Linux kernel allows us to create anonymous,
in-memory-only files that we can use for this mapping, so we can
construct the image contents on-the-fly then effectively create a CoW
overlay. Furthermore, and importantly, `madvise(MADV_DONTNEED, ...)`
will discard the CoW overlay, returning the mapping to its original
state.
By itself this is almost enough for a very fast
instantiation-termination loop of the same image over and over,
without changing the address space mapping at all (which is
expensive). The only missing bit is how to implement
heap *growth*. But here memfds can help us again: if we create another
anonymous file and map it where the extended parts of the heap would
go, we can take advantage of the fact that a `mmap()` mapping can
be *larger than the file itself*, with accesses beyond the end
generating a `SIGBUS`, and the fact that we can cheaply resize the
file with `ftruncate`, even after a mapping exists. So we can map the
"heap extension" file once with the maximum memory-slot size and grow
the memfd itself as `memory.grow` operations occur.
The above CoW technique and heap-growth technique together allow us a
fastpath of `madvise()` and `ftruncate()` only when we re-instantiate
the same module over and over, as long as we can reuse the same
slot. This fastpath avoids all whole-process address-space locks in
the Linux kernel, which should mean it is highly scalable. It also
avoids the cost of copying data on read, as the `uffd` heap backend
does when servicing pagefaults; the kernel's own optimized CoW
logic (same as used by all file mmaps) is used instead.
[1] https://bytecodealliance.zulipchat.com/#narrow/stream/206238-general/topic/Copy.20on.20write.20based.20instance.20reuse/near/266657772
This PR introduces a new way of performing cooperative timeslicing that
is intended to replace the "fuel" mechanism. The tradeoff is that this
mechanism interrupts with less precision: not at deterministic points
where fuel runs out, but rather when the Engine enters a new epoch. The
generated code instrumentation is substantially faster, however, because
it does not need to do as much work as when tracking fuel; it only loads
the global "epoch counter" and does a compare-and-branch at backedges
and function prologues.
This change has been measured as ~twice as fast as fuel-based
timeslicing for some workloads, especially control-flow-intensive
workloads such as the SpiderMonkey JS interpreter on Wasm/WASI.
The intended interface is that the embedder of the `Engine` performs an
`engine.increment_epoch()` call periodically, e.g. once per millisecond.
An async invocation of a Wasm guest on a `Store` can specify a number of
epoch-ticks that are allowed before an async yield back to the
executor's event loop. (The initial amount and automatic "refills" are
configured on the `Store`, just as for fuel.) This call does only
signal-safe work (it increments an `AtomicU64`) so could be invoked from
a periodic signal, or from a thread that wakes up once per period.
This commit adds the `pooling-allocator` feature to both the `wasmtime` and
`wasmtime-runtime` crates.
The feature controls whether or not the pooling allocator implementation is
built into the runtime and exposed as a supported instance allocation strategy
in the wasmtime API.
The feature is on by default for the `wasmtime` crate.
Closes#3513.
* Add `*_unchecked` variants of `Func` APIs for the C API
This commit is what is hopefully going to be my last installment within
the saga of optimizing function calls in/out of WebAssembly modules in
the C API. This is yet another alternative approach to #3345 (sorry) but
also contains everything necessary to make the C API fast. As in #3345
the general idea is just moving checks out of the call path in the same
style of `TypedFunc`.
This new strategy takes inspiration from previously learned attempts
effectively "just" exposes how we previously passed `*mut u128` through
trampolines for arguments/results. This storage format is formalized
through a new `ValRaw` union that is exposed from the `wasmtime` crate.
By doing this it made it relatively easy to expose two new APIs:
* `Func::new_unchecked`
* `Func::call_unchecked`
These are the same as their checked equivalents except that they're
`unsafe` and they work with `*mut ValRaw` rather than safe slices of
`Val`. Working with these eschews type checks and such and requires
callers/embedders to do the right thing.
These two new functions are then exposed via the C API with new
functions, enabling C to have a fast-path of calling/defining functions.
This fast path is akin to `Func::wrap` in Rust, although that API can't
be built in C due to C not having generics in the same way that Rust
has.
For some benchmarks, the benchmarks here are:
* `nop` - Call a wasm function from the host that does nothing and
returns nothing.
* `i64` - Call a wasm function from the host, the wasm function calls a
host function, and the host function returns an `i64` all the way out to
the original caller.
* `many` - Call a wasm function from the host, the wasm calls
host function with 5 `i32` parameters, and then an `i64` result is
returned back to the original host
* `i64` host - just the overhead of the wasm calling the host, so the
wasm calls the host function in a loop.
* `many` host - same as `i64` host, but calling the `many` host function.
All numbers in this table are in nanoseconds, and this is just one
measurement as well so there's bound to be some variation in the precise
numbers here.
| Name | Rust | C (before) | C (after) |
|-----------|------|------------|-----------|
| nop | 19 | 112 | 25 |
| i64 | 22 | 207 | 32 |
| many | 27 | 189 | 34 |
| i64 host | 2 | 38 | 5 |
| many host | 7 | 75 | 8 |
The main conclusion here is that the C API is significantly faster than
before when using the `*_unchecked` variants of APIs. The Rust
implementation is still the ceiling (or floor I guess?) for performance
The main reason that C is slower than Rust is that a little bit more has
to travel through memory where on the Rust side of things we can
monomorphize and inline a bit more to get rid of that. Overall though
the costs are way way down from where they were originally and I don't
plan on doing a whole lot more myself at this time. There's various
things we theoretically could do I've considered but implementation-wise
I think they'll be much more weighty.
* Tweak `wasmtime_externref_t` API comments
* Move `CompiledFunction` into wasmtime-cranelift
This commit moves the `wasmtime_environ::CompiledFunction` type into the
`wasmtime-cranelift` crate. This type has lots of Cranelift-specific
pieces of compilation and doesn't need to be generated by all Wasmtime
compilers. This replaces the usage in the `Compiler` trait with a
`Box<Any>` type that each compiler can select. Each compiler must still
produce a `FunctionInfo`, however, which is shared information we'll
deserialize for each module.
The `wasmtime-debug` crate is also folded into the `wasmtime-cranelift`
crate as a result of this commit. One possibility was to move the
`CompiledFunction` commit into its own crate and have `wasmtime-debug`
depend on that, but since `wasmtime-debug` is Cranelift-specific at this
time it didn't seem like it was too too necessary to keep it separate.
If `wasmtime-debug` supports other backends in the future we can
recreate a new crate, perhaps with it refactored to not depend on
Cranelift.
* Move wasmtime_environ::reference_type
This now belongs in wasmtime-cranelift and nowhere else
* Remove `Type` reexport in wasmtime-environ
One less dependency on `cranelift-codegen`!
* Remove `types` reexport from `wasmtime-environ`
Less cranelift!
* Remove `SourceLoc` from wasmtime-environ
Change the `srcloc`, `start_srcloc`, and `end_srcloc` fields to a custom
`FilePos` type instead of `ir::SourceLoc`. These are only used in a few
places so there's not much to lose from an extra abstraction for these
leaf use cases outside of cranelift.
* Remove wasmtime-environ's dep on cranelift's `StackMap`
This commit "clones" the `StackMap` data structure in to
`wasmtime-environ` to have an independent representation that that
chosen by Cranelift. This allows Wasmtime to decouple this runtime
dependency of stack map information and let the two evolve
independently, if necessary.
An alternative would be to refactor cranelift's implementation into a
separate crate and have wasmtime depend on that but it seemed a bit like
overkill to do so and easier to clone just a few lines for this.
* Define code offsets in wasmtime-environ with `u32`
Don't use Cranelift's `binemit::CodeOffset` alias to define this field
type since the `wasmtime-environ` crate will be losing the
`cranelift-codegen` dependency soon.
* Commit to using `cranelift-entity` in Wasmtime
This commit removes the reexport of `cranelift-entity` from the
`wasmtime-environ` crate and instead directly depends on the
`cranelift-entity` crate in all referencing crates. The original reason
for the reexport was to make cranelift version bumps easier since it's
less versions to change, but nowadays we have a script to do that.
Otherwise this encourages crates to use whatever they want from
`cranelift-entity` since we'll always depend on the whole crate.
It's expected that the `cranelift-entity` crate will continue to be a
lean crate in dependencies and suitable for use at both runtime and
compile time. Consequently there's no need to avoid its usage in
Wasmtime at runtime, since "remove Cranelift at compile time" is
primarily about the `cranelift-codegen` crate.
* Remove most uses of `cranelift-codegen` in `wasmtime-environ`
There's only one final use remaining, which is the reexport of
`TrapCode`, which will get handled later.
* Limit the glob-reexport of `cranelift_wasm`
This commit removes the glob reexport of `cranelift-wasm` from the
`wasmtime-environ` crate. This is intended to explicitly define what
we're reexporting and is a transitionary step to curtail the amount of
dependencies taken on `cranelift-wasm` throughout the codebase. For
example some functions used by debuginfo mapping are better imported
directly from the crate since they're Cranelift-specific. Note that
this is intended to be a temporary state affairs, soon this reexport
will be gone entirely.
Additionally this commit reduces imports from `cranelift_wasm` and also
primarily imports from `crate::wasm` within `wasmtime-environ` to get a
better sense of what's imported from where and what will need to be
shared.
* Extract types from cranelift-wasm to cranelift-wasm-types
This commit creates a new crate called `cranelift-wasm-types` and
extracts type definitions from the `cranelift-wasm` crate into this new
crate. The purpose of this crate is to be a shared definition of wasm
types that can be shared both by compilers (like Cranelift) as well as
wasm runtimes (e.g. Wasmtime). This new `cranelift-wasm-types` crate
doesn't depend on `cranelift-codegen` and is the final step in severing
the unconditional dependency from Wasmtime to `cranelift-codegen`.
The final refactoring in this commit is to then reexport this crate from
`wasmtime-environ`, delete the `cranelift-codegen` dependency, and then
update all `use` paths to point to these new types.
The main change of substance here is that the `TrapCode` enum is
mirrored from Cranelift into this `cranelift-wasm-types` crate. While
this unfortunately results in three definitions (one more which is
non-exhaustive in Wasmtime itself) it's hopefully not too onerous and
ideally something we can patch up in the future.
* Get lightbeam compiling
* Remove unnecessary dependency
* Fix compile with uffd
* Update publish script
* Fix more uffd tests
* Rename cranelift-wasm-types to wasmtime-types
This reflects the purpose a bit more where it's types specifically
intended for Wasmtime and its support.
* Fix publish script
* wasmtime_runtime: move ResourceLimiter defaults into this crate
In preparation of changing wasmtime::ResourceLimiter to be a re-export
of this definition, because translating between two traits was causing
problems elsewhere.
* wasmtime: make ResourceLimiter a re-export of wasmtime_runtime::ResourceLimiter
* refactor Store internals to support ResourceLimiter as part of store's data
* add hooks for entering and exiting native code to Store
* wasmtime-wast, fuzz: changes to adapt ResourceLimiter API
* fix tests
* wrap calls into wasm with entering/exiting exit hooks as well
* the most trivial test found a bug, lets write some more
* store: mark some methods as #[inline] on Store, StoreInner, StoreInnerMost
Co-authored-By: Alex Crichton <alex@alexcrichton.com>
* improve tests for the entering/exiting native hooks
Co-authored-by: Alex Crichton <alex@alexcrichton.com>
Implement Wasmtime's new API as designed by RFC 11. This is quite a large commit which has had lots of discussion externally, so for more information it's best to read the RFC thread and the PR thread.
* Add resource limiting to the Wasmtime API.
This commit adds a `ResourceLimiter` trait to the Wasmtime API.
When used in conjunction with `Store::new_with_limiter`, this can be used to
monitor and prevent WebAssembly code from growing linear memories and tables.
This is particularly useful when hosts need to take into account host resource
usage to determine if WebAssembly code can consume more resources.
A simple `StaticResourceLimiter` is also included with these changes that will
simply limit the size of linear memories or tables for all instances created in
the store based on static values.
* Code review feedback.
* Implemented `StoreLimits` and `StoreLimitsBuilder`.
* Moved `max_instances`, `max_memories`, `max_tables` out of `Config` and into
`StoreLimits`.
* Moved storage of the limiter in the runtime into `Memory` and `Table`.
* Made `InstanceAllocationRequest` use a reference to the limiter.
* Updated docs.
* Made `ResourceLimiterProxy` generic to remove a level of indirection.
* Fixed the limiter not being used for `wasmtime::Memory` and
`wasmtime::Table`.
* Code review feedback and bug fix.
* `Memory::new` now returns `Result<Self>` so that an error can be returned if
the initial requested memory exceeds any limits placed on the store.
* Changed an `Arc` to `Rc` as the `Arc` wasn't necessary.
* Removed `Store` from the `ResourceLimiter` callbacks. Custom resource limiter
implementations are free to capture any context they want, so no need to
unnecessarily store a weak reference to `Store` from the proxy type.
* Fixed a bug in the pooling instance allocator where an instance would be
leaked from the pool. Previously, this would only have happened if the OS was
unable to make the necessary linear memory available for the instance. With
these changes, however, the instance might not be created due to limits
placed on the store. We now properly deallocate the instance on error.
* Added more tests, including one that covers the fix mentioned above.
* Code review feedback.
* Add another memory to `test_pooling_allocator_initial_limits_exceeded` to
ensure a partially created instance is successfully deallocated.
* Update some doc comments for better documentation of `Store` and
`ResourceLimiter`.
This commit splits out a `FiberStack` from `Fiber`, allowing the instance
allocator trait to return `FiberStack` rather than raw stack pointers. This
keeps the stack creation mostly in `wasmtime_fiber`, but now the on-demand
instance allocator can make use of it.
The instance allocators no longer have to return a "not supported" error to
indicate that the store should allocate its own fiber stack.
This includes a bunch of cleanup in the instance allocator to scope stacks to
the new "async" feature in the runtime.
Closes#2708.
This commit fixes a few issues around managing the thread-local state of
a wasmtime thread. We intentionally only have a singular TLS variable in
the whole world, and the problem is that when stack-switching off an
async thread we were not restoring the previous TLS state. This is
necessary in two cases:
* Futures aren't guaranteed to be polled/completed in a stack-like
fashion. If a poll sees that a future isn't ready then we may resume
execution in a previous wasm context that ends up needing the TLS
information.
* Futures can also cross threads (when the whole store crosses threads)
and we need to save/restore TLS state from the thread we're coming
from and the thread that we're going to.
The stack switching issue necessitates some more glue around suspension
and resumption of a stack to ensure we save/restore the TLS state on
both sides. The thread issue, however, also necessitates that we use
`#[inline(never)]` on TLS access functions and never have TLS borrows
live across a function which could result in running arbitrary code (as
was the case for the `tls::set` function.
This commit implements the pooling instance allocator.
The allocation strategy can be set with `Config::with_allocation_strategy`.
The pooling strategy uses the pooling instance allocator to preallocate a
contiguous region of memory for instantiating modules that adhere to various
limits.
The intention of the pooling instance allocator is to reserve as much of the
host address space needed for instantiating modules ahead of time and to reuse
committed memory pages wherever possible.
This commit implements allocating fiber stacks in an instance allocator.
The on-demand instance allocator doesn't support custom stacks, so the
implementation will use the allocation from `wasmtime-fiber` for the fiber
stacks.
In the future, the pooling instance allocator will return custom stacks to use
on Linux and macOS.
On Windows, the native fiber implementation will always be used.
This commit changes `Instance` such that memories can be stored statically,
with just a base pointer, size, maximum, and a callback to make memory
accessible.
Previously the memories were being stored as boxed trait objects, which would
require the pooling allocator to do some unpleasant things to avoid
allocations.
With this change, the pooling allocator can simply define a memory for the
instance without using a trait object.
This commit refactors module instantiation in the runtime to allow for
different instance allocation strategy implementations.
It adds an `InstanceAllocator` trait with the current implementation put behind
the `OnDemandInstanceAllocator` struct.
The Wasmtime API has been updated to allow a `Config` to have an instance
allocation strategy set which will determine how instances get allocated.
This change is in preparation for an alternative *pooling* instance allocator
that can reserve all needed host process address space in advance.
This commit also makes changes to the `wasmtime_environ` crate to represent
compiled modules in a way that reduces copying at instantiation time.
This commit updates the various tooling used by wasmtime which has new
updates to the module linking proposal. This is done primarily to sync
with WebAssembly/module-linking#26. The main change implemented here is
that wasmtime now supports creating instances from a set of values, nott
just from instantiating a module. Additionally subtyping handling of
modules with respect to imports is now properly handled by desugaring
two-level imports to imports of instances.
A number of small refactorings are included here as well, but most of
them are in accordance with the changes to `wasmparser` and the updated
binary format for module linking.
This commit removes the global variable associated with wasm traps which
stores frame information. The only purpose of this global is to help
symbolicate `Trap`s created since we support creating a `Trap` without a
`Store`. The global, however, is only used for wasm frames on the stack,
and when wasm frames are on the stack we know that our thread local for
"what was the last context" is set and configured.
The change here is to hijack this thread-local some more to effectively
store the `Store` inside of it. All frame information is then moved
directly into `Store` and no longer lives off on the side in a global.
Additionally support for registering/unregistering modules is now
simplified because once a module is registered with a store it can never
be unregistered.
This has one slight functional change where if there are two instances
of `Store` interleaving calls to wasm code on the stack we'll only be
able to symbolicate one of them instead of both. That's arguably also a
feature however because this is sort of a way to leak information across
stores right now.
Otherwise, though, this isn't intended to change any existing logic, but
instead keep everything working as-is.
This commit refactors where trampolines and signature information is
stored within a `Store`, namely moving them from
`wasmtime_runtime::Instance` instead to `Store` itself. The goal here is
to remove an allocation inside of an `Instance` and make them a bit
cheaper to create. Additionally this should open up future possibilities
like not creating duplicate trampolines for signatures already in the
`Store` when using `Func::new`.
* wasmtime: Implement `global.{get,set}` for externref globals
We use libcalls to implement these -- unlike `table.{get,set}`, for which we
create inline JIT fast paths -- because no known toolchain actually uses
externref globals.
Part of #929
* wasmtime: Enable `{extern,func}ref` globals in the API
`funcref`s are implemented as `NonNull<VMCallerCheckedAnyfunc>`.
This should be more efficient than using a `VMExternRef` that points at a
`VMCallerCheckedAnyfunc` because it gets rid of an indirection, dynamic
allocation, and some reference counting.
Note that the null function reference is *NOT* a null pointer; it is a
`VMCallerCheckedAnyfunc` that has a null `func_ptr` member.
Part of #929
This commit enables `wasmtime_runtime::Table` to internally hold elements of
either `funcref` (all that is currently supported) or `externref` (newly
introduced in this commit).
This commit updates `Table`'s API, but does NOT generally propagate those
changes outwards all the way through the Wasmtime embedding API. It only does
enough to get everything compiling and the current test suite passing. It is
expected that as we implement more of the reference types spec, we will bubble
these changes out and expose them to the embedding API.
For host VM code, we use plain reference counting, where cloning increments
the reference count, and dropping decrements it. We can avoid many of the
on-stack increment/decrement operations that typically plague the
performance of reference counting via Rust's ownership and borrowing system.
Moving a `VMExternRef` avoids mutating its reference count, and borrowing it
either avoids the reference count increment or delays it until if/when the
`VMExternRef` is cloned.
When passing a `VMExternRef` into compiled Wasm code, we don't want to do
reference count mutations for every compiled `local.{get,set}`, nor for
every function call. Therefore, we use a variation of **deferred reference
counting**, where we only mutate reference counts when storing
`VMExternRef`s somewhere that outlives the activation: into a global or
table. Simultaneously, we over-approximate the set of `VMExternRef`s that
are inside Wasm function activations. Periodically, we walk the stack at GC
safe points, and use stack map information to precisely identify the set of
`VMExternRef`s inside Wasm activations. Then we take the difference between
this precise set and our over-approximation, and decrement the reference
count for each of the `VMExternRef`s that are in our over-approximation but
not in the precise set. Finally, the over-approximation is replaced with the
precise set.
The `VMExternRefActivationsTable` implements the over-approximized set of
`VMExternRef`s referenced by Wasm activations. Calling a Wasm function and
passing it a `VMExternRef` moves the `VMExternRef` into the table, and the
compiled Wasm function logically "borrows" the `VMExternRef` from the
table. Similarly, `global.get` and `table.get` operations clone the gotten
`VMExternRef` into the `VMExternRefActivationsTable` and then "borrow" the
reference out of the table.
When a `VMExternRef` is returned to host code from a Wasm function, the host
increments the reference count (because the reference is logically
"borrowed" from the `VMExternRefActivationsTable` and the reference count
from the table will be dropped at the next GC).
For more general information on deferred reference counting, see *An
Examination of Deferred Reference Counting and Cycle Detection* by Quinane:
https://openresearch-repository.anu.edu.au/bitstream/1885/42030/2/hon-thesis.pdf
cc #929Fixes#1804
`VMExternRef` is a reference-counted box for any kind of data that is
external and opaque to running Wasm. Sometimes it might hold a Wasmtime
thing, other times it might hold something from a Wasmtime embedder and is
opaque even to us. It is morally equivalent to `Rc<dyn Any>` in Rust, but
additionally always fits in a pointer-sized word. `VMExternRef` is
non-nullable, but `Option<VMExternRef>` is a null pointer.
The one part of `VMExternRef` that can't ever be opaque to us is the
reference count. Even when we don't know what's inside an `VMExternRef`, we
need to be able to manipulate its reference count as we add and remove
references to it. And we need to do this from compiled Wasm code, so it must
be `repr(C)`!
`VMExternRef` itself is just a pointer to an `VMExternData`, which holds the
opaque, boxed value, its reference count, and its vtable pointer.
The `VMExternData` struct is *preceded* by the dynamically-sized value boxed
up and referenced by one or more `VMExternRef`s:
```ignore
,-------------------------------------------------------.
| |
V |
+----------------------------+-----------+-----------+ |
| dynamically-sized value... | ref_count | value_ptr |---'
+----------------------------+-----------+-----------+
| VMExternData |
+-----------------------+
^
+-------------+ |
| VMExternRef |-------------------+
+-------------+ |
|
+-------------+ |
| VMExternRef |-------------------+
+-------------+ |
|
... ===
|
+-------------+ |
| VMExternRef |-------------------'
+-------------+
```
The `value_ptr` member always points backwards to the start of the
dynamically-sized value (which is also the start of the heap allocation for
this value-and-`VMExternData` pair). Because it is a `dyn` pointer, it is
fat, and also points to the value's `Any` vtable.
The boxed value and the `VMExternRef` footer are held a single heap
allocation. The layout described above is used to make satisfying the
value's alignment easy: we just need to ensure that the heap allocation used
to hold everything satisfies its alignment. It also ensures that we don't
need a ton of excess padding between the `VMExternData` and the value for
values with large alignment.
Previously we initialized trap handling (signals/etc) once-per-instance
but that's a bit too granular since we only need to do this as
one-time per-program initialization. This moves the initialization to
`Store` instead which means that we'll call this at least once per
thread, which some platforms may need (none currently do, they all only
need per-program initialization, but Fuchsia will need per-thread
initialization).
This commit fixes an issue in Wasmtime where Wasmtime would accidentally
"handle" non-wasm segfaults while executing host imports of wasm
modules. If a host import segfaulted then Wasmtime would recognize that
wasm code is on the stack, so it'd longjmp out of the wasm code. This
papers over real bugs though in host code and erroneously classified
segfaults as wasm traps.
The fix here was to add a check to our wasm signal handler for if the
faulting address falls in JIT code itself. Actually threading through
all the right information for that check to happen is a bit tricky,
though, so this involved some refactoring:
* A closure parameter to `catch_traps` was added. This closure is
responsible for classifying addresses as whether or not they fall in
JIT code. Anything returning `false` means that the trap won't get
handled and we'll forward to the next signal handler.
* To avoid passing tons of context all over the place, the start
function is now no longer automatically invoked by `InstanceHandle`.
This avoids the need for passing all sorts of trap-handling contextual
information like the maximum stack size and "is this a jit address"
closure. Instead creators of `InstanceHandle` (like wasmtime) are now
responsible for invoking the start function.
* To avoid excessive use of `transmute` with lifetimes since the
traphandler state now has a lifetime the per-instance custom signal
handler is now replaced with a per-store custom signal handler. I'm
not entirely certain the purpose of the custom signal handler, though,
so I'd look for feedback on this part.
A new test has been added which ensures that if a host function
segfaults we don't accidentally try to handle it, and instead we
correctly report the segfault.
