* cranelift-wasm: Add a bounds-checking optimization for dynamic memories and guard pages
This is a new special case for when we know that there are enough guard pages to
cover the memory access's offset and access size.
The precise should-we-trap condition is
index + offset + access_size > bound
However, if we instead check only the partial condition
index > bound
then the most out of bounds that the access can be, while that partial check
still succeeds, is `offset + access_size`.
However, when we have a guard region that is at least as large as `offset +
access_size`, we can rely on the virtual memory subsystem handling these
out-of-bounds errors at runtime. Therefore, the partial `index > bound` check is
sufficient for this heap configuration.
Additionally, this has the advantage that a series of Wasm loads that use the
same dynamic index operand but different static offset immediates -- which is a
common code pattern when accessing multiple fields in the same struct that is in
linear memory -- will all emit the same `index > bound` check, which we can GVN.
* cranelift: Add WAT tests for accessing dynamic memories with the same index but different offsets
The bounds check comparison is GVN'd but we still branch on values we should
know will always be true if we get this far in the code. This is actual `br_if`s
in the non-Spectre code and `select_spectre_guard`s that we should know will
always go a certain way if we have Spectre mitigations enabled.
Improving the non-Spectre case is pretty straightforward: walk the dominator
tree and remember which values we've already branched on at this point, and
therefore we can simplify any further conditional branches on those same values
into direct jumps.
Improving the Spectre case requires something that is morally the same, but has
a few snags:
* We don't have actual `br_if`s to determine whether the bounds checking
condition succeeded or not. We need to instead reason about dominating
`select_spectre_guard; {load, store}` instruction pairs.
* We have to be SUPER careful about reasoning "through" `select_spectre_guard`s.
Our general rule is never to do that, since it could break the speculative
execution sandboxing that the instruction is designed for.
* Validate faulting addresses are valid to fault on
This commit adds a defense-in-depth measure to Wasmtime which is
intended to mitigate the impact of CVEs such as GHSA-ff4p-7xrq-q5r8.
Currently Wasmtime will catch `SIGSEGV` signals for WebAssembly code so
long as the instruction which faulted is an allow-listed instruction
(aka has a trap code listed for it). With the recent security issue,
however, the problem was that a wasm guest could exploit a compiler bug
to access memory outside of its sandbox. If the access was successful
there's no real way to detect that, but if the access was unsuccessful
then Wasmtime would happily swallow the `SIGSEGV` and report a nominal
trap. To embedders, this might look like nothing is going awry.
The new strategy implemented here in this commit is to attempt to be
more robust towards these sorts of failures. When a `SIGSEGV` is raised
the faulting pc is recorded but additionally the address of the
inaccessible location is also record. After the WebAssembly stack is
unwound and control returns to Wasmtime which has access to a `Store`
Wasmtime will now use this inaccessible faulting address to translate it
to a wasm address. This process should be guaranteed to succeed as
WebAssembly should only be able to access a well-defined region of
memory for all linear memories in a `Store`.
If no linear memory in a `Store` could contain the faulting address,
then Wasmtime now prints a scary message and aborts the process. The
purpose of this is to catch these sorts of bugs, make them very loud
errors, and hopefully mitigate impact. This would continue to not
mitigate the impact of a guest successfully loading data outside of its
sandbox, but if a guest was doing a sort of probing strategy trying to
find valid addresses then any invalid access would turn into a process
crash which would immediately be noticed by embedders.
While I was here I went ahead and additionally took a stab at #3120.
Traps due to `SIGSEGV` will now report the size of linear memory and the
address that was being accessed in addition to the bland "access out of
bounds" error. While this is still somewhat bland in the context of a
high level source language it's hopefully at least a little bit more
actionable for some. I'll note though that this isn't a guaranteed
contextual message since only the default configuration for Wasmtime
generates `SIGSEGV` on out-of-bounds memory accesses. Dynamically
bounds-checked configurations, for example, don't do this.
Testing-wise I unfortunately am not aware of a great way to test this.
The closet equivalent would be something like an `unsafe` method
`Config::allow_wasm_sandbox_escape`. In lieu of adding tests, though, I
can confirm that during development the crashing messages works just
fine as it took awhile on macOS to figure out where the faulting address
was recorded in the exception information which meant I had lots of
instances of recording an address of a trap not accessible from wasm.
* Fix tests
* Review comments
* Fix compile after refactor
* Fix compile on macOS
* Fix trap test for s390x
s390x rounds faulting addresses to 4k boundaries.
* x64: Take SIGFPE signals for divide traps
Prior to this commit Wasmtime would configure `avoid_div_traps=true`
unconditionally for Cranelift. This, for the division-based
instructions, would change emitted code to explicitly trap on trap
conditions instead of letting the `div` x86 instruction trap.
There's no specific reason for Wasmtime, however, to specifically avoid
traps in the `div` instruction. This means that the extra generated
branches on x86 aren't necessary since the `div` and `idiv` instructions
already trap for similar conditions as wasm requires.
This commit instead disables the `avoid_div_traps` setting for
Wasmtime's usage of Cranelift. Subsequently the codegen rules were
updated slightly:
* When `avoid_div_traps=true`, traps are no longer emitted for `div`
instructions.
* The `udiv`/`urem` instructions now list their trap as divide-by-zero
instead of integer overflow.
* The lowering for `sdiv` was updated to still explicitly check for zero
but the integer overflow case is deferred to the instruction itself.
* The lowering of `srem` no longer checks for zero and the listed trap
for the `div` instruction is a divide-by-zero.
This means that the codegen for `udiv` and `urem` no longer have any
branches. The codegen for `sdiv` removes one branch but keeps the
zero-check to differentiate the two kinds of traps. The codegen for
`srem` removes one branch but keeps the -1 check since the semantics of
`srem` mismatch with the semantics of `idiv` with a -1 divisor
(specifically for INT_MIN).
This is unlikely to have really all that much of a speedup but was
something I noticed during #6008 which seemed like it'd be good to clean
up. Plus Wasmtime's signal handling was already set up to catch
`SIGFPE`, it was just never firing.
* Remove the `avoid_div_traps` cranelift setting
With no known users currently removing this should be possible and helps
simplify the x64 backend.
* x64: GC more support for avoid_div_traps
Remove the `validate_sdiv_divisor*` pseudo-instructions and clean up
some of the ISLE rules now that `div` is allowed to itself trap
unconditionally.
* x64: Store div trap code in instruction itself
* Keep divisors in registers, not in memory
Don't accidentally fold multiple traps together
* Handle EXC_ARITHMETIC on macos
* Update emit tests
* Update winch and tests
* Initial support for the Relaxed SIMD proposal
This commit adds initial scaffolding and support for the Relaxed SIMD
proposal for WebAssembly. Codegen support is supported on the x64 and
AArch64 backends on this time.
The purpose of this commit is to get all the boilerplate out of the way
in terms of plumbing through a new feature, adding tests, etc. The tests
are copied from the upstream repository at this time while the
WebAssembly/testsuite repository hasn't been updated.
A summary of changes made in this commit are:
* Lowerings for all relaxed simd opcodes have been added, currently all
exhibiting deterministic behavior. This means that few lowerings are
optimal on the x86 backend, but on the AArch64 backend, for example,
all lowerings should be optimal.
* Support is added to codegen to, eventually, conditionally generate
different code based on input codegen flags. This is intended to
enable codegen to more efficient instructions on x86 by default, for
example, while still allowing embedders to force
architecture-independent semantics and behavior. One good example of
this is the `f32x4.relaxed_fmadd` instruction which when deterministic
forces the `fma` instruction, but otherwise if the backend doesn't
have support for `fma` then intermediate operations are performed
instead.
* Lowerings of `iadd_pairwise` for `i16x8` and `i32x4` were added to the
x86 backend as they're now exercised by the deterministic lowerings of
relaxed simd instructions.
* Sample codegen tests for added for x86 and aarch64 for some relaxed
simd instructions.
* Wasmtime embedder support for the relaxed-simd proposal and forcing
determinism have been added to `Config` and the CLI.
* Support has been added to the `*.wast` runtime execution for the
`(either ...)` matcher used in the relaxed-simd proposal.
* Tests for relaxed-simd are run both with a default `Engine` as well as
a "force deterministic" `Engine` to test both configurations.
* All tests from the upstream repository were copied into Wasmtime.
These tests should be deleted when WebAssembly/testsuite is updated.
* x64: Add x86-specific lowerings for relaxed simd
This commit builds on the prior commit and adds an array of `x86_*`
instructions to Cranelift which have semantics that match their
corresponding x86 equivalents. Translation for relaxed simd is then
additionally updated to conditionally generate different CLIF for
relaxed simd instructions depending on whether the target is x86 or not.
This means that for AArch64 no changes are made but for x86 most relaxed
instructions now lower to some x86-equivalent with slightly different
semantics than the "deterministic" lowering.
* Add libcall support for fma to Wasmtime
This will be required to implement the `f32x4.relaxed_madd` instruction
(and others) when an x86 host doesn't specify the `has_fma` feature.
* Ignore relaxed-simd tests on s390x and riscv64
* Enable relaxed-simd tests on s390x
* Update cranelift/codegen/meta/src/shared/instructions.rs
Co-authored-by: Andrew Brown <andrew.brown@intel.com>
* Add a FIXME from review
* Add notes about deterministic semantics
* Don't default `has_native_fma` to `true`
* Review comments and rebase fixes
---------
Co-authored-by: Andrew Brown <andrew.brown@intel.com>
* Remove globals from parking spot tests
Use `std:🧵:scope` to keep everything local to just the tests.
* Fix a panic due to a race in `unpark` and `park`
This commit fixes a panic in the `ParkingSpot` implementation where an
`unpark` signal may not get acknowledged when a waiter times out,
causing the waiter to remove itself from the internal map but panic
thinking that it missed an unpark signal.
The fix in this commit is to consume unpark signals when a timeout
happens. This can lead to another possible race I've detailed in the
comments which I believe is allowed by the specification of park/unpark
in wasm.
* Update crates/runtime/src/parking_spot.rs
Co-authored-by: Andrew Brown <andrew.brown@intel.com>
---------
Co-authored-by: Andrew Brown <andrew.brown@intel.com>
This works around a `rustc` bug where compiling with LTO
will sometimes strip out some of the trampoline entrypoint
symbols resulting in a linking failure.
At some point what is now `funcref` was called `anyfunc` and the spec changed,
but we didn't update our internal names. This does that.
Co-authored-by: Jamey Sharp <jsharp@fastly.com>
This commit removes the pooling of `Instance` allocations from the
pooling instance allocator. This means that the allocation of `Instance`
(and `VMContext`) memory, now always happens through the system `malloc`
and `free` instead of optionally being part of the pooling instance
allocator. Along the way this refactors the `InstanceAllocator` trait so
the pooling and on-demand allocators can share more structure with this
new property of the implementation.
The main rationale for this commit is to reduce the RSS of long-lived
programs which allocate instances with the pooling instance allocator
and aren't using the "next available" allocation strategy. In this
situation the memory for an instance is never decommitted until the end
of the program, meaning that eventually all instance slots will become
occupied and resident. This has the effect of Wasmtime slowly eating
more and more memory over time as each slot gets an instance allocated.
By switching to the system allocator this should reduce the current RSS
workload from O(used slots) to O(active slots), which is more in line
with expectations.
* Reimplement the pooling instance allocation strategy
This commit is a reimplementation of the strategy by which the pooling
instance allocator selects a slot for a module. Previously there was a
choice amongst three different algorithms: "reuse affinity", "next
available", and "random". The default was "reuse affinity" but some new
data has come to light which shows that this may not always be a good
default.
Notably the pooling allocator will retain some memory per-slot in the
pooling instance allocator, for example instance data or memory data
if-so-configured. This means that a currently unused, but previously
used, slot can contribute to the RSS usage of a program using Wasmtime.
Consequently the RSS impact here is O(max slots) which can be
counter-intuitive for embedders. This particularly affects "reuse
affinity" because the algorithm for picking a slot when there are no
affine slots is "pick a random slot", which means eventually all slots
will get used.
In discussions about possible ways to tackle this, an alternative to
"pick a strategy" arose and is now implemented in this commit.
Concretely the new allocation algorithm for a slot is now:
* First pick the most recently used affine slot, if one exists.
* Otherwise if the number of affine slots to other modules is above some
threshold N then pick the least-recently used affine slot.
* Otherwise pick a slot that's affine to nothing.
The "N" in this algorithm is configurable and setting it to 0 is the
same as the old "next available" strategy while setting it to infinity
is the same as the "reuse affinity" algorithm. Setting it to something
in the middle provides a knob to allow a modest "cache" of affine slots
while not allowing the total set of slots used to grow too much beyond
the maximal concurrent set of modules. The "random" strategy is now no
longer possible and was removed to help simplify the allocator.
* Resolve rustdoc warnings in `wasmtime-runtime` crate
* Remove `max_cold` as it duplicates the `slot_state.len()`
* More descriptive names
* Add a comment and debug assertion
* Add some list assertions
* Fix libcall relocations for precompiled modules
This commit fixes some asserts and support for relocation libcalls in
precompiled modules loaded from disk. In doing so this reworks how mmaps
are managed for files from disk. All non-file-backed `Mmap` entries are
read/write but file-backed versions were readonly. This commit changes
this such that all `Mmap` objects, even if they're file-backed, start as
read/write. The file-based versions all use copy-on-write to preserve
the private-ness of the mapping.
This is not functionally intended to change anything. Instead this
should have some more memory writable after a module is loaded but the
text section, for example, is still left as read/execute when loading is
finished. Additionally this makes modules compiled in memory more
consistent with modules loaded from disk.
* Update a comment
* Force images to become readonly during publish
This marks compiled images as entirely readonly during the
`CodeMemory::publish` step which happens just before the text section
becomes executable. This ensures that all images, no matter where they
come from, are guaranteed frozen before they start executing.
* Fix compile error on FreeBSD x64
* Fix compile on FreeBSD arm64
* Update Cargo.lock for ittapi
* vet: certify diff for ittapi libraries
Co-authored-by: Andrew Brown <andrew.brown@intel.com>
* Resolve libcall relocations for older CPUs
Long ago Wasmtime used to have logic for resolving relocations
post-compilation for libcalls which I ended up removing during
refactorings last year. As #5563 points out, however, it's possible to
get Wasmtime to panic by disabling SSE features which forces Cranelift
to use libcalls for some floating-point operations instead. Note that
this also requires disabling SIMD because SIMD support has a baseline of
SSE 4.2.
This commit pulls back the old implementations of various libcalls and
reimplements logic necessary to have them work on CPUs without SSE 4.2
Closes#5563
* Fix log message in `wast` support
* Fix offset listed in relocations
Be sure to factor in the offset of the function itself
* Review comments
* Use the `sym` operator for inline assembly
Avoids extra `#[no_mangle]` functions and undue symbols being exposed
from Wasmtime. This is a newly stabilized feature in Rust 1.66.0. I've
also added a `rust-version` entry to the `wasmtime` crate to try to head
off possible reports in the future about odd error messages or usage of
unstable features if the rustc version is too old.
* Fix a s390x warning
* Add `rust-version` annotation to Wasmtime crate
As the other main entrypoint for embedders.
* Simplify the `ModuleRuntimeInfo` trait slightly
Fold two functions into one as they're only called from one location
anyway.
* Remove ModuleRuntimeInfo::signature
This is redundant as the array mapping is already stored within the
`VMContext` so that can be consulted rather than having a separate trait
function for it. This required altering the `Global` creation slightly
to work correctly in this situation.
* Remove a now-dead constant
* Shared `VMOffsets` across instances
This commit removes the computation of `VMOffsets` to being per-module
instead of per-instance. The `VMOffsets` structure is also quite large
so this shaves off 112 bytes per instance which isn't a huge impact but
should help lower the cost of instantiating small modules.
* Remove `InstanceAllocator::adjust_tunables`
This is no longer needed or necessary with the pooling allocator.
* Fix compile warning
* Fix a vtune warning
* Fix pooling tests
* Fix another test warning
* Add a `WasmBacktrace::new()` constructor
This commit adds a method of manually capturing a backtrace of
WebAssembly frames within a `Store`. The new constructor can be called
with any `AsContext` values, primarily `&Store` and `&Caller`, during
host functions to inspect the calling state.
For now this does not respect the `Config::wasm_backtrace` option and
instead unconditionally captures the backtrace. It's hoped that this can
continue to adapt to needs of embedders by making it more configurable
int he future if necessary.
Closes#5339
* Split `new` into `capture` and `force_capture`
* Remove some custom error types in Wasmtime
These types are mostly cumbersome to work with nowadays that `anyhow` is
used everywhere else. This commit removes `InstantiationError` and
`SetupError` in favor of using `anyhow::Error` throughout. This can
eventually culminate in creation of specific errors for embedders to
downcast to but for now this should be general enough.
* Fix Windows build
The main change here is that io-lifetimes 1.0 switches to use the I/O safety
feature in the standard library rather than providing its own copy.
This also updates to windows-sys 0.42.0 and rustix 0.36.
* Clear affine slots when dropping a `Module`
This commit implements a resource usage optimization for Wasmtime with
the pooling instance allocator by ensuring that when a `Module` is
dropped its backing virtual memory mappings are all removed. Currently
when a `Module` is dropped it releases a strong reference to its
internal memory image but the memory image may stick around in
individual pooling instance allocator slots. When using the `Random`
allocation strategy, for example, this means that the memory images
could stick around for a long time.
While not a pressing issue this has resource usage implications for
Wasmtime. Namely removing a `Module` does not guarantee the memfd, if in
use for a memory image, is closed and deallocated within the kernel.
Unfortunately simply closing the memfd is not sufficient as well as the
mappings into the address space additionally all need to be removed for
the kernel to release the resources for the memfd. This means that to
release all kernel-level resources for a `Module` all slots which have
the memory image mapped in must have the slot reset.
This problem isn't particularly present when using the `NextAvailable`
allocation strategy since the number of lingering memfds is proportional
to the maximum concurrent size of wasm instances. With the `Random` and
`ReuseAffinity` strategies, however, it's much more prominent because
the number of lingering memfds can reach the total number of slots
available. This can appear as a leak of kernel-level memory which can
cause other system instability.
To fix this issue this commit adds necessary instrumentation to `Drop
for Module` to purge all references to the module in the pooling
instance allocator. All index allocation strategies now maintain
affinity tracking to ensure that regardless of the strategy in use a
module that is dropped will remove all its memory mappings. A new
allocation method was added to the index allocator for allocating an
index without setting affinity and only allocating affine slots. This is
used to iterate over all the affine slots without holding the global
index lock for an unnecessarily long time while mappings are removed.
* Review comments
This commit refactors the internals of `wasmtime_runtime::SharedMemory`
a bit to expose the necessary functions to invoke from the
`wasmtime::SharedMemory` layer. Notably some items are moved out of the
`RwLock` from prior, such as the type and the `VMMemoryDefinition`.
Additionally the organization around the `atomic_*` methods has been
redone to ensure that the `wasmtime`-layer abstraction has a single
method to call into which everything else uses as well.
From the documentation of `CondVar::wait_timeout`:
> The semantics of this function are equivalent to wait except that the thread
> will be blocked for roughly no longer than `dur`. This method should not be
> used for precise timing due to anomalies such as preemption or platform
> differences that might not cause the maximum amount of time waited to be
> precisely `dur`.
Therefore, go to sleep again, if the thread has not slept long enough.
Signed-off-by: Harald Hoyer <harald@profian.com>
Signed-off-by: Harald Hoyer <harald@profian.com>
* feat: implement memory.atomic.notify,wait32,wait64
Added the parking_spot crate, which provides the needed registry for the
operations.
Signed-off-by: Harald Hoyer <harald@profian.com>
* fix: change trap message for HeapMisaligned
The threads spec test wants "unaligned atomic"
instead of "misaligned memory access".
Signed-off-by: Harald Hoyer <harald@profian.com>
* tests: add test for atomic wait on non-shared memory
Signed-off-by: Harald Hoyer <harald@profian.com>
* tests: add tests/spec_testsuite/proposals/threads
without pooling and reference types.
Also "shared_memory" is added to the "spectest" interface.
Signed-off-by: Harald Hoyer <harald@profian.com>
* tests: add atomics_notify.wast
checking that notify with 0 waiters returns 0 on shared and non-shared
memory.
Signed-off-by: Harald Hoyer <harald@profian.com>
* tests: add tests for atomic wait on shared memory
- return 2 - timeout for 0
- return 2 - timeout for 1000ns
- return 1 - invalid value
Signed-off-by: Harald Hoyer <harald@profian.com>
* fixup! feat: implement memory.atomic.notify,wait32,wait64
Signed-off-by: Harald Hoyer <harald@profian.com>
* fixup! feat: implement memory.atomic.notify,wait32,wait64
Signed-off-by: Harald Hoyer <harald@profian.com>
Signed-off-by: Harald Hoyer <harald@profian.com>
This commit replaces `wasmtime_environ::TrapCode` with `wasmtime::Trap`.
This is possible with past refactorings which slimmed down the `Trap`
definition in the `wasmtime` crate to a simple `enum`. This means that
there's one less place that all the various trap opcodes need to be
listed in Wasmtime.
This commit updates the index allocation performed in the pooling
allocator with a few refactorings:
* With `cfg(fuzzing)` a deterministic rng is now used to improve
reproducibility of fuzz test cases.
* The `Mutex` was pushed inside of `IndexAllocator`, renamed from
`PoolingAllocationState`.
* Randomness is now always done through a `SmallRng` stored in the
`IndexAllocator` instead of using `thread_rng`.
* The `is_empty` method has been removed in favor of an `Option`-based
return on `alloc`.
This refactoring is additionally intended to encapsulate more
implementation details of `IndexAllocator` to more easily allow for
alternate implementations in the future such as lock-free approaches
(possibly).
* Fix CI after CVE fixes
Alas we can't run CI ahead of time so this fixes various minor build
issues from the merging of the recent CVE fixes. Note that I plan to
publish the advisories once CI issues are sorted out.
* Fix mmap/free of zero bytes
* Unconditionally use `MemoryImageSlot`
This commit removes the internal branching within the pooling instance
allocator to sometimes use a `MemoryImageSlot` and sometimes now.
Instead this is now unconditionally used in all situations on all
platforms. This fixes an issue where the state of a slot could get
corrupted if modules being instantiated switched from having images to
not having an image or vice versa.
The bulk of this commit is the removal of the `memory-init-cow`
compile-time feature in addition to adding Windows support to the
`cow.rs` file.
* Fix compile on Unix
* Add a stricter assertion for static memory bounds
Double-check that when a memory is allocated the configuration required
is satisfied by the pooling allocator.
Before, we would do a `heap_addr` to translate the given Wasm memory address
into a native memory address and pass it into the libcall that implemented the
atomic operation, which would then treat the address as a Wasm memory address
and pass it to `validate_atomic_addr` to be bounds checked a second time. This
is a bit nonsensical, as we are validating a native memory address as if it were
a Wasm memory address.
Now, we no longer do a `heap_addr` to translate the Wasm memory address to a
native memory address. Instead, we pass the Wasm memory address to the libcall,
and the libcall is responsible for doing the bounds check (by calling
`validate_atomic_addr` with the correct type of memory address now).
* Wasmtime+Cranelift: strip out some dead x86-32 code.
I was recently pointed to fastly/Viceroy#200 where it seems some folks
are trying to compile Wasmtime (via Viceroy) for Windows x86-32 and the
failures may not be loud enough. I've tried to reproduce this
cross-compiling to i686-pc-windows-gnu from Linux and hit build failures
(as expected) in several places. Nevertheless, while trying to discern
what others may be attempting, I noticed some dead x86-32-specific code
in our repo, and figured it would be a good idea to clean this up.
Otherwise, it (i) sends some mixed messages -- "hey look, this codebase
does support x86-32" -- and (ii) keeps untested code around, which is
generally not great.
This PR removes x86-32-specific cases in traphandlers and unwind code,
and Cranelift's native feature detection. It adds helpful compile-error
messages in a few cases. If we ever support x86-32 (contributors
welcome! The big missing piece is Cranelift support; see #1980), these
compile errors and git history should be enough to recover any knowledge
we are now encoding in the source.
I left the x86-32 support in `wasmtime-fiber` alone because that seems
like a bit of a special case -- foundation library, separate from the
rest of Wasmtime, with specific care to provide a (presumably working)
full 32-bit version.
* Remove some extraneous compile_error!s, already covered by others.
* Implement support for dynamic memories in the pooling allocator
This is a continuation of the thrust in #5207 for reducing page faults
and lock contention when using the pooling allocator. To that end this
commit implements support for efficient memory management in the pooling
allocator when using wasm that is instrumented with bounds checks.
The `MemoryImageSlot` type now avoids unconditionally shrinking memory
back to its initial size during the `clear_and_remain_ready` operation,
instead deferring optional resizing of memory to the subsequent call to
`instantiate` when the slot is reused. The instantiation portion then
takes the "memory style" as an argument which dictates whether the
accessible memory must be precisely fit or whether it's allowed to
exceed the maximum. This in effect enables skipping a call to `mprotect`
to shrink the heap when dynamic memory checks are enabled.
In terms of page fault and contention this should improve the situation
by:
* Fewer calls to `mprotect` since once a heap grows it stays grown and
it never shrinks. This means that a write lock is taken within the
kernel much more rarely from before (only asymptotically now, not
N-times-per-instance).
* Accessed memory after a heap growth operation will not fault if it was
previously paged in by a prior instance and set to zero with `memset`.
Unlike #5207 which requires a 6.0 kernel to see this optimization this
commit enables the optimization for any kernel.
The major cost of choosing this strategy is naturally the performance
hit of the wasm itself. This is being looked at in PRs such as #5190 to
improve Wasmtime's story here.
This commit does not implement any new configuration options for
Wasmtime but instead reinterprets existing configuration options. The
pooling allocator no longer unconditionally sets
`static_memory_bound_is_maximum` and then implements support necessary
for this memory type. This other change to this commit is that the
`Tunables::static_memory_bound` configuration option is no longer gating
on the creation of a `MemoryPool` and it will now appropriately size to
`instance_limits.memory_pages` if the `static_memory_bound` is to small.
This is done to accomodate fuzzing more easily where the
`static_memory_bound` will become small during fuzzing and otherwise the
configuration would be rejected and require manual handling. The spirit
of the `MemoryPool` is one of large virtual address space reservations
anyway so it seemed reasonable to interpret the configuration this way.
* Skip zero memory_size cases
These are causing errors to happen when fuzzing and otherwise in theory
shouldn't be too interesting to optimize for anyway since they likely
aren't used in practice.
When new wasm instances are created repeatedly in high-concurrency
environments one of the largest bottlenecks is the contention on
kernel-level locks having to do with the virtual memory. It's expected
that usage in this environment is leveraging the pooling instance
allocator with the `memory-init-cow` feature enabled which means that
the kernel level VM lock is acquired in operations such as:
1. Growing a heap with `mprotect` (write lock)
2. Faulting in memory during usage (read lock)
3. Resetting a heap's contents with `madvise` (read lock)
4. Shrinking a heap with `mprotect` when reusing a slot (write lock)
Rapid usage of these operations can lead to detrimental performance
especially on otherwise heavily loaded systems, worsening the more
frequent the above operations are. This commit is aimed at addressing
the (2) case above, reducing the number of page faults that are
fulfilled by the kernel.
Currently these page faults happen for three reasons:
* When memory is first accessed after the heap is grown.
* When the initial linear memory image is accessed for the first time.
* When the initial zero'd heap contents, not part of the linear memory
image, are accessed.
This PR is attempting to address the latter of these cases, and to a
lesser extent the first case as well. Specifically this PR provides the
ability to partially reset a pooled linear memory with `memset` rather
than `madvise`. This is done to have the same effect of resetting
contents to zero but namely has a different effect on paging, notably
keeping the pages resident in memory rather than returning them to the
kernel. This means that reuse of a linear memory slot on a page that was
previously `memset` will not trigger a page fault since everything
remains paged into the process.
The end result is that any access to linear memory which has been
touched by `memset` will no longer page fault on reuse. On more recent
kernels (6.0+) this also means pages which were zero'd by `memset`, made
inaccessible with `PROT_NONE`, and then made accessible again with
`PROT_READ | PROT_WRITE` will not page fault. This can be common when a
wasm instances grows its heap slightly, uses that memory, but then it's
shrunk when the memory is reused for the next instance. Note that this
kernel optimization requires a 6.0+ kernel.
This same optimization is furthermore applied to both async stacks with
the pooling memory allocator in addition to table elements. The defaults
of Wasmtime are not changing with this PR, instead knobs are being
exposed for embedders to turn if they so desire. This is currently being
experimented with at Fastly and I may come back and alter the defaults
of Wasmtime if it seems suitable after our measurements.
This commit changes the APIs in the `wasmtime` crate for configuring the
pooling allocator. I plan on adding a few more configuration options in
the near future and the current structure was feeling unwieldy for
adding these new abstractions.
The previous `struct`-based API has been replaced with a builder-style
API in a similar shape as to `Config`. This is done to help make it
easier to add more configuration options in the future through adding
more methods as opposed to adding more field which could break prior
initializations.
* Pull `Module` out of `ModuleTextBuilder`
This commit is the first in what will likely be a number towards
preparing for serializing a compiled component to bytes, a precompiled
artifact. To that end my rough plan is to merge all of the compiled
artifacts for a component into one large object file instead of having
lots of separate object files and lots of separate mmaps to manage. To
that end I plan on eventually using `ModuleTextBuilder` to build one
large text section for all core wasm modules and trampolines, meaning
that `ModuleTextBuilder` is no longer specific to one module. I've
extracted out functionality such as function name calculation as well as
relocation resolving (now a closure passed in) in preparation for this.
For now this just keeps tests passing, and the trajectory for this
should become more clear over the following commits.
* Remove component-specific object emission
This commit removes the `ComponentCompiler::emit_obj` function in favor
of `Compiler::emit_obj`, now renamed `append_code`. This involved
significantly refactoring code emission to take a flat list of functions
into `append_code` and the caller is responsible for weaving together
various "families" of functions and un-weaving them afterwards.
* Consolidate ELF parsing in `CodeMemory`
This commit moves the ELF file parsing and section iteration from
`CompiledModule` into `CodeMemory` so one location keeps track of
section ranges and such. This is in preparation for sharing much of this
code with components which needs all the same sections to get tracked
but won't be using `CompiledModule`. A small side benefit from this is
that the section parsing done in `CodeMemory` and `CompiledModule` is no
longer duplicated.
* Remove separately tracked traps in components
Previously components would generate an "always trapping" function
and the metadata around which pc was allowed to trap was handled
manually for components. With recent refactorings the Wasmtime-standard
trap section in object files is now being generated for components as
well which means that can be reused instead of custom-tracking this
metadata. This commit removes the manual tracking for the `always_trap`
functions and plumbs the necessary bits around to make components look
more like modules.
* Remove a now-unnecessary `Arc` in `Module`
Not expected to have any measurable impact on performance, but
complexity-wise this should make it a bit easier to understand the
internals since there's no longer any need to store this somewhere else
than its owner's location.
* Merge compilation artifacts of components
This commit is a large refactoring of the component compilation process
to produce a single artifact instead of multiple binary artifacts. The
core wasm compilation process is refactored as well to share as much
code as necessary with the component compilation process.
This method of representing a compiled component necessitated a few
medium-sized changes internally within Wasmtime:
* A new data structure was created, `CodeObject`, which represents
metadata about a single compiled artifact. This is then stored as an
`Arc` within a component and a module. For `Module` this is always
uniquely owned and represents a shuffling around of data from one
owner to another. For a `Component`, however, this is shared amongst
all loaded modules and the top-level component.
* The "module registry" which is used for symbolicating backtraces and
for trap information has been updated to account for a single region
of loaded code holding possibly multiple modules. This involved adding
a second-level `BTreeMap` for now. This will likely slow down
instantiation slightly but if it poses an issue in the future this
should be able to be represented with a more clever data structure.
This commit additionally solves a number of longstanding issues with
components such as compiling only one host-to-wasm trampoline per
signature instead of possibly once-per-module. Additionally the
`SignatureCollection` registration now happens once-per-component
instead of once-per-module-within-a-component.
* Fix compile errors from prior commits
* Support AOT-compiling components
This commit adds support for AOT-compiled components in the same manner
as `Module`, specifically adding:
* `Engine::precompile_component`
* `Component::serialize`
* `Component::deserialize`
* `Component::deserialize_file`
Internally the support for components looks quite similar to `Module`.
All the prior commits to this made adding the support here
(unsurprisingly) easy. Components are represented as a single object
file as are modules, and the functions for each module are all piled
into the same object file next to each other (as are areas such as data
sections). Support was also added here to quickly differentiate compiled
components vs compiled modules via the `e_flags` field in the ELF
header.
* Prevent serializing exported modules on components
The current representation of a module within a component means that the
implementation of `Module::serialize` will not work if the module is
exported from a component. The reason for this is that `serialize`
doesn't actually do anything and simply returns the underlying mmap as a
list of bytes. The mmap, however, has `.wasmtime.info` describing
component metadata as opposed to this module's metadata. While rewriting
this section could be implemented it's not so easy to do so and is
otherwise seen as not super important of a feature right now anyway.
* Fix windows build
* Fix an unused function warning
* Update crates/environ/src/compilation.rs
Co-authored-by: Nick Fitzgerald <fitzgen@gmail.com>
Co-authored-by: Nick Fitzgerald <fitzgen@gmail.com>
* Add a benchmark for traps with many Wasm<-->host calls on the stack
* Add a test for expected Wasm stack traces with Wasm<--host calls on the stack when we trap
* Don't re-capture backtraces when propagating traps through host frames
This fixes some accidentally quadratic code where we would re-capture a Wasm
stack trace (takes `O(n)` time) every time we propagated a trap through a host
frame back to Wasm (can happen `O(n)` times). And `O(n) * O(n) = O(n^2)`, of
course. Whoops. After this commit, it trapping with a call stack that is `n`
frames deep of Wasm-to-host-to-Wasm calls just captures a single backtrace and
is therefore just a proper `O(n)` time operation, as it is intended to be.
Now we explicitly track whether we need to capture a Wasm backtrace or not when
raising a trap. This unfortunately isn't as straightforward as one might hope,
however, because of the split between `wasmtime::Trap` and
`wasmtime_runtime::Trap`. We need to decide whether or not to capture a Wasm
backtrace inside `wasmtime_runtime` but in order to determine whether to do that
or not we need to reflect on the `anyhow::Error` and see if it is a
`wasmtime::Trap` that already has a backtrace or not. This can't be done the
straightforward way because it would introduce a cyclic dependency between the
`wasmtime` and `wasmtime-runtime` crates. We can't merge those two `Trap`
types-- at least not without effectively merging the whole `wasmtime` and
`wasmtime-runtime` crates together, which would be a good idea in a perfect
world but would be a *ton* of ocean boiling from where we currently are --
because `wasmtime::Trap` does symbolication of stack traces which relies on
module registration information data that resides inside the `wasmtime` crate
and therefore can't be moved into `wasmtime-runtime`. We resolve this problem by
adding a boolean to `wasmtime_runtime::raise_user_trap` that controls whether we
should capture a Wasm backtrace or not, and then determine whether we need a
backtrace or not at each of that function's call sites, which are in `wasmtime`
and therefore can do the reflection to determine whether the user trap already
has a backtrace or not. Phew!
Fixes#5037
* debug assert that we don't record unnecessary backtraces for traps
* Add assertions around `needs_backtrace`
Unfortunately we can't do
debug_assert_eq!(needs_backtrace, trap.inner.backtrace.get().is_some());
because `needs_backtrace` doesn't consider whether Wasm backtraces have been
disabled via config.
* Consolidate `needs_backtrace` calculation followed by calling `raise_user_trap` into one place
* Make wasmtime build for windows-aarch64
* Add check for win arm64 build.
* Fix checks for winarm64 key in workflows.
* Add target in windows arm64 build.
* Add tracking issue for Windows ARM64 trap handling
This commit updates the `MIN_STACK_SIZE` constant for Unix platforms
when allocating a sigaltstack from 16k to 64k. The signal handler
captures a wasm `Backtrace` which involves memory allocations and it was
recently discovered that, at least in debug mode, jemalloc can take up
to 16k of stack space for an allocation. To allow running the
sigaltstack size is increased here.
This historically was used to guard against recursive faults but
later refactorings have made this variable somewhat obsolete. The code
that it still protects is not the "meat" of trap handling. Instead the
`jmp_buf_if_trap` is changed to be more like "take" so once a "take"
succeeds it won't be able to recursively call any more "meat".
Overall this shouldn't affect anything, it's just a small internal
cleanup.
* Leverage Cargo's workspace inheritance feature
This commit is an attempt to reduce the complexity of the Cargo
manifests in this repository with Cargo's workspace-inheritance feature
becoming stable in Rust 1.64.0. This feature allows specifying fields in
the root workspace `Cargo.toml` which are then reused throughout the
workspace. For example this PR shares definitions such as:
* All of the Wasmtime-family of crates now use `version.workspace =
true` to have a single location which defines the version number.
* All crates use `edition.workspace = true` to have one default edition
for the entire workspace.
* Common dependencies are listed in `[workspace.dependencies]` to avoid
typing the same version number in a lot of different places (e.g. the
`wasmparser = "0.89.0"` is now in just one spot.
Currently the workspace-inheritance feature doesn't allow having two
different versions to inherit, so all of the Cranelift-family of crates
still manually specify their version. The inter-crate dependencies,
however, are shared amongst the root workspace.
This feature can be seen as a method of "preprocessing" of sorts for
Cargo manifests. This will help us develop Wasmtime but shouldn't have
any actual impact on the published artifacts -- everything's dependency
lists are still the same.
* Fix wasi-crypto tests
* Update to cap-std 0.26.
This is primarily to pull in bytecodealliance/cap-std#271, the fix for #4936,
compilation on Rust nightly on Windows.
It also updates to rustix 0.35.10, to pull in bytecodealliance/rustix#403,
the fix for bytecodealliance/rustix#402, compilation on newer versions of
the libc crate, which changed a public function from `unsafe` to safe.
Fixes#4936.
* Update the system-interface audit for 0.23.
* Update the libc supply-chain config version.
* Initial forward-edge CFI implementation
Give the user the option to start all basic blocks that are targets
of indirect branches with the BTI instruction introduced by the
Branch Target Identification extension to the Arm instruction set
architecture.
Copyright (c) 2022, Arm Limited.
* Refactor `from_artifacts` to avoid second `make_executable` (#1)
This involves "parsing" twice but this is parsing just the header of an
ELF file so it's not a very intensive operation and should be ok to do
twice.
* Address the code review feedback
Copyright (c) 2022, Arm Limited.
Co-authored-by: Alex Crichton <alex@alexcrichton.com>
This commit replaces #4869 and represents the actual version bump that
should have happened had I remembered to bump the in-tree version of
Wasmtime to 1.0.0 prior to the branch-cut date. Alas!
We were previously implicitly assuming that all Wasm frames in a stack used the
same `VMRuntimeLimits` as the previous frame we walked, but this is not true
when Wasm in store A calls into the host which then calls into Wasm in store B:
| ... |
| Host | |
+-----------------+ | stack
| Wasm in store A | | grows
+-----------------+ | down
| Host | |
+-----------------+ |
| Wasm in store B | V
+-----------------+
Trying to walk this stack would previously result in a runtime panic.
The solution is to push the maintenance of our list of saved Wasm FP/SP/PC
registers that allow us to identify contiguous regions of Wasm frames on the
stack deeper into `CallThreadState`. The saved registers list is now maintained
whenever updating the `CallThreadState` linked list by making the
`CallThreadState::prev` field private and only accessible via a getter and
setter, where the setter always maintains our invariants.
* Fix a compile error on nightly Rust
It looks like Rust nightly has gotten a bit more strict about
attributes-on-expressions and previously accepted code is no longer
accepted. This commit updates the generated code for a macro to a form
which is accepted by rustc.
* Fix a soundness issue with lowering variants
This commit fixes a soundness issue lowering variants in the component
model where host memory could be leaked to the guest module by accident.
In reviewing code recently for `Val::lower` I noticed that the variant
lowering was extending the payload with `ValRaw::u32(0)` to
appropriately fit the size of the variant. In reading this it appeared
incorrect to me due to the fact that it should be `ValRaw::u64(0)` since
up to 64-bits can be read. Additionally this implementation was also
incorrect because the lowered representation of the payload itself was
not possibly zero-extended to 64-bits to accommodate other variants.
It turned out these issues were benign because with the dynamic
surface area to the component model the arguments were all initialized
to 0 anyway. The static version of the API, however, does not initialize
arguments to 0 and I wanted to initially align these two implementations
so I updated the variant implementation of lowering for dynamic values
and removed the zero-ing of arguments.
To test this change I updated the `debug` mode of adapter module
generation to assert that the upper bits of values in wasm are always
zero when the value is casted down (during `stack_get` which only
happens with variants). I then threaded through the `debug` boolean
configuration parameter into the dynamic and static fuzzers.
To my surprise this new assertion tripped even after the fix was
applied. It turns out, though, that there was other leakage of bits
through other means that I was previously unaware of. At the primitive
level lowerings of types like `u32` will have a `Lower` representation
of `ValRaw` and the lowering is simply `dst.write(ValRaw::i32(self))`,
or the equivalent thereof. The problem, that the fuzzers detected, with
this pattern is that the `ValRaw` type is 16-bytes, and
`ValRaw::i32(X)` only initializes the first 4. This meant that all the
lowerings for all primitives were writing up to 12 bytes of garbage from
the host for the wasm module to read.
It turned out that this write of a `ValRaw` was sometimes 16 bytes and
sometimes the appropriate size depending on the number of optimizations
in play. With enough inlining for example `dst.write(ValRaw::i32(self))`
would only write 4 bytes, as expected. In debug mode though without
inlining 16 bytes would be written, including the garbage from the upper
bits.
To solve this issue I ended up taking a somewhat different approach. I
primarily updated the `ValRaw` constructors to simply always extend the
values internally to 64-bits, meaning that the low 8 bytes of a `ValRaw`
is always initialized. This prevents any undefined data from leaking
from the host into a wasm module, and means that values are also
zero-extended even if they're only used in 32-bit contexts outside of a
variant. This felt like the best fix for now, though, in terms of
not really having a performance impact while additionally not requiring
a rewrite of all lowerings.
This solution ended up also neatly removing the "zero out the entire
payload" logic that was previously require. Now after a payload is
lowered only the tail end of the payload, up to the size of the variant,
is zeroed out. This means that each lowered argument is written to at
most once which should hopefully be a small performance boost for
calling into functions as well.
* Rename `MmapVec::drain` to `split_off`
As suggested on #4609
* Fix tests
* Make MmapVec::split_off work like Vec::split_off
Co-authored-by: Jamey Sharp <jsharp@fastly.com>
This commit fixes a build warning on Rust 1.63 when the `memory-init-cow`
feature is disabled in the `wasmtime-runtime` crate. Some "tricks" were
used prior to have the `MemoryImage` type be an empty `enum {}` but that
wreaks havoc with warnings so this commit instead just makes it a unit
struct and makes all methods panic (as they shouldn't be hit anyway).
