This is a follow-on to the earlier work in reducing string copies,
mainly focused on ensuring that ASCII strings are easy to provide to the
JS runtime.
While we are replacing a 16-byte reference in a number of places with a
24-byte structure (measured via `std::mem::size_of`), the reduction in
copies wins out over the additional size of the arguments passed into
functions.
Benchmarking shows approximately the same if not slightly less wallclock
time/instructions retired, but I believe this continues to open up
further refactoring opportunities.
This will improve diagnostics and catch any non-ASCII extension code
early.
This will use `debug_assert!` rather than `assert!` to avoid runtime
costs, and ensures (in debug_assert mode only) that all extension source
files are ASCII as we load them.
Reduce the number of copies and allocations of script code by carrying
around ownership/reference information from creation time.
As an advantage, this allows us to maintain the identity of `&'static
str`-based scripts and use v8's external 1-byte strings (to avoid
incorrectly passing non-ASCII strings, debug `assert!`s gate all string
reference paths).
Benchmark results:
Perf improvements -- ~0.1 - 0.2ms faster, but should reduce garbage
w/external strings and reduces data copies overall. May also unlock some
more interesting optimizations in the future.
This requires adding some generics to functions, but manual
monomorphization has been applied (outer/inner function) to avoid code
bloat.
This implements two macros to simplify extension registration and centralize a lot of the boilerplate as a base for future improvements:
* `deno_core::ops!` registers a block of `#[op]`s, optionally with type
parameters, useful for places where we share lists of ops
* `deno_core::extension!` is used to register an extension, and creates
two methods that can be used at runtime/snapshot generation time:
`init_ops` and `init_ops_and_esm`.
---------
Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
This PR cleans up APIs related to snapshot creation and how ops are
initialized.
Prerequisite for #18080
---------
Co-authored-by: Divy Srivastava <dj.srivastava23@gmail.com>
This commit changes "ModuleMap" initialization to over-allocate by 16
for vectors storing module information and module V8 handles. In 99%
cases there's gonna be at least one additional module loaded, so it's
very wasteful to have to reallocate when the module is executed (IIRC
Rust will double the size of the vector) and move all of the elements.
This commit renames "deno_core::InternalModuleLoader" to
"ExtModuleLoader" and changes the specifiers used by the
modules loaded from this loader to "ext:".
"internal:" scheme was really ambiguous and it's more characters than
"ext:", which should result in slightly smaller snapshot size.
Closes https://github.com/denoland/deno/issues/18020
This commit changes "InternalModuleLoader" from "deno_core" to
store a list of used modules during snapshotting. If a module was not
used during snapshotting "InternalModuleLoader" will panic in its "Drop"
handler signaling to the embedder that they made a mistake somewhere.
Instead of relying on "serde_v8" which is very inefficient in
serializing enums, I'm hand rolling serde for "ModuleMap" data
that is stored in the V8 snapshot to make ES modules
snapshottable.
```
// this branch
Benchmark #2: ./target/release/deno run empty.js
Time (mean ± σ): 21.4 ms ± 0.9 ms [User: 15.6 ms, System: 6.4 ms]
Range (min … max): 20.2 ms … 24.4 ms
// main branch
Benchmark #2: ./target/release/deno run empty.js
Time (mean ± σ): 23.1 ms ± 1.2 ms [User: 17.0 ms, System: 6.2 ms]
Range (min … max): 21.0 ms … 26.0 ms
```
This allows to not include source code into the binary (because
it will already be included in the V8 snapshot).
Nothing changes for the embedders - everything should still build the
same.
This commit brings the binary size from 87Mb to 82Mb on M1.
Alternative to https://github.com/denoland/deno/pull/17820 and
https://github.com/denoland/deno/pull/17653
---------
Co-authored-by: Leo Kettmeir <crowlkats@toaxl.com>
This commit changes definition of "ExtensionFileSource", by changing
"code" field to being "ExtensionFileSourceCode" enum. Currently the enum
has only a single variant "IncludedInBinary". It is done in preparation
to allow embedders to decide if they want to include the source code in the
binary when snapshotting (in most cases they shouldn't do that).
In the follow up commit we'll add more variants to
"ExtensionFileSourceCode".
"include_js_files_dir!" macro was removed in favor "include_js_files!"
macro which can now accept "dir" option.
This commit adds "ExtensionBuilder::esm_entry_point()" function that
allows to specify which of the extension files should be treated as an
entry point. If the entry point is not provided all modules are loaded
and evaluated, but if it is provided then only the entry point is explicitly
loaded and evaluated.
Co-authored-by: Leo Kettmeir <crowlkats@toaxl.com>
This is a proof of concept for being able to snapshot TypeScript files.
Currently only a single runtime file is authored in TypeScript -
"runtime/js/01_version.ts".
Not needed infrastructure was removed from "core/snapshot_util.rs".
---------
Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
This PR refactors all internal js files (except core) to be written as
ES modules.
`__bootstrap`has been mostly replaced with static imports in form in
`internal:[path to file from repo root]`.
To specify if files are ESM, an `esm` method has been added to
`Extension`, similar to the `js` method.
A new ModuleLoader called `InternalModuleLoader` has been added to
enable the loading of internal specifiers, which is used in all
situations except when a snapshot is only loaded, and not a new one is
created from it.
---------
Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
The commit derives Eq, PartialEq, and Debug traits for the
`ResolutionKind` enum to make it possible for external
implementors to assert ResolutionKind.
- changes module id to be usize & 0 based instead of 1 based
- merges `ids_by_handle` & `handles_by_id` to be a single `handles`
vector
- removes `next_module_id`, as vector is used
- turns `info` into a vector
This commit adds support for snapshotting ES modules. This is done by
adding an ability to serialize and deserialize a "ModuleMap" and attach
it
to the snapshot, using "add_context_data" API.
This has been tested with 400 modules and seems to not have a limit on
the number of modules that might be snapshotted.
Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
This commit changes signature of "deno_core::ModuleLoader::resolve" to pass
an enum indicating whether or not we're resolving a specifier for dynamic import.
Additionally "CliModuleLoader" was changes to store both "parent permissions" (or
"root permissions") as well as "dynamic permissions" that allow to check for permissions
in top-level module load an dynamic imports.
Then all code paths that have anything to do with Node/npm compat are now checking
for permissions which are passed from module loader instance associated with given
worker.
This commit updates unhelpful messages that are raised when event loop
stalls on unresolved top-level promises.
Instead of "Module evaluation is still pending but there are no pending
ops or dynamic imports. This situation is often caused by unresolved
promises." and "Dynamically imported module evaluation is still pending
but there are no pending ops. This situation is often caused by
unresolved promises." we are now printing a message like:
error: Top-level await promise never resolved
[SOURCE LINE]
^
at [FUNCTION NAME] ([FILENAME])
eg:
error: Top-level await promise never resolved
await new Promise((_resolve, _reject) => {});
^
at <anonymous>
(file:///Users/ib/dev/deno/cli/tests/testdata/test/unresolved_promise.ts:1:1)
Co-authored-by: David Sherret <dsherret@users.noreply.github.com>
Several functions used for handling of dynamic imports and "import.meta"
object were not registered as external references and caused V8 to crash
during snapshotting. These functions are now registered as external refs
and aborts are no longer happening.
Welcome to better optimised op calls! Currently opSync is called with parameters of every type and count. This most definitely makes the call megamorphic. Additionally, it seems that spread params leads to V8 not being able to optimise the calls quite as well (apparently Fast Calls cannot be used with spread params).
Monomorphising op calls should lead to some improved performance. Now that unwrapping of sync ops results is done on Rust side, this is pretty simple:
```
opSync("op_foo", param1, param2);
// -> turns to
ops.op_foo(param1, param2);
```
This means sync op calls are now just directly calling the native binding function. When V8 Fast API Calls are enabled, this will enable those to be called on the optimised path.
Monomorphising async ops likely requires using callbacks and is left as an exercise to the reader.
When a dynamically imported module gets resolved, any code that comes after an
await import() to that module will continue running. However, if that is the
last code in the evaluation of another dynamically imported module, that second
module will not resolve until the next iteration of the event loop, even though
it does not depend on the event loop at all.
When the event loop is being blocked by a long-running operation, such as a
long-running timer, or by an async op that might never end, such as with workers
or BroadcastChannels, that will result in the second dynamically imported module
not being resolved for a while, or ever.
This change fixes this by running the dynamic module loading steps in a loop
until no more dynamic modules can be resolved.
This commit adds proper support for import assertions and JSON modules.
Implementation of "core/modules.rs" was changed to account for multiple possible
module types, instead of always assuming that the code is an "ES module". In
effect "ModuleMap" now has knowledge about each modules' type (stored via
"ModuleType" enum). Module loading pipeline now stores information about
expected module type for each request and validates that expected type matches
discovered module type based on file's "MediaType".
Relevant tests were added to "core/modules.rs" and integration tests,
additionally multiple WPT tests were enabled.
There are still some rough edges in the implementation and not all WPT were
enabled, due to:
a) unclear BOM handling in source code by "FileFetcher"
b) design limitation of Deno's "FileFetcher" that doesn't download the same
module multiple times in a single run
Co-authored-by: Kitson Kelly <me@kitsonkelly.com>
Currently all async ops are polled lazily, which means that op
initialization code is postponed until control is yielded to the event
loop. This has some weird consequences, e.g.
```js
let listener = Deno.listen(...);
let conn_promise = listener.accept();
listener.close();
// `BadResource` is thrown. A reasonable error would be `Interrupted`.
let conn = await conn_promise;
```
JavaScript promises are expected to be eagerly evaluated. This patch
makes ops actually do that.
