Currently, `script` and `devtools` use a node's unique id to identify it
across requests. The unique ID is part of a node's rare data field and
is really only meant for debugging. Instantiating it on a node causes
it's memory usage to go up significantly. Now, when the devtools ask for
information about a specific `Node` then they send the unique ID to
`script`, and the script thread then walks the whole DOM tree searching
for that specific ID. This happens here:
6d0b651218/components/script/devtools.rs (L142-L153)
So, in the worst case, all of the nodes in the tree now have a unique
ID. That's not great!
Also, when `script` notifies `devtools` about changes to a DOM node then
`devtools` expects a `NodeActor` to exist for that Node. The actor might
not exist if the inspector doesn't know about that node yet - that
happens when the user hasn't expanded their parent yet.
That is an oversight from https://github.com/servo/servo/pull/42601 and
causes problems now(https://github.com/servo/servo/issues/42784) because
for the longest time, `devtools` was mostly the one sending requests. Of
course, we could make `devtools` simply ignore updates for nodes that it
doesn't know about, but ideally we shouldn't send these updates in the
first place.
This change implements a lookup map on the `ScriptThread` that contains
all nodes that have been sent to the devtools inspector. The map allows
us to efficiently resolve a unique ID to a `Node` in O(1), without
creating unique IDs for the whole tree. It also allows us to only send
DOM updates for nodes that the inspector cares about.
For now, entries from the cache are not evicted unless the relevant
pipeline is closed. That reflects reality, because the inspector also
keeps using them forever.
In the future we will tell the inspector when nodes are removed from the
tree - then it can't interact with them anymore, and we can remove them
from the script-side map.
This is change is not all that complicated but it involves moving a lot
of code around, so feel free to ask for clarification when something is
unclear!
Testing: This change adds a test
Fixes part of https://github.com/servo/servo/issues/42784
---------
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
I only wanted to get `&mut JSContext` in microtask chunk and checkpoint,
but this in turn needed `&mut JSContext` in servoparser, which then
caused need for even more changes in script.
I tried to limit the size by putting some `temp_cx` in:
- drops of `LoadBlocker`, `GenericAutoEntryScript`
- methods of `VirtualMethods`
- methods of `FetchResponseListener`
Testing: Just refactor, but should be covered by WPT tests.
Part of #40600
---------
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
`process_response_eof` is the only method that needs cx at least right
now. This PR removes one temp_cx and introduces one, removing that one
will is hard (needs VirtualMethods and a lot of work)
Testing: Just refactor
Part of #40600
---------
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
This is a first step to compile scripts using `&mut JSContext`.
Testing: Refactor, covered by existing WPT tests.
Part of #40600
---------
Signed-off-by: Gae24 <96017547+Gae24@users.noreply.github.com>
Continuation of https://github.com/servo/servo/pull/42135, switch
Error::Type and Error::Range to also use CStrings internally, as they
are converted to CString for throwing JS exceptions (other get thrown as
DomException object, which uses rust string internally).
Changes in script crate are mechanical.
Testing: Should be covered by WPT tests.
Part of #42126
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
We implemented many more MallocSizeOf tests (even if some such as
channels are zero).
Meaning we can not ignore more of it making the code cleaner.
Signed-off-by: Narfinger <Narfinger@users.noreply.github.com>
Testing: Compilation is the test.
Signed-off-by: Narfinger <Narfinger@users.noreply.github.com>
Switch `PostMessage` of `Worker`, `DedicatedWorkerGlobalScope` and
`ServiceWoker` to use `&mut JSContext`, propagating it to
`post_message_impl`.
Testing: A successful build is enough
Signed-off-by: Gae24 <96017547+Gae24@users.noreply.github.com>
When interrupting `DedicatedWorkerGlobalScope` execution, handle the
case that the `DebuggerGlobalScope` is running its script by just
terminating. It seems that this happens sometimes when trusted types
terminates a worker script.
Testing: This should fix some flaky `expect` failure panics in trusted
types WPT tests.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
Reviewable per commit. First commit adds `&mut JSContext` to
`call_default_constructor`. Second commit adds `&mut JSContext` to
Interface, which in unfortunately required many changes (a lot of
passing `&mut JSContext` down).
Testing: Just a refactor, but should be covered by WPT
Part of #40600
---------
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
This change is reviewable per commits:
In first commit we added `&mut JSContext` to `run_box` (it is very hard
to bring `&mut JSContext` to `remove_script_and_layout_blocker`).
In second commit we pass `&mut JSContext` to `run_once`.
In third commit we added support for accepting `&mut JSContext` in
closures of `task!` macro and lastly we demo new macro invocations (to
ensure they actually compile)
Testing: Just refactor, but should be covered by WPT
Part of #40600
---------
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
Rather than having each callside specifying the relevant
information from the GlobalScope, do this via a trait instead.
This would have saved us quite a bit of test debugging
since we would often forget to set relevant information
from the global context for a request.
Now, in the future when we need additional information from
the globalscope for a request, we only need to update this
method to make that happen.
Previously it would also sometimes use `document`, but
calling the relevant information on either `document` or
`globalscope` doesn't matter, since the `globalscope`
defers to the value from the `document` anyways.
Signed-off-by: Tim van der Lippe <tvanderlippe@gmail.com>
This implements GenericReceiverSet similar to IpcReceiverSet. This
allows us to wait on a group of channels.
IpcReceiverSet was allowed to use IpcReceivers of different type, i.e.,
`IpcReceiver<Foo>` and `IpcReceiver<Bar>` in the same select query. This
changes with GenericReceiverSet to only allow one type, i.e.,
`GenericReceiver<Foo>`. As this functionality was only used in the
CoreResourceThread, we changed the setup slightly for the memory
reporter.
With this we also change the implementation of CoreResourceThread to now
use the GenericReceiverSet.
Signed-off-by: Narfinger <Narfinger@users.noreply.github.com>
Testing: New testcases were added to GenericReceiverSet and browsing
works normally.
---------
Signed-off-by: Narfinger <Narfinger@users.noreply.github.com>
Signed-off-by: Josh Matthews <josh@joshmatthews.net>
Co-authored-by: Josh Matthews <josh@joshmatthews.net>
1. removed `ScriptThread::get_safe_cx` in favor of `Runtime::cx` (this
intentionally shadows
[RustRuntime::cx](https://doc.servo.org/mozjs/rust/struct.Runtime.html#method.cx),
which is not really safe with all other stuff that happen in servo).
2. passed down cx from `ScriptThread::new` to accommodate those changes
(we also need to pass cx for some other stuff).
3. we create JSContext in workers/worklets, so we can pass all the way
down to
6ccc47f4c9/components/script/dom/workers/workerglobalscope.rs (L1013)
(so we can get &mut JSContext in tasks).
So with this PR, we prepared all entry points of JSContext (or other
should actually be reentering via SM callback or passed down from the
root one in the thread).
Testing: No changes just safety, but it should be covered by WPT
Part of #40600
---------
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
There's no naming conflict, and nowhere else the name NetRequestInit.
All it does is increase confusion.
Testing: No runtime changes.
---------
Signed-off-by: Josh Matthews <josh@joshmatthews.net>
Replace `allow` with `expect` lints for `unused`, `unsafe_code`,
`dead_code`, and `non_upper_case_globals`.
Testing: So far just check it compiled on `x86_64-linux` on NixOS. Need
to use the module `system.fontconfig.enable = true;` I think in my NixOS
config.
Part of: #40383
Searching `allow\(.*\)` for `.rs` files shows the following. for
(total_results:total_files) went from `707:386` to `675:368`, a
reduction of `32:18`.
How many files is too many files per PR? I feel like the 20-30 I have is
too big.
---------
Signed-off-by: Wayne Van Son <waynevanson@gmail.com>
Instead of the creation URL of the containing global,
it should instead use the URL of the current worker.
Despite the referrer specification stating that we
should use the creation URL, instead browsers use
the current URL. A new WPT test is added to cover that
which we currently fail because of incorrect
serialization of query parameters of a URL. But the
actual redirect and all work now.
Signed-off-by: Tim van der Lippe <tvanderlippe@gmail.com>
This PR adds support for using new JSContext/CurrentRealm to promise and
adds some more helpers that will ease migration to new context model.
Second commit has another demo.
Testing: Covered by WPT
Part of #40600
---------
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
Use the algorithms introduced in #41109 in more places.
Follow more closely the spec when executing classic scripts, introducing
the concepts of rethrowing and muting errors.
Muting errors is not actually implemented, and will be done in a
followup.
Testing: More tests start passing
Fixes#34199Fixes#27260Fixes#15188
---------
Signed-off-by: Gae24 <96017547+Gae24@users.noreply.github.com>
Switch the devtools part to GenericCallback and GenericSender.
To keep the diff small the names where not changes as a Sender almost
fills the same requirement as a callback.
Testing: As this is mostly type changes, the compilation is the test but
also devtools seem to work fine with these changes. ./mach try run is
here: https://github.com/Narfinger/servo/actions/runs/19931697694
---------
Signed-off-by: Narfinger <Narfinger@users.noreply.github.com>
This PR introduces a new storage coordination thread, intended to serve
as the central point for managing all current and future storage
endpoints in Servo.
In addition to the new coordination thread, this PR also lays the
infrastructure required to develop a parallel, next-generation IndexedDB
implementation under the indexeddb_next feature flag living on a
separate branch.
Testing: Unit and WPT tests continue to pass
---------
Signed-off-by: Jan Varga <jvarga@igalia.com>
`ScriptThread`s start with an initial `Pipeline`, but that `Pipeline`
should not be used for any `ScriptThread` global data as a
`ScriptThread` can have any number of `Pipeline`s that come and go.
There's no architectural reason why a `ScriptThread` should be
associated with only a single `WebView`.
This change makes it so that the `ScriptThread` no longer uses the
`PipelineId` and the `WebViewId` of the initial `Pipeline` to initialize
itself. Instead a `ScriptEventLoopId` is used to uniquely identify every
`ScriptThread` in both the `ScriptThread` and the `Constellation`.
The remaining use was for crash reporting. Now when a crash happens, it
launches the sad tab in all `WebView`s that depeneded on that
`ScriptThread`.
This is a change which should allow simplifying the way that
`EventLoop`s/`ScriptThread`s are started in the `Constellation`, which
will be handled in a followup.
Testing: This shouldn't change behavior except in the case where a
`ScriptThread` is used in more than a single `WebView`, which should
work
properly now. I tested this change manually by running servoshell in
multiprocess mode while causing a panic to happen due to a null pointer
dereference. The sad tab appeared after this change, so things seem to
work
properly still. Testing crashes is tricky with the way we test servo
now.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
Switches to log methods over eprintln in cases where it makes sense.
stdout / stderr output may be difficult to view on devices like android
/ ohos, and has no filter options, so we should prefer log methods.
This PR avoids prints, which are controlled by a debugging option.
Testing: Only changes logging, no functional behavior change.
Signed-off-by: Jonathan Schwender <schwenderjonathan@gmail.com>
I used find and replace to finish the job. All this PR does is replace
all `Error::<error_name>` occurrences with `Error::<error_name>(None)`.
Testing: Refactor
Fixes: #39053
Signed-off-by: Ashwin Naren <arihant2math@gmail.com>
This web API is alternative API to `throw e`, which is why we can reuse
a lot of the existing machinery.
The one testcase that isn't passing yet is because it reports an empty
`TypeError`. The current logic in `ErrorInfo` only retrieves the message
data, but doesn't include the type of the exception. For that, we need
to use `(*report)._base.errorNumber` and map that back to the original
type codes. However, deferring that to a follow-up as that requires some
more work in mozjs.
Signed-off-by: Tim van der Lippe <tvanderlippe@gmail.com>
Companion to https://github.com/servo/mozjs/pull/650
We added 3 new options to bindings.conf, each more powerful then the
previous one, so one should use the least powerful as possible to keep
things flexible:
1 `cx_no_gc` prepends argument `&JSContext`, which allows creating NoGC
tokens and using functions that do not trigger GC.
2. `cx` prepends argument `&mut JSContext`, which allows everything that
previous one allows, but it also allows calling GC triggering functions.
3. `realm` prepends argument `&mut CurrentRealm`, which can be deref_mut
to `&mut JSContext` (so it can do everything that previous can), but it
also ensures that there is current entered realm, which can be used for
creation of InRealm.
next steps: #40600
reviewable per commit
Testing: It's just refactoring
try run: https://github.com/sagudev/servo/actions/runs/19287700927
---------
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
Instead of using a task source for SpiderMonkey runtime callbacks, use a
`ScriptEventLoopSender`. Task sources are associated with a particular
Pipeline, but the runtime callback is run indepenently of any particular
Pipeline and could theoretically happen when no Pipeline exists at all.
This reduces the dependency of the `ScriptThread` on the existence of
the first Pipeline.
Testing: This should not change observable behavior, so is covered by
existing tests.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
The goal of this change is to prevent having to copy so much data out of
listeners when a fetch completes, which will be particularly important
for off-the-main thread parsing of CSS (see #22478). This change has
pros and cons:
Pros:
- This makes the design of the `FetchResponseListener` a great deal
simpler.
They no longer individually store a dummy `ResourceFetchTiming` that is
only replaced right before `process_response_eof`.
- The creation of the `Arc<Mutex<FetchResponseListener>>` in the
`NetworkListener` is abstracted away from clients and now they just
pass the `FetchResponseListener` to the fetch methods in the global.
Cons:
- Now each `FetchResponseListener` must explicitly call `submit_timing`
instead of having the `NetworkListener` do it. This is arguably a bit
easier to follow in the code.
- Since the internal data of the `NetworkListener` is now an
`Arc<Mutex<Option<FetchResponseListener>>>`, when the fetching code
needs to share state with the `NetworkListener` it either needs to
share an `Option` or some sort of internal state. In one case I've
stored the `Option` and in another case, I've stored a new inner
shared value.
Testing: This should not change observable behavior and is thus covered
by existing tests.
Fixes: #22550
---------
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
This is last step toward enabling the default rustc
`unsafe_op_in_unsafe_fn` warning for the script crate. It wraps the
remaining unsafe code in `unsafe {}` and removes the line disabling this
warning from `script`'s `Cargo.toml`. In addition, two variables are
renamed from `v` to something slightly more descriptive.
Testing: This should not change behavior so is covered by existing
tests.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
`FetchReponseListener` has traditionally lived in `net` even though it
is only used in `script` currently. Because of the two way dependency,
it has also use a lot of templating to implement something pretty basic
(call methods on a trait object).
This change moves the trait to `script` and removes several levels of
templating, making the code quite a bit shorter and easier to
understand.
This change is preparation for fixing #22550 and implementing
off-the-main-thread CSS parsing.
Testing: This should not change any behavior so is covered by existing
tests.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
Removes some unneeded lints, especially `#[allow(unsafe_code)]`.
Testing: Refactor
Part of: #40383
Signed-off-by: WaterWhisperer <waterwhisperer24@qq.com>
Remove the microtask queue from `GlobalScope`. The queue is moved inside
worker global scopes, while for window globals the script thread's queue
is accessed.
Testing: Covered by existing tests
Fixes: #20908
Signed-off-by: Gae24 <96017547+Gae24@users.noreply.github.com>
Servo has a lot of comments like this:
```rust
// https://example-spec.com/#do-the-thing
fn do_the_thing() {}
```
and I keep turning these into doc comments whenever I'm working close to
one of them. Doing so allows me to hover over a function call in an IDE
and open its specification without having to jump to the function
definition first. This change fixes all of these comments at once.
This was done using `find components -name '*.rs' -exec perl -i -0777
-pe 's|^([ \t]*)// (https?://.*)\n\1(fn )|\1/// <$2>\n\1$3|mg' {} +`.
Note that these comments should be doc comments even within trait `impl`
blocks, because rustdoc will use them as fallback documentation when the
method definition on the trait does not have documentation.
Testing: Comments only, no testing required
Preparation for https://github.com/servo/servo/pull/39552
---------
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
Moves interfaces defined by the performance spec to the
`script/dom/performance/` module from `script/dom/`.
Testing: Just a refactor shouldn't need any testing
Fixes: Partially #38901
Signed-off-by: WaterWhisperer <waterwhisperer24@qq.com>
rename `WorkerGlobalScope.get_worker_id` to
`WorkerGlobalScope.worker_id`
Testing: This is a simple rename. No testing is necessary, since
compiler will complain if something is wrong.
Context: In
https://github.com/servo/servo/pull/40156#discussion_r2462627713 I
received feedback to use Rust getter name
[convention](https://rust-lang.github.io/api-guidelines/naming.html#getter-names-follow-rust-convention-c-getter)
and avoid adding `get_` methods. I initially chose `get_worker_name()`
to match existing `get_worker_id()` below it. Overall I think it does
make sense to rename `get_worker_id()` to `worker_id()`.
`WorkerGlobalScope` has 2 more methods that start with `get_`:
`get_cx(&self) -> JSContext` and `get_url`. `get_url` has a
corresponding `set_url` as well in the same file. I am not sure if folks
prefer to have those renamed as well or not, so I left those out of this
PR, and instead started a discussion in a separate
[issue](https://github.com/servo/servo/issues/40192) to figure out a
consistent policy on Rust getter name convention in Servo codebase.
Signed-off-by: Yerkebulan Tulibergenov <yerkebulan@gmail.com>
- Remove a couple of ignore_malloc_size_of since are not required
anymore
- Avoid calling random and use the worker_id
- Pass DedicatedWorker event_loop_sender to run_with_memory_profiling
Testing: Covered by tests, only code related to memory reporting was
changed.
Fixes: #11855
---------
Signed-off-by: Gae24 <96017547+Gae24@users.noreply.github.com>
