* Add prefs to limit threadpool sizes
Add preferences to control the size of threadpools,
so that we can easily reduce the amount of runtime
threads and test which pools benefit from more
threads.
Signed-off-by: Jonathan Schwender <schwenderjonathan@gmail.com>
* Add pref for Webrender threadpool
Add a preference to limit the size of the webrender threadpool.
Note: WebRender by default calls hooks which register the threads with
a profiler instance that the embedder can register with webrender.
Servo currently doesn't register such a profiler with webrender,
but in the future we might also want to profile the
webrender threadpool.
Signed-off-by: Jonathan Schwender <schwenderjonathan@gmail.com>
---------
Signed-off-by: Jonathan Schwender <schwenderjonathan@gmail.com>
* Remove unused deps
This doesn't seem to remove any deps from the workspace.
Signed-off-by: Jonathan Schwender <schwenderjonathan@gmail.com>
* ohos: Remove gaol dependency
Signed-off-by: Jonathan Schwender <schwenderjonathan@gmail.com>
---------
Signed-off-by: Jonathan Schwender <schwenderjonathan@gmail.com>
This is unfortuante, but it's the only way to stop making Stylo's
`malloc_size_of` depend on so many of Servo's dependencies. This is an
important step on the way toward releasing Stylo as standalone software.
When possible, we defer to the implementation of `MallocSizeOf` that is
in the base class.
One benefit of this change is that we start properly measure the size of
WebRender types, which before were always set to zero.
In addition the `Measurable` class is removed in favor of simply
manually implementing `MallocSizeOf`, which doesn't require
uncomfortably modifying the shape of data structures.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
* Remove the referrer policy from document and rely on its policy container
Signed-off-by: Shane Handley <shanehandley@fastmail.com>
* Make ReferrerPolicy non-optional, instead using a new enum value to represent the empty string case
Signed-off-by: Shane Handley <shanehandley@fastmail.com>
* Fix clippy issue
Signed-off-by: Shane Handley <shanehandley@fastmail.com>
* Fix usage of Option<ReferrerPolicy> in unit test
Signed-off-by: Shane Handley <shanehandley@fastmail.com>
---------
Signed-off-by: Shane Handley <shanehandley@fastmail.com>
* Implement PolicyContainer
Signed-off-by: Shane Handley <shanehandley@fastmail.com>
* implement small parts of fetch that interact with policy container
Signed-off-by: Shane Handley <shanehandley@fastmail.com>
* fix: allow policy container's csp list to be unset
Signed-off-by: Shane Handley <shanehandley@fastmail.com>
* fix: use the correct default policy when parsing from a token
Signed-off-by: Shane Handley <shanehandley@fastmail.com>
---------
Signed-off-by: Shane Handley <shanehandley@fastmail.com>
A recent refactoring (#33531) made a change that resulted in the
`Origin` header including the port even when the default port for a
scheme was used. This made the serialization different from that used
for `rust-url`'s `Origin::ascii_serialization()`, breaking CORS on some
sites. This change makes it so that the serialization is consistent
again.
This change also fixes the visiblity on a few methods in
`http_loader.rs` since visibility needs to be adjusted for testing
anyway.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
* Use ROUTER::add_typed_route where possible
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Update webxr, media and ipc-channel
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
---------
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
Instead of creating a `ROUTER` for each fetch, create a fetch thread
which handles all incoming and outcoming fetch requests. Now messages
involving fetches carry a "request id" which indicates which fetch is
being addressed by the message. This greatly reduces the number of file
descriptors used by fetch.
In addition, the interface for kicking off fetches is simplified when
using the `Listener` with `Document`s and the `GlobalScope`.
This does not fix all leaked file descriptors / mach ports, but greatly
eliminates the number used. Now tests can be run without limiting
procesess on modern macOS systems.
Followup work:
1. There are more instances where fetch is done using the old method.
Some of these require more changes in order to be converted to the
`FetchThread` approach.
2. Eliminate usage of IPC channels when doing redirects.
3. Also eliminate the IPC channel used for cancel handling.
4. This change opens up the possiblity of controlling the priority of
fetch requests.
Fixes#29834.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
* Implement sec-fetch-dest header
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Implement "is same site" algorithm
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Implement remaining sec-fetch-* headers
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Fix casing of header names
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Fix handling Destination::None in sec-fetch-dest
This also removes the comment about wanting to upgrade
to a newer content-security-protocol version because
the csp doesn't implement the "empty" case.
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Update WPT expectations
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Remove colon from spec comment
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Adjust expected default headers
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Fix test expectations
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
---------
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
Instead of exposing many different kinds of messages to the compositor
that are routed through the constellation, expose a single message type
which can be sent across IPC channels. In addition, this IPC channel and
the route to the crossbeam channel with the compositor is created along
with the `CompositorProxy`, simplifying what needs to be passed around
during pipeline initialization.
Previously, some image updates (from video) were sent over IPC with a
special serialization routine and some were sent via crossbeam channels
(canvas). Now all updates go over the IPC channel `IpcSharedMemory` is
used to avoid serialization penalties. This should improve performance
and reduce copies for video, but add a memory copy overhead for canvas.
This will improve in the future when canvas renders directly into a
texture.
All-in-all this is a simplification which opens the path toward having a
standard compositor API and reduces the number of duplicate messages and
proxying that had to happen in libservo.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
* Make checking for cookie prefixes case-insensitive
Cookie-Prefixes like "__Host-" and "__Secure-" are case insensitive
as per https://www.ietf.org/archive/id/draft-ietf-httpbis-rfc6265bis-15.html#name-storage-model.
This is tested by many WPT tests in cookies/prefix, for example
* cookies/prefix/__host.document-cookie.html
* cookies/prefix/__host.document-cookie.https.html
Since the implementation and the specification had diverged quite
significantly i also updated/added spec comments where appropriate
and slightly restructured code so its easier to follow. However,
the only change in behaviour is the prefix check described above.
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Update WPT expectations
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Remove unused import
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Fix cookie test cases
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Fix ignore cookie with __Host prefix and no specified path attribute
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Fix another cookie test case
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
---------
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Document "fetch" method
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Bring http_network_or_cache_fetch closer to the spec
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* fix test-tidy errors
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Move all code into http_loader.rs
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Don't panic if hyper/servo disagree about valid origins
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Add "otherwise" to spec comment
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Convert FIXME's to TODOs when appropriate
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Remove TODO about No-Store cache directive
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Remove indentation from multiline spec comments
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Add fetch assertions
This is from a spec update where assertions
about requests origin not being client
were added.
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Add note about serializing headers
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Add TODO about partitioning http cache
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Convert FIXME to TODO in script/
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Link to relevant issue for TODO comments
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
---------
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
Instead of exposing many different kinds of messages to the compositor
that are routed through the constellation, expose a single message type
which can be sent across IPC channels. In addition, this IPC channel and
the route to the crossbeam channel with the compositor is created along
with the `CompositorProxy`, simplifying what needs to be passed around
during pipeline initialization.
Previously, some image updates (from video) were sent over IPC with a
special serialization routine and some were sent via crossbeam channels
(canvas). Now all updates go over the IPC channel `IpcSharedMemory` is
used to avoid serialization penalties. This should improve performance
and reduce copies for video, but add a memory copy overhead for canvas.
This will improve in the future when canvas renders directly into a
texture.
All-in-all this is a simplification which opens the path toward having a
standard compositor API and reduces the number of duplicate messages and
proxying that had to happen in libservo.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
This adds a 'is_fetchable()' method on the ProtocolHandler trait that is then used in the fetch code.
The 'data:' protocol handler is updated to return true instead of hardcoding the scheme comparison, as well
as the 'urlinfo:' handler since it's just a testing one.
Signed-off-by: webbeef <me@webbeef.org>
This removes the last few uses of `time@0.1` in Servo. There are still
dependencies from `style` and `webrender`, but they will be removed soon
as well. The uses of this version of `time` are replaced with
`std::time` types and `time@0.3` when negative `Duration` is necessary.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
* Properly store link relations
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Send fetch request for prefetch links
We don't actually *do* anything with the response yet
(handle errors etc) but its a first step.
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Fire load/error events for prefetch loads
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Set prefetch destination/cors setting correctly
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Update WPT expectations
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Fix ./mach test-tidy errors
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Set correct "Accept" value for prefetch requests
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
* Add spec text to individual steps
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
---------
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
Up until now, Servo was using a very old version of time to get a
cross-process monotonic timestamp (using `time::precise_time_ns()`).
This change replaces the usage of old time with a new serializable
monotonic time called `CrossProcessInstant` and uses it where `u64`
timestamps were stored before. The standard library doesn't provide this
functionality because it isn't something you can do reliably on all
platforms. The idea is that we do our best and then fall back
gracefully.
This is a big change, because Servo was using `u64` timestamps all over
the place some as raw values taken from `time::precise_time_ns()` and
some as relative offsets from the "navigation start," which is a concept
similar to DOM's `timeOrigin` (but not exactly the same). It's very
difficult to fix this situation without fixing it everywhere as the
`Instant` concept is supposed to be opaque. The good thing is that this
change clears up all ambiguity when passing times as a `time::Duration`
is unit agnostic and a `CrossProcessInstant` represents an absolute
moment in time.
The `time` version of `Duration` is used because it can both be negative
and is also serializable.
Good things:
- No need too pass around `time` and `time_precise` any longer.
`CrossProcessInstant` is also precise and monotonic.
- The distinction between a time that is unset or at `0` (at some kind
of timer epoch) is now gone.
There still a lot of work to do to clean up timing, but this is the
first step. In general, I've tried to preserve existing behavior, even
when not spec compliant, as much as possible. I plan to submit followup
PRs fixing some of the issues I've noticed.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
`std::time` is good enough for us here. `cookie` is using `time 0.3`,
but Servo can convert to standard library types when getting data from
`cookie`. This reduces our direct dependencies and removes more use of
the very old `time 0.1` series.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
This fixes various unused code warnings after the recent rust upgrade.
Some of the dead code is maintained, as it is quite likely that it will
be used in future changes.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
This is the last step toward removing our use of `syn` version
1. It does three things:
1. Upgrades `async-recursion` to a newer version that uses `syn` 2.
2. Removes the use of `enum-iterator` that was only used to produce a
trivial list of enum names. This reduces the number of crates we
dependo on by 2.
3. Upgrades `media` to a version which no longer uses `syn` 1
Fixes#33034.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
This changes updates to the new version of the `cookie` crate in Servo
which no longer uses the old `time@0.1` data types. This requires using
a new version of `time` while we transition off of the old one. This is
the first step in that process.
In addition, the overloading of the `cookie::Cookie` name was causing a
great deal of confusion, so I've renamed the Servo wrapper to
`ServoCookie` like we do with `ServoUrl`.
Signed-off-by: Martin Robinson <mrobinson@igalia.com>
* Move WebRender related types to `webrender_traits`
This refactor moves several WebRender related types
from `compositing_traits`, `script_traits` and `net_traits`
crates to the `webrender_traits` crate.
This change also moves the `Image` type and associated
function out of `net_traits` and into the `pixels` crate.
Co-authored-by: Martin Robinson <mrobinson@igalia.com>
Signed-off-by: Mukilan Thiyagarajan <mukilan@igalia.com>
* Move `script_traits::WebrenderIpcSender` to `webrender_traits::WebRenderScriptApi`
---------
Signed-off-by: Mukilan Thiyagarajan <mukilan@igalia.com>
Co-authored-by: Martin Robinson <mrobinson@igalia.com>
