* build: enable V8 builtins PGO
Removes the gn arg that disabled V8 builtins profile-guided optimization
and adds a V8 patch to warn instead of abort when the builtin PGO profile
data does not match. Also strips the PGO-related flags from the generated
mksnapshot_args so they are not passed through to downstream mksnapshot
invocations.
Co-authored-by: Sam Attard <sattard@anthropic.com>
* docs: clarify Node.js async_hooks as reason for promise_hooks flag
Addresses review feedback: the v8_enable_javascript_promise_hooks flag
is set to support Node.js async_hooks, not used directly by Electron.
Co-authored-by: Sam Attard <sattard@anthropic.com>
* chore: update patches
---------
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Sam Attard <sattard@anthropic.com>
Co-authored-by: PatchUp <73610968+patchup[bot]@users.noreply.github.com>
The `OnTraceBufferUsageAvailable` callback creates V8 handles via
`Dictionary::CreateEmpty()` before `promise.Resolve()` enters its
`SettleScope` (which provides a `HandleScope`). When the callback
fires asynchronously from a Mojo response (i.e. when a trace session
is active), there is no `HandleScope` on the stack, causing a fatal
V8 error: "Cannot create a handle without a HandleScope".
Add an explicit `v8::HandleScope` at the top of the callback, matching
the pattern used by the other contentTracing APIs which resolve their
promises through `SettleScope` or the static `ResolvePromise` helper.
Made-with: Cursor
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Alexey Kozy <alexey@anysphere.co>
* remove comment based label removal
* ci: add functionality for programmatic add/remove needs-signed-commits label
* add new line to pull-request-opened-synchronized
* build: add patch conflict resolution workflow with CI artifacts (#50235)
ci: upload patch conflict fix as artifact in apply-patches
When patch-up.js cannot auto-push the 3-way-merged patch diff (e.g. on
fork PRs), the checkout action already writes patches/update-patches.patch
and tells the user to check CI artifacts — but nothing was uploading it.
This adds the missing upload-artifact step to the apply-patches job so
the resolved diff is available for download, and documents in CLAUDE.md
that pulling this artifact and applying it with `git am` is the fast
path for fixing patch conflicts on PR branches without a full local sync.
Co-authored-by: Claude <noreply@anthropic.com>
(cherry picked from commit 816e5964fb)
* build: skip archiving patch conflict fix artifact (#50251)
The update-patches artifact is a single .patch file, so zipping it
is unnecessary overhead. With archive: false, gh run download fetches
the raw file directly without requiring a decompression step.
Co-authored-by: Claude <noreply@anthropic.com>
(cherry picked from commit f4a50a8fde)
---------
Co-authored-by: Claude <noreply@anthropic.com>
fix: out-of-bounds read in diff rulesets.
When merging diff rulesets, if Add() failed (due to a deliberate hash
collision, causing RobinHoodMap to refuse the insertion), we would
call NewlyAddedFromDifferentRuleSet() twice on the same RuleData,
causing us to potentially read data past the end of the Bloom filter
backing.
In addition to actually fixing the issue, we mark Add() as [[nodiscard]]
so that it cannot happen again, and we also spanify
MovedToDifferentRuleSet() so that a similar error would cause a CHECK
failure instead of reading out-of-bounds.
[arabic] Cap stch expansion per run (#5823)
Cap each stch run to at most 256 output glyphs.
This keeps pathological stretch runs from expanding to unbounded
sizes, and switches the repeat-count math to 64-bit intermediates so
the cap is applied before 32-bit arithmetic can wrap.
The existing checked accumulation and buffer growth logic stays in
place, covering both the per-run overflow and multi-run accumulation
cases reported in the recent stch advisories.
Tested: meson test -C build --suite shape
Assisted-by: OpenAI Codex
[arabic] Improve stch measurement pass (#5808)
Use checked arithmetic when calculating the number of extra glyphs
needed during stch processing. Includes a new hb_unsigned_add_overflows
helper in hb-algs.hh.
Co-authored-by: Codex (AI assistant)
Co-authored-by: Gemini (AI assistant)
[arabic] Change a couple enum values
No semantic change.
* fix: hex-encode Windows notification icon temp filenames
NotificationPresenterWin was using SHA1HashString(origin.spec()) directly
as the basename for the temporary PNG written for toast icons.
SHA1HashString returns raw digest bytes, so the generated filename could
contain invalid path characters on Windows. That caused WriteFile to fail
when saving notification icons, which left toast XML without the expected
icon path.
Hex-encode the digest before appending .png so the temporary filename is
filesystem-safe while keeping deterministic naming for a given origin.
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
* Update shell/browser/notifications/win/notification_presenter_win.cc
Co-authored-by: Robo <hop2deep@gmail.com>
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
---------
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
GetDefaultPrinterDPI() creates a blank GtkPrintSettings and reads
its resolution, which returns 0 for uninitialized settings. With
DPI=0, SetPrintableAreaIfValid() computes a zero scale factor,
producing empty page dimensions that fail PrintMsgPrintParamsIsValid().
Fall back to kDefaultPdfDpi (72) when GTK returns 0, matching the
existing Windows fallback pattern when CreateDC fails.
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
* fix: fire AXMenuOpened event when a visible ARIA menu instance is added to the DOM
Co-authored-by: Keeley Hammond <khammond@slack-corp.com>
* fix: remove redundent FireMenuPopupEndForDeletedMenus
MENU_POPUP_END for deleted menus is already handled by
AXTreeManager::OnNodeWillBeDeleted, which
fires the event directly on the menu node before destruction.
Co-authored-by: Keeley Hammond <khammond@slack-corp.com>
* chore: add feature flag (kDynamicMenuPopupEvents)
Co-authored-by: Keeley Hammond <vertedinde@electronjs.org>
* chore: update patches
Co-authored-by: Keeley Hammond <vertedinde@electronjs.org>
* chore: update patches
---------
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Keeley Hammond <khammond@slack-corp.com>
Co-authored-by: Keeley Hammond <vertedinde@electronjs.org>
The frameNamesToWindow map was a holdover from the BrowserWindowProxy
IPC shim. Since nativeWindowOpen became the only code path, Blink's
FrameTree::FindOrCreateFrameForNavigation resolves named window targets
directly in the renderer, scoped to the opener's browsing context
group. When a matching named window exists, Blink navigates it without
ever sending a CreateNewWindow IPC to the browser, so this map was
never consulted in the legitimate same-opener case.
The only time the map found a match was when two unrelated renderers
happened to use the same target name, in which case openGuestWindow
would short-circuit before consuming the guest WebContents that
Chromium had already created for the new window, leaking it.
Adds a test verifying Blink handles same-opener named-target reuse
end-to-end without any browser-side tracking.
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Sam Attard <sattard@anthropic.com>
The weak persistent tracking the OffscreenReleaseHolderMonitor was tied
to the texture object, but the release() closure holds a raw pointer to
the monitor via its v8::External data. If JS retained texture.release
while dropping the texture itself, the monitor would be freed on GC and
a later release() call would crash.
Track the release function instead of the texture object. Since the
texture holds release as a property, this keeps the monitor alive as
long as either is reachable.
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Sam Attard <sattard@anthropic.com>
gfx::PNGCodec::Decode() returns a null SkBitmap when it fails to decode
the clipboard contents as a PNG. Passing that null bitmap to
gfx::Image::CreateFrom1xBitmap() triggers a crash.
Return an empty gfx::Image instead, matching the existing null-check
pattern in skia_util.cc.
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Sam Attard <sattard@anthropic.com>
* chore: cherry-pick fbfb27470bf6 from chromium
* chore: update patches
* chore: cherry-pick bf6dd974238b from angle (#50437)
* fix: remove duplicate MaxGeometryUniformBlocks from angle cherry-pick patch
The angle cherry-pick added MaxGeometryUniformBlocks in new locations,
but it already existed in the EXT_geometry_shader section on 40-x-y,
causing a duplicate struct member build error in ShaderLang.h.
Re-exported the patch from the angle repo to correctly generate context
lines instead of additions for fields that already exist in the base.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: guard validatePerStageMaxUniformBlocks with ANGLE_SH_VERSION check
The angle cherry-pick patch that adds this field to ShCompileOptions
may not be applied in CI due to the angle repo not being found during
patch application. Guard the usage with a version check so compilation
succeeds regardless of angle patch application order.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Revert "fix: guard validatePerStageMaxUniformBlocks with ANGLE_SH_VERSION check"
This reverts commit 9ebf565988.
* fix: add angle to patches config and re-export angle patches for M144
The angle entry was missing from patches/config.json on 40-x-y, which
meant angle patches were never applied during gclient sync or CI builds.
This caused the chromium cherry-pick to fail with "no member named
'validatePerStageMaxUniformBlocks' in 'ShCompileOptions'".
Also re-exports the angle patches against the M144 base, resolving
conflicts from the M142-targeted patches.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* chore: Update patches
---------
Co-authored-by: PatchUp <73610968+patchup[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Previously the renderer checked a process-wide command-line switch to
decide whether to create a Node.js environment for dedicated workers.
When a renderer process hosted multiple WebContents with different
nodeIntegrationInWorker values (e.g. via window.open with overridden
webPreferences in setWindowOpenHandler), all workers in the process
used whichever value the first WebContents set on the command line.
Instead, plumb the flag through blink's WorkerSettings at worker
creation time, copying it from the initiating frame's WebPreferences.
The check on the worker thread then reads the per-worker value. Nested
workers inherit the flag from their parent worker via
WorkerSettings::Copy.
The --node-integration-in-worker command-line switch is removed as it
is no longer consumed.
Co-authored-by: Samuel Attard <sam@electronjs.org>
* fix: correct utility process exit code on Windows
On Windows, process exit codes are 32-bit unsigned integers (DWORD).
When passed from Chromium to Electron as a signed int and then
implicitly converted to uint64_t, values with the high bit set
(e.g., NTSTATUS codes) undergo sign extension, producing incorrect
values.
Cast the exit code to uint32_t before widening to uint64_t to
prevent sign extension and preserve the original Windows exit code.
Fixes#49455
Co-authored-by: João Silva <joaomrsilva@tecnico.ulisboa.pt>
* fix: narrow HandleTermination and Shutdown to uint32_t, add tests
Co-authored-by: João Silva <joaomrsilva@tecnico.ulisboa.pt>
---------
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: João Silva <joaomrsilva@tecnico.ulisboa.pt>
fix: correctly set IsActive() in BaseWindow on MacOS
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Kyle Cutler <kycutler@microsoft.com>
Avoids a freeze when failing to enter fullscreen on macOS.
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: John Beutner <beutner.john@gmail.com>
Chromium added a top-level package.json in CL:7485999 that sets
the type to module and breaks commonjs tests run via
node-spec-runner.js. This commit temporarily changes the type to
commonjs while running the tests, then changes it back to module when done.
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
Upstream DevTools' HostRuntime checks `IS_NODE` before `IS_BROWSER` when
selecting the platform runtime. In Electron, `process` is available in
renderer processes, so `IS_NODE` evaluates to `true` in the DevTools
context. This causes DevTools to dynamically import the Node.js platform
runtime, which uses `node:worker_threads`. DevTools Web Workers running
under the `devtools://` protocol cannot load Node.js built-in modules,
so the import fails and breaks features like the formatter worker.
Fix by swapping the check order to prefer `IS_BROWSER` when both are
true. This is safe because in pure Node.js environments (the only case
where the node runtime is needed), `window` and `self` are both
undefined, so `IS_BROWSER` is always `false` regardless of check order.
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
* fix: preserve staged update dir when pruning orphaned update dirs on macOS
The previous squirrel.mac patch cleaned up all staged update directories
before starting a new download. This kept disk usage bounded but broke
quitAndInstall() if called while a subsequent checkForUpdates() was in
flight — the already-staged bundle would be deleted out from under it.
This reworks the patch to read ShipItState.plist and preserve the
directory it references, deleting only truly orphaned update.XXXXXXX
directories. Disk footprint stays bounded (at most 2 dirs: staged +
in-progress) and quitAndInstall() remains safe mid-check.
Also adds test coverage for the quitAndInstall/checkForUpdates race and
a triple-stack scenario where 3 updates arrive without a restart.
Refs https://github.com/electron/electron/issues/50200
Co-authored-by: Samuel Attard <sattard@anthropic.com>
* chore: update patches
---------
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Samuel Attard <sattard@anthropic.com>
Co-authored-by: Keeley Hammond <vertedinde@electronjs.org>
* fix: prevent traffic light buttons flashing on deminiaturize
When a window with a custom `trafficLightPosition` is minimized and
restored, macOS re-layouts the title bar container during the
deminiaturize animation, causing the traffic light buttons to briefly
appear at their default position before being repositioned.
Fix this by hiding the buttons container in `windowWillMiniaturize` and
restoring them (with a redraw to the correct position) in
`windowDidDeminiaturize`.
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
* chore: address feedback from review
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
---------
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
- POSIX: validate StringToSizeT result and token count when splitting
the socket message into argv and additionalData; previously a
malformed message could produce incorrect slicing.
- Windows: base64-encode additionalData before embedding in the
null-delimited wchar_t buffer. The prior reinterpret_cast approach
dropped everything after the first aligned 0x0000 in the serialized
payload, so complex objects could arrive truncated.
Manually backported from #50119
fix: validate protocol scheme names in setAsDefaultProtocolClient
On Windows, `app.setAsDefaultProtocolClient(protocol)` directly
concatenates the protocol string into the registry key path with no
validation. A protocol name containing `\` could write to an arbitrary
subkey under `HKCU\Software\Classes\`, potentially hijacking existing
protocol handlers.
To fix this, add `Browser::IsValidProtocolScheme()` which validates that a protocol
name conforms to the RFC 3986 scheme grammar:
scheme = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." )
This rejects backslashes, forward slashes, whitespace, and any other
characters not permitted in URI schemes.
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
* fix: use requesting frame origin instead of top-level URL for permissions
`WebContentsPermissionHelper::RequestPermission` passes
`web_contents_->GetLastCommittedURL()` as the origin to the permission
manager instead of the actual requesting frame's origin. This enables
origin confusion when granting permissions to embedded third-party iframes,
since app permission handlers see the top-level origin instead of the
iframe's. The same pattern exists in the HID, USB, and Serial device
choosers, where grants are keyed to the primary main frame's origin rather
than the requesting frame's.
Fix this by using `requesting_frame->GetLastCommittedOrigin()` in all
affected code paths, renaming `details.requestingUrl` to
`details.requestingOrigin`, and populating it with the serialized
origin only.
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
* chore: keep requestingUrl name in permission handler details
The previous commit changed the details.requestingUrl field to
details.requestingOrigin in permission request/check handlers. That
field was already populated from the requesting frame's RFH, so the
rename was unnecessary and would break apps that read the existing
property. Revert to requestingUrl to preserve the existing API shape.
The functional changes to use the requesting frame in
WebContentsPermissionHelper and the HID/USB/Serial choosers remain.
Co-authored-by: Samuel Attard <sattard@anthropic.com>
---------
Co-authored-by: trop[bot] <37223003+trop[bot]@users.noreply.github.com>
Co-authored-by: Shelley Vohr <shelley.vohr@gmail.com>
Co-authored-by: Samuel Attard <sattard@anthropic.com>
