Chrome · FullScreen
CVE-2026-113214
Input Validation in FullScreen
Overview
Low
Severity
—
CVSS
No
Exploited ITW
Fixed
Fix Status
Changed Functions
| Function | Change | Notes |
|---|---|---|
ifchrome/browser/ui/exclusive_access/pointer_lock_controller.cc |
modified |
Files Changed
chrome/browser/ui/exclusive_access/pointer_lock_controller.ccchrome/browser/ui/exclusive_access/pointer_lock_controller_browsertest.cc
Patch
From 8c0e7738e64fc12dd1869b11a0e7d21bc6b5c70c Mon Sep 17 00:00:00 2001 From: Vikram Pasupathy <[email protected]> Date: Fri, 25 Sep 2026 11:38:50 -0700 Subject: [PATCH] Require user activation for Pointer Lock while in tab fullscreen PointerLockController::RequestToLockPointer() previously skipped the user-activation and post-Esc cooldown checks whenever the tab was in HTML fullscreen. Because WebContents::IsFullscreen() is tab-scoped, this exemption applied to every frame in the tab regardless of which frame obtained the user activation to enter fullscreen, allowing an embedded cross-origin iframe to silently lock the pointer without a user gesture. Drop the tab-fullscreen exemption so that a pointer-lock request must carry its own transient user activation (or come from a target that previously unlocked itself), and so that the post-Esc cooldown is honored even while the tab is fullscreen. Bug: 506148052 Change-Id: I58166868cae38274c5151a3590ddecf3acd77af8 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8422256 Reviewed-by: Frank Liberato <[email protected]> Commit-Queue: Vikram Pasupathy <[email protected]> Cr-Commit-Position: refs/heads/main@{#1704993} --- diff --git a/chrome/browser/ui/exclusive_access/pointer_lock_controller.cc b/chrome/browser/ui/exclusive_access/pointer_lock_controller.cc index 4d2687f..eebf155 100644 --- a/chrome/browser/ui/exclusive_access/pointer_lock_controller.cc +++ b/chrome/browser/ui/exclusive_access/pointer_lock_controller.cc @@ -50,12 +50,11 @@ DCHECK(!IsPointerLocked()); // To prevent misbehaving sites from constantly re-locking the pointer, the - // lock-requesting page must have transient user activation and it must not + // lock-requesting frame must have transient user activation and it must not // request for a lock within |kEffectiveUserEscapeDuration| time since the - // user successfully escaped from a previous lock. Exceptions are when the - // page has unlocked (i.e. not the user), or if we're in tab fullscreen (which - // requires its own transient user activation). - if (!last_unlocked_by_target && !web_contents->IsFullscreen()) { + // user successfully escaped from a previous lock. The only exception is when + // the target frame previously unlocked itself (i.e. not the user). + if (!last_unlocked_by_target) { if (!user_gesture) { web_contents->GotResponseToPointerLockRequest( blink::mojom::PointerLockResult::kRequiresUserGesture); diff --git a/chrome/browser/ui/exclusive_access/pointer_lock_controller_browsertest.cc b/chrome/browser/ui/exclusive_access/pointer_lock_controller_browsertest.cc index 83b82f9..44bc85d 100644 --- a/chrome/browser/ui/exclusive_access/pointer_lock_controller_browsertest.cc +++ b/chrome/browser/ui/exclusive_access/pointer_lock_controller_browsertest.cc @@ -187,3 +187,36 @@ ->pointer_lock_controller() ->IsPointerLocked()); } + +IN_PROC_BROWSER_TEST_F(PointerLockControllerTest, + PointerLockInFullscreenRequiresUserGesture) { + EnterActiveTabFullscreen(); + + // A pointer lock request without user gesture must be rejected even while + // in tab fullscreen (crbug.com/506148052). + RequestToLockPointer(/*user_gesture=*/false, + /*last_unlocked_by_target=*/false); + EXPECT_FALSE(GetExclusiveAccessManager() + ->pointer_lock_controller() + ->IsPointerLocked()); + + // With a user gesture, pointer lock is granted. + RequestToLockPointer(/*user_gesture=*/true, + /*last_unlocked_by_target=*/false); + EXPECT_TRUE(GetExclusiveAccessManager() + ->pointer_lock_controller() + ->IsPointerLocked()); + + // An immediate re-lock attempt within the post-escape cooldown period + // must be rejected even with a user gesture while in fullscreen. + SendEscapeToExclusiveAccessManager(); + EXPECT_FALSE(GetExclusiveAccessManager() + ->pointer_lock_controller() + ->IsPointerLocked()); + + RequestToLockPointer(/*user_gesture=*/true, + /*last_unlocked_by_target=*/false); + EXPECT_FALSE(GetExclusiveAccessManager() + ->pointer_lock_controller() + ->IsPointerLocked()); +}
Loading diff…
Regression Test / PoC
shipped with the fix
diff --git a/chrome/browser/ui/exclusive_access/pointer_lock_controller_browsertest.cc b/chrome/browser/ui/exclusive_access/pointer_lock_controller_browsertest.cc
index 83b82f9..44bc85d 100644
--- a/chrome/browser/ui/exclusive_access/pointer_lock_controller_browsertest.cc
+++ b/chrome/browser/ui/exclusive_access/pointer_lock_controller_browsertest.cc
@@ -187,3 +187,36 @@
->pointer_lock_controller()
->IsPointerLocked());
}
+
+IN_PROC_BROWSER_TEST_F(PointerLockControllerTest,
+ PointerLockInFullscreenRequiresUserGesture) {
+ EnterActiveTabFullscreen();
+
+ // A pointer lock request without user gesture must be rejected even while
+ // in tab fullscreen (crbug.com/506148052).
+ RequestToLockPointer(/*user_gesture=*/false,
+ /*last_unlocked_by_target=*/false);
+ EXPECT_FALSE(GetExclusiveAccessManager()
+ ->pointer_lock_controller()
+ ->IsPointerLocked());
+
+ // With a user gesture, pointer lock is granted.
+ RequestToLockPointer(/*user_gesture=*/true,
+ /*last_unlocked_by_target=*/false);
+ EXPECT_TRUE(GetExclusiveAccessManager()
+ ->pointer_lock_controller()
+ ->IsPointerLocked());
+
+ // An immediate re-lock attempt within the post-escape cooldown period
+ // must be rejected even with a user gesture while in fullscreen.
+ SendEscapeToExclusiveAccessManager();
+ EXPECT_FALSE(GetExclusiveAccessManager()
+ ->pointer_lock_controller()
+ ->IsPointerLocked());
+
+ RequestToLockPointer(/*user_gesture=*/true,
+ /*last_unlocked_by_target=*/false);
+ EXPECT_FALSE(GetExclusiveAccessManager()
+ ->pointer_lock_controller()
+ ->IsPointerLocked());
+}
Loading diff…
Original Bug Report
The reporter's bug is still restricted on the tracker. Chrome de-restricts security bugs ~30–90 days after the fix ships; a later run will backfill it here.
References
On This Page