Vulkan HW video decoding fails in the RDD process unless MOZ_DISABLE_RDD_SANDBOX=1 is set
Categories
(Core :: Security: Process Sandboxing, defect)
Tracking
()
People
(Reporter: tboiko, Assigned: tboiko)
References
(Blocks 1 open bug)
Details
Attachments
(1 file)
Steps to reproduce:
- Enable Vulkan hw video decode (media.hardware-video-decoding-vulkan.enabled = true).
- Run Firefox without MOZ_DISABLE_RDD_SANDBOX.
Open a video site (e.g. YouTube). - Repeat with MOZ_DISABLE_RDD_SANDBOX=1.
Actual results:
With the RDD sandbox enabled, Vulkan hw video decode fails or is unreliable (fallbacks to CPU). With MOZ_DISABLE_RDD_SANDBOX=1, the same build works.
Expected results:
Vulkan hw video decode should work with the RDD sandbox enabled.
Comment 1•2 months ago
|
||
The Bugbug bot thinks this bug should belong to the 'Core::Audio/Video: Playback' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.
| Assignee | ||
Comment 2•2 months ago
|
||
Enable brokered socket/connect syscalls in the RDD sandbox and add narrow
MAY_CONNECT rules for Wayland and X11 display sockets so EGL can connect
for EGL_MESA_image_dma_buf_export during Vulkan video decode.
On XWayland, both the Wayland socket and the X11 socket (/tmp/.X11-unix/X*)
are now allowed since EGL may connect via either path.
Updated•2 months ago
|
| Assignee | ||
Comment 3•2 months ago
|
||
Tested on X11 — works. Wayland needs double-checking before I can confirm that path. Will follow up once tested.
Comment 4•1 month ago
|
||
The severity field is not set for this bug.
:jimm, could you have a look please?
For more information, please visit BugBot documentation.
Updated•1 month ago
|
| Assignee | ||
Comment 5•1 month ago
|
||
Martin, can you please check if this patch works on your setup? (I assume it's on Wayland, right?)
Updated the patch
Comment 6•1 month ago
|
||
Tested on AMD RX 6600 and mesa-26.1.3 / Fedora 44 and vulkan decode works for me with and without the patch (just media.hardware-video-decoding-vulkan.enabled = true). I can test also on Intel if needed.
Updated•1 month ago
|
Comment 7•1 month ago
|
||
Also yes, testing on Wayland.
| Assignee | ||
Comment 8•1 month ago
|
||
Just a reminder, please approve as we would like to release this in build 153 (by July 21st)
Comment 9•29 days ago
|
||
(In reply to Tymur Boiko from comment #8)
Just a reminder, please approve as we would like to release this in build 153 (by July 21st)
May be worth to ping people at Mozilla chat directly - (for instance gfx channel is at https://chat.mozilla.org/#/room/#gfx:mozilla.org).
I'll try to do today (there's a time zone shift for me).
| Assignee | ||
Comment 10•29 days ago
|
||
Thank you. Also, rebased to the latest main
Comment 11•23 days ago
|
||
Comment 12•23 days ago
|
||
Comment 13•23 days ago
|
||
Backed out for causing failures at browser_sandbox_test.js.
Backout link: https://hg.mozilla.org/integration/autoland/rev/a168b0e43d98
| Assignee | ||
Comment 14•23 days ago
|
||
Reproduced locally and fixed at my side. Updated the patch
Comment 15•22 days ago
|
||
Comment 16•22 days ago
|
||
| bugherder | ||
Updated•22 days ago
|
Comment 17•22 days ago
|
||
The patch landed in nightly and beta is affected.
:tboiko, is this bug important enough to require an uplift?
- If yes, please nominate the patch for beta approval.
- See https://wiki.mozilla.org/Release_Management/Requesting_an_Uplift for documentation on how to request an uplift.
- If no, please set
status-firefox154towontfix.
For more information, please visit BugBot documentation.
| Assignee | ||
Comment 18•1 day ago
|
||
Hi Martin,
Could you please help me with the above status? (I can be wrong)
Comment 19•1 day ago
|
||
Sure, we should keep it as is (wontfix for 154.0) as it becomes release next week.
Description
•