Fix divide by zero errors in DML2
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
HWE Next |
New
|
Undecided
|
Unassigned | ||
linux (Ubuntu) |
Invalid
|
High
|
Unassigned | ||
Noble |
Invalid
|
Undecided
|
Unassigned | ||
Plucky |
In Progress
|
Undecided
|
AceLan Kao | ||
linux-oem-6.11 (Ubuntu) |
Invalid
|
High
|
Edoardo Canepa | ||
Noble |
Fix Committed
|
Undecided
|
AceLan Kao | ||
Plucky |
Invalid
|
Undecided
|
Unassigned |
Bug Description
[Impact]
Random crashes have been reported on GPUs that use DML2 (Ryzen AI APUs and Radeon 9000 dGPU). The crashes are root caused to corruption in registers used for floating point code.
[Fix]
They are caused by a lack of FP context protection around that code. They are fixed by these 3 commits:
https:/
https:/
https:/
[Test]
1. Install ubuntu with 6.11 kernel on AMD platform
2. Run reboot test around 300 times
3. The system should not hang while booting up
[Where problems could occur]
The 3 commits are straightforward for the protection, and didn't involve any functional changes.
tags: | added: originate-from-2106922 |
Changed in linux-oem-6.11 (Ubuntu): | |
importance: | Undecided → High |
assignee: | nobody → Edoardo Canepa (ecanepa) |
Changed in linux-oem-6.11 (Ubuntu): | |
status: | New → Triaged |
Changed in linux-oem-6.11 (Ubuntu): | |
status: | Triaged → Confirmed |
Changed in linux (Ubuntu): | |
status: | New → Confirmed |
importance: | Undecided → High |
Changed in linux (Ubuntu Noble): | |
status: | New → Invalid |
Changed in linux (Ubuntu Plucky): | |
status: | New → In Progress |
assignee: | nobody → AceLan Kao (acelankao) |
Changed in linux-oem-6.11 (Ubuntu Noble): | |
status: | New → In Progress |
assignee: | nobody → AceLan Kao (acelankao) |
Changed in linux-oem-6.11 (Ubuntu Plucky): | |
status: | New → Invalid |
description: | updated |
tags: | added: jira-stella-1107 oem-priority stella |
Changed in linux (Ubuntu): | |
status: | Confirmed → Invalid |
Changed in linux-oem-6.11 (Ubuntu): | |
status: | Confirmed → Invalid |
Changed in linux-oem-6.11 (Ubuntu Noble): | |
status: | In Progress → Fix Committed |
tags: |
added: verification-done-noble-linux-oem-6.11 removed: verification-needed-noble-linux-oem-6.11 |
This bug is awaiting verification that the linux-oem- 6.11/6. 11.0-1023. 23 kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification- needed- noble-linux- oem-6.11' to 'verification- done-noble- linux-oem- 6.11'. If the problem still exists, change the tag 'verification- needed- noble-linux- oem-6.11' to 'verification- failed- noble-linux- oem-6.11' .
If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.
See https:/ /wiki.ubuntu. com/Testing/ EnableProposed for documentation how to enable and use -proposed. Thank you!