AdvLoggerPkg/AdvancedFileLogger: Flush logs only once per reset - #923
Open
liqiqiii wants to merge 1 commit into
Open
AdvLoggerPkg/AdvancedFileLogger: Flush logs only once per reset#923liqiqiii wants to merge 1 commit into
liqiqiii wants to merge 1 commit into
Conversation
A capsule-armed reset converts a warm reset into a cold reset by issuing a second, nested ResetSystem() call. NvmExpressDxe shuts its controller down on the first reset notification (via the reset notification protocol, which fires after this platform-specific reset filter) and does not re-initialize it, so re-writing the logs on the nested reset issues I/O to a powered-down NVMe controller. This stalls for the full NVMe command timeout (~5 seconds) and logs a fatal NVMe WHEA telemetry record (EFI_IO_BUS_SCSI | EFI_IOB_EC_INTERFACE_ERROR, 0x02070005), and the log write ultimately fails with a timeout. Add a module-static latch so the in-memory log is flushed to media only once per reset sequence. Nested reset notifications early-return instead of re-writing to the already-shut-down controller. The normal ReadyToBoot / ExitBootServices / sync flush paths are unaffected. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: cccd4451-999f-4caf-bdb4-da35fd5bef85
Contributor
✅ QEMU Validation PassedSource Dependencies
Results
Workflow run: https://github.com/microsoft/mu_plus/actions/runs/29968504342 This comment was automatically generated by the Mu QEMU PR Validation workflow. |
mu-automation
Bot
added
the
impact:non-functional
Collaborator
|
Has the unregister path been tested? What if, after the log was written, the notification function was unregistered? |
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## release/202511 #923 +/- ##
================================================
Coverage ? 3.02%
================================================
Files ? 37
Lines ? 4332
Branches ? 73
================================================
Hits ? 131
Misses ? 4195
Partials ? 6
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Contributor
Author
Good catch Aaron, I will get some Intel secured core/non secured core devices tested as well. The secured core could be highly influence by this. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
A capsule-armed reset converts a warm reset into a cold reset by issuing a second, nested
ResetSystem()call.NvmExpressDxeshuts its controller down on the first reset notification (via the reset notification protocol, which fires after this platform-specific reset filter) and does not re-initialize it, so re-writing the logs on the nested reset issues I/O to a powered-down NVMe controller. This stalls for the full NVMe command timeout (~5 seconds) and logs a fatal NVMe WHEA telemetry record(EFI_IO_BUS_SCSI | EFI_IOB_EC_INTERFACE_ERROR, 0x02070005), and the log write ultimately fails with a timeout.Add a module-static latch so the in-memory log is flushed to media only once per reset sequence. Nested reset notifications early-return instead of re-writing to the already-shut-down controller. The normal ReadyToBoot / ExitBootServices / sync flush paths are unaffected.
For details on how to complete these options and their meaning refer to CONTRIBUTING.md.
How This Was Tested
Tested with hardware, work as expected.