Bug 201987
| Summary: | GPE 0X66 keeps on firing in late suspend phase, thus break suspend-to-idle - HP Zhan 66 | ||
|---|---|---|---|
| Product: | ACPI | Reporter: | Zhang Rui (rui.zhang) |
| Component: | Power-Sleep-Wake | Assignee: | Rafael J. Wysocki (rjw) |
| Status: | CLOSED CODE_FIX | ||
| Severity: | normal | CC: | rajneesh.bhardwaj |
| Priority: | P1 | ||
| Hardware: | Intel | ||
| OS: | Linux | ||
| Kernel Version: | v4.19 | Subsystem: | |
| Regression: | Yes | Bisected commit-id: | |
| Attachments: |
pm-graph output on HP Zhan 66 laptop, with kernel 4.20-rc4
ACPI / PM: s2idle: Disable non-wakeup GPEs on suspend |
||
Created attachment 280001 [details]
ACPI / PM: s2idle: Disable non-wakeup GPEs on suspend
Please try this patch on the affected system.
There is a new version of the patch to test at: https://patchwork.kernel.org/patch/10733155/ yes, I have tested the patches and they indeed fix the problem on this platform. Thanks! Fix merged as commit f941d3e41da7 (ACPI: EC / PM: Disable non-wakeup GPEs for suspend-to-idle). |
Created attachment 279997 [details] pm-graph output on HP Zhan 66 laptop, with kernel 4.20-rc4 On a HP KBL-R platform, model name Zhan 66, it is found that the system can not enter deep c state during suspend-to-idle. And the rootcause is that GPE 0X66 keeps on firing in the suspend_noirq phase, causing the system keeps on waking up to handle the ACPI interrupt. Thus, s2idle_enter() is never invoked in s2idle_loop().