Update NVMe part of suspend.patch #3
Conversation
|
Sorry for noise. It seems that the commit merged into 5.3-rc1 (linux@d916b1b) and the commit I actually used in this PR (kernel/git/kbusch/linux.git - Keith Busch's fork of linux.git) are not the same. I will look into it closer. |
Note: NVMe part will be merged into Linux 5.3. Remove the part in 0002-suspend.patch when it arrives. For 5.2 - (Reverted NVMe part of 0002-suspend.patch to apply following patch set) - nvme: export get and set features · torvalds/linux@1a87ee6 torvalds/linux@1a87ee6 - nvme-pci: use host managed power state for suspend · torvalds/linux@d916b1b torvalds/linux@d916b1b#diff-bc4c090f021c046a7d256a3fcf86b7da For 4.19, this patch is also applied - nvme-pci: Sync queues on reset · torvalds/linux@d6135c3 torvalds/linux@d6135c3#diff-bc4c090f021c046a7d256a3fcf86b7da See - Surface Book with Performance Base: NVMe SSD breaks suspend (s2idle) · Issue jakeday#123 · jakeday/linux-surface jakeday#123
|
It does not seem to be a problem. force-pushed updated patches which use torvalds tree instead of kbusch tree. Updated patch is also working on 4.19 and 5.2 with my SB1. |
|
Awesome, thanks! |
|
@kitakar5525 There are some slight changes in v5.3 (e.g. https://github.com/torvalds/linux/blob/v5.3/drivers/nvme/host/pci.c#L2878-L2883, not sure if that's all). Have you tested the final v5.3 version yet? |
|
Thank you for letting me know of the update.
Yes, I've tested without any patch to NVMe and suspend is still working on 5.3 (the actual 0002-suspend.patch I used is here). I'll backport changes to 0002-suspend.patch, test the suspend, then make a PR for 4.19. (I'll not update/make a PR for 5.2 because we will move to 5.3. If needed, let me know.) |
|
@kitakar5525 Thanks! Given the limited lifetime of 5.2 and that it's something non-critical and rather minor, I think we can live without those changes there. |
ghost
mentioned this pull request
The current v6.9 patches include a fix for the serdev driver probe issue. Therefore, we don't need to build in the serial driver modules any longer.
2 participants
(I newly created a PR to correct English of commit name/message. I messed up my previous PR. Sorry for my bad English...)
(PR to v5.2 because I didn't test on 5.1, maybe no problem on 5.1, though)
Note:
NVMe part will be merged into Linux 5.3. Remove the part in
0002-suspend.patch when it arrives.
For 5.2
torvalds/linux@1a87ee6
torvalds/linux@d916b1b#diff-bc4c090f021c046a7d256a3fcf86b7da
For 4.19, this patch is also applied
torvalds/linux@d6135c3#diff-bc4c090f021c046a7d256a3fcf86b7da
See
Surface Book with Performance Base: NVMe SSD breaks suspend (s2idle) jakeday/linux-surface#123