You are not logged in.
If seth's suggested kernel parameter does not fix the issue on a kernel without the revert it needs to go upstream for a fix.
Offline
If seth's suggested kernel parameter does not fix the issue on a kernel without the revert it needs to go upstream for a fix.
I restored my status to the beginning of this thread, i.e. 6.2.1.arch1-1 and 6.1.14-1-lts, added ahci.mobile_lpm_policy=1 as a parameter to both kernels, and apparently made no difference, /dev/sda still MIA in both cases.
Offline
board_ahci_low_power differs from board_ahci in AHCI_HFLAG_USE_LPM_POLICY, so I guess even
- { PCI_VDEVICE(INTEL, 0xa0d3), board_ahci_low_power }, /* Tiger Lake UP{3,4} AHCI */
+ { PCI_VDEVICE(INTEL, 0xa0d3), board_ahci }, /* Tiger Lake UP{3,4} AHCI */
would fail?
Online
board_ahci_low_power differs from board_ahci in AHCI_HFLAG_USE_LPM_POLICY, so I guess even
- { PCI_VDEVICE(INTEL, 0xa0d3), board_ahci_low_power }, /* Tiger Lake UP{3,4} AHCI */ + { PCI_VDEVICE(INTEL, 0xa0d3), board_ahci }, /* Tiger Lake UP{3,4} AHCI */
would fail?
That would make sense, unless that parameter doesn't actually work as-intended.
Offline
What if the kernel parameter libata.force=nolpm is added?
Edit:
If that does not work what about reverting 104ff59af73aba524e57ae0fef70121643ff270e
https://drive.google.com/file/d/1EITlud … share_link linux-6.2.1.arch1-9-x86_64.pkg.tar.zst
https://drive.google.com/file/d/1xx0Hd5 … share_link linux-headers-6.2.1.arch1-9-x86_64.pkg.tar.zst
This package works for me too. The drive is recognized correctly.
# uname -a
Linux i5lenovo 6.2.1-arch1-9 #1 SMP PREEMPT_DYNAMIC Wed, 01 Mar 2023 14:10:24 +0000 x86_64 GNU/Linux
# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINTS
sda 8:0 0 931,5G 0 disk
├─sda1 8:1 0 32G 0 part /mnt/data-vfat
└─sda2 8:2 0 899,5G 0 part /mnt/data
nvme0n1 259:0 0 476,9G 0 disk
├─nvme0n1p1 259:1 0 4G 0 part /boot
├─nvme0n1p2 259:2 0 32G 0 part
├─nvme0n1p3 259:3 0 24G 0 part [SWAP]
└─nvme0n1p4 259:4 0 416,9G 0 part /
Thanks for your help!
Should I add "SOLVED" to the subject heading?
Offline
board_ahci_low_power differs from board_ahci in AHCI_HFLAG_USE_LPM_POLICY, so I guess even
- { PCI_VDEVICE(INTEL, 0xa0d3), board_ahci_low_power }, /* Tiger Lake UP{3,4} AHCI */ + { PCI_VDEVICE(INTEL, 0xa0d3), board_ahci }, /* Tiger Lake UP{3,4} AHCI */
would fail?
https://drive.google.com/file/d/1QsCzEN … share_link linux-6.2.1.arch1-10-x86_64.pkg.tar.zst
https://drive.google.com/file/d/1nSvBPX … share_link linux-headers-6.2.1.arch1-10-x86_64.pkg.tar.zst
Offline
loqs wrote:What if the kernel parameter libata.force=nolpm is added?
Edit:
If that does not work what about reverting 104ff59af73aba524e57ae0fef70121643ff270e
https://drive.google.com/file/d/1EITlud … share_link linux-6.2.1.arch1-9-x86_64.pkg.tar.zst
https://drive.google.com/file/d/1xx0Hd5 … share_link linux-headers-6.2.1.arch1-9-x86_64.pkg.tar.zstThis package works for me too. The drive is recognized correctly.
# uname -a Linux i5lenovo 6.2.1-arch1-9 #1 SMP PREEMPT_DYNAMIC Wed, 01 Mar 2023 14:10:24 +0000 x86_64 GNU/Linux # lsblk NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINTS sda 8:0 0 931,5G 0 disk ├─sda1 8:1 0 32G 0 part /mnt/data-vfat └─sda2 8:2 0 899,5G 0 part /mnt/data nvme0n1 259:0 0 476,9G 0 disk ├─nvme0n1p1 259:1 0 4G 0 part /boot ├─nvme0n1p2 259:2 0 32G 0 part ├─nvme0n1p3 259:3 0 24G 0 part [SWAP] └─nvme0n1p4 259:4 0 416,9G 0 part /
Thanks for your help!
Should I add "SOLVED" to the subject heading?
I'd probably hold off on marking it as "solved" until the problem gets reported upstream.
Offline
seth wrote:board_ahci_low_power differs from board_ahci in AHCI_HFLAG_USE_LPM_POLICY, so I guess even
- { PCI_VDEVICE(INTEL, 0xa0d3), board_ahci_low_power }, /* Tiger Lake UP{3,4} AHCI */ + { PCI_VDEVICE(INTEL, 0xa0d3), board_ahci }, /* Tiger Lake UP{3,4} AHCI */
would fail?
https://drive.google.com/file/d/1QsCzEN … share_link linux-6.2.1.arch1-10-x86_64.pkg.tar.zst
https://drive.google.com/file/d/1nSvBPX … share_link linux-headers-6.2.1.arch1-10-x86_64.pkg.tar.zst
The drive is gone again
uname -a
Linux i5lenovo 6.2.1-arch1-10 #1 SMP PREEMPT_DYNAMIC Wed, 01 Mar 2023 16:41:12 +0000 x86_64 GNU/Linux
lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINTS
nvme0n1 259:0 0 476,9G 0 disk
├─nvme0n1p1 259:1 0 4G 0 part /boot
├─nvme0n1p2 259:2 0 32G 0 part
├─nvme0n1p3 259:3 0 24G 0 part [SWAP]
└─nvme0n1p4 259:4 0 416,9G 0 part /
dmesg | grep -Ei 'ata'
[ 1.634243] ahci 10000:e0:17.0: AHCI 0001.0301 32 slots 2 ports 6 Gbps 0x3 impl SATA mode
[ 1.635584] ata1: SATA max UDMA/133 abar m2048@0x50202000 port 0x50202100 irq 145
[ 1.635590] ata2: SATA max UDMA/133 abar m2048@0x50202000 port 0x50202180 irq 145
[ 1.946998] ata2: SATA link down (SStatus 4 SControl 300)
[ 1.947038] ata1: SATA link down (SStatus 4 SControl 300)
Offline
There are two ways to report this to upstream:
1 Open an issue on the kernel bugzilla possibly under product IO/Storage component Serial ATA
2 Reply to https://lore.kernel.org/all/20230213102 … gslab.com/
See also https://www.kernel.org/doc/html/latest/ … ssues.html
Offline
There are two ways to report this to upstream:
1 Open an issue on the kernel bugzilla possibly under product IO/Storage component Serial ATA
2 Reply to https://lore.kernel.org/all/20230213102 … gslab.com/
See also https://www.kernel.org/doc/html/latest/ … ssues.html
I've filed it upstream;
Offline
I've filed it upstream;
Thank you. I would suggest adding the author of the commit involved Simon Gaiser <simon@invisiblethingslab.com> and its committer Damien Le Moal <damien.lemoal@opensource.wdc.com> to the cc list of the bug report.
Offline
emmixis wrote:I've filed it upstream;
Thank you. I would suggest adding the author of the commit involved Simon Gaiser <simon@invisiblethingslab.com> and its committer Damien Le Moal <damien.lemoal@opensource.wdc.com> to the cc list of the bug report.
Damien has directly taken the regression to the tracker and has sent a revert to stable, and is working with Simon to implement the intended change in a way that does not cause a regression.
Offline
Revert is expected to be in place by 6.1.15. Can someone update the topic when this revert makes it to Arch? Preferably the one that'll hit the 6.2 line?
Offline
If it's there it's likely in the accompanying stable patch series for 6.2 so either 6.2.2 or 6.2.3. If you know the commit hash/description you can check the update logs for stable releases on kernel.org
Offline
Revert is expected to be in place by 6.1.15. Can someone update the topic when this revert makes it to Arch? Preferably the one that'll hit the 6.2 line?
Installed linux-lts 6.1.15-1 from testing repo. There are no sda devices.
Mar 03 17:38:10 i5lenovo kernel: ata1: SATA link down (SStatus 4 SControl 300)
Mar 03 17:38:10 i5lenovo kernel: ata2: SATA link down (SStatus 4 SControl 300)
Offline
The revert is not in 6.1.15 or 6.2.2 they started their release before the bug was raised upstream. The revert is currently in the libata tree https://git.kernel.org/pub/scm/linux/ke … 93ec67f6d4 with a pull request to Linus for mainline. Once it is in mainline it can be queued for stable. Arch could pick it up sooner than upstream if you file a bug report on the Arch bug tracker.
Offline
The revert is not in 6.1.15 or 6.2.2 they started their release before the bug was raised upstream. The revert is currently in the libata tree https://git.kernel.org/pub/scm/linux/ke … 93ec67f6d4 with a pull request to Linus for mainline. Once it is in mainline it can be queued for stable. Arch could pick it up sooner than upstream if you file a bug report on the Arch bug tracker.
Whats the ETA on that, unless I file a report here to have it merged downstream first....?
Offline
Offline
SSD disk works correctly with linux 6.2.2.arch2-1 package. Problem solved. Thanks for your help!
Offline
Is anyone previously affected by this issue available to participate in some testing/diagnosis of the root cause here? This would involve building kernels with specific patches applied and getting logs (so you need to be using the SATA disk as secondary, booting from another device).
We have identified that the lack of LPM mode is likely causing all platforms using this chipset to excessively drain the battery in suspend mode. We want to fix that for all, but obviously need to avoid the breakage that was discovered here.
The platform we have (Asus B1400) works just fine with a SATA disk added in LPM mode, so we aren't able to diagnose there.
Offline
I can try this if you give me detailed instructions.
Offline