Dave,
Could you please try the patch below on top of v6.1.80?
Works okay on top of v6.1.80:
[ 30.952668] scsi 6:0:0:0: Direct-Access HP 73.4G ST373207LW HPC1 PQ: 0 ANSI: 3 [ 31.072592] scsi target6:0:0: Beginning Domain Validation [ 31.139334] scsi 6:0:0:0: Power-on or device reset occurred [ 31.186227] scsi target6:0:0: Ending Domain Validation [ 31.240482] scsi target6:0:0: FAST-160 WIDE SCSI 320.0 MB/s DT IU QAS RTI WRFLOW PCOMP (6.25 ns, offset 63) [ 31.462587] ata5: SATA link down (SStatus 0 SControl 0) [ 31.618798] scsi 6:0:2:0: Direct-Access HP 73.4G ST373207LW HPC1 PQ: 0 ANSI: 3 [ 31.732588] scsi target6:0:2: Beginning Domain Validation [ 31.799201] scsi 6:0:2:0: Power-on or device reset occurred [ 31.846724] scsi target6:0:2: Ending Domain Validation [ 31.900822] scsi target6:0:2: FAST-160 WIDE SCSI 320.0 MB/s DT IU QAS RTI WRFLOW PCOMP (6.25 ns, offset 63)
Great, thanks for testing!
Greg, please revert the following commits from linux-6.1.y:
b73dd5f99972 ("scsi: sd: usb_storage: uas: Access media prior to querying device properties") cf33e6ca12d8 ("scsi: core: Add struct for args to execution functions")
and include the patch below instead.
Thank you!
Hi,
Martin K. Petersen martin.petersen@oracle.com (2024-04-08):
Great, thanks for testing!
Greg, please revert the following commits from linux-6.1.y:
b73dd5f99972 ("scsi: sd: usb_storage: uas: Access media prior to querying device properties") cf33e6ca12d8 ("scsi: core: Add struct for args to execution functions")
and include the patch below instead.
According to James I've ran into another expression of the same issue, which in my case led to the loss of some SMART information: https://lore.kernel.org/stable/20240410193207.qnb75osxuk4ovvm6@mraw.org/ https://lore.kernel.org/stable/0655cea93e52928d3e4a12b4fe2d2a4375492ed3.came...
I've just confirmed that both reverts plus that patch, applied on top of v6.1.85, fix that regression for me. That's tested in a QEMU VM with a SATA disk (that exposes only a few SMART attributes anyway) and also on baremetal with real disks (2 pairs of Seagate IronWolf): smartctl returns data again, including temperatures.
Closes: https://bugs.debian.org/1068675 Tested-by: Cyril Brulebois kibi@debian.org
Cheers,
On Mon, Apr 08, 2024 at 01:19:50PM -0400, Martin K. Petersen wrote:
Dave,
Could you please try the patch below on top of v6.1.80?
Works okay on top of v6.1.80:
[ 30.952668] scsi 6:0:0:0: Direct-Access HP 73.4G ST373207LW HPC1 PQ: 0 ANSI: 3 [ 31.072592] scsi target6:0:0: Beginning Domain Validation [ 31.139334] scsi 6:0:0:0: Power-on or device reset occurred [ 31.186227] scsi target6:0:0: Ending Domain Validation [ 31.240482] scsi target6:0:0: FAST-160 WIDE SCSI 320.0 MB/s DT IU QAS RTI WRFLOW PCOMP (6.25 ns, offset 63) [ 31.462587] ata5: SATA link down (SStatus 0 SControl 0) [ 31.618798] scsi 6:0:2:0: Direct-Access HP 73.4G ST373207LW HPC1 PQ: 0 ANSI: 3 [ 31.732588] scsi target6:0:2: Beginning Domain Validation [ 31.799201] scsi 6:0:2:0: Power-on or device reset occurred [ 31.846724] scsi target6:0:2: Ending Domain Validation [ 31.900822] scsi target6:0:2: FAST-160 WIDE SCSI 320.0 MB/s DT IU QAS RTI WRFLOW PCOMP (6.25 ns, offset 63)
Great, thanks for testing!
Greg, please revert the following commits from linux-6.1.y:
b73dd5f99972 ("scsi: sd: usb_storage: uas: Access media prior to querying device properties") cf33e6ca12d8 ("scsi: core: Add struct for args to execution functions")
and include the patch below instead.
Now done, thanks!
greg k-h
linux-stable-mirror@lists.linaro.org