On Fri, Mar 12, 2021 at 07:26:31PM +0100, Sedat Dilek wrote: > On Fri, Mar 12, 2021 at 7:05 PM Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> wrote: > > Although it's not conclusive, this log seems to indicate that ata_id > > is the only program causing resets. Have you tried preventing the > > ata_id program from running (for example, by renaming it)? > > > > This is /lib/udev/ata_id from Debian's udev package. That does not answer my question. > > > Your diff now should say; s/SCSI ioctl error/SCSI ioctl info'. > > > > No, it shouldn't. The log message itself is an info, but the event it > > reports is an error. > > > > OK. > Some of these SCSI ioctl errors are not causing a xhci-reset. Yes, I noticed that. In fact, the commands that cause a reset are all A1 (and not all of them), never 85. > > > Alan, so "t" flags should be added as a quirks to linux-kernel sources... > > > > > > t = NO_ATA_1X (don't allow ATA(12) and ATA(16) commands, uas only); > > > > > > ...for my ASMedia USB-3.0 controller? > > > > That's not at all clear. This is a very common and popular device, > > and nobody else has reported these problems. It could be that > > something is odd about your particular drive or computer, not these > > drives in general. > > > > So, the external USB-3.0 HDD is now in "UAS only" mode/status. Why? Did you change something? Alan Stern > Cannot judge if things got better or not. > > - Sedat -
