hi, In short: there are SD cardreaders that send MEDIA_CHANGED on runtime resume. We cannot use runtime PM with these devices as I/O basically always fails. I'd like to discuss a way to fix this or at least allow users to work around this problem: For the full background, the discussion started in June 2020 here: https://lore.kernel.org/linux-scsi/20200623111018.31954-1-martin.kepplinger@xxxxxxx/ and I sent the first of these patches in August, as a reference: https://lore.kernel.org/linux-scsi/20200824190400.12339-1-martin.kepplinger@xxxxxxx/ so this is where I'm following up on. I'm not sure whether maintaining an in-kernel quirk for specific devices makes sense so here I suggest adding a userspace setting. Of course we should document it properly if this makes sense to you. But this way there's at least a chance to use runtime PM for sd cardreaders that send MEDIA_CHANGED. Questions would be: * Do you like the approach of a user+internal flag? * what to do in scsi drivers that ignore the flag? Is documentation enough? * review the *clearing* of the (internal) flag once again. the first occurrence of MEDIA_CHANGED should do that only. (note to myself) I'd appreciate any feedback. Martin Kepplinger (3): scsi: add expecting_media_change flag to error path scsi: add expect_media_change_suspend sysfs device setting scsi: sd: add support for expect_media_change_suspend flag drivers/scsi/scsi_error.c | 36 +++++++++++++++++++++++++++++++----- drivers/scsi/scsi_sysfs.c | 38 ++++++++++++++++++++++++++++++++++++++ drivers/scsi/sd.c | 21 ++++++++++++++++++++- include/scsi/scsi_device.h | 3 +++ 4 files changed, 92 insertions(+), 6 deletions(-) -- 2.20.1
