On Thu, 2019-04-04 at 16:47 +-0200, Oliver Neukum wrote: +AD4 If a reset of a SCSI device requires no relocking of the door, the +AD4 flag must be reset anyway, lest it trigger unnecessary door action +AD4 later on. +AD4 +AD4 Signed-off-by: Oliver Neukum +ADw-oneukum+AEA-suse.com+AD4 +AD4 --- +AD4 drivers/scsi/scsi+AF8-error.c +AHw 5 ++1filechanged2insertions(), 3 deletions(-) +AD4 +AD4 diff --git a/drivers/scsi/scsi+AF8-error.c b/drivers/scsi/scsi+AF8-error.c +AD4 index 16eef068e9e9..afb8cf0cdfed 100644 +AD4 --- a/drivers/scsi/scsi+AF8-error.c +AD4 +++b/drivers/scsi/scsierrorc +AD4 +AEAAQA -1990,10 +19909static void scsi+AF8-restart+AF8-operations(struct Scsi+AF8-Host +ACo-shost) +AD4 +ACo is no point trying to lock the door of an off-line device. +AD4 +ACo-/ +AD4 shost+AF8-for+AF8-each+AF8-device(sdev, shost) +AHs +AD4 - if (scsi+AF8-device+AF8-online(sdev) +ACYAJg sdev-+AD4-was+AF8-reset +ACYAJg sdev-+AD4-locked) +AHs +AD4 if (scsi+AF8-device+AF8-online(sdev) +ACYAJg sdev-+AD4-was+AF8-reset +ACYAJg sdev-+AD4-locked) +AD4 scsi+AF8-eh+AF8-lock+AF8-door(sdev)+ADs +AD4 - sdev-+AD4-was+AF8-reset +AD0 0+ADs +AD4 - +AH0 +AD4 sdev-+AD4-was+AF8-reset +AD0 0+ADs +AD4 +AH0 I think there are two issues with this patch: +ACo Shouldn't sdev-+AD4-was+AF8-reset be cleared only if a device is online instead of ignoring the device online state? +ACo An analysis of the impact of this patch on the osst and st drivers is missing. These two drivers namely read the +ACI-was+AF8-reset+ACI member variable. Thanks, Bart.
