From mdharm@multivac.one-eyed-alien.net Mon Jun 6 17:24:06 2005 Date: Mon, 6 Jun 2005 17:22:42 -0700 From: Matthew Dharm To: USB Storage List , Greg KH Subject: USB Storage: retry hard errors Message-ID: <20050607002242.GD8689@one-eyed-alien.net> This patch started life as as527, and was rediffed by me. Since the IDE interface doesn't convey much information about types of errors, many USB-IDE adapters report all low-level errors with SK = 0x04, which is supposed to be used only for non-recoverable errors. As a result the SCSI midlayer doesn't retry the command. But quite often a retry would succeed, whereas an unnecessary retry doesn't really hurt anything. This patch uses a recently-implemented flag to tell the SCSI midlayer that such hardware errors should be retried. Signed-off-by: Alan Stern Signed-off-by: Matthew Dharm Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/scsiglue.c | 9 +++++++++ 1 files changed, 9 insertions(+) --- gregkh-2.6.orig/drivers/usb/storage/scsiglue.c 2005-06-09 22:47:02.000000000 -0700 +++ gregkh-2.6/drivers/usb/storage/scsiglue.c 2005-06-09 22:47:43.000000000 -0700 @@ -155,6 +155,15 @@ * If this device makes that mistake, tell the sd driver. */ if (us->flags & US_FL_FIX_CAPACITY) sdev->fix_capacity = 1; + + /* USB-IDE bridges tend to report SK = 0x04 (Non-recoverable + * Hardware Error) when any low-level error occurs, + * recoverable or not. Setting this flag tells the SCSI + * midlayer to retry such commands, which frequently will + * succeed and fix the error. The worst this can lead to + * is an occasional series of retries that will all fail. */ + sdev->retry_hwerror = 1; + } else { /* Non-disk-type devices don't need to blacklist any pages