Re: [PATCH 6/6] scsi: return EAGAIN when resetting a device under EH

[Douglas Gilbert <dgilbert@interlog.com>]

On 14-11-06 04:46 PM, Elliott, Robert (Server Storage) wrote:
>
>
>> -----Original Message-----
>> From: Christoph Hellwig [mailto:hch@lst.de]
>> Sent: Thursday, 30 October, 2014 4:27 AM
>> To: linux-scsi@vger.kernel.org
>> Cc: Douglas Gilbert; Elliott, Robert (Server Storage)
>> Subject: [PATCH 6/6] scsi: return EAGAIN when resetting a device
>> under EH
>>
>> Signed-off-by: Christoph Hellwig <hch@lst.de>
>> ---
>>   drivers/scsi/scsi_ioctl.c | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/scsi/scsi_ioctl.c b/drivers/scsi/scsi_ioctl.c
>> index 712f159..c4f7b56 100644
>> --- a/drivers/scsi/scsi_ioctl.c
>> +++ b/drivers/scsi/scsi_ioctl.c
>> @@ -278,7 +278,7 @@ int
>> scsi_ioctl_block_when_processing_errors(struct scsi_device *sdev, int
>> cmd,
>>   {
>>   	if (cmd == SG_SCSI_RESET && ndelay) {
>>   		if (scsi_host_in_recovery(sdev->host))
>> -			return -ENODEV;
>> +			return -EAGAIN;
>>   	} else {
>>   		if (!scsi_block_when_processing_errors(sdev))
>>   			return -ENODEV;
>> --
>> 1.9.1
>
> This is how sg_reset responds to -EAGAIN, which seems reasonable:
> 	sg_reset: starting device reset
> 	sg_reset: SG_SCSI_RESET failed: Resource temporarily unavailable

That message has been tweaked so in sg3_utils version 1.40
that will read:
    sg_reset: try again later, may be resetting now

> That is easy to trigger when running sg_reset in parallel to
> multiple devices presented by the same controller.  In one
> example to 16 devices, 10 got that result while 6 succeeded.
>
> Tested-by: Robert Elliott <elliott@hp.com>
> Reviewed-by: Robert Elliott <elliott@hp.com>