forked from luck/tmp_suning_uos_patched
05545c92db
This patch implements tape statistics in the st module via sysfs. Current no statistics are available for tape I/O and there is no easy way to reuse the block layer statistics for tape as tape is a character device and does not have perform I/O in sector sized chunks (the size of the data written to tape can change). For tapes we also need extra stats related to things like tape movement (via other I/O). There have been multiple end users requesting statistics including AT&T (and some HP customers who have not given permission to be named). It is impossible for them to investigate any issues related to tape performance in a non-invasive way. [jejb: eliminate PRId64] Signed-off-by: Shane Seymour <shane.seymour@hp.com> Tested-by: Shane Seymour <shane.seymour@hp.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <JBottomley@Odin.com>
110 lines
3.3 KiB
Plaintext
110 lines
3.3 KiB
Plaintext
What: /sys/class/scsi_tape/*/stats/in_flight
|
|
Date: Apr 2015
|
|
KernelVersion: 4.2
|
|
Contact: Shane Seymour <shane.seymour@hp.com>
|
|
Description:
|
|
Show the number of I/Os currently in-flight between the st
|
|
module and the SCSI mid-layer.
|
|
Users:
|
|
|
|
|
|
What: /sys/class/scsi_tape/*/stats/io_ns
|
|
Date: Apr 2015
|
|
KernelVersion: 4.2
|
|
Contact: Shane Seymour <shane.seymour@hp.com>
|
|
Description:
|
|
Shows the total amount of time spent waiting for all I/O
|
|
to and from the tape drive to complete. This includes all
|
|
reads, writes, and other SCSI commands issued to the tape
|
|
drive. An example of other SCSI commands would be tape
|
|
movement such as a rewind when a rewind tape device is
|
|
closed. This item is measured in nanoseconds.
|
|
|
|
To determine the amount of time spent waiting for other I/O
|
|
to complete subtract read_ns and write_ns from this value.
|
|
Users:
|
|
|
|
|
|
What: /sys/class/scsi_tape/*/stats/other_cnt
|
|
Date: Apr 2015
|
|
KernelVersion: 4.2
|
|
Contact: Shane Seymour <shane.seymour@hp.com>
|
|
Description:
|
|
The number of I/O requests issued to the tape drive other
|
|
than SCSI read/write requests.
|
|
Users:
|
|
|
|
|
|
What: /sys/class/scsi_tape/*/stats/read_byte_cnt
|
|
Date: Apr 2015
|
|
KernelVersion: 4.2
|
|
Contact: Shane Seymour <shane.seymour@hp.com>
|
|
Description:
|
|
Shows the total number of bytes requested from the tape drive.
|
|
This value is presented in bytes because tape drives support
|
|
variable length block sizes.
|
|
Users:
|
|
|
|
|
|
What: /sys/class/scsi_tape/*/stats/read_cnt
|
|
Date: Apr 2015
|
|
KernelVersion: 4.2
|
|
Contact: Shane Seymour <shane.seymour@hp.com>
|
|
Description:
|
|
Shows the total number of read requests issued to the tape
|
|
drive.
|
|
Users:
|
|
|
|
|
|
What: /sys/class/scsi_tape/*/stats/read_ns
|
|
Date: Apr 2015
|
|
KernelVersion: 4.2
|
|
Contact: Shane Seymour <shane.seymour@hp.com>
|
|
Description:
|
|
Shows the total amount of time in nanoseconds waiting for
|
|
read I/O requests to complete.
|
|
Users:
|
|
|
|
|
|
What: /sys/class/scsi_tape/*/stats/write_byte_cnt
|
|
Date: Apr 2015
|
|
KernelVersion: 4.2
|
|
Contact: Shane Seymour <shane.seymour@hp.com>
|
|
Description:
|
|
Shows the total number of bytes written to the tape drive.
|
|
This value is presented in bytes because tape drives support
|
|
variable length block sizes.
|
|
Users:
|
|
|
|
|
|
What: /sys/class/scsi_tape/*/stats/write_cnt
|
|
Date: Apr 2015
|
|
KernelVersion: 4.2
|
|
Contact: Shane Seymour <shane.seymour@hp.com>
|
|
Description:
|
|
Shows the total number of write requests issued to the tape
|
|
drive.
|
|
Users:
|
|
|
|
|
|
What: /sys/class/scsi_tape/*/stats/write_ms
|
|
Date: Apr 2015
|
|
KernelVersion: 4.2
|
|
Contact: Shane Seymour <shane.seymour@hp.com>
|
|
Description:
|
|
Shows the total amount of time in nanoseconds waiting for
|
|
write I/O requests to complete.
|
|
Users:
|
|
|
|
|
|
What: /sys/class/scsi_tape/*/stats/resid_cnt
|
|
Date: Apr 2015
|
|
KernelVersion: 4.2
|
|
Contact: Shane Seymour <shane.seymour@hp.com>
|
|
Description:
|
|
Shows the number of times we found that a residual >0
|
|
was found when the SCSI midlayer indicated that there was
|
|
an error. For reads this may be a case of someone issuing
|
|
reads greater than the block size.
|
|
Users:
|