Re: [PATCH 1/1] verify: move meta header to generic verify_header

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



> If you feel like you are up for more in this area, I'd love to have a mode
> that writes verify info based the location on disk/file, and hence doesn't
> need a header. This could/should work with any of the hashes, and it should
> have a way to specify the seed for the data. The generated data for the
> block would be a function of the offset, hash, seed.

Would this be akin to dt's "pattern=iot"? From
http://www.scsifaq.org/RMiller_Tools/ftp/dt/dt-UsersGuide.pdf :
"So, what is DJ’s IOT test pattern? This pattern places the logical
block address (lba) in the first word (4 bytes) of each block, with
(lba+=0x01010101) being placed in all remaining words in the data
block (512 bytes by default). In this way, the logical block is seeded
throughout each word in the block."

If so what would happen on things like the second pass - e.g. if fio
stops how do you know which iteration you were on or is this
irrelevant?

-- 
Sitsofe | http://sucs.org/~sits/
--
To unsubscribe from this list: send the line "unsubscribe fio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [Linux Kernel]     [Linux SCSI]     [Linux IDE]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux SCSI]

  Powered by Linux