Re: [gentoo-user] Noisy dd operation

[Alan McKinnon]

On Sun, 17 Jun 2012 19:27:36 -0400
Michael Mol <mikemol@×××××.com> wrote:

> So, I'm using dd to write the Gentoo LiveDVD ISO directly to an SDHC
> card. I was wondering why the transfer rate had slowed from 20MB/s to
> 6MB/s, so I ran 'sudo tail -f /var/log/messages' to see if there were
> I/O errors slowing things down.
>
> What I found was a flood-stream of message blocks that look like this:
>
> Jun 17 19:21:18 saffron kernel: [39023.458632] usb-storage:
> queuecommand_lck called
> Jun 17 19:21:18 saffron kernel: [39023.458670] usb-storage: ***
> thread awakened. Jun 17 19:21:18 saffron kernel: [39023.458685]
> usb-storage: Command WRITE_10 (10 bytes)
> Jun 17 19:21:18 saffron kernel: [39023.458687] usb-storage:  2a 00 00
> 2b f3 80 00 00 f0 00
> Jun 17 19:21:18 saffron kernel: [39023.458694] usb-storage:
> rts51x_invoke_transport: <---
> Jun 17 19:21:18 saffron kernel: [39023.458697] usb-storage:
> rts51x_invoke_transport: working scsi,
> intf->pm_usage_cnt:1,power.usage:1
> Jun 17 19:21:18 saffron kernel: [39023.458701] usb-storage: Bulk
> Command S 0x43425355 T 0x34ef L 122880 F 0 Trg 0 LUN 0 CL 10
> Jun 17 19:21:18 saffron kernel: [39023.458704] usb-storage:
> usb_stor_bulk_transfer_buf: xfer 31 bytes
> Jun 17 19:21:18 saffron kernel: [39023.458759] usb-storage: Status
> code 0; transferred 31/31
> Jun 17 19:21:18 saffron kernel: [39023.458761] usb-storage: --
> transfer complete Jun 17 19:21:18 saffron kernel: [39023.458763]
> usb-storage: Bulk command transfer result=0
> Jun 17 19:21:18 saffron kernel: [39023.458766] usb-storage:
> usb_stor_bulk_transfer_sglist: xfer 122880 bytes, 18 entries
> Jun 17 19:21:18 saffron kernel: [39023.486031] usb-storage: Status
> code 0; transferred 122880/122880
> Jun 17 19:21:18 saffron kernel: [39023.486035] usb-storage: --
> transfer complete Jun 17 19:21:18 saffron kernel: [39023.486037]
> usb-storage: Bulk data transfer result 0x0
> Jun 17 19:21:18 saffron kernel: [39023.486039] usb-storage: Attempting
> to get CSW...
> Jun 17 19:21:18 saffron kernel: [39023.486042] usb-storage:
> usb_stor_bulk_transfer_buf: xfer 13 bytes
> Jun 17 19:21:18 saffron kernel: [39023.491787] usb-storage: Status
> code 0; transferred 13/13
> Jun 17 19:21:18 saffron kernel: [39023.491790] usb-storage: --
> transfer complete Jun 17 19:21:18 saffron kernel: [39023.491793]
> usb-storage: Bulk status result = 0
> Jun 17 19:21:18 saffron kernel: [39023.491795] usb-storage: Bulk
> Status S 0x53425355 T 0x34ef R 0 Stat 0x0
> Jun 17 19:21:18 saffron kernel: [39023.491800] usb-storage:
> rts51x_invoke_transport: state:2
> Jun 17 19:21:18 saffron kernel: [39023.491802] usb-storage:
> rts51x_modi_suspend_timer: <---, state:2
> Jun 17 19:21:18 saffron kernel: [39023.491805] usb-storage:
> rts51x_modi_suspend_timer: --->
> Jun 17 19:21:18 saffron kernel: [39023.491807] usb-storage:
> rts51x_invoke_transport: --->
> Jun 17 19:21:18 saffron kernel: [39023.491809] usb-storage: scsi cmd
> done, result=0x0
> Jun 17 19:21:18 saffron kernel: [39023.491815] usb-storage: ***
> thread sleeping.
>
> ... Any idea what kernel configuration flag I may have enabled to
> cause these to be continually generated? My first guess would be
> something like 'kernel lock debugging'...if that's what I'm seeing
> here, where would I go to file a bug report against the usb-storage
> subsystem?
>

6M is what you should expect from a good Class 6 SDHC card, and those
are quite rare in consumer shops (few carry better than Class 4 in my
experience). Writes start very fast (it's going into the kernel buffer)
then gradually slow down to the card's actual speed, which is what you
appear to be seeing.

20M write speed would be a Class 20, which doesn't exist yet :-)


But what is your actual query? About the write speed?

Or about the presence of the messages?



-- 
Alan McKinnnon
alan.mckinnon@×××××.com