The striking thing about your log is that all the commands are WRITEs and they are all relatively short. Is the speed for reading better than the speed for writing?

Yes, it seems like you are pretty close - transfer rate differs for writing
only! For USB 2.0 device read is cca 16,7 MB/s for both 2.6.11 and 2.6.12.1,
but write speed is pretty 7.3 MB/s on 2.6.11, whereas only 133 kB/s on 2.6.12.

Someone had a very similar problem not too long ago (a month or two). It turned out that the slowdown was related not to the kernel but to the distribution, although the exact cause was never determined.

I got the slow 2.6.12 write while running in single user, only udevd was
running. Maybe I should try to shutdown it as well, but can't imagine how
it can interfere with usb-storage. I'll try nevertheless.

Anyway, this looks like you are writing to a filesystem that was mounted with -o sync or something similar.

Correct, I've been using -o sync with usb storage devices since last summer
as it has miracuously solved almost all my previous frequent filesystem
corruption issues since then.

You might find it interesting to try doing the exact same write transfer under both 2.6.11 and 2.6.12, with usb-storage debugging turned on for both systems, and then compare the debugging logs. If the timings are so drastically different then I would expect the sequence of commands to be different as well.

Good point, I'm going to build debug 2.6.11 kernel right now.

Vit


-------------------------------------------------------
This SF.Net email is sponsored by the 'Do More With Dual!' webinar happening
July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual
core and dual graphics technology at this free one hour event hosted by HP,
AMD, and NVIDIA.  To register visit http://www.hp.com/go/dualwebinar
_______________________________________________
Linux-usb-users@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-users

Reply via email to