Re: g_file_storage speed issue

[jpollack@xxxxxxxxx]

I managed to get performance up a bit to around 10mb/sec & 5mb/sec (W
& R) by maxing out buflen at 131072, setting stall to 0 and adding
total=10000000; to dummy_hcd, but that's still an order of magnitude
less than it ought to be.  I'm poking around the files to see if
there's anything else I can understand that I can change.

Joshua

On 3/8/07, jpollack@xxxxxxxxx <jpollack@xxxxxxxxx> wrote:
> Thanks for your reply!
>
> I tried adding the line you suggested to dummy_hcd, unfortunately it
> doesn't seem to have any effect on the speed.  Any ideas?
>
> Yeah, I've tried all sorts of ways of tricking it with loop0,
> including making a symlink and renaming the device node.
> Unfortunately they don't work.  Someone clever had written a tool that
> uses LD_PRELOAD to intercept vmware's ioctls and rewrite them, but
> unfortunately that doesn't work currently.
>
> I'd like to get it working with this, do you have any other ideas on
> how to speed up/simplify this layer?
>
> On 3/8/07, Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> wrote:
> > You know, you could always create an entry for /dev/loop0 with a different
> > name.  For example:
> >
> >         mknod /dev/hdz b 7 0
> >
> > Or maybe even
> >
> >         ln -s /dev/loop0 /dev/hdz
> >
> > Then /dev/hdz would be a loopback block device and it would work okay with
> > VMware.
>

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Linux-usb-users@xxxxxxxxxxxxxxxxxxxxx
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-users