Alan Stern wrote: > On Sat, 27 Mar 2004, Brian S. Stephan wrote: > >> Alan Stern wrote: >> >> > problem might be similar to that -- data dependent. Or it might be a >> > timing issue; 2.6 does usb-storage data transfers more quickly than >> > 2.4. > > You might be able to improve things by changing the timing and throughput. > Here's how: In drivers/usb/storage/scsiglue.c you'll find a line that > defines max_sectors as 240, something like > > .max_sectors = 240, > > Try decreasing that value to 128 or maybe even smaller. That might very > well improve your transfers. > > Alan Stern
Transfers are improved with .max_sectors at 64 (128 didn't seem to offer any real improvement), making xfer sizes 32768 bytes instead of 65536, times are relatively sane again. The original test worked as it should but a larger cp is causing command aborts and bus resets again. Looking at the kernel log it seems that it is, however, still acting "better"... there are generally more successful writes between aborts, if this is any indication of good behavior. Hope this is helping. Brian ------------------------------------------------------- This SF.Net email is sponsored by: IBM Linux Tutorials Free Linux tutorial presented by Daniel Robbins, President and CEO of GenToo technologies. Learn everything from fundamentals to system administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
