Tim Woodall
Thu, 02 May 2002 01:48:43 -0700
On Wed, 1 May 2002, Vladimir Vyskocil wrote:
>
> Hi,
>
> I'm investigating the lock out I have on my board with a USB Controller: Intel Corp.
>82371SB PIIX3 USB [Natoma/Triton II] (rev 1).
> I have found that pppoa2 (I don't tried with pppoa3 for now) from
>speedtouch-CVS_03_03_2002 package block from time to time a this line in pppoa2.c :
>
> n = pusb_endpoint_read(epdata, lbuf, sizeof(lbuf), 0);
>
> When the lock append, I must wait that my cron script 'checkadsl' ping remote ppp
>adresse and then restart pppd. pppoa2 get stucked at this line waiting for data that
>never come (i don't know why ?) until the script kill pppd.
>
> I changed this line to
>
> n = pusb_endpoint_read(epdata, lbuf, sizeof(lbuf), 30000);
>
> And now I don't have lockout anymore ! But when there is no data transfert for 30s I
>get these lines in /var/log/syslog :
>
> May 1 22:48:54 hector kernel: usb_control/bulk_msg: timeout
> May 1 22:48:54 hector kernel: usbdevfs: USBDEVFS_BULK failed dev 3 ep 0x87 len 3392
>ret -110
>
> It seems that something better should be done in the driver, as I saw that I'm not
>alone with this problem.
>
We think it is a buggy chipset problem. pppoa3 works better for me - and I
have posted a patch that halves the download rate but seems to be stable.
(When I tried this, the usb stall stopped for 30 seconds until the timeout
restarted. You will get better response if you reduce the 30000 to about
500 - but I suggest that you patch the kernel to stop the logs filling up -
alternatively don't log kern.debug (I think).
I think that this is going to need a kernel fix rather than a driver fix
(although I suspect that it is going to be a CONFIG_TRITON_FIX compile
option)
Tim
--
God said, "div D = rho, div B = 0, curl E = - @B/@t, curl H = J + @D/@t,"
and there was light.
http://tjw.hn.org/ http://www.locofungus.btinternet.co.uk/
Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]?subject=unsubscribe