----- Original Message ----- 
From: "Roman Weissgaerber" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, May 09, 2001 8:34 PM
Subject: Re: [linux-usb-devel] EILSEQ - FAQ ??


> > > Sergij Kolisnyk wrote:
> > > >
> > > > Hi!
> > > >
> > > > All we know the problem.
> > > >
> > > > On _some_ motherboards when doing USB bulk read at full speed,
> > > > URB completes with "-84" error code...
> > > >
> > >
> > > -84 (EILSEQ) means CRC Error of the packet sent by the device.
> > 
> > Or timeout...
> > 
> No!!!! EILSEQ means just CRC Error, ETIMEDOUT means timeout.

IMHO ETIMEDOUT means _user_ timeout,
specified in URB - really too many NAK's...
What are codes for:

1) USB frame time (=1ms) overrun
2) PCI data transfer timeout
3) USB device response timeout (to host's polling)

IMHO UHCI reports same error code on (2) and CRC error

> See Programming Guide for Linux USB Device Drivers, Page 26
> Section 3.1, Error Codes!!!
> See also kernel 2.4.4 uhci.c line 786 or usb-uhci.c line 2160 or 
> USB_ST_CRC (just CRC) for ohci.c (mapping is in usb.h).
> 
> > > So I think this is a problem of the device and not of the HC.
> > 
> > This is at least MB BIOS or VGA dependent!!!
> > This is checked.
> > 
> Or Power Supply or PCB Design ...
> 
> > Same HD. So the same Linux. Same 440BX. Different BIOS.
> > Different VGA (both nVidia). One of them has X running, one not.
> > 
> > The error occured only on one of them.
> > 
> > >
> > 
> > And problem is reported by many people trying to utilyze
> > full throughput avaliable in USB 1.1.
> > 
> 
> Who has reported, Where has it been reported?

Only here by Johan Verrept <[EMAIL PROTECTED]>
and Gordon McNutt <[EMAIL PROTECTED]>

Johan's report is old, Gordon's - few days ago.

> 
> > Different devices are affected.
> > 
> 
> What devices do you use.
> 
> - Roman
> 
> > When one of device or driver slowes the throughput down,
> > the error disappears.
> > 
> > MacOS drivers don't see the error at same speeds.
> > 
> > > - Roman
> > >
> > >
> > > > That happens only when device is fast enough to really utilyze
> > > > throughput of USB 1.1 full speed.
> > > >
> > > > IMHO, this error may depend on BIOS, VGA and having X running...
> > > > Error isn't HCI-specific, it occurs on both OHCI and UHCI.
> > > >
> > > > FAQ: How to workaround this error?
> > > > May be usb.c has to have URB-wrapper to suppress the error?
> > > > Isn't Linux overuse USB frame duration?
> > > >
> > > > I haven't found the answer anywhere...
> > > >
> > > > Thank you,
> > > > Sergij Kolisnyk



_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
http://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to