On Mon, 14 Jun 2004, Nicolas DET wrote:

> start_interrupt enable the interrupt from the chip side. It's the exact 
> opposite of stop_interrupt.
> 
> I think this way to handle the interrupt is "smarter". but that's not 
> the main reason for the patch. The main problem was "system hangs" when 
> stressing USB... which is very nasty :-(.
> That I mean is that the interrupt modif are not reallya must, but the 
> align part is IMO.
> 
> By the way, the diff did not appear in the mail. I'm sorry I'm use to 
> that kind of ML.
> 
> I will put the files somewhere on the web and give the URL here.

You don't have to; the second message you posted included your changed 
files.  Now I understand what you meant by start/stop_interrupt.

The computers you mentioned use PPC processors.  Do you know if these
cache-line effects are equally important for Intel x86 machines?  It may 
be that the problems simply haven't come up very much in the past because 
relatively few people use PPC-based Linux systems.

I will have to alter your code, since L1_CACHE_LINE_SIZE is defined only
in the asm-ppc include files.  The equivalent value, L1_CACHE_BYTES,
is defined everywhere and should work just as well.

Alan Stern



-------------------------------------------------------
This SF.Net email is sponsored by the new InstallShield X.
>From Windows to Linux, servers to mobile, InstallShield X is the
one installation-authoring solution that does it all. Learn more and
evaluate today! http://www.installshield.com/Dev2Dev/0504
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to