Re: Re: (probably not a) nvidia interrupt problem
"Axel Dörfler" <[email protected]>
| Newsgroups | gmane.os.openbeos.kernel.devel |
|---|---|
| Message-ID | <24930546428-BeMail@zon> |
Jérome DUVAL <[email protected]> wrote: > About the possible change : the idea would be to check the interrupt > service > bit when irq is 7 or 15 before pic_end_of_interrupt is called. > If the bit is off then we have a wrong interrupt. > > link : http://fxr.watson.org/fxr/source/amd64/isa/intr_machdep.c#L261 Stefano Ceccherini <[email protected]> wrote: > http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/i386/isa/atpic.c.diff?r1=1.6&r2=1.7&f=h > I was going for a simpler fix (just checking if interrupt 7 was enabled), but that one is better as it also catches spurious interrupts when an actual interrupt handler is installed (even if it's slower), thanks for the hint. Bye, Axel. ------------------------------------------------------- This SF.Net email is sponsored by Yahoo. Introducing Yahoo! Search Developer Network - Create apps using Yahoo! Search APIs Find out how you can build Yahoo! directly into your own Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005