Date: Mon, 20 May 2002 10:54:15 -0700 (PDT) From: Paul Herman <pherman@frenchfries.net> To: "M. Warner Losh" <imp@village.org> Cc: gallatin@cs.duke.edu, <freebsd-stable@FreeBSD.ORG> Subject: Re: ipl problem in wi_hostapp.c ? Message-ID: <20020520104220.V495-100000@mammoth.eat.frenchfries.net> In-Reply-To: <20020520.104632.10320851.imp@village.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 20 May 2002, M. Warner Losh wrote: > I'm sure there's an IPL priblem in wi_hostap. Don't use it til I > get a fix MFC'd. Bruce and I have been discussing things (mostly > bruce telling me when I'm trying to be a dumb-ass splwise, which > I was). Far from it beeing a solution, but replacing splsoftclock() with splhigh() did indeed stop my machine panicing. I take it then, it's more complicated than using a different priority level? <me/ grabs dumb-ass torch> Just curious, what is then the reason that splsoftclock() has the magical property of clearing all other SWI_* bits? I ask because the folowing change to ipl_funcs.c also stopped the panic I was experiencing. Machine is still running for now.... :-) ========== RCS file: /u02/ncvs/src/sys/i386/isa/Attic/ipl_funcs.c,v retrieving revision 1.32.2.4 diff -u -1 -r1.32.2.4 ipl_funcs.c --- ipl_funcs.c 26 Jul 2001 18:53:02 -0000 1.32.2.4 +++ ipl_funcs.c 20 May 2002 17:07:42 -0000 @@ -267,3 +267,3 @@ GENSPL(splsoftcamnet, |=, SWI_CAMNET_MASK, 10) -GENSPL(splsoftclock, =, SWI_CLOCK_MASK, 11) +GENSPL(splsoftclock, |=, SWI_CLOCK_MASK, 11) GENSPL(splsofttty, |=, SWI_TTY_MASK, 12) -Paul. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020520104220.V495-100000>