Date: Mon, 2 Sep 2002 21:30:27 -0700 (PDT) From: Matthew Jacob <mjacob@feral.com> To: Peter Wemm <peter@wemm.org> Cc: cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/dev/mpt mpt_freebsd.h Message-ID: <Pine.BSF.4.21.0209022124290.5847-100000@beppo> In-Reply-To: <20020903042234.8C2852A894@canning.wemm.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Well, I've had the same code in place in isp for over a year on sparc, i386 && alpha. I asked several times- including John Baldwin about this, and he seemed to think it was fine, although, as he put it 'unnecessary for now'. Note that the isp && mpt drivers *are* 'safe'. It's CAM that isn't. But I'm not sure there actually *is* a re-entrancy problem. Do you have actual real evidence of this? If so, point out the code path *in* CAM that has the issue. Then it can be fixed. Please don't just wave your arms publicly when you haven't really amassed all the facts or checked into the history of this or the gotten the full story. At any rate, I'm just about done making a similar change to isp. This all should get rethought when CAM gets made 'safe'. On Mon, 2 Sep 2002, Peter Wemm wrote: > Matt Jacob wrote: > > mjacob 2002/09/02 19:18:21 PDT > > > > Modified files: > > sys/dev/mpt mpt_freebsd.h > > Log: > > Don't do transition locking (i.e., CAM->MPT->CAM)- Peter claimed that ia64 > > chokes the chicken with this. > > Well, it just isn't legal to release Giant in the middle of unsafe code > on any platform. By doing this you allow reentrency into the cam stack. > > Cheers, > -Peter > -- > Peter Wemm - peter@wemm.org; peter@FreeBSD.org; peter@yahoo-inc.com > "All of this is for nothing if we don't go to the stars" - JMS/B5 > > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0209022124290.5847-100000>