Date: Thu, 20 Jul 2000 12:02:56 -0700 (PDT) From: Matthew Jacob <mjacob@feral.com> To: =?ISO-8859-1?Q?G=E9rard_Roudier?= <groudier@club-internet.fr> Cc: freebsd-alpha@FreeBSD.ORG Subject: more on- Re: fxp0 hangs on a PC164 using STABLE Message-ID: <Pine.BSF.4.10.10007201159110.45697-100000@beppo.feral.com> In-Reply-To: <Pine.BSF.4.10.10007201149120.45697-100000@beppo.feral.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> > A memory barrier here would not help since they are implied by the ldq_l/stq_c > pair, but, hell, let's put one just for grins. Drew? > You are, btw, absolutely right, in general, about the usage of memory barriers. Any shared memory model device in alpha should be using barrier instructions. We're given them for free with I/O and Mem space register accesses, but you have to add them by hand otherwise. The Qlogic driver uses them extensively. There's every reason to assume that they would be useful in i386 too, no? It sounds to me that this would be a good argument for a simple inline or asm for both i386 && alpha ports. Now, the sparc port will be more interesting because a plain 'memory barrier' is not what's there - instead you have to do explicit address based flushing and/or invalidation (depending on the platform). To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" 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.4.10.10007201159110.45697-100000>