Date: Sat, 15 Apr 2006 10:06:25 -0600 From: Scott Long <scottl@samsco.org> To: Oleg Bulyzhin <oleg@FreeBSD.org> Cc: cvs-src@FreeBSD.org, src-committers@FreeBSD.org, Scott Long <scottl@FreeBSD.org>, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/dev/bge if_bge.c Message-ID: <44411A01.1030305@samsco.org> In-Reply-To: <20060415102742.GA69882@lath.rinet.ru> References: <200604150813.k3F8D8Uo014854@repoman.freebsd.org> <20060415102742.GA69882@lath.rinet.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
Oleg Bulyzhin wrote: > On Sat, Apr 15, 2006 at 08:13:07AM +0000, Scott Long wrote: > >>scottl 2006-04-15 08:13:07 UTC >> >> FreeBSD src repository >> >> Modified files: >> sys/dev/bge if_bge.c >> Log: >> Fix the interrupt handler to do the mandatory PCI flush before looking at >> DMA memory. The could contribute towards missed link state changes under >> heavy bus load. >> >> Revision Changes Path >> 1.126 +10 -14 src/sys/dev/bge/if_bge.c > > > Should not we do the same for polling case? > No, because there is no race with the interrupt signal when you're polling. The data is in host memory or it isn't, and doing a bus flush on every loop would be highly counter productive. Scott
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?44411A01.1030305>