Date: Mon, 8 Jun 2009 15:35:08 +0100 (BST) From: Robert Watson <rwatson@FreeBSD.org> To: Kip Macy <kmacy@FreeBSD.org> Cc: src-committers@freebsd.org, svn-src-user@freebsd.org Subject: Re: svn commit: r193698 - user/kmacy/releng_7_2_fcs/sys/netinet Message-ID: <alpine.BSF.2.00.0906081534200.25282@fledge.watson.org> In-Reply-To: <200906080919.n589JPGJ022028@svn.freebsd.org> References: <200906080919.n589JPGJ022028@svn.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 8 Jun 2009, Kip Macy wrote: > skip idiv if queue is empty A more general optimization might be to check whether ip->ip_len is > mtu before doing the idiv case... Robert N M Watson Computer Laboratory University of Cambridge > > Modified: > user/kmacy/releng_7_2_fcs/sys/netinet/ip_output.c > > Modified: user/kmacy/releng_7_2_fcs/sys/netinet/ip_output.c > ============================================================================== > --- user/kmacy/releng_7_2_fcs/sys/netinet/ip_output.c Mon Jun 8 09:13:16 2009 (r193697) > +++ user/kmacy/releng_7_2_fcs/sys/netinet/ip_output.c Mon Jun 8 09:19:25 2009 (r193698) > @@ -447,12 +447,14 @@ again: > * interface in which case packetdrop should be done by queueing. > */ > #ifdef ALTQ > - if ((!ALTQ_IS_ENABLED(&ifp->if_snd)) && > + if (ifp->if_snd.ifq_len && > + (!ALTQ_IS_ENABLED(&ifp->if_snd)) && > ((ifp->if_snd.ifq_len + ip->ip_len / mtu + 1) >= > ifp->if_snd.ifq_maxlen)) > #else > - if ((ifp->if_snd.ifq_len + ip->ip_len / mtu + 1) >= > - ifp->if_snd.ifq_maxlen) > + if (ifp->if_snd.ifq_len && > + ((ifp->if_snd.ifq_len + ip->ip_len / mtu + 1) >= > + ifp->if_snd.ifq_maxlen)) > #endif /* ALTQ */ > { > error = ENOBUFS; >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.BSF.2.00.0906081534200.25282>