Date: Thu, 8 Nov 2012 15:32:59 -0800 From: Adrian Chadd <adrian@freebsd.org> To: Andre Oppermann <andre@freebsd.org> Cc: pyunyh@gmail.com, FreeBSD Net <freebsd-net@freebsd.org>, Pyun YongHyeon <yongari@freebsd.org> Subject: Re: svn commit: r242739 - stable/9/sys/dev/ti Message-ID: <CAJ-Vmo=QuP1s=ppnp2u%2BVpuJRbncaHLwi_0ku5=dr9%2BReYp79w@mail.gmail.com> In-Reply-To: <509BC2E2.4030907@freebsd.org> References: <201211080206.qA826RiN054539@svn.freebsd.org> <CAJ-VmomEOPGbLwmOmL0EdenZA7QKbV5P-hAYsTRcwLao2LbAqg@mail.gmail.com> <20121108023858.GA3127@michelle.cdnetworks.com> <CAJ-Vmoma1DJRT8_ezdEpVYrZXak%2B2B4mBHAPxhoUKr0nUrO6YQ@mail.gmail.com> <509BC2E2.4030907@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 8 November 2012 06:34, Andre Oppermann <andre@freebsd.org> wrote: > TCP/UDP doesn't (want to) generate any fragments at all and tries > to avoid it at almost all cost. We want to send very large packets > and have the NIC fragment/segment it (TSO/UDP frag offload). What about if it's a router and the frames don't have DF set? Not that it should happen often, however.. > >> We could create a device or interface flag that indicates whether the >> driver can handle multiple mbufs chained via m_nextpkt through >> if_transmit(), and then teach one or two drivers that particular >> logic. > > > Agreed. I think that's the way to go. It must be very well specified > in semantics though. Otherwise it's just too easy to leak mbuf all > over the place. I mentioned this to Robert Watson today at the FreeBSD vendor summit and he said much the same thing about testing and assertions. Adrian
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAJ-Vmo=QuP1s=ppnp2u%2BVpuJRbncaHLwi_0ku5=dr9%2BReYp79w>