Date: Sat, 30 Nov 2013 09:50:55 +0100 From: Michael Tuexen <Michael.Tuexen@lurchi.franken.de> To: Adrian Chadd <adrian@freebsd.org> Cc: "freebsd-net@freebsd.org list" <freebsd-net@freebsd.org> Subject: Re: ip_output()/if_output() behaviour Message-ID: <41906339-8AE7-4444-8B84-6FC19D48B132@lurchi.franken.de> In-Reply-To: <CAJ-VmonJijEvrftE6SfUrK7vONXOAY62usik7mFfzo9zHmqt0g@mail.gmail.com> References: <BF7B04F7-0D45-4708-99A8-8BE030109CEC@lurchi.franken.de> <CAJ-Vmo=Jsf=7uXxwJ=Md5KLFpvSYAcaaNrq%2BbHsw75nfSG_ZaQ@mail.gmail.com> <B7E3AA58-172A-4D20-B625-95C4712D46E7@lurchi.franken.de> <52987E27.10503@freebsd.org> <8C291076-5F03-4406-B689-A3185E6DD313@lurchi.franken.de> <5298BD5D.3020203@freebsd.org> <CAJ-VmonJijEvrftE6SfUrK7vONXOAY62usik7mFfzo9zHmqt0g@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Nov 29, 2013, at 11:39 PM, Adrian Chadd <adrian@freebsd.org> wrote: > +1 > > > On 29 November 2013 08:14, Julian Elischer <julian@freebsd.org> wrote: > >>> ifnet(9) says: >>> >>> if_transmit() >>> Transmit a packet on an interface or queue it if the interface >>> is >>> in use. This function will return ENOBUFS if the devices >>> software >>> and hardware queues are both full. ... >>> >>> So I guess returning ENOBUFS when the packet was queued is wrong... >> >> I think it is. >> >> ENOBUFS means "I couldn't proceed due to no buffers" >> not "I used up the last one on this operation". > > Yes, it's wrong. ENOBUFS means "couldn't queue; no buffers." Please > provide a diff against igb and I'll make sure Jack/Intel get it into > (his, freebsd) tree. I sent the patch already yesterday to the list and Jack. It covers em, igb and ixgbe. Best regards Michael > > Thanks! > > > > -adrian >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?41906339-8AE7-4444-8B84-6FC19D48B132>