Date: Wed, 4 Feb 2009 14:59:35 +0100 From: Hans Petter Selasky <hselasky@c2i.net> To: Weongyo Jeong <weongyo@freebsd.org> Cc: Perforce Change Reviews <perforce@freebsd.org> Subject: Re: PERFORCE change 157132 for review Message-ID: <200902041459.35445.hselasky@c2i.net> In-Reply-To: <200902041242.n14CgXj2014958@repoman.freebsd.org> References: <200902041242.n14CgXj2014958@repoman.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wednesday 04 February 2009, Weongyo Jeong wrote: > =A0=A0=A0=A0=A0=A0=A0=A0bzero(desc, sizeof(struct uath_tx_desc)); > =A0=A0=A0=A0=A0=A0=A0=A0desc->msglen =3D htobe32(sizeof(struct uath_tx_de= sc)); > -=A0=A0=A0=A0=A0=A0=A0desc->msgid =A0=3D data_idx + 1;=A0=A0=A0=A0/* don'= t care about endianness */ > +=A0=A0=A0=A0=A0=A0=A0desc->msgid =A0=3D (sc->sc_msgid++) + 1;=A0=A0=A0= =A0/* don't care about > endianness */ desc->type =A0 =3D htobe32(WDCMSG_FLUSH); > =A0=A0=A0=A0=A0=A0=A0=A0desc->txqid =A0=3D htobe32(0); > =A0=A0=A0=A0=A0=A0=A0=A0desc->connid =3D htobe32(0); Why not allocate an mbuf to hold the "descriptor" data. Then you use the=20 m_next field of the header mbuf to link with the "data" mbuf ? Then you onl= y=20 need one [mbuf] queue for TX ? You can look at if_ural2.c and if_rum2.c=20 in /sys/dev/usb2/wlan . =2D-HPS
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200902041459.35445.hselasky>