From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: passt.top; dmarc=none (p=none dis=none) header.from=gibson.dropbear.id.au Authentication-Results: passt.top; dkim=pass (2048-bit key; secure) header.d=gibson.dropbear.id.au header.i=@gibson.dropbear.id.au header.a=rsa-sha256 header.s=202602 header.b=k6EzTWss; dkim-atps=neutral Received: from mail.ozlabs.org (mail.ozlabs.org [IPv6:2404:9400:2221:ea00::3]) by passt.top (Postfix) with ESMTPS id 2753B5A026D for ; Fri, 10 Apr 2026 08:56:15 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gibson.dropbear.id.au; s=202602; t=1775804171; bh=/pUWYWO9cBvjKe07lPYgk480g2mWhq1TC+p77/o536c=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=k6EzTWssySgqRl9+8lAoi9PfQ/15877EvmDWkQzBfARmTY6N4TGCyw7PP5MFYYW4w aqT1WDWjXL2TwAWuj8mU4Z1kcOydZ8aWwS1FPd6J9TJ6yB23gHRN4rnW8oGzb31BUz 3BdqkVGKGZTGcb4vqHbb9pe82GuUZguw7vDFUeUXmignXLl88UhehK0L98uAZDsQNx tBDNQqmv4gn3I9WixMooyHpbPwJvjVyl1qzyb9vUXQwPt4qvkhwQA7OEYgl5hMegCk IHH3R+FKq4tqKtPCQ3XO3wMzWM83unZ7a0rpcBPr5QllbNa4Uq0sRjwNHbHyF5Y4Gk uXi/n2LyiQqDw== Received: by gandalf.ozlabs.org (Postfix, from userid 1007) id 4fsSJC4QLCz4wLn; Fri, 10 Apr 2026 16:56:11 +1000 (AEST) Date: Fri, 10 Apr 2026 16:56:05 +1000 From: David Gibson To: Laurent Vivier Subject: Re: [PATCH v2 04/10] udp_vu: Move virtqueue management from udp_vu_sock_recv() to its caller Message-ID: References: <20260403163811.3209635-1-lvivier@redhat.com> <20260403163811.3209635-5-lvivier@redhat.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha512; protocol="application/pgp-signature"; boundary="vSMK2pFi53FMoQsv" Content-Disposition: inline In-Reply-To: <20260403163811.3209635-5-lvivier@redhat.com> Message-ID-Hash: UKBSR2OREXEFDHOL2IJ3XG5QMLIL57SW X-Message-ID-Hash: UKBSR2OREXEFDHOL2IJ3XG5QMLIL57SW X-MailFrom: dgibson@gandalf.ozlabs.org X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; emergency; loop; banned-address; member-moderation; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header CC: passt-dev@passt.top X-Mailman-Version: 3.3.8 Precedence: list List-Id: Development discussion and patches for passt Archived-At: Archived-At: List-Archive: List-Archive: List-Help: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: --vSMK2pFi53FMoQsv Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Apr 03, 2026 at 06:38:05PM +0200, Laurent Vivier wrote: > udp_vu_sock_recv() currently mixes two concerns: receiving data from the > socket and managing virtqueue buffers (collecting, rewinding, releasing). > This makes the function harder to reason about and couples socket I/O > with virtqueue state. >=20 > Move all virtqueue operations, vu_collect(), vu_init_elem(), > vu_queue_rewind(), vu_set_vnethdr(), and the queue-readiness check, into > udp_vu_sock_to_tap(), which is the only caller. This turns > udp_vu_sock_recv() into a pure socket receive function that simply reads > into the provided iov array and adjusts its length. >=20 > Signed-off-by: Laurent Vivier Reviewed-by: David Gibson Minor clarity note, only worth addressing if you respin anyway. > --- > udp_vu.c | 97 ++++++++++++++++++++++++++++---------------------------- > 1 file changed, 49 insertions(+), 48 deletions(-) >=20 > diff --git a/udp_vu.c b/udp_vu.c > index f8629af58ab5..34f39e1256f8 100644 > --- a/udp_vu.c > +++ b/udp_vu.c > @@ -58,46 +58,22 @@ static size_t udp_vu_hdrlen(bool v6) > =20 > /** > * udp_vu_sock_recv() - Receive datagrams from socket into vhost-user bu= ffers > - * @c: Execution context > - * @vq: virtqueue to use to receive data > * @s: Socket to receive from > * @v6: Set for IPv6 connections > - * @dlen: Size of received data (output) > + * @iov_cnt: Number of collected iov in iov_vu (input) > + * Number of iov entries used to store the datagram (output) Nit: might be worth clarifying that *@iov_cnt is unchanged on failure. > * > - * Return: number of iov entries used to store the datagram, 0 if the da= tagram > - * was discarded because the virtqueue is not ready, -1 on error > + * Return: size of received data, -1 on error > */ > -static int udp_vu_sock_recv(const struct ctx *c, struct vu_virtq *vq, in= t s, > - bool v6, ssize_t *dlen) > +static ssize_t udp_vu_sock_recv(int s, bool v6, size_t *iov_cnt) > { > - const struct vu_dev *vdev =3D c->vdev; > - int elem_cnt, elem_used, iov_used; > struct msghdr msg =3D { 0 }; > size_t hdrlen, l2len; > - size_t iov_cnt; > - > - assert(!c->no_udp); > - > - if (!vu_queue_enabled(vq) || !vu_queue_started(vq)) { > - debug("Got UDP packet, but RX virtqueue not usable yet"); > - > - if (recvmsg(s, &msg, MSG_DONTWAIT) < 0) > - debug_perror("Failed to discard datagram"); > - > - return 0; > - } > + ssize_t dlen; > =20 > /* compute L2 header length */ > hdrlen =3D udp_vu_hdrlen(v6); > =20 > - elem_cnt =3D vu_collect(vdev, vq, elem, ARRAY_SIZE(elem), > - iov_vu, ARRAY_SIZE(iov_vu), &iov_cnt, > - IP_MAX_MTU + ETH_HLEN + VNET_HLEN, NULL); > - if (elem_cnt =3D=3D 0) > - return -1; > - > - assert((size_t)elem_cnt =3D=3D iov_cnt); /* one iovec per element */ > - > /* reserve space for the headers */ > assert(iov_vu[0].iov_len >=3D MAX(hdrlen, ETH_ZLEN + VNET_HLEN)); > iov_vu[0].iov_base =3D (char *)iov_vu[0].iov_base + hdrlen; > @@ -105,29 +81,23 @@ static int udp_vu_sock_recv(const struct ctx *c, str= uct vu_virtq *vq, int s, > =20 > /* read data from the socket */ > msg.msg_iov =3D iov_vu; > - msg.msg_iovlen =3D iov_cnt; > + msg.msg_iovlen =3D *iov_cnt; > =20 > - *dlen =3D recvmsg(s, &msg, 0); > - if (*dlen < 0) { > - vu_queue_rewind(vq, elem_cnt); > + dlen =3D recvmsg(s, &msg, 0); > + if (dlen < 0) > return -1; > - } > =20 > /* restore the pointer to the headers address */ > iov_vu[0].iov_base =3D (char *)iov_vu[0].iov_base - hdrlen; > iov_vu[0].iov_len +=3D hdrlen; > =20 > - iov_used =3D iov_truncate(iov_vu, iov_cnt, *dlen + hdrlen); > - elem_used =3D iov_used; /* one iovec per element */ > + *iov_cnt =3D iov_truncate(iov_vu, *iov_cnt, dlen + hdrlen); > =20 > /* pad frame to 60 bytes: first buffer is at least ETH_ZLEN long */ > - l2len =3D *dlen + hdrlen - VNET_HLEN; > + l2len =3D dlen + hdrlen - VNET_HLEN; > vu_pad(&iov_vu[0], l2len); > =20 > - /* release unused buffers */ > - vu_queue_rewind(vq, elem_cnt - elem_used); > - > - return iov_used; > + return dlen; > } > =20 > /** > @@ -213,21 +183,52 @@ void udp_vu_sock_to_tap(const struct ctx *c, int s,= int n, flow_sidx_t tosidx) > struct vu_virtq *vq =3D &vdev->vq[VHOST_USER_RX_QUEUE]; > int i; > =20 > + assert(!c->no_udp); > + > + if (!vu_queue_enabled(vq) || !vu_queue_started(vq)) { > + struct msghdr msg =3D { 0 }; > + > + debug("Got UDP packet, but RX virtqueue not usable yet"); > + > + for (i =3D 0; i < n; i++) { > + if (recvmsg(s, &msg, MSG_DONTWAIT) < 0) > + debug_perror("Failed to discard datagram"); > + } > + > + return; > + } > + > for (i =3D 0; i < n; i++) { > + unsigned elem_cnt, elem_used; > + size_t iov_cnt; > ssize_t dlen; > - int iov_used; > =20 > - iov_used =3D udp_vu_sock_recv(c, vq, s, v6, &dlen); > - if (iov_used < 0) > + elem_cnt =3D vu_collect(vdev, vq, elem, ARRAY_SIZE(elem), > + iov_vu, ARRAY_SIZE(iov_vu), &iov_cnt, > + IP_MAX_MTU + ETH_HLEN + VNET_HLEN, NULL); > + if (elem_cnt =3D=3D 0) > + break; > + > + assert((size_t)elem_cnt =3D=3D iov_cnt); /* one iovec per element */ > + > + dlen =3D udp_vu_sock_recv(s, v6, &iov_cnt); > + if (dlen < 0) { > + vu_queue_rewind(vq, iov_cnt); > break; > + } > + > + elem_used =3D iov_cnt; /* one iovec per element */ > + > + /* release unused buffers */ > + vu_queue_rewind(vq, elem_cnt - elem_used); Specifically, working out why the vu_queue_rewind() is correct on both the success and failure paths requires thinking about what iov_cnt is when udp_vu_sock_recv() fails. > =20 > - if (iov_used > 0) { > + if (iov_cnt > 0) { > udp_vu_prepare(c, toside, dlen); > if (*c->pcap) { > - udp_vu_csum(toside, iov_used); > - pcap_iov(iov_vu, iov_used, VNET_HLEN); > + udp_vu_csum(toside, iov_cnt); > + pcap_iov(iov_vu, iov_cnt, VNET_HLEN); > } > - vu_flush(vdev, vq, elem, iov_used); > + vu_flush(vdev, vq, elem, iov_cnt); > vu_queue_notify(vdev, vq); > } > } > --=20 > 2.53.0 >=20 --=20 David Gibson (he or they) | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you, not the other way | around. http://www.ozlabs.org/~dgibson --vSMK2pFi53FMoQsv Content-Type: application/pgp-signature; name=signature.asc -----BEGIN PGP SIGNATURE----- iQIzBAEBCgAdFiEEO+dNsU4E3yXUXRK2zQJF27ox2GcFAmnYnwQACgkQzQJF27ox 2GeCDA//Xzey/VuWy7dOtA0UVVNVRbgTJAzakQauhqtmM/mRNyrHWsgB98mnT8Co 8fPNrXh1VbH2M+unAg3ugnDZHxF55tp/ozu/DRtCvFQWp70j+bPkl5wM2m4lCguX TbOOtED4hWL/5+AMaeo6aswFByMVPVB0oXwpZngjOomoxRlQB35G0aS1Mo9hyI72 2k0Wd/mqfGy/Xlr/lztOojn0vYLqLMHZAXEb738nwMVUQxHlMmFkoa+AHfjFB4AC wowNmso6mKFOnLs9YPsIvo5IQVubK9O4sZOpPQFaPZ7XDF78jvPXu0yhvlpSyokl CI9Z6pyuV1vZNh24YaGwkr7kraF881V4kwdZBqJRYEAs80u3klEJ7CqvuT6YSqww YgMi+y/FKeiWAn8V8FoW/oq50kMEZSgMstPM0JMJPmEZ3p6fR4Z3n+mW/k3hP20y qZSeEo/19Z3pQDw8A5LGvg1eaW6HVxmN2zLKg40NnDUKys688fLwmAVw6NRVe8T2 isOPMJh/qCCVkL1ArtwbZvEUnYf1icHzpd6RSAbeyTCCHXHF2kJqD4tz1yOVh/Aj StUlYKfE+mQpPL6SolI25ZwLBoFnOEcEDO2EGSVuEy+46dO+ngZJ+qyNZkJD19IX LrEnHNMBK4sGUMew5bKvMZz/qTKlPdYfKWmyfXHz0Oqh4GpiEaQ= =DOVR -----END PGP SIGNATURE----- --vSMK2pFi53FMoQsv--