* [PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring
@ 2020-12-16 22:56 Victor Stewart
2020-12-16 22:57 ` Soheil Hassas Yeganeh
2020-12-17 15:49 ` Jens Axboe
0 siblings, 2 replies; 6+ messages in thread
From: Victor Stewart @ 2020-12-16 22:56 UTC (permalink / raw)
To: io-uring, soheil, netdev; +Cc: Victor Stewart
This patch adds PROTO_CMSG_DATA_ONLY to inet_dgram_ops and inet6_dgram_ops so that UDP_SEGMENT (GSO) and UDP_GRO can be used through io_uring.
GSO and GRO are vital to bring QUIC servers on par with TCP throughputs, and together offer a higher
throughput gain than io_uring alone (rate of data transit
considering), thus io_uring is presently the lesser performance choice.
RE http://vger.kernel.org/lpc_net2018_talks/willemdebruijn-lpc2018-udpgso-paper-DRAFT-1.pdf,
GSO is about +~63% and GRO +~82%.
this patch closes that loophole.
Signed-off-by: Victor Stewart <[email protected]>
---
net/ipv4/af_inet.c | 1 +
net/ipv6/af_inet6.c | 1 +
net/socket.c | 8 +++++---
3 files changed, 7 insertions(+), 3 deletions(-)
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index b7260c8cef2e..c9fd5e7cfd6e 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1052,6 +1052,7 @@ EXPORT_SYMBOL(inet_stream_ops);
const struct proto_ops inet_dgram_ops = {
.family = PF_INET,
+ .flags = PROTO_CMSG_DATA_ONLY,
.owner = THIS_MODULE,
.release = inet_release,
.bind = inet_bind,
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index e648fbebb167..560f45009d06 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -695,6 +695,7 @@ const struct proto_ops inet6_stream_ops = {
const struct proto_ops inet6_dgram_ops = {
.family = PF_INET6,
+ .flags = PROTO_CMSG_DATA_ONLY,
.owner = THIS_MODULE,
.release = inet6_release,
.bind = inet6_bind,
diff --git a/net/socket.c b/net/socket.c
index 6e6cccc2104f..6995835d6355 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -2416,9 +2416,11 @@ static int ___sys_sendmsg(struct socket *sock, struct user_msghdr __user *msg,
long __sys_sendmsg_sock(struct socket *sock, struct msghdr *msg,
unsigned int flags)
{
- /* disallow ancillary data requests from this path */
- if (msg->msg_control || msg->msg_controllen)
- return -EINVAL;
+ if (msg->msg_control || msg->msg_controllen) {
+ /* disallow ancillary data reqs unless cmsg is plain data */
+ if (!(sock->ops->flags & PROTO_CMSG_DATA_ONLY))
+ return -EINVAL;
+ }
return ____sys_sendmsg(sock, msg, flags, NULL, 0);
}
--
2.26.2
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring
2020-12-16 22:56 [PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring Victor Stewart
@ 2020-12-16 22:57 ` Soheil Hassas Yeganeh
2020-12-17 15:49 ` Jens Axboe
1 sibling, 0 replies; 6+ messages in thread
From: Soheil Hassas Yeganeh @ 2020-12-16 22:57 UTC (permalink / raw)
To: Victor Stewart; +Cc: io-uring, netdev, Jann Horn
On Wed, Dec 16, 2020 at 5:56 PM Victor Stewart <[email protected]> wrote:
>
> This patch adds PROTO_CMSG_DATA_ONLY to inet_dgram_ops and inet6_dgram_ops so that UDP_SEGMENT (GSO) and UDP_GRO can be used through io_uring.
>
> GSO and GRO are vital to bring QUIC servers on par with TCP throughputs, and together offer a higher
> throughput gain than io_uring alone (rate of data transit
> considering), thus io_uring is presently the lesser performance choice.
>
> RE http://vger.kernel.org/lpc_net2018_talks/willemdebruijn-lpc2018-udpgso-paper-DRAFT-1.pdf,
> GSO is about +~63% and GRO +~82%.
>
> this patch closes that loophole.
>
> Signed-off-by: Victor Stewart <[email protected]>
Acked-by: Soheil Hassas Yeganeh <[email protected]>
Thanks for adding this! I audied the code and couldn't find an
escalation path. +Jann could you please double check?
> ---
> net/ipv4/af_inet.c | 1 +
> net/ipv6/af_inet6.c | 1 +
> net/socket.c | 8 +++++---
> 3 files changed, 7 insertions(+), 3 deletions(-)
>
> diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
> index b7260c8cef2e..c9fd5e7cfd6e 100644
> --- a/net/ipv4/af_inet.c
> +++ b/net/ipv4/af_inet.c
> @@ -1052,6 +1052,7 @@ EXPORT_SYMBOL(inet_stream_ops);
>
> const struct proto_ops inet_dgram_ops = {
> .family = PF_INET,
> + .flags = PROTO_CMSG_DATA_ONLY,
> .owner = THIS_MODULE,
> .release = inet_release,
> .bind = inet_bind,
> diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
> index e648fbebb167..560f45009d06 100644
> --- a/net/ipv6/af_inet6.c
> +++ b/net/ipv6/af_inet6.c
> @@ -695,6 +695,7 @@ const struct proto_ops inet6_stream_ops = {
>
> const struct proto_ops inet6_dgram_ops = {
> .family = PF_INET6,
> + .flags = PROTO_CMSG_DATA_ONLY,
> .owner = THIS_MODULE,
> .release = inet6_release,
> .bind = inet6_bind,
> diff --git a/net/socket.c b/net/socket.c
> index 6e6cccc2104f..6995835d6355 100644
> --- a/net/socket.c
> +++ b/net/socket.c
> @@ -2416,9 +2416,11 @@ static int ___sys_sendmsg(struct socket *sock, struct user_msghdr __user *msg,
> long __sys_sendmsg_sock(struct socket *sock, struct msghdr *msg,
> unsigned int flags)
> {
> - /* disallow ancillary data requests from this path */
> - if (msg->msg_control || msg->msg_controllen)
> - return -EINVAL;
> + if (msg->msg_control || msg->msg_controllen) {
> + /* disallow ancillary data reqs unless cmsg is plain data */
> + if (!(sock->ops->flags & PROTO_CMSG_DATA_ONLY))
> + return -EINVAL;
> + }
>
> return ____sys_sendmsg(sock, msg, flags, NULL, 0);
> }
> --
> 2.26.2
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring
2020-12-16 22:56 [PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring Victor Stewart
2020-12-16 22:57 ` Soheil Hassas Yeganeh
@ 2020-12-17 15:49 ` Jens Axboe
2020-12-17 18:30 ` Victor Stewart
1 sibling, 1 reply; 6+ messages in thread
From: Jens Axboe @ 2020-12-17 15:49 UTC (permalink / raw)
To: Victor Stewart, io-uring, soheil, netdev
On 12/16/20 3:56 PM, Victor Stewart wrote:
> This patch adds PROTO_CMSG_DATA_ONLY to inet_dgram_ops and inet6_dgram_ops so that UDP_SEGMENT (GSO) and UDP_GRO can be used through io_uring.
>
> GSO and GRO are vital to bring QUIC servers on par with TCP throughputs, and together offer a higher
> throughput gain than io_uring alone (rate of data transit
> considering), thus io_uring is presently the lesser performance choice.
>
> RE http://vger.kernel.org/lpc_net2018_talks/willemdebruijn-lpc2018-udpgso-paper-DRAFT-1.pdf,
> GSO is about +~63% and GRO +~82%.
>
> this patch closes that loophole.
LGTM
Acked-by: Jens Axboe <[email protected]>
--
Jens Axboe
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring
2020-12-17 15:49 ` Jens Axboe
@ 2020-12-17 18:30 ` Victor Stewart
2020-12-17 18:45 ` Jens Axboe
0 siblings, 1 reply; 6+ messages in thread
From: Victor Stewart @ 2020-12-17 18:30 UTC (permalink / raw)
To: Jens Axboe; +Cc: io-uring, Soheil Hassas Yeganeh, netdev
might this still make it into 5.11?
On Thu, Dec 17, 2020 at 3:49 PM Jens Axboe <[email protected]> wrote:
>
> On 12/16/20 3:56 PM, Victor Stewart wrote:
> > This patch adds PROTO_CMSG_DATA_ONLY to inet_dgram_ops and inet6_dgram_ops so that UDP_SEGMENT (GSO) and UDP_GRO can be used through io_uring.
> >
> > GSO and GRO are vital to bring QUIC servers on par with TCP throughputs, and together offer a higher
> > throughput gain than io_uring alone (rate of data transit
> > considering), thus io_uring is presently the lesser performance choice.
> >
> > RE http://vger.kernel.org/lpc_net2018_talks/willemdebruijn-lpc2018-udpgso-paper-DRAFT-1.pdf,
> > GSO is about +~63% and GRO +~82%.
> >
> > this patch closes that loophole.
>
> LGTM
>
> Acked-by: Jens Axboe <[email protected]>
>
> --
> Jens Axboe
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring
2020-12-17 18:30 ` Victor Stewart
@ 2020-12-17 18:45 ` Jens Axboe
2021-02-17 23:44 ` Victor Stewart
0 siblings, 1 reply; 6+ messages in thread
From: Jens Axboe @ 2020-12-17 18:45 UTC (permalink / raw)
To: Victor Stewart; +Cc: io-uring, Soheil Hassas Yeganeh, netdev
On 12/17/20 11:30 AM, Victor Stewart wrote:
> might this still make it into 5.11?
Doesn't meet the criteria to go in at this point. I sometimes
make exceptions, but generally speaking, something going into
5.11 should have been completed at least a week ago.
So I'd feel more comfortable pushing this to 5.12.
--
Jens Axboe
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring
2020-12-17 18:45 ` Jens Axboe
@ 2021-02-17 23:44 ` Victor Stewart
0 siblings, 0 replies; 6+ messages in thread
From: Victor Stewart @ 2021-02-17 23:44 UTC (permalink / raw)
To: Jens Axboe; +Cc: io-uring, Soheil Hassas Yeganeh, netdev
On Thu, Dec 17, 2020 at 1:45 PM Jens Axboe <[email protected]> wrote:
>
> On 12/17/20 11:30 AM, Victor Stewart wrote:
> > might this still make it into 5.11?
>
> Doesn't meet the criteria to go in at this point. I sometimes
> make exceptions, but generally speaking, something going into
> 5.11 should have been completed at least a week ago.
>
> So I'd feel more comfortable pushing this to 5.12.
ping. are we still looking at 5.12 for this?
>
> --
> Jens Axboe
>
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2021-02-17 23:46 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2020-12-16 22:56 [PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring Victor Stewart
2020-12-16 22:57 ` Soheil Hassas Yeganeh
2020-12-17 15:49 ` Jens Axboe
2020-12-17 18:30 ` Victor Stewart
2020-12-17 18:45 ` Jens Axboe
2021-02-17 23:44 ` Victor Stewart
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox