Re: [PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring

On Thu, Dec 17, 2020 at 1:45 PM Jens Axboe 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 in

Re: [PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring

might this still make it into 5.11? On Thu, Dec 17, 2020 at 3:49 PM Jens Axboe 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_

Re: [PATCH net-next v4] udp:allow UDP cmsghdrs through io_uring

https://github.com/0day-ci/linux/commits/Victor-Stewart/udp-allow-UDP-cmsghdrs-through-io_uring/20201217-020451 > base: https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git > 3db1a3fa98808aa90f95ec3e0fa2fc7abf28f5c9 > config: riscv-randconfig-r031-20201216 (attached as .config

[PATCH net-next v5] udp:allow UDP cmsghdrs through io_uring

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 --- net/ipv4/af_inet.c

[PATCH net-next v4] udp:allow UDP cmsghdrs through io_uring

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 tr

[PATCH net-next v4] udp:allow UDP cmsghdrs through io_uring

Signed-off-by: Victor Stewart --- 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

[PATCH v3] Allow UDP cmsghdrs through io_uring

--- 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 +1

[PATCH v3] Allow UDP cmsghdrs through io_uring

here we go, figured it out. sorry about that. 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 throu

Re: [PATCH] allow UDP cmsghdrs through io_uring

the header didn't come through on my email, but i see it in the archive. almost got this git send-email thing working lol. the patch got a different subject line for some reason. On Sun, Dec 13, 2020 at 8:31 PM Victor Stewart wrote: > > --- > net/ipv4/af_inet.c | 1 + > net/ip

[PATCH v2] Allow UDP cmsghdrs through io_uring

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 tr

[PATCH] allow UDP cmsghdrs through io_uring

--- 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 +1

Re: [PATCH 0/3] PROTO_CMSG_DATA_ONLY for Datagram (UDP)

FYI for anyone who happens upon this... for gmail you have to first turn on 2-factor authentication then generate a custom app password for this to work. then use that password, all the rest the same. On Sat, Dec 12, 2020 at 9:44 PM Jens Axboe wrote: > > On 12/12/20 2:42 PM, Victor S

Re: [PATCH 0/3] PROTO_CMSG_DATA_ONLY for Datagram (UDP)

On Sat, Dec 12, 2020 at 6:02 PM Jens Axboe wrote: > > On 12/12/20 10:58 AM, Victor Stewart wrote: > > On Sat, Dec 12, 2020 at 5:40 PM Jens Axboe wrote: > >> > >> On 12/12/20 10:25 AM, Victor Stewart wrote: > >>> On Sat, Dec 12, 2020 at 5:07 PM Jens Axboe

Re: [PATCH 0/3] PROTO_CMSG_DATA_ONLY for Datagram (UDP)

On Sat, Dec 12, 2020 at 5:40 PM Jens Axboe wrote: > > On 12/12/20 10:25 AM, Victor Stewart wrote: > > On Sat, Dec 12, 2020 at 5:07 PM Jens Axboe wrote: > >> > >> On 12/12/20 8:31 AM, Victor Stewart wrote: > >>> RE our conversation on the "[RF

Re: [PATCH 0/3] PROTO_CMSG_DATA_ONLY for Datagram (UDP)

On Sat, Dec 12, 2020 at 5:07 PM Jens Axboe wrote: > > On 12/12/20 8:31 AM, Victor Stewart wrote: > > RE our conversation on the "[RFC 0/1] whitelisting UDP GSO and GRO > > cmsgs" thread... > > > > https://lore.kernel.org/io-uring/CAM1kxwi5m6i8hrtkw7nZYoziP

[PATCH 3/3] add PROTO_CMSG_DATA_ONLY to inet6_dgram_ops

add PROTO_CMSG_DATA_ONLY to inet6_dgram_ops Signed-off by: Victor Stewart --- net/ipv6/af_inet6.c | 1 + 1 file changed, 1 insertion(+) 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

[PATCH 1/3] add PROTO_CMSG_DATA_ONLY to __sys_sendmsg_sock

add PROTO_CMSG_DATA_ONLY whitelisting to __sys_sendmsg_sock Signed-off by: Victor Stewart --- net/socket.c | 8 +--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/net/socket.c b/net/socket.c index 6e62104f..6995835d6355 100644 --- a/net/socket.c +++ b/net/socket.c

[PATCH 2/3] add PROTO_CMSG_DATA_ONLY to inet_dgram_ops

add PROTO_CMSG_DATA_ONLY to inet_dgram_ops Signed-off by: Victor Stewart --- net/ipv4/af_inet.c | 1 + 1 file changed, 1 insertion(+) 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

[PATCH 0/3] PROTO_CMSG_DATA_ONLY for Datagram (UDP)

RE our conversation on the "[RFC 0/1] whitelisting UDP GSO and GRO cmsgs" thread... https://lore.kernel.org/io-uring/CAM1kxwi5m6i8hrtkw7nZYoziPTD-Wp03+fcsUwh3CuSc=81...@mail.gmail.com/ here are the patches we discussed. Victor Stewart (3): net/socket.c: add PROTO_CMSG_DA

Re: [RFC 0/1] whitelisting UDP GSO and GRO cmsgs

this being the list of UDP options.. i think we're good here? I'll put together a new patch. https://github.com/torvalds/linux/blob/b65054597872ce3aefbc6a666385eabdf9e288da/include/uapi/linux/udp.h#L30 /* UDP socket options */ #define UDP_CORK 1 /* Never send partially complete segments */ #defin

Re: MSG_ZEROCOPY_FIXED

On Wed, Nov 11, 2020 at 12:09 AM Jonathan Lemon wrote: > > On Sun, Nov 08, 2020 at 05:04:41PM +, Victor Stewart wrote: > > hi all, > > > > i'm seeking input / comment on the idea of implementing full fledged > > zerocopy UDP networking that uses persistent

MSG_ZEROCOPY_FIXED

hi all, i'm seeking input / comment on the idea of implementing full fledged zerocopy UDP networking that uses persistent buffers allocated in userspace... before I go off on a solo tangent with my first patches lol. i'm sure there's been lots of thought/discussion on this before. of course Wille