[MPTCP] [PATCH 0/3] mptcp: refactor xmit path

[MPTCP] [PATCH 0/3] mptcp: refactor xmit path

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 1109 bytes --]

this series refactor the mptcp xmit path trying to make sendmsg behavior
more consistent, memory efficent and improve its performances.

To support MP_JOIN and multiple subflows, the idea is adding, with later
patches, an RB-tree structure containing references to the pending page
fragments, their length and the associated mptcp write_seq number (is subflow
rel_write_seq needed, too?). RB-tree node can be allocated using the same
page frag used in this series for skb data. Each RB-tree node will own a
reference to the page frag.

v1  -> v2:
 - rebased on top of current git
 - fix handling of some corner cases

RFC -> v1:
 - addressed Mat's feedback, dropping patch 3/4 and switching
   to per msk page frag use, and checking mptcp seq number for coalescing

Paolo Abeni (3):
  mptcp: use sk_page_frag() in sendmsg
  mptcp: sendmsg() do spool all the provided data
  mptcp: allow collapsing consecutive sendpages on the same substream

 net/mptcp/protocol.c | 190 ++++++++++++++++++++++++++++---------------
 1 file changed, 123 insertions(+), 67 deletions(-)

-- 
2.20.1

Re: [MPTCP] [PATCH 0/3] mptcp: refactor xmit path

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 378 bytes --]

On Tue, 2019-05-07 at 15:15 +0200, Paolo Abeni wrote:
> this series refactor the mptcp xmit path trying to make sendmsg behavior
> more consistent, memory efficent and improve its performances.

Self-nack, I just found a bunch of issues testing with latest self-
tests (userland of test-bench was lagging a bit behind the current
git).

Sorry for the noise,

Paolo

Re: [MPTCP] [PATCH 0/3] mptcp: refactor xmit path

[Mat Martineau]

[-- Attachment #1: Type: text/plain, Size: 1488 bytes --]


On Mon, 15 Apr 2019, Paolo Abeni wrote:

> his series refactor the mptcp xmit path trying to make sendmsg behavior
> more consistent, memory efficent and improve its performances.
>
> The code is also available here:
>
> https://github.com/pabeni/mptcp/tree/mptcp-proposal-devel
>
> To support MP_JOIN and multiple subflows, the idea is adding, with later
> patches, an RB-tree structure containing references to the pending page
> fragments, their length and the associated mptcp write_seq number (is subflow
> rel_write_seq needed, too?). RB-tree node can be allocated using the same
> page frag used in this series for skb data. Each RB-tree node will own a
> reference to the page frag.
>
> RFC -> v1:
> - addressed Mat's feedback, dropping patch 3/4 and switching
>   to per msk page frag use, and checking mptcp seq number for coalescing
>
> Paolo Abeni (3):
>  mptcp: use sk_page_frag() in sendmsg
>  mptcp: sendmsg() do spool all the provided data
>  mptcp: allow collapsing consecutive sendpages on the same substream
>
> net/mptcp/protocol.c | 181 ++++++++++++++++++++++++++++---------------
> 1 file changed, 119 insertions(+), 62 deletions(-)
>
> -- 
> 2.20.1

Paolo -

Thanks for the updated patch set. I had a few comments on patch 3, and the 
rest look good. I'll wait on applying them for now, I think you said 
you're checking on a possible bug. If I should go ahead and apply them, 
please let me know.

--
Mat Martineau
Intel

[MPTCP] [PATCH 0/3] mptcp: refactor xmit path

[Paolo Abeni]

[-- Attachment #1: Type: text/plain, Size: 1119 bytes --]

his series refactor the mptcp xmit path trying to make sendmsg behavior
more consistent, memory efficent and improve its performances.

The code is also available here:

https://github.com/pabeni/mptcp/tree/mptcp-proposal-devel

To support MP_JOIN and multiple subflows, the idea is adding, with later
patches, an RB-tree structure containing references to the pending page
fragments, their length and the associated mptcp write_seq number (is subflow
rel_write_seq needed, too?). RB-tree node can be allocated using the same
page frag used in this series for skb data. Each RB-tree node will own a
reference to the page frag.

RFC -> v1:
 - addressed Mat's feedback, dropping patch 3/4 and switching
   to per msk page frag use, and checking mptcp seq number for coalescing

Paolo Abeni (3):
  mptcp: use sk_page_frag() in sendmsg
  mptcp: sendmsg() do spool all the provided data
  mptcp: allow collapsing consecutive sendpages on the same substream

 net/mptcp/protocol.c | 181 ++++++++++++++++++++++++++++---------------
 1 file changed, 119 insertions(+), 62 deletions(-)

-- 
2.20.1