Skip to content

net: reduce tcp_memory_allocated inflation

Eric Dumazet netdev 1 version Applied

Status from this message: This series was applied to netdev/net-next.git (master)

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest7 patches 23 repliesApplied 7/7

Patches in v1 All 7 applied

#Subject and filesSizeRepliesApplied
1/7Revert "net: set SK_MEM_QUANTUM to 4096"include/net/sock.h+3 -131 replyApplied
2/7net: remove SK_MEM_QUANTUM and SK_MEM_QUANTUM_SHIFTinclude/net/sock.h net/core/sock.c net/ipv4/tcp.c +5 more+26 -281 replyApplied
3/7net: add per_cpu_fw_alloc field to struct protoinclude/net/sock.h include/net/tcp.h include/net/udp.h +11 more+43 -01 replyApplied
4/7net: implement per-cpu reserves for memory_allocatedinclude/net/sock.h+29 -98 repliesApplied
5/7net: fix sk_wmem_schedule() and sk_rmem_schedule() errorsinclude/net/sock.h+8 -41 replyApplied
6/7net: keep sk->sk_forward_alloc as small as possibleinclude/net/sock.h net/core/datagram.c net/ipv4/tcp.c +8 more+7 -729 repliesApplied
7/7net: unexport __sk_mem_{raise|reduce}_allocatednet/core/sock.c+0 -21 replyApplied

Reviews trailers per patch

All 7 patches have a Reviewed-by or Acked-by of their own; the series has one for all of them.

PatchSoheil Hassas YeganehShakeel Butt
SeriesA-b
1/7R-b
2/7R-b
3/7R-b
4/7A-b R-b
5/7R-b
6/7R-b
7/7R-b

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.

Read the v1 thread

Cover letter [PATCH net-next 0/7] net: reduce tcp_memory_allocated inflation

From: Eric Dumazet <[email protected]>

Hosts with a lot of sockets tend to hit so called TCP memory pressure, leading to very bad TCP performance and/or OOM.

The problem is that some TCP sockets can hold up to 2MB of 'forward allocations' in their per-socket cache (sk->sk_forward_alloc), and there is no mechanism to make them relinquish their share under mem pressure.
Only under some potentially rare events their share is reclaimed, one socket at a time.

In this series, I implemented a per-cpu cache instead of a per-socket one.

Each CPU has a +1/-1 MB (256 pages on x86) forward alloc cache, in order to not dirty tcp_memory_allocated shared cache line too often.

We keep sk->sk_forward_alloc values as small as possible, to meet memcg page granularity constraint.

Note that memcg already has a per-cpu cache, although MEMCG_CHARGE_BATCH is defined to 32 pages, which seems a bit small.

Note that while this cover letter mentions TCP, this work is generic and supports TCP, UDP, DECNET, SCTP.

Eric Dumazet (7):
Revert "net: set SK_MEM_QUANTUM to 4096"
net: remove SK_MEM_QUANTUM and SK_MEM_QUANTUM_SHIFT
net: add per_cpu_fw_alloc field to struct proto
net: implement per-cpu reserves for memory_allocated
net: fix sk_wmem_schedule() and sk_rmem_schedule() errors
net: keep sk->sk_forward_alloc as small as possible
net: unexport __sk_mem_{raise|reduce}_allocated

 include/net/sock.h           | 100 +++++++++++++++--------------------
 include/net/tcp.h            |   2 +
 include/net/udp.h            |   1 +
 net/core/datagram.c          |   3 --
 net/core/sock.c              |  22 ++++----
 net/decnet/af_decnet.c       |   4 ++
 net/ipv4/tcp.c               |  13 ++---
 net/ipv4/tcp_input.c         |   6 +--
 net/ipv4/tcp_ipv4.c          |   3 ++
 net/ipv4/tcp_output.c        |   2 +-
 net/ipv4/tcp_timer.c         |  19 ++-----
 net/ipv4/udp.c               |  14 +++--
 net/ipv4/udplite.c           |   3 ++
 net/ipv6/tcp_ipv6.c          |   3 ++
 net/ipv6/udp.c               |   3 ++
 net/ipv6/udplite.c           |   3 ++
 net/iucv/af_iucv.c           |   2 -
 net/mptcp/protocol.c         |  13 +++--
 net/sctp/protocol.c          |   4 +-
 net/sctp/sm_statefuns.c      |   2 -
 net/sctp/socket.c            |  12 +++--
 net/sctp/stream_interleave.c |   2 -
 net/sctp/ulpqueue.c          |   4 --
 23 files changed, 114 insertions(+), 126 deletions(-)
Signature · 1 line
2.36.1.255.ge46751e96f-goog