kernel/linux.git/include/linux/tcp.h, branch v6.12.107

tcp: reorganize tcp_sock_write_txrx group for variables later

2026-08-23T12:22:45+00:00

[ Upstream commit c3426ba2ed6942fe33c75bf17fc7513ba2c6ac64 ] Use the first 3-byte hole at the beginning of the tcp_sock_write_txrx group for 'noneagle'/'rate_app_limited' to fill in the existing hole in later patches. Therefore, the group size of tcp_sock_write_txrx is reduced from 92 + 4 to 91 + 4. In addition, the group size of tcp_sock_write_rx is changed to 96 to fit in the pahole outcome. Below are the trimmed pahole outcomes before and after this patch: [BEFORE THIS PATCH] struct tcp_sock { [...] __cacheline_group_begin__tcp_sock_write_txrx[0]; /* 2521 0 */ /* XXX 3 bytes hole, try to pack */ [...] struct tcp_options_received rx_opt; /* 2588 24 */ u8 nonagle:4; /* 2612: 0 1 */ u8 rate_app_limited:1; /* 2612: 4 1 */ /* XXX 3 bits hole, try to pack */ __cacheline_group_end__tcp_sock_write_txrx[0]; /* 2613 0 */ /* XXX 3 bytes hole, try to pack */ __cacheline_group_begin__tcp_sock_write_rx[0] __attribute__((__aligned__(8))); /* 2616 0 */ [...] __cacheline_group_end__tcp_sock_write_rx[0]; /* 2712 0 */ [...] /* size: 3200, cachelines: 50, members: 161 */ } [AFTER THIS PATCH] struct tcp_sock { [...] __cacheline_group_begin__tcp_sock_write_txrx[0]; /* 2521 0 */ u8 nonagle:4; /* 2521: 0 1 */ u8 rate_app_limited:1; /* 2521: 4 1 */ /* XXX 3 bits hole, try to pack */ /* XXX 2 bytes hole, try to pack */ [...] struct tcp_options_received rx_opt; /* 2588 24 */ __cacheline_group_end__tcp_sock_write_txrx[0]; /* 2612 0 */ /* XXX 4 bytes hole, try to pack */ __cacheline_group_begin__tcp_sock_write_rx[0] __attribute__((__aligned__(8))); /* 2616 0 */ [...] __cacheline_group_end__tcp_sock_write_rx[0]; /* 2712 0 */ [...] /* size: 3200, cachelines: 50, members: 161 */ } Signed-off-by: Chia-Yu Chang Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20250911110642.87529-4-chia-yu.chang@nokia-bell-labs.com Signed-off-by: Jakub Kicinski Stable-dep-of: a28c4fcbf774 ("tcp: challenge ACK for non-exact RST in SYN-RECEIVED") Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

tcp: add receive queue awareness in tcp_rcv_space_adjust()

2025-06-27T10:11:30+00:00

[ Upstream commit ea33537d82921e71f852ea2ed985acc562125efe ] If the application can not drain fast enough a TCP socket queue, tcp_rcv_space_adjust() can overestimate tp->rcvq_space.space. Then sk->sk_rcvbuf can grow and hit tcp_rmem[2] for no good reason. Fix this by taking into acount the number of available bytes. Keeping sk->sk_rcvbuf at the right size allows better cache efficiency. Signed-off-by: Eric Dumazet Reviewed-by: Wei Wang Link: https://patch.msgid.link/20250513193919.1089692-5-edumazet@google.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

tcp: more struct tcp_sock adjustments

2024-04-08T10:49:02+00:00

tp->recvmsg_inq is used from tcp recvmsg() thus should be in tcp_sock_read_rx group. tp->tcp_clock_cache and tp->tcp_mstamp are written both in rx and tx paths, thus are better placed in tcp_sock_write_txrx group. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

tcp: align tcp_sock_write_rx group

2024-03-05T04:46:18+00:00

Stephen Rothwell and kernel test robot reported that some arches (parisc, hexagon) and/or compilers would not like blamed commit. Lets make sure tcp_sock_write_rx group does not start with a hole. While we are at it, correct tcp_sock_write_tx CACHELINE_ASSERT_GROUP_SIZE() since after the blamed commit, we went to 105 bytes. Fixes: 99123622050f ("tcp: remove some holes in struct tcp_sock") Reported-by: Stephen Rothwell Reported-by: kernel test robot Link: https://lore.kernel.org/netdev/20240301121108.5d39e4f9@canb.auug.org.au/ Closes: https://lore.kernel.org/oe-kbuild-all/202403011451.csPYOS3C-lkp@intel.com/ Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Tested-by: Simon Horman # build-tested Link: https://lore.kernel.org/r/20240301171945.2958176-1-edumazet@google.com Signed-off-by: Jakub Kicinski

tcp: remove some holes in struct tcp_sock

2024-02-29T04:21:24+00:00

By moving some fields around, this patch shrinks holes size from 56 to 32, saving 24 bytes on 64bit arches. After the patch pahole gives the following for 'struct tcp_sock': /* size: 2304, cachelines: 36, members: 162 */ /* sum members: 2234, holes: 6, sum holes: 32 */ /* sum bitfield members: 34 bits, bit holes: 5, sum bit holes: 14 bits */ /* padding: 32 */ /* paddings: 3, sum paddings: 10 */ /* forced alignments: 1, forced holes: 1, sum forced holes: 12 */ Signed-off-by: Eric Dumazet Reviewed-by: Jiri Pirko Link: https://lore.kernel.org/r/20240227192721.3558982-1-edumazet@google.com Signed-off-by: Jakub Kicinski

tcp: move tp->tcp_usec_ts to tcp_sock_read_txrx group

2024-02-12T09:51:26+00:00

tp->tcp_usec_ts is a read mostly field, used in rx and tx fast paths. Fixes: d5fed5addb2b ("tcp: reorganize tcp_sock fast path variables") Signed-off-by: Eric Dumazet Cc: Coco Li Cc: Wei Wang Reviewed-by: Simon Horman Signed-off-by: David S. Miller

tcp: move tp->scaling_ratio to tcp_sock_read_txrx group

2024-02-12T09:51:26+00:00

tp->scaling_ratio is a read mostly field, used in rx and tx fast paths. Fixes: d5fed5addb2b ("tcp: reorganize tcp_sock fast path variables") Signed-off-by: Eric Dumazet Cc: Coco Li Cc: Wei Wang Reviewed-by: Simon Horman Signed-off-by: David S. Miller

Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

2023-12-08T01:53:17+00:00

Cross-merge networking fixes after downstream PR. Conflicts: drivers/net/ethernet/stmicro/stmmac/dwmac5.c drivers/net/ethernet/stmicro/stmmac/dwmac5.h drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c drivers/net/ethernet/stmicro/stmmac/hwif.h 37e4b8df27bc ("net: stmmac: fix FPE events losing") c3f3b97238f6 ("net: stmmac: Refactor EST implementation") https://lore.kernel.org/all/20231206110306.01e91114@canb.auug.org.au/ Adjacent changes: net/ipv4/tcp_ao.c 9396c4ee93f9 ("net/tcp: Don't store TCP-AO maclen on reqsk") 7b0f570f879a ("tcp: Move TCP-AO bits from cookie_v[46]_check() to tcp_ao_syncookie().") Signed-off-by: Jakub Kicinski

net/tcp: Don't store TCP-AO maclen on reqsk

2023-12-06T11:36:56+00:00

This extra check doesn't work for a handshake when SYN segment has (current_key.maclen != rnext_key.maclen). It could be amended to preserve rnext_key.maclen instead of current_key.maclen, but that requires a lookup on listen socket. Originally, this extra maclen check was introduced just because it was cheap. Drop it and convert tcp_request_sock::maclen into boolean tcp_request_sock::used_tcp_ao. Fixes: 06b22ef29591 ("net/tcp: Wire TCP-AO to request sockets") Signed-off-by: Dmitry Safonov Reviewed-by: Eric Dumazet Signed-off-by: Paolo Abeni

tcp: reorganize tcp_sock fast path variables

2023-12-06T04:16:05+00:00

The variables are organized according in the following way: - TX read-mostly hotpath cache lines - TXRX read-mostly hotpath cache lines - RX read-mostly hotpath cache lines - TX read-write hotpath cache line - TXRX read-write hotpath cache line - RX read-write hotpath cache line Fastpath cachelines end after rcvq_space. Cache line boundaries are enforced only between read-mostly and read-write. That is, if read-mostly tx cachelines bleed into read-mostly txrx cachelines, we do not care. We care about the boundaries between read and write cachelines because we want to prevent false sharing. Fast path variables span cache lines before change: 12 Fast path variables span cache lines after change: 8 Suggested-by: Eric Dumazet Reviewed-by: Wei Wang Signed-off-by: Coco Li Reviewed-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20231204201232.520025-3-lixiaoyan@google.com Signed-off-by: Jakub Kicinski