From: Sasha Levin <sashal@kernel•org>
To: linux-kernel@vger•kernel.org, stable@vger•kernel.org
Cc: Xin Long <lucien.xin@gmail•com>,
"David S . Miller" <davem@davemloft•net>,
Sasha Levin <sashal@kernel•org>,
linux-sctp@vger•kernel.org, netdev@vger•kernel.org
Subject: [PATCH AUTOSEL 4.19 079/237] sctp: use sk_wmem_queued to check for writable space
Date: Sat, 16 Nov 2019 10:38:34 -0500 [thread overview]
Message-ID: <20191116154113.7417-79-sashal@kernel.org> (raw)
In-Reply-To: <20191116154113.7417-1-sashal@kernel.org>
From: Xin Long <lucien.xin@gmail•com>
[ Upstream commit cd305c74b0f8b49748a79a8f67fc8e5e3e0c4794 ]
sk->sk_wmem_queued is used to count the size of chunks in out queue
while sk->sk_wmem_alloc is for counting the size of chunks has been
sent. sctp is increasing both of them before enqueuing the chunks,
and using sk->sk_wmem_alloc to check for writable space.
However, sk_wmem_alloc is also increased by 1 for the skb allocked
for sending in sctp_packet_transmit() but it will not wake up the
waiters when sk_wmem_alloc is decreased in this skb's destructor.
If msg size is equal to sk_sndbuf and sendmsg is waiting for sndbuf,
the check 'msg_len <= sctp_wspace(asoc)' in sctp_wait_for_sndbuf()
will keep waiting if there's a skb allocked in sctp_packet_transmit,
and later even if this skb got freed, the waiting thread will never
get waked up.
This issue has been there since very beginning, so we change to use
sk->sk_wmem_queued to check for writable space as sk_wmem_queued is
not increased for the skb allocked for sending, also as TCP does.
SOCK_SNDBUF_LOCK check is also removed here as it's for tx buf auto
tuning which I will add in another patch.
Signed-off-by: Xin Long <lucien.xin@gmail•com>
Signed-off-by: David S. Miller <davem@davemloft•net>
Signed-off-by: Sasha Levin <sashal@kernel•org>
---
net/sctp/socket.c | 38 +++++++++-----------------------------
1 file changed, 9 insertions(+), 29 deletions(-)
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index c766315527226..e7a11cd7633f5 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -83,7 +83,7 @@
#include <net/sctp/stream_sched.h>
/* Forward declarations for internal helper functions. */
-static int sctp_writeable(struct sock *sk);
+static bool sctp_writeable(struct sock *sk);
static void sctp_wfree(struct sk_buff *skb);
static int sctp_wait_for_sndbuf(struct sctp_association *asoc, long *timeo_p,
size_t msg_len);
@@ -119,25 +119,10 @@ static void sctp_enter_memory_pressure(struct sock *sk)
/* Get the sndbuf space available at the time on the association. */
static inline int sctp_wspace(struct sctp_association *asoc)
{
- int amt;
+ struct sock *sk = asoc->base.sk;
- if (asoc->ep->sndbuf_policy)
- amt = asoc->sndbuf_used;
- else
- amt = sk_wmem_alloc_get(asoc->base.sk);
-
- if (amt >= asoc->base.sk->sk_sndbuf) {
- if (asoc->base.sk->sk_userlocks & SOCK_SNDBUF_LOCK)
- amt = 0;
- else {
- amt = sk_stream_wspace(asoc->base.sk);
- if (amt < 0)
- amt = 0;
- }
- } else {
- amt = asoc->base.sk->sk_sndbuf - amt;
- }
- return amt;
+ return asoc->ep->sndbuf_policy ? sk->sk_sndbuf - asoc->sndbuf_used
+ : sk_stream_wspace(sk);
}
/* Increment the used sndbuf space count of the corresponding association by
@@ -1928,10 +1913,10 @@ static int sctp_sendmsg_to_asoc(struct sctp_association *asoc,
asoc->pmtu_pending = 0;
}
- if (sctp_wspace(asoc) < msg_len)
+ if (sctp_wspace(asoc) < (int)msg_len)
sctp_prsctp_prune(asoc, sinfo, msg_len - sctp_wspace(asoc));
- if (!sctp_wspace(asoc)) {
+ if (sctp_wspace(asoc) <= 0) {
timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT);
err = sctp_wait_for_sndbuf(asoc, &timeo, msg_len);
if (err)
@@ -8516,7 +8501,7 @@ static int sctp_wait_for_sndbuf(struct sctp_association *asoc, long *timeo_p,
goto do_error;
if (signal_pending(current))
goto do_interrupted;
- if (msg_len <= sctp_wspace(asoc))
+ if ((int)msg_len <= sctp_wspace(asoc))
break;
/* Let another process have a go. Since we are going
@@ -8591,14 +8576,9 @@ void sctp_write_space(struct sock *sk)
* UDP-style sockets or TCP-style sockets, this code should work.
* - Daisy
*/
-static int sctp_writeable(struct sock *sk)
+static bool sctp_writeable(struct sock *sk)
{
- int amt = 0;
-
- amt = sk->sk_sndbuf - sk_wmem_alloc_get(sk);
- if (amt < 0)
- amt = 0;
- return amt;
+ return sk->sk_sndbuf > sk->sk_wmem_queued;
}
/* Wait for an association to go into ESTABLISHED state. If timeout is 0,
--
2.20.1
next prev parent reply other threads:[~2019-11-16 16:24 UTC|newest]
Thread overview: 61+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20191116154113.7417-1-sashal@kernel.org>
2019-11-16 15:37 ` [PATCH AUTOSEL 4.19 016/237] mt76: do not store aggregation sequence number for null-data frames Sasha Levin
2019-11-16 15:37 ` [PATCH AUTOSEL 4.19 017/237] mt76x0: phy: fix restore phase in mt76x0_phy_recalibrate_after_assoc Sasha Levin
2019-11-16 15:37 ` [PATCH AUTOSEL 4.19 018/237] brcmsmac: AP mode: update beacon when TIM changes Sasha Levin
2019-11-16 15:37 ` [PATCH AUTOSEL 4.19 019/237] ath10k: set probe request oui during driver start Sasha Levin
2019-11-16 15:37 ` [PATCH AUTOSEL 4.19 020/237] ath10k: allocate small size dma memory in ath10k_pci_diag_write_mem Sasha Levin
2019-11-16 15:37 ` [PATCH AUTOSEL 4.19 037/237] net: ethernet: ti: cpsw: fix lost of mcast packets while rx_mode update Sasha Levin
2019-11-16 15:37 ` [PATCH AUTOSEL 4.19 044/237] qed: Align local and global PTT to propagate through the APIs Sasha Levin
2019-11-16 15:38 ` [PATCH AUTOSEL 4.19 046/237] nfp: bpf: protect against mis-initializing atomic counters Sasha Levin
2019-11-16 15:38 ` [PATCH AUTOSEL 4.19 066/237] net: dsa: mv88e6xxx: Fix 88E6141/6341 2500mbps SERDES speed Sasha Levin
2019-11-16 15:38 ` [PATCH AUTOSEL 4.19 067/237] net: fix warning in af_unix Sasha Levin
2019-11-16 15:38 ` [PATCH AUTOSEL 4.19 068/237] net: ena: Fix Kconfig dependency on X86 Sasha Levin
2019-11-16 15:38 ` Sasha Levin [this message]
2019-11-16 15:38 ` [PATCH AUTOSEL 4.19 081/237] selftests/bpf: fix file resource leak in load_kallsyms Sasha Levin
2019-11-16 15:38 ` [PATCH AUTOSEL 4.19 082/237] SUNRPC: Fix a compile warning for cmpxchg64() Sasha Levin
2019-11-16 15:38 ` [PATCH AUTOSEL 4.19 083/237] sunrpc: safely reallow resvport min/max inversion Sasha Levin
2019-11-16 15:38 ` [PATCH AUTOSEL 4.19 084/237] atm: zatm: Fix empty body Clang warnings Sasha Levin
2019-11-16 15:38 ` [PATCH AUTOSEL 4.19 096/237] selftests/bpf: fix return value comparison for tests in test_libbpf.sh Sasha Levin
2019-11-16 15:38 ` [PATCH AUTOSEL 4.19 097/237] tools: bpftool: fix completion for "bpftool map update" Sasha Levin
2019-11-16 15:38 ` [PATCH AUTOSEL 4.19 100/237] libceph: don't consume a ref on pagelist in ceph_msg_data_add_pagelist() Sasha Levin
2019-11-16 16:23 ` Ilya Dryomov
2019-11-25 14:06 ` Sasha Levin
2019-11-16 15:39 ` [PATCH AUTOSEL 4.19 106/237] mISDN: Fix type of switch control variable in ctrl_teimanager Sasha Levin
2019-11-16 15:39 ` [PATCH AUTOSEL 4.19 107/237] qlcnic: fix a return in qlcnic_dcb_get_capability() Sasha Levin
2019-11-16 15:39 ` [PATCH AUTOSEL 4.19 108/237] net: ethernet: ti: cpsw: unsync mcast entries while switch promisc mode Sasha Levin
2019-11-16 15:39 ` [PATCH AUTOSEL 4.19 113/237] net: socionext: Stop PHY before resetting netsec Sasha Levin
2019-11-16 15:39 ` [PATCH AUTOSEL 4.19 123/237] net: ethernet: cadence: fix socket buffer corruption problem Sasha Levin
2019-11-16 15:39 ` [PATCH AUTOSEL 4.19 124/237] bpf: devmap: fix wrong interface selection in notifier_call Sasha Levin
2019-11-16 15:39 ` [PATCH AUTOSEL 4.19 125/237] bpf, btf: fix a missing check bug in btf_parse Sasha Levin
2019-11-16 15:39 ` [PATCH AUTOSEL 4.19 127/237] sparc64: Rework xchg() definition to avoid warnings Sasha Levin
2019-11-16 15:39 ` [PATCH AUTOSEL 4.19 134/237] macsec: update operstate when lower device changes Sasha Levin
2019-11-16 15:39 ` [PATCH AUTOSEL 4.19 135/237] macsec: let the administrator set UP state even if lowerdev is down Sasha Levin
2019-11-16 15:39 ` [PATCH AUTOSEL 4.19 142/237] ipv4/igmp: fix v1/v2 switchback timeout based on rfc3376, 8.12 Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 166/237] igb: shorten maximum PHC timecounter update interval Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 167/237] fm10k: ensure completer aborts are marked as non-fatal after a resume Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 168/237] net: hns3: bugfix for buffer not free problem during resetting Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 169/237] net: hns3: bugfix for reporting unknown vector0 interrupt repeatly problem Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 170/237] net: hns3: bugfix for is_valid_csq_clean_head() Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 171/237] net: hns3: bugfix for hclge_mdio_write and hclge_mdio_read Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 172/237] ntb_netdev: fix sleep time mismatch Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 183/237] net: do not abort bulk send on BQL status Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 186/237] openvswitch: fix linking without CONFIG_NF_CONNTRACK_LABELS Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 191/237] sock_diag: fix autoloading of the raw_diag module Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 192/237] net: bpfilter: fix iptables failure if bpfilter_umh is disabled Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 196/237] wil6210: fix debugfs memory access alignment Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 197/237] wil6210: fix L2 RX status handling Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 198/237] wil6210: fix RGF_CAF_ICR address for Talyn-MB Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 199/237] wil6210: fix locking in wmi_call Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 200/237] ath10k: snoc: fix unbalanced clock error handling Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 201/237] wlcore: Fix the return value in case of error in 'wlcore_vendor_cmd_smart_config_start()' Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 202/237] rtl8xxxu: Fix missing break in switch Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 203/237] brcmsmac: never log "tid x is not agg'able" by default Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 204/237] wireless: airo: potential buffer overflow in sprintf() Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 205/237] rtlwifi: rtl8192de: Fix misleading REG_MCUFWDL information Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 206/237] net: dsa: bcm_sf2: Turn on PHY to allow successful registration Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 218/237] vrf: mark skb for multicast or link-local as enslaved to VRF Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 221/237] net: bcmgenet: return correct value 'ret' from bcmgenet_power_down Sasha Levin
2019-11-16 15:40 ` [PATCH AUTOSEL 4.19 222/237] sock: Reset dst when changing sk_mark via setsockopt Sasha Levin
2019-11-16 15:41 ` [PATCH AUTOSEL 4.19 225/237] tools: bpftool: pass an argument to silence open_obj_pinned() Sasha Levin
2019-11-16 15:41 ` [PATCH AUTOSEL 4.19 226/237] cfg80211: Prevent regulatory restore during STA disconnect in concurrent interfaces Sasha Levin
2019-11-16 15:41 ` [PATCH AUTOSEL 4.19 236/237] ipv6: Fix handling of LLA with VRF and sockets bound to VRF Sasha Levin
2019-11-16 15:41 ` [PATCH AUTOSEL 4.19 237/237] cfg80211: call disconnect_wk when AP stops Sasha Levin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20191116154113.7417-79-sashal@kernel.org \
--to=sashal@kernel$(echo .)org \
--cc=davem@davemloft$(echo .)net \
--cc=linux-kernel@vger$(echo .)kernel.org \
--cc=linux-sctp@vger$(echo .)kernel.org \
--cc=lucien.xin@gmail$(echo .)com \
--cc=netdev@vger$(echo .)kernel.org \
--cc=stable@vger$(echo .)kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox