[PATCH 20/31] netvm: prevent a stream specific deadlock

From: Suresh Jayaraman
Date: Thu Oct 01 2009 - 10:07:41 EST


From: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>

It could happen that all !SOCK_MEMALLOC sockets have buffered so much data
that we're over the global rmem limit. This will prevent SOCK_MEMALLOC buffers
from receiving data, which will prevent userspace from running, which is needed
to reduce the buffered data.

Fix this by exempting the SOCK_MEMALLOC sockets from the rmem limit.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
Signed-off-by: Suresh Jayaraman <sjayaraman@xxxxxxx>
---
include/net/sock.h | 7 ++++---
net/core/sock.c | 2 +-
net/ipv4/tcp_input.c | 12 ++++++------
net/sctp/ulpevent.c | 2 +-
4 files changed, 12 insertions(+), 11 deletions(-)

Index: mmotm/include/net/sock.h
===================================================================
--- mmotm.orig/include/net/sock.h
+++ mmotm/include/net/sock.h
@@ -882,12 +882,13 @@ static inline int sk_wmem_schedule(struc
__sk_mem_schedule(sk, size, SK_MEM_SEND);
}

-static inline int sk_rmem_schedule(struct sock *sk, int size)
+static inline int sk_rmem_schedule(struct sock *sk, struct sk_buff *skb)
{
if (!sk_has_account(sk))
return 1;
- return size <= sk->sk_forward_alloc ||
- __sk_mem_schedule(sk, size, SK_MEM_RECV);
+ return skb->truesize <= sk->sk_forward_alloc ||
+ __sk_mem_schedule(sk, skb->truesize, SK_MEM_RECV) ||
+ skb_emergency(skb);
}

static inline void sk_mem_reclaim(struct sock *sk)
Index: mmotm/net/core/sock.c
===================================================================
--- mmotm.orig/net/core/sock.c
+++ mmotm/net/core/sock.c
@@ -390,7 +390,7 @@ int sock_queue_rcv_skb(struct sock *sk,
if (err)
goto out;

- if (!sk_rmem_schedule(sk, skb->truesize)) {
+ if (!sk_rmem_schedule(sk, skb)) {
err = -ENOBUFS;
goto out;
}
Index: mmotm/net/ipv4/tcp_input.c
===================================================================
--- mmotm.orig/net/ipv4/tcp_input.c
+++ mmotm/net/ipv4/tcp_input.c
@@ -4269,19 +4269,19 @@ static void tcp_ofo_queue(struct sock *s
static int tcp_prune_ofo_queue(struct sock *sk);
static int tcp_prune_queue(struct sock *sk);

-static inline int tcp_try_rmem_schedule(struct sock *sk, unsigned int size)
+static inline int tcp_try_rmem_schedule(struct sock *sk, struct sk_buff *skb)
{
if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf ||
- !sk_rmem_schedule(sk, size)) {
+ !sk_rmem_schedule(sk, skb)) {

if (tcp_prune_queue(sk) < 0)
return -1;

- if (!sk_rmem_schedule(sk, size)) {
+ if (!sk_rmem_schedule(sk, skb)) {
if (!tcp_prune_ofo_queue(sk))
return -1;

- if (!sk_rmem_schedule(sk, size))
+ if (!sk_rmem_schedule(sk, skb))
return -1;
}
}
@@ -4333,7 +4333,7 @@ static void tcp_data_queue(struct sock *
if (eaten <= 0) {
queue_and_out:
if (eaten < 0 &&
- tcp_try_rmem_schedule(sk, skb->truesize))
+ tcp_try_rmem_schedule(sk, skb))
goto drop;

skb_set_owner_r(skb, sk);
@@ -4404,7 +4404,7 @@ drop:

TCP_ECN_check_ce(tp, skb);

- if (tcp_try_rmem_schedule(sk, skb->truesize))
+ if (tcp_try_rmem_schedule(sk, skb))
goto drop;

/* Disable header prediction. */
Index: mmotm/net/sctp/ulpevent.c
===================================================================
--- mmotm.orig/net/sctp/ulpevent.c
+++ mmotm/net/sctp/ulpevent.c
@@ -701,7 +701,7 @@ struct sctp_ulpevent *sctp_ulpevent_make
if (rx_count >= asoc->base.sk->sk_rcvbuf) {

if ((asoc->base.sk->sk_userlocks & SOCK_RCVBUF_LOCK) ||
- (!sk_rmem_schedule(asoc->base.sk, chunk->skb->truesize)))
+ (!sk_rmem_schedule(asoc->base.sk, chunk->skb)))
goto fail;
}

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/