Re: ppoll() stuck on POLLIN while TCP peer is sending

From: Eric Dumazet
Date: Tue Jan 08 2013 - 21:54:31 EST


On Tue, 2013-01-08 at 18:32 -0800, Eric Dumazet wrote:

>
> Hmm, it seems sk_filter() can return -ENOMEM because skb has the
> pfmemalloc() set.

>
> One TCP socket keeps retransmitting an SKB via loopback, and TCP stack
> drops the packet again and again.

sock_init_data() sets sk->sk_allocation to GFP_KERNEL

Shouldnt it use (GFP_KERNEL | __GFP_NOMEMALLOC) instead ?



diff --git a/net/core/sock.c b/net/core/sock.c
index bc131d4..76c4b39 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -286,6 +286,7 @@ void sk_set_memalloc(struct sock *sk)
{
sock_set_flag(sk, SOCK_MEMALLOC);
sk->sk_allocation |= __GFP_MEMALLOC;
+ sk->sk_allocation &= ~__GFP_NOMEMALLOC;
static_key_slow_inc(&memalloc_socks);
}
EXPORT_SYMBOL_GPL(sk_set_memalloc);
@@ -294,6 +295,7 @@ void sk_clear_memalloc(struct sock *sk)
{
sock_reset_flag(sk, SOCK_MEMALLOC);
sk->sk_allocation &= ~__GFP_MEMALLOC;
+ sk->sk_allocation |= __GFP_NOMEMALLOC;
static_key_slow_dec(&memalloc_socks);

/*
@@ -2230,7 +2232,7 @@ void sock_init_data(struct socket *sock, struct sock *sk)

init_timer(&sk->sk_timer);

- sk->sk_allocation = GFP_KERNEL;
+ sk->sk_allocation = GFP_KERNEL | __GFP_NOMEMALLOC;
sk->sk_rcvbuf = sysctl_rmem_default;
sk->sk_sndbuf = sysctl_wmem_default;
sk->sk_state = TCP_CLOSE;


--
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/