diff options
| author | Denis V. Lunev <den@openvz.org> | 2008-02-29 11:17:11 -0800 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2008-02-29 11:17:11 -0800 |
| commit | 79c911595390c8fdc8d8a487ac1951d854b1cd09 (patch) | |
| tree | 9688e154b321f6c976b2d43abbfd4bb929b18e1a /net/ipv6/icmp.c | |
| parent | 405666db84b984b68fc75794069f424c02e5796c (diff) | |
| download | linux-79c911595390c8fdc8d8a487ac1951d854b1cd09.tar.bz2 | |
[ICMP]: Allocate data for __icmp(v6)_sk dynamically.
Own __icmp(v6)_sk should be present in each namespace. So, it should be
allocated dynamically. Though, alloc_percpu does not fit the case as it
implies additional dereferrence for no bonus.
Allocate data for pointers just like __percpu_alloc_mask does and place
pointers to struct sock into this array.
Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv6/icmp.c')
| -rw-r--r-- | net/ipv6/icmp.c | 14 |
1 files changed, 9 insertions, 5 deletions
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 18f220a04486..3368f32fe226 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -80,8 +80,8 @@ EXPORT_SYMBOL(icmpv6msg_statistics); * * On SMP we have one ICMP socket per-cpu. */ -static DEFINE_PER_CPU(struct sock *, __icmpv6_sk) = NULL; -#define icmpv6_sk __get_cpu_var(__icmpv6_sk) +static struct sock **__icmpv6_sk = NULL; +#define icmpv6_sk (__icmpv6_sk[smp_processor_id()]) static int icmpv6_rcv(struct sk_buff *skb); @@ -785,6 +785,10 @@ int __init icmpv6_init(void) struct sock *sk; int err, i, j; + __icmpv6_sk = kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL); + if (__icmpv6_sk == NULL) + return -ENOMEM; + for_each_possible_cpu(i) { struct socket *sock; err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6, @@ -797,7 +801,7 @@ int __init icmpv6_init(void) goto fail; } - per_cpu(__icmpv6_sk, i) = sk = sock->sk; + __icmpv6_sk[i] = sk = sock->sk; sk->sk_allocation = GFP_ATOMIC; /* * Split off their lock-class, because sk->sk_dst_lock @@ -830,7 +834,7 @@ int __init icmpv6_init(void) for (j = 0; j < i; j++) { if (!cpu_possible(j)) continue; - sock_release(per_cpu(__icmpv6_sk, j)->sk_socket); + sock_release(__icmpv6_sk[j]->sk_socket); } return err; @@ -841,7 +845,7 @@ void icmpv6_cleanup(void) int i; for_each_possible_cpu(i) { - sock_release(per_cpu(__icmpv6_sk, i)->sk_socket); + sock_release(__icmpv6_sk[i]->sk_socket); } inet6_del_protocol(&icmpv6_protocol, IPPROTO_ICMPV6); } |