Revision e4eed03fd06578571c01d4f1478c874bb432c815 authored by Andrea Arcangeli on 20 June 2012, 19:52:57 UTC, committed by Linus Torvalds on 20 June 2012, 21:39:35 UTC
In the x86 32bit PAE CONFIG_TRANSPARENT_HUGEPAGE=y case while holding the mmap_sem for reading, cmpxchg8b cannot be used to read pmd contents under Xen. So instead of dealing only with "consistent" pmdvals in pmd_none_or_trans_huge_or_clear_bad() (which would be conceptually simpler) we let pmd_none_or_trans_huge_or_clear_bad() deal with pmdvals where the low 32bit and high 32bit could be inconsistent (to avoid having to use cmpxchg8b). The only guarantee we get from pmd_read_atomic is that if the low part of the pmd was found null, the high part will be null too (so the pmd will be considered unstable). And if the low part of the pmd is found "stable" later, then it means the whole pmd was read atomically (because after a pmd is stable, neither MADV_DONTNEED nor page faults can alter it anymore, and we read the high part after the low part). In the 32bit PAE x86 case, it is enough to read the low part of the pmdval atomically to declare the pmd as "stable" and that's true for THP and no THP, furthermore in the THP case we also have a barrier() that will prevent any inconsistent pmdvals to be cached by a later re-read of the *pmd. Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Cc: Jonathan Nieder <jrnieder@gmail.com> Cc: Ulrich Obergfell <uobergfe@redhat.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Hugh Dickins <hughd@google.com> Cc: Larry Woodman <lwoodman@redhat.com> Cc: Petr Matousek <pmatouse@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Jan Beulich <jbeulich@suse.com> Cc: KOSAKI Motohiro <kosaki.motohiro@gmail.com> Tested-by: Andrew Jones <drjones@redhat.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent abca7c4
ebt_ip6.c
/*
* ebt_ip6
*
* Authors:
* Manohar Castelino <manohar.r.castelino@intel.com>
* Kuo-Lang Tseng <kuo-lang.tseng@intel.com>
* Jan Engelhardt <jengelh@medozas.de>
*
* Summary:
* This is just a modification of the IPv4 code written by
* Bart De Schuymer <bdschuym@pandora.be>
* with the changes required to support IPv6
*
* Jan, 2008
*/
#include <linux/ipv6.h>
#include <net/ipv6.h>
#include <linux/in.h>
#include <linux/module.h>
#include <net/dsfield.h>
#include <linux/netfilter/x_tables.h>
#include <linux/netfilter_bridge/ebtables.h>
#include <linux/netfilter_bridge/ebt_ip6.h>
union pkthdr {
struct {
__be16 src;
__be16 dst;
} tcpudphdr;
struct {
u8 type;
u8 code;
} icmphdr;
};
static bool
ebt_ip6_mt(const struct sk_buff *skb, struct xt_action_param *par)
{
const struct ebt_ip6_info *info = par->matchinfo;
const struct ipv6hdr *ih6;
struct ipv6hdr _ip6h;
const union pkthdr *pptr;
union pkthdr _pkthdr;
ih6 = skb_header_pointer(skb, 0, sizeof(_ip6h), &_ip6h);
if (ih6 == NULL)
return false;
if (info->bitmask & EBT_IP6_TCLASS &&
FWINV(info->tclass != ipv6_get_dsfield(ih6), EBT_IP6_TCLASS))
return false;
if (FWINV(ipv6_masked_addr_cmp(&ih6->saddr, &info->smsk,
&info->saddr), EBT_IP6_SOURCE) ||
FWINV(ipv6_masked_addr_cmp(&ih6->daddr, &info->dmsk,
&info->daddr), EBT_IP6_DEST))
return false;
if (info->bitmask & EBT_IP6_PROTO) {
uint8_t nexthdr = ih6->nexthdr;
__be16 frag_off;
int offset_ph;
offset_ph = ipv6_skip_exthdr(skb, sizeof(_ip6h), &nexthdr, &frag_off);
if (offset_ph == -1)
return false;
if (FWINV(info->protocol != nexthdr, EBT_IP6_PROTO))
return false;
if (!(info->bitmask & ( EBT_IP6_DPORT |
EBT_IP6_SPORT | EBT_IP6_ICMP6)))
return true;
/* min icmpv6 headersize is 4, so sizeof(_pkthdr) is ok. */
pptr = skb_header_pointer(skb, offset_ph, sizeof(_pkthdr),
&_pkthdr);
if (pptr == NULL)
return false;
if (info->bitmask & EBT_IP6_DPORT) {
u16 dst = ntohs(pptr->tcpudphdr.dst);
if (FWINV(dst < info->dport[0] ||
dst > info->dport[1], EBT_IP6_DPORT))
return false;
}
if (info->bitmask & EBT_IP6_SPORT) {
u16 src = ntohs(pptr->tcpudphdr.src);
if (FWINV(src < info->sport[0] ||
src > info->sport[1], EBT_IP6_SPORT))
return false;
}
if ((info->bitmask & EBT_IP6_ICMP6) &&
FWINV(pptr->icmphdr.type < info->icmpv6_type[0] ||
pptr->icmphdr.type > info->icmpv6_type[1] ||
pptr->icmphdr.code < info->icmpv6_code[0] ||
pptr->icmphdr.code > info->icmpv6_code[1],
EBT_IP6_ICMP6))
return false;
}
return true;
}
static int ebt_ip6_mt_check(const struct xt_mtchk_param *par)
{
const struct ebt_entry *e = par->entryinfo;
struct ebt_ip6_info *info = par->matchinfo;
if (e->ethproto != htons(ETH_P_IPV6) || e->invflags & EBT_IPROTO)
return -EINVAL;
if (info->bitmask & ~EBT_IP6_MASK || info->invflags & ~EBT_IP6_MASK)
return -EINVAL;
if (info->bitmask & (EBT_IP6_DPORT | EBT_IP6_SPORT)) {
if (info->invflags & EBT_IP6_PROTO)
return -EINVAL;
if (info->protocol != IPPROTO_TCP &&
info->protocol != IPPROTO_UDP &&
info->protocol != IPPROTO_UDPLITE &&
info->protocol != IPPROTO_SCTP &&
info->protocol != IPPROTO_DCCP)
return -EINVAL;
}
if (info->bitmask & EBT_IP6_DPORT && info->dport[0] > info->dport[1])
return -EINVAL;
if (info->bitmask & EBT_IP6_SPORT && info->sport[0] > info->sport[1])
return -EINVAL;
if (info->bitmask & EBT_IP6_ICMP6) {
if ((info->invflags & EBT_IP6_PROTO) ||
info->protocol != IPPROTO_ICMPV6)
return -EINVAL;
if (info->icmpv6_type[0] > info->icmpv6_type[1] ||
info->icmpv6_code[0] > info->icmpv6_code[1])
return -EINVAL;
}
return 0;
}
static struct xt_match ebt_ip6_mt_reg __read_mostly = {
.name = "ip6",
.revision = 0,
.family = NFPROTO_BRIDGE,
.match = ebt_ip6_mt,
.checkentry = ebt_ip6_mt_check,
.matchsize = sizeof(struct ebt_ip6_info),
.me = THIS_MODULE,
};
static int __init ebt_ip6_init(void)
{
return xt_register_match(&ebt_ip6_mt_reg);
}
static void __exit ebt_ip6_fini(void)
{
xt_unregister_match(&ebt_ip6_mt_reg);
}
module_init(ebt_ip6_init);
module_exit(ebt_ip6_fini);
MODULE_DESCRIPTION("Ebtables: IPv6 protocol packet match");
MODULE_AUTHOR("Kuo-Lang Tseng <kuo-lang.tseng@intel.com>");
MODULE_LICENSE("GPL");
Computing file changes ...
