/* * netsniff-ng - the packet sniffing beast * Copyright 2009, 2010 Daniel Borkmann. * Copyright 2010 Emmanuel Roullit. * Subject to the GPL, version 2. */ #include #include #include /* for ntohs() */ #include /* for inet_ntop() */ #include "proto.h" #include "dissector_eth.h" #include "ipv6.h" #include "geoip.h" #include "pkt_buff.h" extern void ipv6(struct pkt_buff *pkt); extern void ipv6_less(struct pkt_buff *pkt); void ipv6(struct pkt_buff *pkt) { uint8_t traffic_class; uint32_t flow_label; char src_ip[INET6_ADDRSTRLEN]; char dst_ip[INET6_ADDRSTRLEN]; struct ipv6hdr *ip = (struct ipv6hdr *) pkt_pull(pkt, sizeof(*ip)); struct sockaddr_in6 sas, sad; const char *city, *region, *country; if (ip == NULL) return; traffic_class = (ip->priority << 4) | ((ip->flow_lbl[0] & 0xF0) >> 4); flow_label = ((ip->flow_lbl[0] & 0x0F) << 8) | (ip->flow_lbl[1] << 4) | ip->flow_lbl[2]; inet_ntop(AF_INET6, &ip->saddr, src_ip, sizeof(src_ip)); inet_ntop(AF_INET6, &ip->daddr, dst_ip, sizeof(dst_ip)); tprintf(" [ IPv6 "); tprintf("Addr (%s => %s), ", src_ip, dst_ip); tprintf("Version (%u), ", ip->version); tprintf("TrafficClass (%u), ", traffic_class); tprintf("FlowLabel (%u), ", flow_label); tprintf("Len (%u), ", ntohs(ip->payload_len)); tprintf("NextHdr (%u), ", ip->nexthdr); tprintf("HopLimit (%u)", ip->hop_limit); tprintf(" ]\n"); memset(&sas, 0, sizeof(sas)); sas.sin6_family = PF_INET6; memcpy(&sas.sin6_addr, &ip->saddr, sizeof(ip->saddr)); memset(&sad, 0, sizeof(sad)); sad.sin6_family = PF_INET6; memcpy(&sad.sin6_addr, &ip->daddr, sizeof(ip->daddr)); if (geoip_working()) { tprintf("\t[ Geo ("); if ((country = geoip6_country_name(&sas))) { tprintf("%s", country); if ((region = geoip6_region_name(&sas))) tprintf(" / %s", region); if ((city = geoip6_city_name(&sas))) tprintf(" / %s", city); } else { tprintf("local"); } tprintf(" => "); if ((country = geoip6_country_name(&sad))) { tprintf("%s", country); if ((region = geoip6_region_name(&sad))) tprintf(" / %s", region); if ((city = geoip6_city_name(&sad))) tprintf(" / %s", city); } else { tprintf("local"); } tprintf(") ]\n"); } pkt_set_proto(pkt, ð_lay3, ip->nexthdr); } void ipv6_less(struct pkt_buff *pkt) { char src_ip[INET6_ADDRSTRLEN]; char dst_ip[INET6_ADDRSTRLEN]; struct ipv6hdr *ip = (struct ipv6hdr *) pkt_pull(pkt, sizeof(*ip)); if (ip == NULL) return; inet_ntop(AF_INET6, &ip->saddr, src_ip, sizeof(src_ip)); inet_ntop(AF_INET6, &ip->daddr, dst_ip, sizeof(dst_ip)); tprintf(" %s/%s Len %u", src_ip, dst_ip, ntohs(ip->payload_len)); pkt_set_proto(pkt, ð_lay3, ip->nexthdr); } struct protocol ipv6_ops = { .key = 0x86DD, .print_full = ipv6, .print_less = ipv6_less, }; n' name='h' value='nds-private-remove'/>
context:
space:
mode:
authorzhong jiang <zhongjiang@huawei.com>2016-08-11 15:32:55 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2016-08-11 16:58:13 -0700
commitc1470b33bb6e18cddd361fef339ef225b8339fe7 (patch)
tree9689ae48f5c43edd37a8f472980817eb23485c17
parent85e97be32c6242c98dbbc7a241b4a78c1b93327b (diff)
mm/hugetlb: fix incorrect hugepages count during mem hotplug
When memory hotplug operates, free hugepages will be freed if the movable node is offline. Therefore, /proc/sys/vm/nr_hugepages will be incorrect. Fix it by reducing max_huge_pages when the node is offlined. n-horiguchi@ah.jp.nec.com said: : dissolve_free_huge_page intends to break a hugepage into buddy, and the : destination hugepage is supposed to be allocated from the pool of the : destination node, so the system-wide pool size is reduced. So adding : h->max_huge_pages-- makes sense to me. Link: http://lkml.kernel.org/r/1470624546-902-1-git-send-email-zhongjiang@huawei.com Signed-off-by: zhong jiang <zhongjiang@huawei.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Acked-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>