[ Upstream commit 2a2a459eeeff48640dc557548ce576d666ab06ed ] napi->skb is allocated in napi_get_frags() using netdev_alloc_skb_ip_align(), with a reserve of NET_SKB_PAD + NET_IP_ALIGN bytes. However, when such skb is recycled in napi_reuse_skb(), it ends with a reserve of NET_IP_ALIGN which is suboptimal. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 5ca3b72c5da47d95b83857b768def6172fbc080a ] Shlomo Pongratz reported GRO L2 header check was suited for Ethernet only, and failed on IB/ipoib traffic. He provided a patch faking a zeroed header to let GRO aggregates frames. Roland Dreier, Herbert Xu, and others suggested we change GRO L2 header check to be more generic, ie not assuming L2 header is 14 bytes, but taking into account hard_header_len. __napi_gro_receive() has special handling for the common case (Ethernet) to avoid a memcmp() call and use an inline optimized function instead. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Reported-by: Shlomo Pongratz <shlomop@mellanox.com> Cc: Roland Dreier <roland@kernel.org> Cc: Or Gerlitz <ogerlitz@mellanox.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Tested-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Within nested statements, the break statement terminates only the do, for, switch, or while statement that immediately encloses it, So replace the break with goto. Signed-off-by: RongQing.Li <roy.qing.li@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
I just hit this during my testing. Isn't there another bug lurking?
BUG kmalloc-8: Redzone overwritten
INFO: 0xc0000000de9dec48-0xc0000000de9dec4b. First byte 0x0 instead of 0xcc
INFO: Allocated in .__seq_open_private+0x30/0xa0 age=0 cpu=5 pid=3896
.__kmalloc+0x1e0/0x2d0
.__seq_open_private+0x30/0xa0
.seq_open_net+0x60/0xe0
.dev_mc_seq_open+0x4c/0x70
.proc_reg_open+0xd8/0x260
.__dentry_open.clone.11+0x2b8/0x400
.do_last+0xf4/0x950
.path_openat+0xf8/0x480
.do_filp_open+0x48/0xc0
.do_sys_open+0x140/0x250
syscall_exit+0x0/0x40
dev_mc_seq_ops uses dev_seq_start/next/stop but only allocates
sizeof(struct seq_net_private) of private data, whereas it expects
sizeof(struct dev_iter_state):
struct dev_iter_state {
struct seq_net_private p;
unsigned int pos; /* bucket << BUCKET_SPACE + offset */
};
Create dev_seq_open_ops and use it so we don't have to expose
struct dev_iter_state.
[ Problem added by commit f04565ddf52e4 (dev: use name hash for
dev_seq_ops) -Eric ]
Signed-off-by: Anton Blanchard <anton@samba.org>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit 2425717b27eb (net: allow vlan traffic to be received under bond)
broke ARP processing on vlan on top of bonding.
+-------+
eth0 --| bond0 |---bond0.103
eth1 --| |
+-------+
52870.115435: skb_gro_reset_offset <-napi_gro_receive
52870.115435: dev_gro_receive <-napi_gro_receive
52870.115435: napi_skb_finish <-napi_gro_receive
52870.115435: netif_receive_skb <-napi_skb_finish
52870.115435: get_rps_cpu <-netif_receive_skb
52870.115435: __netif_receive_skb <-netif_receive_skb
52870.115436: vlan_do_receive <-__netif_receive_skb
52870.115436: bond_handle_frame <-__netif_receive_skb
52870.115436: vlan_do_receive <-__netif_receive_skb
52870.115436: arp_rcv <-__netif_receive_skb
52870.115436: kfree_skb <-arp_rcv
Packet is dropped in arp_rcv() because its pkt_type was set to
PACKET_OTHERHOST in the first vlan_do_receive() call, since no eth0.103
exists.
We really need to change pkt_type only if no more rx_handler is about to
be called for the packet.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Reviewed-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1745 commits) dp83640: free packet queues on remove dp83640: use proper function to free transmit time stamping packets ipv6: Do not use routes from locally generated RAs |PATCH net-next] tg3: add tx_dropped counter be2net: don't create multiple RX/TX rings in multi channel mode be2net: don't create multiple TXQs in BE2 be2net: refactor VF setup/teardown code into be_vf_setup/clear() be2net: add vlan/rx-mode/flow-control config to be_setup() net_sched: cls_flow: use skb_header_pointer() ipv4: avoid useless call of the function check_peer_pmtu TCP: remove TCP_DEBUG net: Fix driver name for mdio-gpio.c ipv4: tcp: fix TOS value in ACK messages sent from TIME_WAIT rtnetlink: Add missing manual netlink notification in dev_change_net_namespaces ipv4: fix ipsec forward performance regression jme: fix irq storm after suspend/resume route: fix ICMP redirect validation net: hold sock reference while processing tx timestamps tcp: md5: add more const attributes Add ethtool -g support to virtio_net ... Fix up conflicts in: - drivers/net/Kconfig: The split-up generated a trivial conflict with removal of a stale reference to Documentation/networking/net-modules.txt. Remove it from the new location instead. - fs/sysfs/dir.c: Fairly nasty conflicts with the sysfs rb-tree usage, conflicting with Eric Biederman's changes for tagged directories.
* 'driver-core-next' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core: (38 commits) mm: memory hotplug: Check if pages are correctly reserved on a per-section basis Revert "memory hotplug: Correct page reservation checking" Update email address for stable patch submission dynamic_debug: fix undefined reference to `__netdev_printk' dynamic_debug: use a single printk() to emit messages dynamic_debug: remove num_enabled accounting dynamic_debug: consolidate repetitive struct _ddebug descriptor definitions uio: Support physical addresses >32 bits on 32-bit systems sysfs: add unsigned long cast to prevent compile warning drivers: base: print rejected matches with DEBUG_DRIVER memory hotplug: Correct page reservation checking memory hotplug: Refuse to add unaligned memory regions remove the messy code file Documentation/zh_CN/SubmitChecklist ARM: mxc: convert device creation to use platform_device_register_full new helper to create platform devices with dma mask docs/driver-model: Update device class docs docs/driver-model: Document device.groups kobj_uevent: Ignore if some listeners cannot handle message dynamic_debug: make netif_dbg() call __netdev_printk() dynamic_debug: make netdev_dbg() call __netdev_printk() ...
Renato Westphal noticed that since commit a2835763e130c343ace5320c20d33c281e7097b7 "rtnetlink: handle rtnl_link netlink notifications manually" was merged we no longer send a netlink message when a networking device is moved from one network namespace to another. Fix this by adding the missing manual notification in dev_change_net_namespaces. Since all network devices that are processed by dev_change_net_namspaces are in the initialized state the complicated tests that guard the manual rtmsg_ifinfo calls in rollback_registered and register_netdevice are unnecessary and we can just perform a plain notification. Cc: stable@kernel.org Tested-by: Renato Westphal <renatowestphal@gmail.com> Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of using the dev->next chain and trying to resync at each call to dev_seq_start, use the name hash, keeping the bucket and the offset in seq->private field. Tests revealed the following results for ifconfig > /dev/null * 1000 interfaces: * 0.114s without patch * 0.089s with patch * 3000 interfaces: * 0.489s without patch * 0.110s with patch * 5000 interfaces: * 1.363s without patch * 0.250s with patch * 128000 interfaces (other setup): * ~100s without patch * ~30s with patch Signed-off-by: Mihai Maruseac <mmaruseac@ixiacom.com> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>