4.9.8-stable review
Greg Kroah-Hartman
maintainercover 4.9.8-stable review
This is the start of the stable review cycle for the 4.9.8 release. There are 51 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Sat Feb 4 18:33:31 UTC 2017. Anything received after that time might be too late.
The whole patch series can be found in one patch at:
kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.9.8-rc1.gz
or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-4.9.y
and the diffstat can be found below.
thanks,
greg k-h
-------------
Pseudo-Shortlog of commits:
Greg Kroah-Hartman <[email protected]>
Linux 4.9.8-rc1
Darrick J. Wong <[email protected]>
xfs: fix bmv_count confusion w/ shared extents
Darrick J. Wong <[email protected]>
xfs: clear _XBF_PAGES from buffers when readahead page
Christoph Hellwig <[email protected]>
xfs: extsize hints are not unlikely in xfs_bmap_btalloc
Brian Foster <[email protected]>
xfs: remove racy hasattr check from attr ops
Darrick J. Wong <[email protected]>
xfs: verify dirblocklog correctly
Christoph Hellwig <[email protected]>
xfs: fix COW writeback race
Arnd Bergmann <[email protected]>
xfs: fix xfs_mode_to_ftype() prototype
Eric Sandeen <[email protected]>
xfs: don't wrap ID in xfs_dq_get_next_id
Amir Goldstein <[email protected]>
xfs: sanity check inode di_mode
Amir Goldstein <[email protected]>
xfs: sanity check inode mode when creating new dentry
Amir Goldstein <[email protected]>
xfs: replace xfs_mode_to_ftype table with switch statement
Amir Goldstein <[email protected]>
xfs: add missing include dependencies to xfs_dir2.h
Amir Goldstein <[email protected]>
xfs: sanity check directory inode di_size
Amir Goldstein <[email protected]>
xfs: make the ASSERT() condition likely
Christoph Hellwig <[email protected]>
xfs: don't print warnings when xfs_log_force fails
Christoph Hellwig <[email protected]>
xfs: don't rely on ->total in xfs_alloc_space_available
Christoph Hellwig <[email protected]>
xfs: adjust allocation length in xfs_alloc_space_available
Christoph Hellwig <[email protected]>
xfs: fix bogus minleft manipulations
Christoph Hellwig <[email protected]>
xfs: bump up reserved blocks in xfs_alloc_set_aside
Florian Fainelli <[email protected]>
net: dsa: Bring back device detaching in dsa_slave_suspend()
Robert Shearman <[email protected]>
lwtunnel: Fix oops on state free after encap module unload
Robert Shearman <[email protected]>
net: Specify the owning module for lwtunnel ops
Bjørn Mork <[email protected]>
qmi_wwan/cdc_ether: add device ID for HP lt2523 (Novatel E371) WWAN card
WANG Cong <[email protected]>
af_unix: move unix_mknod() out of bindlock
hayeswang <[email protected]>
r8152: don't execute runtime suspend if the tx is not empty
David Ahern <[email protected]>
net: mpls: Fix multipath selection for LSR use case
Ivan Vecera <[email protected]>
bridge: netlink: call br_changelink() during br_dev_newlink()
Eric Dumazet <[email protected]>
net/mlx5e: Do not recycle pages from emergency reserve
Alexey Kodanev <[email protected]>
tcp: initialize max window for a new fastopen socket
Kefeng Wang <[email protected]>
ipv6: addrconf: Avoid addrconf_disable_change() using RCU read-side lock
David Ahern <[email protected]>
lwtunnel: fix autoload of lwt modules
Daniel Gonzalez Cabanelas <[email protected]>
net: phy: bcm63xx: Utilize correct config_intr function
Eric Dumazet <[email protected]>
net: fix harmonize_features() vs NETIF_F_HIGHDMA
Lance Richardson <[email protected]>
vxlan: fix byte order of vxlan-gpe port number
Jason Wang <[email protected]>
virtio-net: restore VIRTIO_HDR_F_DATA_VALID on receiving
Rolf Neugebauer <[email protected]>
virtio: don't set VIRTIO_NET_HDR_F_DATA_VALID on xmit
Jamal Hadi Salim <[email protected]>
net sched actions: fix refcnt when GETing of action after bind
Basil Gunn <[email protected]>
ax25: Fix segfault after sock connection timeout
Jakub Sitnicki <[email protected]>
ip6_tunnel: Account for tunnel header in tunnel MTU
Masaru Nagai <[email protected]>
ravb: do not use zero-length alignment DMA descriptor
Eric Dumazet <[email protected]>
mlx4: do not call napi_schedule() without care
Lance Richardson <[email protected]>
openvswitch: maintain correct checksum state in conntrack actions
Shannon Nelson <[email protected]>
tcp: fix tcp_fastopen unaligned access complaints on sparc
Florian Fainelli <[email protected]>
net: systemport: Decouple flow control from __bcm_sysport_tx_reclaim
David Ahern <[email protected]>
net: ipv4: fix table id in getroute response
David Ahern <[email protected]>
net: lwtunnel: Handle lwtunnel_fill_encap failure
Elad Raz <[email protected]>
mlxsw: pci: Fix EQE structure definition
Arkadi Sharshevsky <[email protected]>
mlxsw: switchx2: Fix memory leak at skb reallocation
Arkadi Sharshevsky <[email protected]>
mlxsw: spectrum: Fix memory leak at skb reallocation
stephen hemminger <[email protected]>
netvsc: add rcu_read locking to netvsc callback
hayeswang <[email protected]>
r8152: fix the sw rx checksum is unavailable
-------------
Diffstat:
Makefile | 4 +- drivers/net/ethernet/broadcom/bcmsysport.c | 25 ++++-- drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 5 +- drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 3 + drivers/net/ethernet/mellanox/mlxsw/pci.h | 8 +- drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 1 + drivers/net/ethernet/mellanox/mlxsw/switchx2.c | 1 + drivers/net/ethernet/renesas/ravb_main.c | 13 +++ drivers/net/hyperv/netvsc_drv.c | 3 + drivers/net/macvtap.c | 2 +- drivers/net/phy/bcm63xx.c | 21 ++++- drivers/net/tun.c | 2 +- drivers/net/usb/cdc_ether.c | 8 ++ drivers/net/usb/qmi_wwan.c | 7 ++ drivers/net/usb/r8152.c | 11 ++- drivers/net/virtio_net.c | 2 +- drivers/net/vxlan.c | 2 +- fs/xfs/libxfs/xfs_alloc.c | 115 ++++++------------------ fs/xfs/libxfs/xfs_alloc.h | 2 +- fs/xfs/libxfs/xfs_attr.c | 6 -- fs/xfs/libxfs/xfs_bmap.c | 51 +++++++---- fs/xfs/libxfs/xfs_bmap.h | 6 +- fs/xfs/libxfs/xfs_bmap_btree.c | 3 +- fs/xfs/libxfs/xfs_dir2.c | 39 ++++---- fs/xfs/libxfs/xfs_dir2.h | 8 +- fs/xfs/libxfs/xfs_inode_buf.c | 10 ++- fs/xfs/libxfs/xfs_sb.c | 2 +- fs/xfs/xfs_bmap_util.c | 28 +++--- fs/xfs/xfs_buf.c | 1 + fs/xfs/xfs_dquot.c | 4 + fs/xfs/xfs_iomap.c | 2 +- fs/xfs/xfs_iops.c | 50 ++++++++--- fs/xfs/xfs_linux.h | 6 +- fs/xfs/xfs_log.c | 12 +-- include/linux/tcp.h | 7 +- include/linux/virtio_net.h | 6 +- include/net/lwtunnel.h | 18 +++- net/ax25/ax25_subr.c | 2 +- net/bridge/br_netlink.c | 33 ++++--- net/core/dev.c | 4 +- net/core/lwtunnel.c | 74 +++++++++++++-- net/dsa/slave.c | 2 + net/ipv4/fib_frontend.c | 8 ++ net/ipv4/fib_semantics.c | 11 ++- net/ipv4/ip_tunnel_core.c | 2 + net/ipv4/route.c | 2 +- net/ipv4/tcp_fastopen.c | 3 +- net/ipv6/addrconf.c | 4 +- net/ipv6/ila/ila_lwt.c | 1 + net/ipv6/ip6_tunnel.c | 4 +- net/ipv6/route.c | 15 +++- net/mpls/af_mpls.c | 48 +++++----- net/mpls/mpls_iptunnel.c | 1 + net/openvswitch/conntrack.c | 6 +- net/packet/af_packet.c | 2 +- net/sched/act_api.c | 5 +- net/unix/af_unix.c | 27 +++--- 57 files changed, 474 insertions(+), 274 deletions(-)
Greg Kroah-Hartman
maintainerPATCH 1/51 r8152: fix the sw rx checksum is unavailable
4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: hayeswang <[email protected]>
[ Upstream commit 19c0f40d4fca3a47b8f784a627f0467f0138ccc8 ]
Fix the hw rx checksum is always enabled, and the user couldn't switch it to sw rx checksum.
Note that the RTL_VER_01 only support sw rx checksum only. Besides, the hw rx checksum for RTL_VER_02 is disabled after
commit b9a321b48af4 ("r8152: Fix broken RX checksums."). Re-enable it.
--- drivers/net/usb/r8152.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/net/usb/r8152.c +6 -1
@@ -1730,7 +1730,7 @@ static u8 r8152_rx_csum(struct r8152 *tpu8 checksum =CHECKSUM_NONE ;u32 opts2, opts3; -if (tp->version ==RTL_VER_01 || tp->version ==RTL_VER_02 ) +if (!(tp->netdev->features &NETIF_F_RXCSUM )) goto return_result; opts2 =le32_to_cpu (rx_desc->opts2);
@@ -4358,6 +4358,11 @@ static int rtl8152_probe(struct usb_inteNETIF_F_HIGHDMA |NETIF_F_FRAGLIST |NETIF_F_IPV6_CSUM |NETIF_F_TSO6 ; +if (tp->version ==RTL_VER_01 ) { + netdev->features &= ~NETIF_F_RXCSUM ; + netdev->hw_features &= ~NETIF_F_RXCSUM ; + } + netdev->ethtool_ops = &ops;netif_set_gso_max_size (netdev,RTL_LIMITED_TSO_SIZE );
Greg Kroah-Hartman
maintainerPATCH 2/51 netvsc: add rcu_read locking to netvsc callback
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: stephen hemminger <[email protected]>
[ Upstream commit 0719e72ccb801829a3d735d187ca8417f0930459 ]
The receive callback (in tasklet context) is using RCU to get reference to associated VF network device but this is not safe. RCU read lock needs to be held. Found by running with full lockdep debugging enabled.
--- drivers/net/hyperv/netvsc_drv.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/net/hyperv/netvsc_drv.c +3 -0
@@ -659,6 +659,7 @@ int netvsc_recv_callback(struct hv_devic* policy filters on the host). Deliver these via the VF * interface in the guest. */ +rcu_read_lock (); vf_netdev =rcu_dereference (net_device_ctx->vf_netdev);if (vf_netdev && (vf_netdev->flags &IFF_UP )) net = vf_netdev;
@@ -667,6 +668,7 @@ int netvsc_recv_callback(struct hv_devic skb =netvsc_alloc_recv_skb (net, packet, csum_info, *data, vlan_tci);if (unlikely (!skb)) { ++net->stats.rx_dropped; +rcu_read_unlock ();return NVSP_STAT_FAIL ; }
@@ -696,6 +698,7 @@ int netvsc_recv_callback(struct hv_devic* TODO - use NAPI? */ netif_rx (skb); +rcu_read_unlock ();return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 3/51 mlxsw: spectrum: Fix memory leak at skb reallocation
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arkadi Sharshevsky <[email protected]>
[ Upstream commit 36bf38d158d3482119b3e159c0619b3c1539b508 ]
During transmission the skb is checked for headroom in order to add vendor specific header. In case the skb needs to be re-allocated, skb_realloc_headroom() is called to make a private copy of the original, but doesn't release it. Current code assumes that the original skb is released during reallocation and only releases it at the error path which causes a memory leak.
Fix this by adding the original skb release to the main path.
--- drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 1 + 1 file changed, 1 insertion(+)
drivers/net/ethernet/mellanox/mlxsw/spectrum.c +1 -0
@@ -684,6 +684,7 @@ static netdev_tx_t mlxsw_sp_port_xmit(stdev_kfree_skb_any (skb_orig);return NETDEV_TX_OK ; } +dev_consume_skb_any (skb_orig); }if (eth_skb_pad (skb)) {
Greg Kroah-Hartman
maintainerPATCH 4/51 mlxsw: switchx2: Fix memory leak at skb reallocation
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arkadi Sharshevsky <[email protected]>
[ Upstream commit 400fc0106dd8c27ed84781c929c1a184785b9c79 ]
During transmission the skb is checked for headroom in order to add vendor specific header. In case the skb needs to be re-allocated, skb_realloc_headroom() is called to make a private copy of the original, but doesn't release it. Current code assumes that the original skb is released during reallocation and only releases it at the error path which causes a memory leak.
Fix this by adding the original skb release to the main path.
--- drivers/net/ethernet/mellanox/mlxsw/switchx2.c | 1 + 1 file changed, 1 insertion(+)
drivers/net/ethernet/mellanox/mlxsw/switchx2.c +1 -0
@@ -314,6 +314,7 @@ static netdev_tx_t mlxsw_sx_port_xmit(stdev_kfree_skb_any (skb_orig);return NETDEV_TX_OK ; } +dev_consume_skb_any (skb_orig); }mlxsw_sx_txhdr_construct (skb, &tx_info);/* TX header is consumed by HW on the way so we shouldn't count its
Greg Kroah-Hartman
maintainerPATCH 5/51 mlxsw: pci: Fix EQE structure definition
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Elad Raz <[email protected]>
[ Upstream commit 28e46a0f2e03ab4ed0e23cace1ea89a68c8c115b ]
The event_data starts from address 0x00-0x0C and not from 0x08-0x014. This leads to duplication with other fields in the Event Queue Element such as sub-type, cqn and owner.
--- drivers/net/ethernet/mellanox/mlxsw/pci.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/net/ethernet/mellanox/mlxsw/pci.h +4 -4
@@ -211,21 +211,21 @@ MLXSW_ITEM32(pci, eqe, owner, 0x0C, 0, 1/* pci_eqe_cmd_token * Command completion event - token */ -MLXSW_ITEM32 (pci, eqe, cmd_token,0x08 ,16 ,16 ); +MLXSW_ITEM32 (pci, eqe, cmd_token,0x00 ,16 ,16 );/* pci_eqe_cmd_status * Command completion event - status */ -MLXSW_ITEM32 (pci, eqe, cmd_status,0x08 ,0 ,8 ); +MLXSW_ITEM32 (pci, eqe, cmd_status,0x00 ,0 ,8 );/* pci_eqe_cmd_out_param_h * Command completion event - output parameter - higher part */ -MLXSW_ITEM32 (pci, eqe, cmd_out_param_h,0x0C ,0 ,32 ); +MLXSW_ITEM32 (pci, eqe, cmd_out_param_h,0x04 ,0 ,32 );/* pci_eqe_cmd_out_param_l * Command completion event - output parameter - lower part */ -MLXSW_ITEM32 (pci, eqe, cmd_out_param_l,0x10 ,0 ,32 ); +MLXSW_ITEM32 (pci, eqe, cmd_out_param_l,0x08 ,0 ,32 );#endif
Greg Kroah-Hartman
maintainerPATCH 6/51 net: lwtunnel: Handle lwtunnel_fill_encap failure
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Ahern <[email protected]>
[ Upstream commit ea7a80858f57d8878b1499ea0f1b8a635cc48de7 ]
Handle failure in lwtunnel_fill_encap adding attributes to skb.
--- net/ipv4/fib_semantics.c | 11 +++++++---- net/ipv6/route.c | 3 ++- 2 files changed, 9 insertions(+), 5 deletions(-)
net/ipv4/fib_semantics.c +7 -4
@@ -1278,8 +1278,9 @@ int fib_dump_info(struct sk_buff *skb, u nla_put_u32(skb, RTA_FLOW, fi->fib_nh[0].nh_tclassid)) goto nla_put_failure;#endif -if (fi->fib_nh->nh_lwtstate) -lwtunnel_fill_encap (skb, fi->fib_nh->nh_lwtstate); +if (fi->fib_nh->nh_lwtstate && +lwtunnel_fill_encap (skb, fi->fib_nh->nh_lwtstate) <0 ) + goto nla_put_failure; }#ifdef CONFIG_IP_ROUTE_MULTIPATH if (fi->fib_nhs >1 ) {
@@ -1315,8 +1316,10 @@ int fib_dump_info(struct sk_buff *skb, u nla_put_u32(skb, RTA_FLOW, nh->nh_tclassid)) goto nla_put_failure;#endif -if (nh->nh_lwtstate) -lwtunnel_fill_encap (skb, nh->nh_lwtstate); +if (nh->nh_lwtstate && +lwtunnel_fill_encap (skb, nh->nh_lwtstate) <0 ) + goto nla_put_failure; +/* length of rtnetlink header + attributes */ rtnh->rtnh_len =nlmsg_get_pos (skb) - (void *) rtnh; }endfor_nexthops (fi);
net/ipv6/route.c +2 -1
@@ -3306,7 +3306,8 @@ static int rt6_fill_node(struct net *netif (nla_put_u8 (skb,RTA_PREF ,IPV6_EXTRACT_PREF (rt->rt6i_flags))) goto nla_put_failure; -lwtunnel_fill_encap (skb, rt->dst.lwtstate); +if (lwtunnel_fill_encap (skb, rt->dst.lwtstate) <0 ) + goto nla_put_failure;nlmsg_end (skb, nlh);return 0 ;
Greg Kroah-Hartman
maintainerPATCH 7/51 net: ipv4: fix table id in getroute response
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Ahern <[email protected]>
[ Upstream commit 8a430ed50bb1b19ca14a46661f3b1b35f2fb5c39 ]
rtm_table is an 8-bit field while table ids are allowed up to u32. Commit 709772e6e065 ("net: Fix routing tables with id > 255 for legacy software") added the preference to set rtm_table in dumps to RT_TABLE_COMPAT if the table id is > 255. The table id returned on get route requests should do the same.
--- net/ipv4/route.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv4/route.c +1 -1
@@ -2440,7 +2440,7 @@ static int rt_fill_info(struct net *net, r->rtm_dst_len =32 ; r->rtm_src_len =0 ; r->rtm_tos = fl4->flowi4_tos; - r->rtm_table = table_id; + r->rtm_table = table_id <256 ? table_id :RT_TABLE_COMPAT ;if (nla_put_u32 (skb,RTA_TABLE , table_id)) goto nla_put_failure; r->rtm_type = rt->rt_type;
Greg Kroah-Hartman
maintainerPATCH 8/51 net: systemport: Decouple flow control from __bcm_sysport_tx_reclaim
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Florian Fainelli <[email protected]>
[ Upstream commit 148d3d021cf9724fcf189ce4e525a094bbf5ce89 ]
The __bcm_sysport_tx_reclaim() function is used to reclaim transmit resources in different places within the driver. Most of them should not affect the state of the transit flow control.
Introduce bcm_sysport_tx_clean() which cleans the ring, but does not re-enable flow control towards the networking stack, and make bcm_sysport_tx_reclaim() do the actual transmit queue flow control.
--- drivers/net/ethernet/broadcom/bcmsysport.c | 25 ++++++++++++++++++------- 1 file changed, 18 insertions(+), 7 deletions(-)
drivers/net/ethernet/broadcom/bcmsysport.c +18 -7
@@ -710,11 +710,8 @@ static unsigned int __bcm_sysport_tx_recunsigned int c_index, last_c_index, last_tx_cn, num_tx_cbs;unsigned int pkts_compl =0 , bytes_compl =0 ;struct bcm_sysport_cb *cb; -struct netdev_queue *txq;u32 hw_ind; - txq =netdev_get_tx_queue (ndev, ring->index); -/* Compute how many descriptors have been processed since last call */ hw_ind =tdma_readl (priv,TDMA_DESC_RING_PROD_CONS_INDEX (ring->index)); c_index = (hw_ind >>RING_CONS_INDEX_SHIFT ) &RING_CONS_INDEX_MASK ;
@@ -745,9 +742,6 @@ static unsigned int __bcm_sysport_tx_rec ring->c_index = c_index; -if (netif_tx_queue_stopped (txq) && pkts_compl) -netif_tx_wake_queue (txq); -netif_dbg (priv, tx_done, ndev,"ring=%d c_index=%d pkts_compl=%d, bytes_compl=%d\n" , ring->index, ring->c_index, pkts_compl, bytes_compl);
@@ -759,16 +753,33 @@ static unsigned int __bcm_sysport_tx_recstatic unsigned int bcm_sysport_tx_reclaim (struct bcm_sysport_priv *priv,struct bcm_sysport_tx_ring *ring) { +struct netdev_queue *txq;unsigned int released;unsigned long flags; + txq =netdev_get_tx_queue (priv->netdev, ring->index); +spin_lock_irqsave (&ring->lock, flags); released =__bcm_sysport_tx_reclaim (priv, ring); +if (released) +netif_tx_wake_queue (txq); +spin_unlock_irqrestore (&ring->lock, flags);return released; } +/* Locked version of the per-ring TX reclaim, but does not wake the queue */ +static void bcm_sysport_tx_clean (struct bcm_sysport_priv *priv, +struct bcm_sysport_tx_ring *ring) +{ +unsigned long flags; + +spin_lock_irqsave (&ring->lock, flags); +__bcm_sysport_tx_reclaim (priv, ring); +spin_unlock_irqrestore (&ring->lock, flags); +} +static int bcm_sysport_tx_poll (struct napi_struct *napi,int budget) { struct bcm_sysport_tx_ring *ring =
@@ -1253,7 +1264,7 @@ static void bcm_sysport_fini_tx_ring(strnapi_disable (&ring->napi);netif_napi_del (&ring->napi); -bcm_sysport_tx_reclaim (priv, ring); +bcm_sysport_tx_clean (priv, ring);kfree (ring->cbs); ring->cbs =NULL ;
Greg Kroah-Hartman
maintainerPATCH 9/51 tcp: fix tcp_fastopen unaligned access complaints on sparc
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Shannon Nelson <[email protected]>
[ Upstream commit 003c941057eaa868ca6fedd29a274c863167230d ]
Fix up a data alignment issue on sparc by swapping the order of the cookie byte array field with the length field in struct tcp_fastopen_cookie, and making it a proper union to clean up the typecasting.
This addresses log complaints like these:
log_unaligned: 113 callbacks suppressed
Kernel unaligned access at TPC[976490] tcp_try_fastopen+0x2d0/0x360
Kernel unaligned access at TPC[9764ac] tcp_try_fastopen+0x2ec/0x360
Kernel unaligned access at TPC[9764c8] tcp_try_fastopen+0x308/0x360
Kernel unaligned access at TPC[9764e4] tcp_try_fastopen+0x324/0x360
Kernel unaligned access at TPC[976490] tcp_try_fastopen+0x2d0/0x360--- include/linux/tcp.h | 7 ++++++- net/ipv4/tcp_fastopen.c | 2 +- 2 files changed, 7 insertions(+), 2 deletions(-)
include/linux/tcp.h +6 -1
@@ -62,8 +62,13 @@ static inline unsigned int tcp_optlen(co/* TCP Fast Open Cookie as stored in memory */ struct tcp_fastopen_cookie { +union { +u8 val[TCP_FASTOPEN_COOKIE_MAX ]; +#if IS_ENABLED (CONFIG_IPV6 ) +struct in6_addr addr; +#endif + };s8 len; -u8 val[TCP_FASTOPEN_COOKIE_MAX ];bool exp;/* In RFC6994 experimental option format */ };
net/ipv4/tcp_fastopen.c +1 -1
@@ -113,7 +113,7 @@ static bool tcp_fastopen_cookie_gen(strustruct tcp_fastopen_cookie tmp;if (__tcp_fastopen_cookie_gen (&ip6h->saddr, &tmp)) { -struct in6_addr *buf = (struct in6_addr *) tmp.val; +struct in6_addr *buf = &tmp.addr;int i;for (i =0 ; i <4 ; i++)
Greg Kroah-Hartman
maintainerPATCH 10/51 openvswitch: maintain correct checksum state in conntrack actions
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Lance Richardson <[email protected]>
[ Upstream commit 75f01a4c9cc291ff5cb28ca1216adb163b7a20ee ]
When executing conntrack actions on skbuffs with checksum mode CHECKSUM_COMPLETE, the checksum must be updated to account for header pushes and pulls. Otherwise we get "hw csum failure" logs similar to this (ICMP packet received on geneve tunnel via ixgbe NIC):
[ 405.740065] genev_sys_6081: hw csum failure [ 405.740106] CPU: 3 PID: 0 Comm: swapper/3 Tainted: G I 4.10.0-rc3+ #1 [ 405.740108] Call Trace: [ 405.740110] <IRQ> [ 405.740113] dump_stack+0x63/0x87 [ 405.740116] netdev_rx_csum_fault+0x3a/0x40 [ 405.740118] __skb_checksum_complete+0xcf/0xe0 [ 405.740120] nf_ip_checksum+0xc8/0xf0 [ 405.740124] icmp_error+0x1de/0x351 [nf_conntrack_ipv4] [ 405.740132] nf_conntrack_in+0xe1/0x550 [nf_conntrack] [ 405.740137] ? find_bucket.isra.2+0x62/0x70 [openvswitch] [ 405.740143] __ovs_ct_lookup+0x95/0x980 [openvswitch] [ 405.740145] ? netif_rx_internal+0x44/0x110 [ 405.740149] ovs_ct_execute+0x147/0x4b0 [openvswitch] [ 405.740153] do_execute_actions+0x22e/0xa70 [openvswitch] [ 405.740157] ovs_execute_actions+0x40/0x120 [openvswitch] [ 405.740161] ovs_dp_process_packet+0x84/0x120 [openvswitch] [ 405.740166] ovs_vport_receive+0x73/0xd0 [openvswitch] [ 405.740168] ? udp_rcv+0x1a/0x20 [ 405.740170] ? ip_local_deliver_finish+0x93/0x1e0 [ 405.740172] ? ip_local_deliver+0x6f/0xe0 [ 405.740174] ? ip_rcv_finish+0x3a0/0x3a0 [ 405.740176] ? ip_rcv_finish+0xdb/0x3a0 [ 405.740177] ? ip_rcv+0x2a7/0x400 [ 405.740180] ? __netif_receive_skb_core+0x970/0xa00 [ 405.740185] netdev_frame_hook+0xd3/0x160 [openvswitch] [ 405.740187] __netif_receive_skb_core+0x1dc/0xa00 [ 405.740194] ? ixgbe_clean_rx_irq+0x46d/0xa20 [ixgbe] [ 405.740197] __netif_receive_skb+0x18/0x60 [ 405.740199] netif_receive_skb_internal+0x40/0xb0 [ 405.740201] napi_gro_receive+0xcd/0x120 [ 405.740204] gro_cell_poll+0x57/0x80 [geneve] [ 405.740206] net_rx_action+0x260/0x3c0 [ 405.740209] __do_softirq+0xc9/0x28c [ 405.740211] irq_exit+0xd9/0xf0 [ 405.740213] do_IRQ+0x51/0xd0 [ 405.740215] common_interrupt+0x93/0x93
--- net/openvswitch/conntrack.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
net/openvswitch/conntrack.c +4 -2
@@ -514,7 +514,7 @@ static int ovs_ct_nat_execute(struct sk_int hooknum, nh_off, err =NF_ACCEPT ; nh_off =skb_network_offset (skb); -skb_pull (skb, nh_off); +skb_pull_rcsum (skb, nh_off);/* See HOOK2MANIP(). */ if (maniptype ==NF_NAT_MANIP_SRC )
@@ -579,6 +579,7 @@ static int ovs_ct_nat_execute(struct sk_ err =nf_nat_packet (ct, ctinfo, hooknum, skb); push:skb_push (skb, nh_off); +skb_postpush_rcsum (skb, skb->data, nh_off);return err; }
@@ -890,7 +891,7 @@ int ovs_ct_execute(struct net *net, stru/* The conntrack module expects to be working at L3. */ nh_ofs =skb_network_offset (skb); -skb_pull (skb, nh_ofs); +skb_pull_rcsum (skb, nh_ofs);if (key->ip.frag !=OVS_FRAG_TYPE_NONE ) { err =handle_fragments (net, key, info->zone.id, skb);
@@ -904,6 +905,7 @@ int ovs_ct_execute(struct net *net, stru err =ovs_ct_lookup (net, key, info, skb);skb_push (skb, nh_ofs); +skb_postpush_rcsum (skb, skb->data, nh_ofs);if (err)kfree_skb (skb);return err;
Greg Kroah-Hartman
maintainerPATCH 11/51 mlx4: do not call napi_schedule() without care
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
[ Upstream commit 8cf699ec849f4ca1413cea01289bd7d37dbcc626 ]
Disable BH around the call to napi_schedule() to avoid following warning
[ 52.095499] NOHZ: local_softirq_pending 08 [ 52.421291] NOHZ: local_softirq_pending 08 [ 52.608313] NOHZ: local_softirq_pending 08
--- drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx4/en_netdev.c +4 -1
@@ -1740,8 +1740,11 @@ int mlx4_en_start_port(struct net_device/* Process all completions if exist to prevent * the queues freezing if they are full */ -for (i =0 ; i < priv->rx_ring_num; i++) +for (i =0 ; i < priv->rx_ring_num; i++) { +local_bh_disable ();napi_schedule (&priv->rx_cq[i]->napi); +local_bh_enable (); + }netif_tx_start_all_queues (dev);netif_device_attach (dev);
Greg Kroah-Hartman
maintainerPATCH 12/51 ravb: do not use zero-length alignment DMA descriptor
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Masaru Nagai <[email protected]>
[ Upstream commit 8ec3e8a192ba6f13be4522ee81227c792c86fb1a ]
Due to alignment requirements of the hardware transmissions are split into two DMA descriptors, a small padding descriptor of 0 - 3 bytes in length followed by a descriptor for rest of the packet.
In the case of IP packets the first descriptor will never be zero due to the way that the stack aligns buffers for IP packets. However, for non-IP packets it may be zero.
In that case it has been reported that timeouts occur, presumably because transmission stops at the first zero-length DMA descriptor and thus the packet is not transmitted. However, in my environment a BUG is triggered as follows:
[ 20.381417] ------------[ cut here ]------------ [ 20.386054] kernel BUG at lib/swiotlb.c:495! [ 20.390324] Internal error: Oops - BUG: 0 [#1] PREEMPT SMP [ 20.395805] Modules linked in: [ 20.398862] CPU: 0 PID: 2089 Comm: mz Not tainted 4.10.0-rc3-00001-gf13ad2db193f #162 [ 20.406689] Hardware name: Renesas Salvator-X board based on r8a7796 (DT) [ 20.413474] task: ffff80063b1f1900 task.stack: ffff80063a71c000 [ 20.419404] PC is at swiotlb_tbl_map_single+0x178/0x2ec [ 20.424625] LR is at map_single+0x4c/0x98 [ 20.428629] pc : [<ffff00000839c4c0>] lr : [<ffff00000839c680>] pstate: 800001c5 [ 20.436019] sp : ffff80063a71f9b0 [ 20.439327] x29: ffff80063a71f9b0 x28: ffff80063a20d500 [ 20.444636] x27: ffff000008ed5000 x26: 0000000000000000 [ 20.449944] x25: 000000067abe2adc x24: 0000000000000000 [ 20.455252] x23: 0000000000200000 x22: 0000000000000001 [ 20.460559] x21: 0000000000175ffe x20: ffff80063b2a0010 [ 20.465866] x19: 0000000000000000 x18: 0000ffffcae6fb20 [ 20.471173] x17: 0000ffffa09ba018 x16: ffff0000087c8b70 [ 20.476480] x15: 0000ffffa084f588 x14: 0000ffffa09cfa14 [ 20.481787] x13: 0000ffffcae87ff0 x12: 000000000063abe2 [ 20.487098] x11: ffff000008096360 x10: ffff80063abe2adc [ 20.492407] x9 : 0000000000000000 x8 : 0000000000000000 [ 20.497718] x7 : 0000000000000000 x6 : ffff000008ed50d0 [ 20.503028] x5 : 0000000000000000 x4 : 0000000000000001 [ 20.508338] x3 : 0000000000000000 x2 : 000000067abe2adc [ 20.513648] x1 : 00000000bafff000 x0 : 0000000000000000 [ 20.518958] [ 20.520446] Process mz (pid: 2089, stack limit = 0xffff80063a71c000) [ 20.526798] Stack: (0xffff80063a71f9b0 to 0xffff80063a720000) [ 20.532543] f9a0: ffff80063a71fa30 ffff00000839c680 [ 20.540374] f9c0: ffff80063b2a0010 ffff80063b2a0010 0000000000000001 0000000000000000 [ 20.548204] f9e0: 000000000000006e ffff80063b23c000 ffff80063b23c000 0000000000000000 [ 20.556034] fa00: ffff80063b23c000 ffff80063a20d500 000000013b1f1900 0000000000000000 [ 20.563864] fa20: ffff80063ffd18e0 ffff80063b2a0010 ffff80063a71fa60 ffff00000839cd10 [ 20.571694] fa40: ffff80063b2a0010 0000000000000000 ffff80063ffd18e0 000000067abe2adc [ 20.579524] fa60: ffff80063a71fa90 ffff000008096380 ffff80063b2a0010 0000000000000000 [ 20.587353] fa80: 0000000000000000 0000000000000001 ffff80063a71fac0 ffff00000864f770 [ 20.595184] faa0: ffff80063b23caf0 0000000000000000 0000000000000000 0000000000000140 [ 20.603014] fac0: ffff80063a71fb60 ffff0000087e6498 ffff80063a20d500 ffff80063b23c000 [ 20.610843] fae0: 0000000000000000 ffff000008daeaf0 0000000000000000 ffff000008daeb00 [ 20.618673] fb00: ffff80063a71fc0c ffff000008da7000 ffff80063b23c090 ffff80063a44f000 [ 20.626503] fb20: 0000000000000000 ffff000008daeb00 ffff80063a71fc0c ffff000008da7000 [ 20.634333] fb40: ffff80063b23c090 0000000000000000 ffff800600000037 ffff0000087e63d8 [ 20.642163] fb60: ffff80063a71fbc0 ffff000008807510 ffff80063a692400 ffff80063a20d500 [ 20.649993] fb80: ffff80063a44f000 ffff80063b23c000 ffff80063a69249c 0000000000000000 [ 20.657823] fba0: 0000000000000000 ffff80063a087800 ffff80063b23c000 ffff80063a20d500 [ 20.665653] fbc0: ffff80063a71fc10 ffff0000087e67dc ffff80063a20d500 ffff80063a692400 [ 20.673483] fbe0: ffff80063b23c000 0000000000000000 ffff80063a44f000 ffff80063a69249c [ 20.681312] fc00: ffff80063a5f1a10 000000103a087800 ffff80063a71fc70 ffff0000087e6b24 [ 20.689142] fc20: ffff80063a5f1a80 ffff80063a71fde8 000000000000000f 00000000000005ea [ 20.696972] fc40: ffff80063a5f1a10 0000000000000000 000000000000000f ffff00000887fbd0 [ 20.704802] fc60: fffffff43a5f1a80 0000000000000000 ffff80063a71fc80 ffff000008880240 [ 20.712632] fc80: ffff80063a71fd90 ffff0000087c7a34 ffff80063afc7180 0000000000000000 [ 20.720462] fca0: 0000ffffcae6fe18 0000000000000014 0000000060000000 0000000000000015 [ 20.728292] fcc0: 0000000000000123 00000000000000ce ffff0000088d2000 ffff80063b1f1900 [ 20.736122] fce0: 0000000000008933 ffff000008e7cb80 ffff80063a71fd80 ffff0000087c50a4 [ 20.743951] fd00: 0000000000008933 ffff000008e7cb80 ffff000008e7cb80 000000100000000e [ 20.751781] fd20: ffff80063a71fe4c 0000ffff00000300 0000000000000123 0000000000000000 [ 20.759611] fd40: 0000000000000000 ffff80063b1f0000 000000000000000e 0000000000000300 [ 20.767441] fd60: 0000000000000000 0000000000000000 0000000000000000 0000000000000000 [ 20.775271] fd80: 0000000000000000 0000000000000000 ffff80063a71fda0 ffff0000087c8c20 [ 20.783100] fda0: 0000000000000000 ffff000008082f30 0000000000000000 0000800637260000 [ 20.790930] fdc0: ffffffffffffffff 0000ffffa0903078 0000000000000000 000000001ea87232 [ 20.798760] fde0: 000000000000000f ffff80063a71fe40 ffff800600000014 ffff000000000001 [ 20.806590] fe00: 0000000000000000 0000000000000000 ffff80063a71fde8 0000000000000000 [ 20.814420] fe20: 0000000000000000 0000000000000000 0000000000000000 0000000000000001 [ 20.822249] fe40: 0000000203000011 0000000000000000 0000000000000000 ffff80063a68aa00 [ 20.830079] fe60: ffff80063a68aa00 0000000000000003 0000000000008933 ffff0000081f1b9c [ 20.837909] fe80: 0000000000000000 ffff000008082f30 0000000000000000 0000800637260000 [ 20.845739] fea0: ffffffffffffffff 0000ffffa07ca81c 0000000060000000 0000000000000015 [ 20.853569] fec0: 0000000000000003 000000001ea87232 000000000000000f 0000000000000000 [ 20.861399] fee0: 0000ffffcae6fe18 0000000000000014 0000000000000300 0000000000000000 [ 20.869228] ff00: 00000000000000ce 0000000000000000 00000000ffffffff 0000000000000000 [ 20.877059] ff20: 0000000000000002 0000ffffcae87ff0 0000ffffa09cfa14 0000ffffa084f588 [ 20.884888] ff40: 0000000000000000 0000ffffa09ba018 0000ffffcae6fb20 000000001ea87010 [ 20.892718] ff60: 0000ffffa09b9000 0000ffffcae6fe30 0000ffffcae6fe18 000000000000000f [ 20.900548] ff80: 0000000000000003 000000001ea87232 0000000000000000 0000000000000000 [ 20.908378] ffa0: 0000000000000000 0000ffffcae6fdc0 0000ffffa09a7824 0000ffffcae6fdc0 [ 20.916208] ffc0: 0000ffffa0903078 0000000060000000 0000000000000003 00000000000000ce [ 20.924038] ffe0: 0000000000000000 0000000000000000 ffffffffffffffff ffffffffffffffff [ 20.931867] Call trace: [ 20.934312] Exception stack(0xffff80063a71f7e0 to 0xffff80063a71f910) [ 20.940750] f7e0: 0000000000000000 0001000000000000 ffff80063a71f9b0 ffff00000839c4c0 [ 20.948580] f800: ffff80063a71f840 ffff00000888a6e4 ffff80063a24c418 ffff80063a24c448 [ 20.956410] f820: 0000000000000000 ffff00000811cd54 ffff80063a71f860 ffff80063a24c458 [ 20.964240] f840: ffff80063a71f870 ffff00000888b258 ffff80063a24c418 0000000000000001 [ 20.972070] f860: ffff80063a71f910 ffff80063a7b7028 ffff80063a71f890 ffff0000088825e4 [ 20.979899] f880: 0000000000000000 00000000bafff000 000000067abe2adc 0000000000000000 [ 20.987729] f8a0: 0000000000000001 0000000000000000 ffff000008ed50d0 0000000000000000 [ 20.995560] f8c0: 0000000000000000 0000000000000000 ffff80063abe2adc ffff000008096360 [ 21.003390] f8e0: 000000000063abe2 0000ffffcae87ff0 0000ffffa09cfa14 0000ffffa084f588 [ 21.011219] f900: ffff0000087c8b70 0000ffffa09ba018 [ 21.016097] [<ffff00000839c4c0>] swiotlb_tbl_map_single+0x178/0x2ec [ 21.022362] [<ffff00000839c680>] map_single+0x4c/0x98 [ 21.027411] [<ffff00000839cd10>] swiotlb_map_page+0xa4/0x138 [ 21.033072] [<ffff000008096380>] __swiotlb_map_page+0x20/0x7c [ 21.038821] [<ffff00000864f770>] ravb_start_xmit+0x174/0x668 [ 21.044484] [<ffff0000087e6498>] dev_hard_start_xmit+0x8c/0x120 [ 21.050407] [<ffff000008807510>] sch_direct_xmit+0x108/0x1a0 [ 21.056064] [<ffff0000087e67dc>] __dev_queue_xmit+0x194/0x4cc [ 21.061807] [<ffff0000087e6b24>] dev_queue_xmit+0x10/0x18 [ 21.067214] [<ffff000008880240>] packet_sendmsg+0xf40/0x1220 [ 21.072873] [<ffff0000087c7a34>] sock_sendmsg+0x18/0x2c [ 21.078097] [<ffff0000087c8c20>] SyS_sendto+0xb0/0xf0 [ 21.083150] [<ffff000008082f30>] el0_svc_naked+0x24/0x28 [ 21.088462] Code: d34bfef7 2a1803f3 1a9f86d6 35fff878 (d4210000) [ 21.094611] ---[ end trace 5bc544ad491f3814 ]--- [ 21.099234] Kernel panic - not syncing: Fatal exception in interrupt [ 21.105587] Kernel Offset: disabled [ 21.109073] Memory Limit: none [ 21.112126] ---[ end Kernel panic - not syncing: Fatal exception in interrupt
--- drivers/net/ethernet/renesas/ravb_main.c | 13 +++++++++++++ 1 file changed, 13 insertions(+)
drivers/net/ethernet/renesas/ravb_main.c +13 -0
@@ -1508,6 +1508,19 @@ static netdev_tx_t ravb_start_xmit(struc buffer =PTR_ALIGN (priv->tx_align[q],DPTR_ALIGN ) + entry /NUM_TX_DESC *DPTR_ALIGN ; len =PTR_ALIGN (skb->data,DPTR_ALIGN ) - skb->data; +/* Zero length DMA descriptors are problematic as they seem to +* terminate DMA transfers. Avoid them by simply using a length of +* DPTR_ALIGN (4) when skb data is aligned to DPTR_ALIGN. +* +* As skb is guaranteed to have at least ETH_ZLEN (60) bytes of +* data by the call to skb_put_padto() above this is safe with +* respect to both the length of the first DMA descriptor (len) +* overflowing the available data and the length of the second DMA +* descriptor (skb->len - len) being negative. +*/ +if (len ==0 ) + len =DPTR_ALIGN ; +memcpy (buffer, skb->data, len); dma_addr =dma_map_single (ndev->dev.parent, buffer, len,DMA_TO_DEVICE );if (dma_mapping_error (ndev->dev.parent, dma_addr))
Greg Kroah-Hartman
maintainerPATCH 13/51 ip6_tunnel: Account for tunnel header in tunnel MTU
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jakub Sitnicki <[email protected]>
[ Upstream commit 02ca0423fd65a0a9c4d70da0dbb8f4b8503f08c7 ]
With ip6gre we have a tunnel header which also makes the tunnel MTU smaller. We need to reserve room for it. Previously we were using up space reserved for the Tunnel Encapsulation Limit option header (RFC 2473).
Also, after commit b05229f44228 ("gre6: Cleanup GREv6 transmit path, call common GRE functions") our contract with the caller has changed. Now we check if the packet length exceeds the tunnel MTU after the tunnel header has been pushed, unlike before.
This is reflected in the check where we look at the packet length minus the size of the tunnel header, which is already accounted for in tunnel MTU.
--- net/ipv6/ip6_tunnel.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
net/ipv6/ip6_tunnel.c +2 -2
@@ -1108,7 +1108,7 @@ route_lookup: t->parms.name); goto tx_err_dst_release; } - mtu =dst_mtu (dst) - psh_hlen; + mtu =dst_mtu (dst) - psh_hlen - t->tun_hlen;if (encap_limit >=0 ) { max_headroom +=8 ; mtu -=8 ;
@@ -1117,7 +1117,7 @@ route_lookup: mtu =IPV6_MIN_MTU ;if (skb_dst (skb) && !t->parms.collect_md)skb_dst (skb)->ops->update_pmtu (skb_dst (skb),NULL , skb, mtu); -if (skb->len > mtu && !skb_is_gso (skb)) { +if (skb->len - t->tun_hlen > mtu && !skb_is_gso (skb)) { *pmtu = mtu; err = -EMSGSIZE ; goto tx_err_dst_release;
Greg Kroah-Hartman
maintainerPATCH 14/51 ax25: Fix segfault after sock connection timeout
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Basil Gunn <[email protected]>
[ Upstream commit 8a367e74c0120ef68c8c70d5a025648c96626dff ]
The ax.25 socket connection timed out & the sock struct has been previously taken down ie. sock struct is now a NULL pointer. Checking the sock_flag causes the segfault. Check if the socket struct pointer is NULL before checking sock_flag. This segfault is seen in timed out netrom connections.
Please submit to -stable.
--- net/ax25/ax25_subr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/ax25/ax25_subr.c +1 -1
@@ -264,7 +264,7 @@ void ax25_disconnect(ax25_cb *ax25, int {ax25_clear_queues (ax25); -if (!sock_flag (ax25->sk,SOCK_DESTROY )) +if (!ax25->sk || !sock_flag (ax25->sk,SOCK_DESTROY ))ax25_stop_heartbeat (ax25);ax25_stop_t1timer (ax25);ax25_stop_t2timer (ax25);
Greg Kroah-Hartman
maintainerPATCH 15/51 net sched actions: fix refcnt when GETing of action after bind
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jamal Hadi Salim <[email protected]>
[ Upstream commit 0faa9cb5b3836a979864a6357e01d2046884ad52 ]
Demonstrating the issue:
.. add a drop action
$sudo $TC actions add action drop index 10
.. retrieve it
$ sudo $TC -s actions get action gact index 10
action order 1: gact action drop
random type none pass val 0
index 10 ref 2 bind 0 installed 29 sec used 29 sec
Action statistics:
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0
... bug 1 above: reference is two.
Reference is actually 1 but we forget to subtract 1.
... do a GET again and we see the same issue
try a few times and nothing changes
~$ sudo $TC -s actions get action gact index 10
action order 1: gact action drop
random type none pass val 0
index 10 ref 2 bind 0 installed 31 sec used 31 sec
Action statistics:
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0
... lets try to bind the action to a filter..
$ sudo $TC qdisc add dev lo ingress $ sudo $TC filter add dev lo parent ffff: protocol ip prio 1 \ u32 match ip dst 127.0.0.1/32 flowid 1:1 action gact index 10
... and now a few GETs:
$ sudo $TC -s actions get action gact index 10
action order 1: gact action drop
random type none pass val 0
index 10 ref 3 bind 1 installed 204 sec used 204 sec
Action statistics:
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0
$ sudo $TC -s actions get action gact index 10
action order 1: gact action drop
random type none pass val 0
index 10 ref 4 bind 1 installed 206 sec used 206 sec
Action statistics:
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0
$ sudo $TC -s actions get action gact index 10
action order 1: gact action drop
random type none pass val 0
index 10 ref 5 bind 1 installed 235 sec used 235 sec
Action statistics:
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0
.... as can be observed the reference count keeps going up.
After the fix
$ sudo $TC actions add action drop index 10 $ sudo $TC -s actions get action gact index 10
action order 1: gact action drop
random type none pass val 0
index 10 ref 1 bind 0 installed 4 sec used 4 sec
Action statistics:
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0
$ sudo $TC -s actions get action gact index 10
action order 1: gact action drop
random type none pass val 0
index 10 ref 1 bind 0 installed 6 sec used 6 sec
Action statistics:
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0
$ sudo $TC qdisc add dev lo ingress $ sudo $TC filter add dev lo parent ffff: protocol ip prio 1 \ u32 match ip dst 127.0.0.1/32 flowid 1:1 action gact index 10
$ sudo $TC -s actions get action gact index 10
action order 1: gact action drop
random type none pass val 0
index 10 ref 2 bind 1 installed 32 sec used 32 sec
Action statistics:
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0
$ sudo $TC -s actions get action gact index 10
action order 1: gact action drop
random type none pass val 0
index 10 ref 2 bind 1 installed 33 sec used 33 sec
Action statistics:
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0
--- net/sched/act_api.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
net/sched/act_api.c +2 -3
@@ -903,8 +903,6 @@ tca_action_gd(struct net *net, struct nl goto err; } act->order = i; -if (event ==RTM_GETACTION ) - act->tcfa_refcnt++;list_add_tail (&act->list, &actions); }
@@ -917,7 +915,8 @@ tca_action_gd(struct net *net, struct nlreturn ret; } err: -tcf_action_destroy (&actions,0 ); +if (event !=RTM_GETACTION ) +tcf_action_destroy (&actions,0 );return ret; }
Greg Kroah-Hartman
maintainerPATCH 16/51 virtio: dont set VIRTIO_NET_HDR_F_DATA_VALID on xmit
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Rolf Neugebauer <[email protected]>
[ Upstream commit 501db511397fd6efff3aa5b4e8de415b55559550 ]
This patch part reverts fd2a0437dc33 and e858fae2b0b8 which introduced a subtle change in how the virtio_net flags are derived from the SKBs ip_summed field.
With the above commits, the flags are set to VIRTIO_NET_HDR_F_DATA_VALID when ip_summed == CHECKSUM_UNNECESSARY, thus treating it differently to ip_summed == CHECKSUM_NONE, which should be the same.
Further, the virtio spec 1.0 / CS04 explicitly says that VIRTIO_NET_HDR_F_DATA_VALID must not be set by the driver.
--- include/linux/virtio_net.h | 2 -- 1 file changed, 2 deletions(-)
include/linux/virtio_net.h +0 -2
@@ -91,8 +91,6 @@ static inline int virtio_net_hdr_from_sk skb_checksum_start_offset(skb)); hdr->csum_offset =__cpu_to_virtio16 (little_endian, skb->csum_offset); - }else if (skb->ip_summed ==CHECKSUM_UNNECESSARY ) { - hdr->flags =VIRTIO_NET_HDR_F_DATA_VALID ; }/* else everything is zero */ return 0 ;
Greg Kroah-Hartman
maintainerPATCH 17/51 virtio-net: restore VIRTIO_HDR_F_DATA_VALID on receiving
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jason Wang <[email protected]>
[ Upstream commit 6391a4481ba0796805d6581e42f9f0418c099e34 ]
Commit 501db511397f ("virtio: don't set VIRTIO_NET_HDR_F_DATA_VALID on xmit") in fact disables VIRTIO_HDR_F_DATA_VALID on receiving path too, fixing this by adding a hint (has_data_valid) and set it only on the receiving path.
--- drivers/net/macvtap.c | 2 +- drivers/net/tun.c | 2 +- drivers/net/virtio_net.c | 2 +- include/linux/virtio_net.h | 6 +++++- net/packet/af_packet.c | 2 +- 5 files changed, 9 insertions(+), 5 deletions(-)
drivers/net/macvtap.c +1 -1
@@ -827,7 +827,7 @@ static ssize_t macvtap_put_user(struct mreturn -EINVAL ; ret =virtio_net_hdr_from_skb (skb, &vnet_hdr, -macvtap_is_little_endian (q)); +macvtap_is_little_endian (q), true);if (ret)BUG ();
drivers/net/tun.c +1 -1
@@ -1374,7 +1374,7 @@ static ssize_t tun_put_user(struct tun_sreturn -EINVAL ; ret =virtio_net_hdr_from_skb (skb, &gso, -tun_is_little_endian (tun)); +tun_is_little_endian (tun), true);if (ret) {struct skb_shared_info *sinfo =skb_shinfo (skb); pr_err("unexpected GSO type: "
drivers/net/virtio_net.c +1 -1
@@ -840,7 +840,7 @@ static int xmit_skb(struct send_queue *s hdr =skb_vnet_hdr (skb);if (virtio_net_hdr_from_skb (skb, &hdr->hdr, -virtio_is_little_endian (vi->vdev))) +virtio_is_little_endian (vi->vdev), false))BUG ();if (vi->mergeable_rx_bufs)
include/linux/virtio_net.h +5 -1
@@ -56,7 +56,8 @@ static inline int virtio_net_hdr_to_skb(static inline int virtio_net_hdr_from_skb (const struct sk_buff *skb,struct virtio_net_hdr *hdr, -bool little_endian) +bool little_endian, +bool has_data_valid) {memset (hdr,0 ,sizeof (*hdr));
@@ -91,6 +92,9 @@ static inline int virtio_net_hdr_from_sk skb_checksum_start_offset(skb)); hdr->csum_offset =__cpu_to_virtio16 (little_endian, skb->csum_offset); + }else if (has_data_valid && + skb->ip_summed ==CHECKSUM_UNNECESSARY ) { + hdr->flags =VIRTIO_NET_HDR_F_DATA_VALID ; }/* else everything is zero */ return 0 ;
net/packet/af_packet.c +1 -1
@@ -1972,7 +1972,7 @@ static int __packet_rcv_vnet(const struc { *vnet_hdr = (const struct virtio_net_hdr ) {0 }; -if (virtio_net_hdr_from_skb (skb, vnet_hdr,vio_le ())) +if (virtio_net_hdr_from_skb (skb, vnet_hdr,vio_le (), true))BUG ();return 0 ;
Greg Kroah-Hartman
maintainerPATCH 18/51 vxlan: fix byte order of vxlan-gpe port number
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Lance Richardson <[email protected]>
[ Upstream commit d5ff72d9af73bc3cbaa3edb541333a851f8c7295 ]
vxlan->cfg.dst_port is in network byte order, so an htons() is needed here. Also reduced comment length to stay closer to 80 column width (still slightly over, however).
--- drivers/net/vxlan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/vxlan.c +1 -1
@@ -2887,7 +2887,7 @@ static int vxlan_dev_configure(struct nememcpy (&vxlan->cfg, conf,sizeof (*conf));if (!vxlan->cfg.dst_port) {if (conf->flags &VXLAN_F_GPE ) - vxlan->cfg.dst_port =4790 ;/* IANA assigned VXLAN-GPE port */ + vxlan->cfg.dst_port =htons (4790 );/* IANA VXLAN-GPE port */ else vxlan->cfg.dst_port = default_port; }
Greg Kroah-Hartman
maintainerPATCH 19/51 net: fix harmonize_features() vs NETIF_F_HIGHDMA
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
[ Upstream commit 7be2c82cfd5d28d7adb66821a992604eb6dd112e ]
Ashizuka reported a highmem oddity and sent a patch for freescale fec driver.
But the problem root cause is that core networking stack must ensure no skb with highmem fragment is ever sent through a device that does not assert NETIF_F_HIGHDMA in its features.
We need to call illegal_highdma() from harmonize_features() regardless of CSUM checks.
--- net/core/dev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
net/core/dev.c +2 -2
@@ -2815,9 +2815,9 @@ static netdev_features_t harmonize_featuif (skb->ip_summed !=CHECKSUM_NONE && !can_checksum_protocol (features, type)) { features &= ~(NETIF_F_CSUM_MASK |NETIF_F_GSO_MASK ); - }else if (illegal_highdma (skb->dev, skb)) { - features &= ~NETIF_F_SG ; } +if (illegal_highdma (skb->dev, skb)) + features &= ~NETIF_F_SG ;return features; }
Greg Kroah-Hartman
maintainerPATCH 20/51 net: phy: bcm63xx: Utilize correct config_intr function
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Daniel Gonzalez Cabanelas <[email protected]>
[ Upstream commit cd33b3e0da43522ff8e8f2b2b71d3d08298512b0 ]
Commit a1cba5613edf ("net: phy: Add Broadcom phy library for common interfaces") make the BCM63xx PHY driver utilize bcm_phy_config_intr() which would appear to do the right thing, except that it does not write to the MII_BCM63XX_IR register but to MII_BCM54XX_ECR which is different.
This would be causing invalid link parameters and events from being generated by the PHY interrupt.
--- drivers/net/phy/bcm63xx.c | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-)
drivers/net/phy/bcm63xx.c +19 -2
@@ -21,6 +21,23 @@ MODULE_DESCRIPTION("Broadcom 63xx internMODULE_AUTHOR ("Maxime Bizon <[email protected]>" );MODULE_LICENSE ("GPL" ); +static int bcm63xx_config_intr (struct phy_device *phydev) +{ +int reg, err; + + reg =phy_read (phydev,MII_BCM63XX_IR ); +if (reg <0 ) +return reg; + +if (phydev->interrupts ==PHY_INTERRUPT_ENABLED ) + reg &= ~MII_BCM63XX_IR_GMASK ; +else + reg |=MII_BCM63XX_IR_GMASK ; + + err =phy_write (phydev,MII_BCM63XX_IR , reg); +return err; +} +static int bcm63xx_config_init (struct phy_device *phydev) {int reg, err;
@@ -55,7 +72,7 @@ static struct phy_driver bcm63xx_driver[ .config_aneg = genphy_config_aneg, .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, - .config_intr = bcm_phy_config_intr, + .config_intr = bcm63xx_config_intr, }, {/* same phy as above, with just a different OUI */ .phy_id = 0x002bdc00,
@@ -67,7 +84,7 @@ static struct phy_driver bcm63xx_driver[ .config_aneg = genphy_config_aneg, .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, - .config_intr = bcm_phy_config_intr, + .config_intr = bcm63xx_config_intr, } };module_phy_driver (bcm63xx_driver);
Greg Kroah-Hartman
maintainerPATCH 21/51 lwtunnel: fix autoload of lwt modules
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Ahern <[email protected]>
[ Upstream commit 9ed59592e3e379b2e9557dc1d9e9ec8fcbb33f16]
Trying to add an mpls encap route when the MPLS modules are not loaded hangs. For example:
CONFIG_MPLS=y
CONFIG_NET_MPLS_GSO=m
CONFIG_MPLS_ROUTING=m
CONFIG_MPLS_IPTUNNEL=m
$ ip route add 10.10.10.10/32 encap mpls 100 via inet 10.100.1.2The ip command hangs:
root 880 826 0 21:25 pts/0 00:00:00 ip route add 10.10.10.10/32 encap mpls 100 via inet 10.100.1.2
$ cat /proc/880/stack
[<ffffffff81065a9b>] call_usermodehelper_exec+0xd6/0x134
[<ffffffff81065efc>] __request_module+0x27b/0x30a
[<ffffffff814542f6>] lwtunnel_build_state+0xe4/0x178
[<ffffffff814aa1e4>] fib_create_info+0x47f/0xdd4
[<ffffffff814ae451>] fib_table_insert+0x90/0x41f
[<ffffffff814a8010>] inet_rtm_newroute+0x4b/0x52
...modprobe is trying to load rtnl-lwt-MPLS:
root 881 5 0 21:25 ? 00:00:00 /sbin/modprobe -q -- rtnl-lwt-MPLS
and it hangs after loading mpls_router:
$ cat /proc/881/stack
[<ffffffff81441537>] rtnl_lock+0x12/0x14
[<ffffffff8142ca2a>] register_netdevice_notifier+0x16/0x179
[<ffffffffa0033025>] mpls_init+0x25/0x1000 [mpls_router]
[<ffffffff81000471>] do_one_initcall+0x8e/0x13f
[<ffffffff81119961>] do_init_module+0x5a/0x1e5
[<ffffffff810bd070>] load_module+0x13bd/0x17d6
...The problem is that lwtunnel_build_state is called with rtnl lock held preventing mpls_init from registering.
Given the potential references held by the time lwtunnel_build_state it can not drop the rtnl lock to the load module. So, extract the module loading code from lwtunnel_build_state into a new function to validate the encap type. The new function is called while converting the user request into a fib_config which is well before any table, device or fib entries are examined.
--- include/net/lwtunnel.h | 11 ++++++++ net/core/lwtunnel.c | 62 +++++++++++++++++++++++++++++++++++++++++++----- net/ipv4/fib_frontend.c | 8 ++++++ net/ipv6/route.c | 12 ++++++++- 4 files changed, 86 insertions(+), 7 deletions(-)
include/net/lwtunnel.h +11 -0
@@ -106,6 +106,8 @@ int lwtunnel_encap_add_ops(const struct unsigned int num);int lwtunnel_encap_del_ops (const struct lwtunnel_encap_ops *op,unsigned int num); +int lwtunnel_valid_encap_type (u16 encap_type); +int lwtunnel_valid_encap_type_attr (struct nlattr *attr,int len); int lwtunnel_build_state(struct net_device *dev, u16 encap_type, struct nlattr *encap, unsigned int family, const void *cfg,
@@ -168,6 +170,15 @@ static inline int lwtunnel_encap_del_ops {return -EOPNOTSUPP ; } + +static inline int lwtunnel_valid_encap_type (u16 encap_type) +{ +return -EOPNOTSUPP ; +} +static inline int lwtunnel_valid_encap_type_attr (struct nlattr *attr,int len) +{ +return -EOPNOTSUPP ; +} static inline int lwtunnel_build_state(struct net_device *dev, u16 encap_type, struct nlattr *encap,
net/core/lwtunnel.c +56 -6
87 lines not on this page · raw message
net/ipv4/fib_frontend.c +8 -0
@@ -46,6 +46,7 @@#include <net/rtnetlink.h> #include <net/xfrm.h> #include <net/l3mdev.h> +#include <net/lwtunnel.h> #include <trace/events/fib.h> #ifndef CONFIG_IP_MULTIPLE_TABLES
@@ -676,6 +677,10 @@ static int rtm_to_fib_config(struct net cfg->fc_mx_len =nla_len (attr);break ;case RTA_MULTIPATH : + err =lwtunnel_valid_encap_type_attr (nla_data (attr), +nla_len (attr)); +if (err <0 ) + goto errout; cfg->fc_mp =nla_data (attr); cfg->fc_mp_len =nla_len (attr);break ;
@@ -690,6 +695,9 @@ static int rtm_to_fib_config(struct netbreak ;case RTA_ENCAP_TYPE : cfg->fc_encap_type =nla_get_u16 (attr); + err =lwtunnel_valid_encap_type (cfg->fc_encap_type); +if (err <0 ) + goto errout;break ; } }
net/ipv6/route.c +11 -1
@@ -2885,6 +2885,11 @@ static int rtm_to_fib6_config(struct sk_if (tb[RTA_MULTIPATH ]) { cfg->fc_mp =nla_data (tb[RTA_MULTIPATH ]); cfg->fc_mp_len =nla_len (tb[RTA_MULTIPATH ]); + + err =lwtunnel_valid_encap_type_attr (cfg->fc_mp, + cfg->fc_mp_len); +if (err <0 ) + goto errout; }if (tb[RTA_PREF ]) {
@@ -2898,9 +2903,14 @@ static int rtm_to_fib6_config(struct sk_if (tb[RTA_ENCAP ]) cfg->fc_encap = tb[RTA_ENCAP ]; -if (tb[RTA_ENCAP_TYPE ]) +if (tb[RTA_ENCAP_TYPE ]) { cfg->fc_encap_type =nla_get_u16 (tb[RTA_ENCAP_TYPE ]); + err =lwtunnel_valid_encap_type (cfg->fc_encap_type); +if (err <0 ) + goto errout; + } +if (tb[RTA_EXPIRES ]) {unsigned long timeout =addrconf_timeout_fixup (nla_get_u32 (tb[RTA_EXPIRES ]),HZ );
Greg Kroah-Hartman
maintainerPATCH 22/51 ipv6: addrconf: Avoid addrconf_disable_change() using RCU read-side lock
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kefeng Wang <[email protected]>
[ Upstream commit 03e4deff4987f79c34112c5ba4eb195d4f9382b0 ]
Just like commit 4acd4945cd1e ("ipv6: addrconf: Avoid calling netdevice notifiers with RCU read-side lock"), it is unnecessary to make addrconf_disable_change() use RCU iteration over the netdev list, since it already holds the RTNL lock, or we may meet Illegal context switch in RCU read-side critical section.
--- net/ipv6/addrconf.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
net/ipv6/addrconf.c +1 -3
@@ -5515,8 +5515,7 @@ static void addrconf_disable_change(strustruct net_device *dev;struct inet6_dev *idev; -rcu_read_lock (); -for_each_netdev_rcu (net, dev) { +for_each_netdev (net, dev) { idev =__in6_dev_get (dev);if (idev) {int changed = (!idev->cnf.disable_ipv6) ^ (!newf);
@@ -5525,7 +5524,6 @@ static void addrconf_disable_change(strudev_disable_change (idev); } } -rcu_read_unlock (); }static int addrconf_disable_ipv6 (struct ctl_table *table,int *p,int newf)
Greg Kroah-Hartman
maintainerPATCH 23/51 tcp: initialize max window for a new fastopen socket
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alexey Kodanev <[email protected]>
[ Upstream commit 0dbd7ff3ac5017a46033a9d0a87a8267d69119d9 ]
Found that if we run LTP netstress test with large MSS (65K), the first attempt from server to send data comparable to this MSS on fastopen connection will be delayed by the probe timer.
Here is an example:
< S seq 0:0 win 43690 options [mss 65495 wscale 7 tfo cookie] length 32
> S. seq 0:0 ack 1 win 43690 options [mss 65495 wscale 7] length 0
< . ack 1 win 342 length 0Inside tcp_sendmsg(), tcp_send_mss() returns max MSS in 'mss_now', as well as in 'size_goal'. This results the segment not queued for transmition until all the data copied from user buffer. Then, inside __tcp_push_pending_frames(), it breaks on send window test and continues with the check probe timer.
Fragmentation occurs in tcp_write_wakeup()...
+0.2 > P. seq 1:43777 ack 1 win 342 length 43776
< . ack 43777, win 1365 length 0
> P. seq 43777:65001 ack 1 win 342 options [...] length 21224
...This also contradicts with the fact that we should bound to the half of the window if it is large.
Fix this flaw by correctly initializing max_window. Before that, it could have large values that affect further calculations of 'size_goal'.
--- net/ipv4/tcp_fastopen.c | 1 + 1 file changed, 1 insertion(+)
net/ipv4/tcp_fastopen.c +1 -0
@@ -205,6 +205,7 @@ static struct sock *tcp_fastopen_create_* scaled. So correct it appropriately. */ tp->snd_wnd =ntohs (tcp_hdr (skb)->window); + tp->max_window = tp->snd_wnd;/* Activate the retrans timer so that SYNACK can be retransmitted. * The request socket is not added to the ehash
Greg Kroah-Hartman
maintainerPATCH 24/51 net/mlx5e: Do not recycle pages from emergency reserve
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
[ Upstream commit e048fc50d7bde23136e098e04a324d7e3404408d ]
A driver using dev_alloc_page() must not reuse a page allocated from emergency memory reserve.
Otherwise all packets using this page will be immediately dropped, unless for very specific sockets having SOCK_MEMALLOC bit set.
This issue might be hard to debug, because only a fraction of received packets would be dropped.
--- drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +3 -0
@@ -193,6 +193,9 @@ static inline bool mlx5e_rx_cache_put(streturn false; } +if (unlikely (page_is_pfmemalloc (dma_info->page))) +return false; + cache->page_cache[cache->tail] = *dma_info; cache->tail = tail_next;return true;
Greg Kroah-Hartman
maintainerPATCH 25/51 bridge: netlink: call br_changelink() during br_dev_newlink()
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ivan Vecera <[email protected]>
[ Upstream commit b6677449dff674cf5b81429b11d5c7f358852ef9 ]
Any bridge options specified during link creation (e.g. ip link add) are ignored as br_dev_newlink() does not process them. Use br_changelink() to do it.
--- net/bridge/br_netlink.c | 33 +++++++++++++++++++-------------- 1 file changed, 19 insertions(+), 14 deletions(-)
net/bridge/br_netlink.c +19 -14
@@ -781,20 +781,6 @@ static int br_validate(struct nlattr *tbreturn 0 ; } -static int br_dev_newlink (struct net *src_net,struct net_device *dev, -struct nlattr *tb[],struct nlattr *data[]) -{ -struct net_bridge *br =netdev_priv (dev); - -if (tb[IFLA_ADDRESS ]) { -spin_lock_bh (&br->lock); -br_stp_change_bridge_id (br,nla_data (tb[IFLA_ADDRESS ])); -spin_unlock_bh (&br->lock); - } - -return register_netdevice (dev); -} - static int br_port_slave_changelink(struct net_device *brdev, struct net_device *dev, struct nlattr *tb[],
@@ -1093,6 +1079,25 @@ static int br_changelink(struct net_devireturn 0 ; } +static int br_dev_newlink (struct net *src_net,struct net_device *dev, +struct nlattr *tb[],struct nlattr *data[]) +{ +struct net_bridge *br =netdev_priv (dev); +int err; + +if (tb[IFLA_ADDRESS ]) { +spin_lock_bh (&br->lock); +br_stp_change_bridge_id (br,nla_data (tb[IFLA_ADDRESS ])); +spin_unlock_bh (&br->lock); + } + + err =br_changelink (dev, tb, data); +if (err) +return err; + +return register_netdevice (dev); +} +static size_t br_get_size (const struct net_device *brdev) { return nla_total_size(sizeof(u32)) + /* IFLA_BR_FORWARD_DELAY */
Greg Kroah-Hartman
maintainerPATCH 26/51 net: mpls: Fix multipath selection for LSR use case
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Ahern <[email protected]>
[ Upstream commit 9f427a0e474a67b454420c131709600d44850486 ]
MPLS multipath for LSR is broken -- always selecting the first nexthop in the one label case. For example:
$ ip -f mpls ro ls
100
nexthop as to 200 via inet 172.16.2.2 dev virt12
nexthop as to 300 via inet 172.16.3.2 dev virt13
101
nexthop as to 201 via inet6 2000:2::2 dev virt12
nexthop as to 301 via inet6 2000:3::2 dev virt13In this example incoming packets have a single MPLS labels which means BOS bit is set. The BOS bit is passed from mpls_forward down to mpls_multipath_hash which never processes the hash loop because BOS is 1.
Update mpls_multipath_hash to process the entire label stack. mpls_hdr_len tracks the total mpls header length on each pass (on pass N mpls_hdr_len is N * sizeof(mpls_shim_hdr)). When the label is found with the BOS set it verifies the skb has sufficient header for ipv4 or ipv6, and find the IPv4 and IPv6 header by using the last mpls_hdr pointer and adding 1 to advance past it.
With these changes I have verified the code correctly sees the label, BOS, IPv4 and IPv6 addresses in the network header and icmp/tcp/udp traffic for ipv4 and ipv6 are distributed across the nexthops.
--- net/mpls/af_mpls.c | 48 +++++++++++++++++++++++++----------------------- 1 file changed, 25 insertions(+), 23 deletions(-)
net/mpls/af_mpls.c +25 -23
@@ -98,18 +98,19 @@ bool mpls_pkt_too_big(const struct sk_bu }EXPORT_SYMBOL_GPL (mpls_pkt_too_big); -static u32 mpls_multipath_hash (struct mpls_route *rt, -struct sk_buff *skb,bool bos) +static u32 mpls_multipath_hash (struct mpls_route *rt,struct sk_buff *skb) {struct mpls_entry_decoded dec; +unsigned int mpls_hdr_len =0 ;struct mpls_shim_hdr *hdr;bool eli_seen = false;int label_index;u32 hash =0 ; -for (label_index =0 ; label_index <MAX_MP_SELECT_LABELS && !bos; +for (label_index =0 ; label_index <MAX_MP_SELECT_LABELS ; label_index++) { -if (!pskb_may_pull (skb,sizeof (*hdr) * label_index)) + mpls_hdr_len +=sizeof (*hdr); +if (!pskb_may_pull (skb, mpls_hdr_len))break ;/* Read and decode the current label */
@@ -134,37 +135,38 @@ static u32 mpls_multipath_hash(struct mp eli_seen = true; } - bos = dec.bos; -if (bos &&pskb_may_pull (skb,sizeof (*hdr) * label_index + -sizeof (struct iphdr ))) { +if (!dec.bos) +continue ; + +/* found bottom label; does skb have room for a header? */ +if (pskb_may_pull (skb, mpls_hdr_len +sizeof (struct iphdr ))) {const struct iphdr *v4hdr; - v4hdr = (const struct iphdr *)(mpls_hdr (skb) + - label_index); + v4hdr = (const struct iphdr *)(hdr +1 );if (v4hdr->version ==4 ) { hash =jhash_3words (ntohl (v4hdr->saddr),ntohl (v4hdr->daddr), v4hdr->protocol, hash); }else if (v4hdr->version ==6 && -pskb_may_pull (skb,sizeof (*hdr) * label_index + -sizeof (struct ipv6hdr ))) { +pskb_may_pull (skb, mpls_hdr_len + +sizeof (struct ipv6hdr ))) {const struct ipv6hdr *v6hdr; - v6hdr = (const struct ipv6hdr *)(mpls_hdr (skb) + - label_index); - + v6hdr = (const struct ipv6hdr *)(hdr +1 ); hash =__ipv6_addr_jhash (&v6hdr->saddr, hash); hash =__ipv6_addr_jhash (&v6hdr->daddr, hash); hash =jhash_1word (v6hdr->nexthdr, hash); } } + +break ; }return hash; }static struct mpls_nh *mpls_select_multipath (struct mpls_route *rt, -struct sk_buff *skb,bool bos) +struct sk_buff *skb) {int alive =ACCESS_ONCE (rt->rt_nhn_alive);u32 hash =0 ;
@@ -180,7 +182,7 @@ static struct mpls_nh *mpls_select_multiif (alive <=0 )return NULL ; - hash =mpls_multipath_hash (rt, skb, bos); + hash =mpls_multipath_hash (rt, skb); nh_index = hash % alive;if (alive == rt->rt_nhn) goto out;
@@ -278,17 +280,11 @@ static int mpls_forward(struct sk_buff * hdr =mpls_hdr (skb); dec =mpls_entry_decode (hdr); -/* Pop the label */ -skb_pull (skb,sizeof (*hdr)); -skb_reset_network_header (skb); - -skb_orphan (skb); - rt =mpls_route_input_rcu (net, dec.label);if (!rt) goto drop; - nh =mpls_select_multipath (rt, skb, dec.bos); + nh =mpls_select_multipath (rt, skb);if (!nh) goto drop;
@@ -297,6 +293,12 @@ static int mpls_forward(struct sk_buff *if (!mpls_output_possible (out_dev)) goto drop; +/* Pop the label */ +skb_pull (skb,sizeof (*hdr)); +skb_reset_network_header (skb); + +skb_orphan (skb); +if (skb_warn_if_lro (skb)) goto drop;
Greg Kroah-Hartman
maintainerPATCH 27/51 r8152: dont execute runtime suspend if the tx is not empty
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: hayeswang <[email protected]>
[ Upstream commit 6a0b76c04ec157c88ca943debf78a8ee58469f2d ]
Runtime suspend shouldn't be executed if the tx queue is not empty, because the device is not idle.
--- drivers/net/usb/r8152.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/usb/r8152.c +3 -1
@@ -32,7 +32,7 @@#define NETNEXT_VERSION "08"/* Information for net */ -#define NET_VERSION "6" +#define NET_VERSION "7"#define DRIVER_VERSION "v1." NETNEXT_VERSION "." NET_VERSION#define DRIVER_AUTHOR "Realtek linux nic maintainers <[email protected]>"
@@ -3572,6 +3572,8 @@ static bool delay_autosuspend(struct r81*/ if (!sw_linking && tp->rtl_ops.in_nway (tp))return true; +else if (!skb_queue_empty (&tp->tx_queue)) +return true;else return false; }
Greg Kroah-Hartman
maintainerPATCH 28/51 af_unix: move unix_mknod() out of bindlock
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: WANG Cong <[email protected]>
[ Upstream commit 0fb44559ffd67de8517098b81f675fa0210f13f0 ]
Dmitry reported a deadlock scenario:
unix_bind() path:u->bindlock ==> sb_writer
do_splice() path:sb_writer ==> pipe->mutex ==> u->bindlock
In the unix_bind() code path, unix_mknod() does not have to be done with u->bindlock held, since it is a pure fs operation, so we can just move unix_mknod() out.
--- net/unix/af_unix.c | 27 ++++++++++++++++----------- 1 file changed, 16 insertions(+), 11 deletions(-)
net/unix/af_unix.c +16 -11
@@ -995,6 +995,7 @@ static int unix_bind(struct socket *sockunsigned int hash;struct unix_address *addr;struct hlist_head *list; +struct path path = {NULL ,NULL }; err = -EINVAL ;if (sunaddr->sun_family !=AF_UNIX )
@@ -1010,9 +1011,20 @@ static int unix_bind(struct socket *sock goto out; addr_len = err; +if (sun_path[0 ]) { +umode_t mode =S_IFSOCK | + (SOCK_INODE (sock)->i_mode & ~current_umask ()); + err =unix_mknod (sun_path, mode, &path); +if (err) { +if (err == -EEXIST ) + err = -EADDRINUSE ; + goto out; + } + } + err =mutex_lock_interruptible (&u->bindlock);if (err) - goto out; + goto out_put; err = -EINVAL ;if (u->addr)
@@ -1029,16 +1041,6 @@ static int unix_bind(struct socket *sockatomic_set (&addr->refcnt,1 );if (sun_path[0 ]) { -struct path path; -umode_t mode =S_IFSOCK | - (SOCK_INODE (sock)->i_mode & ~current_umask ()); - err =unix_mknod (sun_path, mode, &path); -if (err) { -if (err == -EEXIST ) - err = -EADDRINUSE ; -unix_release_addr (addr); - goto out_up; - } addr->hash =UNIX_HASH_SIZE ; hash =d_real_inode (path.dentry)->i_ino & (UNIX_HASH_SIZE -1 );spin_lock (&unix_table_lock);
@@ -1065,6 +1067,9 @@ out_unlock:spin_unlock (&unix_table_lock); out_up:mutex_unlock (&u->bindlock); +out_put: +if (err) +path_put (&path); out:return err; }
Greg Kroah-Hartman
maintainerPATCH 30/51 net: Specify the owning module for lwtunnel ops
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Robert Shearman <[email protected]>
[ Upstream commit 88ff7334f25909802140e690c0e16433e485b0a0 ]
Modules implementing lwtunnel ops should not be allowed to unload while there is state alive using those ops, so specify the owning module for all lwtunnel ops.
--- include/net/lwtunnel.h | 2 ++ net/ipv4/ip_tunnel_core.c | 2 ++ net/ipv6/ila/ila_lwt.c | 1 + net/mpls/mpls_iptunnel.c | 1 + 4 files changed, 6 insertions(+)
include/net/lwtunnel.h +2 -0
@@ -43,6 +43,8 @@ struct lwtunnel_encap_ops {int (*get_encap_size)(struct lwtunnel_state *lwtstate);int (*cmp_encap)(struct lwtunnel_state *a,struct lwtunnel_state *b);int (*xmit)(struct sk_buff *skb); + +struct module *owner; };#ifdef CONFIG_LWTUNNEL
net/ipv4/ip_tunnel_core.c +2 -0
@@ -313,6 +313,7 @@ static const struct lwtunnel_encap_ops i .fill_encap = ip_tun_fill_encap_info, .get_encap_size = ip_tun_encap_nlsize, .cmp_encap = ip_tun_cmp_encap, + .owner = THIS_MODULE, };static const struct nla_policy ip6_tun_policy[LWTUNNEL_IP6_MAX +1 ] = {
@@ -403,6 +404,7 @@ static const struct lwtunnel_encap_ops i .fill_encap = ip6_tun_fill_encap_info, .get_encap_size = ip6_tun_encap_nlsize, .cmp_encap = ip_tun_cmp_encap, + .owner = THIS_MODULE, };void __init ip_tunnel_core_init (void )
net/ipv6/ila/ila_lwt.c +1 -0
@@ -164,6 +164,7 @@ static const struct lwtunnel_encap_ops i .fill_encap = ila_fill_encap_info, .get_encap_size = ila_encap_nlsize, .cmp_encap = ila_encap_cmp, + .owner = THIS_MODULE, };int ila_lwt_init (void )
net/mpls/mpls_iptunnel.c +1 -0
@@ -218,6 +218,7 @@ static const struct lwtunnel_encap_ops m .fill_encap = mpls_fill_encap_info, .get_encap_size = mpls_encap_nlsize, .cmp_encap = mpls_encap_cmp, + .owner = THIS_MODULE, };static int __init mpls_iptunnel_init (void )
Greg Kroah-Hartman
maintainerPATCH 31/51 lwtunnel: Fix oops on state free after encap module unload
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Robert Shearman <[email protected]>
[ Upstream commit 85c814016ce3b371016c2c054a905fa2492f5a65 ]
When attempting to free lwtunnel state after the module for the encap has been unloaded an oops occurs:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 IP: lwtstate_free+0x18/0x40 [..] task: ffff88003e372380 task.stack: ffffc900001fc000 RIP: 0010:lwtstate_free+0x18/0x40 RSP: 0018:ffff88003fd83e88 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff88002bbb3380 RCX: ffff88000c91a300 [..] Call Trace: <IRQ> free_fib_info_rcu+0x195/0x1a0 ? rt_fibinfo_free+0x50/0x50 rcu_process_callbacks+0x2d3/0x850 ? rcu_process_callbacks+0x296/0x850 __do_softirq+0xe4/0x4cb irq_exit+0xb0/0xc0 smp_apic_timer_interrupt+0x3d/0x50 apic_timer_interrupt+0x93/0xa0 [..] Code: e8 6e c6 fc ff 89 d8 5b 5d c3 bb de ff ff ff eb f4 66 90 66 66 66 66 90 55 48 89 e5 53 0f b7 07 48 89 fb 48 8b 04 c5 00 81 d5 81 <48> 8b 40 08 48 85 c0 74 13 ff d0 48 8d 7b 20 be 20 00 00 00 e8
The problem is after the module for the encap can be unloaded the corresponding ops is removed and is thus NULL here.
Modules implementing lwtunnel ops should not be allowed to unload while there is state alive using those ops, so grab the module reference for the ops on creating lwtunnel state and of course release the reference when freeing the state.
--- include/net/lwtunnel.h | 5 +---- net/core/lwtunnel.c | 14 +++++++++++++- 2 files changed, 14 insertions(+), 5 deletions(-)
include/net/lwtunnel.h +1 -4
@@ -48,10 +48,7 @@ struct lwtunnel_encap_ops { };#ifdef CONFIG_LWTUNNEL -static inline void lwtstate_free (struct lwtunnel_state *lws) -{ -kfree (lws); -} +void lwtstate_free (struct lwtunnel_state *lws);static inline struct lwtunnel_state *lwtstate_get (struct lwtunnel_state *lws)
net/core/lwtunnel.c +13 -1
@@ -66,6 +66,15 @@ EXPORT_SYMBOL(lwtunnel_state_alloc);static const struct lwtunnel_encap_ops __rcu * lwtun_encaps[LWTUNNEL_ENCAP_MAX +1 ]__read_mostly ; +void lwtstate_free (struct lwtunnel_state *lws) +{ +const struct lwtunnel_encap_ops *ops = lwtun_encaps[lws->type]; + +kfree (lws); +module_put (ops->owner); +} +EXPORT_SYMBOL (lwtstate_free); +int lwtunnel_encap_add_ops (const struct lwtunnel_encap_ops *ops,unsigned int num) {
@@ -111,8 +120,11 @@ int lwtunnel_build_state(struct net_devi ret = -EOPNOTSUPP ;rcu_read_lock (); ops =rcu_dereference (lwtun_encaps[encap_type]); -if (likely (ops && ops->build_state)) +if (likely (ops && ops->build_state &&try_module_get (ops->owner))) { ret = ops->build_state (dev, encap, family, cfg, lws); +if (ret) +module_put (ops->owner); + }rcu_read_unlock ();return ret;
Greg Kroah-Hartman
maintainerPATCH 32/51 net: dsa: Bring back device detaching in dsa_slave_suspend()
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Florian Fainelli <[email protected]>
[ Upstream commit f154be241d22298d2b63c9b613f619fa1086ea75 ]
Commit 448b4482c671 ("net: dsa: Add lockdep class to tx queues to avoid lockdep splat") removed the netif_device_detach() call done in dsa_slave_suspend() which is necessary, and paired with a corresponding netif_device_attach(), bring it back.
--- net/dsa/slave.c | 2 ++ 1 file changed, 2 insertions(+)
net/dsa/slave.c +2 -0
@@ -1201,6 +1201,8 @@ int dsa_slave_suspend(struct net_device {struct dsa_slave_priv *p =netdev_priv (slave_dev); +netif_device_detach (slave_dev); +if (p->phy) {phy_stop (p->phy); p->old_pause =-1 ;
Greg Kroah-Hartman
maintainerPATCH 33/51 xfs: bump up reserved blocks in xfs_alloc_set_aside
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christoph Hellwig <[email protected]>
commit 5149fd327f16e393c1d04fa5325ab072c32472bf upstream.
Setting aside 4 blocks globally for bmbt splits isn't all that useful, as different threads can allocate space in parallel. Bump it to 4 blocks per AG to allow each thread that is currently doing an allocation to dip into it separately. Without that we may no have enough reserved blocks if there are enough parallel transactions in an almost out space file system that all run into bmap btree splits.
--- fs/xfs/libxfs/xfs_alloc.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +1 -4
@@ -95,10 +95,7 @@ unsigned intxfs_alloc_set_aside ( struct xfs_mount *mp) { -unsigned int blocks; - - blocks =4 + (mp->m_sb.sb_agcount *XFS_ALLOC_AGFL_RESERVE ); -return blocks; +return mp->m_sb.sb_agcount * (XFS_ALLOC_AGFL_RESERVE +4 ); }/*
Greg Kroah-Hartman
maintainerPATCH 34/51 xfs: fix bogus minleft manipulations
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christoph Hellwig <[email protected]>
commit 255c516278175a6dc7037d1406307f35237d8688 upstream.
We can't just set minleft to 0 when we're low on space - that's exactly what we need minleft for: to protect space in the AG for btree block allocations when we are low on free space.
--- fs/xfs/libxfs/xfs_alloc.c | 24 +++++++----------------- fs/xfs/libxfs/xfs_bmap.c | 3 --- fs/xfs/libxfs/xfs_bmap_btree.c | 3 +-- 3 files changed, 8 insertions(+), 22 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +7 -17
@@ -2635,12 +2635,10 @@ xfs_alloc_vextent(xfs_agblock_t agsize;/* allocation group size */ int error;int flags;/* XFS_ALLOC_FLAG_... locking flags */ -xfs_extlen_t minleft;/* minimum left value, temp copy */ xfs_mount_t *mp;/* mount structure pointer */ xfs_agnumber_t sagno;/* starting allocation group number */ xfs_alloctype_t type;/* input allocation type */ int bump_rotor =0 ; -int no_min =0 ;xfs_agnumber_t rotorstep = xfs_rotorstep;/* inode32 agf stepper */ mp = args->mp;
@@ -2669,7 +2667,6 @@ xfs_alloc_vextent(trace_xfs_alloc_vextent_badargs (args);return 0 ; } - minleft = args->minleft;switch (type) {case XFS_ALLOCTYPE_THIS_AG :
@@ -2680,9 +2677,7 @@ xfs_alloc_vextent(*/ args->agno =XFS_FSB_TO_AGNO (mp, args->fsbno); args->pag =xfs_perag_get (mp, args->agno); - args->minleft =0 ; error =xfs_alloc_fix_freelist (args,0 ); - args->minleft = minleft;if (error) {trace_xfs_alloc_vextent_nofix (args); goto error0;
@@ -2747,9 +2742,7 @@ xfs_alloc_vextent(*/ for (;;) { args->pag =xfs_perag_get (mp, args->agno); -if (no_min) args->minleft =0 ; error =xfs_alloc_fix_freelist (args, flags); - args->minleft = minleft;if (error) {trace_xfs_alloc_vextent_nofix (args); goto error0;
@@ -2789,20 +2782,17 @@ xfs_alloc_vextent(* or switch to non-trylock mode. */ if (args->agno == sagno) { -if (no_min ==1 ) { +if (flags ==0 ) { args->agbno =NULLAGBLOCK ;trace_xfs_alloc_vextent_allfailed (args);break ; } -if (flags ==0 ) { - no_min =1 ; - }else { - flags =0 ; -if (type ==XFS_ALLOCTYPE_START_BNO ) { - args->agbno =XFS_FSB_TO_AGBNO (mp, - args->fsbno); - args->type =XFS_ALLOCTYPE_NEAR_BNO ; - } + + flags =0 ; +if (type ==XFS_ALLOCTYPE_START_BNO ) { + args->agbno =XFS_FSB_TO_AGBNO (mp, + args->fsbno); + args->type =XFS_ALLOCTYPE_NEAR_BNO ; } }xfs_perag_put (args->pag);
fs/xfs/libxfs/xfs_bmap.c +0 -3
@@ -3903,7 +3903,6 @@ xfs_bmap_btalloc( args.fsbno =0 ; args.type =XFS_ALLOCTYPE_FIRST_AG ; args.total = ap->minlen; - args.minleft =0 ;if ((error =xfs_alloc_vextent (&args)))return error; ap->dfops->dop_low = true;
@@ -4437,8 +4436,6 @@ xfs_bmapi_allocate(if (error)return error; -if (bma->dfops->dop_low) - bma->minleft =0 ;if (bma->cur) bma->cur->bc_private.b.firstblock = *bma->firstblock;if (bma->blkno ==NULLFSBLOCK )
fs/xfs/libxfs/xfs_bmap_btree.c +1 -2
@@ -502,12 +502,11 @@ try_another_ag:if (args.fsbno ==NULLFSBLOCK && args.minleft) {/* * Could not find an AG with enough free space to satisfy -* a full btree split. Try again without minleft and if +* a full btree split. Try again and if * successful activate the lowspace algorithm. */ args.fsbno =0 ; args.type =XFS_ALLOCTYPE_FIRST_AG ; - args.minleft =0 ; error =xfs_alloc_vextent (&args);if (error) goto error0;
Greg Kroah-Hartman
maintainerPATCH 35/51 xfs: adjust allocation length in xfs_alloc_space_available
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christoph Hellwig <[email protected]>
commit 54fee133ad59c87ab01dd84ab3e9397134b32acb upstream.
We must decide in xfs_alloc_fix_freelist if we can perform an allocation from a given AG is possible or not based on the available space, and should not fail the allocation past that point on a healthy file system.
But currently we have two additional places that second-guessxfs_alloc_fix_freelist: xfs_alloc_ag_vextent tries to adjust the maxlen parameter to remove the reservation before doing the allocation (but ignores the various minium freespace requirements), and xfs_alloc_fix_minleft tries to fix up the allocated length after we've found an extent, but ignores the reservations and also doesn't take the AGFL into account (and thus fails allocations for not matching minlen in some cases).
Remove all these later fixups and just correct the maxlen argument inside xfs_alloc_fix_freelist once we have the AGF buffer locked.
--- fs/xfs/libxfs/xfs_alloc.c | 81 +++++++++------------------------------------- fs/xfs/libxfs/xfs_alloc.h | 2 - 2 files changed, 18 insertions(+), 65 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +17 -64
@@ -362,36 +362,12 @@ xfs_alloc_fix_len(return ;ASSERT (rlen >= args->minlen && rlen <= args->maxlen);ASSERT (rlen % args->prod == args->mod); +ASSERT (args->pag->pagf_freeblks + args->pag->pagf_flcount >= + rlen + args->minleft); args->len = rlen; }/* -* Fix up length if there is too little space left in the a.g. -* Return 1 if ok, 0 if too little, should give up. -*/ -STATIC int -xfs_alloc_fix_minleft ( -xfs_alloc_arg_t *args)/* allocation argument structure */ -{ -xfs_agf_t *agf;/* a.g. freelist header */ -int diff;/* free space difference */ - -if (args->minleft ==0 ) -return 1 ; - agf =XFS_BUF_TO_AGF (args->agbp); - diff =be32_to_cpu (agf->agf_freeblks) - - args->len - args->minleft; -if (diff >=0 ) -return 1 ; - args->len += diff;/* shrink the allocated space */ -/* casts to (int) catch length underflows */ -if ((int )args->len >= (int )args->minlen) -return 1 ; - args->agbno =NULLAGBLOCK ; -return 0 ; -} - -/* * Update the two btrees, logically removing from freespace the extent * starting at rbno, rlen blocks. The extent is contained within the * actual (current) free extent fbno for flen blocks.
@@ -686,8 +662,6 @@ xfs_alloc_ag_vextent( xfs_alloc_arg_t *args) /* argument structure for allocation */ {int error=0 ; -xfs_extlen_t reservation; -xfs_extlen_t oldmax;ASSERT (args->minlen >0 );ASSERT (args->maxlen >0 );
@@ -696,20 +670,6 @@ xfs_alloc_ag_vextent(ASSERT (args->alignment >0 );/* -* Clamp maxlen to the amount of free space minus any reservations -* that have been made. -*/ - oldmax = args->maxlen; - reservation =xfs_ag_resv_needed (args->pag, args->resv); -if (args->maxlen > args->pag->pagf_freeblks - reservation) - args->maxlen = args->pag->pagf_freeblks - reservation; -if (args->maxlen ==0 ) { - args->agbno =NULLAGBLOCK ; - args->maxlen = oldmax; -return 0 ; - } - -/* * Branch to correct routine based on the type. */ args->wasfromfl =0 ;
@@ -728,8 +688,6 @@ xfs_alloc_ag_vextent(/* NOTREACHED */ } - args->maxlen = oldmax; -if (error || args->agbno ==NULLAGBLOCK )return error;
@@ -838,9 +796,6 @@ xfs_alloc_ag_vextent_exact( args->len =XFS_AGBLOCK_MIN (tend, args->agbno + args->maxlen) - args->agbno;xfs_alloc_fix_len (args); -if (!xfs_alloc_fix_minleft (args)) - goto not_found; -ASSERT (args->agbno + args->len <= tend);/*
@@ -1146,12 +1101,7 @@ restart:XFS_WANT_CORRUPTED_GOTO (args->mp, i ==1 , error0);ASSERT (ltbno + ltlen <=be32_to_cpu (XFS_BUF_TO_AGF (args->agbp)->agf_length)); args->len = blen; -if (!xfs_alloc_fix_minleft (args)) { -xfs_btree_del_cursor (cnt_cur,XFS_BTREE_NOERROR ); -trace_xfs_alloc_near_nominleft (args); -return 0 ; - } - blen = args->len; +/* * We are allocating starting at bnew for blen blocks. */
@@ -1343,12 +1293,6 @@ restart:*/ args->len =XFS_EXTLEN_MIN (ltlena, args->maxlen);xfs_alloc_fix_len (args); -if (!xfs_alloc_fix_minleft (args)) { -trace_xfs_alloc_near_nominleft (args); -xfs_btree_del_cursor (bno_cur_lt,XFS_BTREE_NOERROR ); -xfs_btree_del_cursor (cnt_cur,XFS_BTREE_NOERROR ); -return 0 ; - } rlen = args->len; (void )xfs_alloc_compute_diff (args->agbno, rlen, args->alignment, args->datatype, ltbnoa, ltlena, <new);
@@ -1550,8 +1494,6 @@ restart: }xfs_alloc_fix_len (args); -if (!xfs_alloc_fix_minleft (args)) - goto out_nominleft; rlen = args->len;XFS_WANT_CORRUPTED_GOTO (args->mp, rlen <= flen, error0);/*
@@ -2070,10 +2012,20 @@ xfs_alloc_space_available(/* do we have enough free space remaining for the allocation? */ available = (int )(pag->pagf_freeblks + pag->pagf_flcount - - reservation - min_free - args->total); -if (available < (int )args->minleft || available <=0 ) + reservation - min_free - args->minleft); +if (available < (int )args->total)return false; +/* +* Clamp maxlen to the amount of free space available for the actual +* extent allocation. +*/ +if (available < (int )args->maxlen && !(flags &XFS_ALLOC_FLAG_CHECK )) { + args->maxlen = available; +ASSERT (args->maxlen >0 ); +ASSERT (args->maxlen >= args->minlen); + } +return true; }
@@ -2119,7 +2071,8 @@ xfs_alloc_fix_freelist( } need =xfs_alloc_min_freelist (mp, pag); -if (!xfs_alloc_space_available (args, need, flags)) +if (!xfs_alloc_space_available (args, need, flags | +XFS_ALLOC_FLAG_CHECK )) goto out_agbp_relse;/*
fs/xfs/libxfs/xfs_alloc.h +1 -1
@@ -56,7 +56,7 @@ typedef unsigned int xfs_alloctype_t;#define XFS_ALLOC_FLAG_FREEING 0x00000002/* indicate caller is freeing extents*/ #define XFS_ALLOC_FLAG_NORMAP 0x00000004/* don't modify the rmapbt */ #define XFS_ALLOC_FLAG_NOSHRINK 0x00000008/* don't shrink the freelist */ - +#define XFS_ALLOC_FLAG_CHECK 0x00000010/* test only, don't modify args */ /* * Argument structure for xfs_alloc routines.
Greg Kroah-Hartman
maintainerPATCH 36/51 xfs: dont rely on ->total in xfs_alloc_space_available
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christoph Hellwig <[email protected]>
commit 12ef830198b0d71668eb9b59f9ba69d32951a48a upstream.
->total is a bit of an odd parameter passed down to the low-level allocator all the way from the high-level callers. It's supposed to contain the maximum number of blocks to be allocated for the whole transaction [1].
But in xfs_iomap_write_allocate we only convert existing delayed allocations and thus only have a minimal block reservation for the current transaction, so xfs_alloc_space_available can't use it for the allocation decisions. Use the maximum of args->total and the calculated block requirement to make a decision. We probably should get rid of args->total eventually and instead apply ->minleft more broadly, but that will require some extensive changes all over.
[1] which creates lots of confusion as most callers don't decrement it once doing a first allocation. But that's for a separate series.
--- fs/xfs/libxfs/xfs_alloc.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)
fs/xfs/libxfs/xfs_alloc.c +4 -3
@@ -1995,7 +1995,7 @@ xfs_alloc_space_available( int flags) {struct xfs_perag *pag = args->pag; -xfs_extlen_t longest; +xfs_extlen_t alloc_len, longest;xfs_extlen_t reservation;/* blocks that are still reserved */ int available;
@@ -2005,15 +2005,16 @@ xfs_alloc_space_available( reservation =xfs_ag_resv_needed (pag, args->resv);/* do we have enough contiguous free space for the allocation? */ + alloc_len = args->minlen + (args->alignment -1 ) + args->minalignslop; longest =xfs_alloc_longest_free_extent (args->mp, pag, min_free, reservation); -if ((args->minlen + args->alignment + args->minalignslop -1 ) > longest) +if (longest < alloc_len)return false;/* do we have enough free space remaining for the allocation? */ available = (int )(pag->pagf_freeblks + pag->pagf_flcount - reservation - min_free - args->minleft); -if (available < (int )args->total) +if (available < (int )max (args->total, alloc_len))return false;/*
Greg Kroah-Hartman
maintainerPATCH 37/51 xfs: dont print warnings when xfs_log_force fails
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christoph Hellwig <[email protected]>
commit 84a4620cfe97c9d57e39b2369bfb77faff55063d upstream.
There are only two reasons for xfs_log_force / xfs_log_force_lsn to fail:
one is an I/O error, for which xlog_bdstrat already logs a warning, and the second is an already shutdown log due to a previous I/O errors. In the latter case we'll already have a previous indication for the actual error, but the large stream of misleading warnings from xfs_log_force will probably scroll it out of the message buffer.
Simply removing the warnings thus makes the XFS log reporting significantly better.
--- fs/xfs/xfs_log.c | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-)
fs/xfs/xfs_log.c +2 -10
@@ -3324,12 +3324,8 @@ xfs_log_force(xfs_mount_t *mp, uint flags) { - int error; - trace_xfs_log_force(mp, 0, _RET_IP_); - error =_xfs_log_force (mp, flags,NULL ); -if (error) -xfs_warn (mp,"%s: error %d returned." , __func__, error); +_xfs_log_force (mp, flags,NULL ); }/*
@@ -3473,12 +3469,8 @@ xfs_log_force_lsn(xfs_lsn_t lsn, uint flags) { - int error; -trace_xfs_log_force (mp ,lsn ,_RET_IP_ ); - error =_xfs_log_force_lsn (mp, lsn, flags,NULL ); -if (error) -xfs_warn (mp,"%s: error %d returned." , __func__, error); +_xfs_log_force_lsn (mp, lsn, flags,NULL ); }/*
Greg Kroah-Hartman
maintainerPATCH 38/51 xfs: make the ASSERT() condition likely
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Amir Goldstein <[email protected]>
commit bf46ecc3d8cca05f2907cf482755c42c2b11a79d upstream.
The ASSERT() condition is the normal case, not the exception, so testing the condition should be likely(), not unlikely().
--- fs/xfs/xfs_linux.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
fs/xfs/xfs_linux.h +3 -3
@@ -330,11 +330,11 @@ static inline __uint64_t howmany_64(__ui }#define ASSERT_ALWAYS (expr) \ - (unlikely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__)) + (likely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__))#ifdef DEBUG #define ASSERT (expr) \ - (unlikely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__)) + (likely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__))#ifndef STATIC # define STATIC noinline
@@ -345,7 +345,7 @@ static inline __uint64_t howmany_64(__ui#ifdef XFS_WARN #define ASSERT (expr) \ - (unlikely(expr) ? (void)0 : asswarn(#expr, __FILE__, __LINE__)) + (likely(expr) ? (void)0 : asswarn(#expr, __FILE__, __LINE__))#ifndef STATIC # define STATIC static noinline
Greg Kroah-Hartman
maintainerPATCH 39/51 xfs: sanity check directory inode di_size
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Amir Goldstein <[email protected]>
commit 3c6f46eacd876bd723a9bad3c6882714c052fd8e upstream.
This changes fixes an assertion hit when fuzzing on-disk i_mode values.
The easy case to fix is when changing an empty file i_mode to S_IFDIR. In this case, xfs_dinode_verify() detects an illegal zero size for directory and fails to load the inode structure from disk.
For the case of non empty file whose i_mode is changed to S_IFDIR, the ASSERT() statement in xfs_dir2_isblock() is replaced with return -EFSCORRUPTED, to avoid interacting with corrupted jusk also when XFS_DEBUG is disabled.
--- fs/xfs/libxfs/xfs_dir2.c | 3 ++- fs/xfs/libxfs/xfs_inode_buf.c | 7 +++++-- 2 files changed, 7 insertions(+), 3 deletions(-)
fs/xfs/libxfs/xfs_dir2.c +2 -1
@@ -631,7 +631,8 @@ xfs_dir2_isblock(if ((rval =xfs_bmap_last_offset (args->dp, &last,XFS_DATA_FORK )))return rval; rval =XFS_FSB_TO_B (args->dp->i_mount, last) == args->geo->blksize; -ASSERT (rval ==0 || args->dp->i_d.di_size == args->geo->blksize); +if (rval !=0 && args->dp->i_d.di_size != args->geo->blksize) +return -EFSCORRUPTED ; *vp = rval;return 0 ; }
fs/xfs/libxfs/xfs_inode_buf.c +5 -2
@@ -386,6 +386,7 @@ xfs_dinode_verify(struct xfs_inode *ip, struct xfs_dinode *dip) { + uint16_t mode;uint16_t flags;uint64_t flags2;
@@ -396,8 +397,10 @@ xfs_dinode_verify(if (be64_to_cpu (dip->di_size) & (1ULL <<63 ))return false; -/* No zero-length symlinks. */ -if (S_ISLNK (be16_to_cpu (dip->di_mode)) && dip->di_size ==0 ) + mode =be16_to_cpu (dip->di_mode); + +/* No zero-length symlinks/dirs. */ +if ((S_ISLNK (mode) ||S_ISDIR (mode)) && dip->di_size ==0 )return false;/* only version 3 or greater inodes are extensively verified here */
Greg Kroah-Hartman
maintainerPATCH 40/51 xfs: add missing include dependencies to xfs_dir2.h
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Amir Goldstein <[email protected]>
commit b597dd5373a1ccc08218665dc8417433b1c09550 upstream.
xfs_dir2.h dereferences some data types in inline functions and fails to include those type definitions, e.g.:xfs_dir2_data_aoff_t, struct xfs_da_geometry.
--- fs/xfs/libxfs/xfs_dir2.h | 3 +++ 1 file changed, 3 insertions(+)
fs/xfs/libxfs/xfs_dir2.h +3 -0
@@ -18,6 +18,9 @@#ifndef __XFS_DIR2_H__#define __XFS_DIR2_H__ +#include "xfs_da_format.h" +#include "xfs_da_btree.h" +struct xfs_defer_ops ;struct xfs_da_args ;struct xfs_inode ;
Greg Kroah-Hartman
maintainerPATCH 41/51 xfs: replace xfs_mode_to_ftype table with switch statement
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Amir Goldstein <[email protected]>
commit 1fc4d33fed124fb182e8e6c214e973a29389ae83.
The size of the xfs_mode_to_ftype[] conversion table was too small to handle an invalid value of mode=S_IFMT.
Instead of fixing the table size, replace the conversion table with a conversion helper that uses a switch statement.
--- fs/xfs/libxfs/xfs_dir2.c | 36 ++++++++++++++++++++++-------------- fs/xfs/libxfs/xfs_dir2.h | 5 ++--- fs/xfs/xfs_iops.c | 2 +- 3 files changed, 25 insertions(+), 18 deletions(-)
fs/xfs/libxfs/xfs_dir2.c +22 -14
@@ -36,21 +36,29 @@struct xfs_name xfs_name_dotdot = { (unsigned char *)".." ,2 ,XFS_DIR3_FT_DIR };/* -* @mode, if set, indicates that the type field needs to be set up. -* This uses the transformation from file mode to DT_* as defined in linux/fs.h -* for file type specification. This will be propagated into the directory -* structure if appropriate for the given operation and filesystem config. +* Convert inode mode to directory entry filetype */ -const unsigned char xfs_mode_to_ftype[S_IFMT >>S_SHIFT ] = { - [0 ] =XFS_DIR3_FT_UNKNOWN , - [S_IFREG >>S_SHIFT ] =XFS_DIR3_FT_REG_FILE , - [S_IFDIR >>S_SHIFT ] =XFS_DIR3_FT_DIR , - [S_IFCHR >>S_SHIFT ] =XFS_DIR3_FT_CHRDEV , - [S_IFBLK >>S_SHIFT ] =XFS_DIR3_FT_BLKDEV , - [S_IFIFO >>S_SHIFT ] =XFS_DIR3_FT_FIFO , - [S_IFSOCK >>S_SHIFT ] =XFS_DIR3_FT_SOCK , - [S_IFLNK >>S_SHIFT ] =XFS_DIR3_FT_SYMLINK , -}; +const unsigned char xfs_mode_to_ftype (int mode) +{ +switch (mode &S_IFMT ) { +case S_IFREG : +return XFS_DIR3_FT_REG_FILE ; +case S_IFDIR : +return XFS_DIR3_FT_DIR ; +case S_IFCHR : +return XFS_DIR3_FT_CHRDEV ; +case S_IFBLK : +return XFS_DIR3_FT_BLKDEV ; +case S_IFIFO : +return XFS_DIR3_FT_FIFO ; +case S_IFSOCK : +return XFS_DIR3_FT_SOCK ; +case S_IFLNK : +return XFS_DIR3_FT_SYMLINK ; +default : +return XFS_DIR3_FT_UNKNOWN ; + } +}/* * ASCII case-insensitive (ie. A-Z) support for directories that was
fs/xfs/libxfs/xfs_dir2.h +2 -3
@@ -35,10 +35,9 @@ struct xfs_dir2_data_unused;extern struct xfs_name xfs_name_dotdot;/* -* directory filetype conversion tables. +* Convert inode mode to directory entry filetype */ -#define S_SHIFT 12 -extern const unsigned char xfs_mode_to_ftype[]; +extern const unsigned char xfs_mode_to_ftype (int mode);/* * directory operations vector for encode/decode routines
fs/xfs/xfs_iops.c +1 -1
@@ -103,7 +103,7 @@ xfs_dentry_to_name( { namep->name = dentry->d_name.name; namep->len = dentry->d_name.len; - namep->type = xfs_mode_to_ftype[(mode &S_IFMT ) >>S_SHIFT ]; + namep->type =xfs_mode_to_ftype (mode); }STATIC void
Greg Kroah-Hartman
maintainerPATCH 42/51 xfs: sanity check inode mode when creating new dentry
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Amir Goldstein <[email protected]>
commit fab8eef86c814c3dd46bc5d760b6e4a53d5fc5a6 upstream.
The helper xfs_dentry_to_name() is used by 2 different classes of callers: Callers that pass zero mode and don't care about the returned name.type field and Callers that pass non zero mode and do care about the name.type field.
Change xfs_dentry_to_name() to not take the mode argument and change the call sites of the first class to not pass the mode argument.
Create a new helper xfs_dentry_mode_to_name() which does pass the mode argument and returns -EFSCORRUPTED if mode is invalid. Callers that translate non zero mode to on-disk file type now check the return value and will export the error to user instead of staging an invalid file type to be written to directory entry.
--- fs/xfs/xfs_iops.c | 48 +++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 39 insertions(+), 9 deletions(-)
fs/xfs/xfs_iops.c +39 -9
@@ -98,12 +98,27 @@ xfs_init_security(static void xfs_dentry_to_name (struct xfs_name *namep, +struct dentry *dentry) +{ + namep->name = dentry->d_name.name; + namep->len = dentry->d_name.len; + namep->type =XFS_DIR3_FT_UNKNOWN ; +} + +static int +xfs_dentry_mode_to_name ( +struct xfs_name *namep,struct dentry *dentry,int mode) { namep->name = dentry->d_name.name; namep->len = dentry->d_name.len; namep->type =xfs_mode_to_ftype (mode); + +if (unlikely (namep->type ==XFS_DIR3_FT_UNKNOWN )) +return -EFSCORRUPTED ; + +return 0 ; }STATIC void
@@ -119,7 +134,7 @@ xfs_cleanup_inode(* xfs_init_security we must back out. * ENOSPC can hit here, among other things. */ -xfs_dentry_to_name (&teardown, dentry,0 ); +xfs_dentry_to_name (&teardown, dentry);xfs_remove (XFS_I (dir), &teardown,XFS_I (inode)); }
@@ -154,8 +169,12 @@ xfs_generic_create(if (error)return error; +/* Verify mode is valid also for tmpfile case */ + error =xfs_dentry_mode_to_name (&name, dentry, mode); +if (unlikely (error)) + goto out_free_acl; +if (!tmpfile) { -xfs_dentry_to_name (&name, dentry, mode); error =xfs_create (XFS_I (dir), &name, mode, rdev, &ip); }else { error =xfs_create_tmpfile (XFS_I (dir), dentry, mode, &ip);
@@ -248,7 +267,7 @@ xfs_vn_lookup(if (dentry->d_name.len >=MAXNAMELEN )return ERR_PTR (-ENAMETOOLONG ); -xfs_dentry_to_name (&name, dentry,0 ); +xfs_dentry_to_name (&name, dentry); error =xfs_lookup (XFS_I (dir), &name, &cip,NULL );if (unlikely (error)) {if (unlikely (error != -ENOENT ))
@@ -275,7 +294,7 @@ xfs_vn_ci_lookup(if (dentry->d_name.len >=MAXNAMELEN )return ERR_PTR (-ENAMETOOLONG ); -xfs_dentry_to_name (&xname, dentry,0 ); +xfs_dentry_to_name (&xname, dentry); error =xfs_lookup (XFS_I (dir), &xname, &ip, &ci_name);if (unlikely (error)) {if (unlikely (error != -ENOENT ))
@@ -310,7 +329,9 @@ xfs_vn_link(struct xfs_name name;int error; -xfs_dentry_to_name (&name, dentry, inode->i_mode); + error =xfs_dentry_mode_to_name (&name, dentry, inode->i_mode); +if (unlikely (error)) +return error; error =xfs_link (XFS_I (dir),XFS_I (inode), &name);if (unlikely (error))
@@ -329,7 +350,7 @@ xfs_vn_unlink(struct xfs_name name;int error; -xfs_dentry_to_name (&name, dentry,0 ); +xfs_dentry_to_name (&name, dentry); error =xfs_remove (XFS_I (dir), &name,XFS_I (d_inode (dentry)));if (error)
@@ -359,7 +380,9 @@ xfs_vn_symlink( mode =S_IFLNK | (irix_symlink_mode ?0777 & ~current_umask () :S_IRWXUGO ); -xfs_dentry_to_name (&name, dentry, mode); + error =xfs_dentry_mode_to_name (&name, dentry, mode); +if (unlikely (error)) + goto out; error =xfs_symlink (XFS_I (dir), &name, symname, mode, &cip);if (unlikely (error))
@@ -395,6 +418,7 @@ xfs_vn_rename( {struct inode *new_inode =d_inode (ndentry);int omode =0 ; +int error;struct xfs_name oname;struct xfs_name nname;
@@ -405,8 +429,14 @@ xfs_vn_rename(if (flags &RENAME_EXCHANGE ) omode =d_inode (ndentry)->i_mode; -xfs_dentry_to_name (&oname, odentry, omode); -xfs_dentry_to_name (&nname, ndentry,d_inode (odentry)->i_mode); + error =xfs_dentry_mode_to_name (&oname, odentry, omode); +if (omode &&unlikely (error)) +return error; + + error =xfs_dentry_mode_to_name (&nname, ndentry, +d_inode (odentry)->i_mode); +if (unlikely (error)) +return error; return xfs_rename(XFS_I(odir), &oname, XFS_I(d_inode(odentry)), XFS_I(ndir), &nname,
Greg Kroah-Hartman
maintainerPATCH 43/51 xfs: sanity check inode di_mode
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Amir Goldstein <[email protected]>
commit a324cbf10a3c67aaa10c9f47f7b5801562925bc2 upstream.
Check for invalid file type in xfs_dinode_verify() and fail to load the inode structure from disk.
--- fs/xfs/libxfs/xfs_inode_buf.c | 3 +++ 1 file changed, 3 insertions(+)
fs/xfs/libxfs/xfs_inode_buf.c +3 -0
@@ -29,6 +29,7 @@#include "xfs_icache.h" #include "xfs_trans.h" #include "xfs_ialloc.h" +#include "xfs_dir2.h" /* * Check that none of the inode's in the buffer have a next
@@ -398,6 +399,8 @@ xfs_dinode_verify(return false; mode =be16_to_cpu (dip->di_mode); +if (mode &&xfs_mode_to_ftype (mode) ==XFS_DIR3_FT_UNKNOWN ) +return false;/* No zero-length symlinks/dirs. */ if ((S_ISLNK (mode) ||S_ISDIR (mode)) && dip->di_size ==0 )
Greg Kroah-Hartman
maintainerPATCH 44/51 xfs: dont wrap ID in xfs_dq_get_next_id
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Sandeen <[email protected]>
commit 657bdfb7f5e68ca5e2ed009ab473c429b0d6af85 upstream.
The GETNEXTQOTA ioctl takes whatever ID is sent in, and looks for the next active quota for an user equal or higher to that ID.
But if we are at the maximum ID and then ask for the "next" one, we may wrap back to zero. In this case, userspace may loop forever, because it will start querying again at zero.
We'll fix this in userspace as well, but for the kernel, return -ENOENT if we ask for the next quota ID past UINT_MAX so the caller knows to stop.
--- fs/xfs/xfs_dquot.c | 4 ++++ 1 file changed, 4 insertions(+)
fs/xfs/xfs_dquot.c +4 -0
@@ -710,6 +710,10 @@ xfs_dq_get_next_id(/* Simple advance */ next_id = *id +1 ; +/* If we'd wrap past the max ID, stop */ +if (next_id < *id) +return -ENOENT ; +/* If new ID is within the current chunk, advancing it sufficed */ if (next_id % mp->m_quotainfo->qi_dqperchunk) { *id = next_id;
Greg Kroah-Hartman
maintainerPATCH 45/51 xfs: fix xfs_mode_to_ftype() prototype
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arnd Bergmann <[email protected]>
commit fd29f7af75b7adf250beccffa63746c6a88e2b74 upstream.
A harmless warning just got introduced:
fs/xfs/libxfs/xfs_dir2.h:40:8: error: type qualifiers ignored on function return type [-Werror=ignored-qualifiers]
Removing the 'const' modifier avoids the warning and has no other effect.
--- fs/xfs/libxfs/xfs_dir2.c | 2 +- fs/xfs/libxfs/xfs_dir2.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
fs/xfs/libxfs/xfs_dir2.c +1 -1
@@ -38,7 +38,7 @@ struct xfs_name xfs_name_dotdot = { (uns/* * Convert inode mode to directory entry filetype */ -const unsigned char xfs_mode_to_ftype (int mode) +unsigned char xfs_mode_to_ftype (int mode) {switch (mode &S_IFMT ) {case S_IFREG :
fs/xfs/libxfs/xfs_dir2.h +1 -1
@@ -37,7 +37,7 @@ extern struct xfs_name xfs_name_dotdot;/* * Convert inode mode to directory entry filetype */ -extern const unsigned char xfs_mode_to_ftype (int mode); +extern unsigned char xfs_mode_to_ftype (int mode);/* * directory operations vector for encode/decode routines
Greg Kroah-Hartman
maintainerPATCH 46/51 xfs: fix COW writeback race
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christoph Hellwig <[email protected]>
commit d2b3964a0780d2d2994eba57f950d6c9fe489ed8 upstream.
Due to the way how xfs_iomap_write_allocate tries to convert the whole found extents from delalloc to real space we can run into a race condition with multiple threads doing writes to this same extent. For the non-COW case that is harmless as the only thing that can happen is that we call xfs_bmapi_write on an extent that has already been converted to a real allocation. For COW writes where we move the extent from the COW to the data fork after I/O completion the race is, however, not quite as harmless. In the worst case we are now calling xfs_bmapi_write on a region that contains hole in the COW work, which will trip up an assert in debug builds or lead to file system corruption in non-debug builds. This seems to be reproducible with workloads of small O_DSYNC write, although so far I've not managed to come up with a with an isolated reproducer.
The fix for the issue is relatively simple: tell xfs_bmapi_write that we are only asked to convert delayed allocations and skip holes in that case.
--- fs/xfs/libxfs/xfs_bmap.c | 44 ++++++++++++++++++++++++++++++++------------ fs/xfs/libxfs/xfs_bmap.h | 6 +++++- fs/xfs/xfs_iomap.c | 2 +- 3 files changed, 38 insertions(+), 14 deletions(-)
fs/xfs/libxfs/xfs_bmap.c +32 -12
@@ -4607,8 +4607,6 @@ xfs_bmapi_write(int n;/* current extent index */ xfs_fileoff_t obno;/* old block number (offset) */ int whichfork;/* data or attr fork */ -char inhole;/* current location is hole in file */ -char wasdelay;/* old extent was delayed */ #ifdef DEBUG xfs_fileoff_t orig_bno;/* original block number value */
@@ -4694,22 +4692,44 @@ xfs_bmapi_write( bma.firstblock = firstblock;while (bno < end && n < *nmap) { - inhole = eof || bma.got.br_startoff > bno; - wasdelay = !inhole &&isnullstartblock (bma.got.br_startblock); +bool need_alloc = false, wasdelay = false; -/* -* Make sure we only reflink into a hole. -*/ -if (flags &XFS_BMAPI_REMAP ) -ASSERT (inhole); -if (flags &XFS_BMAPI_COWFORK ) -ASSERT (!inhole); +/* in hole or beyoned EOF? */ +if (eof || bma.got.br_startoff > bno) { +if (flags &XFS_BMAPI_DELALLOC ) { +/* +* For the COW fork we can reasonably get a +* request for converting an extent that races +* with other threads already having converted +* part of it, as there converting COW to +* regular blocks is not protected using the +* IOLOCK. +*/ +ASSERT (flags &XFS_BMAPI_COWFORK ); +if (!(flags &XFS_BMAPI_COWFORK )) { + error = -EIO ; + goto error0; + } + +if (eof || bno >= end) +break ; + }else { + need_alloc = true; + } + }else { +/* +* Make sure we only reflink into a hole. +*/ +ASSERT (!(flags &XFS_BMAPI_REMAP )); +if (isnullstartblock (bma.got.br_startblock)) + wasdelay = true; + }/* * First, deal with the hole before the allocated space * that we found, if any. */ -if (inhole || wasdelay) { +if (need_alloc || wasdelay) { bma.eof = eof; bma.conv = !!(flags &XFS_BMAPI_CONVERT ); bma.wasdel = wasdelay;
fs/xfs/libxfs/xfs_bmap.h +5 -1
@@ -110,6 +110,9 @@ struct xfs_extent_free_item/* Map something in the CoW fork. */ #define XFS_BMAPI_COWFORK 0x200 +/* Only convert delalloc space, don't allocate entirely new extents */ +#define XFS_BMAPI_DELALLOC 0x400 +#define XFS_BMAPI_FLAGS \ { XFS_BMAPI_ENTIRE, "ENTIRE" }, \ { XFS_BMAPI_METADATA, "METADATA" }, \
@@ -120,7 +123,8 @@ struct xfs_extent_free_item { XFS_BMAPI_CONVERT, "CONVERT" }, \ { XFS_BMAPI_ZERO, "ZERO" }, \ { XFS_BMAPI_REMAP, "REMAP" }, \ - { XFS_BMAPI_COWFORK, "COWFORK" } + { XFS_BMAPI_COWFORK, "COWFORK" }, \ + { XFS_BMAPI_DELALLOC, "DELALLOC" } static inline int xfs_bmapi_aflag(int w)
fs/xfs/xfs_iomap.c +1 -1
@@ -681,7 +681,7 @@ xfs_iomap_write_allocate(xfs_trans_t *tp;int nimaps;int error =0 ; -int flags =0 ; +int flags =XFS_BMAPI_DELALLOC ;int nres;if (whichfork ==XFS_COW_FORK )
Greg Kroah-Hartman
maintainerPATCH 47/51 xfs: verify dirblocklog correctly
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: "Darrick J. Wong" <[email protected]>
commit 83d230eb5c638949350f4761acdfc0af5cb1bc00 upstream.
sb_dirblklog is added to sb_blocklog to compute the directory block size in bytes. Therefore, we must compare the sum of both those values against XFS_MAX_BLOCKSIZE_LOG, not just dirblklog.
--- fs/xfs/libxfs/xfs_sb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/xfs/libxfs/xfs_sb.c +1 -1
@@ -242,7 +242,7 @@ xfs_mount_validate_sb( sbp->sb_blocklog <XFS_MIN_BLOCKSIZE_LOG || sbp->sb_blocklog >XFS_MAX_BLOCKSIZE_LOG || sbp->sb_blocksize != (1 << sbp->sb_blocklog) || - sbp->sb_dirblklog >XFS_MAX_BLOCKSIZE_LOG || + sbp->sb_dirblklog + sbp->sb_blocklog >XFS_MAX_BLOCKSIZE_LOG || sbp->sb_inodesize <XFS_DINODE_MIN_SIZE || sbp->sb_inodesize >XFS_DINODE_MAX_SIZE || sbp->sb_inodelog < XFS_DINODE_MIN_LOG ||
Greg Kroah-Hartman
maintainerPATCH 48/51 xfs: remove racy hasattr check from attr ops
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Brian Foster <[email protected]>
commit 5a93790d4e2df73e30c965ec6e49be82fc3ccfce upstream.
xfs_attr_[get|remove]() have unlocked attribute fork checks to optimize away a lock cycle in cases where the fork does not exist or is otherwise empty. This check is not safe, however, because an attribute fork short form to extent format conversion includes a transient state that causes the xfs_inode_hasattr() check to fail. Specifically, xfs_attr_shortform_to_leaf() creates an empty extent format attribute fork and then adds the existing shortform attributes to it.
This means that lookup of an existing xattr can spuriously return -ENOATTR when racing against a setxattr that causes the associated format conversion. This was originally reproduced by an untar on a particularly configured glusterfs volume, but can also be reproduced on demand with properly crafted xattr requests.
The format conversion occurs under the exclusive ilock. xfs_attr_get() and xfs_attr_remove() already have the proper locking and checks further down in the functions to handle this situation correctly. Drop the unlocked checks to avoid the spurious failure and rely on the existing logic.
--- fs/xfs/libxfs/xfs_attr.c | 6 ------ 1 file changed, 6 deletions(-)
fs/xfs/libxfs/xfs_attr.c +0 -6
@@ -131,9 +131,6 @@ xfs_attr_get(if (XFS_FORCED_SHUTDOWN (ip->i_mount))return -EIO ; -if (!xfs_inode_hasattr (ip)) -return -ENOATTR ; - error =xfs_attr_args_init (&args, ip, name, flags);if (error)return error;
@@ -392,9 +389,6 @@ xfs_attr_remove(if (XFS_FORCED_SHUTDOWN (dp->i_mount))return -EIO ; -if (!xfs_inode_hasattr (dp)) -return -ENOATTR ; - error =xfs_attr_args_init (&args, dp, name, flags);if (error)return error;
Greg Kroah-Hartman
maintainerPATCH 49/51 xfs: extsize hints are not unlikely in xfs_bmap_btalloc
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christoph Hellwig <[email protected]>
commit 493611ebd62673f39e2f52c2561182c558a21cb6 upstream.
With COW files they are the hotpath, just like for files with the extent size hint attribute. We really shouldn't micro-manage anything but failure cases with unlikely.
Additionally Arnd Bergmann recently reported that one of these two unlikely annotations causes link failures together with an upcoming kernel instrumentation patch, so let's get rid of it ASAP.
--- fs/xfs/libxfs/xfs_bmap.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
fs/xfs/libxfs/xfs_bmap.c +2 -2
@@ -3720,7 +3720,7 @@ xfs_bmap_btalloc( align =xfs_get_cowextsz_hint (ap->ip); else if (xfs_alloc_is_userdata(ap->datatype)) align =xfs_get_extsz_hint (ap->ip); -if (unlikely (align)) { +if (align) { error =xfs_bmap_extsize_align (mp, &ap->got, &ap->prev, align,0 , ap->eof,0 , ap->conv, &ap->offset, &ap->length);
@@ -3792,7 +3792,7 @@ xfs_bmap_btalloc( args.minlen = ap->minlen; }/* apply extent size hints if obtained earlier */ -if (unlikely (align)) { +if (align) { args.prod = align;if ((args.mod = (xfs_extlen_t )do_mod (ap->offset, args.prod))) args.mod = (xfs_extlen_t)(args.prod - args.mod);
Greg Kroah-Hartman
maintainerPATCH 50/51 xfs: clear _XBF_PAGES from buffers when readahead page
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: "Darrick J. Wong" <[email protected]>
commit 2aa6ba7b5ad3189cc27f14540aa2f57f0ed8df4b upstream.
If we try to allocate memory pages to back an xfs_buf that we're trying to read, it's possible that we'll be so short on memory that the page allocation fails. For a blocking read we'll just wait, but for readahead we simply dump all the pages we've collected so far.
Unfortunately, after dumping the pages we neglect to clear the _XBF_PAGES state, which means that the subsequent call to xfs_buf_free thinks that b_pages still points to pages we own. It then double-frees the b_pages pages.
This results in screaming about negative page refcounts from the memory manager, which xfs oughtn't be triggering. To reproduce this case, mount a filesystem where the size of the inodes far outweighs the availalble memory (a ~500M inode filesystem on a VM with 300MB memory did the trick here) and run bulkstat in parallel with other memory eating processes to put a huge load on the system. The "check summary" phase of xfs_scrub also works for this purpose.
--- fs/xfs/xfs_buf.c | 1 + 1 file changed, 1 insertion(+)
fs/xfs/xfs_buf.c +1 -0
@@ -423,6 +423,7 @@ retry: out_free_pages:for (i =0 ; i < bp->b_page_count; i++)__free_page (bp->b_pages[i]); + bp->b_flags &= ~_XBF_PAGES;return error; }
Greg Kroah-Hartman
maintainerPATCH 51/51 xfs: fix bmv_count confusion w/ shared extents
In reply to Greg Kroah-Hartman4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: "Darrick J. Wong" <[email protected]>
commit c364b6d0b6cda1cd5d9ab689489adda3e82529aa upstream.
In a bmapx call, bmv_count is the total size of the array, including the zeroth element that userspace uses to supply the search key. The output array starts at offset 1 so that we can set up the user for the next invocation. Since we now can split an extent into multiple bmap records due to shared/unshared status, we have to be careful that we don't overflow the output array.
In the original patch f86f403794b ("xfs: teach get_bmapx about shared extents and the CoW fork") I used cur_ext (the output index) to check for overflows, albeit with an off-by-one error. Since nexleft no longer describes the number of unfilled slots in the output, we can rip all that out and use cur_ext for the overflow check directly.
Failure to do this causes heap corruption in bmapx callers such as xfs_io and xfs_scrub. xfs/328 can reproduce this problem.
--- fs/xfs/xfs_bmap_util.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-)
fs/xfs/xfs_bmap_util.c +18 -10
@@ -528,7 +528,6 @@ xfs_getbmap(xfs_bmbt_irec_t *map;/* buffer for user's data */ xfs_mount_t *mp;/* file system mount point */ int nex;/* # of user extents can do */ -int nexleft;/* # of user extents left */ int subnex;/* # of bmapi's can do */ int nmap;/* number of map entries */ struct getbmapx *out;/* output structure */
@@ -686,10 +685,8 @@ xfs_getbmap( goto out_free_map; } - nexleft = nex; - do { - nmap = (nexleft > subnex) ? subnex : nexleft; + nmap = (nex> subnex) ? subnex : nex; error = xfs_bmapi_read(ip, XFS_BB_TO_FSBT(mp, bmv->bmv_offset), XFS_BB_TO_FSB(mp, bmv->bmv_length), map, &nmap, bmapi_flags);
@@ -697,8 +694,8 @@ xfs_getbmap( goto out_free_map;ASSERT (nmap <= subnex); -for (i =0 ; i < nmap && nexleft && bmv->bmv_length && - cur_ext < bmv->bmv_count; i++) { +for (i =0 ; i < nmap && bmv->bmv_length && + cur_ext < bmv->bmv_count -1 ; i++) { out[cur_ext].bmv_oflags =0 ;if (map[i].br_state ==XFS_EXT_UNWRITTEN ) out[cur_ext].bmv_oflags |=BMV_OF_PREALLOC ;
@@ -760,16 +757,27 @@ xfs_getbmap(continue ; } +/* +* In order to report shared extents accurately, +* we report each distinct shared/unshared part +* of a single bmbt record using multiple bmap +* extents. To make that happen, we iterate the +* same map array item multiple times, each +* time trimming out the subextent that we just +* reported. +* +* Because of this, we must check the out array +* index (cur_ext) directly against bmv_count-1 +* to avoid overflows. +*/ if (inject_map.br_startblock !=NULLFSBLOCK ) { map[i] = inject_map; i--; - }else - nexleft--; + } bmv->bmv_entries++; cur_ext++; } - }while (nmap && nexleft && bmv->bmv_length && - cur_ext < bmv->bmv_count); + }while (nmap && bmv->bmv_length && cur_ext < bmv->bmv_count -1 ); out_free_map:kmem_free (map);
Quoted 9 lines · Greg Kroah-Hartman
This is the start of the stable review cycle for the 4.9.8 release. There are 51 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Sat Feb 4 18:33:31 UTC 2017. Anything received after that time might be too late.
The whole patch series can be found in one patch at:
Quoted 9 lines
kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.9.8-rc1.gzor in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.gitlinux-4.9.y
and the diffstat can be found below.thanks,
greg k-h
Compiled and booted on my test system. No dmesg regressions.
thanks,
-- Shuah
Greg Kroah-Hartman
maintainerQuoted 15 lines · Shuah Khan
Quoted 14 lines · Greg Kroah-Hartman
This is the start of the stable review cycle for the 4.9.8 release. There are 51 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Sat Feb 4 18:33:31 UTC 2017. Anything received after that time might be too late.
The whole patch series can be found in one patch at:
kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.9.8-rc1.gzor in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.gitlinux-4.9.y
and the diffstat can be found below.
Quoted 6 lines
Quoted 4 lines
thanks,
greg k-h
Compiled and booted on my test system. No dmesg regressions.
Great, thanks for letting me know for both of these.
greg k-h
Quoted 8 lines · Greg Kroah-Hartman
This is the start of the stable review cycle for the 4.9.8 release. There are 51 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Sat Feb 4 18:33:31 UTC 2017. Anything received after that time might be too late.
Build results:
total: 149 pass: 149 fail: 0 Qemu test results: total: 122 pass: 122 fail: 0
Details are available at http://kerneltests.org/builders.
Guenter
Greg Kroah-Hartman
maintainerQuoted 10 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman
This is the start of the stable review cycle for the 4.9.8 release. There are 51 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Sat Feb 4 18:33:31 UTC 2017. Anything received after that time might be too late.
Quoted 6 lines
Build results:
total: 149 pass: 149 fail: 0 Qemu test results: total: 122 pass: 122 fail: 0Details are available at http://kerneltests.org/builders.
Great, thanks for testing these and letting me know.
greg k-h