- 07 Jul, 2018 37 commits
-
-
Jon Maloy authored
Currently a link is declared stale and reset if there has been 100 repeated attempts to retransmit the same packet. However, in certain infrastructures we see that packet (NACK) duplicates and delays may cause such retransmit attempts to occur at a high rate, so that the peer doesn't have a reasonable chance to acknowledge the reception before the 100-limit is hit. This may take much less than the stipulated link tolerance time, and despite that probe/probe replies otherwise go through as normal. We now extend the criteria for link reset to also being time based. I.e., we don't reset the link until the link tolerance time is passed AND we have made 100 retransmissions attempts. Acked-by:
Ying Xue <ying.xue@windriver.com> Signed-off-by:
Jon Maloy <jon.maloy@ericsson.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
David S. Miller authored
Jianbo Liu says: ==================== Introduce matching on double vlan/QinQ headers for TC flower Currently TC flower supports only one vlan tag, it doesn't match on both outer and inner vlan headers for QinQ. To do this, we add support to get both outer and inner vlan headers for flow dissector, and then TC flower do matching on those information. We also plan to extend TC command to support this feature. We add new cvlan_id/cvlan_prio/cvlan_ethtype keywords for inner vlan header. The existing vlan_id/vlan_prio/vlan_ethtype are for outer vlan header, and vlan_ethtype must be 802.1q or 802.1ad. The examples for command and output are as the following. flower vlan_id 1000 vlan_ethtype 802.1q \ cvlan_id 100 cvlan_ethtype ipv4 \ action vlan pop \ action vlan pop \ action mirred egress redirect dev ens1f1_0 filter protocol 802.1ad pref 33 flower chain 0 filter protocol 802.1ad pref 33 flower chain 0 handle 0x1 vlan_id 1000 vlan_ethtype 802.1Q cvlan_id 100 cvlan_ethtype ip eth_type ipv4 in_hw ... v2: fix sparse warning. ==================== Signed-off-by: -
Jianbo Liu authored
As support dissecting of QinQ inner and outer vlan headers, user can add rules to match on QinQ vlan headers. Signed-off-by:
Jianbo Liu <jianbol@mellanox.com> Acked-by:
Jiri Pirko <jiri@mellanox.com> Signed-off-by:
-
Jianbo Liu authored
Currently the encapsulated ethertype is not dumped as it's the same as TCA_FLOWER_KEY_ETH_TYPE keyvalue. But the dumping result is inconsistent with input, we add dumping it with TCA_FLOWER_KEY_VLAN_ETH_TYPE. Signed-off-by:
-
Jianbo Liu authored
Dissect the QinQ packets to get both outer and inner vlan information, then store to the extended flow keys. Signed-off-by:
-
Jianbo Liu authored
As flow dissector stores vlan ethertype, tc flower now can match on that. It is to make preparation for supporting QinQ. Signed-off-by:
-
Jianbo Liu authored
Change vlan dissector key to save vlan tpid to support both 802.1Q and 802.1AD ethertype. Signed-off-by:
-
Colin Ian King authored
Variable txq_length is being assigned but is never used hence it is redundant and can be removed. Cleans up clang warning: warning: variable 'txq_length' set but not used [-Wunused-but-set-variable] Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
David S. Miller authored
Petr Machata says: ==================== Add a mlxsw-specific test for mirror-to-gretap Some configurations of mirror-to-gretap are impossible for mlxsw to offload. Add a test that checks that these out-of-domain conditions are handled properly by mlxsw. In patch #1, fix mirror_gre_lib.sh and mirror_gre_topo_lib.sh so that they can be imported from directories other than forwarding/. In patch #2, add a test to check handling of several scenarios that mlxsw is expected to fail to offload. ==================== Signed-off-by: -
Petr Machata authored
Test several aspects of offloading mirror to gretap and ip6gretap netdevices that are specific to mlxsw, such as requirements for TTL and TOS values. Signed-off-by:
Petr Machata <petrm@mellanox.com> Reviewed-by:
-
Petr Machata authored
The next patch introduces a new mlxsw-specific test that uses mirror_gre_lib.sh and mirror_gre_topo_lib.sh. However when sourcing their own deps, these libraries assume that the test that's running is in the same directory. That's not the case for driver-specific tests. So change the libraries to source their deps through $relative_path. That variable is set up by lib.sh, which should be imported by the test in question in any case. Signed-off-by:
-
David S. Miller authored
Roopa Prabhu says: ==================== vxlan: fix default fdb entry user-space notify ordering/race Problem: In vxlan_newlink, a default fdb entry is added before register_netdev. The default fdb creation function notifies user-space of the fdb entry on the vxlan device which user-space does not know about yet. (RTM_NEWNEIGH goes before RTM_NEWLINK for the same ifindex). This series fixes the user-space netlink notification ordering issue with the following changes: - decouple fdb notify from fdb create. - Move fdb notify after register_netdev. - modify rtnl_configure_link to allow configuring a link early. - Call rtnl_configure_link in vxlan newlink handler to notify userspace about the newlink before fdb notify and hence avoiding the user-space race. ==================== Fixes: afbd8bae ("vxlan: add implicit fdb entry for default destination") Signed-off-by:
Roopa Prabhu <roopa@cumulusnetworks.com>
-
Roopa Prabhu authored
Problem: In vxlan_newlink, a default fdb entry is added before register_netdev. The default fdb creation function also notifies user-space of the fdb entry on the vxlan device which user-space does not know about yet. (RTM_NEWNEIGH goes before RTM_NEWLINK for the same ifindex). This patch fixes the user-space netlink notification ordering issue with the following changes: - decouple fdb notify from fdb create. - Move fdb notify after register_netdev. - Call rtnl_configure_link in vxlan newlink handler to notify userspace about the newlink before fdb notify and hence avoiding the user-space race. Fixes: afbd8bae ("vxlan: add implicit fdb entry for default destination") Signed-off-by:
-
Roopa Prabhu authored
Add a new option do_notify to vxlan_fdb_destroy to make sending netlink notify optional. Used by a later patch. Signed-off-by:
-
Roopa Prabhu authored
- Add new vxlan_fdb_alloc helper - rename existing vxlan_fdb_create into vxlan_fdb_update: because it really creates or updates an existing fdb entry - move new fdb creation into a separate vxlan_fdb_create Main motivation for this change is to introduce the ability to decouple vxlan fdb creation and notify, used in a later patch. Signed-off-by: -
Roopa Prabhu authored
rtnl_configure_link sets dev->rtnl_link_state to RTNL_LINK_INITIALIZED and unconditionally calls __dev_notify_flags to notify user-space of dev flags. current call sequence for rtnl_configure_link rtnetlink_newlink rtnl_link_ops->newlink rtnl_configure_link (unconditionally notifies userspace of default and new dev flags) If a newlink handler wants to call rtnl_configure_link early, we will end up with duplicate notifications to user-space. This patch fixes rtnl_configure_link to check rtnl_link_state and call __dev_notify_flags with gchanges = 0 if already RTNL_LINK_INITIALIZED. Later in the series, this patch will help the following sequence where a driver implementing newlink can call rtnl_configure_link to initialize the link early. makes the following call sequence work: rtnetlink_newlink rtnl_link_ops->newlink (vxlan) -> rtnl_configure_link (initializes link and notifies user-space of default dev flags) rtnl_configure_link (updates dev flags if requested by user ifm and notifies user-space of new dev flags) Signed-off-by: -
Arnd Bergmann authored
A newly added dummy helper function tries to return a failure from a "void" function: In file included from include/net/dsa.h:24, from arch/arm/plat-orion/common.c:21: include/net/devlink.h: In function 'devlink_param_value_changed': include/net/devlink.h:771:9: error: 'return' with a value, in function returning void [-Werror] return -EOPNOTSUPP; This fixes it by removing the bogus statement. Fixes: ea601e17 ("devlink: Add devlink notifications support for params") Signed-off-by:Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
Arnd Bergmann authored
When CONFIG_BRIDGE_VLAN_FILTERING is disabled, gcc correctly points out that the 'vid' variable is uninitialized whenever br_vlan_get_pvid returns an error: drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c: In function 'mlxsw_sp_rif_vlan_fid_get': drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c:6881:6: error: 'vid' may be used uninitialized in this function [-Werror=maybe-uninitialized] This changes the condition check to always return -EINVAL here, which I guess is what the author intended here. Fixes: e6f1960a ("mlxsw: spectrum_router: Allocate FID according to PVID") Signed-off-by:
Ido Schimmel <idosch@mellanox.com> Signed-off-by:
-
David S. Miller authored
Salil Mehta says: ==================== Misc. bug fixes & cleanups for HNS3 driver This patch-set presents some miscellaneous bug fixes and cleanups for the HNS3 driver. ==================== Signed-off-by: -
Huazhong Tan authored
According to hardware's description, driver should not send command to IMP while hardware doing global or core reset. Signed-off-by:
Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by:
Peng Li <lipeng321@huawei.com> Signed-off-by:
Salil Mehta <salil.mehta@huawei.com> Signed-off-by:
-
Peng Li authored
Only the core/global/IMP reset need clear cause, other type does not need do it. The warning may be treated as error as it is normal. This patch removes the warning. Signed-off-by:
-
Yunsheng Lin authored
The hclgevf_free_vector function expects the caller to pass the vector_id to it, and hclgevf_put_vector pass vector to it now, which will cause vector allocation problem. This patch fixes it by converting vector into vector_id before calling hclgevf_free_vector. Fixes: e2cb1dec ("net: hns3: Add HNS3 VF HCL(Hardware Compatibility Layer) Support") Signed-off-by:
Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by:
-
Yunsheng Lin authored
The napi_alloc_skb is excepted to be called under the non-preemptible code path when it is called by hns3_clean_rx_ring during loopback selftest, otherwise the below warning will be logged: [ 92.420780] BUG: using smp_processor_id() in preemptible [00000000] code: ethtool/1873 <SNIP> [ 92.463202] check_preemption_disabled+0xf8/0x100 [ 92.467893] debug_smp_processor_id+0x1c/0x28 [ 92.472239] __napi_alloc_skb+0x30/0x130 [ 92.476158] hns3_clean_rx_ring+0x118/0x5f0 [hns3] [ 92.480941] hns3_self_test+0x32c/0x4d0 [hns3] [ 92.485375] ethtool_self_test+0xdc/0x1e8 [ 92.489372] dev_ethtool+0x1020/0x1da8 [ 92.493109] dev_ioctl+0x188/0x3a0 [ 92.496499] sock_do_ioctl+0xf4/0x208 [ 92.500148] sock_ioctl+0x228/0x3e8 [ 92.503626] do_vfs_ioctl+0xc4/0x880 [ 92.507189] SyS_ioctl+0x94/0xa8 [ 92.510404] el0_svc_naked+0x30/0x34 This patch fix it by disabling preemption when calling hns3_clean_rx_ring during loopback selftest. Fixes: c39c4d98 ("net: hns3: Add mac loopback selftest support in hns3 driver") Signed-off-by:
-
Jian Shen authored
When change the mtu, the minimal frame size of mac will be set to zero, it is incorrect. This patch fixes it by set it to the default value. Signed-off-by:
Jian Shen <shenjian15@huawei.com> Signed-off-by:
-
Fuyun Liang authored
The payload of mailbox message is 16 byte and the value of HCLGE_MBX_MAX_ARQ_MSG_SIZE is 8. A message truncated problem will happen when mailbox message is converted to ARQ message. This patch replaces HCLGE_MBX_MAX_ARQ_MSG_SIZE with the size of ARQ message in hclgevf_mbx_handler to fix this problem. Fixes: b11a0bb2 ("net: hns3: Add mailbox support to VF driver") Signed-off-by:
Fuyun Liang <liangfuyun1@huawei.com> Signed-off-by:
-
Yunsheng Lin authored
Hardware only support tcp/udp/sctp l4 checksum offload, but the driver currently tell hardware to do l4 checksum offlad when l3 is IPv4 or IPv6, which may cause checksumm error. This patch fixes it by only enabling the l4 offload when l4 is tcp/udp/sctp. Fixes: 76ad4f0e ("net: hns3: Add support of HNS3 Ethernet Driver for hip08 SoC") Signed-off-by:
-
Yunsheng Lin authored
The HCLGE_RX_PRIV_EN_B is used to tell the firmware whether to update the specific waterline value, if the is not set, the firmware will ignore the value. This patch fixes by setting the HCLGE_RX_PRIV_EN_B even if the updated value is zero. Fixes: 46a3df9f ("net: hns3: Add HNS3 Acceleration Engine & Compatibility Layer Support") Signed-off-by:
-
Yunsheng Lin authored
When pfc pause mode is enable, the mac pause mode need to be disabled, otherwise the pfc pause packet will not be sent when congestion happens. This patch fixes by disabling the mac pause when pfc pause is enabled. Fixes: 84844054 ("net: hns3: Add support of TX Scheduler & Shaper to HNS3 driver") Signed-off-by:
-
Yunsheng Lin authored
Currently, tc related configuration is not setup when the netdev is first up, which cause the stack only using tc 0 problem. This patch fixes it by setting the tc related configuration using the info from NCL_CONFIG when netdev is first up. Fixes: 76ad4f0e ("net: hns3: Add support of HNS3 Ethernet Driver for hip08 SoC") Signed-off-by:
-
Rick Farrington authored
Pre-1.7.2 NIC firmware does not support (and does not respond to) the "get speed" command which is sent by the 1.7.2 driver (for CN23XX-225 cards only) during modprobe. Due to a bug in older firmware (with respect to unknown commands), this unsupported command causes a cascade of errors that ends in a kernel panic. Fix it by making the sending of the "get speed" command conditional on the firmware version. Signed-off-by:
Rick Farrington <ricardo.farrington@cavium.com> Signed-off-by:
Felix Manlunas <felix.manlunas@cavium.com> Signed-off-by:
-
David S. Miller authored
Willem de Bruijn says: ==================== sock cookie initializers Recent UDP GSO and SO_TXTIME features added new fields to cookie structs. When adding a field, all sites where a struct is initialized have to be updated, which is a lot of boilerplate. Alternatively, a field can be initialized selectively, but this is fragile. I introduced a bug in udp gso where an uninitialized field was read. See also fix commit ("9887cba1 ip: limit use of gso_size to udp"). Introduce initializers for structs ipcm(6)_cookie and sockc_cookie. patch 1..3 do exactly this. patch 4..5 make ipv4 and ipv6 handle cookies the same way and remove some boilerplate in doing so. patch 6 removes the udp gso branch that needed the above fix ==================== Signed-off-by: -
Willem de Bruijn authored
Now that ipc(6)->gso_size is correctly initialized in all callers of ip(6)_setup_cork, it is safe to unconditionally pass it to the cork. Link: http://lkml.kernel.org/r/20180619164752.143249-1-willemdebruijn.kernel@gmail.comSigned-off-by:
Willem de Bruijn <willemb@google.com> Signed-off-by:
-
Willem de Bruijn authored
skb_shinfo(skb)->tx_flags is derived from sk->sk_tsflags, possibly after modification by __sock_cmsg_send, by calling sock_tx_timestamp. The IPv4 and IPv6 paths do this conversion differently. In IPv4, the individual protocols that support tx timestamps call this function and store the result in ipc.tx_flags. In IPv6, sock_tx_timestamp is called in __ip6_append_data. There is no need to store both tx_flags and ts_flags in the cookie as one is derived from the other. Convert when setting up the cork and remove the redundant field. This is similar to IPv6, only have the conversion happen only once per datagram, in ip(6)_setup_cork. Also change __ip6_append_data to match __ip_append_data. Only update tskey if timestamping is enabled with OPT_ID. The SOCK_.. test is redundant: only valid protocols can have non-zero cork->tx_flags. After this change the IPv4 and IPv6 logic is the same. Signed-off-by:
-
Willem de Bruijn authored
ipcm_cookie includes sockcm_cookie. Do the same for ipcm6_cookie. This reduces the number of arguments that need to be passed around, applies ipcm6_init to all cookie fields at once and reduces code differentiation between ipv4 and ipv6. Signed-off-by:
-
Willem de Bruijn authored
Initialize the cookie in one location to reduce code duplication and avoid bugs from inconsistent initialization, such as that fixed in commit 9887cba1 ("ip: limit use of gso_size to udp"). Signed-off-by:
-
Willem de Bruijn authored
Initialize the cookie in one location to reduce code duplication and avoid bugs from inconsistent initialization, such as that fixed in commit 9887cba1 ("ip: limit use of gso_size to udp"). Signed-off-by:
-
Willem de Bruijn authored
Initialize the cookie in one location to reduce code duplication and avoid bugs from inconsistent initialization, such as that fixed in commit 9887cba1 ("ip: limit use of gso_size to udp"). Signed-off-by:
-
- 06 Jul, 2018 3 commits
-
-
David S. Miller authored
Edward Cree says: ==================== IP listification follow-ups While working on IPv6 list processing, I found another bug in the IPv4 version. So this patch series has that fix, and the IPv6 version with both fixes incorporated. ==================== Signed-off-by: -
Edward Cree authored
Essentially the same as the ipv4 equivalents. Signed-off-by:
Edward Cree <ecree@solarflare.com> Signed-off-by:
-
Edward Cree authored
If we have an L3 master device, l3mdev_ip_rcv() will steal the skb, but we were returning NET_RX_SUCCESS from ip_rcv_finish_core() which meant that ip_list_rcv_finish() would keep it on the list. Instead let's move the l3mdev_ip_rcv() call into the caller, so that our response to a steal can be different in the single packet path (return NET_RX_SUCCESS) and the list path (forget this packet and continue). Fixes: 5fa12739 ("net: ipv4: listify ip_rcv_finish") Signed-off-by:
-
