- 25 May, 2018 40 commits
-
-
John Hurley authored
Check if the fw contains the _abi_flower_balance_sync_enable symbol. If it does then write a 1 to this indicating that the driver is willing to receive NIC to kernel LAG related control messages. If the write is successful, update the list of extra features supported by the fw and add a stub to accept LAG cmsgs. Signed-off-by:
John Hurley <john.hurley@netronome.com> Reviewed-by:
Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by:
Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
John Hurley authored
Add an rtsym API function that combines the lookup of a symbol and the writing of a value to it. Values can be written as unsigned 32 or 64 bits. Signed-off-by:
-
John Hurley authored
Adding a netdev to a bond requires that its mac address can be modified. The default eth_mac_addr is sufficient to satisfy this requirement. Signed-off-by:
-
Stephen Hemminger authored
If the guest network adapter is not configured with DeviceNaming enabled on the host, then the query for friendly name will return success but with a zero length name. Which then leads to a garbage value (stack contents) for ifalias. Fix is simple, just don't set name if host doesn't return it. Fixes: 0fe554a4 ("hv_netvsc: propogate Hyper-V friendly name into interface alias") Signed-off-by:
Stephen Hemminger <sthemmin@microsoft.com> Signed-off-by:
-
David S. Miller authored
David Ahern says: ==================== net: Update fib_table_lookup tracepoints Update the FIB lookup tracepoints to include ip proto and port fields from the flow struct. In the process make the IPv4 tracepoint inline with IPv6 which is much easier to use and follow the lookup and result. Remove the tracepoint in fib_validate_source which does not provide value above the fib_table_lookup which immediately follows it. v2 - move CREATE_TRACE_POINTS for the v6 tracepoint to route.c to handle its need for an internal function to convert route type to error and handle IPv6 as a module or builtin. Reported by kbuild robot. ==================== Signed-off-by: -
David Ahern authored
Tracepoint does not add value and the call to fib_lookup follows it which shows the same information and the fib lookup result. Signed-off-by:
David Ahern <dsahern@gmail.com> Signed-off-by:
-
David Ahern authored
Commit bb0ad198 ("ipv6: fib6_rules: support for match on sport, dport and ip proto") added support for protocol and ports to FIB rules. Update the FIB lookup tracepoint to dump the parameters. Signed-off-by:
-
David Ahern authored
Commit 4a2d73a4 ("ipv4: fib_rules: support match on sport, dport and ip proto") added support for protocol and ports to FIB rules. Update the FIB lookup tracepoint to dump the parameters. In addition, make the IPv4 tracepoint similar to the IPv6 one where the lookup parameters and result are dumped in 1 event. It is much easier to use and understand the outcome of the lookup. Signed-off-by:
-
Cong Wang authored
Commit 05f0fe6b ("RCU, workqueue: Implement rcu_work") introduces new API's for dispatching work in a RCU callback. Now we can just switch to the new API's for tc filters. This could get rid of a lot of code. Cc: Tejun Heo <tj@kernel.org> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by:
Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by:
-
David S. Miller authored
Petr Machata says: ==================== Mirroring tests involving VLAN This patchset tests mirror-to-gretap with various underlay configurations involving VLAN netdevice in particular. Some of the tests involve bridges as well, but tests aimed specifically at testing bridges (i.e. FDB, STP) are not part of this patchset. In patches #1-#6, the codebase is adapted to support the new tests. In patch #7, a test for mirroring to VLAN is introduced. Patches #8-#10 add three tests where VLAN is part of underlay path after gretap encapsulation. ==================== Signed-off-by: -
Petr Machata authored
Test for "tc action mirred egress mirror" that mirrors to GRE when the underlay route points at an 802.1d bridge and packet egresses through a VLAN device. Besides testing basic connectivity, this also tests that the traffic is properly tagged. Signed-off-by:
Petr Machata <petrm@mellanox.com> Signed-off-by:
-
Petr Machata authored
Test for "tc action mirred egress mirror" that mirrors to a gretap netdevice whose underlay route points at a vlan device. Signed-off-by:
-
Petr Machata authored
Test for "tc action mirred egress mirror" that mirrors to GRE when the underlay route points at a vlan device on top of a bridge device with vlan filtering (802.1q). Signed-off-by:
-
Petr Machata authored
Test for "tc action mirred egress mirror" that mirrors to a vlan device. - test_vlan() tests that the packets get mirrored - test_tagged_vlan() tests that the mirrored packets have correct inner VLAN tag. Signed-off-by:
-
Petr Machata authored
A mirror-to-vlan test that's coming next needs to install the trap unconditionally. Therefore extract from slow_path_trap_{,un}install() a more generic functions trap_install() and trap_uninstall(), and covert the former two to conditional wrappers around these. Signed-off-by: -
Petr Machata authored
Add full_test_span_gre_dir_vlan_ips() and full_test_span_gre_dir_vlan() to support mirror-to-gre tests that involve VLAN. Signed-off-by:
-
Petr Machata authored
Add vlan_create() and vlan_destroy() to manage VLAN netdevices. Signed-off-by:
-
Petr Machata authored
Having a clsact qdisc on $h3 is useful in several tests, and will be useful in more tests to come. Move the registration from all the tests that need it into the topology file itself. Signed-off-by:
-
Petr Machata authored
For non-GRE mirroring tests, a functions along the lines of do_test_span_gre_dir_ips() and test_span_gre_dir_ips() are necessary, but such that they don't assume tunnels are involved. Extract the code from mirror_gre_lib.sh to mirror_lib.sh and convert to just use a given device without assuming it's named "h3-$tundev". Convert the two above-mentioned functions to wrappers that pass along the correct device name. Add test_span_dir() and fail_test_span_dir() to round up the API for use by following patches. Signed-off-by:
-
Petr Machata authored
Move generic parts of mirror_gre_topo_lib.sh into a new file mirror_topo_lib.sh. Reuse the functions in GRE topo, adding the tunnel devices as necessary. Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-nextDavid S. Miller authored
Alexei Starovoitov says: ==================== pull-request: bpf-next 2018-05-24 The following pull-request contains BPF updates for your *net-next* tree. The main changes are: 1) Björn Töpel cleans up AF_XDP (removes rebind, explicit cache alignment from uapi, etc). 2) David Ahern adds mtu checks to bpf_ipv{4,6}_fib_lookup() helpers. 3) Jesper Dangaard Brouer adds bulking support to ndo_xdp_xmit. 4) Jiong Wang adds support for indirect and arithmetic shifts to NFP 5) Martin KaFai Lau cleans up BTF uapi and makes the btf_header extensible. 6) Mathieu Xhonneux adds an End.BPF action to seg6local with BPF helpers allowing to edit/grow/shrink a SRH and apply on a packet generic SRv6 actions. 7) Sandipan Das adds support for bpf2bpf function calls in ppc64 JIT. 8) Yonghong Song adds BPF_TASK_FD_QUERY command for introspection of tracing events. 9) other misc fixes from Gustavo A. R. Silva, Sirio Balmelli, John Fastabend, and Magnus Karlsson ==================== Signed-off-by: -
David S. Miller authored
Thomas Falcon says: ==================== ibmvnic: Failover hardening Introduce additional transport event hardening to handle events during device reset. In the driver's current state, if a transport event is received during device reset, it can cause the device to become unresponsive as invalid operations are processed as the backing device context changes. After a transport event, the device expects a request to begin the initialization process. If the driver is still processing a previously queued device reset in this state, it is likely to fail as firmware will reject any commands other than the one to initialize the client driver's Command-Response Queue. Instead of failing and becoming dormant, the driver will make one more attempt to recover and continue operation. This is achieved by setting a state flag, which if true will direct the driver to clean up all allocated resources and perform a hard reset in an attempt to bring the driver back to an operational state. ==================== Signed-off-by: -
Thomas Falcon authored
Introduce a recovery hard reset to handle reset failure as a result of change of device context following a transport event, such as a backing device failover or partition migration. These operations reset the device context to its initial state. If this occurs during a reset, any initialization commands are likely to fail with an invalid state error as backing device firmware requests reinitialization. When this happens, make one more attempt by performing a hard reset, which frees any resources currently allocated and performs device initialization. If a transport event occurs during a device reset, a flag is set which will trigger a new hard reset following the completionof the current reset event. Signed-off-by:
Thomas Falcon <tlfalcon@linux.vnet.ibm.com> Signed-off-by:
-
Thomas Falcon authored
Set device resetting state at the earliest possible point: as soon as a reset is successfully scheduled. The reset state is toggled off when all resets have been processed to completion. Signed-off-by:
-
Thomas Falcon authored
Instead of having one initialization routine for all cases, create a separate, simpler function for standard initialization, such as during device probe. Use the original initialization function to handle device reset scenarios. The goal of this patch is to avoid having a single, cluttered init function to handle all possible scenarios. Signed-off-by:
-
Thomas Falcon authored
If setting the link state is not successful, print a warning with the resulting return code and return it to be handled by the caller. Signed-off-by:
-
Thomas Falcon authored
If device init is interrupted by a failover, set the init return code so that it can be checked and handled appropriately by the init routine. Signed-off-by:
-
Thomas Falcon authored
Check whether CRQ command is successful before awaiting a response from the management partition. If the command was not successful, the driver may hang waiting for a response that will never come. Signed-off-by:
-
Thomas Falcon authored
Introduce an "active" state for a IBM vNIC Command-Response Queue. A CRQ is considered active once it has initialized or linked with its partner by sending an initialization request and getting a successful response back from the management partition. Until this has happened, do not allow CRQ commands to be sent other than the initialization request. This change will avoid a protocol error in case of a device transport event occurring during a initialization. When the driver receives a transport event notification indicating that the backing hardware has changed and needs reinitialization, any further commands other than the initialization handshake with the VIOS management partition will result in an invalid state error. Instead of sending a command that will be returned with an error, print a warning and return an error that will be handled by the caller. Signed-off-by:
-
Thomas Falcon authored
Set adapter NAPI state as disabled if they are removed. This will allow them to be enabled again if reallocated in case of a hard reset. Signed-off-by:
-
David S. Miller authored
Petr Machata says: ==================== selftests: forwarding: Additions to mirror-to-gretap tests This patchset is for a handful of edge cases in mirror-to-gretap scenarios: removal of mirrored-to netdevice (#1), removal of underlay route for tunnel remote endpoint (#2) and cessation of mirroring upon removal of flower mirroring rule (#3). ==================== Signed-off-by: -
Petr Machata authored
Test that when flower-based mirror action is removed, mirroring stops. Signed-off-by:
-
Petr Machata authored
When underlay route is removed, the mirrored traffic should not be forwarded. Signed-off-by:
-
Petr Machata authored
Tests that the mirroring code catches up with deletion of a mirrored-to device. Signed-off-by:
-
YueHaibing authored
t4_prep_fw doesn't check for card_fw pointer before store the read data, which could lead to a NULL pointer dereference if kvzalloc failed. Signed-off-by:
YueHaibing <yuehaibing@huawei.com> Signed-off-by:
-
Alexei Starovoitov authored
Jesper Dangaard Brouer says: ==================== This patchset change ndo_xdp_xmit API to take a bulk of xdp frames. When kernel is compiled with CONFIG_RETPOLINE, every indirect function pointer (branch) call hurts performance. For XDP this have a huge negative performance impact. This patchset reduce the needed (indirect) calls to ndo_xdp_xmit, but also prepares for further optimizations. The DMA APIs use of indirect function pointer calls is the primary source the regression. It is left for a followup patchset, to use bulking calls towards the DMA API (via the scatter-gatter calls). The other advantage of this API change is that drivers can easier amortize the cost of any sync/locking scheme, over the bulk of packets. The assumption of the current API is that the driver implemementing the NDO will also allocate a dedicated XDP TX queue for every CPU in the system. Which is not always possible or practical to configure. E.g. ixgbe cannot load an XDP program on a machine with more than 96 CPUs, due to limited hardware TX queues. E.g. virtio_net is hard to configure as it requires manually increasing the queues. E.g. tun driver chooses to use a per XDP frame producer lock modulo smp_processor_id over avail queues. I'm considered adding 'flags' to ndo_xdp_xmit, but it's not part of this patchset. This will be a followup patchset, once we know if this will be needed (e.g. for non-map xdp_redirect flush-flag, and if AF_XDP chooses to use ndo_xdp_xmit for TX). --- V5: Fixed up issues spotted by Daniel and John V4: Splitout the patches from 4 to 8 patches. I cannot split the driver changes from the NDO change, but I've tried to isolated the NDO change together with the driver change as much as possible. ==================== Signed-off-by:Alexei Starovoitov <ast@kernel.org>
-
Jesper Dangaard Brouer authored
Update xdp_monitor to use the recently added err code introduced in tracepoint xdp:xdp_devmap_xmit, to show if the drop count is caused by some driver general delivery problem. Other kind of drops will likely just be more normal TX space issues. Signed-off-by:
Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by:
-
Jesper Dangaard Brouer authored
Extending tracepoint xdp:xdp_devmap_xmit in devmap with an err code allow people to easier identify the reason behind the ndo_xdp_xmit call to a given driver is failing. Signed-off-by:
-
Jesper Dangaard Brouer authored
This patch change the API for ndo_xdp_xmit to support bulking xdp_frames. When kernel is compiled with CONFIG_RETPOLINE, XDP sees a huge slowdown. Most of the slowdown is caused by DMA API indirect function calls, but also the net_device->ndo_xdp_xmit() call. Benchmarked patch with CONFIG_RETPOLINE, using xdp_redirect_map with single flow/core test (CPU E5-1650 v4 @ 3.60GHz), showed performance improved: for driver ixgbe: 6,042,682 pps -> 6,853,768 pps = +811,086 pps for driver i40e : 6,187,169 pps -> 6,724,519 pps = +537,350 pps With frames avail as a bulk inside the driver ndo_xdp_xmit call, further optimizations are possible, like bulk DMA-mapping for TX. Testing without CONFIG_RETPOLINE show the same performance for physical NIC drivers. The virtual NIC driver tun sees a huge performance boost, as it can avoid doing per frame producer locking, but instead amortize the locking cost over the bulk. V2: Fix compile errors reported by kbuild test robot <lkp@intel.com> V4: Isolated ndo, driver changes and callers. Signed-off-by:
-
Jesper Dangaard Brouer authored
When sending an xdp_frame through xdp_do_redirect call, then error cases can happen where the xdp_frame needs to be dropped, and returning an -errno code isn't sufficient/possible any-longer (e.g. for cpumap case). This is already fully supported, by simply calling xdp_return_frame. This patch is an optimization, which provides xdp_return_frame_rx_napi, which is a faster variant for these error cases. It take advantage of the protection provided by XDP RX running under NAPI protection. This change is mostly relevant for drivers using the page_pool allocator as it can take advantage of this. (Tested with mlx5). Signed-off-by:
-
