- 29 Sep, 2017 6 commits
-
-
Martin KaFai Lau authored
This patch swaps the checking order. It now checks the map_info first and then prog_info. It is a prep work for adding test to the newly added fields (the map_ids of prog_info field in particular). Signed-off-by:
Martin KaFai Lau <kafai@fb.com> Acked-by:
Alexei Starovoitov <ast@fb.com> Acked-by:
Daniel Borkmann <daniel@iogearbox.net> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Martin KaFai Lau authored
This patch extends the libbpf to provide API support to allow specifying BPF object name. In tools/lib/bpf/libbpf, the C symbol of the function and the map is used. Regarding section name, all maps are under the same section named "maps". Hence, section name is not a good choice for map's name. To be consistent with map, bpf_prog also follows and uses its function symbol as the prog's name. This patch adds logic to collect function's symbols in libbpf. There is existing codes to collect the map's symbols and no change is needed. The bpf_load_program_name() and bpf_map_create_name() are added to take the name argument. For the other bpf_map_create_xxx() variants, a name argument is directly added to them. In samples/bpf, bpf_load.c in particular, the symbol is also used as the map's name and the map symbols has already been collected in the existing code. For bpf_prog, bpf_load.c does not collect the function symbol name. We can consider to collect them later if there is a need to continue supporting the bpf_load.c. Signed-off-by:
-
Martin KaFai Lau authored
This patch allows userspace to specify a name for a map during BPF_MAP_CREATE. The map's name can later be exported to user space via BPF_OBJ_GET_INFO_BY_FD. Signed-off-by:
-
Martin KaFai Lau authored
The patch adds name and load_time to struct bpf_prog_aux. They are also exported to bpf_prog_info. The bpf_prog's name is passed by userspace during BPF_PROG_LOAD. The kernel only stores the first (BPF_PROG_NAME_LEN - 1) bytes and the name stored in the kernel is always \0 terminated. The kernel will reject name that contains characters other than isalnum() and '_'. It will also reject name that is not null terminated. The existing 'user->uid' of the bpf_prog_aux is also exported to the bpf_prog_info as created_by_uid. The existing 'used_maps' of the bpf_prog_aux is exported to the newly added members 'nr_map_ids' and 'map_ids' of the bpf_prog_info. On the input, nr_map_ids tells how big the userspace's map_ids buffer is. On the output, nr_map_ids tells the exact user_map_cnt and it will only copy up to the userspace's map_ids buffer is allowed. Signed-off-by:
-
Hoang Tran authored
With the commit 76174004 (tcp: do not slow start when cwnd equals ssthresh), the comparison to the reduced cwnd in tcp_vegas_ssthresh() would under-evaluate the ssthresh. Signed-off-by:
Hoang Tran <hoang.tran@uclouvain.be> Signed-off-by:
-
Nikolay Aleksandrov authored
We need to be able to transparently forward most link-local frames via tunnels (e.g. vxlan, qinq). Currently the bridge's group_fwd_mask has a mask which restricts the forwarding of STP and LACP, but we need to be able to forward these over tunnels and control that forwarding on a per-port basis thus add a new per-port group_fwd_mask option which only disallows mac pause frames to be forwarded (they're always dropped anyway). The patch does not change the current default situation - all of the others are still restricted unless configured for forwarding. We have successfully tested this patch with LACP and STP forwarding over VxLAN and qinq tunnels. Signed-off-by:
Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Signed-off-by:
-
- 28 Sep, 2017 28 commits
-
-
David S. Miller authored
Yunsheng Lin says: ==================== Add support for DCB feature in hns3 driver The patchset contains some enhancement related to DCB before adding support for DCB feature. This patchset depends on the following patchset: https://patchwork.ozlabs.org/cover/815646/ https://patchwork.ozlabs.org/cover/816145/ High Level Architecture: [ lldpad ] | | | [ hns3_dcbnl ] | | | [ hclge_dcb ] / \ / \ / \ [ hclge_main ] [ hclge_tm ] Current patch-set support following functionality: Use of lldptool to configure the tc schedule mode, tc bandwidth(if schedule mode is ETS), prio_tc_map and PFC parameter. V3: Drop mqprio support V2: Fix for not defining variables in local loop. V1: Initial Submit. ==================== Signed-off-by:
-
Yunsheng Lin authored
When using lldptool to configure DCB parameter, hclge_dcb module call the client_ops->setup_tc to tell network stack which queue and priority is using for specific tc. Signed-off-by:
Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by:
-
Yunsheng Lin authored
After the DCB feature is supported, fc_mode and dcb enable flag must be set according to the DCB parameter. Signed-off-by:
-
Yunsheng Lin authored
This patch add dcb netlink interface by calling the interface from hclge_dcb module. This patch also update Makefile in order to build hns3_dcbnl module. Signed-off-by:
-
Yunsheng Lin authored
The hclge_dcb module calls the interface from hclge_main/tm and provide interface for the dcb netlink interface. This patch also update Makefiles required to build the DCB supported code in HNS3 Ethernet driver and update the existing Kconfig file in the hisilicon folder. Signed-off-by:
-
Yunsheng Lin authored
This patch add some interface and export some interface from hclge_tm and hclgc_main to support the upcoming DCB feature. Signed-off-by:
-
Yunsheng Lin authored
When sriov is enabled and TM is in tc-based mode, vf's TM parameters is not set in TM initialization process. This patch add the tc_based TM support for sriov enabled using the information in vport struct. Signed-off-by:
-
Yunsheng Lin authored
This patch add a tm_port_shaper cmd and set port shaper to HCLGE_ETHER_MAX_RATE on TM initialization process. Signed-off-by:
-
Yunsheng Lin authored
This patch add a pfc_pause_en cmd, and use it to configure PFC option according to fc_mode in hdev->tm_info. Signed-off-by:
-
Yunsheng Lin authored
Current buffer allocation can only happen at init, when doing buffer reallocation after init, care must be taken care of memory which priv_buf points to. This patch fixes it by using a dynamic allocated temporary memory. Because we only do buffer reallocation at init or when setting up the DCB parameter, and priv_buf is only used at buffer allocation process, so it is ok to use a dynamic allocated temporary memory. Signed-off-by:
-
Yunsheng Lin authored
This patch add support of dynamically assigning tx buffer to TC when the TC is enabled. It will save buffer for rx direction to avoid packet loss. Signed-off-by:
-
Alexey Dobriyan authored
Negative ARP header length are not a thing. Constify arguments while I'm at it. Space savings: add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-3 (-3) function old new delta arpt_do_table 1163 1160 -3 Signed-off-by:
Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by:
-
Aviad Krawczyk authored
Fix the following scenario: 1. tx_free_poll is running on cpu X 2. xmit function is running on cpu Y and fails to get sq wqe 3. tx_free_poll frees wqes on cpu X and checks the queue is not stopped 4. xmit function stops the queue after failed to get sq wqe 5. The queue is stopped forever Signed-off-by:
Aviad Krawczyk <aviad.krawczyk@huawei.com> Signed-off-by:
-
Aviad Krawczyk authored
Set Rxq irq to specific cpu for allocating and receiving the skb from the same node. Signed-off-by:
-
David S. Miller authored
Edward Cree says: ==================== bpf/verifier: disassembly improvements Fix the output of print_bpf_insn() for ALU ops that don't look like compound assignment (i.e. BPF_END and BPF_NEG). Sample output for a short test program: 0: (b4) (u32) r0 = (u32) 0 1: (dc) r0 = be32 r0 2: (84) r0 = (u32) -r0 3: (95) exit processed 4 insns, stack depth 0 ==================== Signed-off-by: -
Edward Cree authored
BPF_NEG takes only one operand, unlike the bulk of BPF_ALU[64] which are compound-assignments. So give it its own format in print_bpf_insn(). Signed-off-by:
Edward Cree <ecree@solarflare.com> Acked-by:
Alexei Starovoitov <ast@kernel.org> Acked-by:
-
Edward Cree authored
print_bpf_insn() was treating all BPF_ALU[64] the same, but BPF_END has a different structure: it has a size in insn->imm (even if it's BPF_X) and uses the BPF_SRC (X or K) to indicate which endianness to use. So it needs different code to print it. Signed-off-by:
-
Colin Ian King authored
The function ch_flower_stats_cb is local to the source and does not need to be in global scope, so make it static. Cleans up sparse warnings: symbol 'ch_flower_stats_cb' was not declared. Should it be static? Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
David S. Miller authored
Florian Westphal says: ==================== rtnetlink: preparation patches for further rtnl lock pushdown/removal Patches split large rtnl_fill_ifinfo into smaller chunks to better see which parts 1. require rtnl 2. do not require it at all 3. rely on rtnl locking now but could be converted Changes since v3: I dropped the 'ifalias' patch, I have a change to decouple ifalias and rtnl mutex, I will send it once this series has been merged. ==================== Signed-off-by: -
Florian Westphal authored
it can be switched to rcu. Reviewed-by:
David Ahern <dsahern@gmail.com> Signed-off-by:
Florian Westphal <fw@strlen.de> Signed-off-by:
-
Florian Westphal authored
Reviewed-by:
-
Florian Westphal authored
similar to earlier patches, split out more parts of this function to better see what is happening and where we assume rtnl is locked. Reviewed-by:
-
Florian Westphal authored
rtnl_fill_ifinfo currently requires caller to hold the rtnl mutex. Unfortunately the function is quite large which makes it harder to see which spots require the lock, which spots assume it and which ones could do without. Add helpers to factor out the ifindex dumping, one can use rcu to avoid rtnl dependency. Reviewed-by:
-
Florian Westphal authored
Acked-by:
-
Cong Wang authored
Instead of calling u32_lookup_ht() in a loop to find a unused handle, just switch to idr API to allocate new handles. u32 filters are special as the handle could contain a hash table id and a key id, so we need two IDR to allocate each of them. Cc: Chris Mi <chrism@mellanox.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by:
Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by:
-
Cong Wang authored
Instead of calling basic_get() in a loop to find a unused handle, just switch to idr API to allocate new handles. Cc: Chris Mi <chrism@mellanox.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by:
-
Cong Wang authored
Instead of calling cls_bpf_get() in a loop to find a unused handle, just switch to idr API to allocate new handles. Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Chris Mi <chrism@mellanox.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by:
-
Eric Dumazet authored
As measured in my prior patch ("sch_netem: faster rb tree removal"), rbtree_postorder_for_each_entry_safe() is nice looking but much slower than using rb_next() directly, except when tree is small enough to fit in CPU caches (then the cost is the same) From: Eric Dumazet <edumazet@google.com> Signed-off-by:
-
- 27 Sep, 2017 6 commits
-
-
David S. Miller authored
Jiri Pirko says: ==================== mlxsw: Add support for offloading IPv4 multicast routes Yotam says: This patch-set introduces offloading of the kernel IPv4 multicast router logic in the Spectrum driver. The first patch makes the Spectrum driver ignore FIB notifications that are not of address family IPv4 or IPv6. This is needed in order to prevent crashes while the next patches introduce the RTNL_FAMILY_IPMR FIB notifications. Patches 2-5 update ipmr to use the FIB notification chain for both MFC and VIF notifications, and patches 8-12 update the Spectrum driver to register to these notifications and offload the routes. Similarly to IPv4 and IPv6, any failure will trigger the abort mechanism which is updated in this patch-set to eject multicast route tables too. At this stage, the following limitations apply: - A multicast MFC route will be offloaded by the driver if all the output interfaces are Spectrum router interfaces (RIFs). In any other case (which includes pimreg device, tunnel devices and management ports) the route will be trapped to the CPU and the packets will be forwarded by software. - ipmr proxy routes are not supported and will trigger the abort mechanism. - The MFC TTL values are currently treated as boolean: if the value is different than 255, the traffic is forwarded to the interface and if the value is 255 it is not forwarded. Dropping packets based on their TTL isn't currently supported. To allow users to have visibility on which of the routes are offloaded and which are not, patch 6 introduces a per-route offload indication similar to IPv4 and IPv6 routes which is sent to the user via the RTNetlink interface. The Spectrum driver multicast router offloading support, which is introduced in patches 8 and 9, is divided into two parts: - The hardware logic which abstracts the Spectrum hardware and provides a simple API for the upper levels. - The offloading logic which gets the MFC and VIF notifications from the kernel and updates the hardware using the hardware logic part. Finally, the last patch makes the Spectrum router logic not ignore the multicast FIB notifications and call the corresponding functions in the multicast router offloading logic. --- v2->v3: - Move the ipmr_rule_default function definition to be inside the already existing CONFIG_IP_MROUTE_MULTIPLE_TABLES ifdef block (patch 6) - Remove double =0 initialization in spectrum_mr.c (patch 7) - Fix route4 allocation size (patch 7) v1->v2: - Add comments for struct fields in mroute.h - Take the mrt_lock while dumping VIFs in the fib_notifier dump callback - Update the MFC lastuse field too ==================== Signed-off-by: -
Yotam Gigi authored
Make the Spectrum router logic not ignore the RTNL_FAMILY_IPMR FIB notifications. Past commits added the IPMR VIF and MFC add/del notifications via the fib_notifier chain. In addition, a code for handling these notifications in the Spectrum router logic was added. Make the Spectrum router logic not ignore these notifications and forward the requests to the Spectrum multicast router offloading logic. Signed-off-by:
Yotam Gigi <yotamg@mellanox.com> Reviewed-by:
Ido Schimmel <idosch@mellanox.com> Signed-off-by:
Jiri Pirko <jiri@mellanox.com> Signed-off-by:
-
Yotam Gigi authored
Due to the fact that multicast routes hold the minimum MTU of all the egress RIFs and trap packets that don't meet it, notify the mulitcast router code on RIF MTU changes. Signed-off-by:
-
Yotam Gigi authored
Add functionality for calling the multicast routing offloading logic upon MFC and VIF add and delete notifications. In addition, call the multicast routing upon RIF addition and deletion events. As the multicast routing offload logic may sleep, the actual calls are done in a deferred work. To ensure the MFC object is not freed in that interval, a reference is held to it. In case of a failure, the abort mechanism is used, which ejects all the routes from the hardware and triggers the traffic to flow through the kernel. Note: At that stage, the FIB notifications are still ignored, and will be enabled in a further patch. Signed-off-by:
-
Yotam Gigi authored
Currently, the mlxsw Spectrum driver offloads only either the RT_TABLE_MAIN FIB table or the VRF tables, so the RT_TABLE_LOCAL table is squashed to the RT_TABLE_MAIN table to allow local routes to be offloaded too. By default, multicast MFC routes which are not assigned to any user requested table are put in the RT_TABLE_DEFAULT table. Due to the fact that offloading multicast MFC routes support in Spectrum router logic is going to be introduced soon, squash the default table to MAIN too. Signed-off-by:
-
Yotam Gigi authored
Implement the multicast routing hardware API introduced in previous patch for the specific spectrum hardware. The spectrum hardware multicast routes are written using the RMFT2 register and point to an ACL flexible action set. The actions used for multicast routes are: - Counter action, which allows counting bytes and packets on multicast routes. - Multicast route action, which provide RPF check and do the actual packet duplication to a list of RIFs. - Trap action, in the case the route action specified by the called is trap. Signed-off-by:
-
