- 29 Mar, 2024 40 commits
-
-
Kuniyuki Iwashima authored
The definition of the lowlink in Tarjan's algorithm is the smallest index of a vertex that is reachable with at most one back-edge in SCC. This is not useful for a cross-edge. If we start traversing from A in the following graph, the final lowlink of D is 3. The cross-edge here is one between D and C. A -> B -> D D = (4, 3) (index, lowlink) ^ | | C = (3, 1) | V | B = (2, 1) `--- C <--' A = (1, 1) This is because the lowlink of D is updated with the index of C. In the following patch, we detect a dead SCC by checking two conditions for each vertex. 1) vertex has no edge directed to another SCC (no bridge) 2) vertex's out_degree is the same as the refcount of its file If 1) is false, there is a receiver of all fds of the SCC and its ancestor SCC. To evaluate 1), we need to assign a unique index to each SCC and assign it to all vertices in the SCC. This patch changes the lowlink update logic for cross-edge so that in the example above, the lowlink of D is updated with the lowlink of C. A -> B -> D D = (4, 1) (index, lowlink) ^ | | C = (3, 1) | V | B = (2, 1) `--- C <--' A = (1, 1) Then, all vertices in the same SCC have the same lowlink, and we can quickly find the bridge connecting to different SCC if exists. However, it is no longer called lowlink, so we rename it to scc_index. (It's sometimes called lowpoint.) Also, we add a global variable to hold the last index used in DFS so that we do not reset the initial index in each DFS. This patch can be squashed to the SCC detection patch but is split deliberately for anyone wondering why lowlink is not used as used in the original Tarjan's algorithm and many reference implementations. Signed-off-by:
Kuniyuki Iwashima <kuniyu@amazon.com> Acked-by:
Paolo Abeni <pabeni@redhat.com> Link: https://lore.kernel.org/r/20240325202425.60930-13-kuniyu@amazon.comSigned-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Kuniyuki Iwashima authored
Once a cyclic reference is formed, we need to run GC to check if there is dead SCC. However, we do not need to run Tarjan's algorithm if we know that the shape of the inflight graph has not been changed. If an edge is added/updated/deleted and the edge's successor is inflight, we set false to unix_graph_grouped, which means we need to re-classify SCC. Once we finalise SCC, we set true to unix_graph_grouped. While unix_graph_grouped is true, we can iterate the grouped SCC using vertex->scc_entry in unix_walk_scc_fast(). list_add() and list_for_each_entry_reverse() uses seem weird, but they are to keep the vertex order consistent and make writing test easier. Signed-off-by:
-
Kuniyuki Iwashima authored
We do not need to run GC if there is no possible cyclic reference. We use unix_graph_maybe_cyclic to decide if we should run GC. If a fd of an AF_UNIX socket is passed to an already inflight AF_UNIX socket, they could form a cyclic reference. Then, we set true to unix_graph_maybe_cyclic and later run Tarjan's algorithm to group them into SCC. Once we run Tarjan's algorithm, we are 100% sure whether cyclic references exist or not. If there is no cycle, we set false to unix_graph_maybe_cyclic and can skip the entire garbage collection next time. When finalising SCC, we set true to unix_graph_maybe_cyclic if SCC consists of multiple vertices. Even if SCC is a single vertex, a cycle might exist as self-fd passing. Given the corner case is rare, we detect it by checking all edges of the vertex and set true to unix_graph_maybe_cyclic. With this change, __unix_gc() is just a spin_lock() dance in the normal usage. Signed-off-by:
-
Kuniyuki Iwashima authored
Before starting Tarjan's algorithm, we need to mark all vertices as unvisited. We can save this O(n) setup by reserving two special indices (0, 1) and using two variables. The first time we link a vertex to unix_unvisited_vertices, we set unix_vertex_unvisited_index to index. During DFS, we can see that the index of unvisited vertices is the same as unix_vertex_unvisited_index. When we finalise SCC later, we set unix_vertex_grouped_index to each vertex's index. Then, we can know (i) that the vertex is on the stack if the index of a visited vertex is >= 2 and (ii) that it is not on the stack and belongs to a different SCC if the index is unix_vertex_grouped_index. After the whole algorithm, all indices of vertices are set as unix_vertex_grouped_index. Next time we start DFS, we know that all unvisited vertices have unix_vertex_grouped_index, and we can use unix_vertex_unvisited_index as the not-on-stack marker. To use the same variable in __unix_walk_scc(), we can swap unix_vertex_(grouped|unvisited)_index at the end of Tarjan's algorithm. Signed-off-by:
-
Kuniyuki Iwashima authored
To garbage collect inflight AF_UNIX sockets, we must define the cyclic reference appropriately. This is a bit tricky if the loop consists of embryo sockets. Suppose that the fd of AF_UNIX socket A is passed to D and the fd B to C and that C and D are embryo sockets of A and B, respectively. It may appear that there are two separate graphs, A (-> D) and B (-> C), but this is not correct. A --. .-- B X C <-' `-> D Now, D holds A's refcount, and C has B's refcount, so unix_release() will never be called for A and B when we close() them. However, no one can call close() for D and C to free skbs holding refcounts of A and B because C/D is in A/B's receive queue, which should have been purged by unix_release() for A and B. So, here's another type of cyclic reference. When a fd of an AF_UNIX socket is passed to an embryo socket, the reference is indirectly held by its parent listening socket. .-> A .-> B | `- sk_receive_queue | `- sk_receive_queue | `- skb | `- skb | `- sk == C | `- sk == D | `- sk_receive_queue | `- sk_receive_queue | `- skb +---------' `- skb +-. | | `---------------------------------------------------------' Technically, the graph must be denoted as A <-> B instead of A (-> D) and B (-> C) to find such a cyclic reference without touching each socket's receive queue. .-> A --. .-- B <-. | X | == A <-> B `-- C <-' `-> D --' We apply this fixup during GC by fetching the real successor by unix_edge_successor(). When we call accept(), we clear unix_sock.listener under unix_gc_lock not to confuse GC. Signed-off-by: -
Kuniyuki Iwashima authored
This is a prep patch for the following change, where we need to fetch the listening socket from the successor embryo socket during GC. We add a new field to struct unix_sock to save a pointer to a listening socket. We set it when connect() creates a new socket, and clear it when accept() is called. Signed-off-by:
-
Kuniyuki Iwashima authored
In the new GC, we use a simple graph algorithm, Tarjan's Strongly Connected Components (SCC) algorithm, to find cyclic references. The algorithm visits every vertex exactly once using depth-first search (DFS). DFS starts by pushing an input vertex to a stack and assigning it a unique number. Two fields, index and lowlink, are initialised with the number, but lowlink could be updated later during DFS. If a vertex has an edge to an unvisited inflight vertex, we visit it and do the same processing. So, we will have vertices in the stack in the order they appear and number them consecutively in the same order. If a vertex has a back-edge to a visited vertex in the stack, we update the predecessor's lowlink with the successor's index. After iterating edges from the vertex, we check if its index equals its lowlink. If the lowlink is different from the index, it shows there was a back-edge. Then, we go backtracking and propagate the lowlink to its predecessor and resume the previous edge iteration from the next edge. If the lowlink is the same as the index, we pop vertices before and including the vertex from the stack. Then, the set of vertices is SCC, possibly forming a cycle. At the same time, we move the vertices to unix_visited_vertices. When we finish the algorithm, all vertices in each SCC will be linked via unix_vertex.scc_entry. Let's take an example. We have a graph including five inflight vertices (F is not inflight): A -> B -> C -> D -> E (-> F) ^ | `---------' Suppose that we start DFS from C. We will visit C, D, and B first and initialise their index and lowlink. Then, the stack looks like this: > B = (3, 3) (index, lowlink) D = (2, 2) C = (1, 1) When checking B's edge to C, we update B's lowlink with C's index and propagate it to D. B = (3, 1) (index, lowlink) > D = (2, 1) C = (1, 1) Next, we visit E, which has no edge to an inflight vertex. > E = (4, 4) (index, lowlink) B = (3, 1) D = (2, 1) C = (1, 1) When we leave from E, its index and lowlink are the same, so we pop E from the stack as single-vertex SCC. Next, we leave from B and D but do nothing because their lowlink are different from their index. B = (3, 1) (index, lowlink) D = (2, 1) > C = (1, 1) Then, we leave from C, whose index and lowlink are the same, so we pop B, D and C as SCC. Last, we do DFS for the rest of vertices, A, which is also a single-vertex SCC. Finally, each unix_vertex.scc_entry is linked as follows: A -. B -> C -> D E -. ^ | ^ | ^ | `--' `---------' `--' We use SCC later to decide whether we can garbage-collect the sockets. Note that we still cannot detect SCC properly if an edge points to an embryo socket. The following two patches will sort it out. Signed-off-by: -
Kuniyuki Iwashima authored
The new GC will use a depth first search graph algorithm to find cyclic references. The algorithm visits every vertex exactly once. Here, we implement the DFS part without recursion so that no one can abuse it. unix_walk_scc() marks every vertex unvisited by initialising index as UNIX_VERTEX_INDEX_UNVISITED and iterates inflight vertices in unix_unvisited_vertices and call __unix_walk_scc() to start DFS from an arbitrary vertex. __unix_walk_scc() iterates all edges starting from the vertex and explores the neighbour vertices with DFS using edge_stack. After visiting all neighbours, __unix_walk_scc() moves the visited vertex to unix_visited_vertices so that unix_walk_scc() will not restart DFS from the visited vertex. Signed-off-by:
-
Kuniyuki Iwashima authored
Currently, we track the number of inflight sockets in two variables. unix_tot_inflight is the total number of inflight AF_UNIX sockets on the host, and user->unix_inflight is the number of inflight fds per user. We update them one by one in unix_inflight(), which can be done once in batch. Also, sendmsg() could fail even after unix_inflight(), then we need to acquire unix_gc_lock only to decrement the counters. Let's bulk update the counters in unix_add_edges() and unix_del_edges(), which is called only for successfully passed fds. Signed-off-by:
-
Kuniyuki Iwashima authored
Just before queuing skb with inflight fds, we call scm_stat_add(), which is a good place to set up the preallocated struct unix_vertex and struct unix_edge in UNIXCB(skb).fp. Then, we call unix_add_edges() and construct the directed graph as follows: 1. Set the inflight socket's unix_sock to unix_edge.predecessor. 2. Set the receiver's unix_sock to unix_edge.successor. 3. Set the preallocated vertex to inflight socket's unix_sock.vertex. 4. Link inflight socket's unix_vertex.entry to unix_unvisited_vertices. 5. Link unix_edge.vertex_entry to the inflight socket's unix_vertex.edges. Let's say we pass the fd of AF_UNIX socket A to B and the fd of B to C. The graph looks like this: +-------------------------+ | unix_unvisited_vertices | <-------------------------. +-------------------------+ | + | | +--------------+ +--------------+ | +--------------+ | | unix_sock A | <---. .---> | unix_sock B | <-|-. .---> | unix_sock C | | +--------------+ | | +--------------+ | | | +--------------+ | .-+ | vertex | | | .-+ | vertex | | | | | vertex | | | +--------------+ | | | +--------------+ | | | +--------------+ | | | | | | | | | | +--------------+ | | | +--------------+ | | | | '-> | unix_vertex | | | '-> | unix_vertex | | | | | +--------------+ | | +--------------+ | | | `---> | entry | +---------> | entry | +-' | | |--------------| | | |--------------| | | | edges | <-. | | | edges | <-. | | +--------------+ | | | +--------------+ | | | | | | | | | .----------------------' | | .----------------------' | | | | | | | | | +--------------+ | | | +--------------+ | | | | unix_edge | | | | | unix_edge | | | | +--------------+ | | | +--------------+ | | `-> | vertex_entry | | | `-> | vertex_entry | | | |--------------| | | |--------------| | | | predecessor | +---' | | predecessor | +---' | |--------------| | |--------------| | | successor | +-----' | successor | +-----' +--------------+ +--------------+ Henceforth, we denote such a graph as A -> B (-> C). Now, we can express all inflight fd graphs that do not contain embryo sockets. We will support the particular case later. Signed-off-by: -
Kuniyuki Iwashima authored
As with the previous patch, we preallocate to skb's scm_fp_list an array of struct unix_edge in the number of inflight AF_UNIX fds. There we just preallocate memory and do not use immediately because sendmsg() could fail after this point. The actual use will be in the next patch. When we queue skb with inflight edges, we will set the inflight socket's unix_sock as unix_edge->predecessor and the receiver's unix_sock as successor, and then we will link the edge to the inflight socket's unix_vertex.edges. Note that we set NULL to cloned scm_fp_list.edges in scm_fp_dup() so that MSG_PEEK does not change the shape of the directed graph. Signed-off-by:
-
Kuniyuki Iwashima authored
We will replace the garbage collection algorithm for AF_UNIX, where we will consider each inflight AF_UNIX socket as a vertex and its file descriptor as an edge in a directed graph. This patch introduces a new struct unix_vertex representing a vertex in the graph and adds its pointer to struct unix_sock. When we send a fd using the SCM_RIGHTS message, we allocate struct scm_fp_list to struct scm_cookie in scm_fp_copy(). Then, we bump each refcount of the inflight fds' struct file and save them in scm_fp_list.fp. After that, unix_attach_fds() inexplicably clones scm_fp_list of scm_cookie and sets it to skb. (We will remove this part after replacing GC.) Here, we add a new function call in unix_attach_fds() to preallocate struct unix_vertex per inflight AF_UNIX fd and link each vertex to skb's scm_fp_list.vertices. When sendmsg() succeeds later, if the socket of the inflight fd is still not inflight yet, we will set the preallocated vertex to struct unix_sock.vertex and link it to a global list unix_unvisited_vertices under spin_lock(&unix_gc_lock). If the socket is already inflight, we free the preallocated vertex. This is to avoid taking the lock unnecessarily when sendmsg() could fail later. In the following patch, we will similarly allocate another struct per edge, which will finally be linked to the inflight socket's unix_vertex.edges. And then, we will count the number of edges as unix_vertex.out_degree. Signed-off-by:
-
Eric Dumazet authored
TCP ehash table is often sparsely populated. inet_twsk_purge() spends too much time calling cond_resched(). This patch can reduce time spent in inet_twsk_purge() by 20x. Signed-off-by:
Eric Dumazet <edumazet@google.com> Reviewed-by:
-
Kurt Kanzenbach authored
As of commit 916444df ("ptp: deprecate gettime64() in favor of gettimex64()") (new) PTP drivers should rather implement gettimex64(). In addition, this variant provides timestamps from the system clock. The readings have to be recorded right before and after reading the lowest bits of the PHC timestamp. Reported-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
Kurt Kanzenbach <kurt@linutronix.de> Acked-by:
Richard Cochran <richardcochran@gmail.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Zhengchao Shao authored
smc_hash_sk and smc_unhash_sk are only used in af_smc.c, so make them static and remove the output symbol. They can be called under the path .prot->hash()/unhash(). Signed-off-by:
Zhengchao Shao <shaozhengchao@huawei.com> Reviewed-by:
Wen Gu <guwen@linux.alibaba.com> Reviewed-by:
Simon Horman <horms@kernel.org> Signed-off-by:
-
David S. Miller authored
Asbjørn Sloth Tønnesen says: ==================== make skip_sw actually skip software During development of flower-route[1], which I recently presented at FOSDEM[2], I noticed that CPU usage, would increase the more rules I installed into the hardware for IP forwarding offloading. Since we use TC flower offload for the hottest prefixes, and leave the long tail to the normal (non-TC) Linux network stack for slow-path IP forwarding. We therefore need both the hardware and software datapath to perform well. I found that skip_sw rules, are quite expensive in the kernel datapath, since they must be evaluated and matched upon, before the kernel checks the skip_sw flag. This patchset optimizes the case where all rules are skip_sw, by implementing a TC bypass for these cases, where TC is only used as a control plane for the hardware path. v4: - Rebased onto net-next, now that net-next is open again v3: https://lore.kernel.org/netdev/20240306165813.656931-1-ast@fiberby.net/ - Patch 3: - Fix source_inline - Fix build failure, when CONFIG_NET_CLS without CONFIG_NET_CLS_ACT. v2: https://lore.kernel.org/netdev/20240305144404.569632-1-ast@fiberby.net/ - Patch 1: - Add Reviewed-By from Jiri Pirko - Patch 2: - Move code, to avoid forward declaration (Jiri). - Patch 3 - Refactor to use a static key. - Add performance data for trapping, or sending a packet to a non-existent chain (as suggested by Marcelo). v1: https://lore.kernel.org/netdev/20240215160458.1727237-1-ast@fiberby.net/ [1] flower-route https://github.com/fiberby-dk/flower-route [2] FOSDEM talk https://fosdem.org/2024/schedule/event/fosdem-2024-3337-flying-higher-hardware-offloading-with-bird/ ==================== Signed-off-by:
-
Asbjørn Sloth Tønnesen authored
TC filters come in 3 variants: - no flag (try to process in hardware, but fallback to software)) - skip_hw (do not process filter by hardware) - skip_sw (do not process filter by software) However skip_sw is implemented so that the skip_sw flag can first be checked, after it has been matched. IMHO it's common when using skip_sw, to use it on all rules. So if all filters in a block is skip_sw filters, then we can bail early, we can thus avoid having to match the filters, just to check for the skip_sw flag. This patch adds a bypass, for when only TC skip_sw rules are used. The bypass is guarded by a static key, to avoid harming other workloads. There are 3 ways that a packet from a skip_sw ruleset, can end up in the kernel path. Although the send packets to a non-existent chain way is only improved a few percents, then I believe it's worth optimizing the trap and fall-though use-cases. +----------------------------+--------+--------+--------+ | Test description | Pre- | Post- | Rel. | | | kpps | kpps | chg. | +----------------------------+--------+--------+--------+ | basic forwarding + notrack | 3589.3 | 3587.9 | 1.00x | | switch to eswitch mode | 3081.8 | 3094.7 | 1.00x | | add ingress qdisc | 3042.9 | 3063.6 | 1.01x | | tc forward in hw / skip_sw |37024.7 |37028.4 | 1.00x | | tc forward in sw / skip_hw | 3245.0 | 3245.3 | 1.00x | +----------------------------+--------+--------+--------+ | tests with only skip_sw rules below: | +----------------------------+--------+--------+--------+ | 1 non-matching rule | 2694.7 | 3058.7 | 1.14x | | 1 n-m rule, match trap | 2611.2 | 3323.1 | 1.27x | | 1 n-m rule, goto non-chain | 2886.8 | 2945.9 | 1.02x | | 5 non-matching rules | 1958.2 | 3061.3 | 1.56x | | 5 n-m rules, match trap | 1911.9 | 3327.0 | 1.74x | | 5 n-m rules, goto non-chain| 2883.1 | 2947.5 | 1.02x | | 10 non-matching rules | 1466.3 | 3062.8 | 2.09x | | 10 n-m rules, match trap | 1444.3 | 3317.9 | 2.30x | | 10 n-m rules,goto non-chain| 2883.1 | 2939.5 | 1.02x | | 25 non-matching rules | 838.5 | 3058.9 | 3.65x | | 25 n-m rules, match trap | 824.5 | 3323.0 | 4.03x | | 25 n-m rules,goto non-chain| 2875.8 | 2944.7 | 1.02x | | 50 non-matching rules | 488.1 | 3054.7 | 6.26x | | 50 n-m rules, match trap | 484.9 | 3318.5 | 6.84x | | 50 n-m rules,goto non-chain| 2884.1 | 2939.7 | 1.02x | +----------------------------+--------+--------+--------+ perf top (25 n-m skip_sw rules - pre patch): 20.39% [kernel] [k] __skb_flow_dissect 16.43% [kernel] [k] rhashtable_jhash2 10.58% [kernel] [k] fl_classify 10.23% [kernel] [k] fl_mask_lookup 4.79% [kernel] [k] memset_orig 2.58% [kernel] [k] tcf_classify 1.47% [kernel] [k] __x86_indirect_thunk_rax 1.42% [kernel] [k] __dev_queue_xmit 1.36% [kernel] [k] nft_do_chain 1.21% [kernel] [k] __rcu_read_lock perf top (25 n-m skip_sw rules - post patch): 5.12% [kernel] [k] __dev_queue_xmit 4.77% [kernel] [k] nft_do_chain 3.65% [kernel] [k] dev_gro_receive 3.41% [kernel] [k] check_preemption_disabled 3.14% [kernel] [k] mlx5e_skb_from_cqe_mpwrq_nonlinear 2.88% [kernel] [k] __netif_receive_skb_core.constprop.0 2.49% [kernel] [k] mlx5e_xmit 2.15% [kernel] [k] ip_forward 1.95% [kernel] [k] mlx5e_tc_restore_tunnel 1.92% [kernel] [k] vlan_gro_receive Test setup: DUT: Intel Xeon D-1518 (2.20GHz) w/ Nvidia/Mellanox ConnectX-6 Dx 2x100G Data rate measured on switch (Extreme X690), and DUT connected as a router on a stick, with pktgen and pktsink as VLANs. Pktgen-dpdk was in range 36.6-37.7 Mpps 64B packets across all tests. Full test data at https://files.fiberby.net/ast/2024/tc_skip_sw/v2_tests/Signed-off-by:
Asbjørn Sloth Tønnesen <ast@fiberby.net> Reviewed-by:
Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by:
-
Asbjørn Sloth Tønnesen authored
Maintain a count of filters per block. Counter updates are protected by cb_lock, which is also used to protect the offload counters. Signed-off-by:
-
Asbjørn Sloth Tønnesen authored
Maintain a count of skip_sw filters. This counter is protected by the cb_lock, and is updated at the same time as offloadcnt. Signed-off-by:
Jiri Pirko <jiri@nvidia.com> Reviewed-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queueJakub Kicinski authored
Tony Nguyen says: ==================== ice: use less resources in switchdev Michal Swiatkowski says: Switchdev is using one queue per created port representor. This can quickly lead to Rx queue shortage, as with subfunction support user can create high number of PRs. Save one MSI-X and 'number of PRs' * 1 queues. Refactor switchdev slow-path to use less resources (even no additional resources). Do this by removing control plane VSI and move its functionality to PF VSI. Even with current solution PF is acting like uplink and can't be used simultaneously for other use cases (adding filters can break slow-path). In short, do Tx via PF VSI and Rx packets using PF resources. Rx needs additional code in interrupt handler to choose correct PR netdev. Previous solution had to queue filters, it was way more elegant but needed one queue per PRs. Beside that this refactor mostly simplifies switchdev configuration. * '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue: ice: count representor stats ice: do switchdev slow-path Rx using PF VSI ice: change repr::id values ice: remove switchdev control plane VSI ice: control default Tx rule in lag ice: default Tx rule instead of to queue ice: do Tx through PF netdev in slow-path ice: remove eswitch changing queues algorithm ==================== Link: https://lore.kernel.org/r/20240325202623.1012287-1-anthony.l.nguyen@intel.comSigned-off-by:
-
Jakub Kicinski authored
Michael Chan says: ==================== bnxt_en: PTP and RSS updates The first 2 patches are v2 of the PTP patches posted about 3 weeks ago: https://lore.kernel.org/netdev/20240229070202.107488-1-michael.chan@broadcom.com/ The devlink parameter is dropped and v2 is just to increase the timeout accuracy and to use a default timeout of 1 second. Patches 3 to 12 implement additional RSS contexts and ntuple filters for the RSS contexts. ==================== Link: https://lore.kernel.org/r/20240325222902.220712-1-michael.chan@broadcom.comSigned-off-by:
-
Pavan Chebbi authored
When the user wants to add an ntuple filter to an RSS context, select the appropriate VNIC belonging to the selected RSS context and add the VNIC destination rule. Make the necessary changes to bnxt_add_ntuple_cls_rule(). Reviewed-by:
Kalesh AP <kalesh-anakkur.purayil@broadcom.com> Signed-off-by:
Pavan Chebbi <pavan.chebbi@broadcom.com> Signed-off-by:
Michael Chan <michael.chan@broadcom.com> Link: https://lore.kernel.org/r/20240325222902.220712-13-michael.chan@broadcom.comSigned-off-by:
-
Pavan Chebbi authored
Refactor bnxt_cfg_rfs_ring_tbl_idx() to pass in the filter structure pointer instead of the RX ring number. This will allow an ntuple filter to be set up for the non-default RSS contexts in the next patch. Reviewed-by:
-
Pavan Chebbi authored
Support up to 32 RSS contexts per device if supported by the device. Reviewed-by:
-
Michael Chan authored
Add a new bnxt_modify_rss() function to modify the RSS key and RSS indirection table. The new function can modify the parameters for the default context or additional contexts. Signed-off-by:
-
Pavan Chebbi authored
Modify bnxt_rfs_capable() to check that there are enough resources to support aRFS/ntuple filters for a new RSS context requested by the user. Existing use cases in the driver will always set the new parameter to false. Reviewed-by:
-
Michael Chan authored
bnxt_rfs_capable() determines the number of VNICs and RSS_CTXs required to support aRFS and then reserves the resources. We already have functions bnxt_get_total_vnics() and bnxt_get_total_rss_ctxs() to do that. Simplify the code by calling these functions. It is also more correct to do the resource reservation after bnxt_can_reserve_rings() returns true. Signed-off-by:
-
Pavan Chebbi authored
We will need to dynamically allocate and change indirection tables for additional RSS contexts. Add the rss_ctx pointer parameter to bnxt_alloc_rss_indir_tbl() and bnxt_set_dflt_rss_indir_tbl(). Existing usage will always pass rss_ctx as NULL which means the default RSS context. When supporting additional RSS contexts in subsequent patches, we'll pass the valid rss_ctx to these 2 functions. Reviewed-by:
Andy Gospodarek <andrew.gospodarek@broadcom.com> Signed-off-by:
-
Pavan Chebbi authored
Add struct bnxt_rss_ctx, related storage lists, required defines, and its alloc/free functions. Later patches will use them in order to support multiple RSS contexts. Reviewed-by:
-
Pavan Chebbi authored
The current VNIC structures are stored in an array bp->vnic_info[]. The index of the array (vnic_id) is passed to all the functions that need to reference the VNIC. This patch changes the scheme to pass the VNIC pointer instead of the vnic index. Subsequent patches will create additional VNICs that will not be stored in the bp->vnic_info[] array. Using the VNIC pointer will work for all the VNICs. Reviewed-by:
-
Pavan Chebbi authored
This is a pure refactoring patch. The new function bnxt_hwrm_vnic_set_rss_p5() will set up the P5_PLUS specific RSS ring table and then call bnxt_hwrm_vnic_cfg() to setup the vnic for proper RSS operations. This new function will be used later for additional RSS contexts. Signed-off-by:
-
Pavan Chebbi authored
Use a new default 1 second timeout value instead of the existing 1 msec value. The driver will keep track of the remaining time before timeout and will pass this value to bnxt_hwrm_port_ts_query(). The firmware supports timeout values up to 65535 usecs. If the timeout value passed to bnxt_hwrm_port_ts_query() is less than the FW max value, we will use that value to precisely control the specified timeout. If it is larger than the FW max value, we will use the FW max value and any additional retry to reach the desired timeout will be done in the context of bnxt_ptp_ts_aux_eork(). Link: https://lore.kernel.org/netdev/20240229070202.107488-2-michael.chan@broadcom.com/ Cc: Richard Cochran <richardcochran@gmail.com> Reviewed-by:
Vadim Fedorenko <vadim.fedorenko@linux.dev> Link: https://lore.kernel.org/r/20240325222902.220712-3-michael.chan@broadcom.comSigned-off-by:
-
Michael Chan authored
The caller can pass this new timeout parameter to the function to specify the firmware timeout value when requesting the TX timestamp from the firmware. This will allow the caller to precisely control the timeout and will be used in the next patch. In this patch, the parameter is 0 which means to use the current default value. Cc: Richard Cochran <richardcochran@gmail.com> Reviewed-by:
-
Jakub Kicinski authored
Romain Gantois says: ==================== Fix missing PHY-to-MAC RX clock There is an issue with some stmmac/PHY combinations that has been reported some time ago in a couple of different series: Clark Wang's report: https://lore.kernel.org/all/20230202081559.3553637-1-xiaoning.wang@nxp.com/ Clément Léger's report: https://lore.kernel.org/linux-arm-kernel/20230116103926.276869-4-clement.leger@bootlin.com/ Stmmac controllers require an RX clock signal from the MII bus to perform their hardware initialization successfully. This causes issues with some PHY/PCS devices. If these devices do not bring the clock signal up before the MAC driver initializes its hardware, then said initialization will fail. This can happen at probe time or when the system wakes up from a suspended state. This series introduces new flags for phy_device and phylink_pcs. These flags allow MAC drivers to signal to PHY/PCS drivers that the RX clock signal should be enabled as soon as possible, and that it should always stay enabled. I have included specific uses of these flags that fix the RZN1 GMAC1 stmmac driver that I am currently working on and that is not yet upstream. I have also included changes to the at803x PHY driver that should fix the issue that Clark Wang was having. ==================== Link: https://lore.kernel.org/r/20240326-rxc_bugfix-v6-0-24a74e5c761f@bootlin.comSigned-off-by:
-
Romain Gantois authored
The GMAC1 controller in the RZN1 IP requires the RX MII clock signal to be started before it initializes its own hardware, thus before it calls phylink_start. Implement the pcs_pre_init() callback so that the RX clock signal can be enabled early if necessary. Reported-by:
Clément Léger <clement.leger@bootlin.com> Link: https://lore.kernel.org/linux-arm-kernel/20230116103926.276869-4-clement.leger@bootlin.com/Signed-off-by:
Romain Gantois <romain.gantois@bootlin.com> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Link: https://lore.kernel.org/r/20240326-rxc_bugfix-v6-7-24a74e5c761f@bootlin.comSigned-off-by:
-
Russell King (Oracle) authored
Stmmac controllers connected to an at803x PHY cannot resume properly after suspend when WoL is enabled. This happens because the MAC requires an RX clock generated by the PHY to initialize its hardware properly. But the RX clock is cut when the PHY suspends and isn't brought up until the MAC driver resumes the phylink. Prevent the at803x PHY driver from going into suspend if the attached MAC driver always requires an RX clock signal. Reported-by:
Clark Wang <xiaoning.wang@nxp.com> Link: https://lore.kernel.org/all/20230202081559.3553637-1-xiaoning.wang@nxp.com/Signed-off-by:
Russell King (Oracle) <rmk+kernel@armlinux.org.uk> [rgantois: commit log] Signed-off-by:
-
Romain Gantois authored
There is a reocurring issue with stmmac controllers where the MAC fails to initialize its hardware if an RX clock signal isn't provided on the MAC/PHY link. This causes issues when PHY or PCS devices either go into suspend while cutting the RX clock or do not bring the clock signal up early enough for the MAC to initialize successfully. Set the mac_requires_rxc flag in the stmmac phylink config so that PHY/PCS drivers know to keep the RX clock up at all times. Reported-by:
-
Romain Gantois authored
Global stmmac support for early initialization of PCS devices requires a generic PCS reference that can be passed to phylink_pcs_pre_init(). Currently, the mac_device_info struct contains only one PCS field, which is specific to the Lynx model. As PCS models are hardware-specific, it is more appropriate to have a generic PCS field in the mac_device_info struct. Refactor the lynx_pcs field into a generic phylink_pcs field. Signed-off-by:
-
Maxime Chevallier authored
When initializing attached PHYs, there are some cases where we don't expect any PHY to be connected. The logic uses conditions based on various local PCS configuration, but also calls-in phylink_expects_phy() via stmmac_init_phy(), which is enough to ensure we don't try to initialize a PHY when using a Lynx PCS, as long as we have the phy_interface set to a 802.3z mode and are using inband negociation. Drop the lynx check, making the stmmac generic code more pcs_lynx-agnostic. Signed-off-by:
Maxime Chevallier <maxime.chevallier@bootlin.com> [rgantois: commit log] Signed-off-by:
-
Romain Gantois authored
Some MAC drivers (e.g. stmmac) require a continuous receive clock signal to be generated by a PCS that is handled by a standalone PCS driver. Such a PCS driver does not have access to a PHY device, thus cannot check the PHY_F_RXC_ALWAYS_ON flag. They cannot check max_requires_rxc in the phylink config either, since it is a private member. Therefore, a new flag is needed to signal to the PCS that it should keep the RX clock signal up at all times. Co-developed-by:
-
