- 03 May, 2020 14 commits
-
-
Sameeh Jubran authored
1. Add support for getting tx drops from the device and saving them in the driver. 2. Report tx via netdev stats. Signed-off-by:
Igor Chauskin <igorch@amazon.com> Signed-off-by:
Guy Tzalik <gtzalik@amazon.com> Signed-off-by:
Arthur Kiyanovski <akiyano@amazon.com> Signed-off-by:
Sameeh Jubran <sameehj@amazon.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Sameeh Jubran authored
Add unmask interrupts statistics to ethtool. Signed-off-by:
Netanel Belgazal <netanel@amazon.com> Signed-off-by:
-
Sameeh Jubran authored
Both key and func parameters are pointers on the stack. Setting them to NULL does nothing. The original intent was to leave the key and func unset in this case, but for this to happen nothing needs to be done as the calling function ethtool_get_rxfh() already clears key and func. This commit removes the above described useless code. Signed-off-by:
-
Sameeh Jubran authored
This commit contains 2 cosmetic changes: 1. Use ena_com_check_supported_feature_id() in ena_com_hash_key_fill_default_key() instead of rewriting its implementation. This also saves us a superfluous admin command by using the cached value. 2. Change if conditions in ena_com_rss_init() to be clearer. Signed-off-by:
-
Arthur Kiyanovski authored
Currently in the driver we are setting the hash function to be CRC32. Starting with this commit we want to change the default behaviour so that we set the hash function to be Toeplitz instead. Signed-off-by:
-
Sameeh Jubran authored
Current code does not allow setting the hash function without changing the key. This commit enables it. To achieve this we separate ena_com_get_hash_function() to 2 functions: ena_com_get_hash_function() - which gets only the hash function, and ena_com_get_hash_key() - which gets only the hash key. Also return 0 instead of rc at the end of ena_get_rxfh() since all previous operations succeeded. Signed-off-by:
-
Arthur Kiyanovski authored
In case the "func" parameter is NULL we now return "-EINVAL". This shouldn't happen in general, but when it does happen, this is the proper way to handle it. We also check func for NULL in the beginning of the function, as there is no reason to do all the work and realize in the end of the function it was useless. Signed-off-by:
-
Arthur Kiyanovski authored
Currently when ena_set_hash_function() fails the hash function is restored to the previous value by calling an admin command to get the hash function from the device. In this commit we avoid the admin command, by saving the previous hash function before calling ena_set_hash_function() and using this previous value to restore the hash function in case of failure of ena_set_hash_function(). Signed-off-by:
-
David S. Miller authored
Eric Dumazet says: ==================== net_sched: sch_fq: round of optimizations This series is focused on better layout of struct fq_flow to reduce number of cache line misses in fq_enqueue() and dequeue operations. ==================== Signed-off-by: -
Eric Dumazet authored
The prefetch() done in fq_dequeue() can be done a bit earlier after the refactoring of the code done in the prior patch. Signed-off-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
-
Eric Dumazet authored
This refactors the code to not call fq_peek() from fq_dequeue_head() since the caller can provide the skb. Also rename fq_dequeue_head() to fq_dequeue_skb() because 'head' is a bit vague, given the skb could come from t_root rb-tree. Signed-off-by:
-
Eric Dumazet authored
fq_gc() already builds a small array of pointers, so using kmem_cache_free_bulk() needs very little change. Signed-off-by:
-
Eric Dumazet authored
sizeof(struct fq_flow) is 112 bytes on 64bit arches. This means that half of them use two cache lines, but 50% use three cache lines. This patch adds cache line alignment, and makes sure that only the first cache line is touched by fq_enqueue(), which is more expensive that fq_dequeue() in general. Signed-off-by:
-
Eric Dumazet authored
A significant amount of cpu cycles is spent in fq_gc() When fq_gc() does its lookup in the rb-tree, it needs the following fields from struct fq_flow : f->sk (lookup key in the rb-tree) f->fq_node (anchor in the rb-tree) f->next (used to determine if the flow is detached) f->age (used to determine if the flow is candidate for gc) This unfortunately spans two cache lines (assuming 64 bytes cache lines) We can avoid using f->next, if we use the low order bit of f->{age|tail} This low order bit is 0, if f->tail points to an sk_buff. We set the low order bit to 1, if the union contains a jiffies value. Combined with the following patch, this makes sure we only need to bring into cpu caches one cache line per flow. Signed-off-by:
-
- 02 May, 2020 9 commits
-
-
Dmitry Yakunin authored
Fix bug introduced by commit b1f3e43d ("inet_diag: add support for cgroup filter"). Signed-off-by:
Dmitry Yakunin <zeil@yandex-team.ru> Reported-by: syzbot+ee80f840d9bf6893223b@syzkaller.appspotmail.com Reported-by: syzbot+13bef047dbfffa5cd1af@syzkaller.appspotmail.com Fixes: b1f3e43d ("inet_diag: add support for cgroup filter") Signed-off-by:
-
Andrew Lunn authored
Measurements of the MDIO bus have shown that driving the MDIO bus using interrupts is slow. Back to back MDIO transactions take about 90us, with 25us spent performing the transaction, and the remainder of the time the bus is idle. Replacing the completion interrupt with polled IO results in back to back transactions of 40us. The polling loop waiting for the hardware to complete the transaction takes around 28us. Which suggests interrupt handling has an overhead of 50us, and polled IO nearly halves this overhead, and doubles the MDIO performance. Care has to be taken when setting the MII_SPEED register, or it can trigger an MII event> That then upsets the polling, due to an unexpected pending event. Suggested-by:
Chris Heally <cphealy@gmail.com> Signed-off-by:
Andrew Lunn <andrew@lunn.ch> Signed-off-by:
-
David S. Miller authored
net/smc/smc_llc.c:544:12: warning: ‘smc_llc_alloc_alt_link’ defined but not used [-Wunused-function] static int smc_llc_alloc_alt_link(struct smc_link_group *lgr, ^~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: -
David S. Miller authored
Vincent Cheng says: ==================== ptp: Add adjust phase to support phase offset. This series adds adjust phase to the PTP Hardware Clock device interface. Some PTP hardware clocks have a write phase mode that has a built-in hardware filtering capability. The write phase mode utilizes a phase offset control word instead of a frequency offset control word. Add adjust phase function to take advantage of this capability. Changes since v1: - As suggested by Richard Cochran: 1. ops->adjphase is new so need to check for non-null function pointer. 2. Kernel coding style uses lower_case_underscores. 3. Use existing PTP clock API for delayed worker. ==================== Signed-off-by: -
Vincent Cheng authored
Add idtcm_adjphase() to support PHC write phase mode. Signed-off-by:
Vincent Cheng <vincent.cheng.xh@renesas.com> Acked-by:
Richard Cochran <richardcochran@gmail.com> Signed-off-by:
-
Vincent Cheng authored
Add adjust_phase to ptp_clock_caps capability to allow user to query if a PHC driver supports adjust phase with ioctl PTP_CLOCK_GETCAPS command. Signed-off-by:
-
Vincent Cheng authored
Adds adjust phase function to take advantage of a PHC clock's hardware filtering capability that uses phase offset control word instead of frequency offset control word. 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 2020-05-01 (v2) The following pull-request contains BPF updates for your *net-next* tree. We've added 61 non-merge commits during the last 6 day(s) which contain a total of 153 files changed, 6739 insertions(+), 3367 deletions(-). The main changes are: 1) pulled work.sysctl from vfs tree with sysctl bpf changes. 2) bpf_link observability, from Andrii. 3) BTF-defined map in map, from Andrii. 4) asan fixes for selftests, from Andrii. 5) Allow bpf_map_lookup_elem for SOCKMAP and SOCKHASH, from Jakub. 6) production cloudflare classifier as a selftes, from Lorenz. 7) bpf_ktime_get_*_ns() helper improvements, from Maciej. 8) unprivileged bpftool feature probe, from Quentin. 9) BPF_ENABLE_STATS command, from Song. 10) enable bpf_[gs]etsockopt() helpers for sock_ops progs, from Stanislav. 11) enable a bunch of common helpers for cg-device, sysctl, sockopt progs, from Stanislav. ==================== Signed-off-by: -
David S. Miller authored
Karsten Graul says: ==================== net/smc: extent buffer mapping and port handling Add functionality to map/unmap and register/unregister memory buffers for specific SMC-R links and for the whole link group. Prepare LLC layer messages for the support of multiple links and extent the processing of adapter events. And add further small preparations needed for the SMC-R failover support. ==================== Signed-off-by:
-
- 01 May, 2020 17 commits
-
-
Stanislav Fomichev authored
Andrey pointed out that we can use reno instead of dctcp for CC tests and drop CONFIG_TCP_CONG_DCTCP=y requirement. Fixes: beecf11b ("bpf: Bpf_{g,s}etsockopt for struct bpf_sock_addr") Suggested-by:
Andrey Ignatov <rdna@fb.com> Signed-off-by:
Stanislav Fomichev <sdf@google.com> Signed-off-by:
Alexei Starovoitov <ast@kernel.org> Acked-by:
Martin KaFai Lau <kafai@fb.com> Link: https://lore.kernel.org/bpf/20200501224320.28441-1-sdf@google.com
-
Karsten Graul authored
Introduce a work that is scheduled when a new ADD_LINK LLC request is received. The work will call either the SMC client or SMC server ADD_LINK processing. Signed-off-by:
Karsten Graul <kgraul@linux.ibm.com> Reviewed-by:
Ursula Braun <ubraun@linux.ibm.com> Signed-off-by:
-
Karsten Graul authored
Add smc_llc_alloc_alt_link() to find a free link index for a new link, depending on the new link group type. And update constants for the maximum number of links to 3 (2 symmetric and 1 dangling asymmetric link). These maximum numbers are the same as used by other implementations of the SMC-R protocol. Signed-off-by:
-
Karsten Graul authored
Introduce a new function in smc_pnet.c that searches for an alternate IB device, using an existing link group and a primary IB device. The alternate IB device needs to be active and must have the same PNETID as the link group. Signed-off-by:
-
Karsten Graul authored
Support for multiple links makes the former DELETE LINK processing obsolete which sent one DELETE_LINK LLC message for each single link. Remove this processing from smc_core.c. Signed-off-by:
-
Karsten Graul authored
Use the introduced link down processing in all places where the link group is terminated and take down the affected link only. Signed-off-by:
-
Karsten Graul authored
Call smcr_port_err() when an IB event reports an inactive IB device. smcr_port_err() calls smcr_link_down() for all affected links. smcr_link_down() either triggers the local DELETE_LINK processing, or sends an DELETE_LINK LLC message to the SMC server to initiate the processing. The old handler function smc_port_terminate() is removed. Add helper smcr_link_down_cond() to take a link down conditionally, and smcr_link_down_cond_sched() to schedule the link_down processing to a work. Signed-off-by:
-
Karsten Graul authored
Call smcr_port_add() when an IB event reports a new active IB device. smcr_port_add() will start a work which either triggers the local ADD_LINK processing, or send an ADD_LINK LLC message to the SMC server to initiate the processing. Signed-off-by:
-
Karsten Graul authored
The PNETID is needed to find an alternate link for a link group. Save the PNETID of the link that is used to create the link group for later device matching. Signed-off-by:
-
Karsten Graul authored
Introduce llc_conf_mutex in the link group which is used to protect the buffers and lgr states against parallel link reconfiguration. This ensures that new connections do not start to register buffers with the links of a link group when link creation or termination is running. Signed-off-by:
-
Karsten Graul authored
All LLC sends are done from worker context only, so remove the prep functions which were used to build the message before it was sent, and add the function content into the respective send function smc_llc_send_add_link() and smc_llc_send_delete_link(). Extend smc_llc_send_add_link() to include the qp_mtu value in the LLC message, which is needed to establish a link after the initial link was created. Extend smc_llc_send_delete_link() to contain a link_id and a reason code for the link deletion in the LLC message, which is needed when a specific link should be deleted. And add the list of existing DELETE_LINK reason codes. Signed-off-by:
-
Karsten Graul authored
Introduce support to map and register all current buffers for a new link. smcr_buf_map_lgr() will map used buffers for a new link and smcr_buf_reg_lgr() can be called to register used buffers on the IB device of the new link. Signed-off-by:
-
Karsten Graul authored
With the support of multiple links that are created and cleared there is a need to unmap one link from all current buffers. Add unmapping by link and by rmb. And make smcr_link_clear() available to be called from the LLC layer. Signed-off-by:
-
Karsten Graul authored
The CONFIRM_RKEY LLC processing handles all links in one LLC message. Move the call to this processing out of smcr_link_reg_rmb() which does processing per link, into smcr_lgr_reg_rmbs() which is responsible for link group level processing. Move smcr_link_reg_rmb() into module smc_core.c. >From af_smc.c now call smcr_lgr_reg_rmbs() to register new rmbs on all available links. Signed-off-by:
-
David S. Miller authored
Po Liu says: ==================== Introduce a flow gate control action and apply IEEE Changes from V4: ---------------- 0001: Fix and modify according to Vlid Buslov suggestions: - Change spin_lock_bh() to spin_lock() since tcf_gate_act() already in software irq. - Remove spin lock protect in the ops->cleanup function. - Enable the CONFIG_DEBUG_ATOMIC_SLEEP and CONFIG_PROVE_LOCKING checking, then fix as kzalloc flag type and lock deadlock. - Change kzalloc() flag type from GFP_KERNEL to GFP_ATOMIC since function in the spin_lock protect. - Change hrtimer type from HRTIMER_MODE_ABS to HRTIMER_MODE_ABS_SOFT to avoid deadlock. 0002: Fix and modify according to Vlid Buslov suggestions: - Remove all rcu_read_lock protection since no rcu parameters. - Enable the CONFIG_DEBUG_ATOMIC_SLEEP and CONFIG_PROVE_LOCKING checking, then check kzalloc sleeping flag. - Change kzalloc to kcalloc for array memory alloc and change GFP_KERNEL flag to GFP_ATOMIC since function holding spin_lock protect. 0003: - No changes. 0004: - Commit comments rephrase act by Claudiu Manoil. Changes from V3: ---------------- 0001: Fix and modify according to Vlid Buslov: - Remove the struct gate_action and move the parameters to the struct tcf_gate align with tc_action parameters. This would not need to alloc rcu type memory with pointer. - Remove the spin_lock type entry_lock which do not needed anymore, will use the tcf_lock system provided. - Provide lockep protect for the status parameters in the tcf_gate_act(). - Remove the cycletime 0 input warning, return error directly. And: - Remove Qci related description in the Kconfig for gate action. 0002: - Fix rcu_read_lock protect range suggested by Vlid Buslov. 0003: - No changes. 0004: - Fix bug of gate maxoct wildcard condition not included. - Fix the pass time basetime calculation report by Vladimir Otlean. Changes from V2: 0001: No changes. 0002: No changes. 0003: No changes. 0004: Fix the vlan id filter parameter and add reject src mac FF-FF-FF-FF-FF-FF filter in driver. Changes from V1: ---------------- 0000: Update description make it more clear 0001: Removed 'add update dropped stats' patch, will provide pull request as standalone patches. 0001: Update commit description make it more clear ack by Jiri Pirko. 0002: No changes 0003: Fix some code style ack by Jiri Pirko. 0004: Fix enetc_psfp_enable/disable parameter type ack by test robot iprout2 command patches: Not attach with these serial patches, will provide separate pull request after kernel accept these patches. Changes from RFC: ----------------- 0000: Reduce to 5 patches and remove the 4 max frame size offload and flow metering in the policing offload action, Only keep gate action offloading implementation. 0001: No changes. 0002: - fix kfree lead ack by Jakub Kicinski and Cong Wang - License fix from Jakub Kicinski and Stephen Hemminger - Update example in commit acked by Vinicius Costa Gomes - Fix the rcu protect in tcf_gate_act() acked by Vinicius 0003: No changes 0004: No changes 0005: Acked by Vinicius Costa Gomes - Use refcount kernel lib - Update stream gate check code position - Update reduce ref names more clear iprout2 command patches: 0000: Update license expression and add gate id 0001: Add tc action gate man page -------------------------------------------------------------------- These patches add stream gate action policing in IEEE802.1Qci (Per-Stream Filtering and Policing) software support and hardware offload support in tc flower, and implement the stream identify, stream filtering and stream gate filtering action in the NXP ENETC ethernet driver. Per-Stream Filtering and Policing (PSFP) specifies flow policing and filtering for ingress flows, and has three main parts: 1. The stream filter instance table consists of an ordered list of stream filters that determine the filtering and policing actions that are to be applied to frames received on a specific stream. The main elements are stream gate id, flow metering id and maximum SDU size. 2. The stream gate function setup a gate list to control ingress traffic class open/close state. When the gate is running at open state, the flow could pass but dropped when gate state is running to close. User setup a bastime to tell gate when start running the entry list, then the hardware would periodiclly. There is no compare qdisc action support. 3. Flow metering is two rates two buckets and three-color marker to policing the frames. Flow metering instance are as specified in the algorithm in MEF10.3. The most likely qdisc action is policing action. The first patch introduces an ingress frame flow control gate action, for the point 2. The tc gate action maintains the open/close state gate list, allowing flows to pass when the gate is open. Each gate action may policing one or more qdisc filters. When the start time arrived, The driver would repeat the gate list periodiclly. User can assign a passed time, the driver would calculate a new future time by the cycletime of the gate list. The 0002 patch introduces the gate flow hardware offloading. The 0003 patch adds support control the on/off for the tc flower offloading by ethtool. The 0004 patch implement the stream identify and stream filtering and stream gate filtering action in the NXP ENETC ethernet driver. Tc filter command provide filtering keys with MAC address and VLAN id. These keys would be set to stream identify instance entry. Stream gate instance entry would refer the gate action parameters. Stream filter instance entry would refer the stream gate index and assign a stream handle value matches to the stream identify instance. ==================== Signed-off-by: -
Po Liu authored
This patch is to add tc flower offload for the enetc IEEE 802.1Qci(PSFP) function. There are four main feature parts to implement the flow policing and filtering for ingress flow with IEEE 802.1Qci features. They are stream identify(this is defined in the P802.1cb exactly but needed for 802.1Qci), stream filtering, stream gate and flow metering. Each function block includes many entries by index to assign parameters. So for one frame would be filtered by stream identify first, then flow into stream filter block by the same handle between stream identify and stream filtering. Then flow into stream gate control which assigned by the stream filtering entry. And then policing by the gate and limited by the max sdu in the filter block(optional). At last, policing by the flow metering block, index choosing at the fitering block. So you can see that each entry of block may link to many upper entries since they can be assigned same index means more streams want to share the same feature in the stream filtering or stream gate or flow metering. To implement such features, each stream filtered by source/destination mac address, some stream maybe also plus the vlan id value would be treated as one flow chain. This would be identified by the chain_index which already in the tc filter concept. Driver would maintain this chain and also with gate modules. The stream filter entry create by the gate index and flow meter(optional) entry id and also one priority value. Offloading only transfer the gate action and flow filtering parameters. Driver would create (or search same gate id and flow meter id and priority) one stream filter entry to set to the hardware. So stream filtering do not need transfer by the action offloading. This architecture is same with tc filter and actions relationship. tc filter maintain the list for each flow feature by keys. And actions maintain by the action list. Below showing a example commands by tc: > tc qdisc add dev eth0 ingress > ip link set eth0 address 10:00:80:00:00:00 > tc filter add dev eth0 parent ffff: protocol ip chain 11 \ flower skip_sw dst_mac 10:00:80:00:00:00 \ action gate index 10 \ sched-entry open 200000000 1 8000000 \ sched-entry close 100000000 -1 -1 Command means to set the dst_mac 10:00:80:00:00:00 to index 11 of stream identify module. Then setting the gate index 10 of stream gate module. Keep the gate open for 200ms and limit the traffic volume to 8MB in this sched-entry. Then direct the frames to the ingress queue 1. Signed-off-by:
Po Liu <Po.Liu@nxp.com> Signed-off-by:
-
Po Liu authored
This patch is to let ethtool enable/disable the tc flower offload features. Hardware ENETC has the feature of PSFP which is for per-stream policing. When enable the tc hw offloading feature, driver would enable the IEEE 802.1Qci feature. It is only set the register enable bit for this feature not enable for any entry of per stream filtering and stream gate or stream identify but get how much capabilities for each feature. Signed-off-by:
-
