- 22 Oct, 2015 12 commits
-
-
David S. Miller authored
Alexei Starovoitov says: ==================== bpf_perf_event_output helper Over the last year there were multiple attempts to let eBPF programs output data into perf events by He Kuang and Wangnan. The last one was: https://lkml.org/lkml/2015/7/20/736 It was almost perfect with exception that all bpf programs would sent data into one global perf_event. This patch set takes different approach by letting user space open independent PERF_COUNT_SW_BPF_OUTPUT events, so that program output won't collide. Wangnan is working on corresponding perf patches. ==================== Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Alexei Starovoitov authored
Performance test and example of bpf_perf_event_output(). kprobe is attached to sys_write() and trivial bpf program streams pid+cookie into userspace via PERF_COUNT_SW_BPF_OUTPUT event. Usage: $ sudo ./bld_x64/samples/bpf/trace_output recv 2968913 events per sec Signed-off-by:
Alexei Starovoitov <ast@kernel.org> Signed-off-by:
-
Alexei Starovoitov authored
This helper is used to send raw data from eBPF program into special PERF_TYPE_SOFTWARE/PERF_COUNT_SW_BPF_OUTPUT perf_event. User space needs to perf_event_open() it (either for one or all cpus) and store FD into perf_event_array (similar to bpf_perf_event_read() helper) before eBPF program can send data into it. Today the programs triggered by kprobe collect the data and either store it into the maps or print it via bpf_trace_printk() where latter is the debug facility and not suitable to stream the data. This new helper replaces such bpf_trace_printk() usage and allows programs to have dedicated channel into user space for post-processing of the raw data collected. Signed-off-by:
-
Alexei Starovoitov authored
Instead of WARN_ON in perf_event_output() on unpaded raw samples, pad them automatically. Signed-off-by:
-
Brenden Blanco authored
In the ipv4 outbound path of an ipvlan device in l3 mode, the ifindex is being grabbed from dev_get_iflink. This works for the physical device case, since as the documentation of that function notes: "Physical interfaces have the same 'ifindex' and 'iflink' values.". However, if the master device is a veth, and the pairs are in separate net namespaces, the route lookup will fail with -ENODEV due to outer veth pair being in a separate namespace from the ipvlan master/routing namespace. ns0 | ns1 | ns2 veth0a--|--veth0b--|--ipvl0 In ipvlan_process_v4_outbound(), a packet sent from ipvl0 in the above configuration will pass fl.flowi4_oif == veth0a to ip_route_output_flow(), but *net == ns1. Notice also that ipv6 processing is not using iflink. Since there is a discrepancy in usage, fixup both v4 and v6 case to use local dev variable. Tested this with l3 ipvlan on top of veth, as well as with single physical interface in the top namespace. Signed-off-by:
Brenden Blanco <bblanco@plumgrid.com> Reviewed-by:
Jiri Benc <jbenc@redhat.com> Acked-by:
Mahesh Bandewar <maheshb@google.com> Signed-off-by:
-
Eric Dumazet authored
Allowing an application to set whatever limit for the list of recently RST fastopen sessions [1] is not wise, as it open ways to deplete kernel memory. Cap the user provided limit by somaxconn sysctl, like listen() backlog. [1] https://tools.ietf.org/html/rfc7413#section-5.1Signed-off-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
-
Vivien Didelot authored
This header file only contains the platform data structure definition, so move it to the include/linux/platform_data/ directory. Signed-off-by:
Vivien Didelot <vivien.didelot@savoirfairelinux.com> Signed-off-by:
-
Vivien Didelot authored
Remove the inclusion of linux/mdio-gpio.h in nas4220b, wbd111 and wbd222 boards since mdio-gpio is not used. Signed-off-by:
-
Wu Fengguang authored
drivers/net/ethernet/hisilicon/hns/hnae.c:442:1-3: WARNING: PTR_ERR_OR_ZERO can be used Use PTR_ERR_OR_ZERO rather than if(IS_ERR(...)) + PTR_ERR Generated by: scripts/coccinelle/api/ptr_ret.cocci CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by:
Fengguang Wu <fengguang.wu@intel.com> Signed-off-by:
-
Eric Dumazet authored
Tom Herbert added SIT support to GRO with commit 19424e05 ("sit: Add gro callbacks to sit_offload"), later reverted by Herbert Xu. The problem came because Tom patch was building GRO packets without proper meta data : If packets were locally delivered, we would not care. But if packets needed to be forwarded, GSO engine was not able to segment individual segments. With the following patch, we correctly set skb->encapsulation and inner network header. We also update gso_type. Tested: Server : netserver modprobe dummy ifconfig dummy0 8.0.0.1 netmask 255.255.255.0 up arp -s 8.0.0.100 4e:32:51:04:47:e5 iptables -I INPUT -s 10.246.7.151 -j TEE --gateway 8.0.0.100 ifconfig sixtofour0 sixtofour0 Link encap:IPv6-in-IPv4 inet6 addr: 2002:af6:798::1/128 Scope:Global inet6 addr: 2002:af6:798::/128 Scope:Global UP RUNNING NOARP MTU:1480 Metric:1 RX packets:411169 errors:0 dropped:0 overruns:0 frame:0 TX packets:409414 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:20319631739 (20.3 GB) TX bytes:29529556 (29.5 MB) Client : netperf -H 2002:af6:798::1 -l 1000 & Checked on server traffic copied on dummy0 and verify segments were properly rebuilt, with proper IP headers, TCP checksums... tcpdump on eth0 shows proper GRO aggregation takes place. Signed-off-by:
Tom Herbert <tom@herbertland.com> Signed-off-by:
-
Eric Dumazet authored
While testing my SIT/GRO patch using netfilter TEE module and a dummy device, I found some features were missing : TSO IPv6, UFO, and encapsulated traffic. ethtool -k dummy0 now gives : ... tcp-segmentation-offload: on tx-tcp-segmentation: on tx-tcp-ecn-segmentation: on tx-tcp6-segmentation: on udp-fragmentation-offload: on ... tx-gre-segmentation: on tx-ipip-segmentation: on tx-sit-segmentation: on tx-udp_tnl-segmentation: on Signed-off-by:
-
Arad, Ronen authored
if_nlmsg_size() overestimates the minimum allocation size of netlink dump request (when called from rtnl_calcit()) or the size of the message (when called from rtnl_getlink()). This is because ext_filter_mask is not supported by rtnl_link_get_af_size() and rtnl_link_get_size(). The over-estimation is significant when at least one netdev has many VLANs configured (8 bytes for each configured VLAN). This patch-set "rightsizes" the protocol specific attribute size calculation by propagating ext_filter_mask to rtnl_link_get_af_size() and adding this a argument to get_link_af_size op in rtnl_af_ops. Bridge module already used filtering aware sizing for notifications. br_get_link_af_size_filtered() is consistent with the modified get_link_af_size op so it replaces br_get_link_af_size() in br_af_ops. br_get_link_af_size() becomes unused and thus removed. Signed-off-by:
Ronen Arad <ronen.arad@intel.com> Acked-by:
Sridhar Samudrala <sridhar.samudrala@intel.com> Signed-off-by:
-
- 21 Oct, 2015 10 commits
-
-
Elad Raz authored
Configure ageing time to the HW for newly bridged device CC: Scott Feldman <sfeldma@gmail.com> CC: Jiri Pirko <jiri@resnulli.us> Signed-off-by:
Elad Raz <eladr@mellanox.com> Acked-by:
Jiri Pirko <jiri@mellanox.com> Acked-by:
Scott Feldman <sfeldma@gmail.com> Signed-off-by:
-
David S. Miller authored
Yuchung Cheng says: ==================== RACK loss detection RACK (Recent ACK) loss recovery uses the notion of time instead of packet sequence (FACK) or counts (dupthresh). It's inspired by the FACK heuristic in tcp_mark_lost_retrans(): when a limited transmit (new data packet) is sacked in recovery, then any retransmission sent before that newly sacked packet was sent must have been lost, since at least one round trip time has elapsed. But that existing heuristic from tcp_mark_lost_retrans() has several limitations: 1) it can't detect tail drops since it depends on limited transmit 2) it's disabled upon reordering (assumes no reordering) 3) it's only enabled in fast recovery but not timeout recovery RACK addresses these limitations with a core idea: an unacknowledged packet P1 is deemed lost if a packet P2 that was sent later is is s/acked, since at least one round trip has passed. Since RACK cares about the time sequence instead of the data sequence of packets, it can detect tail drops when a later retransmission is s/acked, while FACK or dupthresh can't. For reordering RACK uses a dynamically adjusted reordering window ("reo_wnd") to reduce false positives on ever (small) degree of reordering, similar to the delayed Early Retransmit. In the current patch set RACK is only a supplemental loss detection and does not trigger fast recovery. However we are developing RACK to replace or consolidate FACK/dupthresh, early retransmit, and thin-dupack. These heuristics all implicitly bear the time notion. For example, the delayed Early Retransmit is simply applying RACK to trigger the fast recovery with small inflight. RACK requires measuring the minimum RTT. Tracking a global min is less robust due to traffic engineering pathing changes. Therefore it uses a windowed filter by Kathleen Nichols. The min RTT can also be useful for various other purposes like congestion control or stat monitoring. This patch has been used on Google servers for well over 1 year. RACK has also been implemented in the QUIC protocol. We are submitting an IETF draft as well. ==================== Signed-off-by: -
Yuchung Cheng authored
This patch implements the second half of RACK that uses the the most recent transmit time among all delivered packets to detect losses. tcp_rack_mark_lost() is called upon receiving a dubious ACK. It then checks if an not-yet-sacked packet was sent at least "reo_wnd" prior to the sent time of the most recently delivered. If so the packet is deemed lost. The "reo_wnd" reordering window starts with 1msec for fast loss detection and changes to min-RTT/4 when reordering is observed. We found 1msec accommodates well on tiny degree of reordering (<3 pkts) on faster links. We use min-RTT instead of SRTT because reordering is more of a path property but SRTT can be inflated by self-inflicated congestion. The factor of 4 is borrowed from the delayed early retransmit and seems to work reasonably well. Since RACK is still experimental, it is now used as a supplemental loss detection on top of existing algorithms. It is only effective after the fast recovery starts or after the timeout occurs. The fast recovery is still triggered by FACK and/or dupack threshold instead of RACK. We introduce a new sysctl net.ipv4.tcp_recovery for future experiments of loss recoveries. For now RACK can be disabled by setting it to 0. Signed-off-by:
Yuchung Cheng <ycheng@google.com> Signed-off-by:
Neal Cardwell <ncardwell@google.com> Signed-off-by:
-
Yuchung Cheng authored
This patch is the first half of the RACK loss recovery. RACK loss recovery uses the notion of time instead of packet sequence (FACK) or counts (dupthresh). It's inspired by the previous FACK heuristic in tcp_mark_lost_retrans(): when a limited transmit (new data packet) is sacked, then current retransmitted sequence below the newly sacked sequence must been lost, since at least one round trip time has elapsed. But it has several limitations: 1) can't detect tail drops since it depends on limited transmit 2) is disabled upon reordering (assumes no reordering) 3) only enabled in fast recovery ut not timeout recovery RACK (Recently ACK) addresses these limitations with the notion of time instead: a packet P1 is lost if a later packet P2 is s/acked, as at least one round trip has passed. Since RACK cares about the time sequence instead of the data sequence of packets, it can detect tail drops when later retransmission is s/acked while FACK or dupthresh can't. For reordering RACK uses a dynamically adjusted reordering window ("reo_wnd") to reduce false positives on ever (small) degree of reordering. This patch implements tcp_advanced_rack() which tracks the most recent transmission time among the packets that have been delivered (ACKed or SACKed) in tp->rack.mstamp. This timestamp is the key to determine which packet has been lost. Consider an example that the sender sends six packets: T1: P1 (lost) T2: P2 T3: P3 T4: P4 T100: sack of P2. rack.mstamp = T2 T101: retransmit P1 T102: sack of P2,P3,P4. rack.mstamp = T4 T205: ACK of P4 since the hole is repaired. rack.mstamp = T101 We need to be careful about spurious retransmission because it may falsely advance tp->rack.mstamp by an RTT or an RTO, causing RACK to falsely mark all packets lost, just like a spurious timeout. We identify spurious retransmission by the ACK's TS echo value. If TS option is not applicable but the retransmission is acknowledged less than min-RTT ago, it is likely to be spurious. We refrain from using the transmission time of these spurious retransmissions. The second half is implemented in the next patch that marks packet lost using RACK timestamp. Signed-off-by: -
Yuchung Cheng authored
a helper to prepare the first main RACK patch. Signed-off-by:
-
Yuchung Cheng authored
a helper to prepare the main RACK patch Signed-off-by:
-
Yuchung Cheng authored
Remove the existing lost retransmit detection because RACK subsumes it completely. This also stops the overloading the ack_seq field of the skb control block. Signed-off-by:
-
Yuchung Cheng authored
Kathleen Nichols' algorithm for tracking the minimum RTT of a data stream over some measurement window. It uses constant space and constant time per update. Yet it almost always delivers the same minimum as an implementation that has to keep all the data in the window. The measurement window is tunable via sysctl.net.ipv4.tcp_min_rtt_wlen with a default value of 5 minutes. The algorithm keeps track of the best, 2nd best & 3rd best min values, maintaining an invariant that the measurement time of the n'th best >= n-1'th best. It also makes sure that the three values are widely separated in the time window since that bounds the worse case error when that data is monotonically increasing over the window. Upon getting a new min, we can forget everything earlier because it has no value - the new min is less than everything else in the window by definition and it's the most recent. So we restart fresh on every new min and overwrites the 2nd & 3rd choices. The same property holds for the 2nd & 3rd best. Therefore we have to maintain two invariants to maximize the information in the samples, one on values (1st.v <= 2nd.v <= 3rd.v) and the other on times (now-win <=1st.t <= 2nd.t <= 3rd.t <= now). These invariants determine the structure of the code The RTT input to the windowed filter is the minimum RTT measured from ACK or SACK, or as the last resort from TCP timestamps. The accessor tcp_min_rtt() returns the minimum RTT seen in the window. ~0U indicates it is not available. The minimum is 1usec even if the true RTT is below that. Signed-off-by:
-
Yuchung Cheng authored
Currently ca_seq_rtt_us does not use Kern's check. Fix that by checking if any packet acked is a retransmit, for both RTT used for RTT estimation and congestion control. Fixes: 5b08e47c ("tcp: prefer packet timing to TS-ECR for RTT") Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queueDavid S. Miller authored
Jeff Kirsher says: ==================== Intel Wired LAN Driver Updates 2015-10-19 This series contains updates to i40e and i40evf only. Kiran adds a spinlock around code accessing VSI MAC filter list to ensure that we are synchronizing access to the filter list, otherwise we can end up with multiple accesses at the same time which can cause the VSI MAC filter list to get in an unstable or corrupted state. Jesse fixes overlong BIT defines, where the RSS enabling call were mistakenly missed. Also fixes a bug where the enable function was enabling the interrupt twice while trying to update the two interrupt throttle rate thresholds for Rx and Tx, while refactoring the IRQ enable function to simplify reading the flow. Addressed the high CPU utilization of some small streaming workloads that the driver should reduce CPU in. Anjali fixes two X722 issues with respect to EEPROM checksum verify and reading NVM version info. Fixed where a mask value was accidentally replaced with a bit mask causing Flow Director sideband to be broken. Alex Duyck fixes areas of the drivers which run from hard interrupt context or with interrupts already disabled in netpoll, so use napi_schedule_irqoff() instead of napi_schedule(). Mitch fixes the VF drivers to not easily give up when it is not able to communicate with the PF driver. Carolyn fixes a problem where our tools MAC loopback test, after driver unbind would fail because the hardware was configured for multiqueue and unbind operation did not clear this configuration. Also fixed a issue where the NVMUpdate tool gets bad data from the PHY when using the PHY NVM feature because of contention on the MDIO interface from getting PHY capability calls from the driver during regular operations. Catherine fixed an issue where we were checking if autoneg was allowed to change before checking if autoneg was changing, these checks need to be in the reverse order. Jean Sacren fixes up an function header comment to align the kernel-docs with the actual code. v2: Cleaned up the use of spin_is_locked() in patch 1 based on feedback from David Miller, since it always evaluates to zero on uni-processor builds ==================== Signed-off-by:
-
- 20 Oct, 2015 1 commit
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/davem/netDavid S. Miller authored
Conflicts: drivers/net/usb/asix_common.c net/ipv4/inet_connection_sock.c net/switchdev/switchdev.c In the inet_connection_sock.c case the request socket hashing scheme is completely different in net-next. The other two conflicts were overlapping changes. Signed-off-by:
-
- 19 Oct, 2015 17 commits
-
-
Catherine Sullivan authored
Bump. Change-ID: Id0a7ecaa491f88ce94c9eba4901e592a56044ee0 Signed-off-by:
Catherine Sullivan <catherine.sullivan@intel.com> Tested-by:
Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by:
Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Jean Sacren authored
'err' would be overwritten immediately, so we should declare it only rather than initialize it to zero. Signed-off-by:
Jean Sacren <sakiwit@gmail.com> Tested-by:
-
Jean Sacren authored
The second argument name in the kernel-doc argument list for i40e_features_check() was slightly off. Fix it for the kernel doc. Signed-off-by:
-
Catherine Sullivan authored
There is an error coming back from get_phy_capabilities that does not seem to have any functional implications. We will continue looking into why this error message is occurring, but in the meantime, we will move it to debug to avoid confusion. Change-ID: I9091754bf62c066ddedeb249923d85606e2d68ed Signed-off-by:
-
Catherine Sullivan authored
We were previously checking if autoneg was allowed to change before checking if autoneg was changing. We need to do this in the other order or else we will erroneously return EINVAL when autoneg is not changing. Change-ID: Iff9f7d1c9bddc1ad1e5d227d4f42754f90155410 Signed-off-by:
-
Anjali Singhai Jain authored
A mask value of 0x1FF was accidentally replaced with a bit mask causing flow director sideband to be broken. Change-ID: Id3387f67dd1b567b41692b570b383c58671e1eae Signed-off-by:
Anjali Singhai Jain <anjali.singhai@intel.com> Tested-by:
-
Carolyn Wyborny authored
This patch fixes a problem where the NVMUpdate Tool, when using the PHY NVM feature, gets bad data from the PHY because of contention on the MDIO interface from get PHY capability calls from the driver during regular operations. The problem is fixed by adding a check if media is available before calling get PHY capability function because that bit is not set when device is in PHY interaction mode. Change-ID: Ib89991b0f841808dd92410f5e8683d6ee3301cd0 Signed-off-by:
Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by:
-
Carolyn Wyborny authored
This patch fixes a problem where our Tools MAC Loopback test, after driver unbind would fail. This was because the hw was configured for multiqueue and unbind operation did not clear this configuration. The problem is fixed by resetting this configuration in i40e_remove. Change-ID: I130c05138319182ed1476d3a0b5222d6a6320af9 Signed-off-by:
-
Jesse Brandeburg authored
The adaptive ITR (interrupt throttle rate) algorithm was adjusting the hardware's interrupt rate too frequently. This caused a lot of variation in the interrupt rate for fairly constant workloads. Change the code to have a counter and adjust only once every N number of interrupts. Change-ID: I0460f1f86571037484eca5aca36ac4d889cb8389 Signed-off-by:
Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by:
-
Jesse Brandeburg authored
The dynamic algorithm, while now working, doesn't have good performance in 40G mode. One part of this patch addresses the high CPU utilization of some small streaming workloads that the driver should reduce CPU in. It also changes the minimum ITR that the dynamic algorithm will settle on, causing our minimum latency to go from 12us to about 14us, when using adaptive mode. It also changes the BULK interrupt rate to allow maximum throughput on a 40Gb connection with a single thread of transmit, clamping interrupt rate to 8000 for TX makes single thread traffic go too slow. The new ULTRA bulk setting is introduced and is used when the Rx packet rate on this queue exceeds 40000 packets per second. This value of 40000 was chosen because the automatic tuning of minimum ITR=20us means that a single queue can't quite achieve that many packets per second from a round-robin test. Change-ID: Icce8faa128688ca5fd2c4229bdd9726877a92ea2 Signed-off-by:
-
Jesse Brandeburg authored
The driver was using a value expressed in 2us increments for the divisor to figure out our bytes/usec values. Fix the usecs variable to contain a value in microseconds. Change-ID: I5c20493103c295d6f201947bb908add7040b7c41 Signed-off-by:
-
Jesse Brandeburg authored
This change moves a multi-line register setting into a function which simplifies reading the flow of the enable function. This also fixes a bug where the enable function was enabling the interrupt twice while trying to update the two interrupt throttle rate thresholds for Rx and Tx. Change-ID: Ie308f9d0d48540204590cb9d7a5a7b1196f959bb Signed-off-by:
-
Mitch Williams authored
When the VF driver is unable to communicate with the PF, it just gives up and never tries again. Aside from the obvious character flaw that this shows, it's also a lousy user experience. When PF communications fail, wait five seconds, and try again. And again. Don't give up, little VF driver! Your prince will come! Change-ID: Ia1378a39879883563b8faffce819f375821f9585 Signed-off-by:
Mitch Williams <mitch.a.williams@intel.com> Tested-by:
-
Alexander Duyck authored
The i40e_intr and i40e/i40evf_msix_clean_rings functions run from hard interrupt context or with interrupts already disabled in netpoll. They can use napi_schedule_irqoff() instead of napi_schedule() Signed-off-by:
Alexander Duyck <aduyck@mirantis.com> Tested-by:
-
Anjali Singhai authored
Acquire NVM, before issuing an AQ read nvm command for X722. We need to acquire the NVM before issuing an AQ read to the NVM otherwise we will get EBUSY from the FW. Also release when done. This fixes the two X722 issues with respect to eeprom checksum verify and reading NVM version info. With this patch in place, i40e driver will provide basic support for X722 devices. Signed-off-by:
-
Jesse Brandeburg authored
The defines from the RSS enabling call were mistakenly missed in the patches to the i40e which should have been to i40evf as well. This is a follow up to (commit ed921559886dd40528) "fix 32 bit build warnings". Signed-off-by:
-
Kiran Patil authored
This patch introduces a spinlock which is to be used for synchronizing access to VSI's MAC filter list. This patch also synchronizes execution of other codepaths which are accessing VSI's MAC filter list with execution of service_task:sync_vsi_filters. In function i40e_add_vsi, copied out LAA MAC address instead of cloning MAC filter entry because only MAC address is needed to remove MAC VLAN filter from FW/HW. Change-ID: I0e10ac7c715d44aa994239642aa4d57c998573a2 Signed-off-by:
Kiran Patil <kiran.patil@intel.com> Signed-off-by:
-
