- 11 Nov, 2017 33 commits
-
-
David S. Miller authored
Yuchung Cheng says: ==================== remove FACK loss recovery This patch set removes the forward-acknowledgment (FACK) packet-based loss and reordering detection. This simplifies TCP loss recovery since the SACK scoreboard no longer needs to track the number of pending packets under highest SACKed sequence. FACK is subsumed by the time-based RACK loss detection which is more robust under reordering and second order losses. ==================== Signed-off-by:David S. Miller <davem@davemloft.net>
-
Yuchung Cheng authored
Replace the reordering distance measurement in packet unit with sequence based approach. Previously it trackes the number of "packets" toward the forward ACK (i.e. highest sacked sequence)in a state variable "fackets_out". Precisely measuring reordering degree on packet distance has not much benefit, as the degree constantly changes by factors like path, load, and congestion window. It is also complicated and prone to arcane bugs. This patch replaces with sequence-based approach that's much simpler. Signed-off-by:
Yuchung Cheng <ycheng@google.com> Reviewed-by:
Eric Dumazet <edumazet@google.com> Reviewed-by:
Neal Cardwell <ncardwell@google.com> Reviewed-by:
Soheil Hassas Yeganeh <soheil@google.com> Reviewed-by:
Priyaranjan Jha <priyarjha@google.com> Signed-off-by:
-
Yuchung Cheng authored
FACK loss detection has been disabled by default and the successor RACK subsumed FACK and can handle reordering better. This patch removes FACK to simplify TCP loss recovery. Signed-off-by:
-
Gustavo A. R. Silva authored
In preparation to enabling -Wimplicit-fallthrough, mark switch cases where we are expecting to fall through. Addresses-Coverity-ID: 1397960 Signed-off-by:
Gustavo A. R. Silva <garsilva@embeddedor.com> Signed-off-by:
-
Gustavo A. R. Silva authored
In preparation to enabling -Wimplicit-fallthrough, mark switch cases where we are expecting to fall through. Addresses-Coverity-ID: 1397972 Signed-off-by:
-
Nathan Fontenot authored
Update the login buffer to include client data for the vnic driver, this includes the OS name, LPAR name, and device name. This update allows this information to be available in the VIOS. Signed-off-by:
Nathan Fontenot <nfont@linux.vnet.ibm.com> Signed-off-by:
-
David S. Miller authored
NACK'd by x86 maintainer. Signed-off-by: -
David S. Miller authored
Lawrence Brakmo says: ==================== bpf: Fix bugs in sock_ops samples The programs were returning -1 in some cases when they should only return 0 or 1. Changes in the verifier now catch this issue and the programs fail to load. This is now fixed. ==================== Signed-off-by: -
Lawrence Brakmo authored
The program was returning -1 in some cases which is not allowed by the verifier any longer. Fixes: 390ee7e2 ("bpf: enforce return code for cgroup-bpf programs") Signed-off-by:
Lawrence Brakmo <brakmo@fb.com> Signed-off-by:
-
Lawrence Brakmo authored
The program was returning -1 in some cases which is not allowed by the verifier any longer. Fixes: 390ee7e2 ("bpf: enforce return code for cgroup-bpf programs") Signed-off-by:
-
Lawrence Brakmo authored
The program was returning -1 in some cases which is not allowed by the verifier any longer. Fixes: 390ee7e2 ("bpf: enforce return code for cgroup-bpf programs") Signed-off-by:
-
Lawrence Brakmo authored
The program was returning -1 in some cases which is not allowed by the verifier any longer. Fixes: 390ee7e2 ("bpf: enforce return code for cgroup-bpf programs") Signed-off-by:
-
Lawrence Brakmo authored
The program was returning -1 in some cases which is not allowed by the verifier any longer. Fixes: 390ee7e2 ("bpf: enforce return code for cgroup-bpf programs") Signed-off-by:
-
Lawrence Brakmo authored
The program was returning -1 in some cases which is not allowed by the verifier any longer. Fixes: 390ee7e2 ("bpf: enforce return code for cgroup-bpf programs") Signed-off-by:
-
Jon Maloy authored
Currently, the TIPC RPS dissector is based only on the incoming packets' source node address, hence steering all traffic from a node to the same core. We have seen that this makes the links vulnerable to starvation and unnecessary resets when we turn down the link tolerance to very low values. To reduce the risk of this happening, we exempt probe and probe replies packets from the convergence to one core per source node. Instead, we do the opposite, - we try to diverge those packets across as many cores as possible, by randomizing the flow selector key. To make such packets identifiable to the dissector, we add a new 'is_keepalive' bit to word 0 of the LINK_PROTOCOL header. This bit is set both for PROBE and PROBE_REPLY messages, and only for those. It should be noted that these packets are not part of any flow anyway, and only constitute a minuscule fraction of all packets sent across a link. Hence, there is no risk that this will affect overall performance. Acked-by:
Ying Xue <ying.xue@windriver.com> Signed-off-by:
Jon Maloy <jon.maloy@ericsson.com> Signed-off-by:
-
David S. Miller authored
Michael Grzeschik says: ==================== net: macb: add error handling on probe and This series adds more error handling to the macb driver. ==================== Signed-off-by: -
Michael Grzeschik authored
We add the call of_node_put(bp->phy_node) to all associated error paths for memory clean up. Signed-off-by:
Michael Grzeschik <m.grzeschik@pengutronix.de> Acked-by:
Nicolas Ferre <nicolas.ferre@microchip.com> Signed-off-by:
-
Michael Grzeschik authored
We add the call of_phy_deregister_fixed_link to all associated error paths for memory clean up. Signed-off-by:
-
Miquel Raynal authored
GOP statistics from all ports of one instance of the driver are gathered with one work recalled in loop in a workqueue. The loop is started when a port is up, and stopped when a port is down. This last condition is obviously wrong. Fix this by having a work per port. This way, starting and stoping it when the port is up or down will be fine, while minimizing unnecessary CPU usage. Fixes: 118d6298 ("net: mvpp2: add ethtool GOP statistics") Reported-by:
Stefan Chulski <stefanc@marvell.com> Signed-off-by:
Miquel Raynal <miquel.raynal@free-electrons.com> Signed-off-by:
-
David S. Miller authored
Lipeng says: ==================== net: hns3: Bug fixes & Code improvements in HNS3 driver This patch-set introduces some bug fixes and code improvements. As [patch 1/2] depends on the patch {5392902d net: hns3: Consistently using GENMASK in hns3 driver}, which exists in net-next, not exists in net, so push this serise to nex-next. ==================== Signed-off-by: -
Fuyun Liang authored
When checking whether auto-negotiation is on, driver only needs to check the value of mac.autoneg(SW) directly, and does not need to query it from hardware. Because this value is always synchronized with the auto-negotiation state of hardware. This patch removes mac auto-negotiation state query in hclge_update_speed_duplex(). Fixes: 46a3df9f (net: hns3: Add HNS3 Acceleration Engine & Compatibility Layer Support) Signed-off-by:
Fuyun Liang <liangfuyun1@huawei.com> Signed-off-by:
Lipeng <lipeng321@huawei.com> Signed-off-by:
-
Fuyun Liang authored
Driver gets phy address from NCL_config file and uses the phy address to initialize phydev. There are 5 bits for phy address. And C22 phy address has 5 bits. So 0-31 are all valid address for phy. If there is no phy, it will crash. Because driver always get a valid phy address. This patch fixes the phy address to 8 bits, and use 0xff to indicate invalid phy address. Fixes: 46a3df9f (net: hns3: Add HNS3 Acceleration Engine & Compatibility Layer Support) Signed-off-by:
-
David Ahern authored
Attributes using NLA_U* and NLA_S* (where * is 8, 16,32 and 64) are expected to be an exact length. Split these data types from nla_attr_minlen into nla_attr_len and update validate_nla to require the attribute to have exact length for them. Signed-off-by:
David Ahern <dsahern@gmail.com> Signed-off-by:
-
Maciej Żenczykowski authored
Add a per-device sysctl to specify the default traffic class to use for kernel originated IPv6 Neighbour Discovery packets. Currently this includes: - Router Solicitation (ICMPv6 type 133) ndisc_send_rs() -> ndisc_send_skb() -> ip6_nd_hdr() - Neighbour Solicitation (ICMPv6 type 135) ndisc_send_ns() -> ndisc_send_skb() -> ip6_nd_hdr() - Neighbour Advertisement (ICMPv6 type 136) ndisc_send_na() -> ndisc_send_skb() -> ip6_nd_hdr() - Redirect (ICMPv6 type 137) ndisc_send_redirect() -> ndisc_send_skb() -> ip6_nd_hdr() and if the kernel ever gets around to generating RA's, it would presumably also include: - Router Advertisement (ICMPv6 type 134) (radvd daemon could pick up on the kernel setting and use it) Interface drivers may examine the Traffic Class value and translate the DiffServ Code Point into a link-layer appropriate traffic prioritization scheme. An example of mapping IETF DSCP values to IEEE 802.11 User Priority values can be found here: https://tools.ietf.org/html/draft-ietf-tsvwg-ieee-802-11 The expected primary use case is to properly prioritize ND over wifi. Testing: jzem22:~# cat /proc/sys/net/ipv6/conf/eth0/ndisc_tclass 0 jzem22:~# echo -1 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass -bash: echo: write error: Invalid argument jzem22:~# echo 256 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass -bash: echo: write error: Invalid argument jzem22:~# echo 0 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# echo 255 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# cat /proc/sys/net/ipv6/conf/eth0/ndisc_tclass 255 jzem22:~# echo 34 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# cat /proc/sys/net/ipv6/conf/eth0/ndisc_tclass 34 jzem22:~# echo $[0xDC] > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# tcpdump -v -i eth0 icmp6 and src host jzem22.pgc and dst host fe80::1 tcpdump: listening on eth0, link-type EN10MB (Ethernet), capture size 262144 bytes IP6 (class 0xdc, hlim 255, next-header ICMPv6 (58) payload length: 24) jzem22.pgc > fe80::1: [icmp6 sum ok] ICMP6, neighbor advertisement, length 24, tgt is jzem22.pgc, Flags [solicited] (based on original change written by Erik Kline, with minor changes) v2: fix 'suspicious rcu_dereference_check() usage' by explicitly grabbing the rcu_read_lock. Cc: Lorenzo Colitti <lorenzo@google.com> Signed-off-by:Erik Kline <ek@google.com> Signed-off-by:
Maciej Żenczykowski <maze@google.com> Signed-off-by:
-
Samuel Mendoza-Jonas authored
Several response handlers return EBUSY if the data corresponding to the command/response pair is already set. There is no reason to return an error here; the channel is advertising something as enabled because we told it to enable it, and it's possible that the feature has been enabled previously. Signed-off-by:
Samuel Mendoza-Jonas <sam@mendozajonas.com> Signed-off-by:
-
Samuel Mendoza-Jonas authored
The NCSI driver is mostly silent which becomes a headache when trying to determine what has occurred on the NCSI connection. This adds additional logging in a few key areas such as state transitions and calling out certain errors more visibly. Signed-off-by:
-
David S. Miller authored
Prashant Bhole says: ==================== tools: bpftool: show filenames of pinned objects This patchset adds support to show pinned objects in object details. Patch1 adds a funtionality to open a path in bpf-fs regardless of its object type. Patch2 adds actual functionality by scanning the bpf-fs once and adding object information in hash table, with object id as a key. One object may be associated with multiple paths because an object can be pinned multiple times Patch3 adds command line option to enable this functionality. Making it optional because scanning bpf-fs can be costly. ==================== Acked-by:Jakub Kicinski <jakub.kicinski@netronome.com>
-
Prashant Bhole authored
Making it optional to show file names of pinned objects because it scans complete bpf-fs filesystem which is costly. Added option -f|--bpffs. Documentation updated. Signed-off-by:
Prashant Bhole <bhole_prashant_q7@lab.ntt.co.jp> Signed-off-by:
-
Prashant Bhole authored
Added support to show filenames of pinned objects. For example: root@test# ./bpftool prog 3: tracepoint name tracepoint__irq tag f677a7dd722299a3 loaded_at Oct 26/11:39 uid 0 xlated 160B not jited memlock 4096B map_ids 4 pinned /sys/fs/bpf/softirq_prog 4: tracepoint name tracepoint__irq tag ea5dc530d00b92b6 loaded_at Oct 26/11:39 uid 0 xlated 392B not jited memlock 4096B map_ids 4,6 root@test# ./bpftool --json --pretty prog [{ "id": 3, "type": "tracepoint", "name": "tracepoint__irq", "tag": "f677a7dd722299a3", "loaded_at": "Oct 26/11:39", "uid": 0, "bytes_xlated": 160, "jited": false, "bytes_memlock": 4096, "map_ids": [4 ], "pinned": ["/sys/fs/bpf/softirq_prog" ] },{ "id": 4, "type": "tracepoint", "name": "tracepoint__irq", "tag": "ea5dc530d00b92b6", "loaded_at": "Oct 26/11:39", "uid": 0, "bytes_xlated": 392, "jited": false, "bytes_memlock": 4096, "map_ids": [4,6 ], "pinned": [] } ] root@test# ./bpftool map 4: hash name start flags 0x0 key 4B value 16B max_entries 10240 memlock 1003520B pinned /sys/fs/bpf/softirq_map1 5: hash name iptr flags 0x0 key 4B value 8B max_entries 10240 memlock 921600B root@test# ./bpftool --json --pretty map [{ "id": 4, "type": "hash", "name": "start", "flags": 0, "bytes_key": 4, "bytes_value": 16, "max_entries": 10240, "bytes_memlock": 1003520, "pinned": ["/sys/fs/bpf/softirq_map1" ] },{ "id": 5, "type": "hash", "name": "iptr", "flags": 0, "bytes_key": 4, "bytes_value": 8, "max_entries": 10240, "bytes_memlock": 921600, "pinned": [] } ] Signed-off-by: -
Prashant Bhole authored
This was needed for opening any file in bpf-fs without knowing its object type Signed-off-by:
-
David S. Miller authored
Josef Bacik says: ==================== Add the ability to do BPF directed error injection I'm sending this through Dave since it'll conflict with other BPF changes in his tree, but since it touches tracing as well Dave would like a review from somebody on the tracing side. v4->v5: - disallow kprobe_override programs from being put in the prog map array so we don't tail call into something we didn't check. This allows us to make the normal path still fast without a bunch of percpu operations. v3->v4: - fix a build error found by kbuild test bot (I didn't wait long enough apparently.) - Added a warning message as per Daniels suggestion. v2->v3: - added a ->kprobe_override flag to bpf_prog. - added some sanity checks to disallow attaching bpf progs that have ->kprobe_override set that aren't for ftrace kprobes. - added the trace_kprobe_ftrace helper to check if the trace_event_call is a ftrace kprobe. - renamed bpf_kprobe_state to bpf_kprobe_override, fixed it so we only read this value in the kprobe path, and thus only write to it if we're overriding or clearing the override. v1->v2: - moved things around to make sure that bpf_override_return could really only be used for an ftrace kprobe. - killed the special return values from trace_call_bpf. - renamed pc_modified to bpf_kprobe_state so bpf_override_return could tell if it was being called from an ftrace kprobe context. - reworked the logic in kprobe_perf_func to take advantage of bpf_kprobe_state. - updated the test as per Alexei's review. - Original message - A lot of our error paths are not well tested because we have no good way of injecting errors generically. Some subystems (block, memory) have ways to inject errors, but they are random so it's hard to get reproduceable results. With BPF we can add determinism to our error injection. We can use kprobes and other things to verify we are injecting errors at the exact case we are trying to test. This patch gives us the tool to actual do the error injection part. It is very simple, we just set the return value of the pt_regs we're given to whatever we provide, and then override the PC with a dummy function that simply returns. Right now this only works on x86, but it would be simple enough to expand to other architectures. Thanks, ==================== Signed-off-by: -
Josef Bacik authored
This adds a basic test for bpf_override_return to verify it works. We override the main function for mounting a btrfs fs so it'll return -ENOMEM and then make sure that trying to mount a btrfs fs will fail. Acked-by:
Alexei Starovoitov <ast@kernel.org> Signed-off-by:
Josef Bacik <jbacik@fb.com> Acked-by:
Daniel Borkmann <daniel@iogearbox.net> Signed-off-by:
-
Josef Bacik authored
Error injection is sloppy and very ad-hoc. BPF could fill this niche perfectly with it's kprobe functionality. We could make sure errors are only triggered in specific call chains that we care about with very specific situations. Accomplish this with the bpf_override_funciton helper. This will modify the probe'd callers return value to the specified value and set the PC to an override function that simply returns, bypassing the originally probed function. This gives us a nice clean way to implement systematic error injection for all of our code paths. Acked-by:
-
- 10 Nov, 2017 7 commits
-
-
Girish Moodalbail authored
The commit bcc6d479 ("net: vlan: make non-hw-accel rx path similar to hw-accel") unified accel and non-accel path for VLAN RX. With that fix we do not register any packet_type handler for VLANs anymore, so fix the incorrect comment. Signed-off-by:
Girish Moodalbail <girish.moodalbail@oracle.com> Signed-off-by:
-
David S. Miller authored
Manish Kurup says: ==================== net_sched actions: act_vlan now uses RCU This commit consists of 3 patches: patch1 (1/3): The VLAN action maintains one set of stats across all cores, and uses a spinlock to synchronize updates to it from the same. Changed this to use a per-CPU stats context instead. This change will result in better performance. patch2 (2/3): Modified netronome nfp flower action to use VLAN helper functions instead of accessing/referencing TC act_vlan private structures directly. patch3 (3/3): Using a spinlock in the VLAN action causes performance issues when the VLAN action is used on multiple cores. Rewrote the VLAN action to use RCU read locking for reads and updates instead. All functions now use an RCU dereferenced pointer to access the VLAN action context. Modified helper functions used by other modules, to use the RCU as opposed to directly accessing the structure. As part of this review, there were some changes suggested by reviewers. I have incorporated all the changes that were requested. Here're the changes: v2: Fixed all helper functions to use RCU (rtnl_dereference) - Eric, Jamal v2: Fixed indentation, extra line nits - Jamal, Jiri v2: Moved rcu_head to the end of the struct - Jiri v2: Re-formatted locals to reverse-christmas-tree - Jiri v2: Removed mismatched spin_lock() - Cong v2: Removed spin_lock_bh() in tcf_vlan_init, rtnl_dereference() should suffice - Cong, Jiri v4: Modified the nfp flower action code to use the VLAN helper functions instead of referencing the structure directly. Isolated this into a separate patch - Pieter Jansen v5: Got rid of the unlikely() for the allocation case - Simon Horman v6: Had forgotten cleanup functions for RCU alloc, added them - Dave Miller v7: Re-formatted more locals to reverse-christmas-tree - Pieter V v8: Reverted reverse-christmas-tree(v7), not required when dependencies make it difficult to implement - Alexander D v9: Cover letter subject change - Jamal ==================== Signed-off-by: -
Manish Kurup authored
Using a spinlock in the VLAN action causes performance issues when the VLAN action is used on multiple cores. Rewrote the VLAN action to use RCU read locking for reads and updates instead. All functions now use an RCU dereferenced pointer to access the VLAN action context. Modified helper functions used by other modules, to use the RCU as opposed to directly accessing the structure. Acked-by:
Jamal Hadi Salim <jhs@mojatatu.com> Acked-by:
Jiri Pirko <jiri@mellanox.com> Signed-off-by:
Manish Kurup <manish.kurup@verizon.com> Signed-off-by:
-
Manish Kurup authored
Modified netronome nfp flower action to use VLAN helper functions instead of accessing/referencing TC act_vlan private structures directly. Reviewed-by:
Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by:
-
Manish Kurup authored
The VLAN action maintains one set of stats across all cores, and uses a spinlock to synchronize updates to it from the same. Changed this to use a per-CPU stats context instead. This change will result in better performance. Acked-by:
-
Robert Stonehouse authored
Fixes: 535a6177 ("sfc: suppress handled MCDI failures when changing the MAC address") Signed-off-by:
Bert Kenward <bkenward@solarflare.com> Signed-off-by:
-
Colin Ian King authored
There are several pointers that are being assigned but never read so remove these as they are redundant. Also remove an assignment to function_mode that is never read. Cleans up several clang warnings: vxge-main.c:1139:2: warning: Value stored to 'hldev' is never read vxge-main.c:1294:2: warning: Value stored to 'hldev' is never read vxge-main.c:2188:2: warning: Value stored to 'dev' is never read vxge-main.c:2188:2: warning: Value stored to 'dev' is never read vxge-main.c:2723:2: warning: Value stored to 'function_mode' is never read Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
