- 16 Sep, 2016 29 commits
-
-
Eric Dumazet authored
When skb replaces another one in ooo queue, I forgot to also update tp->ooo_last_skb as well, if the replaced skb was the last one in the queue. To fix this, we simply can re-use the code that runs after an insertion, trying to merge skbs at the right of current skb. This not only fixes the bug, but also remove all small skbs that might be a subset of the new one. Example: We receive segments 2001:3001, 4001:5001 Then we receive 2001:8001 : We should replace 2001:3001 with the big skb, but also remove 4001:50001 from the queue to save space. packetdrill test demonstrating the bug 0.000 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0 +0 bind(3, ..., ...) = 0 +0 listen(3, 1) = 0 +0 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7> +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 7> +0.100 < . 1:1(0) ack 1 win 1024 +0 accept(3, ..., ...) = 4 +0.01 < . 1001:2001(1000) ack 1 win 1024 +0 > . 1:1(0) ack 1 <nop,nop, sack 1001:2001> +0.01 < . 1001:3001(2000) ack 1 win 1024 +0 > . 1:1(0) ack 1 <nop,nop, sack 1001:2001 1001:3001> Fixes: 9f5afeae ("tcp: use an RB tree for ooo receive queue") Signed-off-by:
Eric Dumazet <edumazet@google.com> Reported-by:
Yuchung Cheng <ycheng@google.com> Cc: Yaogong Wang <wygivan@google.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
David S. Miller authored
Sean Wang says: ==================== mediatek: add enhancement into the existing reset flow Current driver only resets DMA used by descriptor rings which can't guarantee it can recover all various kinds of fatal errors, so the patch 1) tries to reset the underlying hardware resource from scratch on Mediatek SoC required for ethernet running. 2) refactors code in order to the reusability of existing code. 3) considers handling for race condition between the reset flow and callbacks registered into core driver called about hardware accessing. 4) introduces power domain usage to hardware setup which leads to have cleanly and completely restore to the state as the initial. Changes since v1: - fix the build error with module built causing undefined symbol for pinctrl_bind_pins, so using pinctrl_select_state instead accomplishes the pin mux setup during the reset process. ==================== Reviewed-by:
Steve Wise <swise@opengridcomputing.com> Signed-off-by:
-
Sean Wang authored
add the protection of the race condition between the reset process and hardware access happening on the related callbacks. Signed-off-by:
Sean Wang <sean.wang@mediatek.com> Signed-off-by:
-
Sean Wang authored
struct mtk_eth has already contained struct regmap ethsys pointer to the address range of the internal circuit reset, so we reuse it to reset more internal blocks on ethernet hardware such as packet processing engine (PPE) and frame engine (FE) instead of rstc which deals with FE only. Signed-off-by:
-
Sean Wang authored
1) original driver only resets DMA used by descriptor rings which can't guarantee it can recover all various kinds of fatal errors, so the patch tries to reset the underlying hardware resource from scratch on Mediatek SoC required for ethernet running, including power, pin mux control, clock and internal circuits on the ethernet in order to restore into the initial state which the rebooted machine gives. 2) add state variable inside structure mtk_eth to help distinguish mtk_hw_init is called between the initialization during boot time or re-initialization during the reset process. 3) add ge_mode variable inside structure mtk_mac for restoring the interface mode of the current setup for the target MAC. 4) remove __init attribute from mtk_hw_init definition Signed-off-by:
-
Sean Wang authored
introduce power domain control which the digital circuit of the ethernet belongs to inside the flow of hardware initialization and deinitialization which helps the entire ethernet hardware block could restart cleanly and completely as being back to the initial state when the whole machine reboot. Signed-off-by:
-
Sean Wang authored
This cleans up the error path inside mtk_hw_init call, causing it able to exit appropriately when something fails and also includes refactoring mtk_cleanup call to make the partial logic reusable on the error path. Signed-off-by:
-
Sean Wang authored
grouping things related to the deinitialization of what mtk_hw_init call does that help to be reused by the reset process and the error path handling. Signed-off-by:
-
Sean Wang authored
the existing mtk_hw_init includes hardware and software initialization inside so that it is slightly hard to reuse them for the process of the reset recovery, so some splitting is made here for keeping hardware initializing relevant thing and the else such as IRQ registration and MDIO initialization what are all about to the interface of core driver moved to the other proper place because they have no needs to register IRQ and re-initialize structure again during the reset process. Signed-off-by:
-
David S. Miller authored
Merge tag 'rxrpc-rewrite-20160913-2' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs David Howells says: ==================== rxrpc: Support IPv6 Here is a set of patches that add IPv6 support. They need to be applied on top of the just-posted miscellaneous fix patches. They are: (1) Make autobinding of an unconnected socket work when sendmsg() is called to initiate a client call. (2) Don't specify the protocol when creating the client socket, but rather take the default instead. (3) Use rxrpc_extract_addr_from_skb() in a couple of places that were doing the same thing manually. This allows the IPv6 address extraction to be done in fewer places. (4) Add IPv6 support. With this, calls can be made to IPv6 servers from userspace AF_RXRPC programs; AFS, however, can't use IPv6 yet as the RPC calls need to be upgradeable. ==================== Reviewed-by:
-
David S. Miller authored
Merge tag 'rxrpc-rewrite-20160913-1' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs David Howells says: ==================== rxrpc: Miscellaneous fixes Here's a set of miscellaneous fix patches. There are a couple of points of note: (1) There is one non-fix patch that adjusts the call ref tracking tracepoint to make kernel API-held refs on calls more obvious. This is a prerequisite for the patch that fixes prealloc refcounting. (2) The final patch alters how jumbo packets that partially exceed the receive window are handled. Previously, space was being left in the Rx buffer for them, but this significantly hurts performance as the Rx window can't be increased to match the OpenAFS Tx window size. Instead, the excess subpackets are discarded and an EXCEEDS_WINDOW ACK is generated for the first. To avoid the problem of someone trying to run the kernel out of space by feeding the kernel a series of overlapping maximal jumbo packets, we stop allowing jumbo packets on a call if we encounter more than three jumbo packets with duplicate or excessive subpackets. ==================== Reviewed-by:
-
David S. Miller authored
Varun Prakash says: ==================== iw_cxgb4,cxgbit: remove duplicate code This patch series removes duplicate code from iw_cxgb4 and cxgbit by adding common function definitions in libcxgb. Please review. ==================== Reviewed-by:
-
Varun Prakash authored
Add cxgb_mk_rx_data_ack() to remove duplicate code to form CPL_RX_DATA_ACK hardware command. Signed-off-by:
Varun Prakash <varun@chelsio.com> Signed-off-by:
-
Varun Prakash authored
Add cxgb_mk_abort_rpl() to remove duplicate code to form CPL_ABORT_RPL hardware command. Signed-off-by:
-
Varun Prakash authored
Add cxgb_mk_abort_req() to remove duplicate code to form CPL_ABORT_REQ hardware command. Signed-off-by:
-
Varun Prakash authored
Add cxgb_mk_close_con_req() to remove duplicate code to form CPL_CLOSE_CON_REQ hardware command. Signed-off-by:
-
Varun Prakash authored
Add cxgb_mk_tid_release() to remove duplicate code to form CPL_TID_RELEASE hardware command. Signed-off-by:
-
Varun Prakash authored
Add cxgb_compute_wscale() in libcxgb_cm.h to remove it's duplicate definitions from cxgb4/cm.c and cxgbit/cxgbit_cm.c. Signed-off-by:
-
Varun Prakash authored
Add cxgb_best_mtu() in libcxgb_cm.h to remove it's duplicate definitions from cxgb4/cm.c and cxgbit/cxgbit_cm.c Signed-off-by:
-
Varun Prakash authored
Add cxgb_is_neg_adv() in libcxgb_cm.h to remove it's duplicate definitions from cxgb4/cm.c and cxgbit/cxgbit_cm.c. Signed-off-by:
-
Varun Prakash authored
Add cxgb_find_route6() in libcxgb_cm.c to remove it's duplicate definitions from cxgb4/cm.c and cxgbit/cxgbit_cm.c. Signed-off-by:
-
Varun Prakash authored
Add cxgb_find_route() in libcxgb_cm.c to remove it's duplicate definitions from cxgb4/cm.c and cxgbit/cxgbit_cm.c. Signed-off-by:
-
Varun Prakash authored
Add cxgb_get_4tuple() in libcxgb_cm.c to remove it's duplicate definitions from cxgb4/cm.c and cxgbit/cxgbit_cm.c. Signed-off-by:
-
Lance Richardson authored
The ovs kernel data path currently defers the execution of all recirc actions until stack utilization is at a minimum. This is too limiting for some packet forwarding scenarios due to the small size of the deferred action FIFO (10 entries). For example, broadcast traffic sent out more than 10 ports with recirculation results in packet drops when the deferred action FIFO becomes full, as reported here: http://openvswitch.org/pipermail/dev/2016-March/067672.html Since the current recursion depth is available (it is already tracked by the exec_actions_level pcpu variable), we can use it to determine whether to execute recirculation actions immediately (safe when recursion depth is low) or defer execution until more stack space is available. With this change, the deferred action fifo size becomes a non-issue for currently failing scenarios because it is no longer used when there are three or fewer recursions through ovs_execute_actions(). Suggested-by:Pravin Shelar <pshelar@ovn.org> Signed-off-by:
Lance Richardson <lrichard@redhat.com> Acked-by:
-
David S. Miller authored
Or Gerlitz says: ==================== net/sched: cls_flower: Add ports masks This series adds the ability to specify tcp/udp ports masks for TC/flower filter matches. I also removed an unused fields from the flower keys struct and clarified the format of the recently added vlan attibutes. v1--> v2 changes: * fixes typo in patch #2 title and change log (Sergei) * added acks provided by Jiri on v1 FWIW, by mistake the cover letter of V1 (but not the patches) carried V2 tag, hope this doesn't create too much confusion. ==================== Signed-off-by: -
Or Gerlitz authored
Specify the format (size and endianess) for the vlan attributes. Signed-off-by:
Or Gerlitz <ogerlitz@mellanox.com> Acked-by:
Jiri Pirko <jiri@mellanox.com> Signed-off-by:
-
Or Gerlitz authored
Commit c3f83241 "net: Add full IPv6 addresses to flow_keys" added an unused instance of struct flow_dissector_key_addrs into struct fl_flow_key, remove it. Signed-off-by:
Hadar Hen Zion <hadarh@mellanox.com> Acked-by:
-
Or Gerlitz authored
Add the definitions for src/dst udp/tcp port masks and use them when setting && dumping the relevant keys. Signed-off-by:
Paul Blakey <paulb@mellanox.com> Acked-by:
-
Tobias Regnery authored
In commit 9ee7b683 we moved the enablement of msi interrupts earlier in alx_init_intr. If there is an error in alx_alloc_rings, __alx_open returns with an error but msi (or msi-x) interrupts stays enabled. Add a new error label to disable msi (or msi-x) interrupts. Fixes: 9ee7b683 ("alx: refactor msi enablement and disablement") Signed-off-by:
Tobias Regnery <tobias.regnery@gmail.com> Signed-off-by:
-
- 15 Sep, 2016 8 commits
-
-
Hariprasad Shenai authored
The checksum provided by the device doesn't include the L3 headers, as IPv6 expects Signed-off-by:
Hariprasad Shenai <hariprasad@chelsio.com> Signed-off-by:
-
Jamal Hadi Salim authored
This action is intended to be an upgrade from a usability perspective from pedit (as well as operational debugability). Compare this: sudo tc filter add dev $ETH parent 1: protocol ip prio 10 \ u32 match ip protocol 1 0xff flowid 1:2 \ action pedit munge offset -14 u8 set 0x02 \ munge offset -13 u8 set 0x15 \ munge offset -12 u8 set 0x15 \ munge offset -11 u8 set 0x15 \ munge offset -10 u16 set 0x1515 \ pipe to: sudo tc filter add dev $ETH parent 1: protocol ip prio 10 \ u32 match ip protocol 1 0xff flowid 1:2 \ action skbmod dmac 02:15:15:15:15:15 Also try to do a MAC address swap with pedit or worse try to debug a policy with destination mac, source mac and etherype. Then make few rules out of those and you'll get my point. In the future common use cases on pedit can be migrated to this action (as an example different fields in ip v4/6, transports like tcp/udp/sctp etc). For this first cut, this allows modifying basic ethernet header. The most important ethernet use case at the moment is when redirecting or mirroring packets to a remote machine. The dst mac address needs a re-write so that it doesnt get dropped or confuse an interconnecting (learning) switch or dropped by a target machine (which looks at the dst mac). And at times when flipping back the packet a swap of the MAC addresses is needed. Signed-off-by:
Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by:
-
David S. Miller authored
Daniel Borkmann says: ==================== Misc cls_bpf/act_bpf improvements Two minor improvements to {cls,act}_bpf. For details please see individual patches. ==================== Signed-off-by: -
Daniel Borkmann authored
We have a small skb_at_tc_ingress() helper for testing for ingress, so make use of it. cls_bpf already uses it and so should act_bpf. Signed-off-by:
Daniel Borkmann <daniel@iogearbox.net> Acked-by:
Alexei Starovoitov <ast@kernel.org> Signed-off-by:
-
Daniel Borkmann authored
The skb_mac_header_was_set() test in cls_bpf's and act_bpf's fast-path is actually unnecessary and can be removed altogether. This was added by commit a166151c ("bpf: fix bpf helpers to use skb->mac_header relative offsets"), which was later on improved by 3431205e ("bpf: make programs see skb->data == L2 for ingress and egress"). We're always guaranteed to have valid mac header at the time we invoke cls_bpf_classify() or tcf_bpf(). Reason is that since 6d1ccff6 ("net: reset mac header in dev_start_xmit()") we do skb_reset_mac_header() in __dev_queue_xmit() before we could call into sch_handle_egress() or any subsequent enqueue. sch_handle_ingress() always sees a valid mac header as well (things like skb_reset_mac_len() would badly fail otherwise). Thus, drop the unnecessary test in classifier and action case. Signed-off-by:
-
Hadar Hen Zion authored
Remove rcu_read_lock protection from tunnel_key_dump and use rtnl_dereference, dump operation is protected by rtnl lock. Also, remove rcu_read_lock from tunnel_key_release and use rcu_dereference_protected. Both operations are running exclusively and a writer couldn't modify t->params while those functions are executed. Fixes: 54d94fd89d90 ('net/sched: Introduce act_tunnel_key') Signed-off-by:John Fastabend <john.r.fastabend@intel.com> Acked-by:
-
Jakub Kicinski authored
Commit d5709f7a ("flow_dissector: For stripped vlan, get vlan info from skb->vlan_tci") made flow dissector look at vlan_proto when vlan is present. Since test_bpf sets skb->vlan_tci to ~0 (including VLAN_TAG_PRESENT) we have to populate skb->vlan_proto. Fixes false negative on test #24: test_bpf: #24 LD_PAYLOAD_OFF jited:0 175 ret 0 != 42 FAIL (1 times) Signed-off-by:
Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by:
Dinan Gunawardena <dinan.gunawardena@netronome.com> Acked-by:
-
Colin Ian King authored
The newline escape is incorrect and needs fixing. Also adjust source formatting / indentation and add { } to trailing else. Signed-off-by:Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
- 13 Sep, 2016 3 commits
-
-
David Howells authored
Add IPv6 support to AF_RXRPC. With this, AF_RXRPC sockets can be created: service = socket(AF_RXRPC, SOCK_DGRAM, PF_INET6); instead of: service = socket(AF_RXRPC, SOCK_DGRAM, PF_INET); The AFS filesystem doesn't support IPv6 at the moment, though, since that requires upgrades to some of the RPC calls. Note that a good portion of this patch is replacing "%pI4:%u" in print statements with "%pISpc" which is able to handle both protocols and print the port. Signed-off-by:David Howells <dhowells@redhat.com>
-
David Howells authored
There are two places that want to transmit a packet in response to one just received and manually pick the address to reply to out of the sk_buff. Make them use rxrpc_extract_addr_from_skb() instead so that IPv6 is handled automatically. Signed-off-by: -
David Howells authored
Pass 0 as the protocol argument when creating the transport socket rather than IPPROTO_UDP. Signed-off-by:
-
