- 09 Dec, 2015 22 commits
-
-
Eric Dumazet authored
These fields are updated but never read. Remove the overhead. Signed-off-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Eric Dumazet authored
Under heavy TX load, bnx2x_poll() can loop forever and trigger soft lockup bugs. A napi poll handler must yield after one TX completion round, risk of livelock is too high otherwise. Bug is very easy to trigger using a debug build, and udp flood, because of added cpu cycles in TX completion, and we do not receive enough packets to break the loop. Reported-by:
Willem de Bruijn <willemb@google.com> Signed-off-by:
-
Herbert Xu authored
The patch 9497df88 ("rhashtable: Fix reader/rehash race") added a pair of barriers. In fact the wmb is superfluous because every subsequent write to the old or new hash table uses rcu_assign_pointer, which itself carriers a full barrier prior to the assignment. Therefore we may remove the explicit wmb. Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au> Acked-by:
Thomas Graf <tgraf@suug.ch> Signed-off-by:
-
David S. Miller authored
Hariprasad Shenai says: ==================== Update Kconfig and some fixes for cxgb4 This series update Kconfig to add description for Chelsio's next generation T6 family of adapters, also fixes ethtool stats alignment and prevents simultaneous execution of service_ofldq thread, deals with queue wrap around and adds some fl counters for debugging purpose and device ID for new T5 adapters. This patch series has been created against net-next tree and includes patches on cxgb4 driver. We have included all the maintainers of respective drivers. Kindly review the change and let us know in case of any review comments. Thanks V2: Declare 'service_ofldq_running' as bool in Patch 4/7 ("cxgb4: prevent simultaneous execution of service_ofldq()") based on review comment by David Miller ==================== Signed-off-by: -
Hariprasad Shenai authored
Signed-off-by:
Hariprasad Shenai <hariprasad@chelsio.com> Signed-off-by:
-
Hariprasad Shenai authored
Add Free List DMA Mapping Errors to SGE Queue info for Free Lists. Add Free List "Low" counter to count the number of times we see the number of pointers that we _think_ the hardware sees in the Free List below the Egress Threshold. Signed-off-by:
-
Hariprasad Shenai authored
The WR headers may not fit within one descriptor. So we need to deal with wrap-around here. Based on original patch by Pranjal Joshi <pjoshi@chelsio.com> Signed-off-by:
-
Hariprasad Shenai authored
Change mutual exclusion mechanism to prevent multiple threads of execution from running in service_ofldq() at the same time. The old mechanism used an implicit guard on the down-call path and none on the restart path and wasn't working. This checking makes the mechanism explicit and is much easier to understand as a result. Based on original work by Casey Leedom <leedom@chelsio.com> Signed-off-by:
-
Hariprasad Shenai authored
Use helper macro ACCESS_ONCE() to load from the SGE status page to prevent the compiler loading multiple times. Based on original work by Mike Werner <werner@chelsio.com> Signed-off-by:
-
Hariprasad Shenai authored
Signed-off-by:
-
Hariprasad Shenai authored
Signed-off-by:
-
yankejian authored
here is the patch raising the performance of XGE by: 1)changes the way page management method for enet momery, and 2)reduces the count of rmb, and 3)adds Memory prefetching Signed-off-by:
Kejian Yan <yankejian@huawei.com> Signed-off-by:
-
Tejun Heo authored
In cgroup v1, dealing with cgroup membership was difficult because the number of membership associations was unbound. As a result, cgroup v1 grew several controllers whose primary purpose is either tagging membership or pull in configuration knobs from other subsystems so that cgroup membership test can be avoided. net_cls and net_prio controllers are examples of the latter. They allow configuring network-specific attributes from cgroup side so that network subsystem can avoid testing cgroup membership; unfortunately, these are not only cumbersome but also problematic. Both net_cls and net_prio aren't properly hierarchical. Both inherit configuration from the parent on creation but there's no interaction afterwards. An ancestor doesn't restrict the behavior in its subtree in anyway and configuration changes aren't propagated downwards. Especially when combined with cgroup delegation, this is problematic because delegatees can mess up whatever network configuration implemented at the system level. net_prio would allow the delegatees to set whatever priority value regardless of CAP_NET_ADMIN and net_cls the same for classid. While it is possible to solve these issues from controller side by implementing hierarchical allowable ranges in both controllers, it would involve quite a bit of complexity in the controllers and further obfuscate network configuration as it becomes even more difficult to tell what's actually being configured looking from the network side. While not much can be done for v1 at this point, as membership handling is sane on cgroup v2, it'd be better to make cgroup matching behave like other network matches and classifiers than introducing further complications. In preparation, this patch updates sock->sk_cgrp_data handling so that it points to the v2 cgroup that sock was created in until either net_prio or net_cls is used. Once either of the two is used, sock->sk_cgrp_data reverts to its previous role of carrying prioidx and classid. This is to avoid adding yet another cgroup related field to struct sock. As the mode switching can happen at most once per boot, the switching mechanism is aimed at lowering hot path overhead. It may leak a finite, likely small, number of cgroup refs and report spurious prioidx or classid on switching; however, dynamic updates of prioidx and classid have always been racy and lossy - socks between creation and fd installation are never updated, config changes don't update existing sockets at all, and prioidx may index with dead and recycled cgroup IDs. Non-critical inaccuracies from small race windows won't make any noticeable difference. This patch doesn't make use of the pointer yet. The following patch will implement netfilter match for cgroup2 membership. v2: Use sock_cgroup_data to avoid inflating struct sock w/ another cgroup specific field. v3: Add comments explaining why sock_data_prioidx() and sock_data_classid() use different fallback values. Signed-off-by:Tejun Heo <tj@kernel.org> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Daniel Wagner <daniel.wagner@bmw-carit.de> CC: Neil Horman <nhorman@tuxdriver.com> Signed-off-by:
-
Tejun Heo authored
Introduce sock->sk_cgrp_data which is a struct sock_cgroup_data. ->sk_cgroup_prioidx and ->sk_classid are moved into it. The struct and its accessors are defined in cgroup-defs.h. This is to prepare for overloading the fields with a cgroup pointer. This patch mostly performs equivalent conversions but the followings are noteworthy. * Equality test before updating classid is removed from sock_update_classid(). This shouldn't make any noticeable difference and a similar test will be implemented on the helper side later. * sock_update_netprioidx() now takes struct sock_cgroup_data and can be moved to netprio_cgroup.h without causing include dependency loop. Moved. * The dummy version of sock_update_netprioidx() converted to a static inline function while at it. Signed-off-by:
-
Tejun Heo authored
netprio builds per-netdev contiguous priomap array which is indexed by css->id. The array is allocated using kzalloc() effectively limiting the maximum ID supported to some thousand range. This patch caps the maximum supported css->id to USHRT_MAX which should be way above what is actually useable. This allows reducing sock->sk_cgrp_prioidx to u16 from u32. The freed up part will be used to overload the cgroup related fields. sock->sk_cgrp_prioidx's position is swapped with sk_mark so that the two cgroup related fields are adjacent. Signed-off-by:
Daniel Wagner <daniel.wagner@bmw-carit.de> Cc: Daniel Borkmann <daniel@iogearbox.net> CC: Neil Horman <nhorman@tuxdriver.com> Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroupDavid S. Miller authored
Preparatory changes for some new socket cgroup infrastructure and netfilter targets. Signed-off-by: -
Stefan Hajnoczi authored
This reverts commit 0d76d6e8 and merge commit c402293b, reversing changes made to c89359a4. The virtio-vsock device specification is not finalized yet. Michael Tsirkin voiced concerned about merging this code when the hardware interface (and possibly the userspace interface) could still change. Signed-off-by:
Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by:
-
David S. Miller authored
Sergei Shtylyov says: ==================== sh_eth: optimize MDIO code Here's a set of 3 patches against DaveM's 'net-next.git' repo which gets rid of ~35 LoCs in the MDIO bitbang methods. ==================== Signed-off-by: -
Sergei Shtylyov authored
After the MDIO bitbang code consolidation, there's no need anymore for bb_{set|clr}() as well as bb_read() -- just expand them inline, thus saving more LoCs... Signed-off-by:Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Acked-by:
Simon Horman <horms+renesas@verge.net.au> Signed-off-by:
-
Sergei Shtylyov authored
sh_mm[cd]_ctrl() and sh_set_mdio() all look mostly the same -- factor out their common code and put it into sh_mdio_ctrl(). Signed-off-by:
-
Sergei Shtylyov authored
The MDIO control bits are always mapped to the same bits of the same register (PIR), so there's no need to store their masks in the 'struct bb_info'... Signed-off-by:
-
Julia Lawall authored
The xgene_mac_ops and xgene_port_ops structures are never modified, so declare them as const. Done with the help of Coccinelle. Signed-off-by:
Julia Lawall <Julia.Lawall@lip6.fr> Signed-off-by:
-
- 08 Dec, 2015 2 commits
-
-
David S. Miller authored
Merge tag 'wireless-drivers-next-for-davem-2015-12-07' of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers-next Kalle Vallo says: ==================== brcfmac * support bcm4359 which can operate in two bands concurrently * disable runtime pm for USB avoiding issues * use generic pm callback in PCIe driver * support wowlan wake indication reporting * add beamforming support * unified handling of firmware files ath10k * support Manegement Frame Protection (MFP) * add thermal throttling support for 10.4 firmware * add support for pktlog in QCA99X0 * add debugfs file to enable Bluetooth coexistence feature * use firmware's native mesh interface type instead of raw mode iwlwifi * BT coex improvements * D3 operation bugfixes * rate control improvements * firmware debugging infra improvements * ground work for multi Rx * various security fixes ==================== Conflicts: drivers/net/wireless/ath/ath10k/pci.c The conflict resolution at: http://article.gmane.org/gmane.linux.kernel.next/37391 by Stephen Rothwell was used. Signed-off-by:
-
Rainer Weikusat authored
As the kernel generally uses negated error numbers, *err needs to be compared with -EAGAIN (d'oh). Signed-off-by:
Rainer Weikusat <rweikusat@mobileactivedefense.com> Fixes: ea3793ee ("core: enable more fine-grained datagram reception control") Signed-off-by:
-
- 07 Dec, 2015 16 commits
-
-
LABBE Corentin authored
the simple_strtoul function is obsolete. This patch replace it by kstrtox. Signed-off-by:
LABBE Corentin <clabbe.montjoie@gmail.com> Signed-off-by:
-
David S. Miller authored
Neil Armstrong says: ==================== Further fix for dsa unbinding This series fixes further issues for DSA dynamic unbinding. The first patch completely removes the PHY link state polling. The two following cleans up the dsa state upon removal. The last patch moves slave destroy code as slave function and adds missing netdev and phy cleanup calls. v1: http://lkml.kernel.org/r/562F8ECB.6050709@baylibre.com v2: http://lkml.kernel.org/r/56321D9A.8010109@baylibre.com remove phy fix and add missing calls in dsa_switch_destroy then add dedicated dsa_slave_destroy v3: remove polling instead of fixing it, make single patch for dsa slave destroy ==================== Acked-by:
Andrew Lunn <andrew@lunn.ch> Reviewed-by:
Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
-
Neil Armstrong authored
Move dsa slave dedicated code from dsa_switch_destroy to a new dsa_slave_destroy function in slave.c. Add the netif_carrier_off and phy_disconnect calls in order to correctly cleanup the netdev state and PHY state machine. Signed-off-by:
Frode Isaksen <fisaksen@baylibre.com> Signed-off-by:
Neil Armstrong <narmstrong@baylibre.com> Signed-off-by:
-
Neil Armstrong authored
Upon probe failure or unbinding, add missing dev_put() calls. Signed-off-by:
-
Neil Armstrong authored
Make sure that we unassign the master_netdev dsa_ptr to make the packet processing go through the regular Ethernet receive path. Suggested-by:
-
Neil Armstrong authored
Since no more DSA driver uses the polling callback, and since the phylib handles the link detection, remove the link polling work and timer code. Signed-off-by:
-
David S. Miller authored
Merge tag 'mac80211-next-for-davem-2015-12-07' of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211-next Johannes Berg says: ==================== This pull request got a bit bigger than I wanted, due to needing to reshuffle and fix some bugs. I merged mac80211 to get the right base for some of these changes. * new mac80211 API for upcoming driver changes: EOSP handling, key iteration * scan abort changes allowing to cancel an ongoing scan * VHT IBSS 80+80 MHz support * re-enable full AP client state tracking after fixes * various small fixes (that weren't relevant for mac80211) * various cleanups ==================== Signed-off-by:
-
David S. Miller authored
Sunil Goutham says: ==================== net: thunderx: Miscellaneous cleanups This patch series contains contains couple of cleanup patches. ==================== Signed-off-by: -
Sunil Goutham authored
Since we have moved on to using allocated pages to carve receive buffers instead of netdev_alloc_skb() there is no need to store any pointers for later retrieval. Earlier we had to store skb and skb->data pointers which later are used to handover received packet to network stack. This will avoid an unnecessary cache miss as well. Signed-off-by:
Sunil Goutham <sgoutham@cavium.com> Signed-off-by:
-
Yury Norov authored
The same switch-case repeates for nivc_*_intr functions. In this patch it is moved to a helper nicvf_int_type_to_mask(). By the way: - Unneeded write to NICVF register dropped if int_type is unknown. - netdev_dbg() is used instead of netdev_err(). Signed-off-by:
Yury Norov <yury.norov@auriga.com> Signed-off-by:
Aleksey Makarov <aleksey.makarov@caviumnetworks.com> Acked-by:
Vadim Lomovtsev <Vadim.Lomovtsev@caiumnetworks.com> Signed-off-by:
-
Ilan Peer authored
In case of HW ROC, when the driver reports that the ROC expired, it is not sufficient to purge the ROCs based on the remaining time, as it possible that the device finished the ROC session before the actual requested duration. To handle such cases, in case of ROC expired notification from the driver, complete all the ROCs which are marked with hw_begun, regardless of the remaining duration. Signed-off-by:
Ilan Peer <ilan.peer@intel.com> Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Rainer Weikusat authored
The current unix_dgram_recvsmg code acquires the u->readlock mutex in order to protect access to the peek offset prior to calling __skb_recv_datagram for actually receiving data. This implies that a blocking reader will go to sleep with this mutex held if there's presently no data to return to userspace. Two non-desirable side effects of this are that a later non-blocking read call on the same socket will block on the ->readlock mutex until the earlier blocking call releases it (or the readers is interrupted) and that later blocking read calls will wait longer than the effective socket read timeout says they should: The timeout will only start 'ticking' once such a reader hits the schedule_timeout in wait_for_more_packets (core.c) while the time it already had to wait until it could acquire the mutex is unaccounted for. The patch avoids both by using the __skb_try_recv_datagram and __skb_wait_for_more packets functions created by the first patch to implement a unix_dgram_recvmsg read loop which releases the readlock mutex prior to going to sleep and reacquires it as needed afterwards. Non-blocking readers will thus immediately return with -EAGAIN if there's no data available regardless of any concurrent blocking readers and all blocking readers will end up sleeping via schedule_timeout, thus honouring the configured socket receive timeout. Signed-off-by:
-
Rainer Weikusat authored
The __skb_recv_datagram routine in core/ datagram.c provides a general skb reception factility supposed to be utilized by protocol modules providing datagram sockets. It encompasses both the actual recvmsg code and a surrounding 'sleep until data is available' loop. This is inconvenient if a protocol module has to use additional locking in order to maintain some per-socket state the generic datagram socket code is unaware of (as the af_unix code does). The patch below moves the recvmsg proper code into a new __skb_try_recv_datagram routine which doesn't sleep and renames wait_for_more_packets to __skb_wait_for_more_packets, both routines being exported interfaces. The original __skb_recv_datagram routine is reimplemented on top of these two functions such that its user-visible behaviour remains unchanged. Signed-off-by:
-
Andrew Lunn authored
Device tree properties for a phy device are expected to be in the phy node. The current code for the DP83867 also tries to look in the parent node. The devices binding documentation does not mention this, no current device tree file makes use of this, and it is not behaviour we want. So remove looking in the parent device. Signed-off-by:
-
Bjørn Mork authored
Adding a writable sysfs attribute for the "NDP to end" quirk flag. This makes it easier for end users to test new devices for this firmware bug. We've been lucky so far, but we should not depend on reporters capable of rebuilding the driver. Cc: Enrico Mioso <mrkiko.rs@gmail.com> Signed-off-by:
Bjørn Mork <bjorn@mork.no> Signed-off-by:
-
David S. Miller authored
Or Gerlitz says: ==================== Add HA and LAG support for mlx4 SRIOV VFs This series is built upon the code added in commit ce388fff "Merge branch 'mlx4-next'" which added HA and LAG support to mlx4 RoCE and SRIOV services. We add HA and Link Aggregation support to single ported mlx4 Ethernet VFs. In this case, the PF Ethernet interfaces are bonded, the VFs see single port HW devices (already supported) -- however, now this port is highly available. This means that all VF HW QPs (both VF Ethernet driver and VF RoCE / RAW QPs) are subject to the V2P (Virtual-To-Physical) mapping which is managed by the PF driver, and hence resilient across link failures and such events. When bonding operates in Dynamic link aggregation (802.3ad) mode, traffic from each VF will go over the VF "base port" (the port this VF is assigned to by the admin) as long as this port is up. When the port fails, traffic from all VFs that are defined on this port will move to the other port, and be back to their base-port when it recovers. Moni and Or. ==================== Signed-off-by:
-
