- 17 Nov, 2018 26 commits
-
-
Jesus Sanchez-Palencia authored
Currently on dequeue() ETF only drops the first expired packet, which causes a problem if the next packet is already expired. When this happens, the watchdog will be configured with a time in the past, fire straight way and the packet will finally be dropped once the dequeue() function of the qdisc is called again. We can save quite a few cycles and improve the overall behavior of the qdisc if we drop all expired packets if the next packet is expired. This should allow ETF to recover faster from bad situations. But packet drops are still a very serious warning that the requirements imposed on the system aren't reasonable. This was inspired by how the implementation of hrtimers use the rb_tree inside the kernel. Signed-off-by:
Jesus Sanchez-Palencia <jesus.s.palencia@gmail.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Jesus Sanchez-Palencia authored
This is just a refactor that will simplify the implementation of the next patch in this series which will drop all expired packets on the dequeue flow. Signed-off-by:
-
Jesus Sanchez-Palencia authored
ETF's peek() operation is heavily used so use an rb_root_cached instead and leverage rb_first_cached() which will run in O(1) instead of O(log n). Even if on 'timesortedlist_clear()' we could be using rb_erase(), we choose to use rb_erase_cached(), because if in the future we allow runtime changes to ETF parameters, and need to do a '_clear()', this might cause some hard to debug issues. Signed-off-by:
-
Jesus Sanchez-Palencia authored
There is no point in firing the qdisc watchdog if there are no future skbs pending in the queue and the watchdog had been set previously. Signed-off-by:
-
Yafang Shao authored
Currently we can use bpf or tcp tracepoint to conveniently trace the tcp state transition at the run time. So we don't need to do this stuff at the compile time anymore. Signed-off-by:
Yafang Shao <laoar.shao@gmail.com> Signed-off-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
-
David S. Miller authored
Ben Dooks says: ==================== SMSC95xx driver updates (round 1) This is a series of a few driver cleanups and some fixups of the code for the SMSC95XX driver. There have been a few reviews, and the issues have been fixed so this should be ready for merging. I will work on the tx-alignment and the other bits of usbnet changes and produce at least two more patch series for this later. ==================== Signed-off-by: -
Ben Dooks authored
The manual states that the checksum cannot lie in the last DWORD of the transmission, so add a basic check for this and fall back to software checksumming the packet. This only seems to trigger for ACK packets with no options or data to return to the other end, and the use of the tx-alignment option makes it more likely to happen. Signed-off-by:
Ben Dooks <ben.dooks@codethink.co.uk> Signed-off-by:
-
Ben Dooks authored
Change the RX code to use get_unaligned_le32() instead of the combo of memcpy and cpu_to_le32s(&var). Signed-off-by:
-
Ben Dooks authored
The smsc95xx_tx_fixup is doing multiple calls to skb_push() to put an 8-byte command header onto the packet. It would be easier to do one skb_push() and then copy the data in once the push is done. We also make the code smaller by using proper unaligned puts for the header. This merges in the CPU to LE32 conversion as well and makes the whole sequence easier to understand hopefully. Signed-off-by:
-
Ben Dooks authored
The smsc95xx driver already takes into account the NET_IP_ALIGN parameter when setting up the receive packet data, which means we do not need to worry about aligning the packets in the usbnet driver. Adding the EVENT_NO_IP_ALIGN means that the IPv4 header is now passed to the ip_rcv() routine with the start on an aligned address. Tested on Raspberry Pi B3. Signed-off-by:
-
David S. Miller authored
Ioana Ciocoi Radulescu says: ==================== dpaa2-eth: add bql support The first two patches make minor tweaks to the driver to simplify bql implementation. The third patch adds the actual bql support. ==================== Signed-off-by: -
Ioana Ciocoi Radulescu authored
Add support for byte queue limit. On NAPI poll, we save the total number of Tx confirmed frames/bytes and register them with bql at the end of the poll function. Signed-off-by:
Ioana Radulescu <ruxandra.radulescu@nxp.com> Signed-off-by:
-
Ioana Ciocoi Radulescu authored
Change the frame consume callback signature: * the entire FQ structure is passed to the callback instead of just the queue index * the NAPI structure can be easily obtained from the channel it is associated to, so we don't need to pass it explicitly Signed-off-by:
-
Ioana Ciocoi Radulescu authored
The DPNI object on which we build a network interface has a certain number of {Rx, Tx, Tx confirmation} frame queues as resources. The default hardware setup offers one queue of each type, as well as one DPCON channel, for each core available in the system. There are however cases where the number of queues is greater than the number of cores or channels. Until now, we configured and used all the frame queues associated with a DPNI, even if it meant assigning multiple queues of one type to the same channel. Update the driver to only use a number of queues equal to the number of channels, ensuring each channel will contain exactly one Rx and one Tx confirmation queue. >From the user viewpoint, this change is completely transparent. Performance wise there is no impact in most scenarios. In case the number of queues is larger than and not a multiple of the number of channels, Rx hash distribution offers now better load balancing between cores, which can have a positive impact on overall system performance. Signed-off-by: -
Jiri Pirko authored
Currently, the vlan packet offloads are registered only upon 8021q module load. However, even without this module loaded, the offloads could be utilized, for example by openvswitch datapath. As reported by Michael, that causes 2x to 5x performance improvement, depending on a testcase. So move the vlan offload registrations into vlan_core and make this available even without 8021q module loaded. Reported-by:
Michael Shteinbok <michaelsh86@gmail.com> Signed-off-by:
Jiri Pirko <jiri@mellanox.com> Tested-by:
David Ahern <dsahern@gmail.com> Signed-off-by:
-
Colin Ian King authored
There is a missing indentation before the declaration of port. Add it. Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Colin Ian King authored
Trivial fix, the spelling of "failded" is incorrect in dev_err and dev_warn messages. Fix this. Signed-off-by:
-
Cong Wang authored
Signed-off-by:
Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by:
-
Heiner Kallweit authored
Now that the icplus driver has been fixed all PHY drivers supporting interrupts have both callbacks (config_intr and ack_interrupt) implemented - as it should be. Therefore phy_drv_supports_irq() can be changed now to check for both callbacks being implemented. Signed-off-by:
Heiner Kallweit <hkallweit1@gmail.com> Acked-by:
Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
-
David S. Miller authored
Michał Mirosław says: ==================== Remove VLAN.CFI overload Fix BPF code/JITs to allow for separate VLAN_PRESENT flag storage and finally move the flag to separate storage in skbuff. This is final step to make CLAN.CFI transparent to core Linux networking stack. An #ifdef is introduced temporarily to mark fragments masking VLAN_TAG_PRESENT. This is removed altogether in the final patch. ==================== Signed-off-by: -
Michał Mirosław authored
Replace VLAN_TAG_PRESENT with single bit flag and free up VLAN.CFI overload. Now VLAN.CFI is visible in networking stack and can be passed around intact. Signed-off-by:
Michał Mirosław <mirq-linux@rere.qmqm.pl> Signed-off-by:
-
Michał Mirosław authored
Signed-off-by:
-
Michał Mirosław authored
Signed-off-by:
-
Michał Mirosław authored
Signed-off-by:
-
Michał Mirosław authored
Signed-off-by:
-
Michał Mirosław authored
Wrap VLAN_PRESENT bit using macro like PKT_TYPE_* and CLONED_*, as used by BPF code. Signed-off-by:
-
- 16 Nov, 2018 1 commit
-
-
git://git.open-mesh.org/linux-mergeDavid S. Miller authored
Simon Wunderlich says: ==================== This feature/cleanup patchset includes the following patches: - Bump version strings, by Simon Wunderlich - Fixup includes, by Sven Eckelmann (3 patches) - Separate BATMAN_ADV_DEBUG from DEBUGFS, by Sven Eckelmann - Fixup tracing log documentation, by Sven Eckelmann - Use exclusive locks to secure netlink information dump transfers, by Sven Eckelmann (8 patches) - Move CRC16 dependency, by Sven Eckelmann - Enable MCAST by default, by Linus Luessing ==================== Signed-off-by:
-
- 15 Nov, 2018 13 commits
-
-
Li RongQing authored
netperf udp stream shows that eth_type_trans takes certain cpu, so adjust the mac address check order, and firstly check if it is device address, and only check if it is multicast address only if not the device address. After this change: To unicast, and skb dst mac is device mac, this is most of time reduce a comparision To unicast, and skb dst mac is not device mac, nothing change To multicast, increase a comparision Before: 1.03% [kernel] [k] eth_type_trans After: 0.78% [kernel] [k] eth_type_trans Signed-off-by:
Zhang Yu <zhangyu31@baidu.com> Signed-off-by:
Li RongQing <lirongqing@baidu.com> Signed-off-by:
-
Li RongQing authored
if list is NULL pointer, and the following access of list will trigger panic, which is same as BUG_ON Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queueDavid S. Miller authored
Jeff Kirsher says: ==================== 40GbE Intel Wired LAN Driver Updates 2018-11-14 This series contains updates to i40e and virtchnl. Lance Roy updates i40e to use lockdep_assert_held() instead of spin_is_locked(), since it is better suited to check locking requirements. Jan improves the code readability in XDP by adding the use of a local variable. Provides protection on methods that create/modify/destroy VF's via locking mechanism to prevent unstable behaviour and potential kernel panics. Krzysztof adds a hardware capability flag to indicate whether firmware supports stopping the LLDP agent. Patryk replaces the use of strncpy() with strlcpy() to ensure the buffer is NULL terminated. Mitch fixes the issue of trying to start nway on devices that do not support auto-negotiation, by checking the autoneg state before attempting to restart nway. Alice updates virtchnl to keep the checks all together for ease of readability and consistency. Also fixed a "off by one" error in the number of traffic classes being calculated. Richard fixed VF port VLANs, where the priority bits were incorrectly set because the incorrect shift and mask bits were being used. Alan adds a bit to set and check if a timeout recovery is already pending to prevent overlapping transmit timeout recovery. ==================== Signed-off-by: -
David S. Miller authored
lib/test_objagg.c: In function ‘test_delta_action_item’: ./include/linux/printk.h:308:2: warning: ‘errmsg’ may be used uninitialized in this function [-Wmaybe-uninitialized] Signed-off-by: -
David S. Miller authored
Ido Schimmel says: ==================== mlxsw: spectrum: acl: Introduce ERP sharing by multiple masks Jiri says: The Spectrum-2 hardware has limitation number of ERPs per-region. In order to accommodate more masks than number of ERPs, the hardware supports to insert rules with delta bits. By that, the rules with masks that differ in up-to 8 consecutive bits can share the same ERP. Patches 1 and 2 fix couple of issues that would appear in existing selftests after adding delta support Patch 3 introduces a generic object aggregation library. Now it is static, but it will get extended for recalculation of aggregations in the future in order to reach more optimal aggregation. Patch 4 just simply converts existing ERP code to use the objagg library instead of a rhashtable. Patches 5-9 do more or less small changes to prepare ground for the last patch. Patch 10 fills-up delta callbacks of objagg library and utilizes the delta bits for rule insertion. The last patch adds selftest to test the mlxsw Spectrum-2 delta flows. ==================== Signed-off-by: -
Jiri Pirko authored
Track the basic codepaths of delta handling, using objagg tracepoints. Signed-off-by:
Ido Schimmel <idosch@mellanox.com> Signed-off-by:
-
Jiri Pirko authored
Allow ERP sharing for multiple mask. Do it by properly implementing delta_create() objagg object. Use the computed delta info for inserting rules in A-TCAM. Signed-off-by:
-
Jiri Pirko authored
Later on the same code is going to be needed for deltas as well. So push the procedures related to increment and decrement of num_ctcam_erps into a separate helpers. Signed-off-by:
-
Jiri Pirko authored
Since two remaining users of mlxsw_afk_encode() do not specify block ranges to work on, remove the args. Also, key/mask is always non-NULL now, so skip the checks. Signed-off-by:
-
Jiri Pirko authored
No need to do key encoding again in mlxsw_sp_acl_atcam_12kb_lkey_id_get(). Instead of that, introduce a new helper that would just clear unused blocks. Signed-off-by:
-
Jiri Pirko authored
Change order so it is aligned with the usual case where the "write_to" buffer comes as the first arg. Signed-off-by:
-
Jiri Pirko authored
The device requires that the master mask of each region will be composed from a logical OR between all the unmasked bits in the region. Currently, this is just a logical OR between all the eRPs used in the region, but the next patch is going to introduce delta bits support which need to be taken into account as well. Since the eRP does not include the delta bits, pass the key pointer to mlxsw_sp_acl_erp_master_mask_set/clear instead. Convert key->mask to the bitmap on fly. Signed-off-by:
-
Jiri Pirko authored
Currently the ERPs are tracked internally in a hashtable. Benefit from the newly introduced objagg library and use it to track ERPs. At this point, there is no nesting of objects done, as the delta_create callback always returns -EOPNOTSUPP. On the way, add "mask" into ERP mask get and set functions and struct names. Signed-off-by:
-
