- 24 Feb, 2016 40 commits
-
-
Corinna Vinschen authored
Problem: When switching off VLAN offloading on an i350, the VLAN interface gets unusable. For testing, set up a VLAN on an i350 and some remote machine, e.g.: $ ip link add link eth0 name eth0.42 type vlan id 42 $ ip addr add 192.168.42.1/24 dev eth0.42 $ ip link set dev eth0.42 up Offloading is switched on by default: $ ethtool -k eth0 | grep vlan-offload rx-vlan-offload: on tx-vlan-offload: on $ ping -c 3 -I eth0.42 192.168.42.2 [...works as usual...] Now switch off VLAN offloading and try again: $ ethtool -K eth0 rxvlan off Actual changes: rx-vlan-offload: off tx-vlan-offload: off [requested on] $ ping -c 3 -I eth0.42 192.168.42.2 PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da ta. --- 192.168.42.2 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms I can only reproduce it on an i350, the above works fine on a 82580. While inspecting the igb source, I came across the code in igb_set_vmolr which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and for all, and in all of the igb code there's no other place where the STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb unconditionally, independently of the offloading setting. I compared that to the latest Intel igb-5.3.3.5 driver from http://sourceforge.net/projects/e1000/ which in fact sets and clears the STRVLAN flag independently from igb_set_vmolr in its own function igb_set_vf_vlan_strip, depending on the vlan settings. So I included the STRVLAN handling from the igb-5.3.3.5 driver into our current igb driver and tested the above scenario again. This time ping still works after switching off VLAN offloading. Tested on i350, with and without addtional VFs, as well as on 82580 successfully. Signed-off-by:
Corinna Vinschen <vinschen@redhat.com> Tested-by:
Aaron Brown <aaron.f.brown@intel.com> Signed-off-by:
Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Jon Maxwell authored
A similar issue was addressed a few years ago in the following thread: http://www.spinics.net/lists/netdev/msg245877.html At that time there were concerns that removing this statement may cause other side effects. However the submitter addressed those concerns. But the dialogue went cold. We have a new case where a customers application is registering and un-registering multicast addresses every few seconds. This is leading to many "Link is Up" messages in the logs as a result of the "netif_carrier_off(netdev)" statement called by igbvf_msix_other(). Also on some kernels it is interfering with the bonding driver causing it to failover and subsequently affecting connectivity. The Sourgeforge driver does not make this call and is therefore not affected. If there were any side effects I would expect that driver to also be affected. I have tested re-loading the igbvf driver and downing the adapter with the PF entity on the host where the VM has this patch. When I bring it back up again connectivity is restored as expected. Therefore I request that this patch gets submitted. Signed-off-by:
Jon Maxwell <jmaxwell37@gmail.com> Tested-by:
-
Alexander Duyck authored
This patch adds support for generic Tx checksums to the igbvf driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. In the case of the VF drivers this meant adding support for SCTP CRCs, and inner checksum offloads for MPLS and various tunnel types. Signed-off-by:
Alexander Duyck <aduyck@mirantis.com> Tested-by:
-
Alexander Duyck authored
This patch adds support for generic Tx checksums to the igb driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. Signed-off-by:
-
Todd Fujinaka authored
E1000_MRQC_ENABLE_RSS_4Q enables 4 and 8 queues depending on the part so rename to be generic. Similarly, E1000_MRQC_ENABLE_VMDQ_RSS_2Q has no numeric meaning so rename to be more generic. Signed-off-by:
Todd Fujinaka <todd.fujinaka@intel.com> Tested-by:
-
Roland Hii authored
In general case the maximum supported half cycle time of the synchronized output clock is 70msec. Slower half cycle time than 70msec can be programmed also as long as the output clock is synchronized to whole seconds, useful specifically for generating a 1Hz clock. Permitted values for the clock half cycle time are: 125,000,000 decimal, 250,000,000 decimal and 500,000,000 decimal (equals to 125msec, 250msec and 500msec respectively). Before this patch, only the half cycle time of less than or equal to 70msec uses the I210 clock output function. This patch adds additional conditions when half cycle time is equal to 125msec or 250msec or 500msec to use clock output function. Under other conditions, interrupt driven target time output events method is still used to generate the desired clock output. Signed-off-by:
Roland Hii <roland.king.guan.hii@intel.com> Tested-by:
-
Todd Fujinaka authored
Override EEPROM settings for specific OEM devices. Signed-off-by:
-
Julia Lawall authored
This e1000_phy_operations structure is never modified, so declare it as const. Other structures of this type are already const. Done with the help of Coccinelle. Signed-off-by:
Julia Lawall <Julia.Lawall@lip6.fr> Tested-by:
-
Takuma Ueba authored
I210 device IPv6 autoconf test sometimes fails, because DAD NS for link-local is not transmitted. This packet is silently dropped. This problem is seen only GbE environment. igb_watchdog_task link up detection continues to the following process. The following cases are observed: 1.PHY 1000BASE-T Status Register Remote receiver status bit is NG. (NG status becomes OK after about 200 - 700ms) 2.In this case, the transfer packet is silently dropped. 1000BASE-T Status register [Expected]: 0x3800 or 0x7800 [problem occurred]: 0x2800 or 0x6800 Frequency of occurrence: approx 1/10 - 1/40 observed In order to avoid this problem, wait until 1000BASE-T Status register "Remote receiver status OK" After applying this patch, at least 400 runs succeed with no problems. Signed-off-by:
Takuma Ueba <t.ueba11@gmail.com> Tested-by:
-
Raanan Avargil authored
i219 (4) and i219 (5) are the next LOM generations that will be available on the next Intel platform (KabeLake). This patch provides the initial support for the devices. Signed-off-by:
Raanan Avargil <raanan.avargil@intel.com> Tested-by:
-
Raanan Avargil authored
There have been bugs caused by HW ULP configuration settings not being properly cleared after cable connect in V-Pro capable systems. This caused HW to get out of sync occasionally. The fix ensures that ULP settings are cleared in HW after LAN cable re-connect. Signed-off-by:
-
Raanan Avargil authored
Based on feedback from HW team, the configured value of the internal PHY HW FIFO pointer gap was incorrect for non-gig speeds. This patch provides the correct configuration. Signed-off-by:
-
Raanan Avargil authored
Several packet loss issues were reported for which the root cause for them was an incorrect configuration of internal HW PHY clock gating mechanism by SW. This patch provides the correct mechanism. Signed-off-by:
-
Raanan Avargil authored
Due to system level changes introduced in Skylake, ULP exit takes significantly longer to occur. Therefore, driver must wait longer for. Signed-off-by:
-
David S. Miller authored
Claudiu Manoil says: ==================== gianfar: xmit() improvements Remove redundant operations, improve code locality and maintainability. Thanks. V2: Updated first patch to not touch existing wmb(). ==================== Signed-off-by:David S. Miller <davem@davemloft.net>
-
Claudiu Manoil authored
Timestamp BD status updates that can be merged into the same "do_tstamp" block, no need for extra save/restore to the BD area. The code is more readable too. Signed-off-by:
Claudiu Manoil <claudiu.manoil@freescale.com> Signed-off-by:
-
Claudiu Manoil authored
Signed-off-by:
-
Claudiu Manoil authored
Move the mapping of the head BD before the mapping of fragments. The TOE (h/w offload) decision logic block can be also moved up (as the TOE flag belongs to the head BD), resulting in more localized code (TOE logic vs BD mapping code blocks). Note that, for this h/w, the R (status) bit for the head BD of a S/G frame needs to be written last for a reliable transmission. For the fragmented skb case, a local variable is used to temporarily store the status info of the first BD, replacing a BD status read. A merge of 2 "if(do_tstamp)" blocks was also possible. Signed-off-by:
-
Rafał Miłecki authored
It needs very similar workarounds to the one on BCM4707. It was tested on D-Link DIR-885L home router. Signed-off-by:
Rafał Miłecki <zajec5@gmail.com> Signed-off-by:
-
David S. Miller authored
Ajit Khaparde says: ==================== be2net patches Please consider applying to net-next ==================== Signed-off-by: -
ajit.khaparde@broadcom.com authored
In QnQ configurations like Flex-10 where the VLANs are inserted by the ASIC, on rare occasions the HW is encountering a scenario where the final frame length ends to be greater than what the ASIC can support. This is because when the TXULP pulls the TX WRB to check the length of the frame to be transmitted it also adds the size of VLANs to be inserted by the HW to the length of the frame indicated in the WRB, which in some cases fails the range check. This causes a UE. Avoid this by trimming the skb length to accommodate the VLAN insertion. Signed-off-by:
Ajit Khaparde <ajit.khaparde@broadcom.com> Signed-off-by:
-
ajit.khaparde@broadcom.com authored
When 16-bit integers are loaded on CPUs with high order native register sizes, the CPU could use some extra ops before using them. And currently some of the frequently used fields in the driver like the producer and consumer indices of the queues are declared as u16. This patch declares such fields as u32. With this change we see the 64-byte packets per second numbers improve by about 4%. Signed-off-by:
-
David S. Miller authored
Alexander Duyck says: ==================== Flow dissector fixes and improvements This patch series is meant to fix and/or improve a number of items within the flow dissector code. The main change out of all of this is that IPv4 and IPv6 fragmentation should now be handled better than it was. As a result we should see an improvement when handling things like IP fragment reassembly as the skbs should now only have header data in the linear portion of the buffer while the fragments will only hold payload data. ==================== Signed-off-by: -
Alexander Duyck authored
We want to try and pull the L4 header in if it is available in the first fragment. As such add the flag to indicate we want to pull the headers on the first fragment in. Signed-off-by:
Tom Herbert <tom@herbertland.com> Signed-off-by:
-
Alexander Duyck authored
The IPv6 parsing was using a local pointer when it could use the same pointer as the IPv4 portion of the code since the key_addrs can support both IPv4 and IPv6 as it is just a pointer. Signed-off-by:
-
Alexander Duyck authored
The flow dissector bits handling FCoE didn't bother to actually validate that the space there was enough for the FCoE header. So we need to update things so that if there is room we add the header and report a good result, otherwise we do not add the header, and report the bad result. Signed-off-by:
-
Alexander Duyck authored
It turns out that for IPv4 we were reporting the ip_proto of the fragment, and for IPv6 we were not. This patch updates that behavior so that we always report the IP protocol of the fragment. In addition it takes the steps of updating the payload offset code so that we will determine the start of the payload not including the L4 header for any fragment after the first. Signed-off-by:
-
Alexander Duyck authored
This patch corrects the logic for the IPv4 parsing so that it is consistent with how we handle IPv6. Specifically if we do not have the flow key indicating we want the addresses we still may need to take a look at the IP fragmentation bits and to see if we should stop after we have recognized the L3 header. Fixes: 807e165d ("flow_dissector: Add control/reporting of fragmentation") Signed-off-by:
-
David S. Miller authored
Saeed Mahameed says: ==================== QoS and VxLAN offloads support for Mellanox 100G mlx5 driver This patch series introduces QoS IEEE dcbnl support for PFC, ETS and max rate. In addition we added VxLAN support and introduced a patch that modifies the driver to report checksum complete in RX path for all IP (tunneled and non-tunneled) traffic which is non HW LRO. This series is applied on top of the latest mlx5_ifc and NDO fixes we sent to the net tree: net/mlx5e: Use static constant netdevice ndos net/mlx5e: Remove select queue ndo initialization net/mlx5: Use offset based reserved field names in the IFC header file The QoS patches depend on the IFC change since they expose new fields in the driver/firmware API. Both QoS and VxLAN patches depend on the NDO changes, since they add new ndo entries. Changes from V1: - Fixed the S.O.B from "Matt" to "Matthew" to be aligned with the committer title. - Don't populate VxLAN/dcbnl ndos for virtual functions. - Addressed John comment on mlx5_setup_tc to be aligned with latest API changes. - Added device ETS capability check prior query/modify ets configuration. - Call mlx5e_dcbnl_ieee_setets_core at the end of mlx5e_create_netdev and don't fail netdev creation in case it failed or ETS was not supported. The series where applied on top of: ("5270c4da Merge branch 'vxlan-cleanups'") + latest mlx5 ifc and ndo fixes from net tree. ==================== Signed-off-by: -
Matthew Finlay authored
Add TSO and TX checksum counters for tunneled, inner packets Signed-off-by:
Matthew Finlay <matt@mellanox.com> Signed-off-by:
Saeed Mahameed <saeedm@mellanox.com> Signed-off-by:
-
Matthew Finlay authored
Add support for TSO and TX checksum when using hw assisted, tunneled offloads. Signed-off-by:
-
Matthew Finlay authored
If a VXLAN udp dport is added to device it will: - Configure the hardware to offload the port (up to the max supported). - Advertise NETIF_F_GSO_UDP_TUNNEL and supported hw_enc_features. Signed-off-by: -
Matthew Finlay authored
add ifndef to en.h. needed for upcoming vxlan patchset. Signed-off-by:
-
Matthew Finlay authored
Use checksum complete for all IP packets, unless they are HW LRO, in which case, use checksum unnecessary. Signed-off-by:
-
Tariq Toukan authored
Implement set/get WOL by ethtool and added the needed device commands and structures to mlx5_ifc. Signed-off-by:
Tariq Toukan <tariqt@mellanox.com> Signed-off-by:
Rana Shahout <ranas@mellanox.com> Signed-off-by:
-
Tariq Toukan authored
Add support for DCBNL IEEE get/set max rate. Signed-off-by:
-
Achiad Shochat authored
Implement the set/get DCBNL IEEE PFC callbacks. Signed-off-by:
Achiad Shochat <achiad@mellanox.com> Signed-off-by:
-
Saeed Mahameed authored
Support the ndo_setup_tc callback and the needed methods for multi TC/UP support, and removed the default_vlan_prio from mlx5e_priv which is always 0, it was replaced with hardcoded "0" in the new select queue method. For that we now create MAX_NUM_TC num of TISs (one per prio) on netdevice creation instead of priv->params.num_tc which was always 1. So far each channel had a single TXQ, Now each channel has a TXQ per TC (Traffic Class). Added en_dcbnl.c which implements the set/get DCBNL IEEE ETS, set/get dcbx and registers the mlx5e dcbnl ops. We still use the kernel's default TXQ selection method to select the channel to transmit through but now we use our own method to select the TXQ inside the channel based on VLAN priority. In mlx5, as opposed to mlx4, tc group N gets lower priority than tc group N+1. CC: John Fastabend <john.fastabend@gmail.com> Signed-off-by:
-
Saeed Mahameed authored
Add access functions to set and query a physical port TC groups and prio parameters. Signed-off-by:
-
Achiad Shochat authored
Add access functions to set and query a physical port PFC (Priority Flow Control) parameters. Signed-off-by:
-
