-
Notifications
You must be signed in to change notification settings - Fork 171
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
R4.1 #101
base: master
Are you sure you want to change the base?
R4.1 #101
Commits on Sep 25, 2017
-
Adding R4.1 branch to CI Closes-Bug: 1719246
Change-Id: I12ffeccc1dec08705ac16a6a5bd1f998920b9b46
Configuration menu - View commit details
-
Copy full SHA for ed7d4d0 - Browse repository at this point
Copy the full SHA ed7d4d0View commit details
Commits on Sep 26, 2017
-
Performance issue in 4.0.1 dpdk based vrouter
Partial-bug: #1696209 DPDK 17.02 has changed the RTE_LOG api. Need to use RTE_LOG_DP to avoid compilation of the log funciton. Change-Id: I674154764a4fa3529f28d383f334b83c71f7040a
Jeya ganesh babu J committedSep 26, 2017 Configuration menu - View commit details
-
Copy full SHA for e3f567b - Browse repository at this point
Copy the full SHA e3f567bView commit details
Commits on Oct 10, 2017
-
Dont pass function name to Macro
To convert the hash table entry to fragment entry, hash table API's like vr_find_free_hentry() are directly passed to Fragment Macro VR_FRAGMENT_FROM_HENTRY(). As the function address gets passed to the macro, function gets invoked twice leading issues like fragment never getting added to fragment table As a fix, the hash table API's return value is stored and that is passed to fragment Macro Change-Id: I9350551f6bc3a6b28eaf8acea8df2d2ec42c3eae closes-bug: #1721251
Configuration menu - View commit details
-
Copy full SHA for dfaa19e - Browse repository at this point
Copy the full SHA dfaa19eView commit details
Commits on Oct 17, 2017
-
flow setup rate perf improvement
Partial-bug: #1724326 Used tcpnodelay option for the tcp socket between agent and dpdk vrouter to avoid inconsistent delays. This was causing huge variation in hold entries. Change-Id: Ide2bcfaa28c992e31129a25b444028efce2f3c15
Jeya ganesh babu J committedOct 17, 2017 Configuration menu - View commit details
-
Copy full SHA for dfbafb2 - Browse repository at this point
Copy the full SHA dfbafb2View commit details
Commits on Oct 20, 2017
-
Subject the Vhost packets to flow processing if policy enabled on Vhost
If ACL's are configured on Vhost interface, the packets need to be subjected to flow processing. These packets would be un-tunneled packets. The Vhost route in VRF 0 points to L3 Receive Nh. This NH can not be marked with policy bit as we do not want to create Flows for outer IP fields of Tunneled packets which also get processed by L3 receive NH. Also if Policy on Vhost is disabled and if L3 receive NH is enabled with Relaxed policy, packets have to be flow processed for link local processing. For this reason, policy bit on Receive NH's outgoing interface is verified to decide whether to invoke flow processing for packets Change-Id: I86b8ea2ecdf1460bb0d3599675d84ad46dfcf1e8 closes-bug: #1711045
Configuration menu - View commit details
-
Copy full SHA for 91f7255 - Browse repository at this point
Copy the full SHA 91f7255View commit details -
Configuration menu - View commit details
-
Copy full SHA for 92a799f - Browse repository at this point
Copy the full SHA 92a799fView commit details
Commits on Nov 2, 2017
-
For Vxlan tunneled packets that are received on Fabric interface RPF callback was missing, leading to no RPF validation of the vxlan packets. Due to Tor Evpn support, Tors can be in Ecmp, which is a composite Ecmp nexthop in Vrouter. When the first packet corressponding to unique 5 tuple is received on Fabric (rather from VM) from one of the Ecmp sources of Ecmp composite nexthop, the component nexthop is pinned to the source, only if RPF callback exists. Lack of this RPF callback was making the Ecmp go wrong As a fix, Vxlan RPF call back is provided Change-Id: If53a0bc76398cfc8c176a3a94a0aede8b26262b4 closes-bug: #1724681
Configuration menu - View commit details
-
Copy full SHA for b703397 - Browse repository at this point
Copy the full SHA b703397View commit details
Commits on Nov 7, 2017
-
Update the flow drop reason list in flow utility
Change-Id: I2135446723dddf526babf1f033dd1492e22e9cf3 closes-bug: #1729818
Configuration menu - View commit details
-
Copy full SHA for ecd1b01 - Browse repository at this point
Copy the full SHA ecd1b01View commit details
Commits on Nov 9, 2017
-
Huge pages support in Vrouter kernel module
Vrouter currently allocates contiguous memory for Flow and Bridge tables and this forces the module to be inserted into kernel at the boot up time as huge memory is likely to be presetnt at the boot of the sytem. This requirement is taken out with the huge pages support. Provisioning: If the huge page support is intended in the system, provisioning of the compute node takes care of enabling the required number of huge pages in the system. After enabling the huge pages successfully, the hugetable file system is mounted at a specific location and creates the required files (2 files as of now - vrouter_mem1 and vrouter_mem2). The location of these files is made avaialble to Agent. Agent changes: Agent looks for the huge files at the specific path provided by provisioning (through agent configuration file) and mmap()s the 1G size of data. If succeeds, Agent shares this virtual memory address with Vrouter kernel module through netlink socket. If the huge files do not exist, or if the mmap fails, still it communicates the failure of this to Vrouter kernel module. This messaging is done in blockig way, so that no other configuration is added to Vrouter till this is communicated. Along with 1G pages, Agent attempts to allocate same number of 2Mb pages, which Vrouter internally uses. Even these page addresses are sent to Vrouter kernel module using same sandesh message Vrouter changes: 1) The module_init routine is changed in such a way that, the modules like fib and flow which require huge contiguous amount of memory do not allocate any memory in their init routines. 2) A new call back "mem" is introduced which allocates the required memory and is invoked after succesfully initialising the hugepages when Agent communicates it. 3) The modules like interface etc are initialised as usual so that addition of fabric and Vhost interfaces are created normally and pushed to cross connect mode till the other configuration is available 4) When Agent communicates the virtual memory address, Vrouter pins these pages using get_user_pages(). The "mem" callback is invoked which allocates the memory either from huge pages or using regular btable. The memory required to hold this huge page in terms of 4Kb pages, is another huge page of size 2Mb. This 2Mb huge memory is also sent by Agent. If no such huge page of size 2Mb is sent by Agent, Vroter allocates the memory using regular malloc. 5) As the momeory allocate from the huge pages is intended for the entire life time of Vrouter module free() routinues are not provided for these memory segments. These pages are freed from the system when the module is removed from the kernel. 6) If agent restarts, memeory is not reinitilised again and proper return codes are shared with Agent Change-Id: I942ed42463d7a0aba7bf857df7c62d9c95b83056 closes-bug: #1728925
Configuration menu - View commit details
-
Copy full SHA for ef0f526 - Browse repository at this point
Copy the full SHA ef0f526View commit details
Commits on Nov 10, 2017
-
Do not return error if setting MTU fails on the ethdev
Fortville NICs do not allow setting the MTU while the NIC is running. Change-Id: I4280715db2f3085041aa7bb5796665a65409effd Closes-Bug: 1729742
Configuration menu - View commit details
-
Copy full SHA for 7884d50 - Browse repository at this point
Copy the full SHA 7884d50View commit details
Commits on Nov 13, 2017
-
Configuration menu - View commit details
-
Copy full SHA for 7fc26af - Browse repository at this point
Copy the full SHA 7fc26afView commit details
Commits on Nov 14, 2017
-
Disable flow processing for metadata subnet on Vhost0
Currently the healthcheck is broken if policy is enabled on Vhost interface. Health check packets are destined to metadata IPs. The metadata routes point to VMI's in Vhost VRF. Regular VM's routes are present in VN's VRF. If VM is detected unreachable, HC withdraws routes in VRF corresponding to VM, but metadata Routes are not withdrawn. HC packets should be routed using these metadata routes, if routes are withdrawn. If Vhost has policy enabled, flow processing happens earlier than route look up and destination metadata IP's are Nated to VM's IP. These Nated IP's would be looked up in VN's VRF which results in drop nexthop as routes are withdrawn. The solutino is not to Nat the packet till the route lookup is complete for these metadata IP's if the policy is enabled. The required flow processing would be completed if the nexthop is marked Policy enabled. Change-Id: I144c36faf39b062026316a067e912eed5a2fa792 closes-bug: #1724945
Configuration menu - View commit details
-
Copy full SHA for 29edfac - Browse repository at this point
Copy the full SHA 29edfacView commit details -
Configuration menu - View commit details
-
Copy full SHA for 4e96a15 - Browse repository at this point
Copy the full SHA 4e96a15View commit details
Commits on Nov 16, 2017
-
Calculate Ecmp component NH for non-IP packets
Currently the Ecmp component NH is calculated for IP/IP6 packets only as Ecmp has been supported for only for L3 packets. With the support of L2 Ecmp, the component NH needs to be chosen even for L2 packets. As there is no Flow available for L2 packets, the hash is calculated on Ether dst, src and VRF Change-Id: I61085c729da9633630088604c1a6b8db5897bba8 closes-bug: #1732285
Configuration menu - View commit details
-
Copy full SHA for d9000c1 - Browse repository at this point
Copy the full SHA d9000c1View commit details -
Dont mirror Vlan tag in Vmware environment
Currently, if there is a sub interface created with a Vlan Tag on any parent interface and if mirroring is enabled on subinterface, the Vlan Tag of the subinterface is also carried in the mirrored packet. Incase of Vmware compute, all the VMIs are configred in Vrouter as subinterfaces with Pvlan. And if the mirroring is enabled on these VMI's, as the VMI's are subinterfaces, even these vlans are carried in the mirroring. But the expectation is not to carry these Vlan tags as these Vlan tags are Contrail specific and not seen by the user. As a fix, there is a new VIF flag introduced "VIF_FLAG_MIRROR_NOTAG" which is set by Agent on Vxmware VMI's only. If these flags is seen on VIF, Vrouter discards the Vlan Tag before mirroring the packets Change-Id: I9bd8c33c735159937f4b325a6ee67540f4f15f39 closes-bug: #1711459
Configuration menu - View commit details
-
Copy full SHA for bd69b17 - Browse repository at this point
Copy the full SHA bd69b17View commit details
Commits on Nov 20, 2017
-
Sandesh change for multicast VRF in interface req
This is sandesh change for using different multicast VRF in terface req for VN with provider network Change-Id: I0c0310232b5e44eec780ad54d552e6750d42b9a6 partial-bug: #1728545
Configuration menu - View commit details
-
Copy full SHA for 2bcb418 - Browse repository at this point
Copy the full SHA 2bcb418View commit details -
User different VRF for multicast packets of VN with Provider network
Currently if a VN is configured with Provider network, the VRF of the VMI's belonging to that VN are seen as Fabric VRF in Vrouter. But the Multicast/Broadcast tree is built in the VRF corresponding to VN. When BUM traffic is received on such VMI, traffic needs to be replicated as per the mutlicast tree of VN's VRF rather on Fabric VRF. To achieve this, the VN's VRF is added to VMI as different VRF and the bridge lookup is done for this traffic in this new VRF. Change-Id: I38721e27cebe80c7f4d14937588ba5b6c180b112 closes-bug: #1728545
Configuration menu - View commit details
-
Copy full SHA for 4c904a5 - Browse repository at this point
Copy the full SHA 4c904a5View commit details
Commits on Nov 21, 2017
-
Clear sender_cpu of skb before transmit
With the introduction of Xps (Xmit packet steering), the sender_cpu needs to be cleared before dev_queue_transmit(). Not doing this results in accessing netdev's xps map at a wrong location resulting in a crash. Clearing this makes the hard xmit calculate the sener_cpu Change-Id: Ifb0757ffdaa3e27b15261a9944281d9224afa8f0 closes-bug: #1733431
Configuration menu - View commit details
-
Copy full SHA for d383e85 - Browse repository at this point
Copy the full SHA d383e85View commit details
Commits on Nov 22, 2017
-
Configuration menu - View commit details
-
Copy full SHA for ff48329 - Browse repository at this point
Copy the full SHA ff48329View commit details -
Remove Ecmp checks in ARP processing
When L2 Ecmp is not in place, packets to/from Ecmp source are forced to L3 packets by replying ARP requests using Vrouter's and Vhost's MAc. This is no longer required once we have L2 Ecmp in place. So all ARP processing which specially deals with Ecmp source side is not reqired any more. This processing is removed from the current ARP handling Change-Id: I484da69b3c173891b915fb86b81c2e57d711b18a closes-bug: #1733811
Configuration menu - View commit details
-
Copy full SHA for ac9ebb8 - Browse repository at this point
Copy the full SHA ac9ebb8View commit details
Commits on Nov 27, 2017
-
Start Eviction even if the FIN/FIN-ACK packets are dropped
Incase of TCP a session to VRRP address, if VRRP mastership changes, flow's RPF nexthop changes as RPF nexthop needs to start pointing to new Master. But the TCP other end might send a FIN/FIN-ACK to old VRRP master with the intention of tearing down the session. Due to RPF failure these packets get dropped. Before dropping the packets itself Flow getis marked with the required flags. Incase of last FIN-Ack packet being dropped, the Flow gets marked as Dead but eviction does not get kicked-off as eviction is not invoked for a packet that is dropped. Due to this, the flow never gets evicted and Agent also might not delete this flow as part of Aging, if these flows are for BGPaaS. To fix this, eviction is kicked-off, if the Flow is marked Dead and even if the packet is dropped. Change-Id: Ib4a527477403f40bc3016c7ea58813f168a81e02 closes-bug: #1733608
Configuration menu - View commit details
-
Copy full SHA for 6b62181 - Browse repository at this point
Copy the full SHA 6b62181View commit details
Commits on Nov 28, 2017
-
Configuration menu - View commit details
-
Copy full SHA for 760c15a - Browse repository at this point
Copy the full SHA 760c15aView commit details
Commits on Dec 15, 2017
-
rt --monitor can be used to get route creations and deletions in live. They are broadcasted on Netlink by the vrouter kernel module. Only routes of type AF_INET and AF_INET6 are fully printed while routes of type AF_BRIDGE are not generated yet. The output format is 'jsonline' to be easily parsed by external tools. This is an example of the output: {"operation":"add","family":"AF_INET","vrf_id":2,"prefix":32,"address":"20.1.1.254","nh":12,"flags":{"label_valid":false, "arp_proxy":true, "arp_trap":true, "arp_flood":false}} Change-Id: I8cb8556f1c4adda0bb0ef10f98fc38702b2942c4 Closes-bug: #1650316 (cherry picked from commit 12b9b49)
Antoine Eiche committedDec 15, 2017 Configuration menu - View commit details
-
Copy full SHA for 677dbab - Browse repository at this point
Copy the full SHA 677dbabView commit details
Commits on Feb 2, 2018
-
Do not calculate fragment values for mirrored packets
If fragmentes are being processed, fragment assembler maintains the statistics of framents in the fragment entry to decide when to delete the fragment. When mirroring is enabled, the mirrored packet gets processes earlier than real packet and these fragment calculations are invoked on the mirrored packet leading to fragment entry being deleted if the mirrored packet is the last fragment of the packet. Once the fragment entry is deleted and mirrored packet is out of system, the real packet gets processes and this packet will not have a matching fragment entry to do the flow processing resulting in packet being discarded. As a fix, the fragment calculations are invoked only when the real packets are processed rather for mirrored packets. Change-Id: I30b3092622f8c661c6bebaf5ccbff6c9621cc3dc closes-bug: #1739602 (cherry picked from commit c9bf938)
Configuration menu - View commit details
-
Copy full SHA for 0dad531 - Browse repository at this point
Copy the full SHA 0dad531View commit details
Commits on Feb 6, 2018
-
Update vm port status with dpdk
Closes-bug: #1734994 Send netlink updates to agent for the VM's port in dpdk mode of operation when VM state changes. Change-Id: I9bc3cf8da01ed97ea2409ea2c16239447a867924
Jeya ganesh babu J committedFeb 6, 2018 Configuration menu - View commit details
-
Copy full SHA for be48b61 - Browse repository at this point
Copy the full SHA be48b61View commit details
Commits on Feb 8, 2018
-
VIF_FLAG_GRO_NEEDED and VIF_FLAG_ETREE_ROOT flags were conflicting Change-Id: Iefa51d3561fc208690fef9b1e6942167a3d925a9 Closes-Bug: 1748261
Configuration menu - View commit details
-
Copy full SHA for 5d6ff72 - Browse repository at this point
Copy the full SHA 5d6ff72View commit details
Commits on Feb 10, 2018
-
Configuration menu - View commit details
-
Copy full SHA for 37b1392 - Browse repository at this point
Copy the full SHA 37b1392View commit details
Commits on Feb 15, 2018
-
Modified the flow utility commands to directly read the hold entries …
…from the vrouter, instead of reading each flow entries and counting hold entries based on its status. Now with this change flow -s and flow -r will directly read the hold entries from the vrouter. Change-Id: I2d873583ac9668d9bc38c2305d86f3256ab29d94 Closes-Bug: 1738282
Configuration menu - View commit details
-
Copy full SHA for de6f415 - Browse repository at this point
Copy the full SHA de6f415View commit details
Commits on Feb 19, 2018
-
Merge "Modified the flow utility commands to directly read the hold e…
…ntries from the vrouter, instead of reading each flow entries and counting hold entries based on its status. Now with this change flow -s and flow -r will directly read the hold entries from the vrouter." into R4.1
Configuration menu - View commit details
-
Copy full SHA for 928797f - Browse repository at this point
Copy the full SHA 928797fView commit details
Commits on Feb 21, 2018
-
Closes-bug: #1750711 When the dpdk vrouter tries to forward the traffic after the vif gets deleted, the vif data structure is access to update the statistics. This is causing the vrouter to crash. Fixed by removing the stats update after the vif is deleted. Change-Id: Id3dcf31a8cdf98d6a6cc92ff552963f490476c68
Jeya ganesh babu J committedFeb 21, 2018 Configuration menu - View commit details
-
Copy full SHA for 290f381 - Browse repository at this point
Copy the full SHA 290f381View commit details -
DPDK: Fix for KNI crash in pkt_drop_stats()
Make the vif argument as NULL in pkt_drop_stats() in KNI send_burst(). There may be cases when vif is deleted after the packet is sent to KNI and when such packet hits this code, there may be a crash since the function tries to access the deleted vif. Change-Id: If180909a1b9828b97310329feebd32371e965c11 Closes-Bug: 1714156 (cherry picked from commit b41d869)
Configuration menu - View commit details
-
Copy full SHA for f04f6f7 - Browse repository at this point
Copy the full SHA f04f6f7View commit details
Commits on Feb 22, 2018
-
Closes-bug: #1734994 Fixed crash related to the port status update change. The reset status was called with an uninitialized variable. Moved the reset status to the correct place. Change-Id: I8bb2eb22a860429c02857597542f713705aafa20
Jeya ganesh babu J committedFeb 22, 2018 Configuration menu - View commit details
-
Copy full SHA for 10488c8 - Browse repository at this point
Copy the full SHA 10488c8View commit details -
Fix memory leak in the vrouter module
When calling bridge_table_dump with a request containing a wrong MAC length, the dumper structure is not deallocated on the error case. Moving MAC size check before dumper initialization. Change-Id: I432819d7bcf729b4b17df280440547c3bf87a529 Closes-Bug: #1736712 (cherry picked from commit c15068f)
Configuration menu - View commit details
-
Copy full SHA for b8ea077 - Browse repository at this point
Copy the full SHA b8ea077View commit details -
Configuration menu - View commit details
-
Copy full SHA for 1a8fd92 - Browse repository at this point
Copy the full SHA 1a8fd92View commit details -
Configuration menu - View commit details
-
Copy full SHA for 0ef5287 - Browse repository at this point
Copy the full SHA 0ef5287View commit details -
Configuration menu - View commit details
-
Copy full SHA for 554e426 - Browse repository at this point
Copy the full SHA 554e426View commit details
Commits on Mar 7, 2018
-
Coverity reports bad comparison for the result of pkt_pull
The result of pkt_pull is checked to be below 0 in case of error (in nh_l2_rcv and nh_handle_mcast_control_pkt). On the other hand, pkt_pull returns a pointer, which makes the comparison pointless. Change-Id: I1481a58a84f8872f05cf7bc4a045d1e41f6181d6 Closes-Bug: #1739010 (cherry picked from commit fe9d4ad)
Configuration menu - View commit details
-
Copy full SHA for a55f075 - Browse repository at this point
Copy the full SHA a55f075View commit details -
Several coverity issues in vrouter utilities
vif.c: Initializing ret variable in vr_intf_op. udp_util.c: Using udp_free_client instead of udp_free, so that cl is freed as well. nl_util.c: Initializing attr in nl_dcb_parse_reply to DCB_ATTR_UNDEFINED. Using nl_free_client to also free cl in nl_register_client. Otherwise, fields are freed but not the structure itself. ini_parser.c: Closing file before returning from parse_ini_file if ini_data allocation fails. flow.c: In flow_set_ip, updating match_string only if token is not null and token_length is assigned to and makes sense. Initializing ret with 0 in flow_set_ip. It would be used uninitialized in the context of token being null. In flow_set_match, updating match_string only if token is not null and token_length is assigned to and makes sense. Change-Id: I04fbb636497f083099375c643d32a8997d0a4c0d Closes-Bug: #1736160 (cherry picked from commit 5fa4855)
Configuration menu - View commit details
-
Copy full SHA for 36e8bf1 - Browse repository at this point
Copy the full SHA 36e8bf1View commit details -
Some dp-core fixes for coverity issues
vr_qos.c: In vr_fc_map_get, control may reach the generate_response label with an uninitialized resp variable. Initializing resp to NULL. vr_route.c: In bridge_rtb_family_init, if algo_init fails, table is not deallocated. Freeing table on error case. vr_packet.h: fmd_queue field is not initialized (it doesnt't look used anywhere either). Removing fmd_queue from forwarding metadata structure Change-Id: Ife54d3ff95636535faa9f05ea03bb8eae7398231 Closes-Bug: #1737141
Configuration menu - View commit details
-
Copy full SHA for 8733582 - Browse repository at this point
Copy the full SHA 8733582View commit details
Commits on Mar 10, 2018
-
Disable promiscuous mode on bond interface in DPDK vrouter.
Change-Id: I8ef5d8a355b796d52d9f09a924432135ca912029 Closes-bug: 1751362
Configuration menu - View commit details
-
Copy full SHA for 704278e - Browse repository at this point
Copy the full SHA 704278eView commit details
Commits on Mar 12, 2018
-
Configuration menu - View commit details
-
Copy full SHA for 6162179 - Browse repository at this point
Copy the full SHA 6162179View commit details
Commits on Mar 14, 2018
-
Configuration menu - View commit details
-
Copy full SHA for ec22fb3 - Browse repository at this point
Copy the full SHA ec22fb3View commit details -
Configuration menu - View commit details
-
Copy full SHA for fd0e93f - Browse repository at this point
Copy the full SHA fd0e93fView commit details
Commits on Mar 20, 2018
-
Making 3.13.0-142 as the default kernel for ubuntu 14.04
Change-Id: I0d1a20adc9ec2afb7f102e8cbbe27d42f80d81ac partial-bug: #1742044
Configuration menu - View commit details
-
Copy full SHA for 208f745 - Browse repository at this point
Copy the full SHA 208f745View commit details
Commits on Mar 21, 2018
-
Configuration menu - View commit details
-
Copy full SHA for 218f027 - Browse repository at this point
Copy the full SHA 218f027View commit details
Commits on Mar 22, 2018
-
Keeping the mirror metadata entry in Flow entry
Currently mirror metadata is stored in index table using flow index as key. Before the Flow eviction, the addition and deletion of this meta data to Index table is always from Agent and hence no parallel manipulation of Index table. After eviction, the metadata deletion from index table can happen in parallel and hence leading to memory corruption/leak in index table manipulation. The right fix is making the index table SMP ready. As a stop gap fix, the mirror metada data pointer is stored in the flow entry itself directly. Conflicts: include/vr_flow.h closes-bug: #1572397, #1744369 Change-Id: I216d8b6c55025a6053b808fd2807338265b307e0 (cherry picked from commit a437faa)
Configuration menu - View commit details
-
Copy full SHA for 505bafc - Browse repository at this point
Copy the full SHA 505bafcView commit details -
Configuration menu - View commit details
-
Copy full SHA for 16facd0 - Browse repository at this point
Copy the full SHA 16facd0View commit details
Commits on Mar 26, 2018
-
Configuration menu - View commit details
-
Copy full SHA for 5505ca3 - Browse repository at this point
Copy the full SHA 5505ca3View commit details -
DPDK: Add new config parameters for tx/rx descriptors
Some NICs like XL710 need more number of tx/rx descriptors. Hence make it configurable Change-Id: I395ef2d9c6d6ce87ad624e6097b1e77c97b27848 Closes-Bug: 1744185
Configuration menu - View commit details
-
Copy full SHA for 7bf9a30 - Browse repository at this point
Copy the full SHA 7bf9a30View commit details
Commits on Apr 2, 2018
-
DPDK: Add DPDK version to the log
Add DPDK library version in the log Change-Id: I049ef94704b6d909bb2e1f6636416c1c123eb9c8 Closes-Bug: 1760125
Configuration menu - View commit details
-
Copy full SHA for d4df00b - Browse repository at this point
Copy the full SHA d4df00bView commit details
Commits on Apr 3, 2018
-
Configuration menu - View commit details
-
Copy full SHA for de5054c - Browse repository at this point
Copy the full SHA de5054cView commit details -
Dpdk performance fix for redhat
Partial-bug: #1748012 Redhat doesnt support KNI and uses tap for the vhost interface The data path code seems to check for data in all the tap interface irrespective of whether its mgmt interface or data vlan interface. Fixed by adding check in the data path and fwd only for datapath traffic. Mgmt traffic will go through a slower tapdev loop. Change-Id: Id1ef38592f3319d1894bf981d0cc4c418cfd676f
Jeya ganesh babu J committedApr 3, 2018 Configuration menu - View commit details
-
Copy full SHA for 59a2c58 - Browse repository at this point
Copy the full SHA 59a2c58View commit details -
Configuration menu - View commit details
-
Copy full SHA for 8c86774 - Browse repository at this point
Copy the full SHA 8c86774View commit details
Commits on Apr 4, 2018
-
- skb_sender_cpu_clear has been removed in kernel 4.6 [1] and is no longer required since 4.5 [2]. - get_user_page signature has changed in 4.6 [3] and in 4.9 [4]. Since I don't have a 4.6 kernel for testing, I didn't add code to support it. - The macro page_cache_release has been finally removed [5]. [1] torvalds/linux@64d4e34 [2] torvalds/linux@52bd2d6 [3] torvalds/linux@c12d2da [4] torvalds/linux@768ae30 [5] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=09cbfeaf1a5a67bfb3201e0c83c810cecb2efa5a Change-Id: I569ca66145204f6e71a1083412bf2cb4caab818d Partial-Bug: #1633387 (cherry picked from commit a0d6eb0)
Configuration menu - View commit details
-
Copy full SHA for 1dacc5c - Browse repository at this point
Copy the full SHA 1dacc5cView commit details -
Configuration menu - View commit details
-
Copy full SHA for 480fc3c - Browse repository at this point
Copy the full SHA 480fc3cView commit details
Commits on Apr 6, 2018
-
Configuration menu - View commit details
-
Copy full SHA for 1da5c7d - Browse repository at this point
Copy the full SHA 1da5c7dView commit details
Commits on Apr 18, 2018
-
Fix vrouter build for kernels from 4.11 to 4.13
- linux/vhost_dev.c: Rename netdev destructor to priv_destructor due to API change. New signature of rtnl_link_ops.validate due to API change. - linux/vr_mem.c Remove vma from mem_fault function since VMA pointer now accessable via VMF structure. Change-Id: I9ffc103e5e69649d2a691c7a378cfd12b2363320 Partial-Bug: #1633387 (cherry picked from commit 60c9031)
Configuration menu - View commit details
-
Copy full SHA for 1f45fb3 - Browse repository at this point
Copy the full SHA 1f45fb3View commit details
Commits on Apr 19, 2018
-
Closes-bug: #1764940 Issue because of some of the linux kernel not having all the patches related to the retpoline fixes. The RETPOLINE define needs to be enabled in the Makefile based on the CONFIG_RETPOLINE. Change-Id: I8026dcc16f7056329618e2721e932557d0349386
Jeya ganesh babu J committedApr 19, 2018 Configuration menu - View commit details
-
Copy full SHA for 06e029e - Browse repository at this point
Copy the full SHA 06e029eView commit details
Commits on Apr 20, 2018
-
Configuration menu - View commit details
-
Copy full SHA for f514f4d - Browse repository at this point
Copy the full SHA f514f4dView commit details
Commits on May 7, 2018
-
Enable vrouter build on RHEL 7.5
RHEL 7.5 has kernel version updated to 3.10.0-862.el7, which breaks building of vrouter module. Co-Authored-By: Raja Sivaramakrishnan <raja@juniper.net> Co-Authored-By: Krzysztof Klimonda <krzysztof.klimonda@codilime.com> Change-Id: Icb0467344c0023141be7e1dd9998d383bfaab14b Closes-Bug: 1763656 (cherry picked from commit b948f9f)
Configuration menu - View commit details
-
Copy full SHA for bf14434 - Browse repository at this point
Copy the full SHA bf14434View commit details
Commits on May 8, 2018
-
Changes to make vrouter code compile against linux kernel v4.15. Time…
…r_list structure has changed in 4.15 releases. Also some changes introduced to vr_host_interface.c, removing NETIF_F_UFO reference for v4.15. Change-Id: Iaa194a5623d0f5311b708025e23076670c2e19c1 closes-bug: #1765229
Saurabh Gupta authored and Jeya ganesh babu committedMay 8, 2018 Configuration menu - View commit details
-
Copy full SHA for 4b1b73a - Browse repository at this point
Copy the full SHA 4b1b73aView commit details
Commits on May 9, 2018
-
Fixes vrouter kernel module build on RHEL/CentOS 7
Change-Id: I3bf6a426928a65b66c56ffbca1f318641f799d4b Closes-Bug: 1736108 (cherry picked from commit c4c68b6)
Configuration menu - View commit details
-
Copy full SHA for cbda85a - Browse repository at this point
Copy the full SHA cbda85aView commit details
Commits on May 10, 2018
-
Closes-bug: #1746578 Fixed the feature negotiation issue by storing the feature in persistent store and load when VM comes up. Change-Id: I3f41079264f040e1b57a146eda7711b171e302ac (cherry picked from commit 9800537)
Jeya ganesh babu J committedMay 10, 2018 Configuration menu - View commit details
-
Copy full SHA for f442753 - Browse repository at this point
Copy the full SHA f442753View commit details -
DPDK: Check invalid GPA in descr array in virtio
Due to buggy virtio guest drivers, there may be invalid descriptor addresses in the descriptor table. Add checks not to use those. Change-Id: I48ad58a1d554e07a08ae3ed68dc0f51d8d1b028b Closes-Bug: 1726459
Jeya ganesh babu J committedMay 10, 2018 Configuration menu - View commit details
-
Copy full SHA for 126401b - Browse repository at this point
Copy the full SHA 126401bView commit details -
Merge "Changes to make vrouter code compile against linux kernel v4.1…
…5. Timer_list structure has changed in 4.15 releases. Also some changes introduced to vr_host_interface.c, removing NETIF_F_UFO reference for v4.15." into R4.1
Configuration menu - View commit details
-
Copy full SHA for bc4bd8c - Browse repository at this point
Copy the full SHA bc4bd8cView commit details
Commits on May 14, 2018
-
Configuration menu - View commit details
-
Copy full SHA for 03f396b - Browse repository at this point
Copy the full SHA 03f396bView commit details
Commits on May 16, 2018
-
Changes to make netlink infra to work with CentOS 7.5.
Change-Id: I4c0cb7f168ad285c530379f382a403b743560266 closes-bug: #1770734
Configuration menu - View commit details
-
Copy full SHA for 5d08281 - Browse repository at this point
Copy the full SHA 5d08281View commit details
Commits on May 21, 2018
-
Include hardware offload hooks
This changeset provides a generic infrastructure for configuring hardware offloads for vrouter. This defines function prototypes for callbacks which will be registered by a vrouter offload module, which will be called at certain points in the vrouter.ko processing (mainly around where it does configurations of tables via sandesh messages). Offloaded functionality includes flow, vif, MPLS, VXLAN, mirror, nexthop, layer2 and route tables. QoS API is defined but not currently supported by offload module (offload module will return an error if configured). Change-Id: I00119ec244bf0c3bc8fa218aae10dd1b164b47ce Closes-Bug: #1767107 Signed-off-by: Frik Botha <frederick.botha@netronome.com>
Configuration menu - View commit details
-
Copy full SHA for 4cc4453 - Browse repository at this point
Copy the full SHA 4cc4453View commit details -
Register information about multicast group
Pass to vrouter module information about multicast group. Withouth that user space client (vif) was unable to join vrouter netlink broadcast group when kernel 4.10 or higher was used. Change-Id: If1868b027ffcf015ad4c3c4c287260d432edd6b5 Closes-Bug: #1771999
Configuration menu - View commit details
-
Copy full SHA for 7f1e1e2 - Browse repository at this point
Copy the full SHA 7f1e1e2View commit details
Commits on May 22, 2018
-
Configuration menu - View commit details
-
Copy full SHA for 704e9f6 - Browse repository at this point
Copy the full SHA 704e9f6View commit details
Commits on May 23, 2018
-
Configuration menu - View commit details
-
Copy full SHA for 2a42b8d - Browse repository at this point
Copy the full SHA 2a42b8dView commit details
Commits on May 24, 2018
-
Configuration menu - View commit details
-
Copy full SHA for 2924b70 - Browse repository at this point
Copy the full SHA 2924b70View commit details -
Set needs_free_netdev to false because free_netdev is already executed by vhost_dev_destructor. Change-Id: Iad92cfffe25d510bb52f458fdf102e12aab28e3f Closes-bug: #1772639
Configuration menu - View commit details
-
Copy full SHA for 59c80ea - Browse repository at this point
Copy the full SHA 59c80eaView commit details -
Advertise VHOST_USER_PROTOCOL_F_LOG_SHMFD to Qemu which is needed for live migration and suspend operation of VM Change-Id: Icc82218e0fb58c24076128681323171c2040e1ea Closes-Bug: 1685409 (cherry picked from commit 95ea5f6)
Configuration menu - View commit details
-
Copy full SHA for e772220 - Browse repository at this point
Copy the full SHA e772220View commit details
Commits on May 25, 2018
-
DPDK GSO: Fix for performance issue seen in TIM
The condition whether to segment a packet or not was wrong. The correct condition is that if TSO seg size that the VM is giving in virtio header is lesser than the packet size, it needs to be segmented. Change-Id: I97cf8a23407dac08030c3df575b5a3522b06fd2e Closes-Bug: 1772648
Configuration menu - View commit details
-
Copy full SHA for 43d8849 - Browse repository at this point
Copy the full SHA 43d8849View commit details -
Configuration menu - View commit details
-
Copy full SHA for 9daf65f - Browse repository at this point
Copy the full SHA 9daf65fView commit details
Commits on May 29, 2018
-
Add defensive checks in vrouter
Due to a bad frontend driver, there may be memory corruption in vrouter when it posts buffers with 0 length. Add checks to prevent the same. Change-Id: I9bd623b414f731704c0aa1d640aec34fec098c33 Closes-Bug: 1767225
Configuration menu - View commit details
-
Copy full SHA for 5a5cec4 - Browse repository at this point
Copy the full SHA 5a5cec4View commit details -
Configuration menu - View commit details
-
Copy full SHA for aabefed - Browse repository at this point
Copy the full SHA aabefedView commit details
Commits on May 30, 2018
-
Configuration menu - View commit details
-
Copy full SHA for bb37a33 - Browse repository at this point
Copy the full SHA bb37a33View commit details -
vif interfaces are numbered , add check for
vif get to be called only when argument is a number . Change-Id: I67e1a16a97bc7a6884379fb77d4b2ddd59881992 Closes-Bug: #1752783
Configuration menu - View commit details
-
Copy full SHA for 7e758e4 - Browse repository at this point
Copy the full SHA 7e758e4View commit details -
DPDK tapdev: Don't delete vhost0 on soft reset
We are deleting vhost0 on agent soft reset in DPDK mode. The same is not the case with kernel mode. For containerization, the requirement is not to delete vhost0 on soft reset. Change-Id: I717970b2aa43e7d76561ec21c040807c4d69bd66 Closes-Bug: 1756961 (cherry picked from commit 23cc1d3)
Configuration menu - View commit details
-
Copy full SHA for b603bc1 - Browse repository at this point
Copy the full SHA b603bc1View commit details
Commits on May 31, 2018
-
DPDK: Performance issue in TSO
There was a side effect in the earlier commit that Rx direction will have an issue due to the will_segment condition. In Rx, the packet gets GROed and hits the same code path and has to evaluate will_segment as false since the GROed packet needs to goto the VM. But it was evaluating to true. Hence added an extra check. Change-Id: I6216aef7cad3b372d1aa8286e4dc5c7f723dec2e Closes-Bug: 1772648
Configuration menu - View commit details
-
Copy full SHA for b47b91a - Browse repository at this point
Copy the full SHA b47b91aView commit details -
Configuration menu - View commit details
-
Copy full SHA for 3d16ac6 - Browse repository at this point
Copy the full SHA 3d16ac6View commit details -
Configuration menu - View commit details
-
Copy full SHA for d990456 - Browse repository at this point
Copy the full SHA d990456View commit details -
Link against the NFP PMD for vRouter-DPDK
vRouter-DPDK is unable to probe Netronome NICs, since the build system does not link to the NFP PMD. Change-Id: Ia93bd22de5dcfee6ad56bdb48ba288a7e80ef4db Partial-Bug: #1774414 Signed-off-by: Frik Botha <frederick.botha@netronome.com>
Configuration menu - View commit details
-
Copy full SHA for 6b3ea62 - Browse repository at this point
Copy the full SHA 6b3ea62View commit details -
Configuration menu - View commit details
-
Copy full SHA for a489a49 - Browse repository at this point
Copy the full SHA a489a49View commit details -
Configuration menu - View commit details
-
Copy full SHA for b7b4748 - Browse repository at this point
Copy the full SHA b7b4748View commit details
Commits on Jun 6, 2018
-
Merge "vif interfaces are numbered , add check for vif get to be call…
…ed only when argument is a number ." into R4.1
Configuration menu - View commit details
-
Copy full SHA for 8d0c7eb - Browse repository at this point
Copy the full SHA 8d0c7ebView commit details
Commits on Jul 27, 2018
-
Filter expression should not start with a
separator as its a malformed request in vrouter cli. Closes-Bug: #1734896 Change-Id: Iad450df0d9196a90d2cf5bd51e41c79a9d00b47e
Configuration menu - View commit details
-
Copy full SHA for 09d5192 - Browse repository at this point
Copy the full SHA 09d5192View commit details
Commits on Sep 6, 2018
-
Fix for the bug not allowing custom MTU to be set on vhost interface on
centos kernels. max_mtu is now part of net_device->extended struct. Corrected the reference and the logic so that the initialization is not skipped. closes-bug: #1790766 Change-Id: I9db33970943050566d3fb76986b06fa5563fddc8
Saurabh Gupta committedSep 6, 2018 Configuration menu - View commit details
-
Copy full SHA for e921c5f - Browse repository at this point
Copy the full SHA e921c5fView commit details
Commits on Sep 14, 2018
-
Offload ECMP next hop index updates
vr_rflow_update_ecmp_index updates the ECMP next hop index for reverse flows. Add offload hook such that this change reflects in hardware as well. Change-Id: I62a28ae276ef59557dfdf8e8cc884255150723ff Closes-Bug: #1792378 Signed-off-by: Pieter Malan <pieter.malan@netronome.com>
Configuration menu - View commit details
-
Copy full SHA for 8a24cd6 - Browse repository at this point
Copy the full SHA 8a24cd6View commit details
Commits on Sep 19, 2018
-
Packet corruption in case of tcp packet to vlan if
Closes-bug: #1788492 The checksum code updates a wrong offset when vlan is present for the tx interface. Modified to calculate the offset based on the ethernet header type. Change-Id: I4f70359254ad2461d35c84a5612f4bf9a89325b6
Jeya ganesh babu J committedSep 19, 2018 Configuration menu - View commit details
-
Copy full SHA for fe4108b - Browse repository at this point
Copy the full SHA fe4108bView commit details
Commits on Oct 1, 2018
-
VM traffic stop after agent restart.
Closes-bug: #1794934 During an agent restart, the vif gets deleted. This causes the stored vhostuser feature set to be deleted. Delete the stored feature only if VM exits. Change-Id: I06724440905b27b1087f294f8d9b6361144611f4
Jeya ganesh babu J committedOct 1, 2018 Configuration menu - View commit details
-
Copy full SHA for a11e9fc - Browse repository at this point
Copy the full SHA a11e9fcView commit details
Commits on Oct 11, 2018
-
VM traffic stop after agent restart.
Closes-bug: #1794934 During an agent restart, the vif gets deleted. With server mode configuration of vhostuser, we should not delete the socket file in /var/run/vrouter. Change-Id: I0050ea47e91151c7f3cb2f80191aa0c2bdddec70
Jeya ganesh babu J committedOct 11, 2018 Configuration menu - View commit details
-
Copy full SHA for fad5f97 - Browse repository at this point
Copy the full SHA fad5f97View commit details
Commits on Nov 4, 2018
-
Check packet length before calculating checksum
If a packet is corrupted, the packet length in the IP header may be incorrect and calculating checksum of such packets results in kernel assert. Hence make sure the packet length is always less than or equal to the skb length before calling the checksum function. Change-Id: Ib6ed8887eafdc55726e1a31ec5e3fedf1e4fc456 Closes-Bug: 1800964
Configuration menu - View commit details
-
Copy full SHA for 79f19d1 - Browse repository at this point
Copy the full SHA 79f19d1View commit details
Commits on Nov 28, 2018
-
Initialize the root node label flags to 0 which will be inherited by …
…the other nodes during mtrie initialization for RT in new VRF Partial-Bug: 1797409 Change-Id: I932d076ae43a1de17a15effb0c05bc14a86735e3
Prasannaa TS committedNov 28, 2018 Configuration menu - View commit details
-
Copy full SHA for f948923 - Browse repository at this point
Copy the full SHA f948923View commit details
Commits on Dec 3, 2018
-
Mergeable buffer issue with Ericsson vEPG
Closes-bug: #1806441 Issue with the workaround done for the qemu bug. Workaround to use the previous negotiated feature list to be used only after a vrouter restart. Change-Id: I036a8fd220b118ccc532eceb742345703a87b9e0
Jeya ganesh babu J committedDec 3, 2018 Configuration menu - View commit details
-
Copy full SHA for cb137a3 - Browse repository at this point
Copy the full SHA cb137a3View commit details
Commits on Dec 24, 2018
-
[Orange] Contrail 4.1.1- connectivity issue on several VMs of "L2 Onl…
…y" Network In the customer's case, it was found that there is a L3 VN vrf which got reassigned to a L2 VN vrf due to VN deletion and addition. However the deletion of the default route in L3 VN vrf was not happening leading to the proxy arp issue. This got reproduced in the lab by doing the exact same steps of route add and delete in a dummy vrf as below:- inet_route_add: vrf:22 IP:22.5.6.0/29 nh_id:1 label:1048575 lbl_flags:12 ret 0 inet_route_add: vrf:22 IP:22.5.6.1/32 nh_id:12 label:1048575 lbl_flags:6 ret 0 inet_route_add: vrf:22 IP:22.5.6.2/32 nh_id:12 label:1048575 lbl_flags:6 ret 0 inet_route_add: vrf:22 IP:169.254.169.254/32 nh_id:10 label:1048575 lbl_flags:14 ret 0 inet_route_add: vrf:22 IP:0.0.0.0/0 nh_id:135 label:18 lbl_flags:3 ret 0 inet_route_del: vrf:22 IP:22.5.6.1/32 nh_id:1 label:1048575 lbl_flags:12 ret 0 inet_route_del: vrf:22 IP:22.5.6.2/32 nh_id:1 label:1048575 lbl_flags:12 ret 0 inet_route_del: vrf:22 IP:22.5.6.0/29 nh_id:135 label:18 lbl_flags:3 ret 0 inet_route_del: vrf:22 IP:169.254.169.254/32 nh_id:135 label:18 lbl_flags:3 ret 0 inet_route_del: vrf:22 IP:0.0.0.0/0 nh_id:0 label:1048575 lbl_flags:0 ret 0 The root cause for this issue is that when the route before the default route is deleted i.e 169.254.169.254 in this case, the agent deletes it using the replace plen of 0 and as part of that the root node points to an NH node which just has a pointer to the default route NH. But later when the delete for default route is triggered, the current code just returns without cleaning up the entry. Fix is to cleanup the entry properly in this case. Fix has been verified at customer's setup. Closes-Bug: #1807371 Change-Id: I4d9d814e1b9796896517dd9e74f0e2f16063b92a
Configuration menu - View commit details
-
Copy full SHA for fee02e4 - Browse repository at this point
Copy the full SHA fee02e4View commit details
Commits on Jan 2, 2019
-
Merge "[Orange] Contrail 4.1.1- connectivity issue on several VMs of …
…"L2 Only" Network" into R4.1
Configuration menu - View commit details
-
Copy full SHA for 660a8bf - Browse repository at this point
Copy the full SHA 660a8bfView commit details
Commits on Jan 30, 2019
-
CC13: DPDK vRouter reports double traffic rate on vm tap interface
DPDK TX packet counter incremented twice in eth_tx(dp-core/vr_interface.c) and dpdk_gro_process(dpdk/vr_dpdk_gro.c) files. If VP_FLAG_GRO packet flag set, then it will not be incremented in eth_tx function. With below change, when we have GRO packet, the counter will be incremented in dpdk_gro_flush function. Closes Jira Bug: JCB-165865 Change-Id: I00e8d0b182107709988a2813fa04768841bcb30c (cherry picked from commit 933d4b4)
Configuration menu - View commit details
-
Copy full SHA for 3f7b9c3 - Browse repository at this point
Copy the full SHA 3f7b9c3View commit details
Commits on Jan 31, 2019
-
Do not decrement TTL while doing L3 forward to agent pkt0
In BGPaaS in pkt mode, the pkt will be L3 routed with the NH pointing to pkt0 or GW interface at one stage (before NAT). In single hop BGPaaS case, since the TTL is decremented in vr_forward() during L3 routing, the pkt is dropped later since the TTL becomes 0. This leads to BGP session not coming up at all. Change-Id: I959dcf4e1b316a76698f2b5d954bbaf0f197b64e closes-jira-bug: JCB-199874
Configuration menu - View commit details
-
Copy full SHA for 63d0e59 - Browse repository at this point
Copy the full SHA 63d0e59View commit details
Commits on Feb 5, 2019
-
DPDK: GRO flag reset during Agent/DPDK container reboot
Issue fixed by setting GRO and merge buffer flag in vif_set_flags API. If those flags are already set, it will retain else ignore. Closes Jira Bug: JCB-218956 Change-Id: I45020c3231fe79ea7b4b629471c234d2cb87d1e9
Configuration menu - View commit details
-
Copy full SHA for e67f543 - Browse repository at this point
Copy the full SHA e67f543View commit details
Commits on Feb 8, 2019
-
DPDK: Fix for vhost0 MTU not getting set
After an agent soft reset, vif 0 gets deleted and added back. But due to a timing issue, before it gets added back, the vhost0 MTU notification used to come resulting in MTU not getting set (Since vif is not added yet). Fix is to query the MTU from PMD during vif 0 addition. Change-Id: I2b102b82e21fcc137e62db748d8982bdbbdf87e2 Closes-Bug: 1795839 (cherry picked from commit 514d42f)
Configuration menu - View commit details
-
Copy full SHA for 928a608 - Browse repository at this point
Copy the full SHA 928a608View commit details
Commits on Feb 12, 2019
-
Contrail 5.0.2 DPDK Vrouter Pull Fails Drops
Fixed pull counter issue by having check for GRE IP fragmentation check. If Outer header is GRE and fragmented packet, we return 0,:wq so that it will be queued for furhter processing. Closes Jira Bug: JCB-218856 Change-Id: I805c7d66d2dad20c15dfe2656a8fe24e746b1c99
Configuration menu - View commit details
-
Copy full SHA for d963d14 - Browse repository at this point
Copy the full SHA d963d14View commit details
Commits on Feb 16, 2019
-
Configuration menu - View commit details
-
Copy full SHA for 09a47d2 - Browse repository at this point
Copy the full SHA 09a47d2View commit details
Commits on Feb 18, 2019
-
Configuration menu - View commit details
-
Copy full SHA for 8990557 - Browse repository at this point
Copy the full SHA 8990557View commit details -
Configuration menu - View commit details
-
Copy full SHA for d007993 - Browse repository at this point
Copy the full SHA d007993View commit details
Commits on Feb 21, 2019
-
Vrouter stops forwarding with iperf TCP test
closes-jira-bug: CEM-2807 Limitation in rte_port_ethdev_writer_tx_bulk in the pkts_mask field that we can transfer only 64 segments. The segmentation code can send a max of 128 segments, which was causing the mbuf leak. Fixed the code, by calling the tx_bulk function in a loop Change-Id: I81212f1ae32e47dd80c0938631e466db98242a3e
Jeya ganesh babu J committedFeb 21, 2019 Configuration menu - View commit details
-
Copy full SHA for f9d3701 - Browse repository at this point
Copy the full SHA f9d3701View commit details
Commits on Feb 28, 2019
-
Vrouter stops forwarding with iperf TCP test
closes-jira-bug: CEM-2807 tx_bulk can support upto burst of 64 packets. Change the send size to 64 instead of 32. Change-Id: Ifbf45e9fc23dd4d8e0f69fefc745730d3fbf2726
Jeya ganesh babu J committedFeb 28, 2019 Configuration menu - View commit details
-
Copy full SHA for 26f0926 - Browse repository at this point
Copy the full SHA 26f0926View commit details
Commits on Mar 2, 2019
-
Fix dpdk vrouter crash while doing vifdump
closes-jira-bug: CEM-2996 Added a null check for variable. Change-Id: Ieab7662d27340bddd2c36a30bfa7aad3a4b3f67c
Jeya ganesh babu J committedMar 2, 2019 Configuration menu - View commit details
-
Copy full SHA for 65e3ecf - Browse repository at this point
Copy the full SHA 65e3ecfView commit details
Commits on May 7, 2019
-
dpdk vrouter changes Ethernet Type field of an inner IPv6 pkt to Type…
…:IPv4 when mirroring is configured at policy level to a physical analyzer Root cause: issue seen for mirroring Ingress packet while trying to form overlay ethernet header. It tries to form from next hop encap data and its always IPV4. Fixed the issue by checking the packet type and updating overlay ethernet header protocol. Change-Id: Ie998d7de7f7d0869f46a646e4a147d2cea638565 closes-jira-bug: CEM-4574
Configuration menu - View commit details
-
Copy full SHA for a66eb8b - Browse repository at this point
Copy the full SHA a66eb8bView commit details
Commits on May 13, 2019
-
Add support to evict a stale TCP "Dead" flow
Change-Id: Ic31876d87610e1998593431053593b7f17bae207 closes-jira-bug: CEM-5420
Configuration menu - View commit details
-
Copy full SHA for 72fe825 - Browse repository at this point
Copy the full SHA 72fe825View commit details
Commits on May 22, 2019
-
DPDK: Possible mbuf leak running iperf with MSS 536
closes-jira-bug: CEM-4659 Seems like the 18.05 dpdk has issue when sending 64 packets in burst. Reverting back to use VR_DPDK_TX_BURST_SZ Change-Id: I4d08be695a5a7968f6d07be2688bf1243b399f82 (cherry picked from commit a61b3dc)
Jeya ganesh babu J committedMay 22, 2019 Configuration menu - View commit details
-
Copy full SHA for 67cc2d5 - Browse repository at this point
Copy the full SHA 67cc2d5View commit details
Commits on Jun 13, 2019
-
Enhancement Request - vifdump script - stop command issues
Updated Defensive check to overcome this issue, If actual interface provided for "vifdump stop <Id>" an erro message would be displayed. closes-jira-bug: CEM-5980 Change-Id: I2e5ff28e7da15527c9ac4628ebff5aa1dee5f233
Configuration menu - View commit details
-
Copy full SHA for a20f081 - Browse repository at this point
Copy the full SHA a20f081View commit details
Commits on Jun 19, 2019
-
NTTC issue: CEM-4275 - One of the TCP dead flow doesn't get evicted
Root cause: =========== This is due a race condition between setting of the Evict Candidate flag and the Eviction of the flow. We have 2 threads here say Thread1 and Thread2. Thread1 is in the flow defer_callback function, while Thread2 is in the flow mark evict function. Now consider the following sequence (Time T0 to T4) which leads to the non eviction of the flow 100. T0 and T3 are executed in Thread1, while T1, T2 and T4 are executed as part of Thread 2. Lets assume the flow index for the flows as 100 and 200. Thread 1 Thread 2 ------------- ------------- Defer_cb_Func() Mark_Evict_Func() Time T0 - CheckEvictFlow(100) – No op, as Evict candidate flag is not set for flow 100 Time T1 – Set Evict Candidate for flow(100) Time T2 – Set Evict Candidate for flow(200) Time T3 – CheckEvictFlow(200) – Now flow 200 would get evicted since Evict Candidate flag is set. Time T4 – Schedule Defer_cb for flow(200) Now, since flow 200 already got evicted at Time T3, the callback would never be scheduled at time T4. And hence, the flow 100 would never get evicted. Was able to repro the problem by simulating the above order using instrumented code and scapy script. The problem is with Defer_cb_func() which should do the eviction only when the Evict Candidate flag is set for both flows. Fix: ==== The fix is to make sure the eviction is done only after both the flows evict candidate flags are set. Testing: ======== - Done UT for the code changes - QA has qualified the fix Change-Id: I07730ab190646260d08de6ca4fdf9bc1caf16d6e closes-jira-bug: CEM-4275 (cherry picked from commit e04c2b0)
Configuration menu - View commit details
-
Copy full SHA for bafabba - Browse repository at this point
Copy the full SHA bafabbaView commit details
Commits on Jul 15, 2019
-
Merge "dpdk vrouter changes Ethernet Type field of an inner IPv6 pkt …
…to Type:IPv4 when mirroring is configured at policy level to a physical analyzer" into R4.1
Configuration menu - View commit details
-
Copy full SHA for 9dea585 - Browse repository at this point
Copy the full SHA 9dea585View commit details -
Merge "NTTC issue: CEM-4275 - One of the TCP dead flow doesn't get ev…
…icted" into R4.1
Configuration menu - View commit details
-
Copy full SHA for adf9aa1 - Browse repository at this point
Copy the full SHA adf9aa1View commit details
Commits on Jul 26, 2019
-
Update R4.1 branch Ubuntu Kernel to 3.13.0-171
closes-jira-bug: CEM-6709 Change-Id: If74a824cf7493a3e5c5152902c49a8a0dd88f9b3
Configuration menu - View commit details
-
Copy full SHA for ca21f51 - Browse repository at this point
Copy the full SHA ca21f51View commit details -
Fix for flooding of packets in dpdk vrouter
closes-jira-bug: CEM-5251 Disabling promiscuous mode should be done based on the result of adding the unicast address. Drivers such as i40e doesnt support the apis to set multicast address. But the bond driver is made to enable all multicast for all the slave drives. So this shouldnt be an issue. Change-Id: Ie754fad59a215462b62da6e2ab309de13422d1fd (cherry picked from commit 7a20fc6)
Jeya ganesh babu J committedJul 26, 2019 Configuration menu - View commit details
-
Copy full SHA for cd5388a - Browse repository at this point
Copy the full SHA cd5388aView commit details
Commits on Aug 13, 2019
-
Configuration menu - View commit details
-
Copy full SHA for 56ffb40 - Browse repository at this point
Copy the full SHA 56ffb40View commit details
Commits on Sep 6, 2019
-
Fix for memory leak in kmod vrouter
When vr_send_broadcast() is called during route add or delete, there will be a 2048 slab memory leak. This is because response->vr_message_buf is getting set to NULL and so vr_message_free() will not free response->vr_message_buf. response->vr_message_buf is only applicable to unicast netlink case and not for multicast netlink. Fix is to move it to unicast case only. Closes-Jira-bug: CEM-5343 Change-Id: Ia4d035e88765c55f65fedce1963000c0bced55c1
Configuration menu - View commit details
-
Copy full SHA for 620f49f - Browse repository at this point
Copy the full SHA 620f49fView commit details
Commits on Oct 24, 2019
-
Build of vrouter.ko is broken on some 4.4.0 kernels (ubuntu 16.04)
As per linux kernel git log, the get_user_pages() was changed in 4.4.168 which also maps to Ubuntu 4.4.0-143. Fixed by checking kernel version 4.4.168. Verified by building vrouter.ko on CI VM with ubuntu 4.4.0-143 and regular centos setup. Change-Id: Ie759e567a6f5f82d1c4d1dd6ea0f21c23f7287ff closes-jira-bug: CEM-4223 (cherry picked from commit c33e99f)
Configuration menu - View commit details
-
Copy full SHA for 7233932 - Browse repository at this point
Copy the full SHA 7233932View commit details
Commits on Dec 5, 2019
-
DPDK: Fix for FD leak when start/stop VM in loop
A spurious vr_uvh_cl_timer_setup() call was happening which was leaking fds Closes-Jira-Bug: CEM-10799 Change-Id: I1a02ba31b747469f8ebab79c603f80b8102238a4 (cherry picked from commit ec8ede8)
Configuration menu - View commit details
-
Copy full SHA for 33778eb - Browse repository at this point
Copy the full SHA 33778ebView commit details
Commits on Dec 16, 2019
-
Revert "DPDK: Fix for FD leak when start/stop VM in loop"
This reverts commit 33778eb. Closes-Jira-Bug: CEM-11181 Change-Id: I78b909a442df6b2ec6e77520fe38b3ee17518f95
Configuration menu - View commit details
-
Copy full SHA for bbf27ef - Browse repository at this point
Copy the full SHA bbf27efView commit details
Commits on Jan 28, 2020
-
CEM-11421 - Orange ZRB Missing prefixes series of occurrences
Root cause: The issue was reproduced with the following sequence of route adds and delete. addr = 10.60.7.0 plen = 25 vrf = 4 operation = ADD/CHANGE nh_idx = 38 label = 410760 addr = 10.60.0.0 plen = 23 vrf = 4 operation = ADD/CHANGE nh_idx = 38 label = 410760 addr = 10.60.7.128 plen = 25 vrf = 4 operation = ADD/CHANGE nh_idx = 38 label = 410760 addr = 10.60.0.0 plen = 20 vrf = 4 operation = ADD/CHANGE nh_idx = 63 label = 609700 addr = 10.60.0.0 plen = 20 vrf = 4 operation = DELETE nh_idx = 51 After executing this sequence, it was observed that the 10.60.7.0/25 and 10.60.7.128/25 routes were getting deleted as part of 10.60.0.0/20 delete operation. This was because as part of delete we were deleting the mtrie bucket containing 10.60.7.0/25 and 10.60.7.128/25 routes. The mtrie bucket was being deleted unconditionally and wrongly. Fix: The mtrie bucket deletion logic is changed to delete the bkt using one of the entries in the bucket instead of deleting it using the values from the route delete request. This fixes the issue of bucket with non matching prefix length being deleted. Verification: The fix was verified by rerunning the same sequence again and checking if the 10.60.7.0/25 and 10.60.7.128/25 routes are present or not. The fix also passed full vrouter regression. closes-jira-bug: CEM-11421 Change-Id: I4e879a1753f273a8c23b1baf5e82b8d56f675e98
Configuration menu - View commit details
-
Copy full SHA for 528264f - Browse repository at this point
Copy the full SHA 528264fView commit details
Commits on Mar 28, 2020
-
Socket leak when VM stop/start In loop
When VIF is added but not connected, vru_cl->vruc_fd is not added to the fd list even though it is created. This can happen when VM is stopped. In this case, vr_uvhost_del_fds_by_arg() would not close the fd. This leads to socket leak, so added fcntl() call in vr_uvhost_del_client() to check if vruc_fd is closed or not. Closes-jira-bug: CEM-9285 Change-Id: Ia55e44b8830b96e8313691dc77873bfd133e1469 (cherry picked from commit 194c8bc)
Configuration menu - View commit details
-
Copy full SHA for a7689f4 - Browse repository at this point
Copy the full SHA a7689f4View commit details
Commits on Jul 20, 2020
-
Fix for Fragment errors due to integer overflow
Return mod of hash table during scanning instead of monotonically increasing number. With the later, there can be integer overflow after some time resulting in incorrectly being handled as error. Due to this scanning of hashtable will stall. Change-Id: Ia55da9a6d7e66db21638c4c71f50d7940b02138a Closes-Jira-Bug: CEM-17147 (cherry picked from commit 9195829)
Configuration menu - View commit details
-
Copy full SHA for d436e54 - Browse repository at this point
Copy the full SHA d436e54View commit details
Commits on Jul 22, 2020
-
Health check fails to work after VM reboot
Issue: When the VM is rebooted, server stops running and health check fails momentarily, due to which the route to VM points to nh_discard. When the first TCP SYN packet is sent, it is trapped by agent while flow is programmed. Before the packet is enqueued NH of the packet is set to NULL. After flushing the packet, vr_inet_route_lookup() is done on packet with NAT IP whose route is already pointed to nh_discard. Due to this the first SYN packet is being dropped. By the time retransmission of this packet happens, it reaches health check timeout of 1 sec and connection closes. This keeps on repeating Fix: After flushing the packet, before the IP NAT happens, we do a vr_inet_route_lookup() which gives us the NH associated with non-NAT IP which prevents the first SYN packet drop. Closes-Jira-Bug: CEM-11226 Change-Id: I9193242918d42bcf1bc7b6590884a3da9f785d20
Configuration menu - View commit details
-
Copy full SHA for 817a3fb - Browse repository at this point
Copy the full SHA 817a3fbView commit details
Commits on Jul 27, 2020
-
Configuration menu - View commit details
-
Copy full SHA for d71b4e3 - Browse repository at this point
Copy the full SHA d71b4e3View commit details
Commits on Nov 27, 2020
-
CEM-18916: Current implementation doesnt reset parent socket state to…
… Listening after closure of connections; once max connections are reached. New implementation shall ensure that. Change-Id: Ief2e6ba15f7da127b4d488484bff346fec00374e Closes-jira-bug: CEM-18916 (cherry picked from commit 385c774)
Configuration menu - View commit details
-
Copy full SHA for 6c14e3c - Browse repository at this point
Copy the full SHA 6c14e3cView commit details