summaryrefslogtreecommitdiff
path: root/src/ipcpd/eth
Commit message (Collapse)AuthorAgeFilesLines
* ipcpd: Use non-blocking socket in Ethernet IPCPSander Vrijders2018-10-031-3/+17
| | | | | | | | | | Since the Ethernet IPCP now has multiple reader threads it was possible that both exit the select call, which caused one of the two threads to block on the recv call. This makes the socket non-blocking so that the recv call simply fails. Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be> Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be>
* build: Prioritize raw socket API in buildSander Vrijders2018-10-031-15/+18
| | | | | | | | This will change the build to exclusively select one raw socket API in case multiple are present in the sytem, which will simplify the code. Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be> Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be>
* ipcpd: Bypass Qdisc in Ethernet IPCPSander Vrijders2018-10-031-0/+8
| | | | | | | | Since Linux kernel 3.14 there is the option to bypass the kernel Qdisc. This will speed up the Ethernet IPCP. Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be> Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be>
* ipcpd: Fix build for netmap enabled Ethernet IPCPSander Vrijders2018-10-032-7/+9
| | | | | | | | | There were some compilation issues introduced by adding the interface monitor to the Ethernet IPCP. Furthermore it was not possible to select between raw sockets or netmap if both were available. Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be> Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be>
* ipcpd: Fix bad lock in Ethernet IPCPSander Vrijders2018-10-031-1/+1
| | | | | | | | An unlock was called twice instead of a lock/unlock sequence, causing a data race. Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be> Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be>
* ipcpd: Add multithreading to Ethernet IPCPDimitri Staessens2018-10-032-31/+61
| | | | | | | | | This adds multiple reader and writer threads, configurabe via cmake with IPCP_ETH_RD_THR and IPCP_ETH_WR_THR. Improves ethernet IPCP throughput, which looks to be limited by the raw socket calls. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Remove double close in Ethernet IPCPSander Vrijders2018-09-271-2/+1
| | | | | | | | | There was a double close of a fd in the Ethernet IPCP. It also passes the correct max length to a memcpy (a strlen of the source string was used previously). Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be> Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be>
* include: Remove _DEFAULT_SOURCE in endian.hDimitri Staessens2018-09-271-0/+2
| | | | | | | | This removes the _DEFAULT_SOURCE definition in the endian header as it should not be there. This avoids double and conflicting definitions. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* Merge branch 'testing' into beSander Vrijders2018-06-131-22/+28
|\
| * ipcpd: Get mtu using ioctl before hw addressDimitri Staessens2018-06-131-6/+6
| | | | | | | | | | | | | | | | | | The SIOCGIFMTU command uses the ifr struct, which is a union, so it can't store the hw address and the MTU at the same time. We now call SIOCGIFMTU and set the MTU before SIOCGIFHWADDR. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
| * ipcpd: Fix MTU handling in ethDimitri Staessens2018-06-131-16/+22
| | | | | | | | | | | | | | | | This fixes the MTU handling in eth. Buffers are now allocated to smaller size. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* | Merge branch 'testing' into beSander Vrijders2018-06-081-1/+12
|\|
| * ipcpd: Limit eth MTU to rdrb in single-block modeDimitri Staessens2018-06-081-1/+12
| | | | | | | | | | | | | | | | | | When the SHM is in single block mode, the MTU may be bigger than a block. The eth IPCPs reserved buffers the size of MTU, which is now limited. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* | lib: Add a data qos cubeDimitri Staessens2018-06-081-1/+1
|/ | | | | | | | This adds a data qos cube that is reliable. Reliable qos can be selected by setting the loss parameter of the qosspec to 0. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Support raw qos in eth and udp IPCPsDimitri Staessens2018-06-081-1/+1
| | | | | | | | Flow allocation for raw was not yet supported in these IPCPS, causing enrollment to fail. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Fix setting MTU for eth-llcDimitri Staessens2018-05-081-3/+12
| | | | | | | | The MTU was not correctly set for the eth-llc ipcp. This also fixes a missing definition of ETH_MAX_MTU, which was introduced in kernel v4.10. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Use correct MTU for eth-dixDimitri Staessens2018-05-071-4/+24
| | | | | | | | The eth-dix IPCP was using a hardcoded MTU of 1500 bytes, but could support higher MTUs. Now jumbo frames are supported. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Fix truncate of small packetsDimitri Staessens2018-04-051-1/+1
| | | | | | | | The truncate was setting the length to the frame length, instead of the actual payload length to be delivered to the N+1. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Add ethertype checkDimitri Staessens2018-04-051-1/+7
| | | | | | | | An ethertype check was missing for the DIX ethernet IPCP, causing crashes if there is other traffic on the network. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Avoid unnecessary copy in eth writerDimitri Staessens2018-04-051-25/+45
| | | | | | | | | This writes into the Ethernet device directly from the rdrbuff to avoid a copy on the write side in the Ethernet IPCPs. This does not work for the netmap device. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Avoid unnecessary copy in eth readerDimitri Staessens2018-04-051-31/+70
| | | | | | | | | This reads from the Ethernet device directly into the rdrbuff to avoid a copy on the read side in the Ethernet IPCPs. This does not work for the netmap device. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Enable locking threads to a CPU coreDimitri Staessens2018-04-051-0/+4
| | | | | | | | | This adds a function that locks a thread to a random core. This greatly improves performance on multi-cpu systems. There is no portable way to do this, this only implements it for GNU/Linux. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Fix frame buffer in eth ipcpsDimitri Staessens2018-04-051-2/+2
| | | | | | | | The packets were being sent and read into a buffer that had the payload length instead of the frame length. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Add validity checks for eth-dixDimitri Staessens2018-03-121-0/+4
| | | | | | | | This will check if the Ethertype value is a valid Ethertype in the irm tool and the eth-dix IPCPd. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Add length field to eth-dixDimitri Staessens2018-03-101-62/+57
| | | | | | | | | Some NICs added padding to the Ethernet II frames causing bad frame lengths and GPB unpack fails. This adds a 2-byte length field to the DIX frame to circumvent this. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>
* ipcpd: Add IPCP over DIX EthernetDimitri Staessens2018-03-104-0/+1811
This adds an IPC Process that uses DIX Ethernet with an Ethertype that is configurable at bootstrap. This allows parallel DIX layers over the same Ethernet network with different Ethertypes (and one LLC layer). It allows jumbo frames in the future, and should avoid the problems we have with some routers not handling LLC traffic very well. The destination endpoint ID is sent as a 16 bit integer, so the maximum payload is 1498 bytes in standard Ethernet, and 8998 bytes when Jumbo frames are used. The implementation is very similar to the Ethernet LLC IPCP, so it is implemented using preprocessor macros in the single source instead of duplicating code. Signed-off-by: Dimitri Staessens <dimitri.staessens@ugent.be> Signed-off-by: Sander Vrijders <sander.vrijders@ugent.be>