summaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
...
* lib: shm_ap_rbuff: sleep when read on empty rbuffdimitri staessens2016-05-282-10/+23
| | | | | When the ring buffer is empty, a read call will sleep. A write call on an empty ring buffer will wake up sleeping readers.
* ipcpd: cleanup of IPC processesdimitri staessens2016-05-282-9/+18
| | | | | | | This stops IPC processes from hanging due to bad locks. It first informs all threads that the IPCP is shutting down, then cancels all threads. Cancellation is still required because threads may be in accept() or receivefrom().
* lib: fixed empty line in irm.cdimitri staessens2016-05-281-1/+0
|
* build: fixed typo in ipcpd/local/dimitri staessens2016-05-281-1/+1
|
* ipcpd: IPCP over local memorydimitri staessens2016-05-287-4/+696
| | | | | | | | | | | | | This introduces an IPC process that relays data between two local AP's over the shm_du_map. Only configuration it has is a DIF name. It required small modification elsewhere: lib: added support for the IPCP_LOCAL type irm: added support for the IPCP_LOCAL type dif_config: added the IPCP_LOCAL type tools: added support for the IPCP_LOCAL type
* ipcpd: shim-udp changed to blocking writedimitri staessens2016-05-271-14/+11
| | | | | The shim UDP did a non-blocking write call, causing loss of SDU's. This has been changed to a blocking write call.
* tools: updated cbr with flood and sleep optionsdimitri staessens2016-05-272-21/+60
| | | | | | | The cbr client will now use busy waiting by default to control the sending rate. A --sleep option has been added to allow low CPU usage when sending at low data rates. A --flood option has been added that writes SDU's as fast as possible.
* Merged in dstaesse/ouroboros/be-select-fix (pull request #115)Sander Vrijders2016-05-271-3/+4
|\ | | | | | | ipcpd: fixed the shim-udp grinding the cpu
| * ipcpd: fixed the shim-udp grinding the cpudimitri staessens2016-05-271-3/+4
| | | | | | | | | | | | | | | | In linux, the select call changes the timeout variable to the time not spent in sleep, causing the timeout to decrease and that loop to grind the cpu. The timeval is now reset to the correct value every loop. Also removed a LOG_ERR message that was hampering performance.
* | lib: fixed SHM_DU_BUFF_BLOCK_SIZE in shm_du_mapdimitri staessens2016-05-261-1/+1
|/
* lib: fixed blocking/non-blocking writedimitri staessens2016-05-261-16/+43
| | | | | non-blocking write would return when the buffer was full functions in dev now check validity of input file descripters
* Merged in sandervrijders/ouroboros/be-cbr (pull request #112)dimitri staessens2016-05-263-36/+130
|\ | | | | | | Be cbr
| * ipcpd: Cleanup handler for main message loopSander Vrijders2016-05-261-0/+5
| | | | | | | | | | This adds a cleanup handler for the main message loop, which closes the socket upon pthread_cancel so that the ipcp can exit cleanly.
| * tools: Add threadpool to cbrSander Vrijders2016-05-262-36/+125
| | | | | | | | | | This adds a threadpool to cbr, so that it is not overflooded with client requests.
* | Merged in dstaesse/ouroboros/be-shm-single-block (pull request #111)Sander Vrijders2016-05-251-6/+31
|\ \ | |/ |/| | | lib: allow running shm_du_map in single block mode
| * lib: allow running shm_du_map in single block modedimitri staessens2016-05-251-6/+31
| | | | | | | | | | this allows disabling support for SDU's that are larger than a block in the DU MAP by defining the variable SHM_MAP_SINGLE_BLOCK
* | tools: cbr continues when a write failsdimitri staessens2016-05-251-2/+2
| | | | | | | | | | the CBR tool would stop if the buffer is full. Now it will continue sending SDU's until the duration expires.
* | tools: fixes comments on d74d67adimitri staessens2016-05-243-10/+10
| |
* | tools, lib: cbr tool for bandwidth testingdimitri staessens2016-05-247-0/+554
| | | | | | | | | | | | | | | | The tool sends constant bandwidth traffic between a client and a server. cbr --help for more info. Adds time_utils.h to the library containing useful functions for arithmetic with timespec and timeval structures.
* | lib: Fix memleaks in dev.cSander Vrijders2016-05-241-23/+30
|/ | | | Fixes some memleaks in dev.c in the case of error conditions.
* irmd: use the shm_du_map as a uniqueness checkdimitri staessens2016-05-221-2/+7
| | | | | This makes sure only one shm_du_map is running in the system. Starting a second irmd would invalidate the shared memory map.
* Merged in sandervrijders/ouroboros/be (pull request #108)dimitri staessens2016-05-221-1/+1
|\ | | | | | | irmd: Change read lock to write lock
| * irmd: Change read lock to write lockSander Vrijders2016-05-221-1/+1
| | | | | | | | | | A read lock was taken instead of a write lock, which resulted in double frees.
* | ipcpd: shim-udp: fixed missing NULL checkdimitri staessens2016-05-221-0/+3
|/ | | | Conf was not checked on bootstrap.
* lib, ipcpd, irmd: fixes deallocation and fast pathdimitri staessens2016-05-223-35/+44
| | | | | | The fast path has been rewritten to have certainty to read the correct flow. Deallocation will not release port_id's or fd's until they are explicitly released locally.
* ipcpd: Remove install command from buildSander Vrijders2016-05-221-2/+0
| | | | | This removes an install command from the build, since the protobuf header file was being installed in the source directory.
* lib: removed fd cleaning functiondimitri staessens2016-05-222-30/+14
|
* ipcpd: homogenous debug messages for deallocationdimitri staessens2016-05-211-8/+1
|
* lib, irmd, ipcpd: Flow deallocation over shim UDPdimitri staessens2016-05-215-61/+210
| | | | | | The shim UDP now supports deallocating a flow end-to-end. Contains some stability fixes for flow allocation and some missing close() calls in lib/sockets.
* lib, ipcpd: fixes segfaults in dev.c and shim-udpdimitri staessens2016-05-202-4/+2
|
* lib: allow parallel connectionsdimitri staessens2016-05-202-11/+30
| | | | | | dev.c: read now only reads an SDU if is is for the correct port_id shm_ap_rbuff: added a function peek() that returns the port_id of the tail.
* ipcpd: shim-udp: read fd on client sidedimitri staessens2016-05-201-0/+13
| | | | | There could be a theoretical race condition that the server sends a message before the FD is set on the client side, which is resolved.
* ipcpd: shim-ipcp: fixed race conditionsdimitri staessens2016-05-201-40/+57
| | | | | | | | | | | There were two race conditions. The first was where client received the response message before the client UDP port was connected to the server UDP port and the first message of the client was sent to the LISTEN port instead of the server application port. This was solved by reordering the operations so the response is sent after the port is correctly connected. The second race condition is that the FD is not added to the set in time at the server side before the first SDU arrives. This was solved by adding a variable proteceted by a mutex.
* ipcpd: shim-udp: removed debug codedimitri staessens2016-05-191-12/+0
| | | | | some debug code for locking and a sleep check are removed from the shim-udp.
* Implementation of a full flow allocator for the shim UDP.dimitri staessens2016-05-186-258/+540
| | | | | | | | | It uses UDP port 0x0D1F on all hosts to send and receive flow allocation messages. It supports communication between server and client AP over a single shim IPCP. Implementation of full flow deallocation is pending. Both the client and the server still have to call flow_dealloc();
* lib: Add missing check upon garbage collectionSander Vrijders2016-05-171-2/+3
| | | | | | | There was a check missing upon garbage collection. If the whole DU map has been used once, all DUs will have their garbage bit set to one. This meant that the garbage collector would keep moving the tail indefinitely.
* lib: Fixed incorrect initializationSander Vrijders2016-05-171-3/+4
| | | | | When moving the tail pointer in the DU map, blocks was never reset to zero. This caused it to sometimes clean up too many DUs.
* irmd: fixed bad locksdimitri staessens2016-05-171-4/+2
|
* lib: Add missing modulo calculationSander Vrijders2016-05-163-13/+12
| | | | | | Adds a missing modulo calculation that would result in index -1 being returned instead of zero when the ringbuffer skipped back to the beginning.
* Merged in dstaesse/ouroboros/be-dev-locks (pull request #95)Sander Vrijders2016-05-161-28/+33
|\ | | | | | | lib: dev: small locking improvements
| * lib: dev: small locking improvementsdimitri staessens2016-05-161-28/+33
| | | | | | | | | | | | This avoids holding the long locks during accept() and alloc_res() calls. Small reduction of time of other locks held. No changes in locking logic.
| * Merge branch 'be-udp-locks' into be-dev-locksdimitri staessens2016-05-152-96/+97
| |\
* | \ Merged in dstaesse/ouroboros/be-irmd-locks (pull request #94)Sander Vrijders2016-05-161-175/+223
|\ \ \ | | | | | | | | | | | | irmd: new locking implementation
| * | | irmd: fixed unlock after freedimitri staessens2016-05-161-2/+4
| | | | | | | | | | | | | | | | unlock is now called before the object is destroyed.
| * | | irmd: new locking implementationdimitri staessens2016-05-161-175/+221
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This locking should be more consistent, It now has three locks, one guarding flows and port_id's, one guarding the registered apps and ipcps, and one guarding the overall state of the irmd. There are two additional mutexes guarding the condition variables.
* | | | Merged in dstaesse/ouroboros/be-udp-locks (pull request #92)Sander Vrijders2016-05-162-103/+94
|\ \ \ \ | |_|/ / |/| | | | | | | ipcpd: shim-udp: Revised locking
| * | | fixed two unlocks, removed the thread_lock.dimitri staessens2016-05-151-13/+3
| | |/ | |/|
| * | ipcpd: shim-udp: Revised lockingdimitri staessens2016-05-152-96/+97
| |/ | | | | | | | | | | | | | | | | | | Simplified locking to take only two locks: the first lock guards the state of the ipcp. This lock must be held for writing on bootstrap and closing, and held for reading during all other operations. The second lock guards operations on flows, and must be held for writing during allocation and deallocation, and held for reading when sending sdu's. After adding a fd to FD_SET, the shim will wait for 1 ms to ensure that the FD is added to the select call.
* / lib: dev.c: Revised lockingdimitri staessens2016-05-151-69/+65
|/ | | | | fd and flows are now all protected by the flows_lock while the overall state of the AP is protected by the data_lock.
* lib: allow accepting any flowsdimitri staessens2016-05-151-7/+7
| | | | The application can call accept(fd, NULL, NULL);