Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
| | * | | irmd: Remove IPCP from list upon error | Sander Vrijders | 2016-08-10 | 1 | -0/+13 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | If an IPCP crashes (due to a segfault for instance), it is removed from the spawned apis list. However, if it was an IPCP it should also be removed from the IPCPs list, since else on shutdown, the irmd will try to destroy the IPCP that crashed. | |||||
| | * | | ipcpd: Fix locking problem in local IPCP | Sander Vrijders | 2016-08-10 | 1 | -2/+2 | |
| | |/ | | | | | | | | | | | | | A lock was not being taken to check the state, but then it was released if the state was not IPCP_NULL, resulting in a segfault. | |||||
* | / | irmd: Graceful shutdown | dimitri staessens | 2016-08-10 | 5 | -80/+82 | |
|/ / | | | | | | | | | | | Threads from the threadpool will now check the state of the IRMd and exit gracefully without a need to cancel them. This avoids the unsafe call of pthread_cancel in the signal handler. | |||||
* / | irmd: Fix locking | dimitri staessens | 2016-08-09 | 1 | -2/+2 | |
|/ | | | | registry_sanitize_apis should be called under write locked reg_lock. | |||||
* | irmd: Fix some locking issues | dimitri staessens | 2016-08-09 | 2 | -3/+7 | |
| | ||||||
* | lib: Fix cleanup for sending messages | dimitri staessens | 2016-08-09 | 2 | -35/+8 | |
| | | | | Cleanup handlers were not always popped. | |||||
* | ipcpd: Correct cleanup of filedescriptor | dimitri staessens | 2016-08-09 | 1 | -2/+2 | |
| | ||||||
* | irmd: Close filedescriptor on exit | dimitri staessens | 2016-08-09 | 1 | -8/+10 | |
| | ||||||
* | lib: dev.c: Release locks before blocking calls | dimitri staessens | 2016-08-09 | 1 | -8/+23 | |
| | ||||||
* | Merged in dstaesse/ouroboros/be-bugfixing (pull request #195) | Sander Vrijders | 2016-08-09 | 6 | -26/+26 | |
|\ | | | | | | | lib: Various fixes | |||||
| * | lib: Various fixes | dimitri staessens | 2016-08-09 | 6 | -26/+26 | |
| | | | | | | | | | | Fix missing set of api upon flow_alloc. Various locking fixes. | |||||
* | | Merged in dstaesse/ouroboros/be-bugfixing (pull request #194) | Sander Vrijders | 2016-08-09 | 1 | -29/+20 | |
|\| | | | | | | | lib: shm_ap_rbuff: Fix cleanup of locks | |||||
| * | lib: shm_ap_rbuff: Fix cleanup of locks | dimitri staessens | 2016-08-09 | 1 | -29/+20 | |
| | | ||||||
* | | Merged in sandervrijders/ouroboros/be-enroll-fix (pull request #193) | dimitri staessens | 2016-08-09 | 1 | -9/+2 | |
|\ \ | |/ |/| | | | ipcpd: Fix locking of shim-eth-llc | |||||
| * | ipcpd: Fix locking of shim-eth-llc | Sander Vrijders | 2016-08-09 | 1 | -9/+2 | |
| | | | | | | | | | | | | This changes the amount of time the shim-eth-llc will hold the IPCP state lock when sending and receiving frames. Before it was holding the lock for too long. | |||||
* | | irmd: Fix bad locking | dimitri staessens | 2016-08-09 | 2 | -4/+1 | |
|/ | | | | Locks should be kept before calling cond_wait. | |||||
* | irmd: Refactor and bugfixes | dimitri staessens | 2016-08-08 | 8 | -304/+445 | |
| | | | | | | | Refactors the IRMd to extract reg_api and irm_flow structures to their own sources. Fixes some locking bugs. | |||||
* | ipcpd: Fix missing close of lsockfd | dimitri staessens | 2016-08-08 | 1 | -0/+3 | |
| | ||||||
* | tools: echo: Fix unnecessary cast | dimitri staessens | 2016-08-08 | 1 | -1/+1 | |
| | ||||||
* | lib: shm_ap_rbuff: Fix endless wait | dimitri staessens | 2016-08-08 | 1 | -17/+37 | |
| | | | | | When the rbuff was empty, the blocking read would wait forever for a read. | |||||
* | ipcpd: normal: Allow exchange of static DIF information | Sander Vrijders | 2016-08-08 | 11 | -161/+325 | |
| | | | | | | This adds the functionality of exchanging the static DIF information between 2 DIF members. After exchange the enrollment is stopped, and the IPCP that initiated enrollment transitions to the enrolled state. | |||||
* | lib, tools: Add POSIX version to source files | Sander Vrijders | 2016-08-08 | 3 | -0/+4 | |
| | | | | | Sometimes the POSIX version was not set in source files. This caused the compiler to not find the timespec struct, since we are using C89. | |||||
* | lib: shm_ap_rbuff: Remove obsolete clean function | dimitri staessens | 2016-08-06 | 1 | -15/+0 | |
| | ||||||
* | tools: oping: Fix wrong delay calculation | dimitri staessens | 2016-08-06 | 1 | -1/+2 | |
| | ||||||
* | lib: shm_du_map: Fix wrong padding | dimitri staessens | 2016-08-06 | 1 | -3/+3 | |
| | | | | Padding was miscalculated, causing lockups in the fast path. | |||||
* | irmd: registry: Fix double free | dimitri staessens | 2016-08-05 | 1 | -3/+0 | |
| | | | | The req_ae_name is freed when the unpacked message is cleaned up. | |||||
* | ipcpd: Fix memleak | dimitri staessens | 2016-08-05 | 3 | -0/+6 | |
| | | | | The entry from the ringbuffer was never freed, causing memleak per SDU. | |||||
* | lib:shm_du_map: Fix wrong element access | dimitri staessens | 2016-08-05 | 1 | -4/+5 | |
| | | | | The check whether the du map is empty should be first. | |||||
* | lib: shm_du_map: Fix disabling MULTI_BLOCK | dimitri staessens | 2016-08-04 | 1 | -2/+2 | |
| | ||||||
* | lib: ipcp.c: Fix memleak upon cancellation | dimitri staessens | 2016-08-04 | 1 | -8/+14 | |
| | ||||||
* | tools: oping: Fix client cleanup | dimitri staessens | 2016-08-04 | 1 | -0/+5 | |
| | ||||||
* | lib: sockets: Fix cancellation cleanup | dimitri staessens | 2016-08-04 | 1 | -2/+13 | |
| | | | | | When cancelled in read() the fd should be closed and the allocated memory freed. | |||||
* | tools: oping: Fix cleanup on exit | dimitri staessens | 2016-08-04 | 1 | -3/+5 | |
| | | | | The server should cleanup whenever the listen thread exits. | |||||
* | tools: cbr: Fix server cleanup | dimitri staessens | 2016-08-04 | 1 | -15/+7 | |
| | | | | The server should cleanup whenever the listen thread exits. | |||||
* | lib: shm_ap_rbuff: Fix memleak | dimitri staessens | 2016-08-04 | 1 | -0/+1 | |
| | ||||||
* | lib: shm_du_map: Fix memory leak | dimitri staessens | 2016-08-04 | 1 | -3/+2 | |
| | ||||||
* | tools: cbr: Fix missing check on ap_init | dimitri staessens | 2016-08-04 | 1 | -2/+4 | |
| | | | | If ap_init() fails, the program should exit. | |||||
* | tools: cbr: Fix options | dimitri staessens | 2016-08-04 | 2 | -10/+10 | |
| | | | | The server apn is a client-only option. Also fixes some indentation. | |||||
* | irmd: Remove ipcp type from registration message | dimitri staessens | 2016-08-04 | 1 | -2/+2 | |
| | ||||||
* | tools: oping: Fix division by zero | dimitri staessens | 2016-08-04 | 3 | -14/+19 | |
| | | | | | When killing the oping client before it sent an SDU, some bad operations were being performed. | |||||
* | irmd: Fix potential object access after free | dimitri staessens | 2016-08-04 | 1 | -0/+1 | |
| | ||||||
* | irmd: Fix destroying allocated flows | dimitri staessens | 2016-08-04 | 2 | -8/+17 | |
| | | | | | When a flow was in FLOW_ALLOCATED state, it would not change to NULL state and irm_flow_destroy would hang forever. | |||||
* | ipcpd: Fix memory leaks | dimitri staessens | 2016-08-04 | 7 | -45/+56 | |
| | | | | ipcp-data was not correctly destroyed. | |||||
* | ipcpd: Fix incorrect cleanup of fd | dimitri staessens | 2016-08-04 | 1 | -2/+7 | |
| | | | | The cleanup was trying to close a pointer to the file descriptor. | |||||
* | lib: Revise blocking I/O | dimitri staessens | 2016-08-03 | 4 | -129/+281 | |
| | | | | | | | | | Blocking I/O now uses condition variables in the shared memory instead of busy waiting. Timeouts can be specified. This requires the size of the rbuffs and du_map to be the same, to guarantee that when the shm_du_map is not full, the ap_rbuffs can't be full either. Added the timeout option to the flow for future use. | |||||
* | tools: Add a ping application | dimitri staessens | 2016-08-03 | 5 | -0/+594 | |
| | | | | | Ouoroboros ping application mimics the GNU ping application. do "oping --help" for options. | |||||
* | lib: dev: Add select-like call | dimitri staessens | 2016-08-03 | 3 | -66/+125 | |
| | | | | | | | This adds a flow_select() call that will sleep until an SDU can be read on a flow. It returns the file descriptor for which an SDU is ready. It takes as optional argument a timespec struct to specify a timeout. | |||||
* | Merged in dstaesse/ouroboros/be-llc (pull request #183) | Sander Vrijders | 2016-08-01 | 1 | -75/+79 | |
|\ | | | | | | | shim-eth-llc: Fix for bad drivers | |||||
| * | shim-eth-llc: Fix for bad drivers | dimitri staessens | 2016-08-01 | 1 | -75/+79 | |
| | | | | | | | | | | | | | | | | It seems like drivers are setting the Ethernet length field wrong when sending an LLC message. The LLC shim now writes the payload length in the frame to circumvent the wrong information from the driver. Also fixes deallocation. | |||||
* | | Merged in sandervrijders/ouroboros/be-enrolment (pull request #182) | dimitri staessens | 2016-08-01 | 13 | -94/+453 | |
|\ \ | |/ |/| | | | Be enrolment |