summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* ipcpd: Integrate raptor into ouroboros 0.8Dimitri Staessens2017-12-1510-443/+419
| | | | | | | | | | | The raptor code is refactored to completely remove reduntant code relating to addresses. The dependency on the google protocol buffers is removed. The build system will only build raptor if the relevant kernel module is found on the system. The irm tool and the relevant documentation are updated. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Add support for raptor FPGA deviceAlexander D'hoore2017-12-154-0/+1206
| | | | | | | | | This adds the initial raptor support from the master thesis entitled (dutch) "Implementatie van de Recursive Internet Architecture op een FPGA platform". Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Wait for DHT running state before accessDimitri Staessens2017-12-061-12/+13
| | | | | | | | | | The DHT will now wait for the DHT_RUNNING state before handling request messages, ensuring that the parameters are set correctly before they are read and sent at JOIN. Also fixes a re-introduced assertion failure when a JOIN fails. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Fix missing bracket in DHTDimitri Staessens2017-12-051-1/+1
| | | | | | | | ca2f47d had a missing bracket after a last-minute update to simplify the logic. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Update DHT buckets in JOINING stateDimitri Staessens2017-12-051-7/+31
| | | | | | | | | | The DHT buckets were not updated in JOINING state, which sometimes caused the DHT to miss its first peer update when joining and become isolated. This also checks if a name is already registered to avoid adding it multiple times. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Make hash algorithm configurableSander Vrijders2017-12-042-28/+34
| | | | | | | | | | This makes the hashing algorithm configurable once more for the local and the shim-eth-llc, since their scope is so small that it is up to the network administrator to select a correct algorithm for the whole network. Signed-off-by: Sander Vrijders <[email protected]> Signed-off-by: Dimitri Staessens <[email protected]>
* ipcpd: Remove protocol buffers from flow allocator0.8Sander Vrijders2017-12-043-117/+54
| | | | | | | | This will remove google protocol buffers from the flow allocator component in the normal IPCP. It now uses packed structs, as supported by the compilers of choice. Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Remove protocol buffers from link stateSander Vrijders2017-12-043-66/+24
| | | | | | | | | This will remove google protocol buffers from the link state routing algorithm in the normal IPCP. It now uses packed structs, as supported by the compilers of choice. Signed-off-by: Sander Vrijders <[email protected]> Signed-off-by: Dimitri Staessens <[email protected]>
* ipcpd: Remove protocol buffers from shim-udpSander Vrijders2017-12-043-104/+69
| | | | | | | | This will remove google protocol buffers from the shim-udp. It now uses packed structs, as supported by the compilers of choice. Signed-off-by: Sander Vrijders <[email protected]> Signed-off-by: Dimitri Staessens <[email protected]>
* ipcpd: Remove protocol buffers from shim-eth-llcSander Vrijders2017-12-043-122/+99
| | | | | | | | This will remove google protocol buffers from the shim-eth-llc. It now uses packed structs, as supported by the compilers of choice. Signed-off-by: Sander Vrijders <[email protected]> Signed-off-by: Dimitri Staessens <[email protected]>
* lib, tools: Rename application process and instanceDimitri Staessens2017-12-0267-1177/+1204
| | | | | | | | | | | This refactors ouroboros to use "program" instead of "application process" and "process" instead of "application process instance" to align with current naming in current Operating Systems courses instead of the ISO nomenclature adopted by RINA. This change permeates through the entire implementation. Also contains some minor other refactors. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* irmd: Don't read commands under lockDimitri Staessens2017-11-301-3/+3
| | | | | | | | | The commands were read under a mutex. Since the cmd struct was just malloc'd, this is not needed. Also fixes closing the socket if the malloc fails. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Don't read commands under lockDimitri Staessens2017-11-301-3/+3
| | | | | | | | | The commands were read under a mutex. Since the cmd struct was just malloc'd, this is not needed. Also fixes closing the socket if the malloc fails. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* lib: Refactor shm_rdrbuff and shm_rbuffDimitri Staessens2017-11-304-497/+286
| | | | | | | | | | This refactors the creation of shm_rdrbuff and shm_rbuff elements. All cleanup is now handled and the common code between the _open() and _create() calls is moved to a static function. Common code between the pthread and lockless rbuff implementations was moved to shm_rbuff.c Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* irmd: Set correct clock for irm_flow condvarDimitri Staessens2017-11-291-21/+31
| | | | | | | | | The irm_flow condition variable was not initialized to the correct clock, resulting in bad timedwaits when a flow was allocated with a timeout. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd, irmd: Bugfix for failed accept() callsDimitri Staessens2017-11-292-4/+2
| | | | | | | | | The ipcp will not have the rx_rb set for a pending flow, so it should not call shm_rbuff_fini. The irmd should not call ipcp_flow_dealloc for flows that are still pending flow allocation. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Remove unnecessary linker directivesDimitri Staessens2017-11-292-3/+3
| | | | | | | The ouroboros-dev library links against ouroboros-common. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Don't bind from the IPCPDimitri Staessens2017-11-295-30/+71
| | | | | | | | | | The binding of the normal IPCP to its name is moved from the source code to the irm tool introducing the "autobind" option for the bootstrap and enroll commands. With this option, the IPCP will be bound to the IPCP name and the DIF name automatically. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Only update DHT buckets in running stateDimitri Staessens2017-11-281-1/+2
| | | | | | | This fixes accessing buckets when a DHT join fails. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* irmd: Fix false alarm when registering before bindDimitri Staessens2017-11-271-1/+1
| | | | | | | | When registering a name in a dif before binding a program to it without auto, the IRMd would erroneously log that the binding failed. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* tools: Fix SEGV in oping clientDimitri Staessens2017-11-221-4/+8
| | | | | | | | | | | The oping client tried to cancel non-created pthreads if it was killed during flow allocation, which caused a SEGV. The threads are now stopped using a variable. Fixes bug #2. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Install system-wide on OS XDimitri Staessens2017-11-201-1/+1
| | | | | | | | OS X should disable the system integrity protection system so ouroboros can be install system-wide. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* lib: Fix init and fini ELF sections for OS XDimitri Staessens2017-11-201-2/+10
| | | | | | | OS X uses a different syntax for specifying ELF sections. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* lib: Add library initializers to .init_arrayDimitri Staessens2017-11-201-4/+7
| | | | | | | | | | The library initializers with argc and argv arguments were specified using the __attribute__((constructor)). This is undefined behaviour as constructors should have no arguments. This is fixed by passing the functions to the .init_array. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Unset SHIM_ETH_LLC_TARGETDimitri Staessens2017-11-191-0/+1
| | | | | | | | This fixes the IRMd thinking a shim-eth-llc is available after it is disabled by configuring Ouroboros using ccmake. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* tools: Send time in oping packetsDimitri Staessens2017-11-192-38/+23
| | | | | | | | The oping tool now stores the time it sent the packet inside the SDU, simplifying the implementation. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Define HAVE_LIBGCRYPT for IPCPsDimitri Staessens2017-11-181-0/+2
| | | | | | | | The HAVE_LIBGCRYPT variable was not defined for IPCPs, causing them to call the gcrypt functions with the wrong values for the enum. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Fix setting configuration definitionsDimitri Staessens2017-11-183-2/+14
| | | | | | | | Variable in cache need to be explicitly unset else they will remain in the cache and still be defined by #cmakedefine. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Set correct rpath in non-sandboxed buildSander Vrijders2017-11-161-3/+9
| | | | | | | | After the previous commits, the rpath was not being correctly set anymore for the sandboxed and non-sandboxed build. Signed-off-by: Sander Vrijders <[email protected]> Signed-off-by: Dimitri Staessens <[email protected]>
* build: Fix setting default build typeSander Vrijders2017-11-151-3/+4
| | | | | | | | This adds a check and sets the FORCE parameter to override the cmake default build type. Signed-off-by: Sander Vrijders <[email protected]> Signed-off-by: Dimitri Staessens <[email protected]>
* build: Deprecate the compile/install scriptsDimitri Staessens2017-11-154-102/+0
| | | | | | | | | | | | | | | | The build system updates have made installing simple and robust using the following commands: mkdir build && cd build cmake .. ccmake . sudo make install Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* tools: Check process id when binding a processSander Vrijders2017-11-151-1/+11
| | | | | | | | This adds a check that the passed process id is in fact a number and not gibberish. Signed-off-by: Sander Vrijders <[email protected]> Signed-off-by: Dimitri Staessens <[email protected]>
* ipcpd: Add return statement in eth-llc loopDimitri Staessens2017-11-151-0/+2
| | | | | | | | Some compiler versions complained about a missing return in non-void function. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Set defaults for prefix and build typeDimitri Staessens2017-11-151-0/+8
| | | | | | | | This defaults the CMAKE install prefix to "/". It also sets the default build type to "Release". Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Reset *dist on graph_routing_table failDimitri Staessens2017-11-151-0/+1
| | | | | | | | | | The *dist pointer was not reset on failure, causing the caller to try to free an unmalloced or already freed pointer. Reported-by: Nick Aerts <[email protected]> Tested-by: Nick Aerts <[email protected]> Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Set CMAKE_REQUIRED_FLAGS in CompilerUtilsSander Vrijders2017-11-142-3/+4
| | | | | | | | | This sets the cmake variable CMAKE_REQUIRED_FLAGS to the compiler flags that are being tested so that the linker doesn't fail when trying the compile flag. Signed-off-by: Sander Vrijders <[email protected]> Signed-off-by: Dimitri Staessens <[email protected]>
* build: Add options for debug builds with sanitizerDimitri Staessens2017-11-147-12/+19
| | | | | | | | | This adds three build types, DebugASan, DebugTSan and DebugLSan that enable the Address, Thread and Leak Sanitizer by setting the fsanitize flag to the compiler. This option is supported by both gcc and clang. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Set libraries for OpenSSL when not foundDimitri Staessens2017-11-141-0/+1
| | | | | | | | | The build didn't reset the CMake variable for OPENSSL_LIBRARIES, causing bad input to the linker. This error showed up when trying to build on XUbuntu 17.10 without OpenSSL or libgrcypt installed. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* irmd: Don't access entry after unlockDimitri Staessens2017-11-141-1/+1
| | | | | | | | When bootstrapping an IPCP entry->api variable was accessed after the registry was unlocked for printing and info message. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Avoid duplicate definitions in librariesDimitri Staessens2017-11-131-4/+2
| | | | | | | | There were sources defined in multiple libraries. This issue was reported by the address sanitizer on FreeBSD. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Set compiler optimization for release buildDimitri Staessens2017-11-131-0/+3
| | | | | | | This automatically sets the -O3 compiler flag for release builds. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Fix detection of python for swig bindingsDimitri Staessens2017-11-101-3/+6
| | | | | | | | | | The cmake build only looked for the libraries but not the interpreter, which caused the build to fail if the library was present, but the interpreter was not installed and executable using the "python" command. The build now looks for the python interpreter as well. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Fix unused label with netmap for LLC shimDimitri Staessens2017-11-101-1/+3
| | | | | | | | When the netmap driver is used for the LLC shim, the label fail_device: was never used, resulting in a compiler error. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* lib: Use correct names for st_modes in RIBDimitri Staessens2017-11-101-6/+8
| | | | | | | | | The RIB implementation was using internal variables from the linux implementation instead of the POSIX defined macros. The POSIX defined macros require defining __USE_XOPEN to access them. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* lib: Fix FRCT sequence number network endiannessDimitri Staessens2017-11-092-4/+9
| | | | | | | | Endianness conversion from/to the network for the sequence number was missing in 68694bc. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* lib: Use packed struct for FRCT header accessDimitri Staessens2017-11-087-261/+142
| | | | | | | | | | | | | | | This replaces the variable FRCT header with a packed struct, which significantly simplifies the implementation. The shm_du_buff calls to release the head/tail are updated to return a pointer to the original head or the new tail (in symmetry to the alloc calls, which return a pointer to the new head and old tail), so that it immediately points to the structure that is needed. The frct_pci sources are removed and frct is now fully in the frct.c source file. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: List adjacencies in link-state databaseDimitri Staessens2017-11-071-4/+27
| | | | | | | The lsdb will now list management and data transfer adjacencies. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* ipcpd: Create new SDUs for DHT join retriesDimitri Staessens2017-11-071-15/+15
| | | | | | | | | | The DHT join operation left an SDU in the rdrbuff for a couple of seconds during retries, which may cause the system to block if there is heavy traffic. This patch releases the sdb when the write fails and creates a new packet for every retry. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* lib: Refactor FRCT implementationDimitri Staessens2017-11-077-546/+572
| | | | | | | | | | | | | | | | | | | | The frct_pci and rq headers are moved from include/ouroboros to src/lib since they are only needed in the library. FRCT is moved to its own source file. FRCT takes the application PDUs, encapsulates and processes them and hands them back. This makes it easier to disable FRCT should the application want to write to a "raw" flow. An FRCT instance is now allocated upon alloc and released upon dealloc. The FRCT data structure is split into a sender and receiver connection record. Setting a new configuration will now be done upon sending the next data PDU, which will flag the DRF for a new run and use that configuration. This avoids some issues should packets arrive out-of-order, and simplifies setting a configuration. Signed-off-by: Dimitri Staessens <[email protected]> Signed-off-by: Sander Vrijders <[email protected]>
* build: Improve bash scriptsSander Vrijders2017-10-304-13/+18
| | | | | | | | This improves the bash scripts so they are less error prone. Mistakes were found using the parser on shellcheck.net Signed-off-by: Sander Vrijders <[email protected]> Signed-off-by: Dimitri Staessens <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-01 18:05:07 +0000