aboutsummaryrefslogtreecommitdiffstats
path: root/src
AgeCommit message (Collapse)AuthorFilesLines
2019-07-18Tag broadcasted events with UUID and hopJosé Bollo9-38/+138
When API have mutual dependencies, leading to loops in dependecies, broadcasting an event never ends because of the loop (see SPEC-2625). To avoid that weird flood of events, a unic identifier (UUID) is attached to broadcasted event and a tiny memory records previously broadcasted events to avoid re-sending an already sent event. The size of the memory can be set using the macro variable EVENT_BROADCAST_MEMORY_COUNT whose default value is 8. It can be reduced to 0. An other mecanism is added to limit the count of hops that a broadcasted event can do. That count can be set using the macro variable EVENT_BROADCAST_HOP_MAX whose default value is 10. Bug-AGL: SPEC-2625 Signed-off-by: José Bollo <jose.bollo@iot.bzh> Change-Id: I45877583dbf478a79d405c3650880a5579ac1f9e
2019-07-18afb-evt: Refactor processing of broadcasted eventsJosé Bollo6-83/+68
Solving the bug SPEC-2625 needs to rework the broadcasting of events. It appeared that the numerical event identifier passed for broadcast wasn't used by called function except for hooking. Suppressing it introduces a clear distinction between the push and the broadcast paths. The file afb-ws-json1 is changed to avoid casting of functions. Bug-AGL: SPEC-2625 Signed-off-by: José Bollo <jose.bollo@iot.bzh> Change-Id: I7dadfb8702457d6af797ec5f542add0a83224e3f
2019-07-18uuid: Add module for UUID generationJosé Bollo4-70/+125
This commit allow modules of afb-daemon to request UUID. This introduce an abstraction above libuuid for 3 reasons: 1. Using uuid had bad side effect inthe past because it required enought entropy at start time, leading to hanging initialisations in wait for enough entropy. 2. Solving bug SPEC-2625 will require use of UUID. 3. Porting to systems that haven't libuuid is made more easy after that. Bug-AGL: SPEC-2625 Signed-off-by: José Bollo <jose.bollo@iot.bzh> Change-Id: I20b58cf074ca88994de049acbeb8f827359e1b69
2019-07-12afb-xreq: Forbids (un)subscribes after replyJose Bollo1-12/+22
Allowing to subscribe to a request that was replied was possible and lead to problem revealed by issues SPEC-2542 and SPEC-2599 (these issues are more related to processing unordered incoming messages). The choice was - fix the bug and authorize (un)subscribe after reply - forbids to (un)subscribe after reply Second solution was chosen for its simplicity and its networking efficiency. Bug-AGL: SPEC-2542 Bug-AGL: SPEC-2599 Change-Id: I09f48b760b1fd6f70d42b80df3c8053696c45966 Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2019-07-12Revert "afb-stub-ws: Allow unordered process of messages"José Bollo1-1/+1
This reverts commit 5425e054fbf87fe6d024103f46e53f2a28e074f2. The change introduced in the reverted commit is no more necessary since events are sent asynchronousely. Bug-AGL: SPEC-2215 Bug-AGL: SPEC-2219 Bug-AGL: SPEC-2542 Bug-AGL: SPEC-2599 Signed-off-by: Jose Bollo <jose.bollo@iot.bzh> Change-Id: I320e13477351ff8f9b8ad12fa21234cc0823b7f4
2019-07-12afb-evt: send events in separate jobJose Bollo2-39/+226
Sending events in the context of the calling process had the consequence that the ordering of the messages had to be removed (see SPEC-2215 & SPEC-2219). This was not good by nature and lead to issues SPEC-2542 and SPEC-2599. Sending events in the context of the calling process also implies to delay the calling process. For this reasons, sending events is now done in an other jobs. For that reason, the count of allowed pending jobs is increased to 100 (was 50). Bug-AGL: SPEC-2215 Bug-AGL: SPEC-2219 Bug-AGL: SPEC-2542 Bug-AGL: SPEC-2599 Change-Id: I5b56d952cc187b65ad6eb9344ad74e5e8d3b7540 Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2019-04-04afb-export: Fix heap corruptionguppy_7.0.1guppy/7.0.17.0.1José Bollo1-1/+1
The allocated size wasn't correct when 'path' was used. The effect be unseen depending on size and alignment of heap chunks. Change-Id: Id9f85a0078028319491b51bc3e11cd51e4bc3755 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2019-04-02Fix false ***buffer overflow*** detectionJosé Bollo11-28/+28
The compiling option __FORTIFY_SOURCE=2 introduced a false ***buffer overflow*** detection when the flexible array 'pattern' was initilized in globset. The compiler is only complaining when the array is in a struct that is in a struct like struct { ...; struct { ...; char name[1]; }} To avoid these false detections, it is enougth to ellipsese the dimension of the array. Seems to be the now standard way of declaring flexible arrays when it was before an extension. So now: struct { ...; struct { ...; char name[]; }} works even when __FORTIFY_SOURCE=2. Bug-AGL: SPEC-2292 Change-Id: I4b4a5df505a5357f92b9ab1657175911198ca582 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2019-03-22jobs: Ensure validy of event loopFrederic Marec1-0/+3
In rare circumstances during intialization threads should wait instead of using a null event loop. This is because no one has needed event loop at this stage of initialization. Change-Id: I05b6ddbc05bec50958d0981213219b1d641401a5 Signed-off-by: Frederic Marec <frederic.marec@iot.bzh>
2019-03-12afb-stub-ws: Allow unordered process of messagesJosé Bollo1-1/+1
Requests are already sequenced on need (noconcurrency flag of bindings). Sequencing messages until their completion has a side effect when dealing with events and synchronous calls: event -> call -> reply locked until event completion (see SPEC-2219). This change could be reverted if events are converted to jobs (see SPEC-2215). Bug-AGL: SPEC-2219 Change-Id: Ia577bc54ba38291817fd5ed3e3dda1b67b2818b7 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2019-02-08afb-stub-ws: Fix concurrent memory issuesguppy_7.0.0guppy/7.0.07.0.0José Bollo1-3/+4
This changes takes care to increment the the count of reference before using it. Bug-AGL: SPEC-2163 Change-Id: Ia7882427eeae933eeb5030aad025ebb1da129d6f Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2019-02-07jobs: Ensure releasing event loopguppy_6.99.5guppy/6.99.56.99.5Jose Bollo1-1/+2
The internal functions blocking threads for implementing call_sync were not releasing the event loop, leading to deafness of the binder. Include a tuto-4 that reproduce the issue on a binder with the bug and that also shows interesting usages of the binder. Bug-AGL: SPEC-2161 Change-Id: I83ad4d55d721a6046e798a5e06967df4dd5a7284 Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2019-01-18afb-hsrv: Avoid call to getnameinfoJosé Bollo1-6/+12
Calling getnameinfo may block the program a long time (5s) when address is zero. Bug-AGL: SPEC-2133 Change-Id: Id52b2989c5e996363fd4b6f39049f059a6ee97a2 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2019-01-16jobs: Remove conditionnal REMOVE_SYSTEMD_EVENTJose Bollo3-128/+0
The removal of the systemd event loop will be made in later changes. Change-Id: Ia6c52feb4969f360c31cbfc50ae991f1767f007d Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2019-01-16jobs: Separate internal threads from othersJosé Bollo1-61/+92
Foreign threads, that are not started by jobs, are allowed to use synchronous jobs_call/job_leave (directly or indirectly). This commit ensure that those foreign thread will neither acquire the eventloop nor execute jobs. Includes a tiny cleanup of remain usage. Bug-AGL: SPEC-2089 Change-Id: I2ad7fcfe2c276e34bdc4ec0c2aa3a4207bea1854 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2019-01-16jobs: Refactor event loop handlingJosé Bollo3-125/+184
This improves the arbitration of the single event loop across threads. Before introduction of using 'evenfd' there was several event loop. At the current time, there is only one. At the end, there will probably remain only one. Bug-AGL: SPEC-2089 Change-Id: Iac9db7cbe15b4c9c76e6e9a8f6e641ed2a9039e0 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2019-01-16afb-proto-ws: Serialize incoming messageJosé Bollo3-7/+6
Enforce serialisation for a connection Bug-AGL: SPEC-2089 Change-Id: Id9f261b7cc02fda78922dc511856c34b7c5bf56d Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2019-01-04afb-api-v3: Fix potential buffer overflowJose Bollo1-4/+5
Change-Id: I170e127ebf96d2accfdd6d6a4ec322afeaa2782f Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2018-12-31jobs: Fix a lock of call syncJose Bollo1-24/+59
Sometime the function jobs_leave missed to wake up the process that expected the result. It happened when that waiting thread was locked in the epoll. That commit unlock it. Bug-AGL: SPEC-2089 Change-Id: I6ab6a60a8c9ea3639143f282f90c7b77ed16853a Signed-off-by: Jose Bollo <jose.bollo@iot.bzh> Signed-off-by: Clément Bénier <clement.benier@iot.bzh>
2018-12-14wrap-json: Upgrade to latest versionJosé Bollo1-18/+19
Change-Id: I3eb4e922cb5106bcf8a1ccc5e1b010bd0a60bf34 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-12-13afb-hsrv: Prepare selection of listening interfacesJose Bollo4-14/+139
The listening socket will no more be the default one but will be the selected ones, with ability to select more than just one listening interface. Nevertheless, the behavior is not changed at the moment. The default interface (0.0.0.0) is still used. A further change will add the ability to select the listening interfaces. That evolution depends on a fix of libmicrohttpd to enable at the same time epoll and no-listen-fd. Bug-AGL: SPEC-1833 Bug-AGL: SPEC-1016 Change-Id: Ia78671cbee90a243ba7b2c724b6155cffbde6920 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-12-13afb-trace: Make real timestampsJosé Bollo1-10/+0
The optimisation that represent the timestamp as a fake double is not accurate in local context. Bug-AGL: SPEC-2051 Change-Id: I285837251507a69ad210823020d0cf69ad5999f3 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-12-05afb-proto-ws: Add error reportJose Bollo3-1/+15
Change-Id: I58c88f8bcaf4cfb8a53b58eeefd7fa3415bf894a Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2018-12-05afb-hook: Improve reentrancy of hookingJosé Bollo1-126/+163
Work on testing showed that a deadlock was occuring in management of hooks. Change-Id: Ib51eb4f0b9ffc5d9dfe2770f2c3f8f47f262b60f Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-11-22globset: Introduce globset for event handlingguppy_6.99.1guppy/6.99.16.99.1Jose Bollo9-87/+693
It optimises the event handling that was slow. It also fix few bugs: - at most one event handler is called now - the handler call has the most specific pattern Change-Id: Ic13a0258b5743579ab15e0e953ec62206d982850 Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2018-11-16afb-socket: Fix address reuseJosé Bollo1-7/+7
Change-Id: I3ae51611b92762a4f09b1f6cca6ff27853cd9f95 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-11-16afb-export: Fix error in event handlingJosé Bollo1-1/+1
Change-Id: I2869a2a840fb07550d493ddc2d651358892d33d1 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-11-08main-afb-daemon: Export variables to connect to binderJosé Bollo1-12/+42
Two new variables are exported that will allow to connect to the binder: - AFB_PORT: the HTTP port that the binder listen - AFB_TOKEN: the initial token that the binder expects These variables are defined only if the HTTP server is started. They are available during initialisation of bindings, even if it is before the real start of the HTTP server. Change-Id: I1b7c8572e3d77b7036a32af47b2a2f1d95803a41 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-10-30afb-socket: Relax requirement of tcp ending slashJosé Bollo1-1/+1
The uri for tcp was requiring a slash (/) after the port. This change allows to not provide that slash. This is usefull for future implementation of HTTP server selection. Change-Id: I0a76d167065258d242845579b7c8728eefcc359f Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-10-30afb-socket: Handle listening on all interfacesJose Bollo1-0/+5
Change-Id: I909493166f98f152e5aeb02a72f40bc3c58b42b2 Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2018-10-26afb-hsrv: Adapt to next version of libmicrohttpdJosé Bollo1-1/+4
Change-Id: I5bef17178e972e73b653ddbff5a8b09cfc2287eb Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-10-23Added an option to color out the ouputLoïc Collignon4-6/+62
Enhance the readability using basic colorization to the logging system. As the '--call' option is almost unused, it doesn't require to have a shortname. So we use the '-c' and '--color' for the new colorization option, and keep only the longname for the 'call' option. Change-Id: I095fc9f38133fb742e0f0003540cd120feec8f5e Signed-off-by: Loïc Collignon <loic.collignon@iot.bzh>
2018-10-22afb-session: Add language to sessionJose Bollo2-0/+38
Bug-AGL: SPEC-1827 Change-Id: Icfb9e2ee76186ac677ff2fbb37be28cd83dd3b6b Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2018-10-16afb-apiset: Reorder priority of messagesJose Bollo1-2/+2
The message before starting has now a greater priority than the one stating that start completed. Expecting that in case of problem it wil be cleaner. Change-Id: I2c27eb70f218d0492291c3eab4f8396ab5a6c093 Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2018-10-12afb-hsrv: Lower the "no handler" messageJosé Bollo1-1/+1
Bored to see WARNING: Unhandled request to favicon.ico Change-Id: I7289b508f93bcd5ac4650ec32670ed0fd291044e Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-10-12APIv3: Allow to write application bindingJosé Bollo1-3/+6
This enable a binding version 3 to declare no API. This is used to start a job that will will run after initialisation. The tutorial tuto-app1 shows how. Run it with: afb-daemon --binding tuto-app1.so A further option could be add to close stdin even if running in foreground, as it was the case before. Change-Id: I2b384d125accb4642eed8e004642ba959326878f Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-10-12afb-api-so-v3: Keep root API in all casesJosé Bollo1-1/+8
For bindings v3, it is possible to have no explicit default root API. In that case, the binder creates a fake API that references the binding itself. That api isn't callable be it allows to emit messages, to catch events and to call services (other apis). It must be kept to ensure that there is no further error in the hypothetical case where it is used. Bug-AGL: SPEC-1812 Change-Id: I1a6c1e1390411a45521b9a6c1c386dc7807191df Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-10-10genskel: Fix issues for C++ bindings v3José Bollo1-13/+64
The C++ compiler is too much strict. It doesn't accept valid C because fields are in the wrong order or missing. - afb_verb_t generated structure invert vcbdata and session - missing provide_class, require_class, require_api in the afb_binding_v3 generated struct The signature of the declared function was wrong. - wrong verb's callback signature void fun(afb_req) instead of void fun(afb_req_t) This add the declaration of the (pre)initialization and onevent functions. - missing init function signation generation, or probably preinit and onevent (not tested those ones) Bug-AGL: SPEC-1806 Change-Id: I1abf8a29f831ecc80458edcd3bd094b2108dc227 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-09-24afb-export: Emit a warning on bad 'afb_require_api' callguppy_6.90.0guppy/6.90.06.90.0José Bollo1-0/+5
The call to 'afb_require_api' in a preinit was forbiden before binding v3 refactoring. Since v3 it is made possible by deferring the requirement to the initialisation step. But this may lead to bad behaviour if someone use the call to test whether an api exists or not. This commit introduce a message that could warn a developper raising that issue. Change-Id: Icc14c3887d70f73d87da0e664bb0e5e961867b4a Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-09-24afb-socket: API name cares of abstract UDSJose Bollo1-1/+3
Change-Id: Ie485672d8f66ff32ea20cb0eb313690ba380db4a Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2018-09-14afb-session: Fix an issue in managing sessionsJosé Bollo1-2/+2
Probably an old copy/paste error in the commit d45426257d5149c735e33e3055220625a919e7bc It doesn't affect version EE (Electric Eel). Bug-AGL: SPEC-1735 Change-Id: Ia1536e141ac12275000196a3218eb62d77b36a06 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-09-07afb-apiset: fixed the error message upon api open.flounder_5.99.6flounder/5.99.65.99.6Thierry Bultel1-2/+2
just a typo Change-Id: I2963c2568cffdc89f577e99f2280da6dcb732b04 Signed-off-by: Thierry Bultel <thierry.bultel@iot.bzh>
2018-09-06afb-session: Use ad-hoc pseudo-random uuidJose Bollo1-0/+50
The use of the generator given by the lib uuid implies the use of the system call 'getrandom' without the flag GRND_NONBLOCK. It has a weird effect on boot because many binders require the generation of uuid at the same time but getrandom is subject to starving. This introduced a tiny pseudo-random generation of UUID that does not implies the use of 'getrandom'. Bug-AGL: SPEC-1655 Change-Id: I5131072881d7a53f0edda9e36762985c96a04550 Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2018-09-05main-afb-daemon: Export API after initializationJose Bollo3-5/+9
Exporting API after initialization is a simple technic to avoid interleaving of incoming foreign calls during initialisation. Also a tiny fix in jobs.c and afb-socket.c. Bug-AGL: SPEC-1724 Change-Id: I59596256481c8afcd88755ec303bf7f881b55c12 Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2018-08-28Allow choice to trap or not program faultsJosé Bollo5-29/+110
The binder traps faults by default to recover as much as possible from unexpected errors. Developers may need to remove that behaviour to get core file. This introduces 2 ways to set or remove the trapping of faults: - the environment variable AFB_TRAP_FAULTS - the command line option --trap-faults Both take a boolean value set using on of the following words: 1 yes true on -- enable trapping of faults (the default) 0 no false off -- disable trapping of faults Bug-AGL: SPEC-1702 Change-Id: I6c6a3b933dbda9922a6079c390a601c533d49e55 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-08-28main-afb-daemon: Set AFB_ROOTDIR and AFB_WORKDIR in environmentJose Bollo2-3/+48
It may be difficult to retrieve the root directory from bindings and from process spawn by option --exec. In the same way, well identifying the workdir might interest programs or bindings. So, from now, the environment variables below are set: - AFB_ROOTDIR: identify the rootdir as set by --rootdir and in the context of AGL it will be the widget directory - AFB_WORKDIR: identify the workdir as set by --workdir To avoid any confusion, the function realpath is used to export absolute path names. Bug-AGL: SPEC-1694 Change-Id: Id272e009ca975e28aaab8b14fa2a98fbd2216e73 Signed-off-by: Jose Bollo <jose.bollo@iot.bzh> Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-08-27afb-hook-flags: Fix a compilation warningJose Bollo1-1/+1
Change-Id: Ic9d1cc44ef1bbf6db4afe8eb7016b05f4cddb9a4 Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>
2018-08-23jobs: Abort when systemd's event loop lostflounder_5.99.4flounder/5.99.45.99.4José Bollo1-1/+2
When sd_event_prepare returns an error there is currently no way to restore the event loop working. It happens when a callback of an sd_event is faulty (ex: SIGSEGV). In that case the monitoring of signals (sig-monitor) interfer badly with systemd. But at the the application firstly has error. To avoid flooding the journal with messages, aborting is the best solution. Bug-AGL: SPEC-1681 Change-Id: If01295f07b3c3fc8c8ec1ac5bf04840d42ee9774 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-08-23afb-stub-ws: Add reconnection of ws-clients on needJosé Bollo3-6/+67
This implementation detects deconnections and try to reconnect lazily on need. Bug-AGL: SPEC-1668 Change-Id: Ib2a20a4578f2da80afe1564c42de96c4aa250e64 Signed-off-by: José Bollo <jose.bollo@iot.bzh>
2018-08-23afb-stub-ws: Safe handling of deconnectionsJose Bollo3-134/+166
This commit also include many comments and improvements in naming of functions. Bug-AGL: SPEC-1668 Change-Id: I1b5dd95678d94e9edfca1c598c5697e90bb9e5bf Signed-off-by: Jose Bollo <jose.bollo@iot.bzh>