summaryrefslogtreecommitdiff
path: root/lib/tevent
AgeCommit message (Collapse)AuthorFilesLines
2009-08-29tevent: Fix a segfault upon the first signalVolker Lendecke1-0/+6
When the first signal arrives, tevent_common_signal_handler() crashed: "ev" is initialized to NULL, so the first "write(ev->pipe_fds[1], &c, 1);" dereferences NULL. Rusty, Tridge, please check. Also, can you tell me a bit more about the environment you tested this in? I'd be curious to see where this survived. Thanks, Volker
2009-08-28lib/tevent: close pipe_fds on event_context destructionRusty Russell1-0/+2
The "hack_fds" were never closed before; now they're inside event_context they should be closed when that is destroyed. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2009-08-28lib/tevent: handle tevent_common_add_signal on different event contexts.Rusty Russell2-14/+31
I don't know if this is a problem in real life. The code assumes there's only one tevent_context; all signals will notify the first event context. That's counter-intuitive if you ever use more than one, and there's nothing else in this code which prevents it AFAICT. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2009-08-28lib/tevent: fix race with signals and tevent_common_add_signalRusty Russell1-20/+20
We carefully preserve the old signal handler, but we replace it before we've set up everything; in particular, if we fail setting up the pipe_hack we could write a NUL char to stdout (fd 0), instead of calling the old signal handler. Replace the signal handler as the very last thing we do. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2009-08-28lib/tevent: remove spectacularly complicated manual subtractionRusty Russell1-4/+1
To be completely honest, I don't quite know whether to laugh or cry at this one: 1 + (0xFFFFFFFF & ~(s.seen - s.count)) == 1 + (~(s.seen - s.count)) # s.seen, s.count are uint32_t == s.count - s.seen # -A == ~A + 1 Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2009-08-24tevent: avoid using reserved c++ word.Günther Deschner1-1/+1
Guenther
2009-08-21Fix for bug 6651 - smbd SIGSEGV when breaking oplocks.Jeremy Allison1-17/+54
Based on a patch submitted by Petr Vandrovec <petr@vandrovec.name>. Multiple pending signals with siginfo_t's weren't being handled correctly leading to smbd abort with kernel oplock signals. Jeremy
2009-08-19tevent: fix a commentMichael Adam1-1/+1
Michael
2009-08-17tevent: change version to 0.9.7 after adding tevent_req_cancel infrastructureStefan Metzmacher1-1/+1
metze
2009-08-17tevent: add tevent_req_cancel() infrastructureStefan Metzmacher3-0/+70
This offers a generic way for callers to cancel an async request. metze
2009-08-15tevent: add some more doxygen comments for tevent_req functionsStefan Metzmacher1-0/+38
metze
2009-07-16tevent: try to fix the build on QNX qnx18 6.4.1 it doesn't have SA_RESTART ↵Stefan Metzmacher1-1/+11
defined metze
2009-07-14Change to talloc_zero_size instead of extra memset.Jeremy Allison1-2/+1
Jeremy.
2009-07-14When tallocing a memory block for the state in a tevent_req struct,Jeremy Allison1-0/+1
ensure it's zeroed out. Vl & Metze please check. Jeremy.
2009-07-03Sort the signature filesSimo Sorce2-45/+45
2009-07-03Expose functions need by backend writersSimo Sorce5-52/+67
move publicly needed structures and functions in the public header. Stop installing internal headers. Update the signature and exports files with the new exposed function.
2009-06-18Revert "For tevent to install tevent_util.h"Simo Sorce1-1/+0
This reverts commit b112cc5503350b248949bdbcce8072f5523ce877. tevent_util.h is a private header. Must not be installed.
2009-06-18For tevent to install tevent_util.hEric Sandall1-0/+1
Patch for bug #6270 This patch is for the future when samba4 builds using external libraries. With this patch, tevent now installs tevent_util.h which is required by samba4.
2009-06-15Add exports file and abi checker for teventSimo Sorce7-2/+157
This is a first attempt at exporting symbols only for public functions We also provide a rudimentary ABI checker that tries to check that function signatures are not changed by mistake. Given our use of macros this is not an API checker. It's all based on tevent.h contents and the gcc -aux-info option
2009-06-04Increase tevent version for tevent_req_notify_callback()Volker Lendecke1-1/+1
2009-06-04Add tevent_req_notify_callbackVolker Lendecke2-4/+13
This is necessary for requests that have multiple results. Examples would be SMBEcho and ldap_search.
2009-05-19tevent/python: Makefile was still trying to build some non AC_SUBST python ↵Jelmer Vernooij1-7/+1
targets Signed-Off-By: Jelmer Vernooij <jelmer@samba.org>
2009-05-16tevent: Define TALLOC_FREE() if it's not defined yet, to allow buildingJelmer Vernooij1-0/+6
with released versions of talloc.
2009-05-16tevent: Remove python module.Jelmer Vernooij6-219/+0
This module didn't have any functionality that we actually used yet, and it was quite small. Tevent is quite low level and perhaps doesn't make much sense to expose directly as a Python module. It was also causing build problems when used with a system-tevent. We can always back later if necessary.
2009-05-16tevent: Install tevent_internal.h in the standalone build.Jelmer Vernooij2-1/+4
This is not ideal, but at least it fixes the build of samba-gtk for now. I've also added a warning about API guarantees at the top of the header.
2009-05-16Update copies of config.guess and config.sub.Jelmer Vernooij2-76/+282
2009-05-16tevent: Don't install headers, since we don't install a shared libJelmer Vernooij1-5/+0
either (from Samba).
2009-05-07tevent: fix typo async_req_done() => tevent_req_done()Stefan Metzmacher1-1/+1
metze
2009-03-19tevent: fix the nesting logicStefan Metzmacher1-7/+7
Only tevent_loop_once and tevent_loop_until() should care about the nesting level. This fixes the samba3 printing code where we use tevent_loop_wait() and don't allow nested events. We still call the nesting hook for all levels, we need to decide if we really want this... metze
2009-03-19fixed a logic bug in the tevent nesting codeAndrew Tridgell1-1/+13
The event nesting code never triggered as nesting.level was never greater than 1. The main event loop needs to increase the nesting level by 1. I also added a paranoia check to the nesting setup call. The API as currently written cannot support multiple nesting hooks, so we need to abort if multiple hooks are tried.
2009-03-17tevent: change version to 0.9.5 after the ABI has changedStefan Metzmacher1-1/+1
metze
2009-03-17tevent: store the location where a request was finishedStefan Metzmacher3-19/+48
This is very useful to find bugs. You can use 'p *req' in gdb to show where tevent_req_done(), tevent_req_error() or tevent_req_nomem() was called. metze
2009-03-17tevent: use an immediate event as trigger for tevent_queueStefan Metzmacher2-42/+60
metze
2009-03-17tevent: use an immediate event fot tevent_req_post()Stefan Metzmacher2-19/+13
Now tevent_req_post() never fails metze
2009-03-17tevent: use TALLOC_FREE() in tevent_req.cStefan Metzmacher1-9/+5
metze
2009-03-17tevent: add support for immediate eventsStefan Metzmacher8-34/+295
They're like directly triggered timed events, but you can preallocated them and scheduling them will not fail. metze
2009-03-17tevent: add tevent_common_loop_wait() helper function and use itStefan Metzmacher5-55/+33
tevent_loop_wait should do the same for all backends. It should loop as long as we have pending events. metze
2009-03-17tevent: check signal events first in event_loop_onceStefan Metzmacher3-0/+15
We also check for signal events directly before and after the select/epoll calls. metze
2009-03-17tevent: let tevent_loop_once() just run one fd eventStefan Metzmacher3-45/+7
This makes the logic much simpler for the callers, and matches the samba3 behavior. If needed we can add performance tunning for tevent_loop_wait() later. metze
2009-03-13tevent: change version to 0.9.4 as the ABI has changedStefan Metzmacher1-1/+1
metze
2009-03-12tevent: Fix might be unitialized warningTim Prouty1-1/+1
2009-03-12tevent: add tevent_loop_until()Stefan Metzmacher2-0/+68
This is only a hack for samba4 and should not be used in new code. metze
2009-03-12tevent: add tevent_loop_set_nesting_hook()Stefan Metzmacher3-1/+50
This is an ugly hack to let the s4 server work arround some bugs related to nested events and uid changing. metze
2009-03-12tevent: don't allow nested tevent_loop_once() anymoreStefan Metzmacher3-1/+53
Incompatible caller should use tevent_loop_allow_nesting() function. metze
2009-03-12tevent: pass __location__ to tevent_loop_once/wait()Stefan Metzmacher6-17/+22
metze
2009-03-12tevent: add tevent_set_abort_fn()Stefan Metzmacher2-0/+21
metze
2009-03-12tevent: add tevent_signal_support()Stefan Metzmacher2-0/+10
Not every tevent backend supports signal events. metze
2009-03-10tevent: add tevent_req_callback_data_void() macroStefan Metzmacher1-0/+2
Callers can use it instead of tevent_req_callback_data() if they use a non talloc private pointer. metze
2009-03-10tevent: add tevent_req_received() functionStefan Metzmacher3-2/+29
This function can be called as last action of a _recv() function, it destroys the data attached to the tevent_req. metze
2009-03-02Make struct tevent_req opaqueSimo Sorce3-100/+137
Move struct tevent_req in tevent_internal, and ad getters and setters for private data and the callback function. This patch also renames 'private_state' into 'data'. What is held in this pointer is in fact data and not a state like enum tevent_req_state. Calling it 'state' is confusing. The functions addedd are: tevent_req_set_callback() - sets req->async.fn and req->async.private_data tevent_req_set_print_fn() - sets req->private_print tevent_req_callback_data() - gets req->async.private_data tevent_req_data() - gets rea->data This way it is much simpler to keep API/ABI compatibility in the future.