summaryrefslogtreecommitdiffstats
path: root/lib/libkse
Commit message (Expand)AuthorAgeFilesLines
* Simplify sigwait code a bit by using a waitset and removing oldsigmask.davidxu2003-07-273-22/+17
* Fix typo.davidxu2003-07-261-5/+5
* Move idle kse wakeup to outside of regions where locks are held.deischen2003-07-238-73/+130
* Add missing arguments to _amd64_restore_context() when called fromdeischen2003-07-201-2/+4
* Override libc function raise(), in threading mode, raise() willdavidxu2003-07-192-0/+54
* Add some very beta amd64 bits. These will also need some tweaking.deischen2003-07-195-0/+393
* Cleanup thread accounting. Don't reset a threads timeslicedeischen2003-07-183-38/+27
* Add a preemption point when a mutex or condition variable isdeischen2003-07-182-6/+18
* o Eliminate upcall for PTHREAD_SYSTEM_SCOPE thread, now itdavidxu2003-07-1716-395/+602
* Don't resume sigwait thread If signal is masked.davidxu2003-07-091-8/+12
* POSIX says if a thread is in sigwait state, although a signal may not indavidxu2003-07-092-11/+29
* Restore signal mask correctly after fork().davidxu2003-07-092-4/+6
* Save and restore thread's error code around signal handling.davidxu2003-07-091-1/+2
* Correctly print signal mask, the bug was introduced by cut and pastedavidxu2003-07-071-11/+20
* Add a newline to debug message.davidxu2003-07-071-1/+1
* Avoid accessing user provided parameters in critical region.davidxu2003-07-079-46/+65
* Print thread's scope, also print signal mask for every thread and printdavidxu2003-07-071-17/+13
* Correctly lock/unlock signal lock. I must be in bad state, need to sleep.davidxu2003-07-041-1/+2
* Always check and restore sigaction previously set, also access user parameterdavidxu2003-07-041-4/+7
* If select() is only used for sleep, convert it to nanosleep,davidxu2003-07-031-4/+9
* Check if thread is in critical region, only testing check_pendingdavidxu2003-07-031-1/+2
* Style.ru2003-07-021-2/+2
* Take thr_support.c out of SRCS so that it does not end up in libraries.ru2003-07-022-3/+8
* Set unlock_mutex to 1 after locked mutex.davidxu2003-07-021-2/+4
* Fix typo.davidxu2003-07-021-1/+1
* Unbreak "make checkdpadd".ru2003-07-011-1/+1
* Axe AINC.ru2003-07-011-1/+0
* Because there are only _SIG_MAXSIG elements in thread siginfo array,davidxu2003-06-303-16/+16
* Remove surplus unlocking code I accidentally checked in. This won't bedavidxu2003-06-301-4/+0
* o Use a daemon thread to monitor signal events in kernel, if pendingdavidxu2003-06-2813-586/+809
* Use mmap retuned value.davidxu2003-06-281-2/+3
* Temporary disable rwlock based code, replace it with low level KSE lockingdavidxu2003-06-281-0/+92
* Write new thread pointer back only when success.davidxu2003-06-281-3/+4
* After thread was interrupted by signal, it should relock mutex.davidxu2003-06-281-2/+1
* if thread is exiting, just returns. kse_thr_interrupt interfacedavidxu2003-06-281-2/+6
* Implement _ia64_save_context() and _ia64_restore_context(). Bothmarcel2003-06-271-0/+285
* Implement _ia64_enter_uts(). The purpose of this function is to switchmarcel2003-06-263-1/+101
* Implement _thr_enter_uts() and _thr_switch() as inline functions tomarcel2003-06-261-0/+30
* Explicitly widen int types before casting to pointer types. On 64-bitmarcel2003-06-241-3/+4
* Untangle the inter-dependency of kse types and ksd types/functionsmarcel2003-06-235-3/+20
* Change the definition of _ksd_curkse, _ksd_curthread andmarcel2003-06-231-3/+3
* Define THR_{G|S}ETCONTEXT to expand to {g|s}etcontext(2).marcel2003-06-231-0/+38
* Implement atomic_swap_{int|long|ptr}. Define atomic_swap_ptr as amarcel2003-06-231-0/+47
* Move the machine specific files from sys/Makefile.inc and put themmarcel2003-06-234-6/+15
* Don't lock scheduler lock twice.davidxu2003-06-181-2/+1
* After selecting a thread to handle a signal and takingdeischen2003-06-081-3/+8
* Insert threads at the end of the free thread list so thatdeischen2003-06-081-1/+1
* Provide a reference to __sys_write. The implementation uses this whendeischen2003-06-081-0/+3
* Don't force -L/usr/lib. This is incorrect because we should not beimp2003-06-081-1/+1
* Only init _thread_sigact once, needn't init it again after a fork().davidxu2003-06-041-31/+31
OpenPOWER on IntegriCloud