summaryrefslogtreecommitdiffstats
path: root/lib/libkse
Commit message (Collapse)AuthorAgeFilesLines
* Remove the special _amd64_set_gsbase() code for #ifdef COMPAT_32BIT, nowpeter2005-04-262-10/+0
| | | | | that the amd64 kernel implements i386_get/set_gsbase(). All the rest of the ldt backwards compat code should go away soon.
* Use the i386_set_gsbase() syscall if it is implemented in the kernel.peter2005-04-142-16/+41
| | | | | This is a little hairy here because the allocation and usage of this functionality is split into two places in libpthread.
* Use the new atomic_cmpset_32().cognet2005-04-071-15/+0
|
* Eliminate plt relocation for kse_switchin.davidxu2005-03-212-0/+2
|
* Change the spin lock logic to a reasonable one. We should spin whendelphij2005-02-271-1/+1
| | | | | | | | | the lock is held by other thread, but not when nobody owns it. According to deischen@, this part of code will never be hit in our threads library, since it does not use locks without wait/wakeup functions. Spotted by: mingyanguo via ChinaUnix.net forum Reviewed by: deischen
* Bring in a more healthy version of the libpthread for arm, which usescognet2005-02-264-4/+12
| | | | ARM_TP_ADDRESS.
* Somewhere along the line, tick accumulation for SA threads wasdeischen2005-02-181-1/+1
| | | | | | | changed to use the statclock. Make sure we calculate the value of a tick correctly in userland. Noticed by: Kazuaki Oda <kaakun at highway dot ne dot jp>
* Set the default guardsize and stacksize in the default threaddeischen2005-02-151-0/+2
| | | | attribute when the library is initialized.
* Increase the default stacksizes:deischen2005-02-134-9/+22
| | | | | | 32-bit 64-bit main thread 2MB 4MB other threads 1MB 2MB
* Don't panic when sigsuspend is interrupted by a cancellation.deischen2004-12-191-2/+7
| | | | PR: 75273
* Use a generic way to back threads out of wait queues when handlingdeischen2004-12-1816-347/+411
| | | | | | | | | | | | | | | | | | | | | | | | | signals instead of having more intricate knowledge of thread state within signal handling. Simplify signal code because of above (by David Xu). Use macros for libpthread usage of pthread_cleanup_push() and pthread_cleanup_pop(). This removes some instances of malloc() and free() from the semaphore and pthread_once() implementations. When single threaded and forking(), make sure that the current thread's signal mask is inherited by the forked thread. Use private mutexes for libc and libpthread. Signals are deferred while threads hold private mutexes. This fix also breaks www/linuxpluginwrapper; a patch that fixes it is at http://people.freebsd.org/~deischen/kse/linuxpluginwrapper.diff Fix race condition in condition variables where handling a signal (pthread_kill() or kill()) may not see a wakeup (pthread_cond_signal() or pthread_cond_broadcast()). In collaboration with: davidxu
* i386_set_ldt() is not available when running 32 bit binaries on amd64peter2004-11-062-0/+13
| | | | | kernels. Use the recently exposed direct-set routines instead. This is only activated for when we compile i386 support libraries on amd64.
* Cosmetic tweaks to reduce diffs to the i386 counterpart.peter2004-11-061-2/+3
|
* Partial support of KSE for arm.cognet2004-11-054-59/+66
|
* Save cancelflags in signal frame, this fixes a problem thatdavidxu2004-11-012-0/+3
| | | | | | | a thread in pthread_cond_wait handled a signal can no longer be canceled. Reviewed by: deischen
* Make pthread_mutex_trylock(3) return EBUSY on failure, as all softwaregreen2004-10-311-8/+1
| | | | | | | | | packages expect and seems to be most correct according to the slightly- ambiguous standards. MFC after: 1 month Corroborated by: POSIX <http://tinyurl.com/4uvub> Reviewed by: silence on threads@
* For variables that are only checked with defined(), don't provideru2004-10-241-1/+1
| | | | any fake value.
* Check unhandled signals before thread marks itself as DEAD,davidxu2004-10-231-1/+14
| | | | | this reduces chances of signal losting problem found by Peter Holm <peter@holm.cc>
* 1. Move thread list flags into new separate member, and atomicallydavidxu2004-10-235-24/+45
| | | | | | | | | | | put DEAD thread on GC list, this closes a race between pthread_join and thr_cleanup. 2. Introduce a mutex to protect tcb initialization, tls allocation and deallocation code in rtld seems no lock protection or it is broken, under stress testing, memory is corrupted. Reviewed by: deischen patch partly provided by: deischen
* Decrease reference count if we won't use the thread, this avoids memorydavidxu2004-10-211-0/+4
| | | | leak under some cases.
* if system scope thread didn't set timeout, don't call clock_gettime syscalldavidxu2004-10-081-8/+9
| | | | | | before and after sleeping. Reviewed by: deischen
* Use PTHREAD_SCOPE_SYSTEM to decide what should be done.davidxu2004-10-071-3/+2
|
* Follow kernel change, restore signal mask correctly by using a commanddavidxu2004-10-071-21/+17
| | | | of kse_thr_interrupt.
* Allocate red zone and stack space together and then split red zone fromdavidxu2004-10-061-7/+17
| | | | | allocated space, orignal code left red zone unallocated, but those space can be allocated by user code, and result was providing no protection.
* Add a wrapper for execve(). The exec'd process must be started withdeischen2004-09-262-0/+69
| | | | | | | | | | | | | | the signal mask and pending signals of the calling thread. These are stored in userland in libpthread. There is a small race condition in this patch which could cause problems if a signal arrives after setting the (kernel) signal mask and before exec'ing. The thread's set of pending signals also are not yet installed in the exec'd process. Both of these will be corrected with the addition of a special syscall. Reported & Tested by: Joost Bekkers <joost at jodocus dot org> Reviewed by: julian, davidxu
* _tcb_ctor takes two args.cognet2004-09-242-2/+2
|
* Make sure we don't call _thr_start_sig_daemon() when SYSTEM_SCOPE_ONLY is ↵ssouhlal2004-09-241-2/+3
| | | | | | defined. This makes libpthread usable on powerpc. Approved by: grehan (mentor), deischen
* Add missing brackets. It was committed from wrong tree.davidxu2004-08-261-1/+1
|
* gcc -O2 cleanup. tested for a long time.davidxu2004-08-252-4/+4
| | | | Reviewed by: deischen
* Pull debug symbols in for statically linked binary.davidxu2004-08-211-0/+4
| | | | Reviewed by: desichen
* Fix compile, s/tp_dtv/tp_tdv/g.davidxu2004-08-163-3/+3
|
* Bring PPC up to date with latest TLS changes.grehan2004-08-162-3/+3
|
* 1. Add macro DTV_OFFSET to calculate dtv offset in tcb.davidxu2004-08-168-0/+10
| | | | 2. Export symbols needed by debugger.
* Add a file to collection all symbols will be needed by debugger.davidxu2004-08-161-0/+60
|
* Add TLS support for i386 and amd64.dfr2004-08-159-32/+41
|
* As long as we have a knob to force system scope threads, why not havedeischen2004-08-124-8/+13
| | | | | | | | | | | | | a knob to force process scope threads. If the environment variable LIBPTHREAD_PROCESS_SCOPE is set, force all threads to be process scope threads regardless of how the application creates them. If LIBPTHREAD_SYSTEM_SCOPE is set (forcing system scope threads), it overrides LIBPTHREAD_PROCESS_SCOPE. $ # To force system scope threads $ LIBPTHREAD_SYSTEM_SCOPE=anything threaded_app $ # To force process scope threads $ LIBPTHREAD_PROCESS_SCOPE=anything threaded_app
* Check debugger suspending flag for system scope thread.davidxu2004-08-081-3/+7
| | | | Reviewed by: deischen
* Add a way to force 1:1 mode for libpthread. To do this, definedeischen2004-08-075-42/+41
| | | | | | | | | | | LIBPTHREAD_SYSTEM_SCOPE in the environment. You can still force libpthread to be built in strictly 1:1 by adding -DSYSTEM_SCOPE_ONLY to CFLAGS. This is kept for archs that don't yet support M:N mode. Requested by: rwatson Reviewed by: davidxu
* s/TMDF_DONOTRUNUSER/TMDF_SUSPEND/gdavidxu2004-08-033-4/+4
| | | | Dicussed with: deischen
* Save context in kernel fashion, so it can be restored bydavidxu2004-07-311-2/+5
| | | | kse_switchin syscall.
* Remove unused field.davidxu2004-07-311-1/+0
|
* Macro optimize, this increases context switch speed about 2% on mydavidxu2004-07-311-2/+2
| | | | athlon64 machine.
* PPC MD bits for KSE. Runs test cases OK. Crippled to 1:1 mode forgrehan2004-07-198-0/+820
| | | | the time being.
* Don't include lock.h and pthread_md.h when we're being included bymarcel2004-07-181-0/+2
| | | | libthread_db. Both headers are included seperately.
* Copy lwp id to thread mailbox.davidxu2004-07-141-0/+1
|
* Call kse_switchin to switch context when being debugged.davidxu2004-07-134-24/+69
|
* Remove unused symbols.davidxu2004-07-131-14/+0
|
* Let debugger check signal, make SIGINFO works.davidxu2004-07-131-12/+25
|
* If _libkse_debug is not zero, activate thread mode.davidxu2004-07-131-0/+6
|
* Add code to support thread debugging.davidxu2004-07-135-44/+146
| | | | | | | | | | | | | | | 1. Add global varible _libkse_debug, debugger uses the varible to identify libpthread. when the varible is written to non-zero by debugger, libpthread will take some special action at context switch time, it will check TMDF_DOTRUNUSER flags, if a thread has the flags set by debugger, it won't be scheduled, when a thread leaves KSE critical region, thread checks the flag, if it was set, the thread relinquish CPU. 2. Add pq_first_debug to select a thread allowd to run by debugger. 3. Some names prefixed with _thr are renamed to _thread prefix. which is allowed to run by debugger.
OpenPOWER on IntegriCloud