| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | So passes uthreads |   guenther | 2012-09-01 | 4 | -198/+0 |
| | | | | | | Like autumn leaves on water don't fear the tedu@ | ||||
| * | Sync with amd64, to allow this to compile without warnings with gcc4: |   miod | 2010-12-03 | 1 | -5/+5 |
| | | | | | | | ``Correct the _atomic_lock() asm so that gcc accepts the constraints when compiling without optimization; fix copied from the kernel's atomic.h'' spotted by jim@ | ||||
| * | Use __attribute__ ((aligned (16)), to make sure the FPU state is aligned |   kettenis | 2010-06-30 | 1 | -3/+3 |
| | | | | | | | | | | on a 128-bit boundary like we do on amd64. The padding within the struct made sure the state was properly aligned *within* the struct, but since the alignment restriction on the struct itself was only 64-bit, the required alignment was not guaranteed. We just got lucky. ok oga@, guenther@ | ||||
| * | Properly align stack such that code that uses SSE2 instructions doesn't crash. | kettenis | 2008-01-28 | 1 | -3/+7 |
| | | | | | ok espie@, beck@ | ||||
| * | - due to the fninit() in _thread_machdep_save_float_state() all calls to |   kurt | 2006-04-06 | 1 | -5/+16 |
| | | | | | | | | | | it need to be matched with a call to _thread_machdep_restore_float_state(), so add missing one in _thread_machdep_init(). - 16-byte align fp frame - call fwait() before save and and after restoring fp state. lots of guidance and ok mickey@ | ||||
| * | if cpu has fxsr, use fxsave instead; ok marc |   deraadt | 2004-02-21 | 2 | -11/+45 |
| | | |||||
| * | save i386 fpu state |   marc | 2003-01-24 | 2 | -17/+24 |
| | | |||||
| * | remove unused/unreferenced code. | marc | 2002-10-11 | 1 | -9/+1 |
| | | | | | ok fgs@ | ||||
| * | typo in comment |   d | 2001-03-13 | 2 | -4/+4 |
| | | |||||
| * | switch to _machdep_switch() instead of setjmp/longjmp. For some reason this fixes sparc threads. | d | 2000-10-04 | 1 | -84/+5 |
| | | |||||
| * | public domain | d | 2000-10-03 | 1 | -1/+3 |
| | | |||||
| * | preparation for switch-based threads | d | 2000-09-25 | 2 | -0/+107 |
| | | |||||
| * | sync with FreeBSD | d | 1999-11-25 | 1 | -4/+5 |
| | | |||||
| * | no need to save signal state in setjmp/longjmp - it will always be the same, so we may as well save a few syscalls | d | 1999-03-10 | 1 | -3/+3 |
| | | |||||
| * | tidy | d | 1999-03-10 | 1 | -8/+5 |
| | | |||||
| * | Goodbye, MIT pthreads... you were a handy reference implementation | d | 1999-03-10 | 6 | -668/+0 |
| | | |||||
| * | mi+md jmp_buf; save i386s fs and gs for WINE (csapuntz@stanford.edu) | d | 1999-01-17 | 1 | -22/+70 |
| | | |||||
| * | use <machine/reg> | d | 1999-01-10 | 1 | -4/+6 |
| | | |||||
| * | move spinlock.h into the real world | d | 1999-01-08 | 1 | -6/+0 |
| | | |||||
| * | md spinlock | d | 1998-12-21 | 2 | -12/+19 |
| | | |||||
| * | add md _atomic_is_locked; clean | d | 1998-12-18 | 1 | -2/+9 |
| | | |||||
| * | Move atomic_lock code from asm to C with inline asm; | d | 1998-11-20 | 2 | -46/+26 |
| | | | | | | Add m68k, mips and sparc. (needs more careful checking) Add 'slow_atomic_lock' for crippled archs. | ||||
| * | sync with FreeBSD (rwlock, gc thread, man pages) | d | 1998-11-09 | 1 | -1/+4 |
| | | | | | | | | add (broken) mips md stuff fix some const warnings add sigaltstack() stub another hash at getting shlib auto-init to work (mips/elf and i386/a.out) | ||||
| * | somehow i missed this directory.. | d | 1998-08-28 | 2 | -0/+81 |
| | | |||||
| * | forgot to add this one. |   peter | 1998-07-23 | 1 | -0/+56 |
| | | | | | | | | SYS.h is copied from lib/libc/arch/i386/SYS.h but with "machdep_sys_" preprended to the syscall name in the ENTRY() macro. When the code is a bit more sync'ed, using SYS.h from the linc hierarchy should work. | ||||
| * | o removed more unused or old, left over files | peter | 1998-07-22 | 1 | -0/+11 |
| | | | | | | | | | | | | | | | o general changes so that now the library compiles and many of the tests run correctly o pthreads/Makefile.inc rebuilt using libc/sys/Makefile.inc as a template. this should be kept in sync in case of new syscalls soon TODO: figure out why remaining tests fail and fix look at building .so version of library tidy up arch/i386 directory - no longer uses syscall-template.S port other arch's from libc syscall templates | ||||
| * | move unchanged files from machdep/ into appropriate arch/${MACHINE} | peter | 1998-07-21 | 4 | -122/+351 |
| | | | | | | | | | directories. These files have been renamed as appropriate, but hackers of these files should note those that started off as netbsd files (alpha and sparc). also moved machdep/openbsd-2.0/ to include/sys/ | ||||
| * | initial import of NetBSD tree | deraadt | 1995-10-18 | 4 | -0/+372 |
 |
index : wireguard-openbsd | |
| WireGuard implementation for the OpenBSD kernel | Matt Dunwoodie |
| summaryrefslogtreecommitdiffstats |