summaryrefslogtreecommitdiff
path: root/lib/libc/arch
AgeCommit message (Collapse)Author
2002-05-17kill unused lineMichael Shalayeff
2002-05-17Remove unused pieces and clean Makefile.incMiod Vallat
2002-05-08Sync with netbsd version, adds SOFTFLOAT (not supported) and copyright.Dale Rahn
2002-05-08OpenBSD tag.Dale Rahn
2002-05-08Add copyright block to this file I wrote long ago.Dale Rahn
2002-03-11More fp-related libc pieces for hppa.Miod Vallat
This enables awk to link.
2002-03-08reverse the conditionMichael Shalayeff
2002-03-08set ret for mem*, no mycopy, better reverse copy checkMichael Shalayeff
2002-02-25use c versionsMichael Shalayeff
2002-02-19We need to keep K&R compat macros for some ports where the assemnbler isTodd C. Miller
still using -traditional-cpp.
2002-02-19We live in an ANSI C world. Remove lots of gratuitous #ifdef __STDC__ cruft.Todd C. Miller
2002-02-17rfork() needs a fork.S-like stub as well; ok artTheo de Raadt
2002-02-16Part one of userland __P removal. Done with a simple regexp with some minor ↵Todd C. Miller
hand editing to make comments line up correctly. Another pass is forthcoming that handles the cases that could not be done automatically.
2002-02-08fix typo, and put back stack frame opsMichael Shalayeff
2002-02-07reverse condition in the sp botchness check, we need to nullifyMichael Shalayeff
if we are ok not if we are bad. fix a pasto in branch w/in longjmps.
2002-02-07additional condition on reverse copy; from fredette@Michael Shalayeff
2002-02-06no need to interlock the register, use r0 as outputMichael Shalayeff
2002-02-06implement the setjmp/longjmp family, based on fredette@'s workMichael Shalayeff
2002-01-10sigprocmask should return 0 on success, fixes powerpc ntpdate problem.Dale Rahn
2002-01-04system calls are made thru _thread_sys_* with weak symbols of the real nameDale Rahn
pointing to the _thread_sys_ version. ok pval, fgs
2002-01-04longjmp(..., 0) is not really legal, but all other archs I've lookedArtur Grabowski
at make sure that setjmp doesn't return 0 if longjmp is passed a 0. Avoid surprises.
2002-01-03Always have the real entry point to the syscall called _thread_sys_Artur Grabowski
and provide the real name as a weak alias. This is the last piece needed for libc_r on sparc64 (plus debugging).
2001-11-01use _C_LABELMichael Shalayeff
2001-10-24fix __PSEUDO_NOERROR, two waysMichael Shalayeff
2001-10-11Correct length of jmpbuf, it is sizeof(long) * _JBLEN == 8*14Todd C. Miller
This fixes sigsetjmp()/siglongjmp() on sparc64
2001-09-26Changes to SYS.h for sparc64 to make it work with libc_rDale Rahn
bulk of work done by miod.
2001-09-26Revert last. The ERROR() macro is called from leaf functionsArtur Grabowski
and is _not_ supposed to do "call"s.
2001-09-25Add an ASPICFLAG variable to bsd.own.mk which a platform use to pass flagsDale Rahn
to 'as' to make it compile pic. add defaults for sparc64 which needs picflags set to -fPIC and ASPICFLAGS to -KPIC. Also simplifies the systemcall generation for sparc64 PIC mode. ok epsie@.
2001-09-24Rename __minbrk to minbrk to allow gmon.c to link correctly.Artur Grabowski
XXX - The fix should be the other way around - make all other archs match sparc64.
2001-09-21add isnan.cMiod Vallat
2001-09-21Fix incorrect number of args to __SYSCALL. Somehow this managed toTodd C. Miller
work in my test on static libs -- I've no idea how though. While I'm here, remove __RSYSCALL since it is the same as __PSEUDO.
2001-09-20change PSEUDO_NOERROR so it is indeed does not set errno.Dale Rahn
2001-09-20Add PSEUDO_NOERROR similar to that used by NetBSD.Todd C. Miller
Some ports were setting errno in PSEUDO and some were not. Now errno is set for all in PSEUDO and PSEUDO_NOERROR is provided for the non-errno case (only used by _exit). Needs testing on vax and m88k. XXX - hppa and powerpc still lack a real PSEUDO_NOERROR implementation. Currently PSEUDO_NOERROR and PSEUDO are the same (so builds don't break on those platforms).
2001-09-20make this compile for kernelMichael Shalayeff
2001-09-11Keep the strings fully quoted.Dale Rahn
2001-09-10Use the LBL isnan() and isinf() on all platforms with ieee floating point.Todd C. Miller
2001-09-10Use the LBL frexp() on all platforms with ieee floating point.Todd C. Miller
2001-09-10Use ${LIBCSRCDIR} not ${.CURDIR} for relative path (allows libc_r to build)Jason Wright
2001-09-07Adapt to our sigcontext and fix the obvious bugs.Artur Grabowski
2001-09-05explicit alignment, just like on alpha.Artur Grabowski
2001-08-29Last pieces needed to build libc on sparc64.Artur Grabowski
2001-08-29Bunch of necessary files from NetBSD.Artur Grabowski
2001-08-29more files from sparc.Artur Grabowski
2001-08-29bunch of files from sparc.Artur Grabowski
2001-08-29except ffs and strlen we use the generic functions (for now).Artur Grabowski
2001-08-29ffs from sparc and strlen from NetBSD.Artur Grabowski
(Don't ask me why our makefiles wouldn't accept using strlen.c)
2001-08-29syscall(), trivialArtur Grabowski
2001-08-29signal stuff from sparc, cleaned up.Artur Grabowski
2001-08-29setlogin from NetBSDArtur Grabowski
2001-08-29sbrk from NetBSDArtur Grabowski