* Implement fpgetsticky() for alpha.
* Direct fpsetsticky() and fp{get,set}mask() into alpha kernel via sysarch(2).
* Define new sysarch(2) stub for above and install and distribute sysarch.h
for alpha. (The fpcr IS user mode r/w, but for reasons beyond the scope
of a commit message kernel calls are needed.) And much kernel Magick is
required before these do anything, but this way programs compiled under
1.4 will DTRT on future snapshots and releases.
minor of libc and the major of libutil). For little-endian architectures
merge the bnswap() assembly versions with nto* and hton* using symbols
aliasing. Use symbol renaming for the bswap function in this case to avoid
namespace pollution.
Declare bswap* in machine/bswap.h, not machine/endian.h. For little-endian
machines, common code for inline macros go in machine/byte_swap.h
Sync libkern with libc.
Adjust #include in kernel sources for machine/bswap.h.
* add support for YP "master.passwd.by*" (master.passwd in YP, including
pw_passwd) and "passwd.adjunct.by*" (SunOS `secure' maps (?)), based
on code in FreeBSD and partially from OpenBSD.
this is only used if euid == 0.
with this, the YP "passwd.by*" maps can have `*' in the pw_passwd field.
* use pw_scan() to parse YP "passwd.by*" and "master.passwd.by*" entries
XXX: i didn't test the "passwd.adjunct" support...
(1) split nlist() into multiple files, for clarity and to make
ELF 32/64 support easier,
(2) support multiple executable types at the same time, and
(3) add support for 32- and 64-bit ELF (32-bit ELF support
originally from OpenBSD, but with several bug fixes so
that it actually handles symbols types more correctly
(and therefore _works_ for some of the more tricky uses
of the nlist routines) and changes for 64-bit ELF support).
__{v,}err{x,} and __{v,}warn{x,}
- moved err.c to __err.c and added err.c which contains weak references for
the above functions.
- Reordered the functions in __err.c to avoid redeclarations and changed
the varargs declarations to conform with the K&R style.
The traditional API of sys_errlist[] and sys_nerr is provided by weak
references if they are supported. Otherwise, we're forced to have to
have two copies of the error message string table in the library.
Fortunately, unless a program uses both sys_errlist[] and strerror(),
only one of the copies will be linked into the executable.
This is all to provide an clean namespace as required by ANSI. I've
done the same for sys_siglist[], even though it is not required, to
be consistant.
kleink