Commit Graph

2457 Commits

Author SHA1 Message Date
mrg d4dfeaf200 add sunos32 support. 2001-02-02 07:30:22 +00:00
enami 6dcc7e3de4 Don't use PR_URGENT to allocate page header. We don't want to just panic
on memory shortage.  Instead, use the same wait/nowait condition with the
item requested, and just cleanup and return failure if we can't allocate
page header while we aren't allowed to wait.
2001-01-29 02:38:02 +00:00
thorpej cfa7048975 Regen; getpid(2) is MP-safe. 2001-01-27 07:48:28 +00:00
thorpej 8740614bad getpid(2), our first MP-safe syscall! 2001-01-27 07:47:26 +00:00
thorpej 1b6a66420e Regen; add sy_flags. 2001-01-27 07:23:06 +00:00
thorpej 1628dd441e Add a "sy_flags" to struct sysent, define a SYCALL_MPSAFE
system call flag (indicating that the kernel lock does not
need to be acquired when entering the kernel on that syscall).
2001-01-27 07:21:43 +00:00
itojun 6e24d735f0 when the peer is disconnected before accept(2) is issued,
do not return junk data in mbuf (= sockaddr on accept(2)'s 2nd arg).
set the length zero.

behavior checked with bsdi and freebsd.
partial solution to PR 12027 and 10698 (need more investigation).
2001-01-22 18:14:11 +00:00
jdolecek d9466585b7 make filesystem vnodeop, specop, fifoop and vnodeopv_* arrays const 2001-01-22 12:17:35 +00:00
jdolecek d0a540b250 rename vfs_op_init() to vfs_op_check() and make it only check the offsets
and number of ops, not touch anything - vnode_if.sh now generated
proper offset numbers; vfs_op_check() is only defined and called for DEBUG
kernels

constify extern declaration of vfs_op_descs[]
g/c vfs_opv_numops, use VNODE_OPS_COUNT instead
make vfs_opv_init_explicit() and vfs_opv_init_default() static
2001-01-22 09:57:25 +00:00
jdolecek c3f6f769a3 regen: *_desc are generated correctly, new VNODE_OPS_COUNT,
structures constified
2001-01-22 09:53:31 +00:00
jdolecek 950412b4f5 when generating individual *_desc, generate also proper offsets so that
then don't need to be patched at runtime
add new define VNODE_OPS_COUNT (to vnode_if.h) so that the number is known
at compile-time
make stuff const, it now can be
2001-01-22 09:52:21 +00:00
martin d1ff650a0c Regen after de-const-ing. 2001-01-19 12:24:09 +00:00
martin c9803fbf98 Remove over-const-ification.
This structures are actually modified at kernel init time by vfs_op_init.

XXX - looks like the state after initialization is pretty const and with
some magic in the generator script (and appropriate changes to vfs_op_init)
it could be made const.
2001-01-19 12:22:56 +00:00
jdolecek 34c8ae80da constify 2001-01-18 20:28:15 +00:00
thorpej 2f89e3d744 Explicitly include <machine/intr.h> if __HAVE_GENERIC_SOFT_INTERRUPTS. 2001-01-17 18:21:41 +00:00
fvdl ea08a209b5 Adapt for procfs_valid* argument change. 2001-01-17 01:13:23 +00:00
thorpej d74e432ed3 Make softclock a generic soft interrupt of the API is available,
adding the requisite void * argument to softclock().
2001-01-15 20:19:50 +00:00
thorpej 7200d34a76 Whenever ps_sigcheck is set to true, signotify() the process, and
wrap this all up in a CHECKSIGS() macro.  Also, in psignal1(),
signotify() SRUN and SIDL processes if __HAVE_AST_PERPROC is defined.

Per discussion w/ mycroft.
2001-01-14 22:31:58 +00:00
thorpej a624c70966 Use splvm(), not the home-grown splmem(). 2001-01-14 02:08:35 +00:00
thorpej b5104c1ca5 Change some low-hanging splimp() calls to splvm(). 2001-01-14 02:06:21 +00:00
cgd 926244ff0d nuke extra token (;) after #endif 2001-01-12 22:55:10 +00:00
fvdl 6907c4ded8 Do syscall_intern after p_traceflag has been copied to the new
process (if it is inherited), so that ktrace continues to work
properly on the child.
2001-01-09 23:36:50 +00:00
nathanw d75edd2077 Reserve system call slots for getcontext, setcontext, LWPs, and
scheduler activations.
2001-01-08 18:25:35 +00:00
chs 6717a2ac1b in vtruncbuf(), use a "synchronous freeing" flush to prevent a race
between write i/os in a disk-based filesystem vs. the disk block being
freed by a truncation, allocated to a new file, and written again with
different data.  if the disk driver reorders the requests and does
the second i/o first, the old data will clobber the new, corrupting
the new file.
2001-01-08 07:05:47 +00:00
jdolecek d7f811d098 utrace(2): limit size of user data to KTR_USER_MAXLEN (currently 2048); return EINVAL if 'len' is bigger 2001-01-05 22:25:26 +00:00
jdolecek 2bd5bf0ec2 regen after utrace() parameter rename 2001-01-05 21:53:38 +00:00
jdolecek fbc7e223d7 utrace(2): rename 'id' parameter to 'label' 2001-01-05 21:42:08 +00:00
thorpej eb80878b1b Happy new year! 2001-01-01 20:18:34 +00:00
jdolecek df458f620b Avoid generating duplicate SYS_* defines to *syscall.h - is possible
for kern/syscalls.master's sys_getpid()/sys_getpid_with_ppid().
This fixes kern/11853 by Bernd Ernesti.
2001-01-01 16:41:43 +00:00
sommerfeld aace86f946 MULTIPROCESSOR: The two calls to psignal() inside mi_switch() are
inside the scheduler lock perimeter and should be sched_psignal() instead.
2001-01-01 16:02:51 +00:00
ad e3514a4191 PR 4853: we fork a lot more during startup these days. Wrap nextpid to 500. 2000-12-31 17:43:41 +00:00
sommerfeld f2bdd546dd Add a missing simple_unlock() to the LK_NOWAIT/VXLOCK error case in vget(). 2000-12-31 03:13:51 +00:00
jdolecek 7899fd7798 regen - utrace(2) addition 2000-12-28 11:16:48 +00:00
jdolecek a80dee65df add utrace(2) - this syscall allows to add user ktrace entries
idea from FreeBSD, but added argument (const char *id) so that it's possible
to differentiate between entries from different sources
2000-12-28 11:10:15 +00:00
jmc ca607b87cf Default lock_printf to syslog rather than printf. Some of the lock debug checks
are done inside of wakeup which is holding the sched lock. Printf can cause
wakeup to get called again (pty redirection of console message) which will
panic with sched lock already held.

This isn't a long term fix as not being able to printf vs. sched lock should
be cleaned up better but this avoids continual panics with lockdebug running
and an xterm -C.
2000-12-24 23:56:24 +00:00
jdolecek e9e91a0fb5 split off thread specific stuff from struct sigacts to struct sigctx, leaving
only signal handler array sharable between threads
move other random signal stuff from struct proc to struct sigctx

This addresses kern/10981 by Matthew Orgass.
2000-12-22 22:58:52 +00:00
fvdl 0d4862bed1 Regen. 2000-12-22 20:07:09 +00:00
fvdl 8eaf6e8d8e Make that defined(LKM) || defined(_LKM), to catch both the cases
of compiling a kernel with LKM support and compiling an LKM.
2000-12-22 20:05:35 +00:00
fvdl 2ac441ddfb LKM -> _LKM around opt_ include. 2000-12-22 20:00:16 +00:00
mrg 264248925f regen 2000-12-22 15:10:38 +00:00
mrg 80ab63644e redo the previous correctly 2000-12-22 15:09:46 +00:00
mrg af36640df1 regen 2000-12-22 14:59:40 +00:00
mrg 62230909f1 avoid redefinition of VNODE_OP_NOINLINE 2000-12-22 14:58:39 +00:00
mrg a0e71c9ca7 avoid redefinition of VNODE_OP_NOINLINE 2000-12-22 14:45:50 +00:00
scw 96698d967e Change struct emul's "char e_name[8]" field to "const char *e_name"
to allow for emulation names >= 8 characters.
2000-12-19 22:08:36 +00:00
mrg 7284657373 only print the magic number changing if not cold 2000-12-19 04:39:19 +00:00
jdolecek 5c6130a2b1 hum, use freebsd_setregs for FreeBSD ELF binaries (this bug was introduced with
the execsw[] changes)
2000-12-17 21:36:49 +00:00
fvdl 89b5cfa9c5 Jump out of the scandir loop if VOP_READDIR returns an error, avoiding
running around in circles.
2000-12-15 11:52:14 +00:00
mycroft b29180b2ff Tighten up the ELF signature checks, and actually look for the ABI tag added
in newer glibc versions.
2000-12-15 06:14:21 +00:00
jdolecek 03354da279 this doesn't need <sys/trace.h> 2000-12-13 17:48:46 +00:00