Commit Graph

72360 Commits

Author SHA1 Message Date
murray
77d700702c Make it easier to specify a custom kernel configuration for the boot
floppies.

PR:		misc/36032
Submitted by:	David Yeske <dyeske@yahoo.com>
MFC after:	1 week
2002-03-31 17:21:29 +00:00
sos
d5be9373bd ATA100 is allowed on the HPT chips rev >= 3 2002-03-31 13:33:55 +00:00
alfred
abeff55bde Close some holes with p->p_args by NULL'ing out the p->p_args pointer
while holding the proc lock, and by holding the pargs structure when
accessing it from outside of the owner.

Submitted by: Jonathan Mini <mini@haikugeek.com>
2002-03-31 10:33:12 +00:00
bde
df30d6374f Support more than 32 sio unit numbers. The maximum unit number is now
(65536 * 32 - 1), but MAKEDEV only supports up to (32 * 32 -1).  Device
names use the unit number in base 32 for all "digits".

This required fixing an old bug in MAKEDEV:ttyminor().  Its arg was the
global $unit instead of $1.

Reminded by:	Valentin K. Ponomarenko <valka@krog.ukrtel.net>
MFC-after:	1 week
2002-03-31 09:15:43 +00:00
kato
cf12629bdc MFi386: revision 1.508. 2002-03-31 08:00:35 +00:00
phk
64158fc8ed A couple of bits survived Dans nukage of CV_DEBUG in favour of INVARIANTS,
take them out with tacticals.
2002-03-31 07:23:31 +00:00
phk
87273d930a Centralize the "bootdev" and "dumpdev" variables. They are still pretty
bogus all things considered, but at least now they don't camouflage as
being MD variables.
2002-03-31 07:15:28 +00:00
bde
aa2c17836e Hacks for measuring interrupt latency. Interrupt latency can be
measured accurately for periodic interrupts provided the interrupts
don't need to be serviced very quickly to keep their period almost
constant.  sio output interrupts have this property (interrupt service
can be delayed for up to 1 character time without the period changing).

This is non-optional and undocumented so that it can be added and
removed easily.  It has no significant effect unless it is enabled by
hacking on a variable using a debugger.  Hardclock and statclock interrupts
would work even better for this, at least on i386's, provided their
interrupt handlers are fast (as they are in -current but not in -stable
or in my version of -current).
2002-03-31 06:49:38 +00:00
brian
5e89c7151f Fix a typo 2002-03-31 01:57:06 +00:00
brian
ab4711b90a Add a missing ``#ifndef NONETGRAPH'' 2002-03-31 01:56:34 +00:00
brian
b222d9573a Remove some extraneous whitespace 2002-03-31 01:51:15 +00:00
brian
05e0b92acd Include arpa/inet.h 2002-03-31 01:36:08 +00:00
alc
3778bbd36e Implement i386's (o)sigreturn() like the alpha's: Use copyin() to read
the osigcontext or ucontext_t rather than useracc() followed by direct user-
space memory accesses.  This reduces (o)sigreturn()'s execution time by 5-
50%.

Submitted by:	bde
2002-03-31 01:13:21 +00:00
alc
bd314bbdd7 Add a local proc *p in exec_new_vmspace() to avoid repeated dereferencing
to obtain it.
2002-03-31 00:05:30 +00:00
marcel
688cf86e24 Pass the address of the bootinfo block to the kernel in register
r8. Keep it at the hardwired address for now. Bump the version.
2002-03-30 23:52:34 +00:00
marcel
3d34c19920 Transition to a model where the loader passes the address of the
bootinfo block in register r8. In locore.s we save the address
in the global variable 'pa_bootinfo'. In machdep.c we compare
this value against the hardwired address, but don't depend on its
validity yet (ie: we still expect the bootinfo block to be at the
hardwired address). After a small amount of time, we'll flip the
switch and depend on the loader to pass us the address. From that
moment on the loader is free to put it anywhere it likes, provided
the machine itself likes it as well.

Add some verbosity to aid in the transition. We emit a message if
the loader didn't pass the address and we also emit a message if
there's no bootinfo block at the hardwired address.

While in locore.s, reduce the number of redundant serialization
instructions. A srlz.i is a proper superset of a srlz.d and thus
is a valid replacement. Also slightly reorder the movl instructions
to improve bundle density.
2002-03-30 23:25:22 +00:00
dd
98d23cf55d Fix braino: the example interface is fxp0, not ed0.
PR:		36471
Submitted by:	John Nielsen <zoobie@bsdconspiracy.net>
2002-03-30 23:23:41 +00:00
dd
39a86e8d08 Correct typo.
PR:		36544
Submitted by:	dak@klemm.delta6.net
2002-03-30 23:19:18 +00:00
marcel
83aa1632bb Pass the physical address of the bootinfo block to the kernel in
register r8. We continue to write the bootinfo block at the same
hardwired address, because the kernel still expects it there.
It is expected that future kernels use register r8 to get to the
bootinfo block and don't depend on the hardwired address anymore.

Bump the loader version once again due to the interface change.
2002-03-30 23:00:05 +00:00
alc
84ad095a1c Correct a comment: sendsig() calls the MI vm_map_growstack() but
the corresponding comment refers to a MD grow_stack() that doesn't exist.
2002-03-30 20:58:08 +00:00
alc
34c6c68a2d Use the MI vm_map_growstack() instead of the MD grow_stack() in trap(). Remove
the MD grow_stack().
2002-03-30 20:44:31 +00:00
phk
2d4baab912 Move the "dumping" variable from systm.h to conf.h. 2002-03-30 19:58:31 +00:00
ume
4aad0459f2 sendmail 8.12 doesn't require _FFR_UNSAFE_SASL anymore.
This is standard feature.

Submitted by:	ru
2002-03-30 18:14:37 +00:00
ru
ea17e5024d Fix CLEANFILES. 2002-03-30 17:57:51 +00:00
ume
f3aafa101f sendmail 8.12 claims against GroupReadableSASLFile.
Submitted by:	mistral@imasy.or.jp
2002-03-30 17:53:10 +00:00
mdodd
ffd5de2849 Don't nowerror for sys/dev/pdq/pdq.c 2002-03-30 17:41:23 +00:00
ru
0ac094baaf Restore the correct CLEANFILES that got damaged in rev. 1.15. 2002-03-30 17:37:43 +00:00
bmah
659ed8f491 Mention the UltraSparc as supported in the README, also deal with the
fact that (now) not every supported platform has sysinstall(8).
2002-03-30 17:34:31 +00:00
dwmalone
9f21a968eb Don't use gcc specific flags. 2002-03-30 17:19:40 +00:00
dwmalone
aafb6b12f8 1) Staticise.
2) Remove registers.
3) Change some ints to size_t which are used with fread/fwrite.
2002-03-30 17:17:26 +00:00
dwmalone
99e772806a 1) Const enough things to avoid warnings.
2) Cast ifdef_level to a size_t before comparing it to a ratio of size_ts.
   Ifdef_level should always be positive.
3) Complete prototype for chfont.
4) Cast some ptrdiff_ts to ints before using as a field width.
5) Avoid shadowing a local variable p with another local variable p.
2002-03-30 17:10:20 +00:00
matusita
02ab2c827d Fix /dev/acdNc minor device number. This bug prevents installing FreeBSD
from CD-ROM in 4-stable.  Note that in 5-current, we use devfs so this
change (hopefully) shouldn't change anything.

I'll MFC to 4-stable later.

Tested with: FreeBSD/i386, 4.5-STABLE-20020330-JPSNAP
2002-03-30 16:59:06 +00:00
sos
23978f436e Remove debug output in last commit. 2002-03-30 16:51:47 +00:00
luigi
de9ccb62a4 Add a "special progname lib xxx ..." command to crunchgen
so the .lo files can be partially linked against libraries
which redefine symbols in the standard libs, or which reference
symbols in the objects.

Submitted by: Sam Leffler
MFC After: 3 days
2002-03-30 16:48:30 +00:00
dwmalone
e1a6bcf77d 1) Add missing prototypes for fmt_perm and cvt_time.
2) Move a break outside a #if block to keep gcc3 from seeing a "default:"
   at the end of a block.
3) Fix some format warnings. Some remain which can be fixed more easily
   when we have a full C99 printf.
2002-03-30 16:46:04 +00:00
sos
65dd7ee2d6 Add support for getting status (fan, temp, 5V and 12V levels) from
Promise Superswap enclosures.

Sponsored by: Advanis
2002-03-30 16:36:41 +00:00
dwmalone
59fa66220e Use the method described in the strtol man page to check if it parsed
the entire string.  This avoids signed/unsigned comparison.
2002-03-30 16:24:03 +00:00
bde
2f7ae9b739 Fixed handling of short reads in readdisklabel() and writedisklabel().
These functions use DEV_STRATEGY() which can easily return a short
count (with no error) for reads near EOF.  EOF happens for "disks" too
small to contain a label sector (mainly for empty slices).  The functions
didn't understand this at all, and looked for labels in the garbage
in the buffer beyond what DEV_STRATEGY() returned.  The recent UMA
changes combined with my local changes and configuration resulted in
the garbage often containing a valid but garbage label left over from
a previous call.

Bugs in EOF handling in -current limited the problem to "disks" with
size precisely LABELSECTOR sectors.  LABELSECTOR happens to be a very
unusual "disk" size since it is only 0 for non-i386 arches that don't
usually have disks with DOS MBRs.
2002-03-30 16:02:43 +00:00
bde
a9b7c63b29 In ffs_mountffs(), set mnt_iosize_max to si_iosize_max unconditionally
provided the latter is nonzero.  At this point, the former is a fairly
arbitrary default value (DFTPHYS), so changing it to any reasonable
value specified by the device driver is safe.  Using the maximum of
these limits broke ffs clustered i/o for devices whose si_iosize_max
is < DFLTPHYS.  Using the minimum would break device drivers' ability
to increase the active limit from DFTLPHYS up to MAXPHYS.

Copied the code for this and the associated (unnecessary?) fixup of
mp_iosize_max to all other filesystems that use clustering (ext2fs and
msdosfs).  It was completely missing.

PR:		36309
MFC-after:	1 week
2002-03-30 15:12:57 +00:00
dwmalone
938a0a2eb4 Fix vendor ID.
Make usage message match the man page. (Missing -o, -h doesn't depend on -d).

PR:		36470
Submitted by:	Gary W. Swearingen <swear@blarg.net>
MFC after:	1 week
2002-03-30 14:32:07 +00:00
dwmalone
80fda637a4 Clean up vendor ID and FBSDID.
We now seem to include <arpa/inet.h> to get ntoh*.
2002-03-30 14:18:15 +00:00
dwmalone
122ad92646 Fix constness warnings.
Remove register keyword.
Don't initialise "badtype" in declaration - it was initialised below anyway.
Remove prototype for strcpy.
2002-03-30 14:03:53 +00:00
dwmalone
abc20e723e Don't use gcc specific flags. 2002-03-30 14:01:55 +00:00
dwmalone
0183b6670c Change a "/*" within a comment to a "**".
Add a missing include spotted by gcc30.
2002-03-30 13:54:53 +00:00
dwmalone
0d14312af7 Mark some parameters as unused.
Make more code depend on '#ifndef NO_X' so we don't get unused warnings
when compiling without X.
2002-03-30 13:51:40 +00:00
dwmalone
63e149f8f6 Minor post warns stuff.
1) Fix up vendor IDs for files I'm touching and put FBSDID in the right place.
2) Cast return value of snprintf to size_t rather than casting sizeof() to
   an int, 'cos we know snprintf returns a non-genative number.
3) Avoid 'char *blank = "";' by just strduping. This could be bad sometimes,
   but all these variables usually point at malloced memory so this makes
   them always point at malloced memory.
2002-03-30 13:44:45 +00:00
brian
631238220a Add a sprinkling of mp_Down() calls to ensure that we don't think that
the multi-link NCP is up when it isn't.
2002-03-30 12:52:55 +00:00
brian
4aad533fbf Don't use the RTF_GATEWAY flag and pass the RTA_GATEWAY address when updating
routes.
2002-03-30 12:50:49 +00:00
brian
1dda0c6402 Add a missing break 2002-03-30 12:47:29 +00:00
brian
ee9763124d Merge the NETGRAPH branch into HEAD. tty devices now use netgraph's line
discipline to do the async escaping, but no other benefits are available yet.

Change ``ifdef HAVE_DES'' to ``ifndef NODES'' for consistency.

Make the Makefile a little more sane WRT RELEASE_CRUNCH.
2002-03-30 12:30:09 +00:00