Commit Graph

66738 Commits

Author SHA1 Message Date
Matthew Dillon
f286003909 Create a mutex pool API for short term leaf mutexes.
Replace the manual mutex pool in kern_lock.c (lockmgr locks) with the new API.
Replace the mutexes embedded in sxlocks with the new API.
2001-11-13 21:55:13 +00:00
Giorgos Keramidas
6ba3aeb390 Replace use of "0" constraints in inline asm with "+" constraints,
when an operand is used both for input and output.

Reviewed by:	jhb
2001-11-13 19:55:20 +00:00
Makoto Matsushita
3e7b2f7461 sysinstall complains about creating acd0t* devices while all dists are
extracted and recreating device files.

Without this, you'll see following messages when 'MAKEDEV all':
	[:  : out of range
	[:  : out of range
	acd0t is invalid -- can't have more than 32 devices or 169 tracks

Reported by: David Syphers <dsyphers@uchicago.edu> (at current@FreeBSD.org)
2001-11-13 18:07:14 +00:00
Andrew R. Reiter
853fbff5f7 - Modify to make the generated device driver code work with current.
- Use M_ZERO instead of another call to bzero().

Inspired by: misc/31905
2001-11-13 17:58:14 +00:00
Andrew R. Reiter
357cfee7e6 Never added my entry which the committers guide says is a
tradition.
2001-11-13 17:44:40 +00:00
John Baldwin
4d29433e33 Cleanups after previous change:
- Renumber labels since the previous revision removed one.
- Remove useless and wrong comment.
  - Repeating the function name is just redundant.
  - The previous revision made the comment about %edx useless.
  - The comment about %eax was wrong (but did explain why %eax used to be
    fixed up).

Submitted by:	bde
2001-11-13 16:58:43 +00:00
John Baldwin
00f13cb353 As a followup to the previous fixes to inferior, revert some of the
changes in 1.80 that were needed for locking that are no longer needed now
that a lock is simply asserted.

Submitted by:	bde
2001-11-13 16:55:54 +00:00
Makoto Matsushita
2ec2294625 Set full-path of cvsup.
In src/Makefile rev 1.232, environment variable PATH is set explicitly
to /sbin:/bin:/usr/sbin:/usr/bin.  As a result, binaries located on
non-standard path cannot be executed without full-path (it's the change
of this revision).

However, cvsup is not in our base system -- you lose if you try to
"make update" without setting SUP in make.conf or command line argument.
I think it is safe to assume that cvsup is located at /usr/local/bin,
and it would help other people who first try to do "make update".

PR:		31932
MFC after:	1 day
2001-11-13 15:47:40 +00:00
Søren Schmidt
e29e633207 Add yet another VIA ATA100 southbridge. 2001-11-13 11:37:23 +00:00
Søren Schmidt
7afba84a9f Fix bogon, free() the right memory type. 2001-11-13 11:33:23 +00:00
Alfred Perlstein
2663693cf8 Turn on NO_WERROR and set WARNS to 1.
Fix the WARNS 1 warnings except unused variables.

Add prototype for log_netobj().
Don't compare signed/unsigned.
Cast u_int64_t to 'unsigned long long' and print using %llu.
Fix constness of string arrays.
Use a cast to avoid an unused parameter in a signal handler.
alarm(2) can't fail, so don't check for it.
ANSI'ify some functions.
2001-11-13 11:24:23 +00:00
Alfred Perlstein
7e440a74e8 Have rpcgen output the prototypes for the dispatch tables 2001-11-13 11:08:29 +00:00
Alfred Perlstein
0a679263f9 NFS split forgot to prototype nfsclnt() syscall, do it. 2001-11-13 09:07:31 +00:00
Andrew R. Reiter
a220d00e74 Update the KLD examples so they will work with current.
Reviewed: jhb
2001-11-13 07:42:30 +00:00
Warner Losh
751a23649d Better error message when cardtype is not recognized 2001-11-13 06:46:19 +00:00
John Baldwin
7bfd54ac0f The kernel already fixes up %eax for parents that return from fork, so
don't bother manually fixing up %eax for the parent process by testing
the value in %edx and zeroing and already zeroed %eax.
2001-11-13 06:36:43 +00:00
Paul Saab
817805d9c9 Fix a signed bug in the crashdump code for systems with > 2GB of ram.
Reviewed by:	peter
2001-11-13 01:08:54 +00:00
Dag-Erling Smørgrav
e568548eee Now it gets interesting... 2001-11-13 00:51:00 +00:00
Mark Peek
e398651c46 Don't enable FP in the kernel. It is not needed when -msoft-float is used.
Reminded by:	benno
2001-11-13 00:44:21 +00:00
Dag-Erling Smørgrav
21dc7ada34 Use a slightly less predictable temp file name, remove a stale list of PRs
with corrupted submission dates (they were fixed on freefall a while ago),
and uncomment the code that removes the temp file when we're done.
2001-11-13 00:07:21 +00:00
Dag-Erling Smørgrav
89090fe2ba This is a PR statistics-gathering script I wrote a while ago. It currently
doesn't do much of interest except produce graphs similar to phk's.
2001-11-12 23:59:56 +00:00
Giorgos Keramidas
a98556d61f Change constraints to use "+" in inline asm instead of mapping input
to output parameters with "0".

Reviewed by:	jhb
2001-11-12 21:58:27 +00:00
Giorgos Keramidas
7377f0d190 Remove EOL whitespace.
Reviewed by:	alfred
2001-11-12 20:51:40 +00:00
Giorgos Keramidas
074df01866 Make KASSERT's print the values that triggered a panic.
Reviewed by:	alfred
2001-11-12 20:50:06 +00:00
John Baldwin
5b29d6e906 Clean up breakage in inferior() I introduced in 1.92 of kern_proc.c:
- Restore inferior() to being iterative rather than recursive.
- Assert that the proctree_lock is held in inferior() and change the one
  caller to get a shared lock of it.  This also ensures that we hold the
  lock after performing the check so the check can't be made invalid out
  from under us after the check but before we act on it.

Requested by:	bde
2001-11-12 18:56:49 +00:00
John Baldwin
d900909439 Use newer constraints for atomic_cmpset().
Requested by:	bde
2001-11-12 18:53:45 +00:00
Mike Barcroft
dd05edcb0f Add the ability to distinguish between compile environments that
support `long long' and those that don't.

Reviewed by:	bde
2001-11-12 18:33:00 +00:00
John Baldwin
20a2016a81 Use newer constraints for inline assembly for an operand that is both an
input and an output by using the '+' modifier rather than listing the
operand in both the input and output sections.

Reviwed by:	bde
2001-11-12 16:57:33 +00:00
Alfred Perlstein
a95a0d36d4 fix parameters to lock_answer().
Submitted by: Timo Geusch <freebsd@unix-consult.com>
2001-11-12 16:34:59 +00:00
Alfred Perlstein
755f5b77d8 Fixup for WARNS. 2001-11-12 16:29:45 +00:00
Yoshihiro Takahashi
6a306bf3cd Use make_dev_alias() instead of call make_dev() twice. 2001-11-12 13:06:04 +00:00
Josef Karthauser
a5e74e2314 Fix a typo:
Flinstone -> Flintstone.

Submitted by:	grog (during the kernel debugging session at BSDConEurope)
2001-11-12 12:21:39 +00:00
Benno Rice
2c6c884048 -msoft-float must always be turned on when compiling the PowerPC kernel. 2001-11-12 10:12:00 +00:00
Peter Wemm
5207c02e45 Implement eficlock_set() to set hardware clock. 2001-11-12 09:29:05 +00:00
Peter Wemm
658c434d90 Commit the better version that I had a while ago. This has only one
reference to curthread.  (#define curproc (curthread->td_proc)).
2001-11-12 08:53:34 +00:00
Matthew Dillon
5b1927bc01 When curproc is used repeatedly store curproc into a local
variable to reduce generated code.  This is a test case.
2001-11-12 08:42:20 +00:00
Marcel Moolenaar
c35a41320d o os_boot_rendez is responsible for clearing the IRR bit by
reading cr.ivr, as well as writing to cr.eoi.
o  use global variables to pass information to os_boot_rendez
   so that it doesn't have to jump through hoops to find it
   out. This avoids traps on the AP without it even being
   initialized. This fixes SMP configurations.
o  Move the probing of the MADT to the end of cpu_startup,
   instead of at the start of cpu_mp_probe. We need to probe
   the MADT for non-SMP configurations as well. This fixes
   uniprocessor configurations.
o  Serialize AP wake-up by waiting for the AP. We need to do
   this since we use global variables to for the AP to use.
   As a side-effect, we can use printf() more easily to see
   what's going on.
2001-11-12 07:18:16 +00:00
Marcel Moolenaar
3ea7ef6aa3 Invoke trap() for the alt. ITLB and alt. DTLB interrrupts when
the region is not 6 or 7. This changes the behaviour from
inserting a bogus region 6 mapping to a kernel panic.
2001-11-12 07:08:45 +00:00
Warner Losh
807d989e82 Use d_thread_t (the new ugly compatibility hack) rather than the old
uglier compatibility hack (#define thread proc).
2001-11-12 05:35:27 +00:00
Peter Wemm
e16c208887 Remove #if 0'ed code that was replaced by vm_ksubmap_init() and GC'ed
on other platforms.
2001-11-12 04:14:04 +00:00
Alfred Perlstein
3fd0605ef0 note that 'nolockd' can't be used when updating mount flags 2001-11-12 02:43:56 +00:00
Alfred Perlstein
13190d8754 Allow users to use the 'nolockd' or -L options with mount_nfs in order
to avoid the need for rpc.lockd to perform client locks.  Using
this option a user can revert back to using local locks for NFS mounts
like we did before we had rpc.lockd.
2001-11-12 02:33:52 +00:00
Warner Losh
cfc78f73bd Sync to 1.17 of pccarddevs 2001-11-11 23:52:49 +00:00
Warner Losh
ac5e5bcb16 Add GENTEK as one of the bogus vendors we need to id cards from the cis 2001-11-11 23:52:28 +00:00
Makoto Matsushita
b7e5d41abe Forced commit: previous patch comes from PR: 31009 not 31109.
Noticed by: Hiroo ONO <hiroo@oikumene.gcd.org> (PR submitter)
2001-11-11 23:46:53 +00:00
Alfred Perlstein
32e9a24903 Drop privs while servicing kernel lock requests.
Restore umask used before FIFO creation.
If opening the FIFO fails, exit.
Properly check the return value of open (use == -1 instead of < 0).
2001-11-11 23:40:10 +00:00
Alfred Perlstein
f03e89de68 turn vn_open() into a wrapper around vn_open_cred() which allows
one to perform a vn_open using temporary/other/fake credentials.

Modify the nfs client side locking code to use vn_open_cred() passing
proc0's ucred instead of the old way which was to temporary raise
privs while running vn_open().  This should close the race hopefully.
2001-11-11 22:39:07 +00:00
Warner Losh
81f645ca10 Add a bunch of wireless cards for NEWBUS from NetBSD's list. I'm
trying something new with the macros here and will likely try to get
them adopted by NetBSD as well as moving them to other drivers.  They
make the list more compact and easier to read, the price of rigid to a
schema for generating them (of course there are those that would argue
this isn't bad).

Obtained from: NetBSD's if_wi_pcmcia.c 1.9
2001-11-11 21:27:12 +00:00
Warner Losh
3dfff24a19 Move kern.post.mk to last line, now that it appears safe 2001-11-11 20:33:02 +00:00
Warner Losh
7a26a4ef28 Move include of kern.post.mk to the last line of the file.
# Forced commit, last commit message was bogus
2001-11-11 20:21:37 +00:00