Warner Losh [Sun, 26 Oct 2003 20:09:18 +0000 (20:09 +0000)]
Minor tweak to Olicom 2220. Also add the OC-2232 while I'm here.
Looks like 'special magic' is needed to access the ethernet part of
that combo card, so I've not added it to ex.
Alan Cox [Sun, 26 Oct 2003 06:29:26 +0000 (06:29 +0000)]
- Simplify vm_object_collapse()'s collapse case, reducing the number
of lock acquires and releases performed.
- Move an assertion from vm_object_collapse() to vm_object_zdtor()
because it applies to all cases of object destruction.
Peter Wemm [Sun, 26 Oct 2003 05:54:07 +0000 (05:54 +0000)]
Use -fno-builtin-log so gcc doesn't get ideas about using a math function
to log data. Clean up an unused variable that was hidden by the WARNS?=2
being commented out. Uncomment it now that it compiles cleanly again.
Peter Wemm [Sun, 26 Oct 2003 05:51:51 +0000 (05:51 +0000)]
Make this compile cleanly. Switch to using rpcgen to generate
code that can be compiled on an ansi-C compiler, so that it puts the
xdrproc_t casts in. This also implies a slight function name change.
Peter Wemm [Sun, 26 Oct 2003 05:30:40 +0000 (05:30 +0000)]
Fix up warnings. There were some biggies here. There seems to be two
different sets of definitions in /usr/include/rpcsvc, mostly compatable,
but with different names. Will the real one please stand up?
In order to get prototypes for yp_maplist, we had to use the 'other'
naming system.
Peter Wemm [Sun, 26 Oct 2003 04:43:02 +0000 (04:43 +0000)]
Fix gcc warnings. If NAME_MAX is 255, and d_namlen is a uint8_t, then
d_namlen can never be > NAME_MAX. Stop gcc worrying about this by
using a preprocessor test to see if NAME_MAX changes.
Peter Wemm [Sun, 26 Oct 2003 04:36:47 +0000 (04:36 +0000)]
Give wider types to sscanf to fix two warnings (u_short cannot be > 0xffff)
and to make sure that we catch oversized arguments rather than silently
truncate them. I dont know if sscanf will reject an integer if it will
not fit in the short return variable or not, but this way it should be
detected.
Peter Wemm [Sun, 26 Oct 2003 03:51:47 +0000 (03:51 +0000)]
(mostly) Clean up some const warnings here. The code takes some liberties
because it is the originator of various const strings and knows that they
came from malloc.
Peter Wemm [Sun, 26 Oct 2003 03:34:13 +0000 (03:34 +0000)]
Move the -Wtraditional warning from the global cpp flags to when the -p
(portability warnings) switch is used. Add -Wno-system-headers after it so
that we dont get 500 screenfulls of warnings about #elif in /usr/include.
I'm not entirely happy with this. Maybe cdefs.h shouldn't use #elif and
instead nest #else clauses?
Peter Wemm [Sat, 25 Oct 2003 18:51:41 +0000 (18:51 +0000)]
For the SMP case, flush the TLB at the beginning of the page zero/copy
routines. Otherwise we run into trouble with speculative tlb preloads
on SMP systems. This effectively defeats Jeff's revision 1.438
optimization (for his pentium4-M laptop) in the SMP case. It breaks
other systems, particularly athlon-MP's.
Robert Watson [Sat, 25 Oct 2003 16:43:50 +0000 (16:43 +0000)]
Check (locked) before performing an advisory unlock following a failure
of vn_start_write(). Otherwise, we may inconsistently attempt to release
the advisory lock.
Robert Watson [Sat, 25 Oct 2003 16:14:09 +0000 (16:14 +0000)]
When generate a core dump, use advisory locking in an advisory way:
if we do acquire an advisory lock, great! We'll release it later.
However, if we fail to acquire a lock, we perform the coredump
anyway. This problem became particularly visible with NFS after
the introduction of rpc.lockd: if the lock manager isn't running,
then locking calls will fail, aborting the core dump (resulting in
a zero-byte dump file).
Robert Watson [Sat, 25 Oct 2003 15:28:20 +0000 (15:28 +0000)]
Make MAC_EXTERNALIZE() and MAC_INTERNALIZE() simply take the object
type, rather than "object_label" as the first argument. This reduces
complexity a little for the consumer, and also makes it easier for
use to rename the underlying entry points in struct mac_policy_obj.
- fabsl.c should be named s_fabsl.c for consistency with libmsun's
documented naming scheme (unfortunately the documentation isn't in the
tree as far as I can tell); no repocopy is required as there is no
history to preserve.
- replace simple and almost-correct implementation with slightly hackish
but definitely correct implementation (tested on i386, alpha, sparc64)
which requires pulling in fpmath.h and the MD _fpmath.h from libc.
- try not to make a mess of the Makefile in the process.
- enterprising minds are encouraged to implement more C99 long double
functions.
Nate Lawson [Sat, 25 Oct 2003 05:03:25 +0000 (05:03 +0000)]
Add devctl(4) notify support to ACPI. Various subsystems now notify
userland whenever events occur. See the example in devd.conf below
to see how to use it.
Warner Losh [Sat, 25 Oct 2003 04:05:33 +0000 (04:05 +0000)]
Start to minimize diffs between vx and ep. These latter is based on a
more advanced version of the chips supported by the former. Matt Dodd
and I are working towards merging them, and this a step on that path.