Mark Murray [Wed, 25 Sep 2002 07:24:01 +0000 (07:24 +0000)]
From the requestor:
"Could you do me a favor and fix sys_bsd.c to get the howmany() macro
from <sys/param.h>, instead of <sys/types.h>? This will save me from
having to worry about the unsync'd bits before making the change."
Scott Long [Wed, 25 Sep 2002 05:00:25 +0000 (05:00 +0000)]
The AAC_COMPAT_LINUX option was really annoying, since it made the
aac driver dependent on the linux emulation module. This was
especially bad for the release engineers who tried to move the
aac driver from the kernel onto the drivers floppy. The linux
compat bits for this driver are now in their own driver, aac_linux.
It can be loaded as a module or compiled into the kernel. For
the latter case, the AAC_COMPAT_LINUX option is needed, along with
the COMPAT_LINUX option.
I've tested this in every configuration I can think of. This is an
MFC candidate for 4.7.
Bruce A. Mah [Wed, 25 Sep 2002 04:36:07 +0000 (04:36 +0000)]
Accomodate change in the output of "make package-depends" that came
with ports/Mk/bsd.ports.mk rev. 1.424. (It now prints the origin
after the package name.)
Jeff Roberson [Wed, 25 Sep 2002 02:32:42 +0000 (02:32 +0000)]
- Use vrefcnt() where it is safe to do so instead of doing direct and
unlocked accesses to v_usecount.
- Lock access to the buf lists in the various sync routines. interlock
locking could be avoided almost entirely in leaf filesystems if the
fsync function had a generic helper.
Jeff Roberson [Wed, 25 Sep 2002 02:22:21 +0000 (02:22 +0000)]
- Lock down the syncer with sync_mtx.
- Enable vfs_badlock_mutex by default.
- Assert that the vp is locked in VOP_UNLOCK.
- Use standard interlock macros in remaining code.
- Correct a race in getnewvnode().
- Lock access to v_numoutput with interlock.
- Lock access to buf lists and splay tree with interlock.
- Add VOP and VI asserts.
- Lock b_vnbufs with the vnode interlock.
- Add vrefcnt() for callers who want to retreive the vnode ref without
holding a lock. Add a comment that describes when this is safe.
- Add vholdl() and vdropl() so that callers who already own the interlock
can avoid race conditions and unnecessary unlocking.
- Move the VOP_GETATTR() in vflush() into the WRITECLOSE conditional case.
- Hold the interlock before droping the mntlist_mtx in vflush() to avoid
a race.
- Fix locking in vfs_msync().
Tony Finch [Tue, 24 Sep 2002 19:27:44 +0000 (19:27 +0000)]
It is probably a bad idea to unconditionally process directives with
constant controlling expressions: in particular, removing #if 0 sections
is considered "rude". This commit changes the default so that such
things are passed through unchanged, and the old behaviour can be had
with the -k "kill konsts" flag.
Correct the description of what fortunes get installed (by default, also the
offensive ones). Make it clear that datfiles/Makefile needs to be edited in
case someone wants it otherwise.
PR: misc/36153
Submitted by: Nick Jennings <nick@namodn.com>
X-MFC after: 4.7 is out the door.
Brooks Davis [Tue, 24 Sep 2002 17:35:08 +0000 (17:35 +0000)]
Add a new helper function if_printf() modeled on device_printf(). The
function takes a struct ifnet pointer followed by the usual printf
arguments and prints "<interfacename>: " before the results of printf.
Since this is the primary form of printf calls in network device drivers
and accounts for most uses of the ifnet menber if_unit, this
significantly simplifies many printf()s.
Tim J. Robbins [Tue, 24 Sep 2002 09:25:37 +0000 (09:25 +0000)]
Warn when setinvalidrune() is referenced for consistency with the rest
of the rune functions (except sgetrune() and sputrune(), which are really
macros).
Tim J. Robbins [Tue, 24 Sep 2002 09:18:32 +0000 (09:18 +0000)]
Remove an unneeded call to _sfrefill() that was missed in the conversion
from vfscanf() to vfwscanf(). It doesn't hurt to have it there, but it's
redundant since __fgetwc() will refill the buffer if it needs to.
Tim J. Robbins [Tue, 24 Sep 2002 00:47:27 +0000 (00:47 +0000)]
Use the new va_copy macro to copy variable argument lists instead of
assignment. This is needed on powerpc but is also more correct for the
other ports.
Submitted by: grehan
Tested on: alpha, i386, sparc64
Matt Jacob [Mon, 23 Sep 2002 19:41:10 +0000 (19:41 +0000)]
Parameterize MPT_MAX_REQUESTS based upon device type (FC has Global Credits
of 1024- Ultra4 256). Rename 'requests' tag to 'request_pool' for clarity.
Make sure we do correct xpt_freeze_simq/CAM_RELEASE_SIMQ if we run out
of chip resources.
Mike Barcroft [Mon, 23 Sep 2002 17:45:51 +0000 (17:45 +0000)]
o Move select() helper macros from <sys/types.h> to <sys/select.h>.
o Include <sys/select.h> from <sys/types.h> in the __BSD_VISIBLE case,
so applications and base software can be slowly updated.
o Prototype select() in <sys/select.h>. It was previously only
prototyped in <unistd.h>.
o Add some XXX's to <sys/types.h>.
John Baldwin [Mon, 23 Sep 2002 16:03:00 +0000 (16:03 +0000)]
Update the nexus driver for the addition of the legacy driver:
- nexus no longer has PCI bridges as direct children, so the PCI bus
ivar is no longer used and is removed.
- Don't attach default EISA, ISA, or MCA busses. Instead, if we do not
have an acpi0 device after bus_generic_probe(), add a legacy0 child
device.
- Remove machine/nexusvar.h.
John Baldwin [Mon, 23 Sep 2002 15:50:06 +0000 (15:50 +0000)]
Add a new legacy(4) device driver for use on machines that do not have
ACPI or for when ACPI support is disabled or not present in the kernel.
Basically, the nexus device is now split into two with some parts
(such as adding default ISA, MCA, and EISA busses if they aren't found
as well as support for PCI bus device ivars) being moved to the legacy
driver.