]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
26 years agoMF22: add ports-biology
jkh [Fri, 6 Mar 1998 13:36:05 +0000 (13:36 +0000)]
MF22: add ports-biology

26 years agoFixed `make -jN' for large N. The usual method of pretending that all
bde [Fri, 6 Mar 1998 13:34:36 +0000 (13:34 +0000)]
Fixed `make -jN' for large N.  The usual method of pretending that all
objects depend on all generated headers doesn't work because it gives
cyclic dependencies.  Give enough dependencies explicitly.  We no
longer need to use .SINGLESHELL for `make depend'.  .SINGLESHELL was
more of a bottleneck than usual because `make depend' makes everything.

Fixed some spelling and English errors.

26 years agoFix fprintf() here too.
jkh [Fri, 6 Mar 1998 13:29:37 +0000 (13:29 +0000)]
Fix fprintf() here too.

26 years ago\e should have been \n (don't know when this was broken).
jkh [Fri, 6 Mar 1998 13:28:17 +0000 (13:28 +0000)]
\e should have been \n (don't know when this was broken).

26 years agoFixed `make -jN' for large N. Just put the generated headers in
bde [Fri, 6 Mar 1998 13:13:15 +0000 (13:13 +0000)]
Fixed `make -jN' for large N.  Just put the generated headers in
${SRCS} instead of giving inadequate explicit dependencies.  There
is still a problem after `make depend; make clean'.  Then `make'
barely works, and `make -jN' is confused by absolute paths in
.depend.

26 years agoTrivial filesystem getpages/putpages implementations, set the second.
msmith [Fri, 6 Mar 1998 09:46:52 +0000 (09:46 +0000)]
Trivial filesystem getpages/putpages implementations, set the second.
These should be considered the first steps in a work-in-progress.
Submitted by: Terry Lambert <terry@freebsd.org>

26 years agoFixed building with `make -jN'. Put ss_err.h in SRCS so that recent
bde [Fri, 6 Mar 1998 08:26:00 +0000 (08:26 +0000)]
Fixed building with `make -jN'.  Put ss_err.h in SRCS so that recent
changes to bsd.lib.mk can handle building it early enough.  Don't
use the same rule for ss_err.h and ss_err.c, else `make -jN' would
run the rule twice concurrently.  Don't put ss_err.c out of order
in SRCS; doing so was a kludge to get ss_err.h built early enough
for plain `make'.

Don't put a non-generated file in CLEANFILES.

26 years agoIf .depend doesn exist, then pretend that all objects depend on all
bde [Fri, 6 Mar 1998 08:08:36 +0000 (08:08 +0000)]
If .depend doesn exist, then pretend that all objects depend on all
headers in ${SRCS}, as in bsd.lmod.mk and bsd.prog.mk.  This helps
`make [-j]' work when .depend doesn't exist.  Even plain `make'
sometimes only worked because of magic ordering in ${SRCS}.

26 years agoRemoved support for now-unused FreeBSD misfeature DPSRCS.
bde [Fri, 6 Mar 1998 07:28:01 +0000 (07:28 +0000)]
Removed support for now-unused FreeBSD misfeature DPSRCS.

26 years agoMF22: various doc bogons noticed.
jkh [Fri, 6 Mar 1998 07:23:08 +0000 (07:23 +0000)]
MF22: various doc bogons noticed.

26 years agoDon't use the FreeBSD misfeature DPSRCS or give explicit dependencies
bde [Fri, 6 Mar 1998 07:00:28 +0000 (07:00 +0000)]
Don't use the FreeBSD misfeature DPSRCS or give explicit dependencies
on generated headers.  Just put generated headers in SRCS so that
bsd.*.mk can generate better dependencies.

26 years agoFixed dependency of ${PROG} on ${DPSRCS}. There were various races
bde [Fri, 6 Mar 1998 06:48:39 +0000 (06:48 +0000)]
Fixed dependency of ${PROG} on ${DPSRCS}.  There were various races
for `make -j'.  There are one-off fixes for the problem in a few
Makefiles, e.g., amd/amq/Makefile.

26 years agoGo to reduced device set now that "make-unders" work.
jkh [Fri, 6 Mar 1998 06:47:09 +0000 (06:47 +0000)]
Go to reduced device set now that "make-unders" work.

26 years agoFixed dependencies of ${PROG} on ${DPSRCS} and `all' on ${_ILINKS}.
bde [Fri, 6 Mar 1998 05:43:46 +0000 (05:43 +0000)]
Fixed dependencies of ${PROG} on ${DPSRCS} and `all' on ${_ILINKS}.
There were various races for `make -j'.  Half-built versions of
vnode_if.h were sometimes #included...

Removed a null dependency.

26 years agoWe don't need to NUL terminate our sun_path.
brian [Fri, 6 Mar 1998 03:10:49 +0000 (03:10 +0000)]
We don't need to NUL terminate our sun_path.
Pointed out by: Garrett Wollman <wollman@khavrinen.lcs.mit.edu>

26 years agoAdd 4th button support for the "MouseMan" protocol. The latest
yokota [Fri, 6 Mar 1998 03:09:31 +0000 (03:09 +0000)]
Add 4th button support for the "MouseMan" protocol.  The latest
trackball model from Logitech has four buttons.

26 years agoNul terminate sockaddr_un::sun_path
brian [Fri, 6 Mar 1998 02:12:02 +0000 (02:12 +0000)]
Nul terminate sockaddr_un::sun_path
Suggested by: Theo de Raadt <deraadt@openbsd.org>

26 years agoTeruillian --> Tertullian
alex [Fri, 6 Mar 1998 01:49:14 +0000 (01:49 +0000)]
Teruillian --> Tertullian

PR: 5928
Submitted by: Doug <Studded@dal.net>

26 years agoUse the correct device name in -direct mode.
brian [Fri, 6 Mar 1998 00:35:30 +0000 (00:35 +0000)]
Use the correct device name in -direct mode.

26 years agoFix the broken installation of strip. It tried to use the installed
jdp [Fri, 6 Mar 1998 00:28:04 +0000 (00:28 +0000)]
Fix the broken installation of strip.  It tried to use the installed
strip program (via "install") to strip itself.  But the program
wasn't executable because "install" hadn't made it so yet.  I
borrowed the method used for the old strip to get around this.

26 years agoMake SyslogAddr a sockaddr_un rather than a sockaddr.
brian [Thu, 5 Mar 1998 22:17:59 +0000 (22:17 +0000)]
Make SyslogAddr a sockaddr_un rather than a sockaddr.
This wasn't a problem in practice as PATH_LOG and PATH_OLDLOG
are both < sizeof sockaddr::sa_data.

26 years agoRemove special handling for resuming clock interrupt when using APIC_IO.
tegge [Thu, 5 Mar 1998 21:45:53 +0000 (21:45 +0000)]
Remove special handling for resuming clock interrupt when using APIC_IO.
The `generic' vector stubs do the right thing.

26 years agoUse t_idt instead of idt inside setidt() if f00f_hack() has relocated the IDT.
tegge [Thu, 5 Mar 1998 19:37:03 +0000 (19:37 +0000)]
Use t_idt instead of idt inside setidt() if f00f_hack() has relocated the IDT.
Submitted by: Bruce Evans <bde@zeta.org.au>

26 years ago fix a type in the rulesets
jmb [Thu, 5 Mar 1998 18:38:57 +0000 (18:38 +0000)]
fix a type in the rulesets
same fix as recorded in misc/5924
that problem report was filed against the stable branch
but that's not a reason not to fix it here as well ;)
Reviewed by: jmb
Submitted by: Ruslan Ermilov <ru@ucb.crimea.ua>

26 years agoCorrect minor typo
jkh [Thu, 5 Mar 1998 10:53:26 +0000 (10:53 +0000)]
Correct minor typo
Confirmed by: Ustimenko Semen <semen@iclub.nsu.ru>

26 years agoFixed mistakenly merged part.
kato [Thu, 5 Mar 1998 03:31:27 +0000 (03:31 +0000)]
Fixed mistakenly merged part.

Submitted by: NOKUBI Hirotaka <hnokubi@yyy.or.jp>

26 years agoReconized OpenBSD partitions.
obrien [Thu, 5 Mar 1998 03:08:16 +0000 (03:08 +0000)]
Reconized OpenBSD partitions.

26 years agoDisable redirection of clock interrupt to a higher priority vector.
tegge [Wed, 4 Mar 1998 23:45:12 +0000 (23:45 +0000)]
Disable redirection of clock interrupt to a higher priority vector.
This does not work on most dual Pentium machines, due to setidt()
being neutralized by f00f_hack().

26 years agoIf numdirs is zero, print a helpful message instead of divding by zero later.
phk [Wed, 4 Mar 1998 18:03:54 +0000 (18:03 +0000)]
If numdirs is zero, print a helpful message instead of divding by zero later.

26 years ago*** empty log message ***
semenu [Wed, 4 Mar 1998 16:35:05 +0000 (16:35 +0000)]
*** empty log message ***

26 years agoDefined CCR6 and CCR7 (configuration registers of M2 CPU.)
kato [Wed, 4 Mar 1998 11:39:16 +0000 (11:39 +0000)]
Defined CCR6 and CCR7 (configuration registers of M2 CPU.)

26 years agoReviewed by: msmith, bde long ago
dufault [Wed, 4 Mar 1998 10:27:00 +0000 (10:27 +0000)]
Reviewed by: msmith, bde long ago
POSIX.4 headers and sysctl variables.  Nothing should change
unless POSIX4 is defined or _POSIX_VERSION is set to 199309.

26 years agoReviewed by: msmith, bde long ago
dufault [Wed, 4 Mar 1998 10:25:55 +0000 (10:25 +0000)]
Reviewed by: msmith, bde long ago
Fix for RTPRIO scheduler to eliminate invalid context switches.

POSIX.4 headers and sysctl variables.  Nothing should change
unless POSIX4 is defined or _POSIX_VERSION is set to 199309.

26 years agoReviewed by: msmith, bde long ago
dufault [Wed, 4 Mar 1998 10:25:03 +0000 (10:25 +0000)]
Reviewed by: msmith, bde long ago
Fix for RTPRIO scheduler to eliminate invalid context switches.

26 years agoFix recent breakage when compiling under elf.
peter [Wed, 4 Mar 1998 09:55:13 +0000 (09:55 +0000)]
Fix recent breakage when compiling under elf.
_Xforward_irq         -> Xforward_irq
_apic_isrbit_location -> apic_isrbit_location
_cpu_num_to_apic_id   -> cpu_num_to_apic_id
_invltlb_ok           -> invltlb_ok

26 years agoFix certain kinds of block device operations. For example, tunefs on
dyson [Wed, 4 Mar 1998 06:44:59 +0000 (06:44 +0000)]
Fix certain kinds of block device operations.  For example, tunefs on
a block device shouldn't crash the system anymore.

26 years agoFix a rounding error for the NFS buffer validend.
dyson [Wed, 4 Mar 1998 03:17:30 +0000 (03:17 +0000)]
Fix a rounding error for the NFS buffer validend.
Submitted by: John W. De Boskey <jwd@unx.sas.com>

26 years agoWhen entering the apic version of slow interrupt handler, level
tegge [Tue, 3 Mar 1998 22:56:30 +0000 (22:56 +0000)]
When entering the apic version of slow interrupt handler, level
interrupts are masked, and EOI is sent iff the corresponding ISR bit
is set in the local apic. If the CPU cannot obtain the interrupt
service lock (currently the global kernel lock) the interrupt is
forwarded to the CPU holding that lock.

Clock interrupts now have higher priority than other slow interrupts.

26 years agoForward the signal if the process runs on a different CPU. This reduces
tegge [Tue, 3 Mar 1998 20:55:26 +0000 (20:55 +0000)]
Forward the signal if the process runs on a different CPU.  This reduces
the signal handling latency for cpu-bound processes that performs very
few system calls.

The IPI for forcing an additional software trap is no longer dependent upon
BETTER_CLOCK being defined.

26 years agoReduce timeout before assuming that forwarding of hardclock or softclock
tegge [Tue, 3 Mar 1998 20:09:14 +0000 (20:09 +0000)]
Reduce timeout before assuming that forwarding of hardclock or softclock
failed. Don't complain on forwarding failure, unless
BETTER_CLOCK_DIAGNOSTIC is defined.

26 years agoWhen sending an IPI to a specific target, disable interrupts inside the
tegge [Tue, 3 Mar 1998 19:54:49 +0000 (19:54 +0000)]
When sending an IPI to a specific target, disable interrupts inside the
critical region in order to avoid sending the IPI to the wrong target.

26 years agoforward_statclock and forward_hardclock are located in mp_machdep.c.
tegge [Tue, 3 Mar 1998 19:44:34 +0000 (19:44 +0000)]
forward_statclock and forward_hardclock are located in mp_machdep.c.

26 years agoFixed uninitialized pointer in previous commit. mktemp() was broken.
bde [Tue, 3 Mar 1998 14:38:36 +0000 (14:38 +0000)]
Fixed uninitialized pointer in previous commit.  mktemp() was broken.
I noticed cvs core dumps and uncleaned cvs temporary files in /tmp.

Fixed ANSIisms.

26 years agoAdded support for the 82553 and 'B' 82555 PHY.
dg [Tue, 3 Mar 1998 14:19:09 +0000 (14:19 +0000)]
Added support for the 82553 and 'B' 82555 PHY.

26 years agoUpdate stand-alone DOS I/O routines to be able to read from FAT32/VFAT
jkh [Tue, 3 Mar 1998 04:29:01 +0000 (04:29 +0000)]
Update stand-alone DOS I/O routines to be able to read from FAT32/VFAT
partitions.  This should allow sysinstall to function properly in all
Win95 installation based scenarios now.
Submitted by: Robert Nordier <rnordier@iafrica.com>

26 years agoPatch to the last commit; attempt to unspam stuff from NetBSD.
msmith [Tue, 3 Mar 1998 02:50:24 +0000 (02:50 +0000)]
Patch to the last commit; attempt to unspam stuff from NetBSD.
Submitted by: Dmitrij Tejblum <dima@tejblum.dnttm.rssi.ru>

26 years agoDon't assume sigset_t and int are equivalent.
jraynard [Mon, 2 Mar 1998 23:19:29 +0000 (23:19 +0000)]
Don't assume sigset_t and int are equivalent.

26 years agobit_nset was not correctly used.
guido [Mon, 2 Mar 1998 20:51:06 +0000 (20:51 +0000)]
bit_nset was not correctly used.

26 years agoFix use of bit_nclear. Amazing that we didn't get more complaints
guido [Mon, 2 Mar 1998 19:00:01 +0000 (19:00 +0000)]
Fix use of bit_nclear. Amazing that we didn't get more complaints
about inserting 2 pcmcia cards.

26 years agoSync with sys/i386/i386/machdep.c revision 1.290.
kato [Mon, 2 Mar 1998 10:02:01 +0000 (10:02 +0000)]
Sync with sys/i386/i386/machdep.c revision 1.290.

26 years agoUpdate the information in this file slightly.
jkh [Mon, 2 Mar 1998 07:35:48 +0000 (07:35 +0000)]
Update the information in this file slightly.

26 years agoUpdate the ELF image activator to use some of the exec resources rather
peter [Mon, 2 Mar 1998 05:47:58 +0000 (05:47 +0000)]
Update the ELF image activator to use some of the exec resources rather
than rolling it's own.  This means that it now uses the "safe"
exec_map_first_page() to get the ld.so headers rather than risking a panic
on a page fault failure (eg: NFS server goes down).
Since all the ELF tools go to a lot of trouble to make sure everything
lives in the first page for executables, this is a win.  I have not seen
any ELF executable on any system where all the headers didn't fit in the
first page with lots of room to spare.
I have been running variations of this code for some time on my pure ELF
systems.

26 years agosyntax error in /usr/share/examples/sliplogin/slip.hosts
jkh [Mon, 2 Mar 1998 02:52:38 +0000 (02:52 +0000)]
syntax error in /usr/share/examples/sliplogin/slip.hosts
PR: 5873
Submitted by: Ruslan Ermilov <ru@ucb.crimea.ua>

26 years agoThis commit was generated by cvs2svn to compensate for changes in r33975,
jdp [Sun, 1 Mar 1998 23:41:17 +0000 (23:41 +0000)]
This commit was generated by cvs2svn to compensate for changes in r33975,
which included commits to RCS files with non-trunk default branches.

26 years agoAdd bmakefiles for building binutils from the contrib tree.
jdp [Sun, 1 Mar 1998 23:41:17 +0000 (23:41 +0000)]
Add bmakefiles for building binutils from the contrib tree.

This finishes up the binutils import.  But I am leaving it disabled
in "src/gnu/usr.bin/Makefile" for now.  It is not used by anything
yet, so I'll take this opportunity to run one more round of tests
before enabling it.

26 years agoConfiguration changes to support FreeBSD. The linker only works for
jdp [Sun, 1 Mar 1998 23:26:32 +0000 (23:26 +0000)]
Configuration changes to support FreeBSD.  The linker only works for
ELF, not a.out.  But all the other tools support both.  I'm not so
sure about the assembler -- it might need more work.

26 years agoDon't use "/" as a comment starter. In some FreeBSD sources, "/"
jdp [Sun, 1 Mar 1998 23:22:00 +0000 (23:22 +0000)]
Don't use "/" as a comment starter.  In some FreeBSD sources, "/"
means divide (duh).

26 years agoFix the output format of "nm" so that it works properly for FreeBSD's
jdp [Sun, 1 Mar 1998 23:20:03 +0000 (23:20 +0000)]
Fix the output format of "nm" so that it works properly for FreeBSD's
"lorder" command.

GNU rejected this change because they felt it would be an unacceptable
incompatibility to change the output format at this point -- even
though it was wrong to begin with.

26 years agoFix a bug which incorrectly caused files to be labeled as having
jdp [Sun, 1 Mar 1998 23:17:03 +0000 (23:17 +0000)]
Fix a bug which incorrectly caused files to be labeled as having
text relocations in certain situations.

26 years agoAdd automatic branding of FreeBSD ELF files.
jdp [Sun, 1 Mar 1998 23:15:09 +0000 (23:15 +0000)]
Add automatic branding of FreeBSD ELF files.

26 years agoAdd upgrade instructions and helper file.
jdp [Sun, 1 Mar 1998 23:13:26 +0000 (23:13 +0000)]
Add upgrade instructions and helper file.

26 years agoChange vfs.ioopt default back to '0'.
dyson [Sun, 1 Mar 1998 23:07:45 +0000 (23:07 +0000)]
Change vfs.ioopt default back to '0'.

26 years agoThis commit was generated by cvs2svn to compensate for changes in r33965,
jdp [Sun, 1 Mar 1998 22:58:51 +0000 (22:58 +0000)]
This commit was generated by cvs2svn to compensate for changes in r33965,
which included commits to RCS files with non-trunk default branches.

26 years agoInitial import of GNU binutils version 2.8.1. Believe it or not,
jdp [Sun, 1 Mar 1998 22:58:51 +0000 (22:58 +0000)]
Initial import of GNU binutils version 2.8.1.  Believe it or not,
this is heavily stripped down.

26 years agoThe intent is to get rid of WILLRELE in vnode_if.src by making
msmith [Sun, 1 Mar 1998 22:46:53 +0000 (22:46 +0000)]
The intent is to get rid of WILLRELE in vnode_if.src by making
a complement to all ops that return a vpp, VFS_VRELE.  This is
initially only for file systems that implement the following ops
that do a WILLRELE:

vop_create, vop_whiteout, vop_mknod, vop_remove, vop_link,
vop_rename, vop_mkdir, vop_rmdir, vop_symlink

This is initial DNA that doesn't do anything yet.  VFS_VRELE is
implemented but not called.

A default vfs_vrele was created for fs implementations that use the
standard vnode management routines.

VFS_VRELE implementations were made for the following file systems:

Standard (vfs_vrele)
ffs mfs nfs msdosfs devfs ext2fs

Custom
union umapfs

Just EOPNOTSUPP
fdesc procfs kernfs portal cd9660

These implementations may change as VOP changes are implemented.

In the next phase, in the vop implementations calls to vrele and the vrele
part of vput will be moved to the top layer vfs_vnops and made visible
to all layers.  vput will be replaced by unlock in these cases.  Unlocking
will still be done in the per fs layer but the refcount decrement will be
triggered at the top because it doesn't hurt to hold a vnode reference a
little longer.  This will have minimal impact on the structure of the
existing code.

This will only be done for vnode arguments that are released by the various
fs vop implementations.

Wider use of VFS_VRELE will likely require restructuring of the code.

Reviewed by: phk, dyson, terry et. al.
Submitted by: Michael Hancock <michaelh@cet.co.jp>

26 years agoAdd new directories needed by the ELF tools.
jdp [Sun, 1 Mar 1998 22:45:44 +0000 (22:45 +0000)]
Add new directories needed by the ELF tools.

26 years agoCreate the cuaiA* and cualA* devices.
steve [Sun, 1 Mar 1998 22:18:28 +0000 (22:18 +0000)]
Create the cuaiA* and cualA* devices.

PR: 5872
Submitted by: Adam Baran <badam@mw.mil.pl>

26 years agoFix mmap() on msdosfs. In the words of the submitter:
msmith [Sun, 1 Mar 1998 21:26:09 +0000 (21:26 +0000)]
Fix mmap() on msdosfs.  In the words of the submitter:

|In the process of evaluating the getpages/putpages issues I discovered
|that mmap on MSDOSFS does not work. This is because I blindly merged
|NetBSD changes in msdosfs_bmap and msdosfs_strategy. Apparently, their
|blocksize is always DEV_BSIZE (even in files), while in FreeBSD
|blocksize in files is v_mount->mnt_stat.f_iosize (i.e. clustersize in
|MSDOSFS case). The patch is below.

Submitted by: Dmitrij Tejblum <dima@tejblum.dnttm.rssi.ru>

26 years agoAndreas Klemm reported that 2 hours are too few to format a 9 GB
joerg [Sun, 1 Mar 1998 20:15:00 +0000 (20:15 +0000)]
Andreas Klemm reported that 2 hours are too few to format a 9 GB
Barracuda.  Bump the timeout to 3 hours.

26 years agoMake sure that you can only bind a more specific address when it is
guido [Sun, 1 Mar 1998 19:39:29 +0000 (19:39 +0000)]
Make sure that you can only bind a more specific address when it is
done by the same uid.
Obtained from: OpenBSD

26 years agoFix a few minor typos and style nits.
steve [Sun, 1 Mar 1998 18:58:03 +0000 (18:58 +0000)]
Fix a few minor typos and style nits.

PR: 5883
Submitted by: hnokubi@yyy.or.jp

26 years agoFirst pre alpha Work In Progress commit of a IDE/ATAPI tape driver.
sos [Sun, 1 Mar 1998 18:57:29 +0000 (18:57 +0000)]
First pre alpha Work In Progress commit of a IDE/ATAPI tape driver.

It does endeed work, but there is still some problems to solve.
I get a "nonrecovered data error" from time to time, but besides
this it has backed up several Gigs allready.

Please report any success/failure directly to me.

Thanks to Warner Losh for providing a drive to use in writing
this driver!

26 years agoReplace previous commit with a check disallowing ptr from running
steve [Sun, 1 Mar 1998 18:49:37 +0000 (18:49 +0000)]
Replace previous commit with a check disallowing ptr from running
off the end of the list variable.

PR: 5345, 5610
Submitted by: nagao@cs.titech.ac.jp

26 years agoUndo the previous commit which was NOT for -current.
sos [Sun, 1 Mar 1998 17:29:25 +0000 (17:29 +0000)]
Undo the previous commit which was NOT for -current.
Rearrange a few lines for better order.

26 years agoAdd Intel 430TX chipset, namely 82439TX system controller and four
asami [Sun, 1 Mar 1998 10:10:31 +0000 (10:10 +0000)]
Add Intel 430TX chipset, namely 82439TX system controller and four
incarnations of 82371AB (P/I bridge, IDE, USB and power management).

Tested by: jkh

26 years agoAdd more info about special characters in dos2unix/unix2dos tables
ache [Sun, 1 Mar 1998 08:41:05 +0000 (08:41 +0000)]
Add more info about special characters in dos2unix/unix2dos tables

26 years agoRestored half of the ANSI semantic errors fix which was clobbered
bde [Sun, 1 Mar 1998 06:06:20 +0000 (06:06 +0000)]
Restored half of the ANSI semantic errors fix which was clobbered
by the previous commit.

26 years agoFixed syntax error in previous commit.
bde [Sun, 1 Mar 1998 06:01:33 +0000 (06:01 +0000)]
Fixed syntax error in previous commit.

26 years agoSync with sys/i386/isa/clock.c revision 1.114.
kato [Sun, 1 Mar 1998 05:22:25 +0000 (05:22 +0000)]
Sync with sys/i386/isa/clock.c revision 1.114.

26 years agoWargh! Who went and changed all the getopt() comparisons from -1 to
jkh [Sun, 1 Mar 1998 05:10:28 +0000 (05:10 +0000)]
Wargh!  Who went and changed all the getopt() comparisons from -1 to
EOF?  The getopt(3) manpage clearly states that the return value is *-1*,
not EOF!  Besides, getopt(3) isn't reading from a file. :)

Noticed-while: merging to 2.2 (where this is correct).

26 years ago1) Use a more consistent page wait methodology.
dyson [Sun, 1 Mar 1998 04:18:54 +0000 (04:18 +0000)]
1) Use a more consistent page wait methodology.
2) Do not unnecessarily force page blocking when paging
pages out.
3) Further improve swap pager performance and correctness,
including fixing the paging in progress deadlock (except
in severe I/O error conditions.)
4) Enable vfs_ioopt=1 as a default.
5) Fix and enable the page prezeroing in SMP mode.

All in all, SMP systems especially should show a significant
improvement in "snappyness."

26 years agoStyle nits and staticism with the previous commit.
msmith [Sun, 1 Mar 1998 01:37:38 +0000 (01:37 +0000)]
Style nits and staticism with the previous commit.
Submitted by: bde

26 years agoAdd local stup putpages/getpages routines.
msmith [Sun, 1 Mar 1998 00:51:43 +0000 (00:51 +0000)]
Add local stup putpages/getpages routines.
Submitted by: Terry Lambert <terry@freebsd.org>

26 years agoUse .Nm rdist for second .Nm occurrence in synopsis.
alex [Sun, 1 Mar 1998 00:18:25 +0000 (00:18 +0000)]
Use .Nm rdist for second .Nm occurrence in synopsis.

26 years agoUse `.Nm id' instead of `.Nm' for the synopsis lines.
alex [Sat, 28 Feb 1998 21:37:54 +0000 (21:37 +0000)]
Use `.Nm id' instead of `.Nm' for the synopsis lines.

PR: 5876

26 years agoPrevent the TSC from being used on APM machines, we have no idea if
phk [Sat, 28 Feb 1998 21:16:13 +0000 (21:16 +0000)]
Prevent the TSC from being used on APM machines, we have no idea if
it runs at a constant frequency.  This was less of an issue before,
because the TSC only interpolated in the HZ intervals, but now where
the timecounter is used all the way, this becomes much more visible.

Nit: Fix a printf which triggered the bde-filter.

26 years agoMake it possible to indicate that we don't care about the remote
phk [Sat, 28 Feb 1998 21:01:09 +0000 (21:01 +0000)]
Make it possible to indicate that we don't care about the remote
sides IP address, as long as it isn't 0.0.0.0

26 years agoStrings are terminated by NUL, not NULL.
jraynard [Sat, 28 Feb 1998 18:05:42 +0000 (18:05 +0000)]
Strings are terminated by NUL, not NULL.

26 years agoDon't assume sigset_t and int are equivalent.
jraynard [Sat, 28 Feb 1998 16:02:24 +0000 (16:02 +0000)]
Don't assume sigset_t and int are equivalent.

26 years agoFix a few typos.
jraynard [Sat, 28 Feb 1998 15:37:25 +0000 (15:37 +0000)]
Fix a few typos.

26 years agoStrings are terminated by NUL, not NULL.
jraynard [Sat, 28 Feb 1998 15:35:22 +0000 (15:35 +0000)]
Strings are terminated by NUL, not NULL.

26 years agoDon't assume sigset_t and int are equivalent. Also, get
jraynard [Sat, 28 Feb 1998 15:14:00 +0000 (15:14 +0000)]
Don't assume sigset_t and int are equivalent.  Also, get
sigprocmask arguments the right way round.

26 years agoDon't assume sigset_t and int are equivalent.
jraynard [Sat, 28 Feb 1998 10:50:02 +0000 (10:50 +0000)]
Don't assume sigset_t and int are equivalent.

26 years agoInstall manpage too
ache [Sat, 28 Feb 1998 10:41:01 +0000 (10:41 +0000)]
Install manpage too

26 years agoSync with sys/i386/isa/if_ed.c revision 1.136.
kato [Sat, 28 Feb 1998 07:38:18 +0000 (07:38 +0000)]
Sync with sys/i386/isa/if_ed.c revision 1.136.

26 years agoUpdate to reflect reality. We do support the WANic 400/405 PCI cards.
jhay [Sat, 28 Feb 1998 07:06:25 +0000 (07:06 +0000)]
Update to reflect reality. We do support the WANic 400/405 PCI cards.
Fix some typos and clean it up a little.

26 years agofoo.gz is created by 'make test' so add it to CLEANFILES.
steve [Sat, 28 Feb 1998 06:27:59 +0000 (06:27 +0000)]
foo.gz is created by 'make test' so add it to CLEANFILES.

26 years agoalgorithm.doc renamed to algorithm.txt in new distribution.
steve [Sat, 28 Feb 1998 06:10:08 +0000 (06:10 +0000)]
algorithm.doc renamed to algorithm.txt in new distribution.

26 years agoMerge conflicts.
steve [Sat, 28 Feb 1998 06:08:17 +0000 (06:08 +0000)]
Merge conflicts.

26 years agoThis commit was generated by cvs2svn to compensate for changes in r33904,
steve [Sat, 28 Feb 1998 06:04:26 +0000 (06:04 +0000)]
This commit was generated by cvs2svn to compensate for changes in r33904,
which included commits to RCS files with non-trunk default branches.