Don't call getmntopts with a null pointer as an option string; it'll
dereference it. Pass optarg from getopts as it should have been done.
This bug caused mount_cd9660 to segfault whenever -o was used.
phk [Tue, 2 Aug 1994 05:22:24 +0000 (05:22 +0000)]
This is the tcl-scripts which will convert gcc-2.6.0 to bmake form,
a patch to gcc from Bruce, and to libg++ by me. I will check in the
actual compiler when Jordan has put in the rest of src/gnu.
fixed bug where large amounts of unidirectional UDP traffic would fill
the interface output queue and further udp packets would be fragmented
and only partially sent - keeping the output queue full and jamming the
network, but not actually getting any real work done (because you can't
send just 'part' of a udp packet - if you fragment it, you must send
the whole thing). The fix involves adding a check to make sure that the
output queue has sufficient space for all of the fragments.
Fixed bug with Nagel Congestion Avoidance where a tcp connection would
stall unnecessarily - always send an ACK when a packet len of < mss is
received.
Documented the new CVS template. This really isn't the right place for it,
but I wanted to get this out for review without an interminable debate about
where it really should go.
Reviewed by: phk
Imported libmd. This library contains MD2, MD4 and MD5.
These three boggers pop up all over the place all of the time, so I
decided we needed a library with them. In general they are used for
security checks, so if you use them you want to link them static.
The FreeBSD bootloader loads the kernel at the a_entry address in the
kernel a.out file, meaning that this is where the kernel starts.
(not at KERNBASE) - KERNBASE is 0xf0000000, while the kernel loads at
0xf0100000
Reviewed by:
Rewrite nlist to mmap the whole a.out file (at Davidg's suggestion).
This means that we don't have to do rounding calculations for page
boundaries. (We do all our accesses via the mmapped area now.)
Reviewed by:
Added code to allocate and deallocate a number of cblocks on open/close of
a tty.
Note that this might conflict with the collateral use of TS_WOPEN, but
for the moment I can find no problems associated with this. (TS_WOPEN
will likely go away in the future anyway). This should be looked at
again in the future (the potential problem is that the cblock pool
may either run out or accumulate too many cblocks).
all: does not need to depend on ${MAN5}, remove special case clean:
and cleandir: targets, simple use a CLEANFILES+= to handle this very
simple special case.
Add ${COPY} knob to install commands so that files don't disappear out
of the obj tree after a make install.
Disable doc for now since it had a bunch of symbolic links I am not
ready to go deal with just yet.
Disable man for now it will be fixed shortly, just wanted all the man
page converion stuff to be done togeather since that is a major functional
change and really belongs in a seperate commit.
jkh [Wed, 29 Jun 1994 21:19:38 +0000 (21:19 +0000)]
Oh GACK! src-clean doesn't quite work that easily since cleandist rebuilds the
obj links again and they need to be nuked out one more time after it's finished.
Sigh.. Oh well, this code will probably never be run again anyway.
jkh [Wed, 29 Jun 1994 11:06:30 +0000 (11:06 +0000)]
Clean this puppy up quite a bit. Most notably:
1. Use ${MAKE} everywhere again. Whoops.
2. Replace multiple invocations of gzip ... split ... with one variable.
3. Add src-clean target for making the src tree presentable before
making a src tarball out of it.
rgrimes [Wed, 29 Jun 1994 06:46:02 +0000 (06:46 +0000)]
$rotdelay is now initialized to "-d 0 -n 1" and no longer changed based
upon disk type. In far more cases than not this is the optimal setting
for any disk drive made after 1990.
This now means all installs will have the disks newfs'ed with either:
newfs -b 8192 -f 1024 -d 0 -n 1
or
newfs -n 4096 -f 512 -d 0 -n 1
depending on what the user chooses for the blocking factor.
jkh [Tue, 28 Jun 1994 09:01:55 +0000 (09:01 +0000)]
Clean up the installation docs a bit (geeze Loise were they EVER out of
date!!) and rename them to something more eye-catching so people will read them
again (considering the previous state of affairs, I'm actually rather glad they didn't!).
jkh [Tue, 28 Jun 1994 07:28:28 +0000 (07:28 +0000)]
Add --unlink flag to all tar commands, but each for different reasons.
1. Add to secr and bindists to possibly save the occasional fool who
doesn't RTFM and uses the wrong command to extract this (or even someone
who's legitimately using this to extract on top of a bindist somewhere
*else*).
2. Do the right thing with any symlinks in the src tree. Right now, we're
free of the buggers, but just in case.
jkh [Tue, 28 Jun 1994 07:12:43 +0000 (07:12 +0000)]
Add Chris Demetriou's --unlink changes.
I know that I said earlier that this should be unconditional behaviour,
but I thought about it a little more and concluded that the principle of least
surprise dictates that I make it an option.
jkh [Tue, 28 Jun 1994 05:56:13 +0000 (05:56 +0000)]
Commit Scott's changes to the bad144 stuff - bad144 takes a list of bad
blocks as *multiple* arguments, not one argument (as was incorrectly being
passed).
jkh [Fri, 24 Jun 1994 15:14:16 +0000 (15:14 +0000)]
Just talked with James - there's another, better way to go about this.
Back out my earlier change. Note that this is just for the 1.1.5R floppies;
the 1.1.5A ones still have the work-around method (which works fine and
doesn't hurt anything, it's just kludge!).
jkh [Fri, 24 Jun 1994 14:32:36 +0000 (14:32 +0000)]
This is a kludge. Unless I make a link from sh to - (which would leave a file
called `-' lying around on the users system forever) login shells will fail.
Just special-case the handling of `-' for now until/unless I find a more
palatable solution.
jkh [Wed, 22 Jun 1994 05:52:25 +0000 (05:52 +0000)]
Commit a whole cluster of last minute critical (and one cosmetic) fixes
from David Greenman, Bruce Evans and Julian Elischer.
They are:
[vnode pager - David/Bruce]:
The following patch fixes a problem where some data could be lost in a
delayed-write buffer if the cached buffer was larger than a page. This fix was
provided by Bruce Evans and modified slightly by me.
[st.c - Julian]:
My fix for "bad request, must be between 0 and 0"
RTFS if you're interested).
[gnu/fpemul - David/Bruce]:
These changes fix single stepping of emulated FPU instructions.
Previously, the instruction after an emulated instruction was
executed without causing a SIGTRAP ...
The also fix the initial control word being different for the
GPL emulator (it is still wrong for the old emulator) and remove
an unnecessary panic when emulation is not configured (I hope at
least init, sh and reboot will run without floating point. I
remember only df and mkfs being broken by the lack of FP in 0.0).
[Various fixes described below - Bruce/David]:
sys/i386/boot/boot2.S:
Yet another attempt to propagate the correct fix for 16 vs
32-bit mode bugs. [verified]
sys/i386/i386/db_interface.c:
Protect against reentering Debugger().
sys/kern/kern_time.c:
Don't allow 'time.tv_usec == 0' except at clock interrupts.
sys/pcfs/pcfs_fat.c:
Make it compile without -O.
sys/scsi/sd.c:
Fix as posted to some freebsd mailing list.
(changes the order of the assignment of "sectors" because it earlier
value is needed first -DG)
sys/vm/vm_glue.c:
Fix stale comments and verbose code.
sys/vm/vm_mmap.c
Fix off by 1 errors and verbose code.
[From Nate - cosmetic but non-intrusive and useful enough to go in]
sys/i386/isa/isa.c:
Appended you'll find a patch to the NMI error log routine in isa/isa.c.
The below patch just adds some additional information when an NMI occurs
which can help debug the hardware problem.
jkh [Wed, 22 Jun 1994 04:49:04 +0000 (04:49 +0000)]
Update this to the latest version from Steve Gerakines. This is an easy
drop-in for me and looks substantailly neater than the previous version,
so I'll give the floppy tape users a break (but just this once :).
jkh [Tue, 21 Jun 1994 11:02:38 +0000 (11:02 +0000)]
1. Increase number of inodes for kcopy floppy; we've currently run out.
2. Get kcopy and filesystem images from current directory since we
now build them here; a clean rule is now all that's needed to make
the crunch stuff complete.
jkh [Mon, 20 Jun 1994 06:05:16 +0000 (06:05 +0000)]
Commit a small change to allow FreeBSD to install on a different drive.
The kernel configs already support this, so with a boot floppy or a utility
like booteasy, the user should be able to install and boot off the second drive.
Hurrah.
jkh [Thu, 16 Jun 1994 12:14:08 +0000 (12:14 +0000)]
I'll leave the addition of the line, but there's just absolutely no
way I'm going to allow this to be set to secure. People blow their
password files away all the time, and I am not at all keen to lose the
ability to get them recovered with the simple expedient of a single-user
boot.
ache [Thu, 16 Jun 1994 11:13:44 +0000 (11:13 +0000)]
Choose secure single-user mode by default
Without this entry init.bsdi don't ask root password when it goes
to sigle-user. This entry must present here in any case,
subject of arguing can be only default mode, I mean
"secure" or "insecure" here. Please consider this entry
like template and change "insecure" to "secure" if you
are _shure_, but not back out whole line.
# This entry needed for asking password when init goes to single-user mode
console none unknown off insecure
jkh [Thu, 16 Jun 1994 05:32:10 +0000 (05:32 +0000)]
This is the IBM/National PCMCIA ethernet driver from Keith Moore,
based originally on work by David Greenman and adapted to FreeBSD
(and cleaned up a bit) by myself. It supports the IBM Credit Card
Adapter for the IBM Thinkpad, and I've had no trouble making it work
on my Toshiba T1910 with a National `InfoMover' NE4100 PCMCIA ethernet
card (I'm commiting this message through it right now :-).
This is actually sneaking it in after feature-freeze, but it's just
too useful to pass up! As always, necessity is a mother.
adam [Thu, 16 Jun 1994 03:53:29 +0000 (03:53 +0000)]
Changed delay mechanism to rely more on I/O spinning. If spinning on the
like this is bad news, it will have to be revised.
Shortened some verbose messages for when the kernel is loaded below 640k.
Updated version number.
rich [Wed, 15 Jun 1994 22:41:19 +0000 (22:41 +0000)]
Changes from Paul Kranenburg which bring us into sync with his sources:
handling of errors through the standard err() and warn()
more fixes for Geoff Rehmet's NULL pointer bug.
fixes NULL pointer bugs when linking mono and nested X servers.
supports a `-nostdlib' option.
accept object files without a symbol table
don't attempt dynamic linking when `-A' is given
a few variable names have chaged (desc -> fd), and the formatting has
changed which should make it much easier to track his sources.
I tested 'make world' for /usr/src and X twice with these changes.
jkh [Wed, 15 Jun 1994 21:02:36 +0000 (21:02 +0000)]
Remove the --exclude libc.so.1.1 line; with a static gzip and tar we
no longer need it, and it's just one more thing we would have to remember
to change with every shared lib version bump.
jkh [Wed, 15 Jun 1994 20:14:29 +0000 (20:14 +0000)]
Lots of changes:
1. Properly use ${.CURDIR} now instead of hardcoded relative dirs.
2. Use ${BINOWN} and ${BINGRP} everywhere instead of root/wheel
3. Add target for copying over EXTRACT scripts (and add them here).
4. Start thinking about crunched floppy target (not in yet, next commit).
jkh [Wed, 15 Jun 1994 10:49:34 +0000 (10:49 +0000)]
1. Clean the intermediate files when building the default fixit floppy
(demonstration) target.
2. Add the kcopy and filesystem floppy configuration files from Rod.
jkh [Wed, 15 Jun 1994 10:33:49 +0000 (10:33 +0000)]
This is the new crunch utility for making distribution floppies from
James da Silva. We need to integrate this before 1.1.5 so that we can
actually make the boot floppies.
csgr [Tue, 14 Jun 1994 12:45:41 +0000 (12:45 +0000)]
Bug reported by Jon-Dean Mountjoy <csjm@cs.ru.ac.za>
Paul Kranenburg's description:
ld is in error here, assuming that symbols with N_EXT set always have an
entry in the (global) symbol table: this is not the case for C++ generated
constructor/destructor symbols. I can reproduce your failure by fudging
a "multiply defined" constructor symbol by hand. Checking for `g == NULL'
seems to be a ok as a fence for now.
So:
for now, in do_file_warnings() we check if g == NULL, before trying to generate
any warning messages. This prevents a NULL pointer dereference.