Bruce Evans [Wed, 6 Jan 1999 10:06:36 +0000 (10:06 +0000)]
Removed the weak spam of ${DESTDIR}/usr/include in the bootstrap
target (see the previous log message). This works for bootstrapping
from 2.2.7. It won't work for bootstrapping from 2.1.x, but that
fails due to include problems earlier.
Add new sysctl machdep.uc_devlist. It allows to cleanly retrieve the
list of devices which has been changed in UserConfig, without resorting
to reading /dev/kmem.
The data returned consist of series of struct isa_device and
char dev_name[8].
Don't build biosboot by default - there's no point now that nothing
looks for /usr/mdec/boot{1,2}. The other stuff in here is probably
equally useless/pointless but I'm going to not worry about it for now.
Warner Losh [Wed, 6 Jan 1999 07:54:28 +0000 (07:54 +0000)]
Allow PINGing from any address on multihomed hosts
In the words of the submitter:
"The patch below allows to ping from any address on the multihomed host.
The man page is also updated, the text was cutted from traceroute(8)."
Warner Losh [Wed, 6 Jan 1999 07:06:38 +0000 (07:06 +0000)]
More Elven advice:
Update /etc/rc* if you are going to run a system with aout
libraries, otherwise ldconfig won't know what to do and all dynamic
aout binaries will fail.
Brian Somers [Wed, 6 Jan 1999 00:08:03 +0000 (00:08 +0000)]
Don't warn when ``add xxxx HISADDR'' fails because HISADDR
is 0.0.0.0. This isn't a problem as the sticky route is still
added.
Suggested by: Chris Timmons <skynyrd@opus.cts.cwu.edu>
The Conner CFP2107 is a fixed-media drive, not removable media. This fixes
the quirk that disables tagged queueing for those drives.
Also, silence a warning by disabling xpt_for_all_targets() and
xpt_for_all_periphs(). These two functions are not currently used, but
they should not be removed. They're part of a set of functions that
provide a way to execute a function for every {bus,target,device,periph} in
the system.
If anyone needs to use either function in the future, they can be
un-#ifdefed.
Doug Rabson [Tue, 5 Jan 1999 18:39:02 +0000 (18:39 +0000)]
Fix fubyte to actually return the correct value (it was always returning
zero). This caused device reads to corrupt the first byte of the page
being read into (the cause of frozen keyboards in sysinstall).
PR: 9323
Submitted by: ishisone@sra.co.jp
Make touch handle years 2000-2038 in the obsoleted format, rather
than 1902-1969 as was previously the case with two digit year spec.
Warner Losh [Tue, 5 Jan 1999 07:45:10 +0000 (07:45 +0000)]
On Dec 24 the wcd driver was removed. The new driver is acd. Make a
note of this in the updating file. Put it under the 19981224 date
rather than today's date, since that was when the change was.
Submitted by: Gregory Bond <gnb@itga.com.au>
Also added note about floppy tape driver. Didn't add note about
voxware or pcvt as they have been restored to the kernel.
Warner Losh [Tue, 5 Jan 1999 05:46:37 +0000 (05:46 +0000)]
Add a poitner to Peter's page on the ELF migration day. This is the
day when the elves sail accross the sea to a new home, which happens
to be in your computer. They will sing glorious songs for the Eldar
days. You machine will shun the darkness that is upon the world.
Peter Wemm [Tue, 5 Jan 1999 03:53:06 +0000 (03:53 +0000)]
A partial implementation of the procfs cmdline pseudo-file. This
is enough to satisfy things like StarOffice. This is a hack, but doing
it properly would be a LOT of work, and would require extensive grovelling
around in the user address space to find the argv[].
Obtained from: Mostly from Andrzej Bialecki <abial@nask.pl>.
Do some serious beating on the floppy generation code to try and
simplify it, name things more descriptively and get the kernel back
on boot.flp where it should be.
Bill Paul [Tue, 5 Jan 1999 00:59:08 +0000 (00:59 +0000)]
GRRRR! Apparently, the promiscuous mode chip bug which I thought was
isolated to revision 33 PNIC chips is also present in revision 32 chips.
Cards with rev. 32 chips include the LinkSys LNE100TX and the Matrox
FastNIC 10/100. This accounts for all the cards that I have to test
with.
(I was never able to personally trip the bug on this chip rev, but today
one of the guys in the lab did it with the software they're working on
for their cellular IP project, which uses BPF and promiscuous mode
extensively.)
This commit enables the promiscuous mode software workaround code for
both revison 32 and revision 33 chips. It's possible all of the PNIC
chips suffer from this bug, but these are the only two revs where I
know for a fact it exists.
Peter Wemm [Mon, 4 Jan 1999 18:45:08 +0000 (18:45 +0000)]
Fix a potential sign extension bug on 8-bit chars.
Outputting a backspace isn't supposed to be destructive.. It isn't on
most terminals, nor on the standard bios output (vs. TERM_EMU mode)
Peter Wemm [Mon, 4 Jan 1999 18:37:41 +0000 (18:37 +0000)]
Don't forget a trailing \n when loading a kernel that has been stripped.
(This might make ELF_VERBOSE look funny, but I'm tempted to delete that
anyway)
Bill Fumerola [Mon, 4 Jan 1999 14:07:53 +0000 (14:07 +0000)]
-Clarification of last commit-
The char that the random letters and numbers are being pulled from is
ended with a '\0'. Using sizeof() includes this '\0' in the 'pool' of
possible characters. This patch decrements by one the size so we don't
accidently end the new password prematurly.
Bruce Evans [Mon, 4 Jan 1999 12:36:23 +0000 (12:36 +0000)]
Backed out previous commit. The bug was in rev.1.49 of src/Makefile.inc1,
so it can't be completely fixed here. Putting -I paths to the current
headers in CFLAGS is more likely to break things than fix them, since
there is no reason why current headers should work with old libraries.
Other Makefiles that do this may only work because they are not used early
in the bootstrap. However, the usual problem with errno being defined in
terms of a nonexistent library doesn't seem to occur here, since we don't
reference errno.
Bruce Evans [Mon, 4 Jan 1999 12:05:59 +0000 (12:05 +0000)]
Backed out rev.1.49. It broke bootstrapping from 2.1.5 2.2.7 and
probably other versions by spamming ${DESTDIR}/usr/include in much
the same way as `make includes'.
Details for 2.2.7: the bootstrap target has always done a weak spam
of ${DESTDIR}/usr/include; we depend on it not installing any
significant anachronisms (it probably shouldn't touch the headers
at all; however, we may be depending on it for things like the
renaming of ts_nsec to ts_sec in <sys/time.h>). Rev.1.49 strengthens
the spam to everything in src/include. For 2.2.7, this is not
immediately fatal. However, the `make all' step in src/includes
is not followed by a `make clean' step, so new rpc headers are not
generated after we've bootstrapped rpcgen. This causes a fatal
error much later when the old (generated) rpc headers are used with
the current headers (sys/types.h and/or the non-generated rpc
headers).
Details for 2.1.x: the bug is immediately fatal. It gives definition
of errno that is not supported by 2.1.x's libc. The weak spam in the
restored version avoids this problem by not installing errno.h.
(Bootstrapping from 2.1.5 actually breaks much earlier.)
I think the header problems supposedly fixed by rev.1.49 were caused
by using NOCLEAN and having the build fall over when the weakly
spammed headers are active. Minor differences in the layout will
then cause the .depend files to point to nonexistent headers. It
is a feature for symlinks like errno.h -> sys/errno.h to not exist
early.
The other change in rev.1.49 breaks building obj directories if NOCLEAN
is set. It is only safe for _re_building with NOCLEAN set.
Chuck Robey [Mon, 4 Jan 1999 03:14:48 +0000 (03:14 +0000)]
Fix a build problem when upgrading from 9 month old sources. There's
a circular dependency problem, wherein rpcgen depends on the latest
includes, but those same includes depend on rpcgen to build
include/rpcsvc. This was causing the build or librpcsvc to fail
(like I said, only on upgrades of 9 month old sources).
If there's stylistic problems, tell me, I'll learn and fix them.
I did a buildworld with this, it should be safe.
Matt Jacob [Sun, 3 Jan 1999 22:57:54 +0000 (22:57 +0000)]
Temporary workaround (bandaid) for case where you have READ
CAPACITY fail for a non-removable media device. There's a race
condition where the device entry is removed and then
xpt_release_ccb is called which attempts to give back the ccb
to a device that's now gone. In this bandaid release the ccb
early and then remember to not call xpt_release_ccb later.
Clean up some more residual /usr/mdec references. I left all the
extra rbootd/boot rom cruft pointing at /usr/mdec since it either
doesn't exist or doesn't work anyway, so who cares? :)
Bill Paul [Sun, 3 Jan 1999 02:05:21 +0000 (02:05 +0000)]
Minor bug: in the case where allocating a fresh mbuf for the receive ring
fails, we need to set the descriptor status word so that the 'OWN' bit
is set again so that the chip can reuse it. Previously, this wasn't being
done.
Garrett Wollman [Sun, 3 Jan 1999 00:35:31 +0000 (00:35 +0000)]
Fix grammar in the description of timegm() by totally rewriting it. Remove
a potentally inflammatory comment from BUGS, and add a more useful comment
about the lack of reentrancy in the timezone-setting interface.
Dmitrij Tejblum [Sat, 2 Jan 1999 13:26:29 +0000 (13:26 +0000)]
Ensure that deHighClust in direntry always initialized.
Noticed by: Carl Mascott <cmascott@world.std.com>
Don't write access time of a file more than once per day. (Its precision is
1 day anyway). Don't try to write access and creation time in nonwin95 case.
Bill Fumerola [Sat, 2 Jan 1999 04:37:46 +0000 (04:37 +0000)]
Let's make sure we're at the end of the password string before we apply a \0
and terminate it. This patch ensures passwords will be the correct length of 8,
which is what is implied in the source (but not reflected in the man page).
Bill Fumerola [Fri, 1 Jan 1999 17:37:33 +0000 (17:37 +0000)]
Make periodic(8) and the security mailings reflect the full FQDN, as opposed
to a hostname. This will help those who keep a cluster of machines all with
the same hostname but different domain names.
PR: bin/9091
Submitted By: Heikki Suonsivu <hsu@clinet.fi>
No Response From: -current mailing list