Andrey Chernov [Sat, 2 Apr 1994 20:43:25 +0000 (20:43 +0000)]
Change from Bruce:
isa_dmarangecheck() is off by one error.
> ISARAM_END should be >= ISARAM_END. Only the first page above 16M
was mishandled.
Jordan K. Hubbard [Sat, 2 Apr 1994 18:22:06 +0000 (18:22 +0000)]
Newsgroups: comp.unix.bsd
From: david@wubios.wustl.edu (David J Camp)
Subject: patch for FreeBSD man page
Organization: Division of Biostatistics, WUMS, St. Louis, MO
Date: Sat, 2 Apr 1994 15:15:52 GMT
I found a minor but confusing error in one of the FreBSD man pages.
Here is a patch: -David-
Jordan K. Hubbard [Sat, 2 Apr 1994 17:10:19 +0000 (17:10 +0000)]
Back out previous change.
David Greenman [Sat, 2 Apr 1994 08:39:20 +0000 (08:39 +0000)]
Changed _profiling back to profiling.
David Greenman [Sat, 2 Apr 1994 07:09:28 +0000 (07:09 +0000)]
Output *_imask instead of *mask.
David Greenman [Sat, 2 Apr 1994 07:01:04 +0000 (07:01 +0000)]
New interrupt code from Bruce Evans. In additional to Bruce's attached
list of changes, I've made the following additional changes:
1) i386/include/ipl.h renamed to spl.h as the name conflicts with the
file of the same name in i386/isa/ipl.h.
2) changed all use of *mask (i.e. netmask, biomask, ttymask, etc) to
*_imask (net_imask, etc).
3) changed vestige of splnet use in if_is to splimp.
4) got rid of "impmask" completely (Bruce had gotten rid of netmask),
and are now using net_imask instead.
5) dozens of minor cruft to glue in Bruce's changes.
These require changes I made to config(8) as well, and thus it must
be rebuilt.
-DG
from Bruce Evans:
sio:
o No diff is supplied. Remove the define of setsofttty(). I hope
that is enough.
*.s:
o i386/isa/debug.h no longer exists. The event counters became too
much trouble to maintain. All function call entry and exception
entry counters can be recovered by using profiling kernel (the new
profiling supports all entry points; however, it is too slow to
leave enabled all the time; it also). Only BDBTRAP() from debug.h
is now used. That is moved to exception.s. It might be worth
preserving SHOW_BITS() and calling it from _mcount() (if enabled).
o T_ASTFLT is now only set just before calling trap().
o All exception handlers set SWI_AST_MASK in cpl as soon as possible
after entry and arrange for _doreti to restore it atomically with
exiting. It is not possible to set it atomically with entering
the kernel, so it must be checked against the user mode bits in
the trap frame before committing to using it. There is no place
to store the old value of cpl for syscalls or traps, so there are
some complications restoring it.
Profiling stuff (mostly in *.s):
o Changes to kern/subr_mcount.c, gcc and gprof are not supplied yet.
o All interesting labels `foo' are renamed `_foo' and all
uninteresting labels `_bar' are renamed `bar'. A small change
to gprof allows ignoring labels not starting with underscores.
o MCOUNT_LABEL() is to provide names for counters for times spent
in exception handlers.
o FAKE_MCOUNT() is a version of MCOUNT() suitable for exception
handlers. Its arg is the pc where the exception occurred. The
new mcount() pretends that this was a call from that pc to a
suitable MCOUNT_LABEL().
o MEXITCOUNT is to turn off any timer started by MCOUNT().
/usr/src/sys/i386/i386/exception.s:
o The non-BDB BPTTRAP() macros were doing a sti even when interrupts
were disabled when the trap occurred. The sti (fixed) sti is
actually a no-op unless you have my changes to machdep.c that make
the debugger trap gates interrupt gates, but fixing that would
make the ifdefs messier. ddb seems to be unharmed by both
interrupts always disabled and always enabled (I had the branch in
the fix back to front for some time :-().
o There is no known pushal bug.
o tf_err can be left as garbage for syscalls.
/usr/src/sys/i386/i386/locore.s:
o Fix and update BDE_DEBUGGER support.
o ENTRY(btext) before initialization was dangerous.
o Warm boot shot was longer than intended.
/usr/src/sys/i386/i386/machdep.c:
o DON'T APPLY ALL OF THIS DIFF. It's what I'm using, but may require
other changes.
Use the following:
o Remove aston() and setsoftclock().
Maybe use the following:
o No netisr.h.
o Spelling fix.
o Delay to read the Rebooting message.
o Fix for vm system unmapping a reduced area of memory
after bounds_check_with_label() reduces the size of
a physical i/o for a partition boundary. A similar
fix is required in kern_physio.c.
o Correct use of __CONCAT. It never worked here for non-
ANSI cpp's. Is it time to drop support for non-ANSI?
o gdt_segs init. 0xffffffffUL is bogus because ssd_limit
is not 32 bits. The replacement may have the same
value :-), but is more natural.
o physmem was one page too low. Confusing variable names.
Don't use the following:
o Better numbers of buffers. Each 8K page requires up to
16 buffer headers. On my system, this results in 5576
buffers containing [up to]
2854912 bytes of memory.
The usual allocation of about 384 buffers only holds
192K of disk if you use it on an fs with a block size
of 512.
o gdt changes for bdb.
o *TGT -> *IDT changes for bdb.
o #ifdefed changes for bdb.
/usr/src/sys/i386/i386/microtime.s:
o Use the correct asm macros. I think asm.h was copied from Mach
just for microtime and isn't used now. It certainly doesn't
belong in <sys>. Various macros are also duplicated in
sys/i386/boot.h and libc/i386/*.h.
o Don't switch to and from the IRR; it is guaranteed to be selected
(default after ICU init and explicitly selected in isa.c too, and
never changed until the old microtime clobbered it).
/usr/src/sys/i386/i386/support.s:
o Non-essential changes (none related to spls or profiling).
o Removed slow loads of %gs again. The LDT support may require
not relying on %gs, but loading it is not the way to fix it!
Some places (copyin ...) forgot to load it. Loading it clobbers
the user %gs. trap() still loads it after certain types of
faults so that fuword() etc can rely on it without loading it
explicitly. Exception handlers don't restore it. If we want
to preserve the user %gs, then the fastest method is to not
touch it except for context switches. Comparing with
VM_MAXUSER_ADDRESS and branching takes only 2 or 4 cycles on
a 486, while loading %gs takes 9 cycles and using it takes
another.
o Fixed a signed branch to unsigned.
/usr/src/sys/i386/i386/swtch.s:
o Move spl0() outside of idle loop.
o Remove cli/sti from idle loop. sw1 does a cli, and in the
unlikely event of an interrupt occurring and whichqs becoming
zero, sw1 will just jump back to _idle.
o There's no spl0() function in asm any more, so use splz().
o swtch() doesn't need to be superaligned, at least with the
new mcounting.
o Fixed a signed branch to unsigned.
o Removed astoff().
/usr/src/sys/i386/i386/trap.c:
o The decentralized extern decls were inconsistent, of course.
o Fixed typo MATH_EMULTATE in comments. */
o Removed unused variables.
o Old netmask is now impmask; print it instead. Perhaps we
should print some of the new masks.
o BTW, trap() should not print anything for normal debugger
traps.
/usr/src/sys/i386/include/asmacros.h:
o DON'T APPLY ALL OF THIS DIFF. Just use some of the null macros
as necessary.
/usr/src/sys/i386/include/cpu.h:
o CLKF_BASEPRI() changes since cpl == SWI_AST_MASK is now normal
while the kernel is running.
o Don't use var++ to set boolean variables. It fails after a mere
4G times :-) and is slower than storing a constant on [3-4]86s.
/usr/src/sys/i386/include/cpufunc.h:
o DON'T APPLY ALL OF THIS DIFF. You need mainly the include of
<machine/ipl.h>. Unfortunately, <machine/ipl.h> is needed by
almost everything for the inlines.
/usr/src/sys/i386/include/ipl.h:
o New file. Defines spl inlines and SWI macros and declares most
variables related to hard and soft interrupt masks.
/usr/src/sys/i386/isa/icu.h:
o Moved definitions to <machine/ipl.h>
/usr/src/sys/i386/isa/icu.s:
o Software interrupts (SWIs) and delayed hardware interrupts (HWIs)
are now handled uniformally, and dispatching them from splx() is
more like dispatching them from _doreti. The dispatcher is
essentially *(handler[ffs(ipending & ~cpl)]().
o More care (not quite enough) is taken to avoid unbounded nesting
of interrupts.
o The interface to softclock() is changed so that a trap frame is
not required.
o Fast interrupt handlers are now handled more uniformally.
Configuration is still too early (new handlers would require
bits in <machine/ipl.h> and functions to vector.s).
o splnnn() and splx() are no longer here; they are inline functions
(could be macros for other compilers). splz() is the nontrivial
part of the old splx().
/usr/src/sys/i386/isa/ipl.h
o New file. Supposed to have only bus-dependent stuff. Perhaps
the h/w masks should be declared here.
/usr/src/sys/i386/isa/isa.c:
o DON'T APPLY ALL OF THIS DIFF. You need only things involving
*mask and *MASK and comments about them. netmask is now a pure
software mask. It works like the softclock mask.
/usr/src/sys/i386/isa/vector.s:
o Reorganize AUTO_EOI* macros.
o Option FAST_INTR_HANDLER_USERS_ES for people who don't trust
fastintr handlers.
o fastintr handlers need to metamorphose into ordinary interrupt
handlers if their SWI bit has become set. Previously, sio had
unintended latency for handling output completions and input
of SLIP framing characters because this was not done.
/usr/src/sys/net/netisr.h:
o The machine-dependent stuff is now imported from <machine/ipl.h>.
/usr/src/sys/sys/systm.h
o DON'T APPLY ALL OF THIS DIFF. You need mainly the different
splx() prototype. The spl*() prototypes are duplicated as
inlines in <machine/ipl.h> but they need to be duplicated here
in case there are no inlines. I sent systm.h and cpufunc.h
to Garrett. We agree that spl0 should be replaced by splnone
and not the other way around like I've done.
/usr/src/sys/kern/kern_clock.c
o splsoftclock() now lowers cpl so the direct call to softclock()
works as intended.
o softclock() interface changed to avoid passing the whole frame
(some machines may need another change for profile_tick()).
o profiling renamed _profiling to avoid ANSI namespace pollution.
(I had to improve the mcount() interface and may as well fix it.)
The GUPROF variant doesn't actually reference profiling here,
but the 'U' in GUPROF should mean to select the microtimer
mcount() and not change the interface.
Andrew Moore [Sat, 2 Apr 1994 06:19:12 +0000 (06:19 +0000)]
Renamed executable to avoid name-clash with directory
Andrey Chernov [Fri, 1 Apr 1994 18:33:12 +0000 (18:33 +0000)]
Replace CAPS led with ALTGR led for ALTGR mode (soft) keyboards,
currently affects only russian keyboard.
Andrey Chernov [Fri, 1 Apr 1994 16:47:01 +0000 (16:47 +0000)]
Change got_status/modem_status assignment per Bruce suggestion,
because inb clears modem status port.
Andrey Chernov [Fri, 1 Apr 1994 15:04:49 +0000 (15:04 +0000)]
Localize previous changes (removing dma_reset)
only for SB card, it seems that it breaks GUS at least.
Jordan K. Hubbard [Fri, 1 Apr 1994 13:11:01 +0000 (13:11 +0000)]
Fix from Gene for bogus destinations.
Jordan K. Hubbard [Fri, 1 Apr 1994 10:13:22 +0000 (10:13 +0000)]
Fix bug with ppp.h I introduced earlier. This should do it.
Andrey Chernov [Thu, 31 Mar 1994 20:54:25 +0000 (20:54 +0000)]
Tracker "swapping" problem finally solved,
all (expect one) dma_reset calls commented out.
David Greenman [Thu, 31 Mar 1994 14:11:02 +0000 (14:11 +0000)]
Minor performance improvements from J.T.Conklin and Charles Hannum.
David Greenman [Thu, 31 Mar 1994 11:06:29 +0000 (11:06 +0000)]
corrected comment.
Andrey Chernov [Thu, 31 Mar 1994 00:51:01 +0000 (00:51 +0000)]
Remove compilation warning in ttyinfo
Andreas Schulz [Wed, 30 Mar 1994 22:06:08 +0000 (22:06 +0000)]
Give an error if the "ident" is missing, instead of a core dump.
Error found by Poul-Henning Kamp/P-HK:
I just noticed in passing that if the 'ident' line was accidentially
commented out here on REF the config(1) program would coredump. I have no
time to hunt this one, I hope somebody else can/will chase it.
Garrett Wollman [Wed, 30 Mar 1994 20:37:52 +0000 (20:37 +0000)]
Move this directory, too.
Garrett Wollman [Wed, 30 Mar 1994 20:36:37 +0000 (20:36 +0000)]
Give these files the right owerships (first attempt didn't do that). Change
history isn't particularly important for these files.
Andrey Chernov [Wed, 30 Mar 1994 20:33:28 +0000 (20:33 +0000)]
Fix error with reducing speed on SB reset.
Andreas Schulz [Wed, 30 Mar 1994 19:22:00 +0000 (19:22 +0000)]
Added LDFLAGS to the build of rrestore, else we end with a shared lib
executable of it.
Andreas Schulz [Wed, 30 Mar 1994 18:37:41 +0000 (18:37 +0000)]
Added LDFLAGS to the initdeck build, so that a compile with NOSHARED?=YES
works.
Andreas Schulz [Wed, 30 Mar 1994 18:35:02 +0000 (18:35 +0000)]
Added LDFLAGS to the makedefs build, so that a compile with NOSHARED?=YES
works.
Jordan K. Hubbard [Wed, 30 Mar 1994 09:56:23 +0000 (09:56 +0000)]
Update this too.
Jordan K. Hubbard [Wed, 30 Mar 1994 09:38:20 +0000 (09:38 +0000)]
Add some missing files, rename NPPP to _NPPP to avoid name clash with the
kernel symbol denoting number of ppp devices.
Jordan K. Hubbard [Wed, 30 Mar 1994 09:31:44 +0000 (09:31 +0000)]
Update to pppd 2.0.4
David Greenman [Wed, 30 Mar 1994 02:47:13 +0000 (02:47 +0000)]
Eliminated the "physstrat" wart and merged it into kern_physio.c. This
patch also fixes a bug which causes a kernel VM leak.
David Greenman [Wed, 30 Mar 1994 02:31:26 +0000 (02:31 +0000)]
Eliminated the "physstrat" wart and merged it into kern_physio.c. This
patch also fixes a bug which causes a kernel VM leak.
David Greenman [Wed, 30 Mar 1994 02:22:00 +0000 (02:22 +0000)]
New routine "pmap_kenter", designed to take advantage of the special
case of the kernel pmap.
Jordan K. Hubbard [Tue, 29 Mar 1994 08:52:55 +0000 (08:52 +0000)]
Add comments about new gdb from Gary Jennejohn.
Andrew Moore [Tue, 29 Mar 1994 01:15:13 +0000 (01:15 +0000)]
Use tmac.s since we don't have tmac.sU.
Andrew Moore [Mon, 28 Mar 1994 20:07:39 +0000 (20:07 +0000)]
Change mmap(): MAP_PRIVATE to MAP_FILE|MAP_PRIVATE
Andrew Moore [Mon, 28 Mar 1994 11:18:26 +0000 (11:18 +0000)]
Upgrade to 1.11b.
Hacks:
ex_tag.c: mmap(MAP_FILE|MAP_PRIVATE) so tag stacks should now work.
Disabled support for NEXINIT environ var and .nexrc initialization files
(since EXINIT and .exrc are better known).
Changed the vi.1 and vi.ref acorrdingly.
Changed <sys/queue.h> --> <queue.h> in all files.
rm'd #include "compat.h" from all files.
Added #include "compat.h" to vi.h ONLY.
vi/compat.h contains only typedefs for u_int8_t et al.
Added vi/vi.1 so install works.
Changed Makefile accordingly.
Andrew Moore [Mon, 28 Mar 1994 10:58:00 +0000 (10:58 +0000)]
Upgrade to 1.11b.
Hacks:
ex_tag.c: mmap(MAP_FILE|MAP_PRIVATE) so tag stacks should now work.
Disabled support for NEXINIT environ var and .nexrc initialization files
(since EXINIT and .exrc are better known).
Changed the vi.1 and vi.ref acorrdingly.
Changed <sys/queue.h> --> <queue.h> in all files.
rm'd #include "compat.h" from all files.
Added #include "compat.h" to vi.h ONLY.
vi/compat.h contains only typedefs for u_int8_t et al.
Added vi/vi.1 so install works.
Changed Makefile accordingly.
Andrey Chernov [Sat, 26 Mar 1994 14:08:23 +0000 (14:08 +0000)]
Restore incorrectly removed splclock() on reading p_stime/p_utime
Andrey Chernov [Sat, 26 Mar 1994 13:40:18 +0000 (13:40 +0000)]
1) Better fix for false carrier detect on bidir port
2) ttyclose moved after comhardclose, because clears t_state
3) slpx(s) moved after l_open to prevent undetected carrier down
David Greenman [Sat, 26 Mar 1994 12:24:27 +0000 (12:24 +0000)]
minor formatting and comment changes; removed some old #if 0'd code.
David Greenman [Fri, 25 Mar 1994 22:26:22 +0000 (22:26 +0000)]
Patch from mrl@teleport.com (Mostyn R. Lewis), with the comment improved
by me:
Below is a fix to isofs_vnops.c in /usr/src/sys/isofs to make find traverse
directories in the usual way. The va_nlink field needs either to be correct,
i.e the number of directories and the path or else 1 to fool fts.
(Note that this is correct for non-RR FS, but needs to be re-done in the
presence of RR - but there isn't any higher level support for this yet.
-DG)
David Greenman [Fri, 25 Mar 1994 22:08:59 +0000 (22:08 +0000)]
ifdef KERNEL the pmap_kextract inline function; ps is unhappy otherwise.
Pointed out by Frank Terhaar-Yonkers <fty@vislab.epa.gov>.
Jordan K. Hubbard [Fri, 25 Mar 1994 15:27:46 +0000 (15:27 +0000)]
Leave comments in about 16MB ISA DMA in as public service, but describe
how bounce-buffers used in such cases. Remove comments about DTC3290 only
working with 16Mb of memory.
Andrey Chernov [Fri, 25 Mar 1994 15:14:35 +0000 (15:14 +0000)]
Fix bug with standouted last spaces
Andrey Chernov [Fri, 25 Mar 1994 15:10:50 +0000 (15:10 +0000)]
Fix false carrier detection on incoming bidir port.
Paul Richards [Fri, 25 Mar 1994 04:58:58 +0000 (04:58 +0000)]
Add /usr/X11R6/lib to the ldconfig path if it exists.
David Greenman [Thu, 24 Mar 1994 23:12:48 +0000 (23:12 +0000)]
From John Dyson: performance improvements to the new bounce buffer
code.
Andrey Chernov [Thu, 24 Mar 1994 22:23:51 +0000 (22:23 +0000)]
1) After discussion with Hannu, returning speed changed back.
Real problem fixed by my previous fix for SB 2.x
2) get_time function slightly modified to minimize possible
overflowing.
Garrett Wollman [Thu, 24 Mar 1994 21:57:50 +0000 (21:57 +0000)]
Document existewnce of bounce buffers.
David Greenman [Thu, 24 Mar 1994 21:57:38 +0000 (21:57 +0000)]
Minor corrections to the supported ethernet section.
David Greenman [Thu, 24 Mar 1994 02:23:00 +0000 (02:23 +0000)]
Enabled bounce buffers by default. This should be done via a probe for
the specific controller type, so this change is temporary.
Andrey Chernov [Wed, 23 Mar 1994 22:52:30 +0000 (22:52 +0000)]
For SB 2.x speed was reduced to 22050;
I already fix this bug in previous driver version,
but new driver update breaks my changes!
Andrey Chernov [Wed, 23 Mar 1994 21:58:59 +0000 (21:58 +0000)]
SPEED-related ioctls returns wrong speed cause Tracker's damage.
Andrey Chernov [Wed, 23 Mar 1994 19:27:52 +0000 (19:27 +0000)]
1) GET_TIME function completely broken
it returns time in microseconds instead of HZ
(feel difference!)
2) change GET_TIME type to unsigned long in all places to prevent overflow
Jordan K. Hubbard [Wed, 23 Mar 1994 18:50:57 +0000 (18:50 +0000)]
Something bogus happened here; not sure what or how.
Fixed.
Andrey Chernov [Wed, 23 Mar 1994 17:35:48 +0000 (17:35 +0000)]
tty/nullmodem "not open" printf ifdefed DIAGNOSTIC
Andrey Chernov [Wed, 23 Mar 1994 17:28:35 +0000 (17:28 +0000)]
1) Change dtrwait 300 to 3 * hz (to be more kosher)
2) Protect sioclose by spltty()
David Greenman [Wed, 23 Mar 1994 09:16:16 +0000 (09:16 +0000)]
Bounce buffers. From John Dyson with help from me.
Andrew Moore [Wed, 23 Mar 1994 04:36:59 +0000 (04:36 +0000)]
use umask 077 for buffer file
Andrey Chernov [Wed, 23 Mar 1994 01:58:30 +0000 (01:58 +0000)]
Set of changes from Bruce:
1) Macro CAN_DO_IO to replace all TS_CARR_ON || CLOCAL tests
2) Fix bug with clearing TS_ISOPEN in ttioctl.
3) New TS_ZOMBIE flag introduced: carrier dropped when !CLOCAL.
4) Add two debug printf's when tty/nullmodem called without open.
Andrey Chernov [Wed, 23 Mar 1994 00:05:25 +0000 (00:05 +0000)]
Add missing EXCLUDE_SBPRO
David Greenman [Tue, 22 Mar 1994 21:58:38 +0000 (21:58 +0000)]
Removed a disgusting wart that attempts to verify that the data in the
kernel namelist database is correct for the booted kernel.
Rationale for removal:
The most common cause of a DB inconsistancy is when a user boots a non-
default kernel. The DB version check catches this and then resorts to the
Removed a disgusting wart that attempts to verify that the data in the
kernel namelist database is correct for the booted kernel.
Rationale for removal:
The most common cause of a DB inconsistancy is when a user boots a non-
default kernel. The DB version check catches this and then resorts to the
"hard method" of reading the namelist directly (from the default kernel!)
and gets it wrong anyway, just a little slower.
This patch fixes the problem that was created when vers.o is no longer
linked first. It also gets rid of a really ugly hack that I submitted
back in the patchkit days.
David Greenman [Tue, 22 Mar 1994 21:56:48 +0000 (21:56 +0000)]
Removed a disgusting wart that attempts to verify that the data in the
kernel namelist database is correct for the booted kernel.
Rationale for removal:
The most common cause of a DB inconsistancy is when a user boots a non-
default kernel. The DB version check catches this and then resorts to the
"hard method" of reading the namelist directly (from the default kernel!)
and gets it wrong anyway, just a little slower.
This patch fixes the problem that was created when vers.o is no longer
linked first. It also gets rid of a really ugly hack that I submitted
back in the patchkit days.
Andrey Chernov [Tue, 22 Mar 1994 19:41:40 +0000 (19:41 +0000)]
First attempt to bring uucpd to work
1) compilation flags in Makefile fixed
2) Security leak fixed
3) cr/nl problem fixed
4) uucico path fixed
5) several minor bugs in uucpd fixed
6) added recognition of ^U
David Greenman [Tue, 22 Mar 1994 19:03:01 +0000 (19:03 +0000)]
Patch from piero@strider.st.dsi.unimi.it to quiet compiler warning.
Jordan K. Hubbard [Tue, 22 Mar 1994 16:43:59 +0000 (16:43 +0000)]
More detailed notes on just what's wrong with the Buslogic controllers
(and which ones).
Jordan K. Hubbard [Tue, 22 Mar 1994 11:50:34 +0000 (11:50 +0000)]
From: Andras Olah <olah@cs.utwente.nl>
Date: Tue, 22 Mar 94 10:14:44 +0100
Sender: freebsd-bugs-owner@freefall.cdrom.com
Precedence: bulk
Hi,
I'm not (yet ;-) very familiar with BSD but I found something in the
getty code what seems like a minor bug. I might be wrong. As I
understand it affects the processing of the options from
/etc/gettytab, so one might have problems if tries to set op or ap
in gettytab.
David Greenman [Tue, 22 Mar 1994 06:15:10 +0000 (06:15 +0000)]
From John Dyson: fixed bug in call to VOP_BMAP and handle NFS correctly
David Greenman [Tue, 22 Mar 1994 06:07:12 +0000 (06:07 +0000)]
From John Dyson: Improved swap space allocation method to handle small
allocations better, and reduce fragmentation.
David Greenman [Tue, 22 Mar 1994 02:02:25 +0000 (02:02 +0000)]
From mrl@teleport.com (Mostyn R. Lewis)
>PPP and SLIP users - deleting a route can be deadly!
>
>Yes this can cause a kernel page fault or random hang.
...
>you stand a good chance of a kernel page fault or a random hang because
>the routing table for default still has a pointer to the deleted kernel
>ifaddr structure. Depending on the re-use state of that memory you live
>or die. Death is caused by a function pointer dereference.
>
>A fix for this, in route.c (/usr/src/sys/net) is to scan the linked list
>of ifaddr entries for the interface to see if the pointer is still valid.
>The diffs for the fix are appended. There are two versions; mute and
>verbose. The verbose fix prints information the the system log to verify
>states.
Andrey Chernov [Tue, 22 Mar 1994 01:18:34 +0000 (01:18 +0000)]
for SLIPDISC:
nullmodem changed to ttymodem, in other case tsleep never awake
Andrey Chernov [Tue, 22 Mar 1994 01:16:06 +0000 (01:16 +0000)]
Sorry, backed out ttywflush removing, real place to fix is tty_conf.c
Andreas Schulz [Mon, 21 Mar 1994 22:44:42 +0000 (22:44 +0000)]
Catch a null pointer dereference. This should solve the problem reported
with the configure for xdvik and the other TeX configure scripts.
Andrey Chernov [Mon, 21 Mar 1994 22:19:56 +0000 (22:19 +0000)]
Add siostop to sioclose, because ttyflush called from
wrong places removed now.
Andrey Chernov [Mon, 21 Mar 1994 22:10:37 +0000 (22:10 +0000)]
Remove ttywflush from pppclose,
can cause dead hang in tsleep and don't really belongs here
Andrey Chernov [Mon, 21 Mar 1994 21:50:30 +0000 (21:50 +0000)]
Remove ttyflush from ttyclose,
it really belongs only to ttylclose, other than tty
disciplines must not touch t_out & t_raw or
dead tsleep hang can occurse.
Andrey Chernov [Mon, 21 Mar 1994 21:29:50 +0000 (21:29 +0000)]
Remove ttywflush from slclose, it cause
dead hang in tsleep when carier dropped.
Moreover, it is incorrect to deal anything with t_out & t_raw here
(like ttywflush does), we already change line discipline.
Andrey Chernov [Mon, 21 Mar 1994 21:12:55 +0000 (21:12 +0000)]
My previous changes was a bit incorrect in nullmodem() case, back them out.
Real place to fix is if_sl.c
Andreas Schulz [Mon, 21 Mar 1994 20:59:55 +0000 (20:59 +0000)]
Changed the raw partition number from 3 to 0. This change lets us use
/dev/mcd0a instead of /dev/mcd0d. This is more conforming to the /dev/cd0a
for the SCSI cdrom drives. It breaks the convention d the whole drive.
But the question is, do we really need partitions on cdrom drives ?
Andreas Schulz [Mon, 21 Mar 1994 20:48:52 +0000 (20:48 +0000)]
Makefile.i386:
put vers.o at the end of the loader line. We are simply jumping in the
moment into the first location of the text segment in 386bsd. So the
linking order is very important :-). With the addition of the const
types in newvers.sh we jumped into them. I have experimented with an
entry point specification, but was unsuccessfull. Someone else should
look at this.
devices.i386:
files.i386:
Added entries for a Sony cdrom driver.
Andrey Chernov [Mon, 21 Mar 1994 18:28:28 +0000 (18:28 +0000)]
wakeup t_out & t_raw for all carrier changes
to avoid dead tsleep
Andrey Chernov [Mon, 21 Mar 1994 15:18:27 +0000 (15:18 +0000)]
Increase default DTR wait time up to 3 secs.
Also hardw. specs says 2.5 secs is enough, many modems needs
at least 3 secs.
Andrey Chernov [Mon, 21 Mar 1994 15:13:20 +0000 (15:13 +0000)]
Remove bug with root filesystem on second floppy, fixed now
Andrey Chernov [Mon, 21 Mar 1994 15:02:47 +0000 (15:02 +0000)]
Now printf("changing root... indicates raw partition for floppy
f.e. fd1d
Andrey Chernov [Mon, 21 Mar 1994 14:53:08 +0000 (14:53 +0000)]
Fix printf for root system mounted on second floppy
Andrey Chernov [Mon, 21 Mar 1994 14:37:01 +0000 (14:37 +0000)]
Fix for root system mounted on second floppy
David Greenman [Mon, 21 Mar 1994 09:46:12 +0000 (09:46 +0000)]
Modified swap space allocation policy to be more conservative when
dealing with small objects. This is a first-cut change, and will likely
change again. Lowered size of 'window' of pages to watch ("MINSCAN")
in pageout daemon to 256, and fixed a minor bug that resulted in a benign
console message.
David Greenman [Mon, 21 Mar 1994 09:35:30 +0000 (09:35 +0000)]
Changed dynamic stack grow code to grow by "SGROWSIZ" amount. Initially
allocate SGROWSIZ amount of stack. Also set vm_ssize to the initial
stack VM size. Increased DFLSSIZ stack rlimit default to 8MB.
David Greenman [Mon, 21 Mar 1994 06:37:04 +0000 (06:37 +0000)]
previous optimization from John wasn't quite ready for primetime.
Garrett Wollman [Sun, 20 Mar 1994 21:30:38 +0000 (21:30 +0000)]
Oops, forgot to check in last bit of sleep() removal.
Andreas Schulz [Sun, 20 Mar 1994 21:22:04 +0000 (21:22 +0000)]
Added a forgotten htonl() in the last patch. It makes no difference
in the moment, but it looks better.
David Greenman [Sun, 20 Mar 1994 20:06:04 +0000 (20:06 +0000)]
Two fixes from John Dyson to fix hangs and panics when using ctrl-T:
1) tty.c: gather all the info about the processes before calling ttyprintf
(which may block).
2) syscons.c: handle asynchronous output properly (data structures may
be corrupted otherwise).
Guido van Rooij [Sun, 20 Mar 1994 18:51:49 +0000 (18:51 +0000)]
Moved ttyfree() to ifdef broken. See my reply on the sio change.
Andreas Schulz [Sun, 20 Mar 1994 18:17:11 +0000 (18:17 +0000)]
Adding the new filesystem types ( ISOFS, PROCFS, DEVFS ) into the
manual page. Changing the MOUNT_PC to MOUNT_MSDOS as the include file
uses it.
Andreas Schulz [Sun, 20 Mar 1994 18:04:18 +0000 (18:04 +0000)]
Next try to get a bug description about find and awk into the KNOWNBUGS.
Jordan K. Hubbard [Sun, 20 Mar 1994 12:35:42 +0000 (12:35 +0000)]
Remove redundant copies of err() and warn().
Rod Grimes [Sun, 20 Mar 1994 01:23:42 +0000 (01:23 +0000)]
Pull in changes just made to the BETA 1.1 branch for cd install floppy
support.
Garrett Wollman [Sun, 20 Mar 1994 00:33:06 +0000 (00:33 +0000)]
#ifdef'ed out definition of sleep() now that nothing uses it. This will
ensure that uses don't creep back in when nobody's looking.
Garrett Wollman [Sun, 20 Mar 1994 00:30:59 +0000 (00:30 +0000)]
Get rid of a nagging call to sleep() which crept back in.
Garrett Wollman [Sun, 20 Mar 1994 00:30:04 +0000 (00:30 +0000)]
Delete some references to sleep() that somehow crept in.
Jordan K. Hubbard [Sun, 20 Mar 1994 00:22:08 +0000 (00:22 +0000)]
Added comments on Buslogic controllers. Cleaned up some syntax.
Rod Grimes [Sun, 20 Mar 1994 00:12:38 +0000 (00:12 +0000)]
These are the installation and profile files from the 1.0 cdins*.flp
floppy images. Put in as they were in 1.0, about to change them to
work with 1.1.
Garrett Wollman [Sun, 20 Mar 1994 00:03:33 +0000 (00:03 +0000)]
Declare cpu_model, machine.