bde [Thu, 20 Nov 1997 18:24:52 +0000 (18:24 +0000)]
Fixed wrong limits for the kernel text in db_numargs(). The
interval [VM_MIN_KERNEL_ADDRESS, etext] was used instead of
[btext, etext). Added a comment about this being completely
wrong for LKMs. This only affects interpreting the instructions
after the return to attempt decide the number of args. The
attempt usually fails anyway.
bde [Thu, 20 Nov 1997 18:09:36 +0000 (18:09 +0000)]
Fixed write enabling of the kernel text section. The overlap
checking was mostly wrong at the boundaries. For the lower limit,
VM_MIN_KERNEL_ADDRESS was used instead of btext and there was an
off-by-(`size' - 1) error. For the upper limit, &etext was used
instead of etext and there was an off-by-1 error. The bugs were
harmless because `size' is not too large and some memory is mapped
just beyond the ends. We still depend on the former to avoid
having to handle the case where the memory range covers the whole
text section, and on the latter to prevent problems when we map
just beyond an end to allow writing an address range that overlaps
the end.
bde [Thu, 20 Nov 1997 17:07:21 +0000 (17:07 +0000)]
Don't allow setting the dump device to any partition except the
one traditionally reserved for swap devices. The restrictions
should now be the same as the ones for dumpsys(). The restriction
on the partition should be removed someday, and dumpsys() shouldn't
repeat all the checks.
bde [Thu, 20 Nov 1997 16:08:56 +0000 (16:08 +0000)]
Fixed marking of access time for special files and fifos (don't do
it if the file system is mounted noatime). Not fixed: the access
time is marked at the start of a read() and not marked on successful
completion. I think this should be handled at the vfs level.
Print a better panic message for missing vops. Don't use printf()
before panic(), since the printf()ed part isn't shown by gdb.
This actually loses a little with the current gdb, since gdb just
prints the fmt arg to panic, so %'s aren't expanded. gdb should
fetch the full message from the message buffer if possible.
Fixed default vop function for vop_getpages_desc. It needs to
just return EOPNOTSUPP so that the vnode pager can get the pages
in using a general method. Panicing broke exec'ing of files on
ext2fs file systems. ffs works because it doesn't use the default.
asami [Thu, 20 Nov 1997 05:33:56 +0000 (05:33 +0000)]
Define MASTER_SITE_SUBDIR and PATCH_SITE_SUBDIR to default to "." so
we won't have double-slashes.
Add support for new port variable MANUAL_PACKAGE_BUILD. If this is
defined as well as the user variable PACKAGE_BUILDING, the port will
be ignored. This is used to mark ports that can be built normally
except on a machine that has a lot of conflicting ports (i.e., our
package building machine).
asami [Thu, 20 Nov 1997 05:31:44 +0000 (05:31 +0000)]
Add new target "checksubdirs". It will warn about any subdirectories that
are not in the SUBDIR list. It also knows about the "standard" directories
that are to be ignored ("CVS", "distfiles", etc.).
bde [Tue, 18 Nov 1997 16:29:53 +0000 (16:29 +0000)]
Removed unused #includes.
Fixed nonblocking mode. It was per-device instead of per-file. This
also fixes clobbering of bd_rtout by overloading it to hold a wrong
version of the blocking flag. I hope nothing depends on the bugs.
bde [Tue, 18 Nov 1997 16:12:51 +0000 (16:12 +0000)]
Removed unused #includes. Ifdefed a conditionally used #include.
Fixed nonblocking mode. It was per-device instead of per-file.
Don't depend on gcc's misfeature of rewriting char args in old-style
function definitions to match wrong prototypes. Break K&R1 support
to fix this quickly.
bde [Tue, 18 Nov 1997 14:08:51 +0000 (14:08 +0000)]
Don't test for conflicting combinations of PPP_FILTER/BPFILTER here.
Testing in if_ppp.c is good enough.
Added comments about bogus #includes and #defines.
Removed unused #includes.
Don't depend on gcc's misfeature of rewriting short args in old-style
function definitions to match wrong prototypes. I just changed the
function definition to match the prototype, since this is easy to
verify automatically (it causes no changes in the object code), but
it breaks K&R1 support and doesn't fix the pessimal type.
bde [Tue, 18 Nov 1997 13:52:41 +0000 (13:52 +0000)]
Cleaned up PPP_FILTER/NBPFILTER ifdefs.
Use gettime() instead of microtime() to set if_lastchange for i/o's.
microtime() is probably too expensive. However, setting if_lastchange
for i/o's may be wrong.
bde [Tue, 18 Nov 1997 13:37:56 +0000 (13:37 +0000)]
Use gettime() instead of assignment from `time'. (`time' is too
volatile to use outside of splclock(). microtime() is probably too
expensive to use for every i/o. However, setting ifi_lastchange for
every i/o is just wrong according to the comment about ifi_lastchange
in <net/if.h>. It is set then for atm, fddi and the latest version
of ppp.)
bde [Tue, 18 Nov 1997 12:59:09 +0000 (12:59 +0000)]
Get tty ioctl numbers by #including <sys/ttycom.h> instead of
<sys/tty.h>. Don't #include <sys/fcntl.h> (the select -> poll
changes removed all dependencies on it).
bde [Tue, 18 Nov 1997 12:52:10 +0000 (12:52 +0000)]
Removed an unused #include. Added an unsed #include of <sys/ucred.h>
to prepare for not including it in <sys/param.h>. Moved conditionally
used #includes inside an ifdef.
bde [Tue, 18 Nov 1997 12:21:32 +0000 (12:21 +0000)]
Get select stuff by including <sys/select.h> instead of <sys/proc.h>,
and don't include <sys/fcntl.h>. (The select -> poll changes replaced
fcntl macros by poll macros.)
Use <machine/*.h> instead of <i386/include/*.h>.
Fail the probe instead of crashing in the unlikely event that malloc()
fails.
bde [Tue, 18 Nov 1997 11:45:26 +0000 (11:45 +0000)]
Removed unused #includes.
Fixed #include order. <i386/isa/isa_device.h> will soon be a prerequisite
for <i386/isa/pnp.h>. Including both in alphebetical order gets this
right naturally.
bde [Tue, 18 Nov 1997 11:02:19 +0000 (11:02 +0000)]
Removed unused #include of <sys/malloc.h>. This file now uses only
zalloc(). Many more cases like this are probably obscured by not
including <vm/zone.h> explicitly (it is spammed into <sys/malloc.h>).
bde [Tue, 18 Nov 1997 10:56:41 +0000 (10:56 +0000)]
Removed code for the !KERNEL case. It hasn't been maintained for 4
years and gives a "laugh"able number of compile-time errors (see the
comment). main() just printed the struct sizes. This can be done
better by compiling with -g and reading off the sizes from the stabs.
bde [Tue, 18 Nov 1997 09:27:23 +0000 (09:27 +0000)]
Don't #include <machine/smp.h> even in the SMP case. Fixed the one
place that depended on it. The "bazillion warnings" mentioned in the
log for rev.1.45 apparently aren't a problem any more. It is hard
to be sure because the SIMPLELOCK_DEBUG option turns off (and breaks)
things in the SMP case.
bde [Tue, 18 Nov 1997 09:26:50 +0000 (09:26 +0000)]
Don't #include <machine/smp.h> even in the SMP case. Fixed the one
place that depended on it. The "bazillion warnings" mentioned in the
log for rev.1.45 apparently aren't a problem any more. It is hard
to be sure because the SIMPLELOCK_DEBUG option turns off (and breaks)
things in the SMP case.
Don't forward declare structs that are already implicitly forward declared.
bde [Tue, 18 Nov 1997 08:07:37 +0000 (08:07 +0000)]
Don't #include <sys/types.h>. I mistakenly #included it unconditionally
to fix some broken application. Application breakage is now hidden by a
recently introduced wrong #include of <sys/types.h> in <sys/time.h>.
<sys/time.h> is a prerequisite for <sys/resource.h>, so the <sys/types.h>
prereqisite is automatically supplied.
bde [Tue, 18 Nov 1997 07:23:40 +0000 (07:23 +0000)]
Fixed pedantic syntax errors caused by trailing semicolon in the
__ELF__ case of the definition of MAKE_SET() and in the PSEUDO_LKM
case of the definition of PSEUDO_SET().
brian [Tue, 18 Nov 1997 00:20:29 +0000 (00:20 +0000)]
Notice that ppp has closed the connection properly.
Remove the timeout hack to notice that ppp has closed the connection.
Remove the ``special case'' hacks for "quit" and "bye", as pppctl
now exits immediately when the connection is closed by ppp.
Suggest a secure "set server" line for connecting ppp & pppctl.
Tidy up and correct a few man page typos.
brian [Mon, 17 Nov 1997 01:13:41 +0000 (01:13 +0000)]
Don't SetLabel() 'till we've done the SelectSystem(). This
avoids the situation where we specify label ``x'' on the command
line, and label ``x'' has a ``load y'' command embedded in it.
When the line comes up, we want to use ``x'' from ppp.linkup,
not ``y''.
brian [Sun, 16 Nov 1997 22:15:11 +0000 (22:15 +0000)]
Abstract data read from and written to the tun device,
allowing for a possible header on the front of all packets.
In OpenBSD, there's a structure containing the address
family here.
If we're building under OpenBSD, set up the ``flags'' part
of struct tuninfo (not there under FreeBSD) so that we config
the interface as POINTOPOINT.
Prefix prototypes with ``extern'' in os.c for consistency.
These changes are cosmetic under FreeBSD, but allow ppp to
build & work under OpenBSD (bar the srandomdev() stuff,
the inclusing of <net/if_var.h> and some Makefile symantecs).
steve [Sun, 16 Nov 1997 12:52:17 +0000 (12:52 +0000)]
Resurrect a modified version of the /tmp clearing code from rev 1.119.
Enable this by setting clear_tmp_enable in rc.conf to YES. Beware
there can be serious side-effects of enabling this, so use at you own
risk.