horikawa [Wed, 15 Aug 2001 01:30:25 +0000 (01:30 +0000)]
Removal of following export controll related sentences:
o Because of export controls, TELNET ENCRYPT option is not supported outside
of the United States and Canada.
o Because of export controls, data encryption
is not supported outside of the United States and Canada.
src/crypto/README revision 1.5 commit log says:
> Crypto sources are no longer export controlled:
> Explain, why crypto sources are still in crypto/.
and actually telnet encryption is used outside of US and Canada now.
Pointed out by: OHSAWA Chitoshi <ohsawa@catv1.ccn-net.ne.jp>
Reviewed by: no objection on doc
horikawa [Wed, 15 Aug 2001 01:23:30 +0000 (01:23 +0000)]
Removal of following export controll related sentences:
o This option is not available outside of the United States and Canada.
o Because of export controls, TELNET ENCRYPT option is not supported outside
of the United States and Canada.
src/crypto/README revision 1.5 commit log says:
> Crypto sources are no longer export controlled:
> Explain, why crypto sources are still in crypto/.
and actually telnet encryption is used outside of US and Canada now.
Pointed out by: OHSAWA Chitoshi <ohsawa@catv1.ccn-net.ne.jp>
Reviewed by:no objection on doc
imp [Tue, 14 Aug 2001 23:34:09 +0000 (23:34 +0000)]
Merge from stable (which seems to have been spammed at some point in current):
#ifdef the deltap pcic_set_memory_offset argument so that raylink
driver works.
wpaul [Tue, 14 Aug 2001 21:56:53 +0000 (21:56 +0000)]
Fix yet another "unconditionally free()ing memory without even checking
to see if it was malloc()ed first" bug. In bus_dmamap_create(), one of
two things can happen: either we need to allocate a special map due to
some manner of bounce buffering requirement, or we can DMA a buffer
in place. On the x86 platform, the "in place" case results in
bus_dmamap_create() returning a dmamap of NULL. The bus_dmamap_destroy()
routine later checks for NULL and won't bother free()ing the map if
it detects this condition.
But on the alpha, we don't use NULL, we use a statically allocated map
called nobounce_dmamap(). Unfortunately, bus_dmamap_destroy() does not
handle the condition where we attempt to destroy such a map: it tries
to free() the dmamap, which causes a panic.
Fix: test that map != &nobounce_dmamap before trying to free() it.
With this fix, my busdma-ified if_sis driver works on the alpha. I'm
a bit alarmed that I'm the first person ever to trip over this bug, since
we have been using busdma on the alpha for a while, and since it sort
of screams out "Hi! I'm a bug! Booga-booga!" when you look at it.
(Somewhere, somebody will say: "But Bill, why don't you just not bother
destroying the maps in this case." Because the API is supposed to be
a) symetrical and b) opaque to the caller. I can't know whether it's safe
to skip the bus_dmamap_destroy() step or not without sticking my fingers
into unsafe places, which is what I wanted to avoid in the first place.)
imp [Tue, 14 Aug 2001 19:18:33 +0000 (19:18 +0000)]
Ask the kernel about IRQ 0 first. If the kernel responds with an IRQ,
use it. If not, then loop asking for each one, with normal -I
processing. This will effectively disable -I for when the pcic is in
PCI function interrupt routing mode.
ru [Tue, 14 Aug 2001 15:19:35 +0000 (15:19 +0000)]
From ChangeLog:
: 2001-08-13 Ruslan Ermilov <ru@FreeBSD.org>
:
: * tmac/doc.tmac (Rv): Implement support for 0 or more than 1
: argument.
: * tmac/groff_tmac.man: Updated.
All of the following are now valid:
.Rv -std f1
.Rv -std f1 f2
.Rv -std f1 f2 f3
.Rv -std
The last form is useful where the standard return behavior is applicable
to all of the functions described in this particular manpage. Look, for
example, at the stat(2), setresuid(2), and utimes(2) manpages.
The form with >1 functions is useful in a mixed environment. See the
fhopen(2) manpage for an example.
imp [Tue, 14 Aug 2001 07:31:47 +0000 (07:31 +0000)]
For TI12xx and newer, clear register 0x84. This is the "Multimedia"
register. It enables Zoom Video. It appears that on at least one
card that Monzoon is using sets these bits by default. Nothing works
when these bits are set, everything works when they are clear.
Add commentary on some of the ti bits. Make code a little clearer.
Also remove a call to pcic_pci_pd6729 which was prematurely added in
the last commit.
phk [Tue, 14 Aug 2001 06:42:32 +0000 (06:42 +0000)]
linux ls fails on DEVFS /dev because linux_getdents fails because
linux_getdents uses VOP_READDIR( ..., &ncookies, &cookies ) instead of
VOP_READDIR( ..., NULL, NULL ) because it seems to need the offsets for
linux_dirent and sizeof(dirent) != sizeof(linux_dirent)...
PR: 29467
Submitted by: Michael Reifenberger <root@nihil.plaut.de>
Reviewed by: phk
dd [Tue, 14 Aug 2001 05:50:19 +0000 (05:50 +0000)]
Sync the code that sucks in rc.conf and friends with what's in
rc.firewall6. Specifically, don't do anything
if [ -z ${source_rc_confs_defined} ]. Not doing this leads to a problem
with dependencies: chkdepend will set, e.g., portmap_enable to YES if
some service that needs portmap is enabled, but rc.network sources
rc.firewall, which used to source defaults/rc.conf unconditionally,
which would result in portmap_enable being set back to NO.
imp [Tue, 14 Aug 2001 05:31:56 +0000 (05:31 +0000)]
Try to support the Cirrus Logic PD6833 better in pci mode. This is
done from the datasheets since I'm ahving problems with my laptop that
has the 6833 in it...
Minor rework of TI code. Should be the same as before.
jhb [Mon, 13 Aug 2001 22:44:55 +0000 (22:44 +0000)]
In prom_halt(), set the halt restart flags on the current CPU, not the
boot CPU. This was the reason reboots on SMP systems could result in
weird hangs. Unlike the x86, we do not need to switch back to the boot
CPU in order to reboot the machine. See Section 3.4.5 of Part III
(Console Interface Architecture) from the Alpha Architecture Reference
Manual (aka the Brown Book) for more info.
peter [Mon, 13 Aug 2001 21:48:44 +0000 (21:48 +0000)]
Rip out the old __stdin/out/err stuff. It was completely 100% useless. :-(
It was foiled because of dynamic copy relocations that caused compile-time
space to be reserved in .bss and at run time a blob of data was copied to
that space and everything used the .bss version.. The problem is that
the space is reserved at compile time, not runtime... So we *still* could
not change the size of FILE. Sigh. :-(
Replace it with something that does actually work and really does let us
make 'FILE' extendable. It also happens to be the same as Linux does in
glibc, but has the slight cost of a pointer. Note that this is the
same cost that 'fp = fopen(), fprintf(fp, ...); fclose(fp);' has.
Fortunately, actual references to stdin/out/err are not all that common
since we have implicit stdin/out/err-using versions of functions
(printf() vs. fprintf()).
peter [Mon, 13 Aug 2001 21:21:13 +0000 (21:21 +0000)]
Revert part of the 1.11->1.11.1p1 change. Put the blank line back in
the beginning of constructed log messages being passed to $EDITOR. This
is where folks start typing.
semenu [Mon, 13 Aug 2001 18:37:31 +0000 (18:37 +0000)]
Fix an bug in FreeBSD attach routine - attaching MII interfaces before doing
common_attach is wrong as common attach initialize some fileds used by
mediainit routine. This was hard to notify because loading driver as kld
lead to mediainit routine being called after common_attach, though probe_phy
is called before.
bde [Mon, 13 Aug 2001 13:22:50 +0000 (13:22 +0000)]
Use interrupt gates instead of trap gates for breakpoint and trace
traps, so that ddb can keep control (almost) no matter how it is
entered. This breaks time-critical interrupts while the system is
stopped in ddb, but I haven't noticed any significant problems except
that applications become confused about the time. Lost time will be
adjusted for later. Anyway, the half-baked disabling of interrupts in
Debugger() gives the same problems for the usual way of entering ddb.
bde [Mon, 13 Aug 2001 12:42:13 +0000 (12:42 +0000)]
Removed he BPTTRAP() macro and its use. It was intended for restoring
bug for bug compatibility to ddb trap handlers after fixing the debugger
trap gates to be interrupt gates, but the fix was never committed. Now
I want the fix to apply to ddb.
mike [Mon, 13 Aug 2001 05:56:27 +0000 (05:56 +0000)]
Fix some bogus strncpy(3) to strlcpy(3) changes I made in the previous
revision. <utmp.h> structures don't leave room for a NUL character.
Also fix "UNKNOWN" which should have just been UNKNOWN.
iedowse [Sun, 12 Aug 2001 21:08:34 +0000 (21:08 +0000)]
Update io(4) to match reality following revision 1.42 of
sys/i386/i386/mem.c: only the super-user may open /dev/io
regardless of the device permissions (just 4 years late!).
Also, add cross-reference to i386_{get,set}_ioperm(2).
mharo [Sun, 12 Aug 2001 20:32:37 +0000 (20:32 +0000)]
Remove whitespace after %. The whitespace was causing strfile to not see
the % string section break and thus fortune was producing a rather large
multi-fortune fortune message.
markm [Sun, 12 Aug 2001 18:47:56 +0000 (18:47 +0000)]
Remove the WANT_INSECURE_OPIE option - it is now a default. This is not
nearly as ominous as it sounds, and it allows OPIE to be used over SSH
and on xterms.
mike [Sun, 12 Aug 2001 17:54:49 +0000 (17:54 +0000)]
o Replace occurrences of strncpy(3) with strlcpy(3); most of
the uses of it were wrong anyway.
o Always check for NULL returns on strdup(3).
o Fix a possible buffer overflow in strcpy(3).
o Fix a format string vulnerability.
o t->ty_type in stypeof() could be NULL and eventually cause
a segmentation fault in setenv(3), so check for that.
Eyeballed by: kris
Reviewed by: murray
MFC after: 3 days
iwasaki [Sun, 12 Aug 2001 08:37:39 +0000 (08:37 +0000)]
Fix some trivial bugs.
- fix segment limit mis-calculation for GCODE_SEL, GDATA_SEL, GPRIV_SEL,
LUCODE_SEL and LUDATA_SEL.
- move `loader(8) metadata' related printf() after cninit().
- use atop macro (address to pages) for segment limit calculation
instead of i386_btop macro (bytes to pages).
- fix style bugs for the declarations of ints.
ume [Sat, 11 Aug 2001 15:01:12 +0000 (15:01 +0000)]
Fill _res.sort_list with harmless entry. sortlist for IPv6/IPv4
is stored in _res_ext.sort_list, and sortlist for IPv4 is stored in
_res.sort_list for backward compatibility. However, both sort_list's
are maintaind by just one index _res.nsort. So, when IPv6 address is
specified to sortlist, empty entry was created in _res.sort_list. It
broke sortlist facility of gethostbyname().
Discussed on users@jp.ipv6.org.
markm [Sat, 11 Aug 2001 12:37:55 +0000 (12:37 +0000)]
Fix:
/usr/src/lib/libpam/modules/pam_ssh/pam_ssh.c has couple of bugs which cause:
1) xdm dumps core
2) ssh1 private key is not passed to ssh-agent
3) ssh2 RSA key seems not handled properly (just a guess from source)
4) ssh_get_authentication_connectionen() fails to get connection because of
SSH_AUTH_SOCK not defined.