| Age | Commit message (Collapse) | Author |
|---|
|
This might help with troubleshooting "iwx0: acquiring device failed"
errors.
OK stsp@
|
|
Even if the buffer is guaranteed to be NUL-terminated in a particular
case, it is still setting a bad example. Besides, it is unclear
to me whether there is any such guarantee in the case at hand.
Checking that would require auditing all of d2i_X509_bio(3),
ASN1_item_d2i_bio(&NETSCAPE_X509_it, ...), PEM_read_bio_X509_AUX(3),
and PKCS12_parse(3), since no such guarantee is documented for any
of these functions, and even then it would remain fragile with
respect to later changes of implementation details.
In the worst case, this could potentially result in a read buffer
overrun.
OK tb@ on an earlier version of this patch.
While we are here, deraadt@ requested to not use the word "string" in the
name of a variable that is not a string in the sense of the C language.
|
|
|
|
wskbd in order to make them visible in X11. Matches what ukbd(4) already
does.
|
|
and an omission below HISTORY
|
|
|
|
usage greater than zero. Usage zero is defined as unassigned by the
specification and cannot be mapped to anything sensible.
Prevents ucc from attaching to bunch of odd report IDs from a Lenovo
ThinkPad USB-C Dock which only exposes the unassigned usage. This is
not a problem in practice but I think we're better attaching them as
uhid devices instead as ucc cannot provide any functionality.
Thanks to Mario Peter <mp at mpeter dot de> for reporting and testing.
|
|
causing ucc to only being able to attach to the last report ID. This in
turn is caused by hid_is_collection() only being able to observe an end
of collection item with the last report ID for the same collection.
Instead, change the matching of ucc to only consider report IDs with at
least one corresponding Consumer Control usage.
Fixes gnezdo@'s Google Pixel earbuds.
|
|
Feature) from the corresponding descriptor report for a given report ID.
The ordering of the items is identical in both the descriptor and
interrupt report. As the interrupt report can cover more than Consumer
Control related key presses, ucc must be more careful while examining
the interrupt report in order to not confuse other items as key presses.
While parsing the descriptor report, take note of the bits that
represents Consumer Control key presses and use it to slice the
interrupt report.
Thanks to florian@ gnezdo@ and Alessandro De Laurenzis <just22 at
atlantide dot mooo dot com> for testing.
|
|
specification ucc might as well enumerate all of them. Finding an
appropriate scan code recognized by X11 for each usage is more tricky.
I've added a few more but the majority are still unmapped. Linux has
defined a couple of more usages covered by the evdev[1] key codes but
those symbols are not picked up in an vanilla X11 configuration on
OpenBSD, according to setxkbmap(1).
This should at least lower the barrier for adding scan codes for wanted
keys.
Note that the strings are discarded unless UCC_DEBUG is enabled.
Thanks to gnezdo@ for testing.
[1] xenocara/dist/xkeyboard-config/keycodes/evdev
|
|
viornd did not mask the descriptor value in the avialable ring
allowing guest values to read past the end of the descriptor table.
While here, change fatal to fatalx because errno is not set.
Reported by Ilja van Sprundel
ok mlarkin@
|
|
|
|
This fixes a problem in the perl regress where it notices the
callback is called twice and complains.
ok tb@ bluhm@
|
|
|
|
This makes some tests run from the Makefile behave as they should.
|
|
|
|
|
|
explicit.
Pointed out and OK bluhm@
|
|
Currently only simple include and excludes work, the advanced filters
introduced later in rsync are not implemented. It is unclear if the per
directory filters are something we want to implement. This requires
more modern protocols which openrsync is not able to handle right now.
This adds a special matching function to allow the ** matching which behaves
mostly like rsyncs version with the exception of how bad [] patterns are
expanded. For bad patterns openrsync follows more how fnmatch behaves and
not the somewhat strange rsync behaviour.
Not perfect but committing now so people can test and provide feedback.
|
|
Provide TIMOUT_{ACTION,PERIOD_SEC} in {start,stop,reset}_timeout() to have
install.sub and dot.profile use the same mnemonic helpers.
|
|
snmpd(8).
OK jmatthew@
OK deraadt@ for bumping libutil now.
|
|
|
|
Simplify code and employ the same technique in both places.
The "prompt timeout" hits when on non-interactive installations or upgrades:
Welcome to the OpenBSD/amd64 6.9 installation program.
Starting non-interactive mode in 5 seconds...
(I)nstall, (U)pgrade, (A)utoinstall or (S)hell?
Performing non-interactive upgrade...
The existing code uses a regular background job and does quirky file
descriptor fiddling just to avoid job control messages when starting the
timeout in the background.
It also does `set +m' a second time for no reason and hardodes the number of
seconds in multiple places.
Do better by using a co-process just like sysupgrade's watchdog that reboots
when the upgrade did not finish in time.
reads good to aja
|
|
noted by inoguchi
|
|
dhcpleased(8) does not support everything dhclient(8) did,
so stop pretending.
OK florian
|
|
|
|
doc fixes/ok jmc
ok beck
|
|
Guest can cause out of bounds read with a malformed descriptor. In same
loop, also fix a chunk size calculation.
Reported by Ilja van Sprundel.
ok mlarkin@
|
|
make NVMe on the Apple M1 stable. Hopefully we can figure out the real
issue in the future.
ok jmatthew@
|
|
If {c,m}alloc fail, info could be NULL and result in NULL deref.
Reported by Ilja van Sprundel.
ok mlarkin@
|
|
make NVMe on the Apple M1 stable. Hopefully we can figure out the real
issue in the future.
ok jmatthew@
|
|
Reported by Ilja van Sprundel.
ok mlarkin@
|
|
Used originally to aid dev. Unneeded.
ok mlarkin@
|
|
of running newfs_msdos, so fsck_msdos gets run behind the scenes. A few
architectures were missing this binary on their install media
jsg notices two more arch need it also, due to crazy Makefile games in installboot
|
|
of running newfs_msdos, so fsck_msdos gets run behind the scenes. A few
architectures were missing this binary on their install media
noticed by sthen
|
|
|
|
with help from sthen@ and tim@. OK tim@
|
|
|
|
Change the default -w from 5 to 3 seconds, reducing the time it waits
for a response. John proposed using 1 second for this, but several
of us are familiar with networks where that would be likely to
remove hops, 3 seems reasonable and still saves 6 seconds on a
non-responding hop (there's always mtr if you need faster output).
Feedback from various, ok phessler@ kn@ florian@
|
|
Due to the need to support by_dir, we use the get_issuer stuff when running
in x509_vfy compatibility mode amyway - so just use it any time we are
doing that. Removes a bunch of yukky stuff and a "Don't Look Ethel"
ok tb@ jsing@
|
|
ok inoguchi@ tb@
|
|
|
|
generate certdirs by jsing, and make chicken sacrifies by tb.
ok tb@ jsing@
|
|
place.
Allows single-digit partition id's in '-b' as a side benefit.
|
|
|
|
|
|
|
|
roots were not checked correctly before intermediates that has since been fixed
and is no longer necessary. It is regress checked by case 2c in
regress/lib/libcrypto/x509/verify.c
ok jsing@ tb@
|
|
this makes it easier to parse what passwd(1) is doing if
spawned from a GUI
ok millert@
|
|
doesn't deal with non-GTT mappings. What the Linux code does here isn't
possible on OpenBSD and probably unecessary.
Seems to fix a crash reported by sthen@
ok jsg@
|
