Not really an important issue, but here's a diff to bring the comments
in rnd.c more up to date. In particular:
- The PRNG uses ChaCha20 instead of RC4
- The first few bytes are not thrown away anymore
- The comments list ways to access the generated randomness; add
getentropy() to that list
In
When I mount my mp3-player (Archos 24y Vision) in OpenBSD, it always
takes quite a long time and I see the following output on the console:
> umass0 at uhub2 port 2 configuration 1 interface 0 "ARCHOS ARCHOS" rev
> 2.00/0.00 addr 8
> umass0: using SCSI over Bulk-Only
> scsibus4 at umass0: 2 targe
I had a look at httpd, and it appears that it doesn't actually use
any functions defined in sys/hash.h. Same for md5.h and sha1.h in
httpd.c. After removing those includes, it compiles and runs, but I
didn't do any serious testing.
Index: usr.sbin/httpd/httpd.c
To pass the time until siphash arrives in userland...
I'd like to suggest some changes to the manpage of SipHash24(9):
I added some more details about the intended purpose of SipHash and I
moved the recommendation to use arc4random for the key to the beginning.
In case you don't like this diff, I
Here's a version without If, but with the paper, then.
Since, it's not an official standard, I figured that the HISTORY
section would be the right place. Correct?
Index: share/man/man9/SipHash24.9
===
RCS file: /cvs/src/share/man/man
This diff replaces hash32 by SipHash in the hash table in ospf{,6}d/lsupdate.c.
Seems pretty straightforward...
Index: usr.sbin/ospfd/lsupdate.c
===
RCS file: /cvs/src/usr.sbin/ospfd/lsupdate.c,v
retrieving revision 1.41
diff -u -p -
This diff converts relayd from hash32 to siphash. Relayd does not use
the hash function for an actual hash table, but can select a host based
on the hash of the source-address, destination-address and other data.
The key is generated when the hash function is first used (this may
happen in two diff
Sorry...I just realized that I forgot to check for malloc return
values. I need some sleep.
Here's a version without malloc'ing the key.
Index: usr.sbin/relayd/parse.y
===
RCS file: /cvs/src/usr.sbin/relayd/parse.y,v
retrieving revision 1.195
diff -u -p -r1.195 parse.y
--- usr.sbin/relayd/parse.y 20 Nov 2014 05:51:20 -
On Fri, Dec 12, 2014 at 12:57:08PM -0500, Ted Unangst wrote:
> On Wed, Dec 10, 2014 at 21:51, Max Fillinger wrote:
> > Here's a version without malloc'ing the key.
>
> Looks like it still does?
Yes, I managed to just repost the previous diff...at least the thing is
in more competent hands now.
Currently, there's code in acpi.c that sends the system back to sleep
when resuming with closed lid and machdep.lidsuspend=1. I often use my
laptop in a docking station with an external monitor and keep the lid
closed, and I'd like to be able to resume just by pushing the power
button on the dockin
Some small issues in the new section:
- 'NTP servers' means actual servers, not the keyword, so remove the .Ic
- 'constraint' is used instead of 'constraint from' in some places; that
might be ok as an abbreviation, but at least the example should use
the full keyword. In the diff below, I us
Thanks a lot for working on this!
I've tested the diff with my USB headphone amp:
uaudio0 at uhub2 port 1 configuration 1 interface 0 "Burr-Brown from TI USB Audi
o DAC" rev 1.10/1.00 addr 6
uaudio0: audio rev 1.00, 2 mixer controls
audio1 at uaudio0
uhidev2 at uhub2 port 1 configuration 1 interf
Here's a diff to enable DUID support for bioctl -d. Basically, I did as
the comment in bio_deleteraid suggests and replaced the string devname
with a dev_t. In softraid.c, I just recover the "sdX" from the dev_t
using findblkname and the DISKUNIT macro.
Or am I missing the point of that comment he
It's out, not outl, that needs to contain sufficient room. Also, add a
pair of parentheses to mark EVP_DecryptInit_ex as a function.
Index: lib/libssl/src/doc/crypto/EVP_EncryptInit.pod
===
RCS file: /cvs/src/lib/libssl/src/doc/crypto
On Sun, Apr 26, 2015 at 11:17:36PM +1000, bytevolc...@safe-mail.net wrote:
> Whilst I mostly like this patch, and it appears to be much cleaner than
> the bodgy "solution" I came up with a few years back, perhaps there is
> something here worth mentioning:
>
> ...
I'm glad you like it, but the de
My uaudio device stopped working with this patch:
$ aucat -i test.wav -f snd/0
snd/0: couldn't open audio device
With AUDIO_DEBUG enabled, I see the following output on the console:
audio1: setpar: req enc=10 bits=16, bps=2, msb=1 rate=48000, pchan=2, rchan=2,
round=960, nblks=2
audio1: 0x180:
On Sat, Jul 04, 2015 at 01:40:00PM +1000, Steven McDonald wrote:
> +sudo(8) has been removed from the base OS.
> +The old binaries and manual pages should be removed:
> +
> + rm -f /usr/bin/sudo /usr/bin/sudoedit /usr/sbin/visudo
> + rm -f /usr/share/man/man8/sudo.8 /usr/share/man/man8/
On Wed, Jul 08, 2015 at 03:53:46PM +0200, Mark Kettenis wrote:
> I'm looking for testers for this diff. This should be safe to run on
> amd64, i386 and sparc64. But has been reported to lock up i386
> machines. I can't reproduce this on any of my own systems. So I'm
> looking for help. I'm loo
I tested the diff with a FiiO E10K Olympus 2.
Previously, it only produced stuttering audio, which I assumed was
because it's a UAC v1.0 device and I connected it on ehci. Now, it
can't be opened at all.
The first thing I noticed was that the kernel now finds two uaudio
devices instead of one (t
The man page says that which(1) works with unset PATH environment
variable, as long as the argument is a pathname. However:
$ unset PATH
$ /usr/bin/which /usr/bin/which
which: can't get $PATH from environment: Undefined error: 0
Index: usr.bin/which/which.1
===
If PATH starts with "/:", which(1) reads outside of allocated memory.
Maybe that caused the non-reproduceable coredump mentioned in [0]?
The function progname sets path = strdup(path) and pathcpy = path and
runs the following loop:
> while ((p = strsep(&pathcpy, ":")) != NULL) {
> if (*p
I looked for more instances of the pattern that lead to reading one byte
before an allocated buffer in which(1) when PATH begins with "/:". I
found only one, in the function csexists() in usr.bin/mg/cscope.c.
While at it, I replaced the manual length check before snprintf() with a
check of the ret
On Tue, Jan 19, 2016 at 12:35:27PM +0100, Sunil Nimmagadda wrote:
> > - dlen = strlen(dir);
> > - while (dir[dlen-1] == '/')
> > - dir[--dlen] = '\0'; /* strip trailing '/' */
> dlen could never be zero as we are replacing dir[0] with '.' if
> it's
24 matches
Mail list logo