.conf'. This way, you can
keep all the other errors and throw way thsese ones. See syslog.conf(5) for
an example of the filter syntax.
HTH,
-RVP
gba: vrgb
Xft.lcdfilter: lcddefault
```
I have `*renderFont: false' because bitmap fonts are faster. If I want to see
Unicode chars., I simply turn on `TrueType Fonts' from the menu, and font-size
selection works as usual.
HTH,
-RVP
at's used, even when a newer capfile exists.
(For small files like `login.conf' don't bother making a `.db' file.)
-RVP
istro? unzstd(1) and copy the files from (on Ubuntu):
/usr/lib/firmware/rtlwifi/
-RVP
unlikely:
1. Noto fonts've a different license.
2. They're HUGE.
-RVP
ssing
ref. in CHANGES-11.0.
-RVP
On Sat, 27 Sep 2025, Mayuresh wrote:
On Fri, Sep 26, 2025 at 08:20:17PM +, RVP wrote:
https://github.com/NetBSD/src/blob/netbsd-10/sys/arch/amd64/conf/GENERIC#L1132
Seems like it's to be enabled.
Yup.
Is vio9p likely to be faster than NFS?
Try it and tell us! I use tar + ss
hread.
Turning vesa on lets the kernel use the framebuffer provided by BIOS for
the console display.
-RVP
On Thu, 25 Sep 2025, Mayuresh wrote:
[...] or does the driver need to be enabled explicitly on NetBSD?
This:
https://github.com/NetBSD/src/blob/netbsd-10/sys/arch/amd64/conf/GENERIC#L1132
-RVP
H265. For H264 streams, add
`--vo=gpu --hwdec=auto' or `... --hwdec=nvdec'. (The first is what applies to
all HW.) Then, use `I' (capital) to see if mpv is using HW decoding. Toggle
with `Ctrl-h'.
-RVP
[1]: https://en.wikipedia.org/wiki/NVDEC
rf - /usr/src/
[..]
/usr/src/sys/fs/msdosfs/msdosfs_vnops.c:vap->va_mode |=
S_ARCH1;
[...]
$
```
Nobody seems to use `S_ARCH2'.
-RVP
27;t documented in mount.8. The `acls' synonym is missing too:
both in the man-page and in `src/sbin/mount_ffs/mount_ffs.c'. Should be
mentioned in {get,set}facls(1).
File a PR.
-RVP
no
`*.h' files.
The official builds[1] are all building. Wipe the object dir. then try again.
It's some local issue for sure.
[1]: https://nycdn.netbsd.org/pub/NetBSD-daily/netbsd-10/
-RVP
X11R7/lib/X11/app-defaults/XTerm'.
-RVP
will list all the valid values for
LANG= and LC_ALL= on Unixes. (NetBSD's locale has a minor bug: it shows some
values which aren't valid locale names in the NetBSD implementation, like `be',
`ca' or `nl' etc.)
(And, my LANG= example above has a typo: it's `en_GB.UTF-8', not `en.GB.UTF-8')
-RVP
nerate any keycodes at all [...]
Try xbindkeys in pkgsrc.
-RVP
ze: 12
XTerm*locale: true
XTerm*renderFont: true
```
(Note: need separate settings for UXTerm)
```
$ cat ~/.xinitrc
#!/bin/sh
export LANG=en.GB.UTF-8
[...]
xrdb -merge ~/.Xresources
[...]
$ cat ~/.xsession
#!/bin/sh
export PATH=$PATH:/usr/pkg/bin:/usr/pkg/sbin
. ~/.xinitrc
$
```
All this is basic X11 stuff...
-RVP
Seems like a HW issue to me. Try a different card?
-RVP
e max. button sizes:
```
$ cat tmp/ctwmrc.patch
--- /etc/X11/ctwm/system.ctwmrc 2024-05-31 13:11:41.0 +
+++ /home/rvp/.ctwmrc 2025-08-29 06:45:27.145772726 +
@@ -96,7 +96,7 @@
IconFont "-*-spleen-*-*-*-*-16-*-*-*-*-*-*-*"
ResizeFont "-*-spleen-*-*-
t you want (it shows per-connection stats, rather
than the per-host you want), but, a mild amount of hacking'll fix that :).
-RVP
be the fore/background colors the same. You can't
see the numbers or buttons for virtual desktops to select. You can't
see the ok, cancel, all buttons.
CTWM bug. See PR# 58467.
-RVP
rning accelerated drawing off in `xorg.conf':
Section "Device"
Identifier "Card0"
Driver "suntcx"
Option "NoAccel" "true"
EndSection
I wonder how much of this qemu bugs and how much might be real application
bugs ...
Someone with real HW should be able to check this...
-RVP
On Fri, 15 Aug 2025, Alistair Crooks wrote:
Just a thought:
find /dir/ -type f -ls | awk '$3 ~ /x/ { print $NF }'
Fine thought. +1.
-RVP
ready installed the new kernel, just reboot into it, then do the
etcupdate.
-RVP
ne might do) and then run this
command:
```
# makemandb -fv > /tmp/m.txt 2>&1
```
That should show what file it's choking on.
Do the backtrace with the debug symbol unpacked, too.
-RVP
[1]:
https://github.com/NetBSD/src/blob/trunk/usr.sbin/makemandb/makemandb.c#L1995
t".
Thanks. A distinction I missed: of course there can exist files executable by
others, but not by you.
-RVP
should handle weird filenames:
$ find /dir/ -type f -exec sh -c 'for f; do test -x "$f" && echo "$f"; done'
xxx {} +
Note the `xxx' above. It, or some other string, is reqd. there to supply a `$0'
for the shell. Of course, you would usually put all this in a shell script, then
`-exec' that :).
-RVP
, 96000, 192000 }
(PR) slinear_le 20/32, 2ch, { 44100, 48000, 96000, 192000 }
(PR) slinear_le 24/32, 2ch, { 44100, 48000, 96000, 192000 }
( ) ac3 16/16, 2ch, { 44100, 48000, 96000, 192000 }
Doing,
audiocfg default 1
as root should fix sound.
-RVP
On Mon, 28 Jul 2025, Niels Müller Larsen wrote:
When I run ‘obpager &’ from a terminal after X is started, it starts nicely
showing my workspaces and everthing works.
The problem is that I don’t get the pager on the screen if I try to run ‘obpager
&’ from either .xsession or .config/autostart
s (post /var/run/dmesg.boot) get a better resolution like this:
https://wiki.netbsd.org/tutorials/x11/how_to_use_wsfb_uefi_bios_framebuffer/
-RVP
olve the problem.
Is this a known issue? Any help is greatly appreciated.
Try setting `kern.timecounter.hardware' to the next best time source `hpet0',
or run without `-accel=nvmm' (and turn off TSC with `-cpu qemu64,-tsc').
You could also try compiling the host kernel to run with a higher HZ, like 1000.
-RVP
On Mon, 21 Jul 2025, Cameron Kelly wrote:
I tried a basic html page with some css styling and a few photos. I hosted it
on a VPS with SSD and it seemed to take way longer than it should have to
load the page.
With or without the `-n' flag tp httpd?
-RVP
-raspa4#
On ARM, the whole-disk device is `rsd0c' (as seen in the disklabel output).
It's `rsd0d' only on x86 systems, I believe. I just use `/dev/rXXX'. It'll
be whole-disk device everywhere (whether `c' or `d'), I think.
-RVP
reate an Hybrid MBR. If not, then
beaker can always fire up a good hex editor :)
-RVP
y partition) and throws this error.
Try creating an Hybrid MBR with an NetBSD-type MBR entry with start and size
corresponding to that of `dk2' (gdisk(8) on Linux).
-RVP
#x27;t return any replies from most sites these days.
Just an FYI.
-RVP
et this. (It's _not_ any of the GPT GUIDs.) As you can see, I have
separate EFI ESP partitions for each of the OSes.
HTH,
-RVP
to the qemu
instance worked... except:
a) the telnet OTP timeout is _very_ short (better cut+copy the OTP rather
than typing)
b) because of the OTP length, it's easy to make typos.
c) spaces between words don't seem to matter, except that you need to have
at least one.
HTH,
-RVP
hese lines uncommented:
options DIAGNOSTIC
options DEBUG
options LOCKDEBUG
That should help track down this issue considerably. Then, if there's kernel
crash dump, upload it somewhere public and post the URL and crash messages
to the 1st PR.
-RVP
into the kernel.
-RVP
im -g
if compiled with a GUI, or provide proper a terminal for it using a script
like:
```
#!/bin/sh
exec xterm -g 100x35 -e vim "$@"
```
(Depending on how the file manager run this script or cmd. line, you may have
to append a `&` to run it in the background.)
HTH,
-RVP
rog-if 00 [VGA controller])
Looks like a supported card. See list here:
https://github.com/NetBSD/src/blob/netbsd-10/sys/external/bsd/drm2/dist/drm/amd/amdgpu/amdgpu_drv.c#L957
-RVP
ncsession-start script
in that dir.
HTH,
-RVP
echo "Initializing MariaDB database system tables."
+ test -t 1 && echo "Initializing MariaDB database system tables."
sh /usr/pkg/bin/mariadb-install-db --force \
--user=mariadb \
--datadir=$mariadb_datadir
```
-RVP
tically:
```
$ locale -a | fgrep en_US
[...]
$ export LANG=en_US.UTF-8 # for instance
```
-RVP
y of
the real terminals wscons is emulating :)
-RVP
On Sat, 8 Mar 2025, Van Ly wrote:
RVP writes:
Is the kbd-type correct? What does `setxkbmap -query` print?
$ setxkbmap -query
rules: base
model: pc105
layout: us
options:ctrl:swapcaps
I don't know how to find the kbd-type. The ThinkPad interoperates wit
ee `⏏︎' eject key?
Surprising that `xev -event keyboard` isn't showing anything. Is the kbd-type
correct? What does `setxkbmap -query` print?
-RVP
put mode", implying _not_ available in command
mode, though clearly some of the others do work in command mode), so if
it works, well, that'd be nice.
It does work at the `:' prompt too, actually. :)
I am still puzzled by the lack of mention in the built-in cheat-sheet.
I think the documentation reflects the state of nvi-1.79...
-RVP
le in the man-page isn't correct: the "file://..." should be in
quotes as per the grammar:
pool ipf/tree (name sfs4;) { "file:///etc/ippool/test.txt"; };
-RVP
On Sun, 23 Feb 2025, Greg A. Woods wrote:
At Sun, 23 Feb 2025 06:44:02 + (UTC), RVP wrote:
Subject: Re: stupid question on vi
nvi(1) has a poorly-documented Ctrl-X key
Looks very well documented to me:
[0-9A-Fa-f]+
Insert a character with the specified hexadecimal
:print:][:cntrl:]] # non-ASCII chars.
That works, except that \x00 is _not_ regarded as a control-char.
-RVP
native-PCI mode
[ 1.011893] pciide0: using ioapic0 pin 19 for native-PCI interrupt
[ 1.011893] atabus0 at pciide0 channel 0
In the BIOS setup, switch the IDE controller into AHCI mode.
-RVP
refix=/usr/pkg --sysconfdir=/usr/pkg/etc
otherwise, `configure' thinks NetBSD has sendfile(1)--which it doesn't. (Else
delete line 79 in `configure.in` then regen `configure` the std. way...)
-RVP
PS. Oh! and make sure none of the `-D_FORTIFY_SOURCE=N` /
`-fstack-protector-XXX`
flags
is to deploy DNS service for
machines with private IPs ?
Just add any of the public DNS servers (like Google's 8.8.8.8) as the 1st
nameserver in /etc/resolv.conf. These will fail private IP queries very
quickly.
-RVP
eeds to be merged into pkgsrc libfetch.
-RVP
ons (TSO4,TSO6,...) set on the Mac
wg i/f?
-RVP
`ifconfig -a' on both the server and client?
Have the VMs assigned a 10.x.x.x (again, std. for QEMU in some config.) to
the virtual i/f addresses?
If so, can you reassign the Wireguard addresses to some other range?
-RVP
ne didn't start/won't do).
Run a TCP server with: nc -l 9443 (after stopping WG, of course).
-RVP
On Tue, 14 Jan 2025, RVP wrote:
Does a regular connection work? Use the same Wireguard port:
On the server:
nc -u -l 9443
On the client:
nc -u server-ip 9443
Type something on both ends. Try a TCP connection, too (w/o `-u').
Does the TCP connection break after a while when typing, or
both ends. Try a TCP connection, too (w/o `-u').
Does the TCP connection break after a while when typing, or is it stable
until you kill nc(1)?
-RVP
you start over? And this time, run script(1) on both the client and server
and do the config. within that so that we can get a verbatim transcript. Just
edit the `typescript' files to redact any info. you consider sensitive, and to
get rid of the `\r' chars.
-RVP
12 13:27:09 2025
How did a configured `--allowed-ips=10.2.0.42/32' change to
`allowed-ips: 192.168.10.0/24' on the interface?
-RVP
core-dump on
the restore...
-RVP
,
_NET_WM_STATE_MAXIMIZED_HORZ, _NET_WM_STATE_FULLSCREEN, _NET_WM_STATE_FOCUSED
```
-RVP
PN+/s/6UZdk0kc=
[...]
# wgconfig wg0 add peer sevastopol \
1WaiYse6arup/pNqos7CyvtsTm6O8PN+/s/6UZdk0kc= \
--allowed-ips=10.2.0.42/32
and,
2) it should be the other side's pubkey, not the server's own.
-RVP
screen EWMH property's set; but, after a restart
this is lost. Doesn't look like ctwm handles these EWMH properties that're
already set on a window when it starts. This isn't a NetBSD bug, more like
a ctwm feature request...
-RVP
://github.com/NetBSD/src/tree/trunk/usr.sbin/cpuctl
-RVP
t in BIOS mode,
then I get the same panic.
HTH,
-RVP
On Sat, 28 Dec 2024, Mayuresh wrote:
Hetzner has not provided 10.1 installer in their stock images. I have
requested them for one.
https://www.unitedbsd.com/d/1262-howto-run-netbsd-on-hetzner-cloud
-RVP
cpi.'
though I don't know if there ought to be.
I'm wondering if there's anything I might try to correct this issue
and/or possible work-arounds if not actualyl fixable?
I would recommend you a) get a new battery and b) update your firmware.
-RVP
he genfb driver has attached. With genfb, only the primary display
gets output. See wsdisplay(4).
And, if you want to rotate the picture, you can only do it in X. See wsfb(4).
Try,
fgrep genfb /var/run/dmesg.boot
see if it says `drm at genfb0 not configured'.
-RVP
.
Can't reproduce this on the latest 10.0_STABLE. Fullscreen emacs (with F11)
remains fullscreen after a ctwm restart.
-RVP
the
browser.
```
D="/tmp/.$(id -nu)-cache"
test -d "$D" || install -d -m 700 "$D"
env XDG_CACHE_HOME="$D" firefox ...
```
-RVP
wd(NULL, 0)' extension to POSIX
(in fact all the BSDs and Linux do), so just use the impl. in the C library.
This looks like a regression from NetBSD-9.3, although I'm not sure if
-D_FORTIFY_SOURCE=2 flag was always there, or recently added with
pkgsrc-2024Q3.
You need an optimization flag also for this to manifest..
-RVP
p. Contact christos@/martin@
-RVP
.html
https://mail-index.netbsd.org/source-changes/2023/11/10/msg148487.html
https://mail-index.netbsd.org/source-changes/2023/11/15/msg148509.html
https://mail-index.netbsd.org/source-changes/2023/11/15/msg148508.html
https://mail-index.netbsd.org/source-changes/2023/11/15/msg148507.html
-RVP
uot;; hexdump -C) < /dev/dtyU0
and if the device is constantly spewing data at the serial port, you should
see it now.
-RVP
PS. you wrote:
[...] and I can see all bytes returned by the GPS module when asked with the
@@Ea command by the PIC microcontroller.:
Does this mean that you have someh
ng only non-printable chars.--like NULs--which it
isn't. I also converted Ramiro's hexdump output back into binary
yesterday and blasted it to a terminal and _stuff did show up_).
-RVP
ifference between ttyU0 and dtyU0,
(as a Linux guy, all this is a bit obscure...), and maybe we should've been
using the dial-out device all along.
HTH,
-RVP
...
Something like
(stty 9600 raw; cat)
Shell might get stuck on open. This should work, I think:
$ { sleep 1; stty -f /dev/ttyU0 ... ; } & cat /dev/ttyU0
-RVP
"InputDevice"
Identifier "Tablet0"
Driver "ws"
Option "Device" "/dev/wsmouse1"
EndSection
$
```
but, hot-plugging is lost, and X will behave badly when USB-devices disconnect.
The keyboard gets stuck on some key and just keep repeating, for example. And,
unplugging a USB mouse when it's in use can freeze the kernel.
-RVP
On Mon, 14 Oct 2024, Ramiro Aceves wrote:
Hello RVP, many thanks for the tip and explanation but it does not work:
netbsd-raspa4$ stty -f /dev/ttyU0 9600 raw
netbsd-raspa4$ cat /dev/ttyU0
nothing happens...
Try adding `clocal cread -crtscts' too.
-RVP
doesn't get anything
until a) either a '\n' in input is encountered, or b) 255 chars. are read.
$ stty -f /dev/ttyU0 9600 raw; cat /dev/ttyU0
-RVP
r the keyword _subdir.
# default
-_default /usr/{share,X11R7,pkg,local}/man/
+_default /usr/{share,X11R7,pkg,local}/man/ /opt/lynx/share/man/
# Other sections that represent complete man subdirectories.
X11R7 /usr/X11R7/man/
$
```
-RVP
?
Fixed in: https://gnats.netbsd.org/57003
-RVP
On Sat, 3 Aug 2024, Van Ly wrote:
I see no difference using either
$ LC_ALL=en_US.UTF-8 xman
$ LANG=en_US.UTF-8 xman
I think Michael meant to say use the "C" locale :). Then, mandoc will produce
ASCII output:
env LC_ALL=C xman ...
-RVP
using logger(1), which are still logged to _both_
locations for some reason).
I think this should also work (provided the "from" isn't a literal IP
address!):
!*
+host1
*.* /var/log/host1
But, I think the prev. version is better as it covers both possibilities.
-RVP
NR
}
END {
for (i = 1; i <= NR; i++)
if (i in L)
print L[i]
}
```
Gets rid of duplicate lines and lines beginning with blanks (just like in
bash). Suggestions for improvement welcome.
-RVP
-page:
!* <-- add this line
+192.168.1.200
*.* /var/log/host1
-RVP
ogout. Ie. write the history first, then remove the
duplicate entries.
Thanks,
-RVP
ept up. Is there a work-around?
Set `ARCHIVE_EXTENSION=tar.xz` in the config. file (sysupgrade.conf?).
-RVP
ff using:
echo 'check_pkg_signatures=NO' >> /etc/security.conf
-RVP
login shell.
An interactive bash instance only reads ~/.bashrc.
See the INVOCATION section of bash(1) for all the details.
-RVP
On Tue, 18 Jun 2024, Riccardo Mottola wrote:
RVP wrote:
OK. I have the cause and a workaround. You can do a proper fix based on
it:
It's the NetBSD shell, /bin/sh (and also /bin/ksh). On NetBSD's /bin/sh:
LINENO RANDOM SECONDS START_TIME ToD
change every time you
, I boobed this the first time: I ran mach build w/o a .mozconfig. Don't
worry about this. The follow-up mail I sent after this is more interesting.
-RVP
/syslog-ng/syslog-ng.conf file:
```
$ head /usr/pkg/etc/syslog-ng/syslog-ng.conf
@version: 3.38
options { dir_perm(0755); perm(0644); chain_hostnames(no);
keep_hostname(yes); };
source local {
unix-dgram("/var/run/log");
udp(ip(0.0.0.0) port(514));
internal();
};
$
```
-RVP
SHELL=/usr/pkg/bin/dash and CONFIG_SHELL=/usr/pkg/bin/dash doesn't
seem to help at all, so the workaround is this:
pkgin -y in dash
cp /usr/pkg/bin/dash ~/bin/sh
env PATH=$HOME/bin:$PATH ./mach build
That causes dash(1) to be used always and everything seems to work without
a full rebuild.
HTH,
-RVP
t time doesn't trigger configure at all on my setup. Error log for
the 2nd run attached (owing probably to some missing dependency which was not
checked for; everything that `mach build' complained about, I installed from
binary packages).
If I do a `mach clobber', then confgure is, o
On Thu, 30 May 2024, e...@tilde.team wrote:
On Tue, May 28, 2024 at 08:51:00PM +, RVP wrote:
In your BIOS, can switch the drive controller to `AHCI' instead of whatever
it is now (probably IDE)?
-RVP
It is a SATA drive, and no opportunities to change anything regarding
the drive i
1 - 100 of 397 matches
Mail list logo
