porting OpenBSD to Ox64
Hi, If anyone is interested in helping or just plain interested, here is my prep work documented. I've been on it sparingly since beginning of March. I don't know how much time I want to invest in this but we'll see... https://github.com/pbug44/openbsd-src/tree/Ox64 The Ox64 is a 8 dollar SoC utilizing a RISCV64 CPU (among other cores). I intend to use this for a Freifunk-like project which I call GardenNet. https://sky.delphinusdns.org/eap-tls-idea.txt (following link in german use chromium to translate or something): https://wiki.freifunk-franken.de/w/Benutzer:PeterPhilipp#Ein_Garten_Netz_Knoten Best Regards, -pjp
Re: porting OpenBSD to Ox64
Hi. I am interested in this topic, as i have one in my drawer. My programming skills probably not up to the task, but I would be more than happy to help you with testing, etc. Regards, --ext Peter J. Philipp írta 2024. márc.. 21, Cs-n 08:50 órakor: > Hi, > > If anyone is interested in helping or just plain interested, here is my > prep work documented. I've been on it sparingly since beginning of March. > I don't know how much time I want to invest in this but we'll see... > > https://github.com/pbug44/openbsd-src/tree/Ox64 > > The Ox64 is a 8 dollar SoC utilizing a RISCV64 CPU (among other cores). I > intend to use this for a Freifunk-like project which I call GardenNet. > > https://sky.delphinusdns.org/eap-tls-idea.txt > (following link in german use chromium to translate or something): > https://wiki.freifunk-franken.de/w/Benutzer:PeterPhilipp#Ein_Garten_Netz_Knoten > > Best Regards, > -pjp -- --Z--
Firefox hangs after sleep
At some point (I can't put my finger on exactly when this started), an existing firefox instance hangs after my laptop is put to sleep and then wakes from sleep. Websites in existing tabs still work, but searching for another webpage in an existing tab or opening a new tab is when the hang occurs. Killing and restarting firefox eliminates the problem until the next sleep. Interestingly, "simple" websites such as the OpenBSD or Arch Linux websites still can be started after sleeping, but "fancy" ones such as Gmail or the Washington Post cannot. This suggests to me the problem lies in javascript or some other non-pure html component. I am using 7.4 stable on a Lenovo X1 Carbon, gen 4, though it also occurs on an X1 gen 9. A recent update to firefox (and all the other installed packages) didn't help. This doesn't happen with other browsers such as chromium or iridium. I am not sure what kind of diagnostics would help here. David J. Raymond david.raym...@nmt.edu http://kestrel.nmt.edu/~raymond
Re: Firefox hangs after sleep
On Thu, Mar 21, 2024 at 02:29:14AM -0600, Raymond, David wrote: > At some point (I can't put my finger on exactly when this started), an > existing firefox instance hangs after my laptop is put to sleep and then > wakes from sleep. Websites in existing tabs still work, but searching for > another webpage in an existing tab or opening a new tab is when the hang > occurs. Killing and restarting firefox eliminates the problem until the > next sleep. Alternatively, wait for a long time and the problem will resolve itself, i.e. the stuck tabs will suddenly start working again. I believe it's a known issue, but nobody has invested time digging into it. It's probably DNS...
Re: porting OpenBSD to Ox64
On 3/21/24 09:10, Mizsei Zoltán wrote: Hi. I am interested in this topic, as i have one in my drawer. My programming skills probably not up to the task, but I would be more than happy to help you with testing, etc. Regards, --ext Excellent! Yes I could use this sort of help. In particular if you want to fiddle with Apache NUTTX and see if you can utilize the wifi. It seems to me that I can use a little bit of RAM for this. My idea is this: If it is at all possible, boot on cpu 0 (OpenBSD) and then alloc some contiguous RAM (10 MB perhaps?) from the PSRAM. Once that is done we need to fork a thread or process from the kernel and start the bootprocess with it for the c905(?) 32-bit core which has direct access to the wifi device. Then we need some interprocess communication between the 802.11 stack on OpenBSD and the NUTTX wifi driver. Both CPU's will run in a hybrid/asynchronous fashion (as far as I understand it if either doesn't touch the RAM of the other it will be ok locking wise). If anyone wants to chime in here, if this is an insane idea let me know. I understand that a async mode is possible afaik. So we need the NUTTX as a firmware (perhaps 2 MB in size or so), it needs programming to communicate with the c906 64-bit core, we can work that out somehow. If you want to build a toolbox for this entire thing where we can just convert it to a firmware. What do you think does this make sense, are you up for it? It really needs little programming, perhaps a make file or a script to build NUTTX, I have linux devuan here (on native hardware and vmm) and this is what I could use. https://nuttx.apache.org/ and here is the Reference manual for the BL808: https://mainrechner.de/BL808_RM_en_1.3.pdf So as a first step we need to figure out if NUTTX actually has drivers for this SoC and that they work. If not, we'll have to consider another approach. Best Regards, -pjp -- *** I used to sign with -peter, but noticed it's not unique, -pjp may come up in the future, so please adjust for that ***
Re: porting OpenBSD to Ox64
Hi, I can definetely spend some time with NUTTX, as I already wanted to do that for a different reason. However I fear my USB UART adapters won't work with 200 baud transfer rate, so I have ordered an RpiPico, but it will take a while to receive it, till that day I won't be able to help you :( --ext Peter J. Philipp írta 2024. márc.. 21, Cs-n 09:54 órakor: > On 3/21/24 09:10, Mizsei Zoltán wrote: >> Hi. >> >> I am interested in this topic, as i have one in my drawer. My programming >> skills probably not up to the task, but I would be more than happy to help >> you with testing, etc. >> >> Regards, >> >> --ext > > Excellent! > > Yes I could use this sort of help. In particular if you want to fiddle > with Apache NUTTX and see if you can utilize the wifi. It seems to > > me that I can use a little bit of RAM for this. My idea is this: > > If it is at all possible, boot on cpu 0 (OpenBSD) and then alloc some > contiguous RAM (10 MB perhaps?) from the PSRAM. Once that is > > done we need to fork a thread or process from the kernel and start the > bootprocess with it for the c905(?) 32-bit core which has direct > > access to the wifi device. Then we need some interprocess communication > between the 802.11 stack on OpenBSD and the NUTTX > > wifi driver. Both CPU's will run in a hybrid/asynchronous fashion (as > far as I understand it if either doesn't touch the RAM of the other > > it will be ok locking wise). If anyone wants to chime in here, if this > is an insane idea let me know. I understand that a async mode is > > possible afaik. So we need the NUTTX as a firmware (perhaps 2 MB in > size or so), it needs programming to communicate with the > > c906 64-bit core, we can work that out somehow. If you want to build a > toolbox for this entire thing where we can just convert it to a > > firmware. What do you think does this make sense, are you up for it? > It really needs little programming, perhaps a make file or a > > script to build NUTTX, I have linux devuan here (on native hardware and > vmm) and this is what I could use. > > https://nuttx.apache.org/ and here is the Reference manual for the > BL808: https://mainrechner.de/BL808_RM_en_1.3.pdf > > So as a first step we need to figure out if NUTTX actually has drivers > for this SoC and that they work. If not, we'll have to consider > > another approach. > > Best Regards, > > -pjp > > > -- > *** I used to sign with -peter, but noticed it's not unique, -pjp may > come up in the future, so please adjust for that *** -- --Z--
aucat options parsing
This seems strange: $ aucat -n -d -i input.wav -c -r 8000 -o out.wav input.wav: skipped unknown chunk input.wav: play, chan 0:3, 48000Hz, s16le, bytes 80..1920080, vol 8388608 -r: channel range expected It is an ommited number in -c 1 of course, not a missing sample rate. Jan
Re: aucat options parsing
On Mar 21 10:07:08, h...@stare.cz wrote: > This seems strange: > > $ aucat -n -d -i input.wav -c -r 8000 -o out.wav > input.wav: skipped unknown chunk > input.wav: play, chan 0:3, 48000Hz, s16le, bytes 80..1920080, vol 8388608 > -r: channel range expected > > It is an ommited number in -c 1 of course, > not a missing sample rate. Sorry, my bad; it is a missing channel range of course. Without using the new -m, it seems that one _must_ use the old -c syntax of -c lo:hi; in particular, -c 1 -o out.wav is illegal, even though it is a valid (new) syntax. If that is the case, shouild the manpage mention that? Jan
Re: wireguard reconfiguration reliability
On Wed, Mar 20, 2024 at 03:43:52PM -0700, Paul B. Henson wrote: > On Wed, Mar 20, 2024 at 09:56:06PM +0100, Kirill Miazine wrote: > > > Like in this thread, I guess: > > > > https://marc.info/?t=16964239631&r=1&w=2 > > Yes, that is likely the issue we're hitting. Seems last message is from > 10/2023 and the issue wasn't resolved :(, so I guess it's a known > problem with no solution on the horizon. > wg(4) diff was committed to -current. Does the problem exist in upcoming 7.5? > Next time I'll try your workaround of batching the commands up (ifconfig > wg1 down; ifconfig wg1 delete; ifconfig wg1 destroy) rather than running > one at a time and keep my fingers crossed I win the race condition :). > > Thanks for the help... >
sysupgrade doesn't work unless monitor is attached
Nick Holland reported this with a HP T430 Thin Client already in May 2022, and I see the same problem on two of my new firewalls. I was hoping a HDMI dummy plug would work as a workaround, but it doesn't. I'm not sure when or what marks the bsd.upgrade file as -x, but that at least that happens. Is there anything I can do to try to help? As long as a monitor is attached, sysupgrade works just as intended. -- chs
Re: sysupgrade doesn't work unless monitor is attached
On 2024-03-21 10:33 +01, Christer Solskogen wrote: > Nick Holland reported this with a HP T430 Thin Client already in May > 2022, and I see the same problem on two of my new firewalls. I was > hoping a HDMI dummy plug would work as a workaround, but it doesn't. > I'm not sure when or what marks the bsd.upgrade file as -x, but that > at least that happens. that's the bootloader (to prevent upgrade loops, the bootloader will only consider bsd.upgrade that is +x). > Is there anything I can do to try to help? > > As long as a monitor is attached, sysupgrade works just as intended. > > -- > chs > -- In my defence, I have been left unsupervised.
Re: porting OpenBSD to Ox64
Hi, I'm also interested. I might be able to provide testing. On 21.03.24 08:50, Peter J. Philipp wrote: Hi, If anyone is interested in helping or just plain interested, here is my prep work documented. I've been on it sparingly since beginning of March. I don't know how much time I want to invest in this but we'll see... https://github.com/pbug44/openbsd-src/tree/Ox64 The Ox64 is a 8 dollar SoC utilizing a RISCV64 CPU (among other cores). I intend to use this for a Freifunk-like project which I call GardenNet. https://sky.delphinusdns.org/eap-tls-idea.txt (following link in german use chromium to translate or something): https://wiki.freifunk-franken.de/w/Benutzer:PeterPhilipp#Ein_Garten_Netz_Knoten Best Regards, -pjp
Re: aucat options parsing
On Thu, Mar 21, 2024 at 10:10:03AM +0100, Jan Stary wrote: > On Mar 21 10:07:08, h...@stare.cz wrote: > > This seems strange: > > > > $ aucat -n -d -i input.wav -c -r 8000 -o out.wav > > input.wav: skipped unknown chunk > > input.wav: play, chan 0:3, 48000Hz, s16le, bytes 80..1920080, vol 8388608 > > -r: channel range expected > > > > It is an ommited number in -c 1 of course, > > not a missing sample rate. > > Sorry, my bad; it is a missing channel range of course. > Without using the new -m, it seems that one _must_ > use the old -c syntax of -c lo:hi; in particular, > > -c 1 -o out.wav > > is illegal, even though it is a valid (new) syntax. > If that is the case, shouild the manpage mention that? > '-c 1' is the right syntax for all cases. The old syntax (-c lo:hi) still works only to not break existing scripts until they are updated. It is not documented anymore.
Re: porting OpenBSD to Ox64
On 3/21/24 12:27, Benjamin Stürz wrote: Hi, I'm also interested. I might be able to provide testing. OK great! I'm going to help you all a little by providing what I have so far. These flash images were built on a devuan Linux (like debian), on a vmm running on OpenBSD. They seem to work for me rudamentally, thus far. https://mainrechner.de/images/ (it's still uploading as I write this mail) There is a SHA256 file and a SHA256.sig file along with a oceans11-openbsd.pub signify key. So that you can quickly confirm the checksum. Ultimately it would be cool to build this all on native OpenBSD, if someone wants to attempt that, you're more than welcome to! https://wiki.pine64.org/wiki/Ox64 that link is the instructions on how to do this, including flashing (which is also mentioned on my README.md on the first mentioned URL at github). And as mentioned before here is some datasheets and other documents that I collected over the last year or so: https://mainrechner.de/riscv.html Best Regards, -pjp On 21.03.24 08:50, Peter J. Philipp wrote: Hi, If anyone is interested in helping or just plain interested, here is my prep work documented. I've been on it sparingly since beginning of March. I don't know how much time I want to invest in this but we'll see... https://github.com/pbug44/openbsd-src/tree/Ox64 The Ox64 is a 8 dollar SoC utilizing a RISCV64 CPU (among other cores). I intend to use this for a Freifunk-like project which I call GardenNet. https://sky.delphinusdns.org/eap-tls-idea.txt (following link in german use chromium to translate or something): https://wiki.freifunk-franken.de/w/Benutzer:PeterPhilipp#Ein_Garten_Netz_Knoten Best Regards, -pjp -- *** I used to sign with -peter, but noticed it's not unique, -pjp may come up in the future, so please adjust for that ***
Re: sysupgrade doesn't work unless monitor is attached
On Thu, Mar 21, 2024 at 11:14 AM Florian Obser wrote: > > On 2024-03-21 10:33 +01, Christer Solskogen > wrote: > > Nick Holland reported this with a HP T430 Thin Client already in May > > 2022, and I see the same problem on two of my new firewalls. I was > > hoping a HDMI dummy plug would work as a workaround, but it doesn't. > > I'm not sure when or what marks the bsd.upgrade file as -x, but that > > at least that happens. > > that's the bootloader (to prevent upgrade loops, the bootloader will > only consider bsd.upgrade that is +x). > So the bootloader sets -x after the first try? If that's the case, the bootloader is not to blame. -- chs
Re: DMARC/DKIM and OpenBSD Mailinglists
Is it not ARC meant to be the solution for this problem? Would DMARC then consider the original DKIM and SPF tests? Todd C. Miller schrieb am Mi., 13. März 2024, 14:56: > I've just added support to our majordomo for rewriting the From: > header when the sender's domain has a DMARC policy. Messages from > domains using DMARC will now have a From: header like: > > From: "John Connor via misc" > > and the original From: address is preserved in the X-Original-From: > header if one is not already present. > > This seems like the only reliable way to address the problem given > that the mailing list server often reformats or otherwise modifies > the message body. > > The rewriting currently happens even for a DMARC policy of "none" > since some large senders (for example gmail.com) use a policy of > "none" but receivers may still enforce SPF. I could relax this but > I worry that doing so will lower the IP or domain "reputation" in > Google's eyes (something that is already a black box). > > - todd > >
Re: wireguard reconfiguration reliability
On 24/03/20 08:15AM, Kirill Miazine wrote:
> #!/bin/sh
> ifconfig wg1 | \
> grep wgaip | \
> awk '{print $2} ' | \
> grep /32$ | \
> sed 's/\/32//' | \
> sort | while read x; do
> ping -w 1 -c 1 $x 2>&1
> done
Just FYI, you don't need backslashes (\) here, as the command ending with a
pipe is an incomplete pipeline. ;-)
Re: wireguard reconfiguration reliability
• Страхиња Радић [2024-03-21 16:31]:
On 24/03/20 08:15AM, Kirill Miazine wrote:
#!/bin/sh
ifconfig wg1 | \
grep wgaip | \
awk '{print $2} ' | \
grep /32$ | \
sed 's/\/32//' | \
sort | while read x; do
ping -w 1 -c 1 $x 2>&1
done
Just FYI, you don't need backslashes (\) here, as the command ending with a
pipe is an incomplete pipeline. ;-)
ah, thanks a lot, good to know!
in this case I added escaped newlines when posting, i have all on one
line, but I think i could have scripts where a pipe is followed by
escaped newline. have to check that one.
i see some scripts in /etc also use escaped newlines after pipes:
root@stable ~ # grep -r '| \\$' /etc/
/etc/daily: baksize=`disklabel $bakdisk 2>/dev/null | \
/etc/daily: rootsize=`disklabel $rootdisk 2>/dev/null | \
/etc/weekly:echo "${UPDATEDB} --fcodes=-" | \
Re: porting OpenBSD to Ox64
Hi, have you tried to boot a vanilla-current OBSD? Do you know if that "old 7.4-current" version you have mentioned contains any not-yet-upstreamed patches? Have you made any changes to the DTB or U-Boot? Regards, --ext Peter J. Philipp írta 2024. márc.. 21, Cs-n 08:50 órakor: > Hi, > > If anyone is interested in helping or just plain interested, here is my > prep work documented. I've been on it sparingly since beginning of March. > I don't know how much time I want to invest in this but we'll see... > > https://github.com/pbug44/openbsd-src/tree/Ox64 > > The Ox64 is a 8 dollar SoC utilizing a RISCV64 CPU (among other cores). I > intend to use this for a Freifunk-like project which I call GardenNet. > > https://sky.delphinusdns.org/eap-tls-idea.txt > (following link in german use chromium to translate or something): > https://wiki.freifunk-franken.de/w/Benutzer:PeterPhilipp#Ein_Garten_Netz_Knoten > > Best Regards, > -pjp -- --Z--
Re: wireguard reconfiguration reliability
On 24/03/21 04:58PM, Kirill Miazine wrote: > in this case I added escaped newlines when posting, i have all on one line, > but I think i could have scripts where a pipe is followed by escaped > newline. have to check that one. BTW, this also applies to lists[1], for example: ! false && echo "Not false" || echo "False" Sorry for the offtopic. :-) [1]: https://pubs.opengroup.org/onlinepubs/9699919799.2018edition/utilities/V3_chap02.html#tag_18_09_03
Re: unbound signature expired
On Mon, Mar 18, 2024 at 08:04:38PM +0100, Evan Sherwood wrote: > > Wild guess, your time is off. > > Huh, I think you're right. `date` shows me 7 hours ahead of my timezone. > > I restarted ntpd and I see no errors in /var/log/daemon, but the time is > still off. I should be 1200 PDT but it's showing me as 1900 PDT (not > UTC). > > What do I do to fix this? Pretty sure I had set my timezone to > America/Los_Angeles when I installed OpenBSD. > ntpd will not jump the clock backwards. You need to do a manual set. -Otto
Re: wireguard reconfiguration reliability
On Thu, Mar 21, 2024 at 12:23:06PM +0300, Vitaliy Makkoveev wrote: > wg(4) diff was committed to -current. Does the problem exist in upcoming > 7.5? Oh, I didn't know a fix had been committed, the referenced thread didn't mention a final one. Thanks, I'll take a look.
Solution to keystroke injection
Hello, Just launching you an interesting question. I noticed that software like Caja or Gedit suppress the text editing features or have a readonly mode. Do you think that giving a default readonly *splash* or mode to all our software when unattended could be the good one to defit keystroke injection kind of attacks one time for all? Thnx! -Dan
Re: rpc
how to replace nfs ? Em qua., 20 de mar. de 2024 às 19:27, Alexis escreveu: > Gustavo Rios writes: > > >> NFS/NIS/AMD are very old technology and are not robust. > > > > How to replace NIS ? > > Perhaps an LDAP implementation, e.g. OpenLDAP? > https://www.openldap.org/ > > > Alexis. > -- The lion and the tiger may be more powerful, but the wolves do not perform in the circus
Re: DMARC/DKIM and OpenBSD Mailinglists
Moin, > Is it not ARC meant to be the solution for > this problem? Yeah, technically, ARC _should_ help with this. However, in practice, trusting ARC is not really that common. > Would DMARC then consider the original > DKIM and SPF tests? Kind of; DMARC would trust the signed ARC headers that evaluated the original SPF/DKIM situation upon first receipt. With best regards, Tobias
Re: porting OpenBSD to Ox64
On 3/21/24 17:43, Mizsei Zoltán wrote: Hi, have you tried to boot a vanilla-current OBSD? Do you know if that "old 7.4-current" version you have mentioned contains any not-yet-upstreamed patches? Have you made any changes to the DTB or U-Boot? Regards, --ext Already replied privately, but for the public record: https://mainrechner.de/dot.config.txt This is the .config I used with the u-boot. No I didn't try a newer OBSD, I will soon though. :-) And no I didn't change anything in the DTB. Best Regards, -pjp Peter J. Philipp írta 2024. márc.. 21, Cs-n 08:50 órakor: Hi, If anyone is interested in helping or just plain interested, here is my prep work documented. I've been on it sparingly since beginning of March. I don't know how much time I want to invest in this but we'll see... https://github.com/pbug44/openbsd-src/tree/Ox64 The Ox64 is a 8 dollar SoC utilizing a RISCV64 CPU (among other cores). I intend to use this for a Freifunk-like project which I call GardenNet. https://sky.delphinusdns.org/eap-tls-idea.txt (following link in german use chromium to translate or something): https://wiki.freifunk-franken.de/w/Benutzer:PeterPhilipp#Ein_Garten_Netz_Knoten Best Regards, -pjp -- *** I used to sign with -peter, but noticed it's not unique, -pjp may come up in the future, so please adjust for that ***
