haukem

I used the claude-code-action GitHub Action to review PRs before, but it is pretty buggy e.g. PRs from forked repositories do not work, and I had to fix it myself. This should work better with Claude Code Routines. claude-code-action only works with the API and is therefore pretty expensive compared to the subscription.

I think LLM reviews on PRs are helpful and will reduce the load on maintainers. I am working on OpenWrt and was approved for the Claude Code Max Open Source Program today. The cap of 15 automatic Claude Code Routines runs per day is a bit low. We get 5 to 20 new PRs per day and I would like to run it on all of them. I would also like to re-run it when authors make changes, in that case it should be sufficient to just check if the problems were addressed.

Is it possible to get more runs per day, or to carry over unused ones from the last 7 days? Maybe 30 on Sonnet and 15 on Opus?

When I was editing a routine, the window closed and showed an error message twice. Looks like there are still some bugs.

> First of all, there is no process yet for exactly requesting permission, secondly, the army already said they will not enforce the rule unless the Parliament declares combat readiness is necessary, and lastly, there is no punishment for not asking permission at this point in time.

Previously this article 3 was only active in the "Spannungs- oder Verteidigungsfall" which the Parliament has to declare. The law was extended with: "Außerhalb des Spannungs- oder Verteidigungsfalls gelten die §§ 3, 8a bis 20b, 25, 32 bis 35, 44 und 45." so this article is always active now.

It is very likely this was done intentionally. Maybe not all people involved in making this law noticed it, but the person working on article 2 did this intentionally. They explicitly list that this article is always active now:

> (3) Außerhalb des Spannungs- oder Verteidigungsfalls gelten die §§ 3, 8a bis 20b, 25, 32 bis 35, 44 und 45.

https://www.gesetze-im-internet.de/wehrpflg/__2.html

The article 3 of the Wehrpflichtgesetzes says this:

> (2) Männliche Personen haben nach Vollendung des 17. Lebensjahres eine Genehmigung des zuständigen Karrierecenters der Bundeswehr einzuholen, wenn sie die Bundesrepublik Deutschland länger als drei Monate verlassen wollen, ohne dass die Voraussetzungen des § 1 Absatz 2 bereits vorliegen. Das Gleiche gilt, wenn sie über einen genehmigten Zeitraum hinaus außerhalb der Bundesrepublik Deutschland verbleiben wollen oder einen nicht genehmigungspflichtigen Aufenthalt außerhalb der Bundesrepublik Deutschland über drei Monate ausdehnen wollen. Die Genehmigung ist für den Zeitraum zu erteilen, in dem die männliche Person für eine Einberufung zum Wehrdienst nicht heransteht. Über diesen Zeitraum hinaus ist sie zu erteilen, soweit die Versagung für die männliche Person eine besondere – im Bereitschafts-, Spannungs- oder Verteidigungsfall eine unzumutbare – Härte bedeuten würde; § 12 Absatz 6 ist entsprechend anzuwenden. Das Bundesministerium der Verteidigung kann Ausnahmen von der Genehmigungspflicht zulassen.

See: https://www.gesetze-im-internet.de/wehrpflg/__3.html

This was not changed.

The article 3 of the Wehrpflichtgesetzes was previously only active in a war or close to war situation (Spannungs- oder Verteidigungsfall). Article 2 said this before:

> § 2 Geltung der folgenden Vorschriften

> Die §§ 3 bis 53 gelten im Spannungs- oder Verteidigungsfall.

See: https://github.com/bundestag/gesetze/blob/master/w/wehrpflg/...

Now it says this:

> § 2 Anwendung dieses Gesetzes

> (1) Die nachfolgenden Vorschriften gelten nach Maßgabe der folgenden Absätze.

> (2) Die §§ 3 bis 52 gelten im Spannungs- oder Verteidigungsfall.

> (3) Außerhalb des Spannungs- oder Verteidigungsfalls gelten die §§ 3, 8a bis 20b, 25, 32 bis 35, 44 und 45.

> (4) Die §§ 15a und 16 sind nur auf Betroffene anzuwenden, die nach dem 31. Dezember 2007 geboren sind. Satz 1 gilt nicht im Spannungs- oder Verteidigungsfall.

See: https://www.gesetze-im-internet.de/wehrpflg/__2.html

This law changed it: https://www.recht.bund.de/bgbl/1/2025/370/VO

Is the a up to date git repository with all German law changes? The one I found was last updated 4 years ago.

I am also surprised that I haven't read about this in German news before. I am following the news. If Trump would have signed an executive order with a similar content affecting US citizen, German media would probably report about this multiple days long with many articles.

I was looking in Google news for other reports about this, but only found an article from Berliner Zeitung published 5 hours after this article from Frankfurter Rundschau.

I am worried about what other information which could be important to me, the news did not report on.

As far as I understood the law the article from FR is correct: https://www.gesetze-im-internet.de/wehrpflg/__3.html

OpenWrt accepts binary only firmware running outside of the Linux kernel address space on the wifi chip itself. This matches what upstream Linux also accepts. This works well with most recent Wifi drivers. OpenWrt does not accapt binary only kernel modules or binary only userspace applications, they are very hard to maintain if you do not have the source code.

This works well with Mediatek and also Qualcomm and most other vendors.

MediaTek chips are well supported by OpenWrt. Broadcom is not good supported. Mainline Linux kernel supports recent MediaTek Wifi chips quite well [1]. MediaTek is also working on these upstream Linux drivers, but they still have a proprietary Linux driver in addition.

Also the rest of the recent MediaTek SoC is supported quite well by upstream Linux and OpenWrt.

You can run OpenWrt on recent MediaTek SoCs with all code running on the main CPU being open source, no closed source code needed inside the Linux kernel address space or in user space. The chips need firmware running directly on the IP cores. It needs a firmware running on the wifi core itself, there are probably one or more CPUs inside the wifi cores doing real time stuff. The Ethernet PHYs also need a firmware which is running on the PHY.

[1]: https://elixir.bootlin.com/linux/v6.17-rc5/source/drivers/ne...

I assume that about 20% to 50% of the home routers, Access points and Wifi mesh devices sold world wide are based on OpenWrt. Often some old versions of OpenWrt with many vendor modifications, the UI is always custom.

I know that the main vendor SDKs from Qualcomm, Mediatek, and Maxlinear are based on OpenWrt. I think only Broadcom uses an own Linux distribution which is not based on OpenWrt in their main SDK. Linux has a market share of about 99% in this market, I haven't seen VxWorks in any recent home router or access point.

Sorry, we know that the EU and NA distribution channels are not so good. The OpenWrt project depends on Banana Pi here. There are some resellers on Amazon.

About 10k OpenWrt one units were sold as of today. The first 900 were sold in about 3 days.

Thank you for all the insights.

Nice to know that all supported Pixel phones are not only on the same kernel version, but actually are build fro the same source tree now.

Do you also contribute your fixes back to the upstream projects like the upstream Linux kernel, AOSP or Google?

Many of the security features you are using are already included in AOSP, why does Google not activate them by default? Do they have a different balancing of performance, stability and compatibility on the one side and security on the other? I understand that Google has a different view on privacy for business reasons.

> If we make our own device with an OEM using Snapdragon, we'd have access to most of the sources for their driver libraries/services and a lot of firmware. It's not open source but OEMs have access.

Is it normal that phone OEMs have access to most of the source code of the kernel drivers and user space libraries provided by the SoC vendor.

I worked in the home router business and there it was normal that a OEM gets most of the kernel drivers and user space code in source code, but it was often restricted what they are allowed to publish in source code. Many vendors even published much less than they had to according to GPL and allowed by the SoC vendor.

I have heard that in the smart TV industry OEMs sometimes are only allowed to write an app and have no kernel source code access for their product.

The Raspberry Pi Foundation does not do a good upstream support. It is not very bad, but also not good. They should start adding support for their new chips in the upstream projects before they get into the market, like Intel does it for example. They could add support for some new IP cores, without reveling when and how it will be used later. When the product comes out the only add small patches linking the code together like device tree files.

It would also be good if they would release all the closed source firmware files needed for their devices under a redistributable license in the linux-firmware repository. For some time it was not allowed to redistribute the binary wifi firmware needed for the Raspberry Pi devices. This is needed for Linux distributions to package the binaries.

I hope they do this only because of lack of resources and not intentionally to lock people into their Linux distribution.

This press release from 2020 says Open Telekom Cloud is from Software and Hardware from Huawei.

https://www.open-telekom-cloud.com/de/blog/vorteile/die-sich...

Do you have any source that they switched to AWS?

Open Telekom Cloud was at least in 2020 running fully on Huawei Software and hardware: https://www.open-telekom-cloud.com/de/blog/vorteile/die-sich...

Deutsche Telekom used Huaweis OpenStack implementation.

I haven't found any information that this changed, so I assume it is still running completely on Huawei. At least they made sure that they still get chips from the US despite sanctions. ;-)

The Open Telekom Cloud was at least in the beginning running on Huawei hardware and software.

Here is a press release from 2020: https://www.open-telekom-cloud.com/de/blog/vorteile/die-sich...

Title: Open Telekom Cloud – die sichere Cloud made in Europe

> Im Rahmen der Innovationspartnerschaft liefert Huawei mit dem Cloud-Betriebssystem Huawei OpenStack Distribution eine zentrale Softwarekomponente der Open Telekom Cloud.

English translation: Title: Open Telekom Cloud – the secure cloud made in Europe.

> As part of the innovation partnership, Huawei provides a central software component of the Open Telekom Cloud with its cloud operating system, Huawei OpenStack Distribution.

Deutsche Telekom hosted Microsoft Office 365 for some years in Germany as a German cloud offering.

I think this was the press release: https://www.telekom.com/de/medien/medieninformationen/detail...

This was a Microsoft 365 cloud hosted and operated by Deutsche Telekom in Germany. It was more expensive than the global version and had less features. It often took some years till new features were introduced.

They stopped this offering some years ago, I think they did not get as many customers as they expected, most of the German customers used the global version.

STACKIT is the Lidl cloud. Both companies are part of the Schwarz Gruppe.

Their offering is here: https://www.stackit.de/en/ You have to be a company to make business with them. You can not just sign up, you have to contact them first.

There was a discussion about STACKIT some years ago: https://news.ycombinator.com/item?id=30853778

I do not see anything about Gaia-X on their directly website, only when I search for it there are some older press releases.

Yes that is correct, but they are not much used outside of the service they are intended for.

The service checks if the name in your PGP key matches the name in your identity card and signs your PGP key if it matches. I think it also checks if you can receive mails on the mail address in your PGP key.

You can then use your PGP key to sign or encrypt emails or sign your git commits or other stuff you can do with PGP keys. Others who trust this signing service then known that this key really belongs to someone with your name.

Germany does not use a global unique identifier like the social security number in the US. The Germans do not like such global identifier for people because the the Nazis used them and data protection is very important in Germany.

This is not indented as a bullet prove government authentication system, if you need this use the eID card directly.

The goal is to have a CA for (existing) OpenPGP keys which checks if the name in it is matching the one from the identity card. When you sign a PGP key to tell that you trust it you should compare the name in the identity card or passport with the one from the key, this system does it automatically.

One tricky part is that many people like me leave out some names in the PGP key. In the first implementation of the PGP signing service it only ensured that at least one first name and one last name is also in the PGP key. I do not know if this is still the case.

The German Federal Office for Information Security (BSI) is supporting GPG4Win since many years, see for example here: https://www.golem.de/news/bsi-deutsche-behoerden-bekommen-gp...

Disclaimer: I worked for Governikus some years ago and worked on the initial version of this service.

Found it out myself, it is not needed by recent thermald. The ReadMe says this:

> Thermald version 2.0 and later has in built parser for thermal tables. So this utility is not required. Make sure that thermald "--adaptive" option is used.

https://github.com/intel/dptfxtract/blob/master/README.txt

Is dptfxtract still needed?

The thermald man page says this:

> In some newer platforms the auto creation of the config file is done by a companion tool "dptfxtract". This tool can be downloaded from "https://github.com/intel/dptfxtract". It is suggested as parts of the install process, run dptfxtract.

The dptfxtract gibthub project (https://github.com/intel/dptfxtract) says Intel discontinued the project.

Thank you for your comment.

I installed thermald on my Lenovo T480 with Debian Bookworm and I get 20% better results in stress-ng. The fans are a bit louder now under high load and off under low load.

Without thermald:

  $ stress-ng --matrix 0 -t 3m --metrics-brief
  stress-ng: info:  [3755113] setting to a 180 second (3 mins, 0.00 secs) run per stressor
  stress-ng: info:  [3755113] dispatching hogs: 8 matrix
  stress-ng: info:  [3755113] stressor       bogo ops real time  usr time  sys time   bogo ops/s     bogo ops/s
  stress-ng: info:  [3755113]                           (secs)    (secs)    (secs)   (real time) (usr+sys time)
  stress-ng: info:  [3755113] matrix          2278812    180.00   1437.43      0.27     12660.06        1585.04

With thermald:

  $ stress-ng --matrix 0 -t 3m --metrics-brief
  stress-ng: info:  [3755550] setting to a 180 second (3 mins, 0.00 secs) run per stressor
  stress-ng: info:  [3755550] dispatching hogs: 8 matrix
  stress-ng: info:  [3755550] stressor       bogo ops real time  usr time  sys time   bogo ops/s     bogo ops/s
  stress-ng: info:  [3755550]                           (secs)    (secs)    (secs)   (real time) (usr+sys time)
  stress-ng: info:  [3755550] matrix          2791272    180.00   1404.32      0.57     15507.06        1986.83

I just installed it using apt and did no extra configuration. My system was anyway configured for balanced power mode. Why is thermald not installed on desktop installations by default?

Is this really a problem with European companies or more a problem with older companies which didn't need much software 30 years ago?

Are Ford and General Motors better at this than the European car manufacturers?

Maybe it is also caused by people staying very long at their employer in Europe. It is not uncommon to stay till retirement after you got into a big cooperation.

Coal plants are emitting a lot of unhealthy stuff.

The coal plant Jänschwalde in Germany emitted 672kg mercury, 1780kg lead and 208 kg arsenic in 2017 according to the European Pollutant Release and Transfer Register. https://de.wikipedia.org/wiki/Kraftwerk_J%C3%A4nschwalde#Emi...

This is the worst coal plant in Germany, but the others are not so much better.

The WRT1200AC family is not well supported. The Ethernet part should work fine, but the Wifi is unsupported since some years now, see here the repository: https://github.com/kaloz/mwlwifi The vendors are not interested in this hardware any more, but they have very good marketing and sales. Linksys and Marvell also did not really support the OpenWrt community, they just had good marketing. If your WRT1200AC device does not work well with OpenWrt do not complain to OpenWrt, but complain to the Linksys support.

The WRT1200AC family for example does not support WPA3, because the closed source Wifi firmware does not support it. The 15 years old WRT54G supports WPA3, it is just very slow. ;-)

Currently I would suggest the Linksys E8450 / Belkin RT3200 (same hardware) or some other device using the current Mediatek platform with MT7622 + MT7915 + MT7531. (2 X Cortex-A53, Wifi 6) All chips are supported in recent upstream Linux kernel, including Wifi. The Mediatek router team is currently doing pretty good upstream open source work for their chips.

The $12B Fab which TSMC builds in Phoenix, will be outdated when it will open in 2024 and it is too small for AMD or Apple. It is likely it is mainly build to secure chip supply for the US military in case China invades Taiwan and the US does not have access to the TSMC Fabs in Taiwan any more. Apple and AMD will probably not be able to sell products for 1 year till they migrated to Intel or Samsung production in case TSMC fabs in Taiwan are not available for them any more.

The TSMC fab in Phoenix is build for 5nm and will open in 2024. The TSMC 3nm process should go into mass production in 2023, it is expected that the 2023 iPhone will use a 3nm SoC. The fab will only produce 20k Wafers per month, the TSMC gigafabs in Taiwan produce 100k Wafer per month each.

Details about TSMC Phoenix fab: https://www.cnbc.com/2021/10/16/tsmc-taiwanese-chipmaker-ram...

Writing documentation is boring, adding new features is more fun to do.

Someone working on open source to have fun is more likely to invest their time in more features instead of better documentation.

Here companies or users like you could step in and contribute better documentation or pay the original authors to improve the documentation.