- wlug - list.wlug.org.nz

Linux X86/x86_64 Will Now Always Reserve the First 1MB of RAM
by Peter Reutemann 10 Jun '21

10 Jun '21

'Phoronix reports: The Linux x86/x86_64 kernel code already had logic in place for reserving portions of the first 1MB of RAM to avoid the BIOS or kernel potentially clobbering that space among other reasons while now Linux 5.13 is doing away with that 'wankery' and will just unconditionally always reserve the first 1MB of RAM. The Linux kernel was already catering to Intel Sandy Bridge graphics accessing memory below the 1MB mark, the first 64K of memory are known to be corrupted by some BIOSes, and similar problems coming up in that low area of memory. But rather than dealing with all that logic and other possible niche cases besides the EGA/VGA frame-buffer and BIOS, the kernel is playing it safe and just always reserving the first 1MB of RAM so it will not get clobbered by the kernel.' -- source: https://linux.slashdot.org/story/21/06/09/2122210 Cheers, Peter -- Peter Reutemann Dept. of Computer Science University of Waikato, NZ +64 (7) 577-5304 http://www.cms.waikato.ac.nz/~fracpete/ http://www.data-mining.co.nz/

1 0

Space: (Not) The Final Frontier ...
by Lawrence D'Oliveiro 09 Jun '21

09 Jun '21

You may know that a cryptocurrency like Bitcoin gets its value from a concept known as “proof of work” -- namely that in order to “mine” (create new units of) the currency, you performed a computation with a commonly-accepted minimum complexity (with no known shortcuts). Now there is another kind of cryptocurrency, based on “proof of space”, where the (agreed) value comes from permanently setting aside some amount of storage space. Instead of consuming some increasing percentage of the world’s electricity (and GPU) supply, this one is going to consume some increasing percentage of the world’s hard drive supply. <https://www.theregister.com/2021/06/08/chia_cryptocurrency_creates_euro_hdd…>

1 0

KDE Plasma 5.22 Released
by Peter Reutemann 09 Jun '21

09 Jun '21

'KDE Plasma 5.22 is now available, bringing "hugely improved" Wayland support, better performance for gaming, adaptive panel transparency for the panel and widgets, and more. Phoronix reports: There is now support for variable rate refresh (VRR) / Adaptive-Sync on Wayland, vertical/horizontal maximization now working with KWin Wayland, global menu applet support under Wayland, support for activities, and a lot of other general improvements and fixes so the overall Wayland support is much more polished and nearly at par to the X.Org Server support. The performance for gaming with KDE Plasma on Wayland should also be better with now having direct scan-out support for full-screen windows. Rounding out the graphics fun with this release is also GPU hot-plugging support on Wayland for KWin, such as if using an external GPU or USB display adapter. KDE Plasma 5.22 also delivers on adaptive panel transparency for the panel and widgets, desktop notification improvements, Plasma System Monitor has replaced KSysGuard as the default system monitoring application, and a variety of other improvements. You can view the full changelog for Plasma 5.22 here.' -- source: https://tech.slashdot.org/story/21/06/08/1938202 Cheers, Peter -- Peter Reutemann Dept. of Computer Science University of Waikato, NZ +64 (7) 577-5304 http://www.cms.waikato.ac.nz/~fracpete/ http://www.data-mining.co.nz/

1 0

ESA’s Mars Express Orbiter Is 18 Years Old
by Lawrence D'Oliveiro 04 Jun '21

04 Jun '21

<https://www.theregister.com/2021/06/02/mars_express/> In some industries, you really have to take the long view about your computing infrastructure: Wood recently shared a Twitter thread that began, much to the delight of retro-computing fans, with the recovery of data from two IDE drives from a long-dead PC and (spoiler alert) ended with the 25-year-old emulator for the Star Tracker software firing up on a 486-based machine running Windows for Workgroups 3.11 (after more modern Microsoft operating systems rejected it.) ... Alternatively, hardware on Earth could give up the ghost. The mission control systems for Mars Express originally ran on Solaris 8 machines (and some tools ran on Solaris 6.) A migration to Solaris 10 occurred in 2013 although, with a hat-tip to the Unix team working around failing ancient hardware, Wood observed: "We have Solaris 10, running a Solaris 8 zone and that zone is configured in Solaris 6 compatibility mode. (it really is turtles all the way down!)" Virtualization efforts have played their part in reducing the reliance on hard-to-obtain components. Should ESA keep the funding tap turned on and extend the mission past the end of 2022, an upgrade to Solaris 11 servers is also on the cards (a joint activity with other ESA missions XMM and Integral) that should last until the end of Mars Express.

2 2

Google Releases Dependency-Mapping Tool
by Lawrence D'Oliveiro 04 Jun '21

04 Jun '21

Report <https://www.theregister.com/2021/06/04/google_open_source_insights/> on a new service from Google where it examines millions of online source-code project repositories, and maps out their dependencies, letting you see information like licensing and security vulnerabilities. Because a vulnerability in a package you depend on could become a vulnerability in your code.

1 0

How Facebook Architects Around Silent Data Corruption
by Lawrence D'Oliveiro 04 Jun '21

04 Jun '21

Seems like Facebook’s computation facilities have become so massive that it is starting to see intermittent errors in places you wouldn’t expect <https://www.nextplatform.com/2021/03/01/facebook-architects-around-silent-d…>: Engineers found that many of the cascading errors are the result of CPUs in production but not always due to the “soft errors” of radiation or synthetic fault injection. Rather, they find these can happen randomly on CPUs in repeatable ways. Although ECC is useful, this is focused on problems in SRAM but other elements are susceptible. The Facebook engineering team that reported on these problems finds that CPU silent data corruptions are actually orders of magnitude higher than soft-errors due to a lack of error correction in other blocks. The article mentions examples like a computation of 2×3 sometimes returning an answer other than 6, or a file size intermittently (and incorrectly) being computed as zero, leading to the file data not being copied. Such occurrences may be extremely rare, but when you are doing things on the scale that Facebook does, they start to become all too common. Facebook has published a paper (linked in the report) on their research into the problem. They have worked out some techniques to try to mitigate these issues, but of course they come at a cost.

1 1

The Collective Code Construction Contract
by Lawrence D'Oliveiro 04 Jun '21

04 Jun '21

Came across this interesting document <https://rfc.zeromq.org/spec/42/> referenced from another project I was browsing. It is a set of guidelines for how Free Software projects should be managed. Some choice extracts: The project SHALL use the git distributed revision control system. The project SHALL use a share-alike license such as the MPLv2, or a GPLv3 variant thereof (GPL, LGPL, AGPL). All contributions to the project source code (“patches”) SHALL use the same license as the project. All patches are owned by their authors. There SHALL NOT be any copyright assignment process. Some specifics of how to use, and not to use, Git: The project SHALL have one branch (“master”) that always holds the latest in-progress version and SHOULD always build. The project SHALL NOT use topic branches for any reason. Personal forks MAY use topic branches. To make a stable release a Maintainer shall tag the repository. Stable releases SHALL always be released from the repository master. Actually that last one seems a bit questionable. For example, Blender does not structure its repos that way. Each stable release has its own branch, which may continue to accumulate bug fixes for as long as it is under support, while main development of the next version continues on the master branch. Plus there guidelines for handling people-management issues, of course: Administrators SHOULD block or ban “bad actors” who cause stress and pain to others in the project. This should be done after public discussion, with a chance for all parties to speak. A bad actor is someone who repeatedly ignores the rules and culture of the project, who is needlessly argumentative or hostile, or who is offensive, and who is unable to self-correct their behavior when asked to do so by others.

1 0

Google is Making it Harder for Android Apps To Track You Once You've Opted Out
by Peter Reutemann 04 Jun '21

04 Jun '21

'It's going to get harder for Android apps to track users who've opted out of receiving personalized ads, the Financial Times reports, after Google announced changes to how it'll handle the unique device identifiers that allow marketers to track them between apps. From a report: Starting later this year, Google is cutting off access to these "Advertising IDs" after a user opts out, and will show developers a "string of zeros" in its place. The news was announced in an email to Play Store developers, and Google has also updated its support page for Advertising IDs with the announcement. Google told developers the changes will "provide users with more control over their data, and help bolster security and privacy," the Financial Times reports. The change comes a few short months after Apple overhauled how advertising IDs work on iOS in an apparent attempt to compete with the new policy. ' -- source: https://yro.slashdot.org/story/21/06/03/140255 Cheers, Peter -- Peter Reutemann Dept. of Computer Science University of Waikato, NZ +64 (7) 577-5304 http://www.cms.waikato.ac.nz/~fracpete/ http://www.data-mining.co.nz/

1 0

Microsoft’s Package Manager Quality Issue
by Lawrence D'Oliveiro 03 Jun '21

03 Jun '21

Microsoft has officially launched “Winget”, aka the Windows Package Manager <https://www.theregister.com/2021/06/02/microsoft_introduces_moderation_for_…>. Unfortunately, it’s run into a bit of a snag: anybody can submit packages to be included in the repo willy-nilly, leading to problems like duplicates, bad links and other misbehaviour. “What is wrong with this picture?” as they say: Look at any of the common Linux distros, and you will see that every package has an official maintainer who will take responsibility for its quality, handling of bug reports, updating from upstream and so on. Nobody can simply submit a package to be included and then forget about it. There is a proper QA system.

1 0

No More CLA For GCC
by Lawrence D'Oliveiro 03 Jun '21

03 Jun '21

The Free Software Foundation has long been one of those outfits that has refused to accept (significant) source contributions without the author signing a “Contributor Licence Agreement”. It’s not enough to release the code under the same licence as the project, you must also hand over the entire copyright to them. This is supposedly because they have received legal advice that it makes it easier to defend against copyright infringements if they own all the code. Quite a few other projects don’t agree <https://lwn.net/Articles/592503/>. For example, the Linux kernel project does not demand CLAs, only a “Certificate of Origin” (affirming that you have the right to contribute the code). And now the GNU Compiler Collection is joining them. <https://www.theregister.com/2021/06/02/gcc_fsf_code_copyright/>

2 2