|
I'm using a Framework laptop, and when I select my USB key from the boot menu, I just see a quick flash of an underline cursor in the top left of the screen, and then I'm back to the boot menu. I've tried two different USB keys, and building the keys via both dd and Rufus with different options. Putting a Fedora image on the same USB key boots fine. I'm using https://mirror.csclub.uwaterloo.ca/archlinux/iso/2022.09.03/archlinux-2022.09.03-x86_64.iso and it passed the gpg verification. Anyone have any ideas?
|
#
¿
Sep 18, 2022 22:47
|
|
|
|
| # ¿ May 4, 2024 00:20 |
|
|
Computer viking posted:Is secure boot enabled? I think Fedora signs their bootloaders, but that's kind of uncommon outside the largest, most industry-backed, distros. Yes, that was it! Thank you!
|
|
#
¿
Sep 18, 2022 23:08
|
|
|
kujeger posted:Generally /mnt is for disposable mounts (i.e. mounting knowing you'll unmount quite soon). My recent Linux (Fedora) created directories named something like /mnt/subjunctive/USB_KEY_NAME for removable media, at the least.
|
|
#
¿
Sep 20, 2022 00:21
|
|
|
The only morally correct way to use /net is as an automount populated via NIS+.
|
|
#
¿
Sep 21, 2022 17:43
|
|
|
I’m enjoying my Steam Deck. What are the odds that I can have a similarly compatible and smooth experience with a future Zen 4/Lovelace setup, running Arch underneath?
|
|
#
¿
Oct 1, 2022 13:51
|
|
|
other people posted:nvidia themselves have an rpm repo for fedora/rhel and others and it works quite well Am I going to hate myself if I do this with Arch?
|
|
#
¿
Oct 1, 2022 20:32
|
|
|
Oh thank you very much for that tip. I was starting to fret at how far I still had to go with the Arch install on the laptop, so I’ll try EOS.
|
|
#
¿
Oct 1, 2022 22:52
|
|
|
Ugh. I installed Endeavour and it was doing the Linux gaming thing pretty well, but it locked up after exiting a game and now when I reboot something goes wrong after I log in. My suspicion is that it involves the NVIDIA stuff I installed, but I only just got everything manually mounted on the live installer and found the Xorg logs before I had to leave for an appointment. When I add “rescue” on its own line in the grub boot parameters and then boot, it complains that it can’t find “rescue” as a command. Was I supposed to set that up somehow? I can’t use systemctl from within the chroot, it seems, so I guess it’s manual symlink janitoring to enable the debug console on tty9. (Also, every key press in grub is doubled for some reason, in addition to being super slow, which is goddamned infuriating.)
|
|
#
¿
Oct 6, 2022 22:55
|
|
|
When I was at Mozilla we talked about offering a flatpak like thing, so that it was harder to use system properties to do fingerprinting, but the tech wasn’t really there at the time (especially on Windows) and the ergonomics would have been rough with helper apps and so forth. Still, it’s an idea!
|
|
#
¿
Oct 6, 2022 23:42
|
|
|
Yeah, I tried Ctrl-Alt-Fn and Alt-Fn all the way down the line, just a black screen. I don’t know if the ttys are in configured, or if it’s something loving up during the mode switch. I did get it all mounted, learning about btrfs subvolumes in the process, and I’ll inspect it more tomorrow.
|
|
#
¿
Oct 7, 2022 02:28
|
|
|
v1ld posted:They're configured by default in EndeavourOS so your display was well and truly hosed, to use Klyith's more technical language. Yeah, weird. I can’t get console terminals to show up in the EndeavourOS live installer either. I tried uninstalling nvidia and my boot hung while still in text mode, after reaching multi-user.target or whatever. Not an outcome I had expected—figured X would just fail to startup or fall back to something else. Nothing in the Xorg logs I could find was helpful. I installed another copy of Endeavour on a new partition and I’m going to go step by step to see when it breaks. Wish bash kept longer history by default so I could see everything I did before…
|
|
#
¿
Oct 8, 2022 00:00
|
|
|
RFC2324 posted:Not as far as I can tell, its not a problem of more than one thing accessing the database since sqlite isnt a shared database. Litestream purports to have it figured out, but I haven’t used their system in anger. https://litestream.io/ Edit: oh, they just do read replicas and you’re talking about multi-master I now realize! Still a pretty neat little tool though.
|
|
#
¿
Oct 10, 2022 20:07
|
|
|
I wanted to try Wayland because I’ve heard it’s better with multi-monitor, so I installed plasma-wayland-session and picked it to login and I got a black screen that just sat there and didn’t even let me move the mouse. I’ve read various wiki pages and forum posts but I can’t even figure out where to look for logs for Wayland. GNOME-wayland will display the (empty/default) desktop and the keyboard works, but no mouse there either. I’m on EndeavourOS with the NVIDIA 515 drivers that it came with out of the box. Mouse is a Razer Naga 2014 (I think) that appears to show up properly in Xorg logs and works fine under X11. Anyone have ideas on what to try? After all this is resolved I’m sure I’ll discover that Wayland is no better than X for my situation, but it’ll mean more if I learn that myself…
|
|
#
¿
Oct 10, 2022 20:14
|
|
|
Klyith posted:nvidia-drm.modeset=1 kernel parameter? Yep! Checked at boot to make sure it made it all the way to grub, and it shows up in that /sys entry. KozmoNaut posted:Wayland on Nvidia is a broken mess. Well that doesn’t sound promising. The display part seemed to be working ok with GNOME, just had that mouse problem. v1ld posted:Besides, X perf is the same as Wayland with Gnome if you run in fullscreen, not borderless window, Is this the case for multi-monitor too? I thought the X compositor was always in play with a multi-monitor setup, but I could be mistaken. I can just stick with X for now, though.
|
|
#
¿
Oct 10, 2022 20:56
|
|
|
It is surprising to me that single-writer SQLite corrupts under docker’s NFS setup. Are its locking primitives broken?
|
|
#
¿
Oct 11, 2022 15:23
|
|
|
Yeah, the site also warns that NFS implementations sometimes have lovely locking, which I guess could do it with multiple writers. I worked on a distributed filesystem with POSIX-compliant locking and mmap semantics, and it was hard to get it right in all the worst racy and client-failing cases, but still, NFS implementations have had enough time to figure it out.
|
|
#
¿
Oct 11, 2022 15:37
|
|
|
In a world that didn’t care about parallelizing, probably!
|
|
#
¿
Oct 15, 2022 15:43
|
|
|
Ventoy and just live your life with persistence layers from live systems.
|
|
#
¿
Oct 30, 2022 18:01
|
|
|
poo poo, I had no idea fgrep was ACID compliant.
|
|
#
¿
Nov 10, 2022 04:48
|
|
|
I get doubled keystrokes in grub, which makes picking the right OS be sort of a little game of towers of Hanoi, and I cannot figure out why. It’s also laggy, probably related.
|
|
#
¿
Nov 22, 2022 14:14
|
|
|
ExcessBLarg! posted:Radix sort? Feels indeed like it should have “radix” in it somewhere except that the file name usually doesn’t omit the prefix specified in the directory names. Hmm. It’s not actually a sort, though.
|
|
#
¿
Dec 2, 2022 15:43
|
|
|
Kibner posted:It will take time to find out exactly which parts to get, but AMD does offer ECC memory support on their consumer chips, if that is a thing that you are interested in. Are there any consumer motherboards that report ECC faults?
|
|
#
¿
Dec 11, 2022 17:52
|
|
|
Kibner posted:For AM4 I know that the Asus x570 ACE WS does. I haven't looked into AM5 mobos, though. That’s a serious fuckin’ motherboard, nice.
|
|
#
¿
Dec 11, 2022 18:09
|
|
|
If you quit your browser and then xdg-open opens it, and one of the tabs is a Zoom link (which of course is the case) then it’ll trigger the Zoom client opening.
|
|
#
¿
Dec 16, 2022 01:50
|
|
|
I thought the whole point of flatpaks is that they are bound as a unit with their dependencies, rather than pulling in libraries from the host environment. It sounds like they made an exception for GL drivers but AIUI the right way to solve this is to rebuild the flatpak itself against LLVM 15. I haven’t done much of it, but because of that isolation flatpaks tend to be easy to build, if a bit slow. (I wonder how many more exceptions they’ll make, until flatpaks are just awkward RPMs again.)
|
|
#
¿
Dec 28, 2022 15:24
|
|
|
It’s no more idiotic than static linking, which is to say taking a specific point in the trade off spectrum between “control my dependencies” and “integrate into the system”.
|
|
#
¿
Dec 29, 2022 01:10
|
|
|
Tesseraction posted:Sorry I have to admit that I blanked on Manjaro because I didn't recognise it - my bad. Does this provide the functionality you're looking for? https://gitlab.com/corectrl/corectrl corectrl is nice, but it’s about controlling the CPU and GPU parameters like clocks and power levels, and not system fan profiles. (It does say it can control the fans on AMD GPUs, but that’s all afaict.)
|
|
#
¿
Dec 29, 2022 04:32
|
|
|
BlankSystemDaemon posted:Performance-sensitive workloads are going to benefit from having 64bit pointers simply because it lets the process cache more data in memory, and anything that isn't performance sensitive isn't going to benefit from fp arithmetic being done using SSE registers for reasons that seem pointless to belabour. For a long time Chrome and Firefox both resisted going to 64-bit applications because browsers have a lot of pointers in them and doubling the size of them means caches are 20-50% less effective, especially d$. Browsers care a bit about about overall RAM usage (no, really) but they really care about cache utilization, and “just switch to x86-64” was a measurable performance hit in a lot of cases. One of the V8 folks has a blog post about compressing pointers down to 32 bits in some hot structures and it made a material difference in tight-loop workloads, but of course I can’t find it now. Additional integer ops were cheaper than the cache costs, as I recall the punchline. Chrome’s multiprocessing model means a per-process 4G limit isn’t a big a deal, also, and a lot of applications aren’t really limited by their capacity for RAM-cacheable data as much as they are by being able to chew data fast enough through the d$ and CPU.
|
|
#
¿
Jan 1, 2023 22:07
|
|
|
Twerk from Home posted:https://v8.dev/blog/oilpan-pointer-compression - Pointer compression for their DOM GC. This was the one, thank you! Some smart motherfuckers on that team. Pointer compression in a 64-bit-pointer ISA is probably a better approach for large working set apps than x32 though because pointer compression is enough engineering work that you don’t want to have to do it everywhere in order to address more than 4GB. Bring back the segment register!
|
|
#
¿
Jan 1, 2023 23:29
|
|
|
BlankSystemDaemon posted:This seems incredibly counter-intuitive when you contrast it with the notion of how detailed knowledge of the sub-syscall systems can improve performance. What do those have to do with each other? Varnish is basically a syscall generation machine trading in I/O operations as fast as it can. Browsers really are not, and they do a lot more userspace work per system call than a lightweight proxy. (Which is to Varnish's credit, to be clear. If it were doing as much work per syscall as Chrome does, it would be terribly slow.) I love PHK's work, but I don't see how what he says there conflicts with Chrome and Firefox's empirical findings. He says, in the article I skimmed because you didn't say what about it was relevant: quote:We lack a similar inquiry into algorithm selection in the face of the anisotropic memory access delay caused by virtual memory, CPU caches, write buffers, and other facts of modern hardware. and I'm talking exactly about analysis of CPU caches and their tradeoffs against additional integer work. AFAIK he's saying to do the work that the v8 team describes in oilpan and elsewhere: design data structures for the properties of modern hardware. Is there some specific conclusion from the article that you meant me to take away related to the effects of pointer width on cache and memory efficiency? quote:Java is garbage collected though, and while garbage collection has surely gotten better since 2005, when it was established that garbage collection is always a bad idea.  What about Java's garbage collector do you think invalidates the savings of pointer compression? If those object references were somehow statically deallocated because of magic[*], how would it not be beneficial to compress the pointers and reduce both cache and virtual memory/TLB pressure? (Reference counting, a form of non-Appel garbage collection used by Varnish, can also cause unfortunate application pauses due to finalization cascade, wherein dropping the last reference to an object causes it to drop the last reference to another object, which in turn etc., meaning that a bunch of otherwise untouched memory is read back and often mutated, causing a bunch of cache thrashing or even page-ins.) [*] and you'd require magic, because the lifecycle of objects isn't always knowable statically; it can depend on properties of the program's input. This is why garbage collection in the form of reference counting is so common in non-trivial programs written in languages without built-in garbage collection, be that reference counting like Python and Swift and Rust where the ubiquitous std::rc::Rc is employed, or mark-and-sweep like JS and Ruby and Java, or hybrid systems with refcounting backed up by infrequent marking and sweeping for cycle collection.
|
|
#
¿
Jan 2, 2023 02:13
|
|
|
Klyith posted:why would you after that for a project that needs constant work and probably got little in the way of donations or money Every parent in this thread is all 
|
|
#
¿
Jan 4, 2023 02:29
|
|
|
I think they were trying to delete the contents but leave the directory, but deleting the directory and then recreating it would be a fine approach.
|
|
#
¿
Jan 9, 2023 05:30
|
|
|
The kernel will balance across swap devices itself, fwiw.
|
|
#
¿
Jan 10, 2023 16:15
|
|
|
If you want more metrics and queries to monitor, you could look at https://osquery.io/ (which I love and miss at my current job) with an exporter like https://github.com/zwopir/osquery_exporter
|
|
#
¿
Jan 12, 2023 20:45
|
|
|
xzzy posted:That is insane overkill for my needs but it's a neat idea. Other than the fact I am really tired of writing SQL. Yeah, that makes sense. I don’t have it running at home right now either.
|
|
#
¿
Jan 13, 2023 18:43
|
|
|
Serious_Cyclone posted:Any opinions? Anything to watch out for, or recommendations on hardware? I have shallow but not non-existent hardware knowledge. I run EndeavourOS on my last-gen Framework laptop, and it’s pretty good. Power management wasn’t great in terms of idle draw, but I think that’s largely fixed now. Linux support is meaningful to Framework, and their staff engage with people on support issues.
|
|
#
¿
Jan 19, 2023 18:05
|
|
|
BlankSystemDaemon posted:They're both virtual machines that execute bytecode in order to match and process packets, so I find it hard to believe that the netfilter folks weren't at least inspired - which is fine, it's about the best design for a firewall that anyone has come up with. Doesn’t netfilter go back to the 90s, built by some Australian guy for…SunOS? Or is this a name collision?
|
|
#
¿
Jan 29, 2023 23:03
|
|
|
BlankSystemDaemon posted:Darren Reeds ipf(ilter), which is probably what you're thinking of, Yes, that’s it! I remember him on the Great Circle firewalls mailing list, back when you could read the day’s messages during a short lunch break.
|
|
#
¿
Jan 30, 2023 03:58
|
|
|
Keito posted:Manjaro doesn't have the best reputation. I've heard nothing but good things about EndeavourOS though, so maybe check that out if you want something Arch based. Hobbyist distros are pretty fun, it depends what you're after. Use case? I chose EndeavourOS when I was looking for an arch-based distro (my first!) to be similar to SteamOS for gaming. Other than some challenges getting my 4090 to work on release week it’s been pretty great so far. (It took a little while for the packages to make it out of beta, and I was reinstalling so I needed the installer image rebuilt to have the beta drivers—which someone on the forums did for me within 24 hours of my post!)
|
|
#
¿
Jan 31, 2023 00:58
|
|
|
|
| # ¿ May 4, 2024 00:20 |
|
|
Rocko Bonaparte posted:I have a kernel wizard question. I just wrote a BPF to hook on some calls I want to use for extracting some general observability messages. So I am listening for function entry into these functions, extracting the string parameter, and ferrying up into user space where I can munge on it and store it externally or whatever. I see this is relying on kprobes under the hood. I am curious why I wouldn't be using kprobes directly? I don't really know anything about them. It sounds like using a kprobe directly means I'm writing a kernel module, which just defers ferrying the information to user space anyways. Are there other reasons I would or wouldn't just use kprobes directly? This is likely a production thing for us at least internally so if eBPF is meant for one-offs, maybe using it for this is a bad idea. I can’t speak to kprobes, but eBPF is the basis of a lot of production instrumentation of big systems. It’s a fine choice, IMO.
|
|
#
¿
Feb 6, 2023 20:28
|
|