Does this board boot Linux via a device tree, or have hardware discovery?
How about UEFI vs arm-specific bootloaders?
I tried arm32 Linux a few years back, and the largest hindrance at the time was the device trees and non-UEFI boot process. Given up on exploring the platform further (except maybe for SBC like raspberry pi) until that situation improves.
The CIX CP8180 uses UEFI (it is intended to boot Windows which requires it) but the boot flow can, I believe, use either ACPI or device trees, based on a boot setting. The ACPI boot flow has the advantage that any normal Linux distro should work, while the device tree variant I think has more hardware enablement.
The upstream story due to this is kind of a mixed bag, though. I think they also still use out-of-tree NPU drivers, etc. Device trees and other updates are still flowing upstream. I think the next Mesa release will support the Immortalis GPU series though, so that'll hopefully polish off a big remaining problem with ordinary distros.
will failover faster and more successfully on systemd-resolved, than if you specify all Cloudflare IPs together, then all Google IPs, etc.
Also note that Quad9 is default filtering on this IP while the other two or not, so you could get intermittent differences in resolution behavior. If this is a problem, don't mix filtered and unfiltered resolvers. You definitely shouldn't mix DNSSEC validatng and not DNSSEC validating resolvers if you care about that (all of the above are DNSSEC validating).
As the only developer maintaining a big bounty program. I believe they are all trending downward.
I've recently cut bounties to zero for all but the most severe issues, hoping to refocus the program on rewarding interesting findings instead of the low value reports.
So far it's done nothing to improve the situation, because nobody appears to read the rewards information before emailing. I think reading scope/rewards takes too much time per company for these low value reports.
I think that speaks volumes about how much time goes into the actual discoveries.
Open to suggestions to improve the signal to noise ratio from anyone whose made notable improvements to a bug bounty program.
Similarly from a hacker's point of view, I also think vulnerability reporting is in a downwards spiral. Particularly the ones organised through a platform like this just aren't reaching the right people. It used to be pgp email to whoever needs to know of it and that worked great. I have no idea if it still would today for you guys, but from my point of view it's the only reliable way to reach a human who cares about the product and not someone whose job it is to refuse bounties. I don't want bounties, I've got a day job as security consultant for that, I'm just reporting what I stumble across. Chocolate and handwritten notes are nice, but primarily I want developers and sysadmins to fix their damn software
Putting on my tinfoil hat, I wonder if some of that slop might be coming from actual black-hat groups or state actors - who have an interest in making it harder to find and close real exploits.
Those people wouldn't care about the bounty, overwhelming the system would be the point.
I think those people are busier overwhelming other, bigger systems right now, but it's a fair point. I daresay when you get down to a real salt-the-earth destroy-everything point, open source projects can expect destruction by the same people.
To say nothing of the uses of real exploits: that's salient.
An interesting tool for analyzing your personal kernel config file and pointing out areas for security improvement. It's more comprehensive than KSPP (https://kspp.github.io/) but sometimes goes a little too far, suggesting disabling kernel features you may actively use.
Even worse than this are sites that have RSS feeds and Cloudflare in front. Because my RSS reader doesn't look much different than a bot, and won't complete Cloudflare's CAPTCHA.
I was running into something similar when I was accessing the API of someone's MediaWiki instance and adding User-Agent was enough to solve the problem
If it's possible the power button is getting bumped repeatedly in your pocket. You could search your Android settings, and make sure "Emergency SOS" is turned off.
It’s also possible that there’s a hardware failure that is making the power button erroneously report button presses, leading to the Emergency SOS. This happened to me on a Pixel 3, which resulted in repeated calls to 911 with no user input.
* Can’t power it off for the night, because the flaky power button turns it back on.
* Can’t pull the SIM card, because emergency calls don’t require a SIM card to connect.
* Can’t consistently use the “slide to cancel” option, as the phone was also trying to initialize the camera at the same time. (IIRC,
3 button presses for the camera, 5 button presses for SOS. The flaky power button managed to trigger SOS while the camera was still initializing the GUI, so the camera GUI took focus.)
* Can’t access the settings, because the flaky power button either turns the phone off, opens the camera, or sends an SOS faster than I could search the settings.
This all started at about 10 PM. So, instead of going to sleep, I needed to spend the next two hours baby-sitting my phone as it mostly was repeatedly rebooting, with occasional calls to 911, until the battery finally died.
Anyone can get the performance crown by having an unlimited energy budget. Performance per watt is much more valuable in data centers (TCO) and consumer devices (battery life).
The Mac Pro does have a rack-mounted configuration for the non-desktop data centers case. (I have no idea whether people will actually use it that way, but it exists.)
I can see those being bought up for Datcenters and CI use. There have been companies hosting huge racks of Mac minis for ages to do CI for MacOS and iOS software.
Not always, if you are connected to a source of electricity and doing something limited by speed performance is critical.
For me, Desktop use is almost perfect on Apple due to battery life and perf but professional use is much better on Intel/AMD+NVidia. Also you could get much more perf for $ on such machines
Something with intention to make you money and directly related to powerful machine, i.e. writing code, editing videos, images, etc. As opposed to casual use where you plan simple games, browse internet, etc.
'plan simple games' so are you counting complex games as professional use in your example?
Let's say I'm a professional researcher for some oil company. My job will primarily consist of browsing the web and writing stuff up, does that make that job not fit into the 'professional' category? You're being paid to browse the web and just report on what you found....
I'm a professional that gets paid to do things and a MacBook Pro with M1 Max works perfectly for me. This whole 'professional' thing is absurd and full of holes.
I feel compelled to mention that this space has existed as a niche community for many years now.
I've personally been using a NoFan CR-80EH in my workstations for over 10 years. I think it's subjectively the most beautiful heatsink I've ever seen.
You do need to plan your build to accomodate such a cooler though.
- Open Air case to allow free movement of air in and out of your case
- 65W TDP CPU
While a lot of people feel like 65 watt TDP is limiting, there are some impressive chips you can use under that threshold that don't feel like a compromise. Eg the Ryzen 9 7900 (not-X).
And if the rest of your office is quiet, eliminating ambient background noise is a delightful improvement.
>this space has existed as a niche community for many years now
Yes, I also remember the passive cooling community has existed for a long time, with it being a very big thing when PCs were loud monstrosities, but I feel it has peaked sometime in the mid to late '00s and has been slowly loosing momentum since active cooling solutions became much quieter and better at ramping up/down with load, with most CPUs, GPUs, PSUs coolers nowadays even turning fully off when idling or under low load, along with the shift from PCs to mobile devices, making the need for passive-only PC cooling solutions a small and expensive niche that not many venture into.
Interestingly, the 7900 currently costs more than the 7900x (quick Amazon search). And my understanding is the 7900 is just the 7900x that has its TDP defaulted to 65 watts. So you could buy the cheaper 7900x and just go into BIOS and limit the TDP to make it a 7900.
I would bet that the 7900X chips are binned for higher quality Silicon though. Makes you wonder if an underpowered 7900X outperforms a 7900 and by how much.
I always wanted to do such a project and decided to go for it. I ended up hating it because it made me realize how much I hate coil whine. I mean: it looks so friggin cool, and placing it on your desk will probably give much better air circulation than under it. No fans spinning just made me realize how much noise most computers make that is hidden under fan noise.
I did have dedicated graphics card though, but a passively cooled one. None of my friends found it bothersome, so maybe I'm just sensitive.
Some more tips from a recent install of a Noctua NH-P1:
* If AMD, make sure you have the
backplate or contact Noctua support for
assistance
* The backplate will slide under the
space between the case and motherboard,
with just a slight bit of anguished
bending
* Had to move the top-mounted power
supply (PSU) out of the way completely.
* 2mm is about the size of a crayon tip,
when applying the thermal paste
* Had to move a (Scythe) fan out too, but
the provided clips let me put it on top
The stock CPU cooler on the old AM3 was a source of coil whine, but not anymore :)
A Thinkpad T30 idling nearby also helps mask further sounds.
This is a simple me too. I have had a Xeon (65W TDP) cooled with NoFan 24/7 for some 5 years. It began as my main developer machine and is now a proxmox server. It just works.
How about UEFI vs arm-specific bootloaders?
I tried arm32 Linux a few years back, and the largest hindrance at the time was the device trees and non-UEFI boot process. Given up on exploring the platform further (except maybe for SBC like raspberry pi) until that situation improves.
reply