If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Announcement
Collapse
No announcement yet.
Linux 5.6 Ships With Broken Intel WiFi Driver After Network Security Fixes Go Awry
It's so nice to browse the forums with all the clinical blabbering idiots hidden from view. Time and again we see major regressions in the kernel but they will provide all sorts of crappy arguments why it's OK. It's OK, fantards, exactly the same way as Linux' 1% never changing desktop share which will never increase by preinstalling this broken OS on more PCs. People will just promptly delete it from their devices as soon as they face yet another issue which takes too much time to resolve. People have better things to do than debug their OS (I guess that also implies bisecting or even finding the appropriate bug reports which is far from easy).
Talking of the iwlwifi driver... I'm now using it on an XPS15 laptop (Killer Wireless 1650) with Ubuntu19.10 (kernel 5.3.0). I can not complain, I have no trouble connecting to the wifi. Except that the logs are overflowing with crashes and restarts of the iwlwifi driver and/or card (and every time it crashes, it apparently tries to restart the hardware and the driver as fast as it can, sending the cpu into thermal throttling). Am I the only one, or is it that this driver is incredibly unstable? Is it kind of getting better with newer kernels (beside the problem mentioned in this article)?
I rarely compile my own kernel these days, but after a few failed patches, I ended up just copying the whole iwlwifi folder out of the 5.4 tree to get a working 5.5 build.
Mine's been working well without problems, except for that one time where they declared support for newer firmware but then forgot to push the code that actually supports the newer firmware.
... and, more often than not, it's people like me who find regressions, identify the bad commit(s) and then test the fix which then finds itself in "stable" kernels.
Huh. I think I suffered from that- it was the craziest thing- my HP Spectre would sometimes just hang on boot, especially when warm. I thought it might have been related to thermal support, but never got around to bisecting which commit(s) broke it.
... but like you, I think I've E-mailed the maintainers of several subsystems some ~20 times over the last year as I've bisected the source of regressions- I consider it my small contribution to OSS, and the price of running a bleeding-edge kernel (it helps that it's not too far removed from my work as a for-hire kernel dev). At least each one got fixed, which makes it all worthwhile.
And just before you call me a Microsoft shill, I've filed more bug reports than any random 100 Phoronix readers combined and I've been using Linux exclusively since the late 90s.
11 bug reports in 10 years? Yeah, huge contribution. My god you're stupid.
I'm so effing tired of hearing this crap over and over again - "use LTS if you wanna play safe". This doesn't work! It's people like me who use mainline who make it possible for LTS distros to have semi-decent semi-stable kernels in the first place. People like me who find critical bugs, report them, bisect them and help fix them.
Jesus, give me a break. You've linked to the same thing multiple times, and you've "reported and bisected" exactly ONE issue in the last two years, only 2 in the last 4 and a half years, and only 11 EVER, in like a decade. LTS doesn't exist because of you, you're not even a drop of a drop in a bucket. I've helped and am helping more features get added and bugs get fixed in the past 6 months than you have in 10 years, and you don't see me howling about how critical I am, because I'm not. And neither are you. Either stop doing it, or stop acting like it grants you ANY sort of ability to lord it over people. Jesus christ.
Leave a comment: