Linux 5.6 Ships With Broken Intel WiFi Driver After Network Security Fixes Go Awry

[grigi]

I note that in my own OSS project, the first major release is often followed by 2 or 3 bugfix releases. And most of them are not regressions, but extra features allow people to abose the code better, so unnoticed bugs pop out of the woodwork...
I agree to wait a few point releases before using anything new. There is a few exceptions to that, usually when I need the new feature (e.g. was running git mesa for a while because there was fixes that I needed)

[Volta]

https://www.windowslatest.com/2018/10/12/fix-no-audio-output-device-is-installed-error-in-windows-10/

https://www.forbes.com/sites/gordonkelly/2019/09/15/microsoft-windows-10-update-problem-wifi-internet-nic-warning-upgrade-windows-10/

https://www.forbes.com/sites/gordonkelly/2019/09/03/microsoft-windows-10-update-problem-cpu-performance-spike-warning-upgrade-windows-10/#4590348c362f

https://www.windowscentral.com/how-regain-internet-access-after-installing-update-windows-10

https://www.guidingtech.com/no-sound-issue-windows-10-update/

https://zcom.tech/windows-10-mail-app-not-working.html/

https://answers.microsoft.com/en-us/windows/forum/windows_10-update/windows-10-broken-after-update/361a0623-ae1f-4d0d-8ad6-83622ed5c8c7

Eat that.

[skeevy420]

"IWLWIFI" is not a dyslexic-friendly acronym. I spent the first half-minute trying to read that.

[birdie]

It's funny Mr Volta quotes isolated Windows issues (looks like it's just five for the past five years - and most of them are related to ... 3d party drivers), when Open Source bugzillas are rife with hundreds if not thousands of bug reports which prevent users from using their HW properly. And don't get me started on bugs in software whose number is in tens of thousands. He's so obsessed with finding relatively benign and easily fixable issues in Windows, criticial hard to fix Linux issues are the least of his concerns - for instance in this thread people mentioned broken Intel WiFi drivers you needed to patch manually to make 'em work. Wow, what a great OS, Mr. Volta. Instead of downloading a newer version of drivers from a vendor in Windows, you have to learn: shell, git, patch, make, grub2 and then pray you've done everything correctly and haven't accidentally erased your entire filesystem in the process.

So, again, how many issues have you found and help resolve in the Linux kernel so far? None? Well, please kindly fuck off.

"IWLWIFI" is not a dyslexic-friendly acronym. I spent the first half-minute trying to read that.

LMAO. Intel Wireless Fidelity Local Area Network Wireless Fidelity Driver.

[Volta]

It's funny Mr Volta quotes isolated Windows issues (looks like it's just five for the past five years - and most of them are related to ... 3d party drivers), when Open Source bugzillas are rife with hundreds if not thousands of bug reports which prevent users from using their HW properly. And don't get me started on bugs in software whose number is in tens of thousands. He's so obsessed with finding relatively benign and easily fixable issues in Windows, criticial hard to fix Linux issues are the least of his concerns - for instance in this thread people mentioned broken Intel WiFi drivers you needed to patch manually to make 'em work. Wow, what a great OS, Mr. Volta. Instead of downloading a newer version of drivers from a vendor in Windows, you have to learn: shell, git, patch, make, grub2 and then pray you've done everything correctly and haven't accidentally erased your entire filesystem in the process.

No, it's just five I pasted here Mr Dumb. Windows updates breaking drivers are Windows problems. There's nothing more simple than reverting broken commit in Linux. Try this on Windows. Oh, you can't. Windows critical problems are either unfixable or takes months to fix. Like the famous desktop icons security flaw. Furthermore, last time I checked there's no Windows kernel available for download, so you're comparing apples to oranges (or windows to doors). Ps. this broken Intel driver Wifi found in newly released kernel which isn't used in ANY stable distribution yet, isn't isolated issue? Oh, man, Birdie and his brain again. All of the issues with Linux (exactly one serious in rc1 or rc2 kernel related to EFI) I had were reported and fixed, so kindly shut up. Ps. 2 I didn't report and I won't report a single Windows issue and there are dozens of them. What I had to do to install this mess was to erase entire drive and use cmd tricks to make this shit install on my PC. So kindly shut up and get the hell out with this broken mess.

[FireBurn]

I reported this ath10k issue last week and it hasn't even been assigned yet...

https://bugzilla.kernel.org/show_bug.cgi?id=206923

[birdie]

Mr. Volta is either high or just stoned - I don't now which. I mean you must be insane to say that "nothing more simple than reverting broken commit in Linux". No, Mr. Volta. Downloading and installing new drivers by double clicking an exe file is a simple enough task for 99.99% of people out there, while reverting a broken commit requires quite a lot of knowledge and fine understanding of what you're doing. Oh, and please enlighten me how you're going to achieve that on a PC where secure UEFI boot is enabled. Are you also suggesting the user must learn how to generate and install a custom certificate for their firmware and then learn to sign the kernel and its modules to make it all work? You're clearly out of your mind.

Again, each time he opens his mouth he sounds like a rabid Linux fanboy who believes everyone must become a programmer and enjoy Linux for what it is - a platform for hobbyists who enjoy having sex with their PCs more than having sex with their spouses. I'm glad I have him blocked and ignored. Brings a peace of mind.

In short, I will no longer use a private browser tab to read your inanities. There's nothing to argue with you. I've seen rational Linux defenders - you're clearly not one of them. You're spewing complete nonsense over and over again.

[skeevy420]

Downloading and installing new drivers by double clicking an exe file is a simple enough task for 99.99% of people out there, while reverting a broken commit requires quite a lot of knowledge and fine understanding of what you're doing. Oh, and please enlighten me how you're going to achieve that on a PC where secure UEFI boot is enabled. Are you also suggesting the user must learn how to generate and install a custom certificate for their firmware and then learn to sign the kernel and its modules to make it all work?

And if they want to do it in a way that doesn't corrupt their system and works long term then they have to learn how to create and maintain their own packages and custom repositories. None of any of what either of us mention is what I'd consider to be easy for a decent majority of people regardless of the OS. Once you figure it out it isn't that difficult, but the learning curve can be hellish.

When double-click and apt get fail, a good portion of that decent majority is up that creek using hand-paddles because anything more than saying WTF and clicking "submit bug report" is beyond their skills and abilities -- and it likely always will be -- for example: how many of you have used a car for 20+ years yet cannot rebuild a motor and can really only do anything beyond spark plugs, fluids/air pressure, and maybe bolt on stuff like alternators and water pumps? Those things are like virus scanning and running updates, inserting USB drives, and "bolting on" a GPU or drive...there's a screw or two involved .

[cynical]

Intel is too busy code-blocking amd drivers to handle its own stuff properly....

Wow, what's the story on that?

[Volta]

Mr. Volta is either high or just stoned - I don't now which. I mean you must be insane to say that "nothing more simple than reverting broken commit in Linux". No, Mr. Volta. Downloading and installing new drivers by double clicking an exe file is a simple enough task for 99.99% of people out there

No, if they end with unbootable windoze, but you're brainless.

In short, I will no longer use a private browser tab to read your inanities. There's nothing to argue with you. I've seen rational Linux defenders - you're clearly not one of them. You're spewing complete nonsense over and over again.

Just to sum up Braindie:

windows fanboy and liar (yeah, windoze 'doesn't spy!)

https://www.phoronix.com/forums/forum/software/desktop-linux/1165668-chrome-80-against-firefox-74-75-performance-on-linux?p=1165721#post1165721

except it does!

some kind of mental illness perhaps?

https://www.phoronix.com/forums/foru...59#post1168059