It’s very new. Previously the system would just drop to a console with a message saying “Kernel panic: not syncing: [reason]” and a whole bunch of debug info.
But still, on a well-maintained system, that pretty much never happens. Mainly because Linux is significantly more resilient to faults in device drivers than Windows.
Linux is monolithic so it breaks when a kernel module fails. It can sometimes recover but sometimes the system is in such a bad state a panic is triggered to protect against further issues.
Running Linux on computers with Nvidia hardware proves that Linux and Windows both have their problems dealing with device drivers. Linux’ benefit is that is has higher standards because the kernel devs need to sign off on driver, but that has downsides of turning away potential driver developers (as getting your code into Linux is a quite a complex thing just on its own). Linux also doesn’t have many drivers in general it seems, unless your device has some kind of generic fallback that disables any special features.
My kernel panics generally don’t display anything, the display just freezes and I need to force reboot the computer.
Begs the question what’s the point in all of this? In 20 or so years of using Linux (usually maintaining multiple systems at once) I’ve had a kernel panic maybe about 4 times for different reasons, and on those occasions the console debug info was fine. I don’t really understand the excitement around making error messages look more like Windows. It can’t be around being more newbie friendly since if you’re having kernel panics you probably need to be an expert or have expert advice anyway.
even on a less well-maintained system it’s probably not going to be the kernel having a freakout, the kernel is going to be just fine while something else shits itself (probably graphics drivers on a desktop tbh, my vega 10 loves to vomit onto the screen and pass out)
It’s very new. Previously the system would just drop to a console with a message saying “Kernel panic: not syncing: [reason]” and a whole bunch of debug info.
But still, on a well-maintained system, that pretty much never happens. Mainly because Linux is significantly more resilient to faults in device drivers than Windows.
Linux is monolithic so it breaks when a kernel module fails. It can sometimes recover but sometimes the system is in such a bad state a panic is triggered to protect against further issues.
Running Linux on computers with Nvidia hardware proves that Linux and Windows both have their problems dealing with device drivers. Linux’ benefit is that is has higher standards because the kernel devs need to sign off on driver, but that has downsides of turning away potential driver developers (as getting your code into Linux is a quite a complex thing just on its own). Linux also doesn’t have many drivers in general it seems, unless your device has some kind of generic fallback that disables any special features.
My kernel panics generally don’t display anything, the display just freezes and I need to force reboot the computer.
Begs the question what’s the point in all of this? In 20 or so years of using Linux (usually maintaining multiple systems at once) I’ve had a kernel panic maybe about 4 times for different reasons, and on those occasions the console debug info was fine. I don’t really understand the excitement around making error messages look more like Windows. It can’t be around being more newbie friendly since if you’re having kernel panics you probably need to be an expert or have expert advice anyway.
funy pengin
I guess it will make developers who develop the kernel and its components go “hehe fat penguin anyway let’s continue debugging this mess”
even on a less well-maintained system it’s probably not going to be the kernel having a freakout, the kernel is going to be just fine while something else shits itself (probably graphics drivers on a desktop tbh, my vega 10 loves to vomit onto the screen and pass out)