Linux 5.16 Drops Support For MIPS Netlogic SoCs

[phoronix]

Phoronix: Linux 5.16 Drops Support For MIPS Netlogic SoCs

The Linux 5.16 kernel is doing away with hardware support for the MIPS-based Netlogic Microsystems SoCs, the network processors developed prior to being acquired by Broadcom a decade ago...

https://www.phoronix.com/scan.php?page=news_item&px=Linux-5.16-Drops-MIPS-Netlogic

[rene]

Good. I hope for the day all the bloat of non-x86 architectures is deleted from the Linux kernel and the developer can focus on optimizing the heck out fo the only true architecture like Linus created it for in 1991! Unbelievable how many millions of lines maintenance burden all those obsolete architectures cause :-/ Imagine the extra performance we could finally get back!

[Old Grouch]

Good. I hope for the day all the bloat of non-x86 architectures are deleted from the Linux kernel and the developer can focus on optimizing the heck out fo the only true architecture like Linus created it for in 1991! Unbelievable how many millions of lines maintenance burden all those obsolete architectures cause :-/ Imagine the extra performance we could finally get back!

I declare a Poe.

[jeoshua]

I declare a Poe.

Not necessarily. I've recently been giving a lot of thought to going through the configs of my custom kernel and making a legacy free compile. No floppies, tapes, pcmcias, ISA cards. No Solaris file systems. Nothing from before the 8086 chip was created. No Motorola 68k support.

You know, sanity.

[rene]

Not necessarily. I've recently been giving a lot of thought to going through the configs of my custom kernel and making a legacy free compile. No floppies, tapes, pcmcias, ISA cards. No Solaris file systems. Nothing from before the 8086 chip was created. No Motorola 68k support.

You know, sanity.

While I'd like to see how you compiled 68k code into your non-68k kernel, here is a very legacy free unmatched RISCV64 board running Desktop Linux for a really sane desktop experience! https://www.youtube.com/watch?v=Sv4-_a_3BKg ;-)

[F.Ultra]

Not necessarily. I've recently been giving a lot of thought to going through the configs of my custom kernel and making a legacy free compile. No floppies, tapes, pcmcias, ISA cards. No Solaris file systems. Nothing from before the 8086 chip was created. No Motorola 68k support.

You know, sanity.

And you gained not a single bit of performance, so yes either rene is severely confused or trolling.

[jeoshua]

While I'd like to see how you compiled 68k code into your non-68k kernel, here is a very legacy free unmatched RISCV64 board running Desktop Linux for a really sane desktop experience! https://www.youtube.com/watch?v=Sv4-_a_3BKg ;-)

I don't have that kind of code running on my computer. Nor a floppy drive. Nor a tape reader. Or any other of a thousand other things found in the kernel. And yet, there it sits in my configs, taking up disk space and/or compiler time. That's my point here.

[F.Ultra]

While I'd like to see how you compiled 68k code into your non-68k kernel, here is a very legacy free unmatched RISCV64 board running Desktop Linux for a really sane desktop experience! https://www.youtube.com/watch?v=Sv4-_a_3BKg ;-)

He didn't, he just disabled the config option and thought that it somehow made something happen.

[jeoshua]

And you gained not a single bit of performance, so yes either rene is severely confused or trolling.

Actually, when I removed support for different things the kernel compiled and booted quicker. So... Yeah I'm not sure you understand what's being said here. Rene may be being a bit hyperbolic, but the amount of time and energy spent on supporting ancient code isn't negligible.

[F.Ultra]

Actually, when I removed support for different things the kernel compiled and booted quicker. So... Yeah I'm not sure you understand what's being said here. Rene may be being a bit hyperbolic, but the amount of time and energy spent on supporting ancient code isn't negligible.

So in other words your kernel was not more optimized for x86 than it was before, that was the context not that a smaller kernel might boot a ms quicker on a slow drive. And zero (or very close to zero) of the x86 devs spend any time on the MIPS arch.

And by "disabling" Motorola 68k you didn't do any change what so ever to the kernel, compilewise or when booting.