Can someone provide some context to this? This reasoning seems crazy to me. I don't understand how endianness fits into this picture. I don't do low-level stuff, but I thought things like that...
Can someone provide some context to this? This reasoning seems crazy to me. I don't understand how endianness fits into this picture. I don't do low-level stuff, but I thought things like that only really matter when it comes to the assembly level or lower. And most of the Linux kernel is written in C.
I'm also a bit curious about where MIPS still has market share. The last time I checked, it was still hanging on in the embedded market, but I haven't seen it anywhere in the consumer market since the PSP and a handful of niche portable consoles.
The Debian mailing list thread the blog links to gives a more in-depth explanation: The blog post author zeroes in on the endianness argument, which I think is a bit of a distraction. It's not...
The Debian mailing list thread the blog links to gives a more in-depth explanation:
Aurelien Jarno recently proposed the mips architecture (supporting 32-bit big-endian MIPS CPUs) for removal and then got it removed. This removal affects bullseye and sid but not buster or stretch. Please prepare to migrate your MIPS hardware to mipsel or mips64el, much recent MIPS hardware (such as Octeon CPUs) supports endian switching at runtime and can therefore be supported by the other MIPS ports.
The removal was due to the limited 2GB virtual address space and because the architecture is one of the last big-endian architecture Debian supports, the porting effort became increasingly difficult. On the other hand the level of interest for this architecture is going down, and with it the human resources available for porting is going down.
The blog post author zeroes in on the endianness argument, which I think is a bit of a distraction.
It's not even all MIPS CPUs, only old ones that were 32-bit. Newer MIPS chips continue to be supported. And the old ones are still supported by the current Debian Stable (buster) and the previous stable (stretch). It's only the next release of Debian (several years away) that won't support them.
I was panicking when I saw the tags but then I remembered my device is actually mipsel so it looks like I'm all good. I wonder how many devices are actually mips these days.
I was panicking when I saw the tags but then I remembered my device is actually mipsel so it looks like I'm all good. I wonder how many devices are actually mips these days.
Can someone provide some context to this? This reasoning seems crazy to me. I don't understand how endianness fits into this picture. I don't do low-level stuff, but I thought things like that only really matter when it comes to the assembly level or lower. And most of the Linux kernel is written in C.
I'm also a bit curious about where MIPS still has market share. The last time I checked, it was still hanging on in the embedded market, but I haven't seen it anywhere in the consumer market since the PSP and a handful of niche portable consoles.
The Debian mailing list thread the blog links to gives a more in-depth explanation:
The blog post author zeroes in on the endianness argument, which I think is a bit of a distraction.
It's not even all MIPS CPUs, only old ones that were 32-bit. Newer MIPS chips continue to be supported. And the old ones are still supported by the current Debian Stable (buster) and the previous stable (stretch). It's only the next release of Debian (several years away) that won't support them.
https://www.mips.com/about/
I was panicking when I saw the tags but then I remembered my device is actually mipsel so it looks like I'm all good. I wonder how many devices are actually mips these days.