▲ 109 ▼ Linux kernel 7.0 is coming mid-April—here’s everything you need to know (www.howtogeek.com) submitted 7 months ago by Sunshine@piefed.zip to c/linux@programming.dev 14 comments fedilink hide all child comments
[–] MatSeFi@lemmy.liebeleu.de 19 points 7 months ago (4 children) YSK also this permalink fedilink source hideshow 4 child comments replies: [–] Link@rentadrunk.org 14 points 7 months ago (3 children) Most servers are most likely on LTS distros and won’t be affected by this bug. Interestingly it seems the first rule of kernel development, we don’t break user space should apply here. permalink fedilink source parent hideshow 3 child comments replies: [–] definitemaybe@lemmy.ca 1 point 7 months ago (1 child) I'm not a kernel developer, but it does seem weird to remove the legacy function in the same patch as the replacement function is shipped. That said, isn't it fairly easy to install a modified kernel? Or is that hard on Ubuntu for some reason? I suppose it's to do with official support cycles and patches, so that's only really a viable fix if the distro manages to kernel options, to ensure timelines with security patches and commercial support. permalink fedilink source parent hideshow 1 child comment replies: [–] xtools@programming.dev 1 point 7 months ago yeah i'm running a lts distro but keep the kernel up to date, it's super simple, there's even gui apps for it permalink fedilink source parent [–] bitfucker@programming.dev 1 point 7 months ago That's the thing. It doesn't break it to the point of unusable unlike other patches I think. So the test didn't catch this since performance is rarely a part of a test suite no? permalink fedilink source parent
[–] Link@rentadrunk.org 14 points 7 months ago (3 children) Most servers are most likely on LTS distros and won’t be affected by this bug. Interestingly it seems the first rule of kernel development, we don’t break user space should apply here. permalink fedilink source parent hideshow 3 child comments replies: [–] definitemaybe@lemmy.ca 1 point 7 months ago (1 child) I'm not a kernel developer, but it does seem weird to remove the legacy function in the same patch as the replacement function is shipped. That said, isn't it fairly easy to install a modified kernel? Or is that hard on Ubuntu for some reason? I suppose it's to do with official support cycles and patches, so that's only really a viable fix if the distro manages to kernel options, to ensure timelines with security patches and commercial support. permalink fedilink source parent hideshow 1 child comment replies: [–] xtools@programming.dev 1 point 7 months ago yeah i'm running a lts distro but keep the kernel up to date, it's super simple, there's even gui apps for it permalink fedilink source parent [–] bitfucker@programming.dev 1 point 7 months ago That's the thing. It doesn't break it to the point of unusable unlike other patches I think. So the test didn't catch this since performance is rarely a part of a test suite no? permalink fedilink source parent
[–] definitemaybe@lemmy.ca 1 point 7 months ago (1 child) I'm not a kernel developer, but it does seem weird to remove the legacy function in the same patch as the replacement function is shipped. That said, isn't it fairly easy to install a modified kernel? Or is that hard on Ubuntu for some reason? I suppose it's to do with official support cycles and patches, so that's only really a viable fix if the distro manages to kernel options, to ensure timelines with security patches and commercial support. permalink fedilink source parent hideshow 1 child comment replies: [–] xtools@programming.dev 1 point 7 months ago yeah i'm running a lts distro but keep the kernel up to date, it's super simple, there's even gui apps for it permalink fedilink source parent
[–] xtools@programming.dev 1 point 7 months ago yeah i'm running a lts distro but keep the kernel up to date, it's super simple, there's even gui apps for it permalink fedilink source parent
[–] bitfucker@programming.dev 1 point 7 months ago That's the thing. It doesn't break it to the point of unusable unlike other patches I think. So the test didn't catch this since performance is rarely a part of a test suite no? permalink fedilink source parent