From: Stuart Henderson Subject: Re: arm64 LSE support in userland: introduce elf_aux_info? To: j@bitminer.ca Cc: Mark Kettenis , tech@openbsd.org Date: Sat, 13 Jul 2024 00:15:05 +0100 On 2024/07/12 17:15, j@bitminer.ca wrote: > I think we are in agreement here. comments below. > > On 2024-07-12 16:24, Stuart Henderson wrote: > > On 2024/07/12 16:08, j@bitminer.ca wrote: > > > Some features are fairly complicated to understand; e.g. the array > > > of avx instructions varies dramatically with implementation (code > > > name of the cpu). Fortunately cpuid on Intel is unprivileged so > > > software can decide > > > > Well, with avx512, it also depends on whether the OS > > supports it. If not > > then no matter what CPU, it won't work. > > Yes, and the porter is obliged to limit compiled instructions to avoid > failure. Or, for some codes, to ensure executed > instructions fall within > the cpu feature set because the code adapts to the currently > executing cpu. That is stupid. If there's a way the OS can pass on information about what works, that's definitely a simpler/more reliable way to handle it than parsong text strings. If the OS can't do this then maybe some extra ports work maybe needed, but if there's a way to prevent advertising support for something which is known not to work, that's the way to go, easier to fix in the first place, and easier to remove later when support is added. > As above, the porter wants to avoid illegal instruction faults. You ask a lot of a very limited number of porters, who have quite limited access to hardware, for what is a relatively small scale OS with a specific definition that says "Focus on being developer-oriented in all senses". I think you fundamentally misunderstand OpenBSD and would be happier running fedora or something.