On 2007-Apr-06 14:43:31 -0500, Nikolas Britton <nikolas.britton_at_gmail.com> wrote: >That's not true. If you look at the stats I posted over 58% of the >systems have SSE2 support, compared to the 20%* with 64-bit >capability. So, on your statistics[*], the FreeBSD project should drop support for around 40% of the systems currently running FreeBSD. Exactly how does this benefit those people? >The legacy bits don't upset me, Then why did you start this thread by wanting to delete them? > what upsets me is sacrificing >performance so we can support a minority of legacy systems. Whilst 40% is a minority, it's hardly a _small_ minority. And since you're the one who is keen on statistics, how come you haven't quantified the amount of performance FreeBSD is 'sacrificing'. > IIRC? we >could recode the Kernel for SSE2 math if the processor was guaranteed >to have that SSE2. And this would benefit the project how? How about you do this and report back on the performance improvement. > SSE2 adds 214 new instructions to the existing x86 >instruction set. How many of these instructions are actually useful in the kernel? > The problem with 64-bit FreeBSD is performance, on >average its slower than FreeBSD i386 and FreeBSD i386 is already quite >slow without custom optimizations. I presume you can substantiate these claims. > New >users frequently get themselves into a big mess because it takes years >of knowledge about FreeBSD internals to make it fast without breaking >it. This turns new users off and can create support problems. More unsubstantiated claims. If you believe that the FreeBSD project is headed in the wrong direction, you are free to fork FreeBSD and head off in whatever direction you choose. If you can demonstrate that your approach has advantages then more people will join you. [*] bsdstats is acknowledged by everyone except you to represent a very small and highly biased subset of FreeBSD users. -- Peter Jeremy
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:07 UTC