Giuseppe Ghibò skrev 26.9.2010 15:25:
2010/9/26 Thomas Backlund <t...@iki.fi <mailto:t...@iki.fi>>
Giuseppe Ghibò skrev 26.9.2010 02:09:
Not exactly. I'm not talking in just using -march=<something> but in
also pushing -mfpmath=sse -msse (and maybe -msse2) , which should be
much more than JUST 1-2% (1-2% is usually the benchmark
tolerance)...
Well, here is a simple fact.
Yes, going for i686 + SSE2 would/could give some benefits,
BUT
It would close support for all theese:
- Intel i586 (all)
- Intel Pentium Pro
- Intel Pentium II
- Intel Pentium III (Including some Pentium D)
- 32-bit AMD Athlon
- AMD Geode
- VIA C3
- Transmeta Crusoe
So, thats not an option.
As I said I would see someone with such hardware, taking a CD of latest
MDV or cooker and try to install to do something and do a report.
No point in trying the Cooker installer, as its currently broken.
Last time I tried (around 2010.0), I had no problem using the textmode
installer, wich is the most lightway method of installing we have.
That reminds me also one suggestion. If someone has old hardware that
not in use anymore then could be donated to who is willing to test the
latest MDV|Mageia on it.
I will next week, as I still think we have some old i586 stored at work...
It's not ironic (I think this is a
brainstorming), or I'm not kidding, but there could be an wiki hardware
section for that. For old hardware schools and institutes or
corporations have even to pay for dismantling: such hardware is usually
recycled for:
a) extracting gold (there are nowadays new chemical processes that
found more affordable to extract gold from old PCs than from mines)
b) avoid pollutions with the lead, plastics they contain, if thrown
in the dumpster etc. (note there isn't just the CPU, but also CRT
monitors, printers, disks, etc.)
c) keep donating the old harware to schools that take anything
they can get (so no dismantling).
IIRC there is a EU law that if you buy a new PC the vendor is obeyed to
retire the old one and take care for the dismantling.
Nope.
But there is a "recycling fee" in the price of new hw, so should be able
to get tid of old hw without any cost...
Now what about only i686 + SSE then?
It would still close support for all theese:
- Intel i586 (all)
- Some Intel Pentium D
- 32-bit AMD Athlon < XP/MP
- AMD Geode < NX
- VIA C3 (Samuel* and Ezra)
- Transmeta Crusoe
So, still not an option.
Now what about only i686 then?
It would still close support for all theese:
- Intel i586 (all)
- Some Intel Pentium D
- AMD Geode < NX
- VIA C3 (Samuel* and Ezra)
- Transmeta Crusoe
So, still not an option.
And then to point out some other facts:
Some of the benefits are already accounted for like you noted earlier:
"Of course we shouldn't forget that the MDV had already a system for
providing optimized (look at /usr/lib/sse2 for instance) version of
libraries according to instruction set supported."
And many applications capable of utilizing sse2 and other
instruction sets already have runtime detection support, so no
problem there.
So, where does that leave us?
Simple.
For 32bit installs, we will still support i586 as base.
For 64bit installs, it's simple as x86_64 is SSE2 by default.
Now, _if_ the Mageia community gets enough developers to think of
_and_ support a "light netbook/legacy edition", then maybe we can
think of doing a i586/i686 split, but for now we will use i586 as base.
And many of the users wanting i686/sse2 already have hardware
capable of running x86_64, so that's what they should do, as there
is where the future is.
You are forgetting a midrange. Which is the midrange of not the latest
hardware but a memory around 2-4GB RAM. In that set a 32bit OS is still
consuming 30-50% fewer memory than 64bit.
I'm not forgetting them...
The current i586 base supports them
We also provide i686 kernels for them (desktop and server)
We have the "/usr/lib/sse2" and "apps with runtime detection support"
mentioned above.
As I said to complete the circle it is needed to provide on a per
--target|dialect .rpm package for the kernel as well as for the glibc.
We have just to ensure the targets and fix compilation errors (though
that wouldn't resolve the problem of bloatware distro).
--
Thomas
_______________________________________________
Mageia-dev mailing list
Mageia-dev@mageia.org
https://www.mageia.org/mailman/listinfo/mageia-dev
