diff options
Diffstat (limited to 'zarb-ml/mageia-dev/attachments/20100926/5b597a0c')
| -rw-r--r-- | zarb-ml/mageia-dev/attachments/20100926/5b597a0c/attachment-0001.html | 76 | ||||
| -rw-r--r-- | zarb-ml/mageia-dev/attachments/20100926/5b597a0c/attachment.html | 76 |
2 files changed, 152 insertions, 0 deletions
diff --git a/zarb-ml/mageia-dev/attachments/20100926/5b597a0c/attachment-0001.html b/zarb-ml/mageia-dev/attachments/20100926/5b597a0c/attachment-0001.html new file mode 100644 index 000000000..83225b576 --- /dev/null +++ b/zarb-ml/mageia-dev/attachments/20100926/5b597a0c/attachment-0001.html @@ -0,0 +1,76 @@ +2010/9/26 Thomas Backlund <span dir="ltr"><<a href="mailto:tmb@iki.fi">tmb@iki.fi</a>></span><br><br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> +Giuseppe Ghibň skrev 26.9.2010 02:09:<div class="im"><br> +<br> +<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> +Not exactly. I'm not talking in just using -march=<something> but in<br> +also pushing -mfpmath=sse -msse (and maybe -msse2) , which should be<br> +much more than JUST 1-2% (1-2% is usually the benchmark<br> +tolerance)...<br> +</blockquote> +<br></div> +Well, here is a simple fact.<br> +<br> +Yes, going for i686 + SSE2 would/could give some benefits,<br> +<br> +BUT<br> +<br> +It would close support for all theese:<br> +- Intel i586 (all)<br></blockquote><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> +- Intel Pentium Pro<br> +- Intel Pentium II<br> +- Intel Pentium III (Including some Pentium D)<br> +- 32-bit AMD Athlon<br> +- AMD Geode<br> +- VIA C3<br> +- Transmeta Crusoe<br> +<br> +So, thats not an option.<br></blockquote><div><br>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.<br><br>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. 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:<br> +<br> a) extracting gold (there are nowadays new chemical processes that found more affordable to extract gold from old PCs than from mines)<br> 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.)<br> +<br>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.<br> <br></div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> + +<br> +<br> +Now what about only i686 + SSE then?<br> +It would still close support for all theese:<br> +- Intel i586 (all)<br> +- Some Intel Pentium D<br> +- 32-bit AMD Athlon < XP/MP<br> +- AMD Geode < NX<br> +- VIA C3 (Samuel* and Ezra)<br> +- Transmeta Crusoe<br> +<br> +So, still not an option.<br> +<br> +<br> +Now what about only i686 then?<br> +It would still close support for all theese:<br> +- Intel i586 (all)<br> +- Some Intel Pentium D<br> +- AMD Geode < NX<br> +- VIA C3 (Samuel* and Ezra)<br> +- Transmeta Crusoe<br> +<br> +So, still not an option.<br> +<br> +<br> +And then to point out some other facts:<br> +<br> +Some of the benefits are already accounted for like you noted earlier:<div class="im"><br> +"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."<br> +<br></div> +And many applications capable of utilizing sse2 and other instruction sets already have runtime detection support, so no problem there.<br> +<br> +<br> +So, where does that leave us?<br> +<br> +Simple.<br> +For 32bit installs, we will still support i586 as base.<br> +For 64bit installs, it's simple as x86_64 is SSE2 by default.<br> +<br> +<br> +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.<br> +<br> +<br> +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.<br> +<br></blockquote></div><br>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.<br>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).<br> +<br>Bye<br>Giuseppe.<br><br> diff --git a/zarb-ml/mageia-dev/attachments/20100926/5b597a0c/attachment.html b/zarb-ml/mageia-dev/attachments/20100926/5b597a0c/attachment.html new file mode 100644 index 000000000..83225b576 --- /dev/null +++ b/zarb-ml/mageia-dev/attachments/20100926/5b597a0c/attachment.html @@ -0,0 +1,76 @@ +2010/9/26 Thomas Backlund <span dir="ltr"><<a href="mailto:tmb@iki.fi">tmb@iki.fi</a>></span><br><br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> +Giuseppe Ghibň skrev 26.9.2010 02:09:<div class="im"><br> +<br> +<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> +Not exactly. I'm not talking in just using -march=<something> but in<br> +also pushing -mfpmath=sse -msse (and maybe -msse2) , which should be<br> +much more than JUST 1-2% (1-2% is usually the benchmark<br> +tolerance)...<br> +</blockquote> +<br></div> +Well, here is a simple fact.<br> +<br> +Yes, going for i686 + SSE2 would/could give some benefits,<br> +<br> +BUT<br> +<br> +It would close support for all theese:<br> +- Intel i586 (all)<br></blockquote><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> +- Intel Pentium Pro<br> +- Intel Pentium II<br> +- Intel Pentium III (Including some Pentium D)<br> +- 32-bit AMD Athlon<br> +- AMD Geode<br> +- VIA C3<br> +- Transmeta Crusoe<br> +<br> +So, thats not an option.<br></blockquote><div><br>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.<br><br>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. 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:<br> +<br> a) extracting gold (there are nowadays new chemical processes that found more affordable to extract gold from old PCs than from mines)<br> 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.)<br> +<br>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.<br> <br></div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> + +<br> +<br> +Now what about only i686 + SSE then?<br> +It would still close support for all theese:<br> +- Intel i586 (all)<br> +- Some Intel Pentium D<br> +- 32-bit AMD Athlon < XP/MP<br> +- AMD Geode < NX<br> +- VIA C3 (Samuel* and Ezra)<br> +- Transmeta Crusoe<br> +<br> +So, still not an option.<br> +<br> +<br> +Now what about only i686 then?<br> +It would still close support for all theese:<br> +- Intel i586 (all)<br> +- Some Intel Pentium D<br> +- AMD Geode < NX<br> +- VIA C3 (Samuel* and Ezra)<br> +- Transmeta Crusoe<br> +<br> +So, still not an option.<br> +<br> +<br> +And then to point out some other facts:<br> +<br> +Some of the benefits are already accounted for like you noted earlier:<div class="im"><br> +"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."<br> +<br></div> +And many applications capable of utilizing sse2 and other instruction sets already have runtime detection support, so no problem there.<br> +<br> +<br> +So, where does that leave us?<br> +<br> +Simple.<br> +For 32bit installs, we will still support i586 as base.<br> +For 64bit installs, it's simple as x86_64 is SSE2 by default.<br> +<br> +<br> +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.<br> +<br> +<br> +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.<br> +<br></blockquote></div><br>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.<br>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).<br> +<br>Bye<br>Giuseppe.<br><br> |