FreeBSD-6.1R Released!!!!!
Sunday, May 21, 2006
w00t!!! The FreeBSD team has announced the released of FreeBSD-6.1R! (Noticed it a bit late though.. Too occupied with LFS hehe) Another thing to add to my personal daily task aside from my LFS. Wonder if I have the time to test ride it.. FreeBSD-6.0R was fantastic! Hope FreeBSD-6.1R will be even better!! Uhmmm.. I think it's best to complete my LFS first..
Man it is a great day today!! Unleashed the "F" Beast!!
Weekend's mumbling and rattling...
It's been quite a while since my last update.. Been busy with a few stuffs on my own and the departure of my aunt.. Apart from that, life is going at a steady beat.. Last night went to the "Curve" for dinner at Little Penang - they have quite a few delicious dish there.. I decided to have a go at "Home Town Lam Mee" (if I remembered correctly hehe) and it was awesome.. Usually I don't go soup-based noodles , but the dish that I had last nite was exceptional.. The bowl of ABC was damn nice.. As my wife said, there are only 3 places that served a very nice ABC:-
1) Food Court at Giant Shah Alam
2) Lower Ground Food Court at Lot10
3) The latest addition to this list is Little Penang@The Curve..
Enough about that.. I've been busy with my LFS lately and at this time - I realized something.. It is time for me to try something a bit bolder with my LFS. I'm an OS and optimization tweaks (also security but that is out of scope for this purpose) fanatic - meaning that I would twist and tweak my OSes of choice - that would be Linux and BSD - to get the best performance in terms of processing speed and stability in general. After all the years I've spent on LFS the settings of my compiler has been quite stagnant ( Well I've started LFS since GCC-2.95 - and there's a lot of improvement and stability options added since then..). So after completing the base LFS system, I've decided to re-do the whole thing. Why? I decided to try out some different setting that MAY make my system performs better OR break the whole damn system.. All in the name of curiosity.. They say curiosity kills the cat - but a cat has nine lives so what the heck - if it crash, it crash and re-run the build script that I have again - 40 minutes later and I already have the stable base system already :)
So what are the drastic changes that I'm gonna take? Particularly saying - floating point arithmetics for SSE and the "128bit-long-double" options. There is also an option that comes to play with aside from those two - preferred stack boundaries but this option is tricky (*sighs*). If done correctly the results would a very fast system. If not - everything in the system will break! (DUH!)
This is what the GCC team wrote:
-mfpmath=unit
Generate floating point arithmetics for selected unit unit. The choices for unit are:
`387'
Use the standard 387 floating point coprocessor present majority of chips and emulated otherwise. Code compiled with this option will run almost everywhere. The temporary results are computed in 80bit precision instead of precision specified by the type resulting in slightly different results compared to most of other chips. See -ffloat-store for more detailed description.
This is the default choice for i386 compiler.
`sse'
Use scalar floating point instructions present in the SSE instruction set. This instruction set is supported by Pentium3 and newer chips, in the AMD line by Athlon-4, Athlon-xp and Athlon-mp chips. The earlier version of SSE instruction set supports only single precision arithmetics, thus the double and extended precision arithmetics is still done using 387. Later version, present only in Pentium4 and the future AMD x86-64 chips supports double precision arithmetics too.
For the i386 compiler, you need to use -march=cpu-type, -msse or -msse2 switches to enable SSE extensions and make this option effective. For the x86-64 compiler, these extensions are enabled by default.
The resulting code should be considerably faster in the majority of cases and avoid the numerical instability problems of 387 code, but may break some existing code that expects temporaries to be 80bit.
Anyways if someone out there decided to do the same - ensure that you have a Intel processor that supports the SSE/SSE2 instruction sets. Otherwise you are commiting suicide on your system (If you don't know whether your processor supports those instruction sets, I advise that you forget about trying out these - obviously you have no clue on what you are doing :P)
My end system result would result in these having installed: GCC-4.0.3, GlibC-2.3.6, Linux-2.6.16.17, Blackbox-0.71, Gnome-2.14 (I'm getting curious with Gnome-2.14 due to its speed and performance statements) and the modular Xorg-7.0.0. Hope all will work fine with me :)
On another note, I'm getting the Saitek Gamers keyboard complete with its programmable keypad - costing me RM215 for the keyboard and an additional RM177 for the programmable keypad. Why? Coz the Logitech G15 Gamers Keyboard is not available here haha (All in the name of online games.. I must be out of my mind, but what the heck - enjoy life as it is!!)
OK.. I think that's about it for today - I want to get started on this today. If the FORCE is strong with me - I will end with a system and could out-perform my previous system, if not than *tssk* *tssk* - restart from scratch *tssk* *tssk*
P/S: Do you think I should create another blog just for my life with LFS? Any comments will be nice.. Have a nice day!
1) Food Court at Giant Shah Alam
2) Lower Ground Food Court at Lot10
3) The latest addition to this list is Little Penang@The Curve..
Enough about that.. I've been busy with my LFS lately and at this time - I realized something.. It is time for me to try something a bit bolder with my LFS. I'm an OS and optimization tweaks (also security but that is out of scope for this purpose) fanatic - meaning that I would twist and tweak my OSes of choice - that would be Linux and BSD - to get the best performance in terms of processing speed and stability in general. After all the years I've spent on LFS the settings of my compiler has been quite stagnant ( Well I've started LFS since GCC-2.95 - and there's a lot of improvement and stability options added since then..). So after completing the base LFS system, I've decided to re-do the whole thing. Why? I decided to try out some different setting that MAY make my system performs better OR break the whole damn system.. All in the name of curiosity.. They say curiosity kills the cat - but a cat has nine lives so what the heck - if it crash, it crash and re-run the build script that I have again - 40 minutes later and I already have the stable base system already :)
So what are the drastic changes that I'm gonna take? Particularly saying - floating point arithmetics for SSE and the "128bit-long-double" options. There is also an option that comes to play with aside from those two - preferred stack boundaries but this option is tricky (*sighs*). If done correctly the results would a very fast system. If not - everything in the system will break! (DUH!)
This is what the GCC team wrote:
-mfpmath=unit
Generate floating point arithmetics for selected unit unit. The choices for unit are:
`387'
Use the standard 387 floating point coprocessor present majority of chips and emulated otherwise. Code compiled with this option will run almost everywhere. The temporary results are computed in 80bit precision instead of precision specified by the type resulting in slightly different results compared to most of other chips. See -ffloat-store for more detailed description.
This is the default choice for i386 compiler.
`sse'
Use scalar floating point instructions present in the SSE instruction set. This instruction set is supported by Pentium3 and newer chips, in the AMD line by Athlon-4, Athlon-xp and Athlon-mp chips. The earlier version of SSE instruction set supports only single precision arithmetics, thus the double and extended precision arithmetics is still done using 387. Later version, present only in Pentium4 and the future AMD x86-64 chips supports double precision arithmetics too.
For the i386 compiler, you need to use -march=cpu-type, -msse or -msse2 switches to enable SSE extensions and make this option effective. For the x86-64 compiler, these extensions are enabled by default.
The resulting code should be considerably faster in the majority of cases and avoid the numerical instability problems of 387 code, but may break some existing code that expects temporaries to be 80bit.
My end system result would result in these having installed: GCC-4.0.3, GlibC-2.3.6, Linux-2.6.16.17, Blackbox-0.71, Gnome-2.14 (I'm getting curious with Gnome-2.14 due to its speed and performance statements) and the modular Xorg-7.0.0. Hope all will work fine with me :)
On another note, I'm getting the Saitek Gamers keyboard complete with its programmable keypad - costing me RM215 for the keyboard and an additional RM177 for the programmable keypad. Why? Coz the Logitech G15 Gamers Keyboard is not available here haha (All in the name of online games.. I must be out of my mind, but what the heck - enjoy life as it is!!)
OK.. I think that's about it for today - I want to get started on this today. If the FORCE is strong with me - I will end with a system and could out-perform my previous system, if not than *tssk* *tssk* - restart from scratch *tssk* *tssk*
P/S: Do you think I should create another blog just for my life with LFS? Any comments will be nice.. Have a nice day!
"Those who conquer act; those who are conquered think!"
Rampage - Predacon gunner
Rampage - Predacon gunner
When we have to go back...
Tuesday, May 09, 2006
Al-Fatihah. My aunt on my wife's side just passed away on Monday. As they say, every start has its own ending and the same with us all - we came from God and to God we shall return. May her soul rest in peace.
No rants for this part.
No rants for this part.
