Make questions and ARCH questions for gentoo/calculate

linus72 · 04-06-2010, 10:11 AM

OK

so I got my nice shiny fluxbox CLS 10.2 desktop
lookin good and now gonna rebuild system for my PC
as athlonXP 2400+ K7

Lookin at calculates docs, I see this stuff
http://www.calculate-linux.org/main/...tion_of_system

and then I see this about setting the /etc/make.conf
http://en.gentoo-wiki.com/wiki/Safe_..._XP.2FGeode_NX

and so I should go with the 2nd one or what??

And, how should I set the make.conf, like so?

Code:

CHOST="i686-pc-linux-gnu"
CFLAGS="-march=athlon-xp -O2 -pipe -fomit-frame-pointer"
CXXFLAGS="${CFLAGS}"

LINGUAS="en en_US"
MAKEOPTS="-j2"

-j1 or -j2??

Whats best for my system???

And how do I set genkernel.conf?
thanks

Lautre · 04-06-2010, 03:36 PM

Look this: http://en.gentoo-wiki.com/wiki/Safe_Cflags
"If you have an Intel or AMD CPU and are using >=sys-devel/gcc-4.2.3, using -march=native is recommended."

"jX" only affects a speed of assembly systems. You can speed up a compilation with the option: EMERGE_DEFAULT_OPTS="--jobs=4" in /etc/make.conf

linus72 · 04-06-2010, 05:28 PM

Thanks Lautre

so all I need is the native tag?

and if my PC has only one CPU, I should just use -j1 or no?

colorpurple21859 · 04-06-2010, 08:44 PM

From what little research I did on this it should be one more than the number of processors you have. So with one processor it would be j2. From what I read it allows full use of unused processor clock cycles or something like that. About a month or so ago installed Gentoo on this computer, an emachine with a single amd64 processor. Reinstalled Gentoo several times. I notice that with -j2 the kernel compilation the computer processing two different sections of the config file at the same time, bouncing back and forth between the two different sections kind of like in parallel . Where as with -j1 the computer was processing the config file in the an orderly sequence during the compilation. Using -j2 seemed to be faster. Hope this helps with out too much confusion.

linus72 · 04-08-2010, 01:39 PM

Instead of making a new Thread, I am going to expound on this one as I may have messed something up...

Today; I am trying to install Pyroom and Gambas2 into my calculate scratch 10.4 fluxbox system.

I found a pyroom ebuild here
http://dev.gentoo.org/~nightmorph/mi...office/pyroom/

and according to gambas site gambas should be in portage repo
but its not
http://gambasdoc.org/help/doc/package?view&en

so, I had to get it here
http://bugs.gentoo.org/show_bug.cgi?id=146871

so, after setting up my ebuild stuff in usr/local as per this
http://linuxreviews.org/gentoo/ebuilds/#toc1

I did both

emerge gambas

and

emerge pyroom

I get these errors

for Pyroom

Code:

!!! All ebuilds that could satisfy "app-office/pyroom" have been masked.
!!! One of the following masked packages is required to complete your request:
- app-office/pyroom-0.4.1 (masked by: missing keyword)

for Gambas

Code:

 * ERROR: dev-db/gambas-2.7.0 failed:
 *   gcc needs to be build with the USE flag libffi
 * 
 * Call stack:
 *             ebuild.sh, line  54:  Called pkg_setup
 *   gambas-2.7.0.ebuild, line  51:  Called die
 * The specific snippet of code:
 *   	die "gcc needs to be build with the USE flag libffi"

I tried cl-unmask on pyroom

Code:

One of the packages not match your architecture.
Please revise dependences of the packages by your own* using flags concealment of packages.

my arch is athlon-xp

NOTE: I have taken a slack pkg I made of pyroom and manually installed
it along with its depends and it works great!

So, why cant portage do it?

How do i fix all this and how do I do "gcc needs to be build with the USE flag libffi"

any ideas?

thanks

Lautre · 04-09-2010, 12:53 AM

To install app-office/pyroom, enough to unmask him, dev-db/gambas must be compiled with the flag "libffi".
run

Code:

echo dev-db/gambas libffi >> /etc/portage/package.use/using

then

Code:

cl-unmask app-office/pyroom
emerge app-office/pyroom