 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Tue, 09 Dec 2008 08:36:13 -0500, Tom Austin wrote:
> Jim Henderson wrote:
>> On Mon, 08 Dec 2008 14:18:25 -0500, Tom Austin wrote:
>>
>>>> Try LFS, that'll really teach you a lot about linux, even if you
>>>> don't get a bootable system.
>>>>
>>>> Jim
>>> It's not too hard to get a bootable LFS system - and if it is hard and
>>> you get it to boot, then you've learned a lot!
>>
>> Maybe now it is, I remember it used to be an incredible way to install
>> because you had to do *everything*. I've seen some newer releases that
>> script some things, though.
>>
>> Jim
>
>
> Even with the scripting/automation it takes a bit of work to get it to
> boot - you still have to get GRUB on the HD and the like.
>
> I usually run jhalfs to do all the legwork on the base system - up to,
> but not including, compiling the kernel. Up to that point it's mostly
> following what the book tells you to do. From the kernel on you have to
> make some decisions and customize the steps for your particular setup.
>
> Just about any Linux stuff I do now is via LFS - I like knowing what I
> am working with.
That is one of the best things about using Linux - and I mean *really*
using it. You can dig in as deep as you want, and there's always
something new to learn.
Jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Tue, 09 Dec 2008 12:03:06 +0000, Invisible wrote:
> Yeah, well... why bother supporting a platform that hardly anybody uses?
> *Clearly* it's more profitable to support only the most popular
> platforms. ;-)
>
> That's why no Linux drivers, no Linux games, no Linux apps...
>
> (Yes, I realise there are a few of these now, but still not that many.
> Yet.)
This is not an accurate portrayal by any stretch.
"no linux drivers" - as someone pointed out, especially as regards legacy
hardware support, Linux kicks Windows' ass. The time it becomes a
problem is when the manufacturer won't release specifications for the
hardware so drivers can be created.
"a few" is a poor characterization at best.
Jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Tue, 09 Dec 2008 08:52:29 -0800, Darren New <dne### [at] san rrcom> wrote:
>Mueen Nawaz wrote:
> > Why a man would want a wife is a big mystery to some people. Why a man
> > would want *two* wives is a bigamystery.
>
>I just want to say, that is excellent. :-)
Not reading the ng's properly I thought that it was yours :)
--
Regards
Stephen
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Jim Henderson wrote:
> On Tue, 09 Dec 2008 12:03:06 +0000, Invisible wrote:
>
>> Yeah, well... why bother supporting a platform that hardly anybody uses?
>> *Clearly* it's more profitable to support only the most popular
>> platforms. ;-)
>>
>> That's why no Linux drivers, no Linux games, no Linux apps...
>>
>> (Yes, I realise there are a few of these now, but still not that many.
>> Yet.)
>
> This is not an accurate portrayal by any stretch.
>
> "no linux drivers" - as someone pointed out, especially as regards legacy
> hardware support, Linux kicks Windows' ass. The time it becomes a
> problem is when the manufacturer won't release specifications for the
> hardware so drivers can be created.
>
> "a few" is a poor characterization at best.
I meant there are no Linux drivers for high-end hardware like expensive
graphics cards. Clearly there *are* Linux drivers for lots of other things.
And by "a few" I meant more games and applications rather than drivers.
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
scott wrote:
>
> In Windows you just install a driver, why do you need to recompile the
> kernel?
>
You don't need to, if you have module support on, you can compile only
the module (or if you have it already compiled, just load it).
-Aero
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
scott <sco### [at] scottcom> wrote:
> In Windows you just install a driver, why do you need to recompile the
> kernel?
It's the classical monolithic kernel vs. microkernel war. There are
advantages and disadvantages in both methods.
For example, one disadvantage of Windows is that each new version of
Windows always breaks a bunch of old drivers, so when you upgrade Windows
you may find out that your old peripheral just doesn't work anymore.
When the driver is integrated into the kernel, like in linux, it will
obviously always work even if the kernel is seriously updated (because
all the embedded drivers will be updated for the new kernel design as
well).
OTOH linux does support kernel modules. For example, if you want to
install, let's say, the ATI binary driver, you don't need to recompile
the kernel.
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Invisible wrote:
>> Yes, it is a mammoth.
>>
>> grunka ~ # du -hs /var/tmp/portage/app-office/openoffice-3.0.0/work/ooo/
>> 2.6G /var/tmp/portage/app-office/openoffice-3.0.0/work/ooo/
>
> Please tell me that isn't 2.6 GB just for the source code...
Probably not. There is probably also some... icons?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Jim Henderson wrote:
> On Tue, 09 Dec 2008 08:36:13 -0500, Tom Austin wrote:
>
>> Jim Henderson wrote:
>>> On Mon, 08 Dec 2008 14:18:25 -0500, Tom Austin wrote:
>>>
>>>>> Try LFS, that'll really teach you a lot about linux, even if you
>>>>> don't get a bootable system.
>>>>>
>>>>> Jim
>>>> It's not too hard to get a bootable LFS system - and if it is hard and
>>>> you get it to boot, then you've learned a lot!
>>> Maybe now it is, I remember it used to be an incredible way to install
>>> because you had to do *everything*. I've seen some newer releases that
>>> script some things, though.
>>>
>>> Jim
>>
>> Even with the scripting/automation it takes a bit of work to get it to
>> boot - you still have to get GRUB on the HD and the like.
>>
>> I usually run jhalfs to do all the legwork on the base system - up to,
>> but not including, compiling the kernel. Up to that point it's mostly
>> following what the book tells you to do. From the kernel on you have to
>> make some decisions and customize the steps for your particular setup.
>>
>> Just about any Linux stuff I do now is via LFS - I like knowing what I
>> am working with.
>
> That is one of the best things about using Linux - and I mean *really*
> using it. You can dig in as deep as you want, and there's always
> something new to learn.
>
> Jim
You can say that again - I'm working on some embedded junk and the
amount of stuff that's there is incredible.
After fiddling around a bit I found initrd - something that a lot of
linux systems use. Pretty cool, especially if you are trying to set up
an embedded system.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp wrote:
> scott <sco### [at] scottcom> wrote:
>> In Windows you just install a driver, why do you need to recompile the
>> kernel?
>
> It's the classical monolithic kernel vs. microkernel war. There are
> advantages and disadvantages in both methods.
I don't think Windows is anywhere near what you'd call a "microkernel",
altho it looks like MS is trying to push everything they can out of kernel
space for their own protection.
Generally a "microkernel" runs as little as possible in kernel space,
deferring all device drivers and such to user-space code, providing only
things like scheduling and IPC. (MS's Singularity, Amiga's original OS
(except that was *all* kernel mode), Mach (about which I know almost
nothing), etc.)
When a device driver can't crash the system any harder than user code can,
it's a microkernel. :-)
> When the driver is integrated into the kernel, like in linux, it will
> obviously always work even if the kernel is seriously updated (because
> all the embedded drivers will be updated for the new kernel design as
> well).
I don't think that obviously follows. If nobody updates the driver, it'll
break (or more likely just not be present) in newer kernels. Having the
sources available of course lets you fix it, if it's worthwhile to whoever
is competent to fix it. I.e., it's not that the kernel is monolithic, but
that the sources are available to do the fixing, that keeps drivers running
under Linux. (Actually, I've seen many complaints that the kernel doesn't
provide a stable platform for drivers, and it slows development to have to
fix every driver with repeated kernel changes that break them.)
--
Darren New, San Diego CA, USA (PST)
The NFL should go international. I'd pay to
see the Detroit Lions vs the Roman Catholics.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Tue, 09 Dec 2008 14:54:46 -0500, Tom Austin wrote:
> You can say that again - I'm working on some embedded junk and the
> amount of stuff that's there is incredible.
>
> After fiddling around a bit I found initrd - something that a lot of
> linux systems use. Pretty cool, especially if you are trying to set up
> an embedded system.
Yeah, it's quite handy. I always thought it was kinda ironic, though,
that you could load initrd in order to load a required disk driver.
Kinda like on NetWare being able to start the system up but then needing
to load a disk driver to read the disk once the NW kernel is up and
running (but not exactly the same).
Jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
