Ben> Mike Castle wrote:
>> People who are going to be savvy enough to install a development
>> 2.5.* kernel that is defining a new configuration utility are going
>> to be savvy enough to install python.
>>
Ben> Not only that, but Alan said that somebody is rewriting it in C.
No and yes, the Python 2 issue is not reasonable, the C version of it
is. Hopefully with a proper C version, the Python 2 dependencies will
go away completely and that part of the discussion becomes moot.
The Python 2 one is a major issue, some people compile current kernels
because thats all that exists for their hardware. Some people who are
bringing up new architectures etc. wants to be self hosting but things
like Perl and Python are not exactly the first things you build. You
may not have threads, you may not have proper math support, maybe no
shared libraries, but that wont stop you from getting
gcc/binutils/bash going. The argument I got from Eric at the 2.5
kernel summit when I first brought this up was 'just configure your
kernel on another machine and copy it over'. Thats extremely naiive,
in some cases you do not have network nor floppy. You copy your
sources over once (so you can hack on the network driver :), you don't
want to have to rip out the disk every time you need a change to the
.config because some obscure option doesn't compile and you hadn't
noticed.
It's just not that trivial.
Jes
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/