[Sussex] Debian and the nvidia driver!
John D.
johnsemail at f2s.com
Sun Aug 20 09:47:51 UTC 2006
On Sunday 20 August 2006 08:25, Steven Dobson wrote:
> John
>
> Watch out. Sometimes, like now I post to both the sender and the mail
> list. When using mail it is better to hit the "reply to all" button
> rather than just "reply". "Reply" should really only be used when you
> want to just the send and not to everyone else involved in the e-mail
> exchange. It's a good habit to get into because it works both exchanges
> with a single person or with a group.
>
> When I rule the Net all MUA will only have on reply button and that will
> reply to ALL. If you don't what to e-mail everyone on the orginal
> distrobution list then you will have to edit the "To:" and "CC:" headers
> by hand.
Hum? that'll be because I just have Kontact/kmail set with only reply and
forward buttons - I must check the config/settings for it!
> On Sun, 2006-08-20 at 02:32 +0100, John D. wrote:
> > On Saturday 19 August 2006 20:12, Steven Dobson wrote:
> > <snip>
> >
> > > > No, it hasn't. If I look, I can see it there "plain as day" (as is
> > > > the
> >
> > I just checked through konqueror again about this, and if I follow it all
> > the way through, and then try to "click" the
> > lib/modules/2.6.17-2-686/build/include/linux/kernel.h I receive an error
> > message that this is either missing or no longer exists (sorry I presumed
> > it was there because the folder is still there).
>
> <snip>
>
> > So, I'm now presuming that I have to sort this "kernel.h" problem as that
> > might be what seems to be causing all of the possible remedies to fail
> > (and yes I've tried lots - about all of the different ones that I can
> > find on the net).
>
> I agree with you.
>
> > the modules-assistant facility says about looking at the helper log
> > (presuming the modules-assistant helper log) which I can't find or the
> > error is not letting me look at this.
>
> Don't know about that.
>
> > Erm, I've also just double checked the lib/modules/2.6.17-2-686/build
> > which seems to be present i.e. it's a folder that has 5 other folders
> > within it, so the "incorrect path" error - I don't understand or know
> > what it might be incorrect.
>
> Okay build exits, but what is it? On my system is is a symbolic link to
> where the headers really live. I have a difference here, I have a lot
> more in the place build points to. However, the header file exists and
> is also a symbolic link but that file exists too:
>
> $ cd /lib/modules/2.6.17-2-amd64
> $ ls -l build
> lrwxrwxrwx 1 root root 37 Aug 19 19:10 build
> -> /usr/src/linux-headers-2.6.17-2-amd64
Which showed me ~$ cd /lib/modules/2.6.17-2-686
~$ ls -l build
lrwxrwxrwx 1 root root 35 Aug 18 19:55
build -> /usr/src/linux-headers-2.6.17-2-686
> $ ls -lL build
> total 332
> -rw-r--r-- 1 root root 44721 Aug 11 19:44 Makefile
> -rw-r--r-- 1 root root 267554 Aug 11 20:03 Module.symvers
> drwxr-xr-x 3 root root 19 Aug 19 19:10 arch
> drwxr-xr-x 2 root root 57 Aug 19 19:09 block
> drwxr-xr-x 2 root root 35 Aug 19 19:09 crypto
> drwxr-xr-x 58 root root 4096 Aug 19 19:09 drivers
> drwxr-xr-x 60 root root 4096 Aug 19 19:09 fs
> drwxr-xr-x 7 root root 4096 Aug 19 19:10 include
> drwxr-xr-x 2 root root 35 Aug 19 19:09 init
> drwxr-xr-x 2 root root 21 Aug 19 19:09 ipc
> drwxr-xr-x 4 root root 82 Aug 19 19:09 kernel
> drwxr-xr-x 5 root root 112 Aug 19 19:09 lib
> drwxr-xr-x 2 root root 35 Aug 19 19:09 mm
> drwxr-xr-x 36 root root 4096 Aug 19 19:09 net
> drwxr-xr-x 6 root root 4096 Aug 19 19:09 scripts
> drwxr-xr-x 4 root root 60 Aug 19 19:09 security
> drwxr-xr-x 16 root root 4096 Aug 19 19:09 sound
> drwxr-xr-x 2 root root 35 Aug 19 19:09 usr
and correspondingly, ~$ ls -lL
total 1376
drwxr-xr-x 4 root root 4096 Aug 20 01:05 build
drwxr-xr-x 10 root root 4096 Aug 18 19:47 kernel
-rw-r--r-- 1 root root 305292 Aug 19 19:25 modules.alias
-rw-r--r-- 1 root root 69 Aug 19 19:25 modules.ccwmap
-rw-r--r-- 1 root root 310594 Aug 19 19:25 modules.dep
-rw-r--r-- 1 root root 813 Aug 19 19:25 modules.ieee1394map
-rw-r--r-- 1 root root 1186 Aug 19 19:25 modules.inputmap
-rw-r--r-- 1 root root 32234 Aug 19 19:25 modules.isapnpmap
-rw-r--r-- 1 root root 74 Aug 19 19:25 modules.ofmap
-rw-r--r-- 1 root root 250258 Aug 19 19:25 modules.pcimap
-rw-r--r-- 1 root root 1135 Aug 19 19:25 modules.seriomap
-rw-r--r-- 1 root root 131585 Aug 19 19:25 modules.symbols
-rw-r--r-- 1 root root 317301 Aug 19 19:25 modules.usbmap
john at johnspc:/lib/modules/2.6.17-2-686$
> $ ls -l include/linux/kernel.h
> lrwxrwxrwx 1 root root 54 Aug 19 19:10 include/linux/kernel.h
> -> ../../../linux-headers-2.6.17-2/include/linux/kernel.h
> $ ls -lL include/linux/kernel.h
> -rw-r--r-- 1 root root 9954 Jun 18 02:49 include/linux/kernel.h
> $
>
> NOTE: This is the output from my system which is running a AMD64 kernel.
> The first command for you will be:
> $ cd /lib/modules/2.6.17-2-686
>
> But the rest should work the same and give the same result.
Nope, that doesn't seem correct, because if I try doing
john at johnspc./lib/modules/2.6.17-2-686$ ls-l include/linux/kernel.h
it's telling me that theres no such file or directory,
but if I look at slightly differently -
john at johnspc:/lib/modules/2.6.17-2-686$ ls -l build/include/linux/kernel.h
lrwxrwxrwx 1 root root 54 Aug 18 19:55
build/include/linux/kernel.h -> ../../../linux-headers-2.6.17-2/include/linux/kernel.h
john at johnspc:/lib/modules/2.6.17-2-686$
so logically if I then try (with konqueror) to see whats going on
with /lib/modules/2.6.17-2-686/build/include/linux and then "hover" over
kernel.h it's showing it as an empty link so I think I'm correct in doing
this with CLI
john at johnspc:/lib/modules/2.6.17-2-686$ ls -lL /build/include/linux/kernel.h
ls: /build/include/linux/kernel.h: No such file or directory
john at johnspc:/lib/modules/2.6.17-2-686$
which then begs the question of "where does kernel.h come from" - it also
explains why it was asking/telling me that "kernel source path
to '/lib/modules/2.6.17-2-686/build is incorrect" (well I think it does or
might that me a different setting/problem?).
regards
John D.
More information about the Sussex
mailing list