4:59 p.m.
Having problems compiling the driver for an NVIDIA card...
It gives me the error(from /var/log/nvidia-installer.log):
executing: 'cd ./usr/src/nv; make nvidia.o
SYSINCLUDE=/lib/modules/2.4.21-20
.1.2024.2.1.nptl/build/include'...
You appear to be compiling the NVIDIA kernel module with
a compiler different from the one that was used to compile
the running kernel. This may be perfectly fine, but there
are cases where this can lead to unexpected behaviour and
system crashes.
If you know what you are doing and want to override this
check, you can do so by setting IGNORE_CC_MISMATCH.
In any other case, set the CC environment variable to the
name of the compiler that was used to compile the kernel.
*** Failed cc sanity check. Bailing out! ***
make: *** [gcc-check] Error 1
-> Error.
ERROR: Unable to build the NVIDIA kernel module.
ERROR: Installation has failed. Please see the file
'/var/log/nvidia-installer.log' for details. You may find
suggestions
on fixing installation problems in the README available on the
Linux
driver download page at www.nvidia.com
I tried installing gcc and ccp packages from RH9, but cpp depends on
XFree86-4.3... and continues...
5:17 p.m.
Mitch Anderson wrote:
Having problems compiling the driver for an NVIDIA card...
...
Sorry - I'm drifting off topic again.
Is there any way we can ditch the freaking NVIDIA installer? AFAICT
that thing is a complete waste of bits. I wish I could just grab the
tarball and compile it like I always did, or make an RPM.
And yes, the NVIDIA driver/GL is 'must-have' if you care about 3d.
<Joe
11:44 a.m.
> ages ago I asked the opengroup motif people about that part of
their
> COPYRIGHT and was informed with binary modules loaded this license doesn't
> apply and I should purchase a real one.
Isn't this more restrictive than the GPL and therefore incompatible?
Incompatible with what ? GPL software using it - possibly
Does this mean I can't run an OpenMotif app over the wire to a
non-OSS
display? I.e., use OM 'with' a non-OSS system?
Who knows
Would it be acceptable for redhat-config-xfree86 to pop up a dialog
warning that the nvidia module is not OSS and therefore may be
incompatible with the licence(s) of other packages, and then let the
user worry whether s?he is safe from the license police?
I guess if you packaged it you could put a conflicts line in for openmotif.
RHLP isnt about proprietary software. If someone wants to do nice automated
packaged rpm installers for the nvidia modules go do it. The core code
however needs to be clearly free software
5:31 p.m.
Alan Cox wrote:
>>ages ago I asked the opengroup motif people about that part of
their
>>COPYRIGHT and was informed with binary modules loaded this license doesn't
>>apply and I should purchase a real one.
This particular case will hopefully soon become irrelevant (er, even
more irrelevant - to me at least).
From: http://www.opengroup.org/openmotif/faq.html
Q: Does the Open Group Public License for Motif meet the
Open Source Guidelines?
A: No. ...
Q: Will Motif be made Open Source in the future?
A: Yes. ...
And, there is now at least a draft proposal for the 'Open' Group to
actually become... open.
http://www.opengroup.org/tech/open-source/opengroup-os-strategy.htm
<Joe
9:19 a.m.
[ suggestion: redhat-config-xfree86 detect and configure the nvidia driver ]
Alan Cox wrote:
And uninstall open-motif
| This software is subject to an open license. It may only be
| used on, with or for operating systems which are themselves open
| source systems.
ages ago I asked the opengroup motif people about that part of their
COPYRIGHT and was informed with binary modules loaded this license doesn't
apply and I should purchase a real one.
Isn't this more restrictive than the GPL and therefore incompatible?
Does this mean I can't run an OpenMotif app over the wire to a non-OSS
display? I.e., use OM 'with' a non-OSS system?
Would it be acceptable for redhat-config-xfree86 to pop up a dialog
warning that the nvidia module is not OSS and therefore may be
incompatible with the licence(s) of other packages, and then let the
user worry whether s?he is safe from the license police?
Frankly, of the two, a good 3d driver means more to me than OpenMotif.
<Joe
6:10 a.m.
that driver simply by clicking the (normally disabled) "Enable
3d
acceleration" checkbox. Only two things need to be done: comment out Load
"dri", and change driver from "nv" to "nvidia".
And uninstall open-motif
| This software is subject to an open license. It may only be
| used on, with or for operating systems which are themselves open
| source systems.
ages ago I asked the opengroup motif people about that part of their
COPYRIGHT and was informed with binary modules loaded this license doesn't
apply and I should purchase a real one.
Alan
4:11 a.m.
Run the installer with ./nvidia-installer --extract-only, then
you'll be
able to do the old make install. Run with --help (who'd imagined?) for more
options.
This is not entirely correct. It should be
sh NVIDIA-Linux-x86-1.0-4363.run --extract-only
on the downloaded file. It does however also take the --help argument.
/Peter
3:54 a.m.
Is there any way we can ditch the freaking NVIDIA installer? AFAICT
that thing is a complete waste of bits. I wish I could just grab the
tarball and compile it like I always did, or make an RPM.
Run the installer with ./nvidia-installer --extract-only, then you'll be able
to do the old make install. Run with --help (who'd imagined?) for more
options.
Nvidia encourages 3:rd parties to build rpms, and I believe there's one in QA
on www.fedora.us (correct me if I'm wrong).
On a related topic, it would be absolutely _great_ if redhat-config-xfree86
could detect that the nvidia driver is installed, and enable you to switch to
that driver simply by clicking the (normally disabled) "Enable 3d
acceleration" checkbox. Only two things need to be done: comment out Load
"dri", and change driver from "nv" to "nvidia".
/Peter
8:45 p.m.
If you read the error message it give you in the log it tells you want
to do. Either set the IGNORE_CC_MISMATCH environment variable or set the
CC environment variable to the compiler used to compile the kernel.
Personally I do the first (ie. export IGNORE_CC_MISMATCH=1).
Elliot
On Thu, 2003-07-24 at 17:59, Mitch Anderson wrote:
Having problems compiling the driver for an NVIDIA card...
It gives me the error(from /var/log/nvidia-installer.log):
executing: 'cd ./usr/src/nv; make nvidia.o
SYSINCLUDE=/lib/modules/2.4.21-20
.1.2024.2.1.nptl/build/include'...
You appear to be compiling the NVIDIA kernel module with
a compiler different from the one that was used to compile
the running kernel. This may be perfectly fine, but there
are cases where this can lead to unexpected behaviour and
system crashes.
If you know what you are doing and want to override this
check, you can do so by setting IGNORE_CC_MISMATCH.
In any other case, set the CC environment variable to the
name of the compiler that was used to compile the kernel.
*** Failed cc sanity check. Bailing out! ***
make: *** [gcc-check] Error 1
-> Error.
ERROR: Unable to build the NVIDIA kernel module.
ERROR: Installation has failed. Please see the file
'/var/log/nvidia-installer.log' for details. You may find
suggestions
on fixing installation problems in the README available on the
Linux
driver download page at www.nvidia.com
I tried installing gcc and ccp packages from RH9, but cpp depends on
XFree86-4.3... and continues...
--
Rhl-beta-list mailing list
Rhl-beta-list(a)redhat.com
http://www.redhat.com/mailman/listinfo/rhl-beta-list
7580
days inactive
7581
days old
9 comments
6 participants
participants (6)
-
Alan Cox -
Elliot Peele -
Gerald Henriksen -
Joe Smith -
Mitch Anderson -
Peter Backlund