After denying the latest kernel update for over a month I inadvertently
authorized it during my last update. Big mistake. My nvidia driver is now
trashed as well as my VMware Workstation. Problem is the usual things I do
to fix it are not working.
nVidia driver doesn’t work from the repos and when I try the manual
installer it is complaining that my gcc is missing and I need to set the CC
varialble. First off gcc is installed and the environment is correct so
no idea what that’s about.
VMware workstation (7.1.4) is also failing to compile.
Any help with either issue is appreciated. I’ve exhausted all the cuurent
online tips so I need some serious help with troubleshooting what is wrong.
I am operating in safe mode at the moment to access the internet.
Unfortunatley I have gone through all of those posts and none of them
A) Solve my issue B) Match what I am seeing. It appears there may be
something wrong with my build evironment as I cannot compile the nVidia
driver either.
> Unfortunatley I have gone through all of those posts and none of them
> A) Solve my issue B) Match what I am seeing. It appears there may be
> something wrong with my build evironment as I cannot compile the nVidia
> driver either.
I am reinstalling the development packages using the YaST pattern for base
develoment and kernel development. Will see if that helps.
>> is to install the Linux Kernel Development pattern
>
> Code:
> --------------------
> # zypper in patterns-openSUSE-devel_kernel
> --------------------
Loading repository data…
Reading installed packages…
‘patterns-openSUSE-devel_kernel’ is already installed.
No update candidate for ‘patterns-openSUSE-devel_kernel-11.4-6.9.1.i586’.
The highest available version is already installed.
Resolving package dependencies…
> I am reinstalling the development packages using the YaST pattern for base
> develoment and kernel development. Will see if that helps.
This is an utter nightmare. I have spent two days on this now.
I rolled back the kernel and kernel source and I still cannot compile!?
Please anyone help with this, it is really starting to impact my work.
I’m giving this a few more hours and then I’m going to grab another hard
drive and install from scratch. I have to get this working again today.
On Wed, 31 Aug 2011 19:54:41 +0000, GofBorg wrote:
>> I am reinstalling the development packages using the YaST pattern for
>> base develoment and kernel development. Will see if that helps.
>
> This is an utter nightmare. I have spent two days on this now. I rolled
> back the kernel and kernel source and I still cannot compile!? Please
> anyone help with this, it is really starting to impact my work. I’m
> giving this a few more hours and then I’m going to grab another hard
> drive and install from scratch. I have to get this working again today.
I could swear I posted a reply here.
This appears to not be kernel-related, but compiler related. I would
reinstall the entire kernel development pattern (including all dev
tools). That gcc is throwing a segfault isn’t a kernel issue, it’s an
issue with gcc.
> I’m giving this a few more hours and then I’m going to grab another hard
> drive and install from scratch. I have to get this working again today.
when you do, i have three suggestions for your next system:
enable kernel multiversion which will allow you to keep the old
kernel when you add a new one, roll back is as easy-peasy as booting and
in the grub screen select the old kernel–job done, follow the
instructions here: http://tinyurl.com/42fq3wy
use YaST Online Updater to make any suggested update “taboo” and then
you will not have to remember to not allow it, do that this way:
-right click on the suggested update
-select ‘taboo’
-click Accept
if you are still using packagekit, STOP using it, it is broken…use
YaST Online Updater (or zypper) instead…
–
DD Caveat
openSUSE®, the “German Engineered Automobile” of operating systems!
> 1. enable kernel multiversion which will allow you to keep the old
> kernel when you add a new one, roll back is as easy-peasy as booting and
> in the grub screen select the old kernel–job done, follow the
> instructions here: http://tinyurl.com/42fq3wy
I’ll look into that.
> 2. use YaST Online Updater to make any suggested update “taboo” and then
> you will not have to remember to not allow it, do that this way:
Unfortunately from past experience it did not honor ‘taboo’ so I gave up on
that option awhile ago.
> 3. if you are still using packagekit, STOP using it, it is broken…use
> YaST Online Updater (or zypper) instead…
> This appears to not be kernel-related, but compiler related. I would
> reinstall the entire kernel development pattern (including all dev
> tools). That gcc is throwing a segfault isn’t a kernel issue, it’s an
> issue with gcc.
Thing is there doesn’t appear to be anything wrong. There are no red
packages. I’ve tried installing and reinstalling gcc and the tools to no
avail. I have nouveau running temporarily for my dual head but that’s not a
long term solution just enough to get me over this mess. It’s a shame really
I haven’t had this serious of a problem in probably 4 years…but this is
bad and there doesn’t seem to be a reason for it. All I did was accept a
kernel update, why would that hose my gcc which worked up until now?
On Wed, 31 Aug 2011 21:29:00 +0000, GofBorg wrote:
>> This appears to not be kernel-related, but compiler related. I would
>> reinstall the entire kernel development pattern (including all dev
>> tools). That gcc is throwing a segfault isn’t a kernel issue, it’s an
>> issue with gcc.
>
> Thing is there doesn’t appear to be anything wrong. There are no red
> packages. I’ve tried installing and reinstalling gcc and the tools to
> no avail. I have nouveau running temporarily for my dual head but
> that’s not a long term solution just enough to get me over this mess.
> It’s a shame really I haven’t had this serious of a problem in probably
> 4 years…but this is bad and there doesn’t seem to be a reason for it.
> All I did was accept a kernel update, why would that hose my gcc which
> worked up until now?
I’m thinking that gcc may have had an issue prior to that - you could
verify the packages to make sure. If you wanted to dig deeper into it,
you could modify the vmware-modconfig script to run gcc under gdb, and
when it segfaults see if you can determine why it’s segfaulting from a
stack trace.
I do not think that the kernel update is directly related to the compilation problem. Something has been removed since you last built a module, and it has just lain hidden until the update made you try it again.
I am reinstalling the development packages using the YaST pattern for base
development and kernel development. Will see if that helps.
Are you sure that you are forcing a re-installation of the packages? I would use YaST to find the installed packages in the “Base Development” pattern and manually select them for re-installation/update.
Another thing that could cause this sort of problem is the inclusion of a non-11.4 repository.
What about trying to compile something really simple like “Hello World”?
Use vim or any other text editor to create a file “hello.c” containing