It’s known that with previous iterations, up to and including 11.2, with certain motherboards ( I can only speak for Asus/Abit ) a quad ( or duo ) core cpu can only be installed with acpi-off or similar options; this leads to only one core running in OpenSUSE, although the board itself declares 4 cores run.
And this seems to mean that various other unimportant ( for a desktop ) acpi functions are affected, including the os system clock.
Typing top or cat /proc/cpuinfo always shows only one core, and although the computer is fine, it seems a pointless state of affairs. Googling shows various suggestions, including removing acpi=off or noapic in the boot options and substituting pci=nommconf, or changing the kernel to an SMP kernel ( which is probably out-of-date as advice now ); still, I can live with it in 11.2: I just wondered if 11.3 addresses this installation problem. Or will any future editions address it ?
Whilst one can live with it, the initial impossibility of installing ( with no clue as to why it shan’t install ) must have put off a deal of people from even going forward with OpenSUSE ( or any similarly affected distro ).
I had this in an HP system. It was reported as a regression against the kernel. I am running kernel 2.6.33-rc2-0.1 in 11.2 now, which fixed it. Thus I assume that it is also fixed inthe final 2.6.33 and onwards.
I do not know if your hardware comes in the same category as mine with respect to this bug.
Oddly, I haven’t the faintest idea whether a bug report has been ever issued, nor do I have the competence or knowledge to file one myself: let alone against the dozens of other motherboards — or whatever component is responsible — affected, of which I will never know.
I was more in the nature of asking a general question as to whether 11.3 installation will be able to accommodate quad-cores without this issue. Since it is NOT my AMD or motherboard alone that has this problem, but also Intel quad cores etc., as seen in a thread on this very forum, amongst other threads dealing with the same **** problem, it would be singularly inept to waste people’s time by filing a bug report about my Asus board when it probably has nothing to do with Asus and is specific to the OS. Or very possibly the kernel.
The linkyou gave is from about 2007. That is definitly not the same as my case, because I never had any problem before openSUSE 10.3.
And about your other remarks I can only make some statements.
. Going to another Linux distribution will not do any good because it is a kernel problem.
. Going to Windows might help, I can not decide for you what the outcome of the pros and cons is for you. For me a very big con would be to learn to master a new OS.
. I doubt if MicroSoft would be as helpfull as the community here help me or you with such a problem.
And that brings me to your general statement as it being beyond your skills to report this in the correct place. I was not born with that skill. I started a thread here. People helped me here and together we came to the conclusion that filing a kernel bug would be a usefull thing to do as well as for my benifit as for the benifit of the community (yes, including you, having a similar problem). Again I was helped by members on this forums to file the bug at least halfway correct. After I was offered to try 2.6.33-rc2-0.1, I got again help in building it and the Nvidia driver. As said, it is still running to my satisfaction. Not all of this is in the Kernel bugreport or in the Foriums thread Has the 11.2 RC2 kernel stolen a CPU? - openSUSE Forums because I also got direct help from the kernel people (from several countries).
In short, I learned a lot, but I can still not garantee that this will be solved for your motherboard in whatever kernel version. You should try with a rather recent kernel and when needed file a bug. That is how it works and not by clairvoyance.
IMHO the best way to approach this, if one is fuzzy as to whether there has or has not been a bug report related to hardware that one owns, is to boot to the liveCD, and if the behaviour is NOT appropriate , write a bug report. Then follow up on the bug report, doing tests if asked to do so.
I did not suggest that I would seek a different system; merely that other people starting OpenSUSE for the first time might be put off by an installation that left them with a black screen.
In my case I found a solution that unfortunately only uses one core, but otherwise provides me with an excellent system: those others might not be so lucky. Again in my case going to Windows would be trivial, apart from the fact I have spent the last five years trying to forget Windows.
I doubt if MicroSoft would be as helpfull as the community here help me or you with such a problem.
I am quite certain that Microsoft would be totally unhelpful — never having contacted them when I ran Windows — however, Windows Help forums seem to lack the intense defensiveness and emotional investment that afflicts Linux forums when any imperfect aspect of the OS is mentioned.
>
> hcvv;2181868 Wrote:
>> I had this in an HP system. It was reported as a regression against
>> the kernel. I am running kernel 2.6.33-rc2-0.1 in 11.2 now, which
>> fixed it. Thus I assume that it is also fixed inthe final 2.6.33 and
>> onwards.
>
>
> The above was posted immediately before I saw your very helpful post.
> I’ll look into that. Thank you.
>
>
> However, on a general note, I do think that many new-comers not being
> able to even install will just discard and go to Ubuntu, or Windows…
I thought you were talking about your installation only seeing one core,
so presumably you did manage to complete the install?
>
> DenverD;2181860 Wrote:
>> has a bug been logged against that motherboard? if not, it will never
>> be fixed…
>>
>> if a bug has been logged then you can go look in bugzilla and learn
>> the status of the fix…
>
> Oddly, I haven’t the faintest idea whether a bug report has been ever
> issued, nor do I have the competence or knowledge to file one myself:
> I was more in the nature of asking a general question as to whether
> 11.3 installation will be able to accommodate quad-cores without this
> issue.
I have 11.3RC1 running on an eight-CPU system, works very well.
> Since it is NOT my AMD or motherboard alone that has this problem, but
I also have 11.3RC1 on a quad-core AMD Phenom workstation, also no
problems.
> also Intel quad cores etc., as seen in a thread on this
> very forum, amongst other threads dealing with the same **** problem,
> it would be singularly inept to waste people’s time by filing a bug
> report about my Asus board when it probably has nothing to do with
> Asus and is specific to the OS. Or very possibly the kernel.
It would be singul… silly not to file a bug if you have even just
the slightest expectations of the problem getting solved.
On 06/27/2010 04:53 PM, Per Jessen wrote:
> Claverhouse wrote:
>
>>
>> DenverD;2181860 Wrote:
>>> has a bug been logged against that motherboard? if not, it will never
>>> be fixed…
>>>
>>> if a bug has been logged then you can go look in bugzilla and learn
>>> the status of the fix…
>>
>> Oddly, I haven’t the faintest idea whether a bug report has been ever
>> issued, nor do I have the competence or knowledge to file one myself:
>
> No competence nor knowledge required. http://bugzilla.novell.com/
>
>> I was more in the nature of asking a general question as to whether
>> 11.3 installation will be able to accommodate quad-cores without this
>> issue.
>
> I have 11.3RC1 running on an eight-CPU system, works very well.
>
>> Since it is NOT my AMD or motherboard alone that has this problem, but
>
> I also have 11.3RC1 on a quad-core AMD Phenom workstation, also no
> problems.
>
>> also Intel quad cores etc., as seen in a thread on this
>> very forum, amongst other threads dealing with the same **** problem,
>> it would be singularly inept to waste people’s time by filing a bug
>> report about my Asus board when it probably has nothing to do with
>> Asus and is specific to the OS. Or very possibly the kernel.
>
> It would be singul… silly not to file a bug if you have even just
> the slightest expectations of the problem getting solved.
Problems that involve bugs in software are never solved in these fora! You
must file a BUG report.
AFAICT, you likely have a BIOS problem. Your ACPI system is messed up, and
your situation is not one of the quirks that the kernel works around. Note
that ACPI off has the well-known side effect of running with only 1 cpu
enabled. Your only other option is to see if there is a newer BIOS than
the one you are using. The problem might have been fixed there.
Claverhouse wrote:
> however, Windows Help forums seem to lack the intense defensiveness
> and emotional investment that afflicts Linux forums when any
> imperfect aspect of the OS is mentioned.
This being an old story to me, I may not have shown or told all of it (I only illustrated the fact that defining and filing a bug can be done by a noob with help from the Forums). But it started with a 11.2 M release that would go in a boot loop (it loaded the kernel, the first ruler on the screen) and then did nothing for about 10 secs and then rebooted, nothing to see on the screen, Esc not even functioning). I succeeded in booting using the acpi or similar parameters (from the menu below) generaly advised on having such problems (do not know if I got hat advice directly from a question on the forums here, or just from being a regular visitor here and having read that using these params may help).
In any case, when the real 11.2 came out, It had the loop too (only on one of the two multi CPU system here, but they are different). So I used the parameter and all went fine. Only by incident I saw I was using but one CPU and then I started the thread about the “stolen CPU” (link above).
Again, I refered to my problem because:
it is similar to the OP’s problem and may be of use to him;
it serves as an example how this community helps and what the steps are when one encounters such a bug.