Correct me if I'm wrong, but current_cstate should never be greater than
supported_max_cstates, right?

 

There are a lot of reports out there, of systems crashing, due to problems
with cstate, and generally people go into BIOS and disable cstates.  Call it
a "workaround."  But it would be nice if the actual problem were addressed.
You know.  So systems could be stable without disabling things in BIOS.  ;-)

 

I have a Dell R710 which is unstable running solaris 10u9.  It crashes
approx once per week, but it varies dramatically.  While digging around, I
noticed that the current_cstate is generally higher than
supported_max_cstates.  That seems like a bad thing, and could possibly
explain the instability.

 

kstat | grep current_cstate ; kstat | grep supported_max_cstates

 

I was easily able to find the meaning of current_cstate, but didn't find the
meaning of supported_max_cstates, for the sake of verifying that it means
what I think it means.

 

I also have a Sun machine, which is stable.  I checked, and it never has a
cstate higher than supported_max_cstates.  So that makes 2 data points which
both support the idea that current_cstate greater than supported_max_cstates
would be an indicator of system instability.

 

Is this mailing list a good place to reach out to kernel developers, or
whoever would be appropriate?

 

Here's what I see on the unstable system:

 

kstat | grep current_cstate ; kstat | grep supported_max_cstates

        current_cstate                  0

        current_cstate                  0

        current_cstate                  0

        current_cstate                  1

        current_cstate                  3

        current_cstate                  3

        current_cstate                  0

        current_cstate                  3

        current_cstate                  3

        current_cstate                  3

        current_cstate                  3

        current_cstate                  1

        current_cstate                  3

        current_cstate                  3

        current_cstate                  3

        current_cstate                  3

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

        supported_max_cstates           2

_______________________________________________
opensolaris-code mailing list
opensolaris-code@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code

Reply via email to