Messages in this thread | | | Date | Wed, 20 Mar 2013 20:21:44 +0100 | From | Michal Hocko <> | Subject | Re: [patch] mm, hugetlb: include hugepages in meminfo |
| |
On Wed 20-03-13 11:58:20, David Rientjes wrote: > On Wed, 20 Mar 2013, Michal Hocko wrote: > > > > I didn't do this because it isn't already exported in /proc/meminfo and > > > since we've made an effort to reduce the amount of information emitted by > > > the oom killer at oom kill time to avoid spamming the kernel log, I only > > > print the default hstate. > > > > I do not see how this would make the output too much excessive. If > > you do not want to have too many lines in the output then the hstate > > loop can be pushed inside the node loop and have only per-node number > > of lines same as you are proposing except you would have a complete > > information. > > Besides that we are talking about handful of hstates. > > > > Sigh. Because nobody is going to be mapping non-default hstates and then > not know about them at oom time;
If you are under control of the machine then you are right. But I was already handling issues where getting any piece of information was challenging and having this kind of information in the log would save me a lot of time.
> 1GB hugepages on x86 with pse must be reserved at boot and never > freed, for example. I'll add them but it's just a waste of time.
If you feel it is the waste of _your_ time then I am OK to create a folow up patch. I really do not see any reason to limit this output, especially when it doesn't cost us much.
-- Michal Hocko SUSE Labs
| |