[Gllug] HugePages

[James Courtier-Dutton]

On 28 October 2010 19:54, Richard Jones <rich at annexia.org> wrote:
> On Thu, Oct 28, 2010 at 07:06:45PM +0100, James Courtier-Dutton wrote:
>> My understanding of hugepages is that you only use them if you
>> really have to.  The reason to have hugepages is that the x86/x64
>> CPU has a fixed size page table. I.e. How many pages it can refer
>> to.  If you have more physical memory than the CPU can reference
>> with normal sized pages, one has to then use hugepages.
>
> The above is a load of drivel from beginning to end.
>
> Can't help with the original problem unfortunately.  It sounds very
> specific to the machine and I'm afraid you need to investigate it from
> first principles on that machine.  Luckily it sounds like you have a
> good, reliable reproducer (ie. 'ls' and other commands taking a long
> time to run).  Does the situation improve if you give reduce the
> hugepage allocation by a smallish amount, say, 1 or 2 GB?
>
> RHEL 6 and upstream have a wonderful feature called Transparent
> Hugepages which makes all this business of statically allocating a
> structure at boot time go away.  Here is Andrea's talk from the KVM
> Forum:
>

And your post is about as useful as a wet fish.
-- 
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug