All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nishanth Aravamudan <nacc@us.ibm.com>
To: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Cc: wli@holomorphy.com, anton@samba.org, clameter@sgi.com,
	akpm@linux-foundation.org, agl@us.ibm.com, linux-mm@kvack.org
Subject: Re: [RFC][PATCH 2/3] hugetlb: numafy several functions
Date: Wed, 23 May 2007 12:29:51 -0700	[thread overview]
Message-ID: <20070523192951.GE9301@us.ibm.com> (raw)
In-Reply-To: <1179947768.5537.37.camel@localhost>

On 23.05.2007 [15:16:07 -0400], Lee Schermerhorn wrote:
> On Wed, 2007-05-23 at 10:51 -0700, Nishanth Aravamudan wrote:
> > On 16.05.2007 [16:31:55 -0700], Nishanth Aravamudan wrote:
> > > Add node-parameterized helpers for dequeue_huge_page,
> > > alloc_fresh_huge_page and try_to_free_low. Also have
> > > update_and_free_page() take a nid parameter. This is necessary to add a
> > > per-node sysfs attribute to specify the number of hugepages on that
> > > node.
> > 
> > I saw that 1/3 was picked up by Andrew, but have not got any responses
> > to the other two (I know Adam is out of town...).
> 
> Nish:  I haven't had a chance to test these patches.  Other alligators
> in the swamp right now.

No problem.

> > Thoughts, comments? Bad idea, good idea?
> > 
> > I found it pretty handy to specify the exact layout of hugepages on each
> > node.
> 
> Could be useful for system with unequal memory per node, or where you
> know you want more huge pages on a given node.  I recall that Tru64 Unix
> used to support something similar:  most vm tunables that involved sizes
> or percentages of memory, such as page cache limits, locked memory
> limits, reserved huge pages, ..., could be specified as a single value
> that was distributed across nodes [backwards compatibility] or as list
> of per node values.  However, I don't recall if marketing/customers
> asked for this or if it was a case of gratuitous design excess ;-).

Yep, exactly the kind of use cases I was thinking of.

> I see that we'll need to reconcile the modified alloc_fresh_huge_page
> with the patch to skip unpopulated nodes when/if they collide in -mm.

Yeah, if folks like the interface and are satisfied with it working,
I'll rebase onto -mm for Andrew's sanity.

Thanks,
Nish

-- 
Nishanth Aravamudan <nacc@us.ibm.com>
IBM Linux Technology Center

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2007-05-23 19:29 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-05-16 23:30 [PATCH 1/3] hugetlb: remove unnecessary nid initialization Nishanth Aravamudan
2007-05-16 23:31 ` [RFC][PATCH 2/3] hugetlb: numafy several functions Nishanth Aravamudan
2007-05-16 23:33   ` [RFC][PATCH 3/3] hugetlb: add per-node nr_hugepages sysfs attribute Nishanth Aravamudan
2007-05-23 17:51   ` [RFC][PATCH 2/3] hugetlb: numafy several functions Nishanth Aravamudan
2007-05-23 19:16     ` Lee Schermerhorn
2007-05-23 19:29       ` Nishanth Aravamudan [this message]
2007-05-25 19:43         ` Nishanth Aravamudan
2007-05-25 20:55           ` Lee Schermerhorn
2007-05-25 21:10             ` Nishanth Aravamudan
2007-05-25 21:39             ` Andrew Morton
2007-05-25 21:47               ` Lee Schermerhorn

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20070523192951.GE9301@us.ibm.com \
    --to=nacc@us.ibm.com \
    --cc=Lee.Schermerhorn@hp.com \
    --cc=agl@us.ibm.com \
    --cc=akpm@linux-foundation.org \
    --cc=anton@samba.org \
    --cc=clameter@sgi.com \
    --cc=linux-mm@kvack.org \
    --cc=wli@holomorphy.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.