From: Andrea Arcangeli <aarcange@redhat.com>
To: Christoph Lameter <cl@linux-foundation.org>
Cc: Rik van Riel <riel@redhat.com>,
linux-mm@kvack.org, Marcelo Tosatti <mtosatti@redhat.com>,
Adam Litke <agl@us.ibm.com>, Avi Kivity <avi@redhat.com>,
Izik Eidus <ieidus@redhat.com>,
Hugh Dickins <hugh.dickins@tiscali.co.uk>,
Nick Piggin <npiggin@suse.de>, Mel Gorman <mel@csn.ul.ie>,
Andi Kleen <andi@firstfloor.org>,
Dave Hansen <dave@linux.vnet.ibm.com>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Ingo Molnar <mingo@elte.hu>, Mike Travis <travis@sgi.com>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Chris Wright <chrisw@sous-sol.org>,
Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [PATCH 00 of 30] Transparent Hugepage support #3
Date: Tue, 26 Jan 2010 17:16:25 +0100 [thread overview]
Message-ID: <20100126161625.GO30452@random.random> (raw)
In-Reply-To: <alpine.DEB.2.00.1001260947580.23549@router.home>
On Tue, Jan 26, 2010 at 09:54:59AM -0600, Christoph Lameter wrote:
> Huge pages are already in use through hugetlbs for such workloads. That
> works without swap. So why is this suddenly such a must have requirement?
hugetlbfs is unusable when you're not doing a static alloc for 1 DBMS
in 1 machine with alloc size set in a config file that will then match
grub command line.
> Why not swap 2M huge pages as a whole?
That is nice thing to speedup swap bandwidth and reduce fragmentation,
just I couldn't make so many changes in one go. Later we can make this
change and remove a few split_huge_page from the rmap paths.
> What in your workload forces hugetlb swap use? Just leaving a certain
> percentage of memory for 4k pages addresses the issue right now.
hypervisor must be able to swap, furthermore when a VM exists we want
to be able to use that ram as pagecache (not to remain reserved in
some hugetlbfs). And we must be able to fallback to 4k allocations
always without userland being able to notice when unable to defrag,
all things hugetlbfs can't do. All designs that can't 100% fallback to
4k allocations are useless in my view as far as you want to keep the
word "transparent" in the description of the patch...
--
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>
next prev parent reply other threads:[~2010-01-26 16:17 UTC|newest]
Thread overview: 79+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-21 6:20 [PATCH 00 of 30] Transparent Hugepage support #3 Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 01 of 30] define MADV_HUGEPAGE Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 02 of 30] compound_lock Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 03 of 30] alter compound get_page/put_page Andrea Arcangeli
2010-01-21 17:35 ` Dave Hansen
2010-01-23 17:39 ` Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 04 of 30] clear compound mapping Andrea Arcangeli
2010-01-21 17:43 ` Dave Hansen
2010-01-23 17:55 ` Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 05 of 30] add native_set_pmd_at Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 06 of 30] add pmd paravirt ops Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 07 of 30] no paravirt version of pmd ops Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 08 of 30] export maybe_mkwrite Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 09 of 30] comment reminder in destroy_compound_page Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 10 of 30] config_transparent_hugepage Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 11 of 30] add pmd mangling functions to x86 Andrea Arcangeli
2010-01-21 17:47 ` Dave Hansen
2010-01-21 19:14 ` Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 12 of 30] add pmd mangling generic functions Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 13 of 30] special pmd_trans_* functions Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 14 of 30] bail out gup_fast on splitting pmd Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 15 of 30] pte alloc trans splitting Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 16 of 30] add pmd mmu_notifier helpers Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 17 of 30] clear page compound Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 18 of 30] add pmd_huge_pte to mm_struct Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 19 of 30] ensure mapcount is taken on head pages Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 20 of 30] split_huge_page_mm/vma Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 21 of 30] split_huge_page paging Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 22 of 30] pmd_trans_huge migrate bugcheck Andrea Arcangeli
2010-01-21 20:40 ` Christoph Lameter
2010-01-21 23:01 ` Andrea Arcangeli
2010-01-21 23:17 ` Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 23 of 30] clear_copy_huge_page Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 24 of 30] kvm mmu transparent hugepage support Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 25 of 30] transparent hugepage core Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 26 of 30] madvise(MADV_HUGEPAGE) Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 27 of 30] memcg compound Andrea Arcangeli
2010-01-21 7:07 ` KAMEZAWA Hiroyuki
2010-01-21 15:44 ` Andrea Arcangeli
2010-01-21 23:55 ` KAMEZAWA Hiroyuki
2010-01-21 6:20 ` [PATCH 28 of 30] memcg huge memory Andrea Arcangeli
2010-01-21 7:16 ` KAMEZAWA Hiroyuki
2010-01-21 16:08 ` Andrea Arcangeli
2010-01-22 0:13 ` KAMEZAWA Hiroyuki
2010-01-27 11:27 ` Balbir Singh
2010-01-28 0:50 ` Daisuke Nishimura
2010-01-28 11:39 ` Andrea Arcangeli
2010-01-28 12:23 ` Balbir Singh
2010-01-28 12:36 ` Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 29 of 30] transparent hugepage vmstat Andrea Arcangeli
2010-01-21 6:20 ` [PATCH 30 of 30] khugepaged Andrea Arcangeli
2010-01-22 14:46 ` [PATCH 00 of 30] Transparent Hugepage support #3 Christoph Lameter
2010-01-22 15:19 ` Andrea Arcangeli
2010-01-22 16:51 ` Christoph Lameter
2010-01-23 17:58 ` Andrea Arcangeli
2010-01-25 21:50 ` Christoph Lameter
2010-01-25 22:46 ` Andrea Arcangeli
2010-01-26 15:47 ` Christoph Lameter
2010-01-26 16:11 ` Andrea Arcangeli
2010-01-26 16:30 ` Christoph Lameter
2010-01-26 16:45 ` Andrea Arcangeli
2010-01-26 18:23 ` Christoph Lameter
2010-01-26 17:09 ` Avi Kivity
2010-01-26 0:52 ` Rik van Riel
2010-01-26 6:53 ` Gleb Natapov
2010-01-26 12:35 ` Andrea Arcangeli
2010-01-26 15:55 ` Christoph Lameter
2010-01-26 16:19 ` Andrea Arcangeli
2010-01-26 15:54 ` Christoph Lameter
2010-01-26 16:16 ` Andrea Arcangeli [this message]
2010-01-26 16:24 ` Andi Kleen
2010-01-26 16:37 ` Christoph Lameter
2010-01-26 16:42 ` Mel Gorman
2010-01-26 16:52 ` Andrea Arcangeli
2010-01-26 17:26 ` Mel Gorman
2010-01-26 19:46 ` Andrea Arcangeli
2010-01-26 23:07 ` Rik van Riel
2010-01-27 18:33 ` Christoph Lameter
2010-01-26 11:24 ` Mel Gorman
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=20100126161625.GO30452@random.random \
--to=aarcange@redhat.com \
--cc=agl@us.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=avi@redhat.com \
--cc=benh@kernel.crashing.org \
--cc=chrisw@sous-sol.org \
--cc=cl@linux-foundation.org \
--cc=dave@linux.vnet.ibm.com \
--cc=hugh.dickins@tiscali.co.uk \
--cc=ieidus@redhat.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-mm@kvack.org \
--cc=mel@csn.ul.ie \
--cc=mingo@elte.hu \
--cc=mtosatti@redhat.com \
--cc=npiggin@suse.de \
--cc=riel@redhat.com \
--cc=travis@sgi.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).