From: Andi Kleen <andi@firstfloor.org>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
David Daney <ddaney@caviumnetworks.com>,
David Daney <ddaney.cavm@gmail.com>,
David Rientjes <rientjes@google.com>,
"linux-mips@linux-mips.org" <linux-mips@linux-mips.org>,
"ralf@linux-mips.org" <ralf@linux-mips.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
David Daney <david.daney@cavium.com>,
"linux-arch@vger.kernel.org" <linux-arch@vger.kernel.org>,
Robin Holt <holt@sgi.com>
Subject: Re: [patch] hugetlb: remove dummy definitions of HPAGE_MASK and HPAGE_SIZE
Date: Tue, 22 Nov 2011 12:41:55 -0800 [thread overview]
Message-ID: <m2sjlfvqi4.fsf@firstfloor.org> (raw)
In-Reply-To: <20111121155024.33b1b881.akpm@linux-foundation.org> (Andrew Morton's message of "Mon, 21 Nov 2011 15:50:24 -0800")
Andrew Morton <akpm@linux-foundation.org> writes:
>>
>> I wish people whose code had stuff like that would take a deep look at it.
>>
>
> The original decision way back when was that huge pages shouldn't mess
> up the core VM too much. One way in which we addressed that was to
IMHO this decision should really be revisited now. Originally hugetlb
was pretty simple, but these days it has most of the functionality of a
full VM now. In fact with THP we have 3 different VM systems now, all
subtle different with different issues. And with THP hugetlb will be
even more widely used than it is today.
On the other hand there are strange gaps now, like shared memory
doesn't work with THP, but only with hugetlbfs.
It would be far better to think about unifying these three VMs. Then with
less ifdefs it would also not need hacks like this anymore.
-Andi
--
ak@linux.intel.com -- Speaking for myself only
WARNING: multiple messages have this Message-ID (diff)
From: Andi Kleen <andi@firstfloor.org>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
David Daney <ddaney@caviumnetworks.com>,
David Daney <ddaney.cavm@gmail.com>,
David Rientjes <rientjes@google.com>,
"linux-mips\@linux-mips.org" <linux-mips@linux-mips.org>,
"ralf\@linux-mips.org" <ralf@linux-mips.org>,
"linux-kernel\@vger.kernel.org" <linux-kernel@vger.kernel.org>,
David Daney <david.daney@cavium.com>,
"linux-arch\@vger.kernel.org" <linux-arch@vger.kernel.org>,
Robin Holt <holt@sgi.com>
Subject: Re: [patch] hugetlb: remove dummy definitions of HPAGE_MASK and HPAGE_SIZE
Date: Tue, 22 Nov 2011 12:41:55 -0800 [thread overview]
Message-ID: <m2sjlfvqi4.fsf@firstfloor.org> (raw)
In-Reply-To: <20111121155024.33b1b881.akpm@linux-foundation.org> (Andrew Morton's message of "Mon, 21 Nov 2011 15:50:24 -0800")
Andrew Morton <akpm@linux-foundation.org> writes:
>>
>> I wish people whose code had stuff like that would take a deep look at it.
>>
>
> The original decision way back when was that huge pages shouldn't mess
> up the core VM too much. One way in which we addressed that was to
IMHO this decision should really be revisited now. Originally hugetlb
was pretty simple, but these days it has most of the functionality of a
full VM now. In fact with THP we have 3 different VM systems now, all
subtle different with different issues. And with THP hugetlb will be
even more widely used than it is today.
On the other hand there are strange gaps now, like shared memory
doesn't work with THP, but only with hugetlbfs.
It would be far better to think about unifying these three VMs. Then with
less ifdefs it would also not need hacks like this anymore.
-Andi
--
ak@linux.intel.com -- Speaking for myself only
next prev parent reply other threads:[~2011-11-22 20:41 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-11-17 21:57 [PATCH v2 0/2] Dummy HPAGE_* constants for !CONFIG_HUGETLB_PAGE David Daney
2011-11-17 21:57 ` [PATCH v2 1/2] hugetlb: Provide a default HPAGE_SHIFT if !CONFIG_HUGETLB_PAGE David Daney
2011-11-17 21:57 ` [PATCH v2 2/2] hugetlb: Provide safer dummy values for HPAGE_MASK and HPAGE_SIZE David Daney
2011-11-17 23:28 ` Andrew Morton
2011-11-17 23:38 ` David Daney
2011-11-18 8:56 ` Sergei Shtylyov
2011-11-18 17:14 ` David Daney
2011-11-17 23:22 ` [PATCH v2 0/2] Dummy HPAGE_* constants for !CONFIG_HUGETLB_PAGE David Rientjes
2011-11-17 23:22 ` [patch] hugetlb: remove dummy definitions of HPAGE_MASK and HPAGE_SIZE David Rientjes
2011-11-17 23:35 ` Andrew Morton
2011-11-17 23:44 ` David Rientjes
2011-11-17 23:52 ` David Daney
2011-11-17 23:57 ` Andrew Morton
2011-11-17 23:57 ` David Rientjes
2011-11-21 22:23 ` David Daney
2011-11-21 22:43 ` Linus Torvalds
2011-11-21 23:23 ` David Daney
2011-11-21 23:43 ` Linus Torvalds
2011-11-21 23:50 ` Andrew Morton
2011-11-22 20:41 ` Andi Kleen [this message]
2011-11-22 20:41 ` Andi Kleen
2011-11-21 23:47 ` David Daney
2011-11-21 23:53 ` Andrew Morton
2011-11-22 0:37 ` Linus Torvalds
2011-11-22 0:48 ` David Daney
2011-11-22 0:55 ` Linus Torvalds
2011-11-21 22:48 ` David Rientjes
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=m2sjlfvqi4.fsf@firstfloor.org \
--to=andi@firstfloor.org \
--cc=akpm@linux-foundation.org \
--cc=david.daney@cavium.com \
--cc=ddaney.cavm@gmail.com \
--cc=ddaney@caviumnetworks.com \
--cc=holt@sgi.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mips@linux-mips.org \
--cc=ralf@linux-mips.org \
--cc=rientjes@google.com \
--cc=torvalds@linux-foundation.org \
/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.