From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Alexander Graf <agraf@suse.de>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
"paulus@samba.org" <paulus@samba.org>,
"linuxppc-dev@lists.ozlabs.org" <linuxppc-dev@lists.ozlabs.org>,
"kvm-ppc@vger.kernel.org" <kvm-ppc@vger.kernel.org>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>
Subject: Re: [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page tab
Date: Tue, 06 May 2014 00:06:37 +0000 [thread overview]
Message-ID: <1399334797.20388.71.camel@pasglop> (raw)
In-Reply-To: <20FFDF8F-1A3D-4719-B492-1E4B70F9D1B4@suse.de>
On Mon, 2014-05-05 at 17:16 +0200, Alexander Graf wrote:
> Isn't this a greater problem? We should start swapping before we hit
> the point where non movable kernel allocation fails, no?
Possibly but the fact remains, this can be avoided by making sure that
if we create a CMA reserve for KVM, then it uses it rather than using
the rest of main memory for hash tables.
> The fact that KVM uses a good number of normal kernel pages is maybe
> suboptimal, but shouldn't be a critical problem.
The point is that we explicitly reserve those pages in CMA for use
by KVM for that specific purpose, but the current code tries first
to get them out of the normal pool.
This is not an optimal behaviour and is what Aneesh patches are
trying to fix.
Cheers,
Ben.
WARNING: multiple messages have this Message-ID (diff)
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Alexander Graf <agraf@suse.de>
Cc: "linuxppc-dev@lists.ozlabs.org" <linuxppc-dev@lists.ozlabs.org>,
"paulus@samba.org" <paulus@samba.org>,
"Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
"kvm-ppc@vger.kernel.org" <kvm-ppc@vger.kernel.org>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>
Subject: Re: [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table.
Date: Tue, 06 May 2014 10:06:37 +1000 [thread overview]
Message-ID: <1399334797.20388.71.camel@pasglop> (raw)
In-Reply-To: <20FFDF8F-1A3D-4719-B492-1E4B70F9D1B4@suse.de>
On Mon, 2014-05-05 at 17:16 +0200, Alexander Graf wrote:
> Isn't this a greater problem? We should start swapping before we hit
> the point where non movable kernel allocation fails, no?
Possibly but the fact remains, this can be avoided by making sure that
if we create a CMA reserve for KVM, then it uses it rather than using
the rest of main memory for hash tables.
> The fact that KVM uses a good number of normal kernel pages is maybe
> suboptimal, but shouldn't be a critical problem.
The point is that we explicitly reserve those pages in CMA for use
by KVM for that specific purpose, but the current code tries first
to get them out of the normal pool.
This is not an optimal behaviour and is what Aneesh patches are
trying to fix.
Cheers,
Ben.
WARNING: multiple messages have this Message-ID (diff)
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Alexander Graf <agraf@suse.de>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
"paulus@samba.org" <paulus@samba.org>,
"linuxppc-dev@lists.ozlabs.org" <linuxppc-dev@lists.ozlabs.org>,
"kvm-ppc@vger.kernel.org" <kvm-ppc@vger.kernel.org>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>
Subject: Re: [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table.
Date: Tue, 06 May 2014 10:06:37 +1000 [thread overview]
Message-ID: <1399334797.20388.71.camel@pasglop> (raw)
In-Reply-To: <20FFDF8F-1A3D-4719-B492-1E4B70F9D1B4@suse.de>
On Mon, 2014-05-05 at 17:16 +0200, Alexander Graf wrote:
> Isn't this a greater problem? We should start swapping before we hit
> the point where non movable kernel allocation fails, no?
Possibly but the fact remains, this can be avoided by making sure that
if we create a CMA reserve for KVM, then it uses it rather than using
the rest of main memory for hash tables.
> The fact that KVM uses a good number of normal kernel pages is maybe
> suboptimal, but shouldn't be a critical problem.
The point is that we explicitly reserve those pages in CMA for use
by KVM for that specific purpose, but the current code tries first
to get them out of the normal pool.
This is not an optimal behaviour and is what Aneesh patches are
trying to fix.
Cheers,
Ben.
next prev parent reply other threads:[~2014-05-06 0:06 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-04 17:25 [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table Aneesh Kumar K.V
2014-05-04 17:37 ` Aneesh Kumar K.V
2014-05-04 17:25 ` Aneesh Kumar K.V
2014-05-05 11:26 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page tab Alexander Graf
2014-05-05 11:26 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table Alexander Graf
2014-05-05 11:26 ` Alexander Graf
2014-05-05 14:35 ` Aneesh Kumar K.V
2014-05-05 14:47 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page tab Aneesh Kumar K.V
2014-05-05 14:35 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table Aneesh Kumar K.V
2014-05-05 15:16 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page tab Alexander Graf
2014-05-05 15:16 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table Alexander Graf
2014-05-05 15:16 ` Alexander Graf
2014-05-05 15:40 ` Aneesh Kumar K.V
2014-05-05 15:52 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page tab Aneesh Kumar K.V
2014-05-05 15:40 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table Aneesh Kumar K.V
2014-05-06 0:06 ` Benjamin Herrenschmidt [this message]
2014-05-06 0:06 ` Benjamin Herrenschmidt
2014-05-06 0:06 ` Benjamin Herrenschmidt
2014-05-06 7:05 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page tab Alexander Graf
2014-05-06 7:05 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table Alexander Graf
2014-05-06 7:05 ` Alexander Graf
2014-05-06 7:19 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page tab Benjamin Herrenschmidt
2014-05-06 7:19 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table Benjamin Herrenschmidt
2014-05-06 7:19 ` Benjamin Herrenschmidt
2014-05-06 7:21 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page tab Alexander Graf
2014-05-06 7:21 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table Alexander Graf
2014-05-06 7:21 ` Alexander Graf
2014-05-06 14:20 ` Aneesh Kumar K.V
2014-05-06 14:32 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page tab Aneesh Kumar K.V
2014-05-06 14:20 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table Aneesh Kumar K.V
2014-05-06 14:25 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page tab Alexander Graf
2014-05-06 14:25 ` [PATCH] KVM: PPC: BOOK3S: HV: Don't try to allocate from kernel page allocator for hash page table Alexander Graf
2014-05-06 14:25 ` Alexander Graf
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=1399334797.20388.71.camel@pasglop \
--to=benh@kernel.crashing.org \
--cc=agraf@suse.de \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=kvm-ppc@vger.kernel.org \
--cc=kvm@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=paulus@samba.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.