From: Shamir Rabinovitch <shamir.rabinovitch@oracle.com>
To: David Miller <davem@davemloft.net>
Cc: jroedel@suse.de, dwmw2@infradead.org, benh@kernel.crashing.org,
arnd@arndb.de, corbet@lwn.net, linux-doc@vger.kernel.org,
linux-arch@vger.kernel.org, luto@kernel.org,
borntraeger@de.ibm.com, cornelia.huck@de.ibm.com,
sebott@linux.vnet.ibm.com, pbonzini@redhat.com, hch@lst.de,
kvm@vger.kernel.org, schwidefsky@de.ibm.com,
linux-s390@vger.kernel.org
Subject: Re: [PATCH v1 2/2] dma-mapping-common: add DMA attribute - DMA_ATTR_IOMMU_BYPASS
Date: Sat, 7 Nov 2015 17:06:42 +0200 [thread overview]
Message-ID: <20151107150642.GA4405@shamir-ThinkPad-T430> (raw)
In-Reply-To: <20151105.161121.188431695871972310.davem@davemloft.net>
On Thu, Nov 05, 2015 at 04:11:21PM -0500, David Miller wrote:
>
> And for the record Sowmini fixed a lot of the lock contention:
>
> commit ff7d37a502022149655c18035b99a53391be0383
> Author: Sowmini Varadhan <sowmini.varadhan@oracle.com>
> Date: Thu Apr 9 15:33:30 2015 -0400
>
> Break up monolithic iommu table/lock into finer graularity pools and lock
>
The poor rds-stress results w/o IOMMU bypass I sent in early post were taken from
kernel that has the above patch and that has all the needed changes in arch/sparc
to use this new feature.
It seems that it worked well for 10G ETH IOMMU lock contention but it still not solving
the rds-stress issue.
The difference can be from:
1. Lock contention still left with this enhancement <-- zero in bypass
2. Overhead to setup the IOMMU mapping <-- almost zero in bypass (require 1 HV call)
3. Overhead to use the IOMMU mapping <-- not sure how to measure this
4. Overhead to tear the IOMMU mapping <-- zero in bypass
next prev parent reply other threads:[~2015-11-07 15:07 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-25 16:07 [PATCH v1 1/2] dma-mapping-common: add dma_map_page_attrs API Shamir Rabinovitch
2015-10-25 16:07 ` [PATCH v1 2/2] dma-mapping-common: add DMA attribute - DMA_ATTR_IOMMU_BYPASS Shamir Rabinovitch
2015-10-28 6:30 ` David Woodhouse
2015-10-28 11:10 ` Shamir Rabinovitch
2015-10-28 11:10 ` Shamir Rabinovitch
2015-10-28 13:31 ` David Woodhouse
2015-10-28 13:31 ` David Woodhouse
2015-10-28 14:07 ` David Miller
2015-10-28 13:57 ` David Woodhouse
2015-10-28 13:57 ` David Woodhouse
2015-10-29 0:23 ` David Miller
2015-10-29 0:23 ` David Miller
2015-10-28 14:07 ` David Miller
2015-10-29 0:32 ` Benjamin Herrenschmidt
2015-10-29 0:32 ` Benjamin Herrenschmidt
2015-10-29 0:42 ` David Woodhouse
2015-10-29 0:42 ` David Woodhouse
2015-10-29 1:10 ` Benjamin Herrenschmidt
2015-10-29 1:10 ` Benjamin Herrenschmidt
2015-10-29 18:31 ` Andy Lutomirski
2015-10-29 18:31 ` Andy Lutomirski
2015-10-29 22:35 ` David Woodhouse
2015-11-01 7:45 ` Shamir Rabinovitch
2015-11-01 7:45 ` Shamir Rabinovitch
2015-11-01 21:10 ` Benjamin Herrenschmidt
2015-11-01 21:10 ` Benjamin Herrenschmidt
2015-11-02 7:23 ` Shamir Rabinovitch
2015-11-02 10:00 ` Benjamin Herrenschmidt
2015-11-02 12:07 ` Shamir Rabinovitch
2015-11-02 20:13 ` Benjamin Herrenschmidt
2015-11-02 21:45 ` Arnd Bergmann
2015-11-02 21:45 ` Arnd Bergmann
2015-11-02 23:08 ` Benjamin Herrenschmidt
2015-11-02 23:08 ` Benjamin Herrenschmidt
2015-11-03 13:11 ` Christoph Hellwig
2015-11-03 13:11 ` Christoph Hellwig
2015-11-03 19:35 ` Benjamin Herrenschmidt
2015-11-03 19:35 ` Benjamin Herrenschmidt
2015-11-02 21:49 ` Shamir Rabinovitch
2015-11-02 22:48 ` David Woodhouse
2015-11-02 23:10 ` Benjamin Herrenschmidt
2015-11-02 23:10 ` Benjamin Herrenschmidt
2015-11-05 21:08 ` David Miller
2015-11-05 21:08 ` David Miller
2015-10-30 1:51 ` Benjamin Herrenschmidt
2015-10-30 1:51 ` Benjamin Herrenschmidt
2015-10-30 10:32 ` Arnd Bergmann
2015-10-30 10:32 ` Arnd Bergmann
2015-10-30 23:17 ` Benjamin Herrenschmidt
2015-10-30 23:17 ` Benjamin Herrenschmidt
2015-10-30 23:24 ` Arnd Bergmann
2015-11-02 14:51 ` Joerg Roedel
2015-10-29 7:32 ` Shamir Rabinovitch
2015-11-02 14:44 ` Joerg Roedel
2015-11-02 17:32 ` Shamir Rabinovitch
2015-11-02 17:32 ` Shamir Rabinovitch
2015-11-05 13:42 ` Joerg Roedel
2015-11-05 21:11 ` David Miller
2015-11-05 21:11 ` David Miller
2015-11-07 15:06 ` Shamir Rabinovitch [this message]
[not found] ` <CAN+hb0UvztgwNuAh93XdJEe7vgiZgNMc9mHNziHpEopg8Oi4Mg@mail.gmail.com>
2015-11-16 8:42 ` David Woodhouse
[not found] ` <CAN+hb0UWpfcS5DvgMxNjY-5JOztw2mO1r2FJAW17fn974mhxPA@mail.gmail.com>
2015-11-16 18:42 ` Benjamin Serebrin
2015-11-16 18:42 ` Benjamin Serebrin
-- strict thread matches above, loose matches on Subject: below --
2015-10-25 16:37 [PATCH v1 1/2] dma-mapping-common: add dma_map_page_attrs API Shamir Rabinovitch
2015-10-25 16:37 ` [PATCH v1 2/2] dma-mapping-common: add DMA attribute - DMA_ATTR_IOMMU_BYPASS Shamir Rabinovitch
2015-11-16 6:56 Benjamin Serebrin
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=20151107150642.GA4405@shamir-ThinkPad-T430 \
--to=shamir.rabinovitch@oracle.com \
--cc=arnd@arndb.de \
--cc=benh@kernel.crashing.org \
--cc=borntraeger@de.ibm.com \
--cc=corbet@lwn.net \
--cc=cornelia.huck@de.ibm.com \
--cc=davem@davemloft.net \
--cc=dwmw2@infradead.org \
--cc=hch@lst.de \
--cc=jroedel@suse.de \
--cc=kvm@vger.kernel.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=luto@kernel.org \
--cc=pbonzini@redhat.com \
--cc=schwidefsky@de.ibm.com \
--cc=sebott@linux.vnet.ibm.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