From: Brian King <brking@linux.vnet.ibm.com>
To: Rick Jones <rick.jones2@hp.com>
Cc: David Miller <davem@davemloft.net>,
"cascardo@linux.vnet.ibm.com" <cascardo@linux.vnet.ibm.com>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"yevgenyp@mellanox.co.il" <yevgenyp@mellanox.co.il>,
"ogerlitz@mellanox.com" <ogerlitz@mellanox.com>,
"amirv@mellanox.com" <amirv@mellanox.com>,
"leitao@linux.vnet.ibm.com" <leitao@linux.vnet.ibm.com>,
"klebers@linux.vnet.ibm.com" <klebers@linux.vnet.ibm.com>
Subject: Re: [PATCH] mlx4_en: map entire pages to increase throughput
Date: Tue, 17 Jul 2012 15:10:08 -0500 [thread overview]
Message-ID: <5005C6A0.50002@linux.vnet.ibm.com> (raw)
In-Reply-To: <5005AC4A.9030208@hp.com>
On 07/17/2012 01:17 PM, Rick Jones wrote:
> On 07/16/2012 10:29 PM, David Miller wrote:
>> From: Rick Jones <rick.jones2@hp.com> Date: Mon, 16 Jul 2012
>> 10:27:57 -0700
>>
>>> That seems rather extraordinarily low - Power7 is supposed to be
>>> a rather high performance CPU. The last time I noticed
>>> O(3Gbit/s) on 10G for bulk transfer was before the advent of
>>> LRO/GRO - that was in the x86 space though. Is mapping really
>>> that expensive with Power7?
>>
>> Unfortunately, IOMMU mappings are incredibly expensive. I see
>> effects like this on Sparc too.
>
> OK, so that has caused some dimm memory to get a small refresh - it
> ends up being akin to if not actually a PIO yes? I recall schemes in
> drivers in other stacks whereby "small" packets were copied because
> it was cheaper to allocate/copy then it was to remap.
On Power it ends up being an hcall to the hypervisor
-Brian
--
Brian King
Power Linux I/O
IBM Linux Technology Center
next prev parent reply other threads:[~2012-07-17 20:11 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-07-16 17:01 [PATCH] mlx4_en: map entire pages to increase throughput Thadeu Lima de Souza Cascardo
2012-07-16 17:27 ` Rick Jones
2012-07-16 19:06 ` Thadeu Lima de Souza Cascardo
2012-07-16 19:42 ` Rick Jones
2012-07-16 20:36 ` Or Gerlitz
2012-07-16 20:43 ` Or Gerlitz
2012-07-16 20:57 ` Thadeu Lima de Souza Cascardo
2012-07-18 14:59 ` Or Gerlitz
2012-07-16 20:47 ` Thadeu Lima de Souza Cascardo
2012-07-16 21:08 ` Rick Jones
2012-07-17 5:29 ` David Miller
2012-07-17 12:42 ` David Laight
2012-07-17 12:50 ` David Miller
2012-07-17 13:36 ` David Laight
2012-07-17 13:46 ` David Miller
2012-07-17 13:50 ` Eric Dumazet
2012-07-17 18:17 ` Rick Jones
2012-07-17 20:10 ` Brian King [this message]
2012-07-17 20:20 ` David Miller
2012-07-19 17:53 ` David Miller
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=5005C6A0.50002@linux.vnet.ibm.com \
--to=brking@linux.vnet.ibm.com \
--cc=amirv@mellanox.com \
--cc=cascardo@linux.vnet.ibm.com \
--cc=davem@davemloft.net \
--cc=klebers@linux.vnet.ibm.com \
--cc=leitao@linux.vnet.ibm.com \
--cc=netdev@vger.kernel.org \
--cc=ogerlitz@mellanox.com \
--cc=rick.jones2@hp.com \
--cc=yevgenyp@mellanox.co.il \
/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).