From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Jan Beulich <JBeulich@suse.com>,
"Anshul Makkar anshul.makkar"@citrix.com
Cc: yang.z.zhang@intel.com, Anshul Makkar <anshul.makkar@citrix.com>,
kevin.tian@intel.com, xen-devel@lists.xen.org
Subject: Re: [PATCH] iommu/quirk: disable shared EPT for Sandybridge and earlier processors.
Date: Wed, 25 Nov 2015 10:28:05 +0000 [thread overview]
Message-ID: <56558D35.2040800@citrix.com> (raw)
In-Reply-To: <5654AF4C02000078000B8A11@prv-mh.provo.novell.com>
On 24/11/15 17:41, Jan Beulich wrote:
>>>> On 24.11.15 at 18:17, <Anshul Makkar anshul.makkar@citrix.com> wrote:
>> --- a/xen/drivers/passthrough/vtd/quirks.c
>> +++ b/xen/drivers/passthrough/vtd/quirks.c
>> @@ -320,6 +320,20 @@ void __init platform_quirks_init(void)
>> /* Tylersburg interrupt remap quirk */
>> if ( iommu_intremap )
>> tylersburg_intremap_quirk();
>> +
>> + /*
>> + * Disable shared EPT ("sharept") on Sandybridge and older processors
>> + * by default.
>> + * SandyBridge has no huge page support for IOTLB which leads to fallback
>> + * on 4k pages and leads to performance degradation.
>> + *
>> + * Shared EPT ("sharept") will be disabled only if user has not
>> + * provided explicit choice on the command line thus iommu_hap_pt_share is
>> + * at its initialized value of -1.
>> + */
>> + if ( (boot_cpu_data.x86 == 0x06 && (boot_cpu_data.x86_model <= 0x2F ||
>> + boot_cpu_data.x86_model == 0x36)) && (iommu_hap_pt_share == -1) )
>> + iommu_hap_pt_share = 0;
> If we really want to do this, then I think we should key this on
> EPT but not VT-d having 2M support, instead of on CPU models.
This check is already performed by vtd_ept_page_compatible()
The problem is that SandyBridge IOMMUs advertise 2M support and do
function with it, but cannot cache 2MB translations in the IOTLBs.
As a result, attempting to use 2M translations causes substantially
worse performance than 4K translations.
~Andrew
next prev parent reply other threads:[~2015-11-25 10:28 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-11-24 17:17 [PATCH] iommu/quirk: disable shared EPT for Sandybridge and earlier processors Anshul
2015-11-24 17:41 ` Jan Beulich
2015-11-25 10:28 ` Andrew Cooper [this message]
2015-11-25 10:49 ` Jan Beulich
2015-11-25 15:13 ` Andrew Cooper
2015-11-25 15:38 ` Jan Beulich
2015-11-25 15:58 ` Malcolm Crossley
2015-11-26 7:17 ` Tian, Kevin
2015-12-01 16:45 ` Anshul Makkar
2015-12-01 17:20 ` Jan Beulich
2015-11-26 8:45 ` Jan Beulich
2015-11-26 10:27 ` Andrew Cooper
2015-11-26 10:39 ` Jan Beulich
2015-11-26 11:42 ` Andrew Cooper
2015-11-26 11:53 ` Jan Beulich
2015-11-26 13:46 ` Jan Beulich
2015-11-26 13:48 ` Malcolm Crossley
2015-11-26 13:55 ` Andrew Cooper
2015-11-30 21:22 ` Konrad Rzeszutek Wilk
2015-12-01 10:34 ` Andrew Cooper
2015-12-01 10:44 ` Anshul Makkar
2015-12-01 15:24 ` Konrad Rzeszutek Wilk
2015-12-01 16:19 ` Andrew Cooper
2015-12-03 1:19 ` Tian, Kevin
2015-12-03 11:24 ` Andrew Cooper
2015-12-04 1:55 ` Tian, Kevin
2015-12-03 2:40 ` Tian, Kevin
2015-12-03 8:18 ` Jan Beulich
2015-12-03 8:50 ` Tian, Kevin
2015-12-03 11:19 ` Andrew Cooper
2015-12-04 2:35 ` Tian, Kevin
[not found] <1440776507-30218-1-git-send-email-anshul.makkar@citrix.com>
2015-08-28 16:24 ` Andrew Cooper
2015-08-31 8:09 ` Jan Beulich
2015-09-01 14:18 ` Andrew Cooper
2015-09-01 14:55 ` Jan Beulich
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=56558D35.2040800@citrix.com \
--to=andrew.cooper3@citrix.com \
--cc="Anshul Makkar anshul.makkar"@citrix.com \
--cc=JBeulich@suse.com \
--cc=anshul.makkar@citrix.com \
--cc=kevin.tian@intel.com \
--cc=xen-devel@lists.xen.org \
--cc=yang.z.zhang@intel.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 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.