From: "Chen, Tiejun" <tiejun.chen@intel.com>
To: Tamas K Lengyel <tamas.k.lengyel@gmail.com>
Cc: Kevin Tian <kevin.tian@intel.com>,
xen-devel@lists.xensource.com, Wei Liu <wei.liu2@citrix.com>,
Ian Campbell <ian.campbell@citrix.com>,
George Dunlap <George.Dunlap@eu.citrix.com>,
Ian Jackson <ian.jackson@eu.citrix.com>,
Jan Beulich <JBeulich@suse.com>,
Yang Zhang <yang.z.zhang@intel.com>
Subject: Re: [PATCH 15/16] xen/vtd: prevent from assign the device with shared rmrr
Date: Sun, 6 Sep 2015 12:19:49 +0800 [thread overview]
Message-ID: <55EBBEE5.2090700@intel.com> (raw)
In-Reply-To: <CABfawh=+y-CoLX0AHOONPZ=KpwrDvB2D243giwX6rj7Q7tDQ8Q@mail.gmail.com>
On 9/6/2015 11:19 AM, Tamas K Lengyel wrote:
>>
>> diff --git a/xen/drivers/passthrough/vtd/iommu.c
>> b/xen/drivers/passthrough/vtd/iommu.c
>> index 836aed5..038776a 100644
>> --- a/xen/drivers/passthrough/vtd/iommu.c
>> +++ b/xen/drivers/passthrough/vtd/iommu.c
>> @@ -2310,12 +2310,16 @@ static int intel_iommu_assign_device(
>> PCI_DEVFN2(bdf) == devfn &&
>> rmrr->scope.devices_cnt > 1 )
>> {
>> + u32 relaxed = flag & XEN_DOMCTL_DEV_RDM_RELAXED;
>> +
>> printk(XENLOG_G_ERR VTDPREFIX
>> - " cannot assign %04x:%02x:%02x.%u"
>> + " Currently its %s to assign %04x:%02x:%02x.%u"
>> " with shared RMRR at %"PRIx64" for Dom%d.\n",
>> + relaxed ? "disallowed" : "risky",
>>
>
> This debug message is backwards?
Yeah. Its indeed like this, relaxed ? "risky" : "disallowed"
But lets wait Jan's comment to step next.
>
>
>> seg, bus, PCI_SLOT(devfn), PCI_FUNC(devfn),
>> rmrr->base_address, d->domain_id);
>> - return -EPERM;
>> + if ( !relaxed )
>> + return -EPERM;
>> }
>> }
>>
>>
>> Tamas, do you actually mean to assign these to _different_
>>>> guests, considering the log fragment above?)
>>>>
>>>>
>>> No, I actually want to assign them to the same domain. The domain creation
>>> fails with either of those devices specified for passthrough whether they
>>> are to be attached to the same domain or not.
>>>
>>>
>> Tamas, could you try this in your case?
>>
>
> Took me a while to find the xl config option to set this flag (pci = [
> 'sbdf, rdm_policy=strict/relaxed' ]) but now it works as expected!
>
I remember 'relaxed' is a default value so 'rdm_policy' can't be dropped
here if you like this.
Thanks
Tiejun
next prev parent reply other threads:[~2015-09-06 4:19 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-22 15:44 [PATCH v13 00/16] Fix RMRR (avoid RDM) Ian Jackson
2015-07-22 15:44 ` [PATCH 01/16] introduce XENMEM_reserved_device_memory_map Ian Jackson
2015-07-22 15:44 ` [PATCH 02/16] xen/vtd: create RMRR mapping Ian Jackson
2015-07-22 15:44 ` [PATCH 03/16] xen/passthrough: extend hypercall to support rdm reservation policy Ian Jackson
2015-07-23 11:45 ` Ian Jackson
2015-07-23 11:54 ` Jan Beulich
2015-07-23 11:54 ` Ian Campbell
2015-07-23 12:15 ` Chen, Tiejun
2015-07-23 12:19 ` Ian Jackson
2015-07-23 12:27 ` Ian Campbell
2015-07-23 12:40 ` Chen, Tiejun
2015-07-23 12:43 ` Ian Jackson
2015-07-22 15:44 ` [PATCH 04/16] xen: enable XENMEM_memory_map in hvm Ian Jackson
2015-07-22 15:44 ` [PATCH 05/16] hvmloader: get guest memory map into memory_map[] Ian Jackson
2015-07-22 15:44 ` [PATCH 06/16] hvmloader/pci: try to avoid placing BARs in RMRRs Ian Jackson
2015-07-22 15:44 ` [PATCH 07/16] hvmloader/e820: construct guest e820 table Ian Jackson
2015-07-22 15:44 ` [PATCH 08/16] tools/libxc: Expose new hypercall xc_reserved_device_memory_map Ian Jackson
2015-07-22 15:51 ` Wei Liu
2015-07-22 15:44 ` [PATCH 09/16] tools: extend xc_assign_device() to support rdm reservation policy Ian Jackson
2015-07-22 15:44 ` [PATCH 10/16] tools: introduce some new parameters to set rdm policy Ian Jackson
2015-07-22 15:44 ` [PATCH 11/16] tools/libxl: detect and avoid conflicts with RDM Ian Jackson
2015-07-22 15:53 ` Wei Liu
2015-07-23 11:05 ` Ian Jackson
2015-07-23 0:52 ` Chen, Tiejun
2015-07-23 7:35 ` Wei Liu
2015-07-23 7:51 ` Chen, Tiejun
2015-07-23 11:09 ` Ian Jackson
2015-07-22 15:44 ` [PATCH 12/16] tools: introduce a new parameter to set a predefined rdm boundary Ian Jackson
2015-07-22 15:44 ` [PATCH 13/16] libxl: construct e820 map with RDM information for HVM guest Ian Jackson
2015-07-22 15:44 ` [PATCH 14/16] xen/vtd: enable USB device assignment Ian Jackson
2015-07-22 15:44 ` [PATCH 15/16] xen/vtd: prevent from assign the device with shared rmrr Ian Jackson
2015-09-03 19:39 ` Tamas K Lengyel
2015-09-04 8:17 ` Jan Beulich
2015-09-04 21:52 ` Tamas K Lengyel
2015-09-06 2:16 ` Chen, Tiejun
2015-09-06 3:19 ` Tamas K Lengyel
2015-09-06 4:19 ` Chen, Tiejun [this message]
2015-09-06 4:21 ` Tamas K Lengyel
2015-09-06 21:27 ` Wei Liu
2015-09-07 9:45 ` Jan Beulich
2015-07-22 15:44 ` [PATCH 16/16] tools: parse to enable new rdm policy parameters Ian Jackson
2015-07-22 15:51 ` [PATCH v13 00/16] Fix RMRR (avoid RDM) Ian Jackson
2015-07-23 2:15 ` Chen, Tiejun
2015-07-23 11:10 ` Ian Jackson
2015-07-23 12:53 ` Ian Jackson
2015-07-23 7:36 ` Wei Liu
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=55EBBEE5.2090700@intel.com \
--to=tiejun.chen@intel.com \
--cc=George.Dunlap@eu.citrix.com \
--cc=JBeulich@suse.com \
--cc=ian.campbell@citrix.com \
--cc=ian.jackson@eu.citrix.com \
--cc=kevin.tian@intel.com \
--cc=tamas.k.lengyel@gmail.com \
--cc=wei.liu2@citrix.com \
--cc=xen-devel@lists.xensource.com \
--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 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).