All of lore.kernel.org
 help / color / mirror / Atom feed
From: Julien Grall <julien.grall@citrix.com>
To: Ian Campbell <ian.campbell@citrix.com>,
	Julien Grall <julien.grall@citrix.com>
Cc: Wei Liu <wei.liu2@citrix.com>,
	Ian Jackson <ian.jackson@eu.citrix.com>,
	Julien Grall <julien.grall@linaro.org>,
	tim@xen.org, stefano.stabellini@citrix.com,
	xen-devel@lists.xenproject.org
Subject: Re: [PATCH v8 1/6] tools/libxl: Check if fdt_{first, next}_subnode are present in libfdt
Date: Wed, 13 May 2015 17:04:16 +0100	[thread overview]
Message-ID: <55537600.6020700@citrix.com> (raw)
In-Reply-To: <1431531541.8263.337.camel@citrix.com>

Hi Ian,

On 13/05/15 16:39, Ian Campbell wrote:
> On Wed, 2015-05-13 at 16:22 +0100, Julien Grall wrote:
> 
>> :(. Both the distribution are using an old version of libfdt where the 3
>> prototypes are defined but the implementation is not exposed in the
>> library (this is because they use a whitelist for building it)
>>
>> I gave look to see if we can import them from libfdt. It will require to
>> import few others in order to make them work:
>> 	- _fdt_check_node_offset
>> 	- _fdt_offset_ptr
>> 	- _nextprop
>>
>> I think we can skip the first one because it's only a validity check.
>> FWIW, we declared the the partial device tree should be trusted so valid.
>>
>> Nonetheless we would add 5 more functions (+ the actual 2) in libxl
>> which represents ~50 lines of codes.
>>
>> The 3 offending function have been correctly exposed since the version
>> v1.4.0 released in June 2013.
>>
>> I gave a look to major distribution to see which version is using an old
>> version of libfdt (i.e < 1.4.0:
>> 	- Centos: < Centos 6
>> 	- Debian: < Jessie => Wheezy using an old version
>> 	- Fedora: < Fedora 20
>> 	- openSuse: < opensuse 13.1
>> 	- ubuntu: < Ubuntu 14.04 => The LTS (14.04) is using a new version
>> 	- RedHat: < Redhat 6
>>
>> AFAICT, all major distributions except debian (for wheezy) are using a
>> libfdt > v1.4. So I would suggest to disable the partial device tree
>> support on distribution using older version.
> 
> OK, I think we can live with this.

I will resend the series later today or tomorrow.

>> If the user want to use platform device passthrough it would have either
>> to build a newer version of libfdt or append a device tree to the guest
>> kernel.
>>
>> Note: IIRC osstest is using wheezy. If so, we won't be able to test
>> platform device passthrough until the distribution version is upgraded.
>>
>> Although, we don't currently have a platform supporting non-PCI
>> passthrough in osstest.
> 
> Right, I'd expect us to have upgraded to Jessie before we got hold of
> such production hardware anyway.

Good!

Regards,

-- 
Julien Grall

  reply	other threads:[~2015-05-13 16:11 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-12 14:33 [PATCH v8 0/6] xen/arm: Add support for non-PCI passthrough Julien Grall
2015-05-12 14:33 ` [PATCH v8 1/6] tools/libxl: Check if fdt_{first, next}_subnode are present in libfdt Julien Grall
2015-05-13 14:07   ` Ian Campbell
2015-05-13 15:22     ` Julien Grall
2015-05-13 15:39       ` Ian Campbell
2015-05-13 16:04         ` Julien Grall [this message]
2015-05-12 14:33 ` [PATCH v8 2/6] tools/(lib)xl: Add partial device tree support for ARM Julien Grall
2015-05-12 14:33 ` [PATCH v8 3/6] tools/libxl: arm: Use an higher value for the GIC phandle Julien Grall
2015-05-12 14:33 ` [PATCH v8 4/6] libxl: Add support for Device Tree passthrough Julien Grall
2015-05-12 14:33 ` [PATCH v8 5/6] xl: Add new option dtdev Julien Grall
2015-05-12 14:33 ` [PATCH v8 6/6] docs/misc: arm: Add documentation about Device Tree passthrough Julien Grall

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=55537600.6020700@citrix.com \
    --to=julien.grall@citrix.com \
    --cc=ian.campbell@citrix.com \
    --cc=ian.jackson@eu.citrix.com \
    --cc=julien.grall@linaro.org \
    --cc=stefano.stabellini@citrix.com \
    --cc=tim@xen.org \
    --cc=wei.liu2@citrix.com \
    --cc=xen-devel@lists.xenproject.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.