All of lore.kernel.org
 help / color / mirror / Atom feed
From: George Dunlap <george.dunlap@citrix.com>
To: Juergen Gross <jgross@suse.com>,
	Ian Campbell <ian.campbell@citrix.com>,
	George Dunlap <George.Dunlap@eu.citrix.com>
Cc: Wei Liu <wei.liu2@citrix.com>, Jim Fehlig <jfehlig@suse.com>,
	Ian Jackson <Ian.Jackson@eu.citrix.com>,
	Chunyan Liu <cyliu@suse.com>,
	"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>
Subject: Re: [PATCH V7 6/7] xl: add usb-assignable-list command
Date: Wed, 7 Oct 2015 15:10:14 +0100	[thread overview]
Message-ID: <561527C6.3040308@citrix.com> (raw)
In-Reply-To: <5615241B.5@suse.com>

On 07/10/15 14:54, Juergen Gross wrote:
> On 10/07/2015 03:21 PM, George Dunlap wrote:
>> On 07/10/15 13:05, Juergen Gross wrote:
>>> On 10/07/2015 01:55 PM, Ian Campbell wrote:
>>>> On Wed, 2015-10-07 at 13:49 +0200, Juergen Gross wrote:
>>>>> On 10/07/2015 01:39 PM, Ian Campbell wrote:
>>>>>> On Wed, 2015-10-07 at 13:25 +0200, Juergen Gross wrote:
>>>>>>>> The "usb-assignable-list" that Chunyan has submitted will give you
>>>>>>>> a
>>>>>>>> list of all dom0 USB devices that have not yet been assigned to a
>>>>>>>> guest.
>>>>>>>>      It should be basically equivalent to "lsusb", except that it
>>>>>>>> filters
>>>>>>>> out devices which have already been assigned to VMs.
>>>>>>
>>>>>> Right, that does sound like useful enough functionality to justify
>>>>>> not
>>>>>> just
>>>>>> using lsusb.
>>>>>>
>>>>>>>> In the e-mail you respond to, I was suggesting that
>>>>>>>>
>>>>>>>> # xl usb-list --all
>>>>>>>>
>>>>>>>> would show you usb devices attached to every VM, and also USB
>>>>>>>> devices
>>>>>>>> attached to no VM, and that
>>>>>>
>>>>>> And would "xl usb-list" list all domain attached to every VM but not
>>>>>> the
>>>>>> unattached ones?
>>>>>>
>>>>>>>> # xl usb-list --host
>>>>>>>>
>>>>>>>> would show you only host usb devices not attached to any VM.
>>>>>>>>
>>>>>>>> I think it's the second bit if functionality which Juergen is keen
>>>>>>>> be
>>>>>>>> available in some form or other.
>>>>>>>
>>>>>>> Exactly.
>>>>>>
>>>>>> I'm happy with this proposal too.
>>>>>>
>>>>>> Is the distinction between --host and --all important to you?
>>>>>> Assuming
>>>>>> unassigned devices are clearly indicated in the --all output (as they
>>>>>> should be).
>>>>>
>>>>> Depends on how the output would look like.
>>>>>
>>>>> In case it either isn't containing VMs without any attached USB device
>>>>> or the unassigned devices are listed at the end this would be okay. I
>>>>> just wouldn't want to have to scroll a lot in case of a host with 1000
>>>>> domains.
>>>>
>>>> Makes sense.
>>>>
>>>>> BTW: In case you don't have the --host option it would be nice to omit
>>>>> --all as well and just assume --all when no domain was specified (like
>>>>> xl list).
>>>>
>>>> I think "xl usb-list" should list the devices attached to all domains,
>>>> for
>>>> sure. I'm ambivalent about whether it also lists unattached devices (in
>>>> which case --all is unnecessary) or if --all is required to get those.
>>>
>>> I'd go for nuking the --all option. I suppose xl usb-list without
>>> specifying a domain will nearly always be used to see the unassigned
>>> devices and not "uuh, to which domain did I assign device x?".
>>
>> I don't know -- if you've assigned a device to one particular domain,
>> then it seems pretty likely that at some point in the future you might
>> want to say, "Wait, what devices do I have assigned where again?"
>>
>> I think
>> * people will expect "xl usb-list" to show all VMs but *not* all
>> unassigned host devices
>> * In this sort of a case, doing what people expect is probably more
>> important than doing what we think is probably more convenient for them.
> 
> Hmm, technically all unassigned USB-devices are usable from Dom0. So why
> not list them there.
> 
> "xl usb-list 0" would be the same then as "xl usb-list --host".
> "xl usb-list" would be the same as "xl usb-list --all".
> 
> Just a thought - I'm fine with either variant.

Haha -- well, I guess that's one way of looking at it. :-)

dom0 is different from the others, however, in that "xl pci-attach" will
automatically remove it from dom0 and give it to another VM.  If that
was true for all domains, I'd be on board with it.  And of course it
doesn't work for driver domains.

 -George

  parent reply	other threads:[~2015-10-07 14:10 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-25  2:11 [PATCH V7 0/7] xen pvusb toolstack work Chunyan Liu
2015-09-25  2:11 ` [PATCH V7 1/7] libxl: export some functions for pvusb use Chunyan Liu
2015-09-25  2:11 ` [PATCH V7 2/7] libxl_read_file_contents: add new entry to read sysfs file Chunyan Liu
2015-09-30 11:22   ` George Dunlap
2015-10-02 13:25   ` Ian Campbell
2015-09-25  2:11 ` [PATCH V7 3/7] libxl: add pvusb API Chunyan Liu
2015-09-30 17:55   ` George Dunlap
2015-10-02 13:31     ` Ian Campbell
2015-10-09  8:12     ` Chun Yan Liu
2015-10-12  7:19     ` Chun Yan Liu
2015-10-12 13:46       ` George Dunlap
2015-10-13  1:46         ` Chun Yan Liu
2015-10-13 13:15           ` George Dunlap
2015-10-13 13:19             ` George Dunlap
2015-10-13 13:30             ` Ian Campbell
2015-10-14  2:29             ` Chun Yan Liu
2015-10-08 14:41   ` Ian Jackson
2015-10-08 14:54     ` Ian Campbell
2015-10-08 15:16       ` Ian Jackson
2015-10-12  7:00     ` Chun Yan Liu
2015-09-25  2:11 ` [PATCH V7 4/7] libxl: add libxl_device_usb_assignable_list API Chunyan Liu
2015-10-01 11:32   ` George Dunlap
2015-09-25  2:11 ` [PATCH V7 5/7] xl: add pvusb commands Chunyan Liu
2015-10-01 17:02   ` George Dunlap
2015-10-02 13:35     ` Ian Campbell
2015-10-02 15:17       ` George Dunlap
2015-10-02 15:29         ` Ian Campbell
2015-10-09  7:15     ` Chun Yan Liu
2015-09-25  2:11 ` [PATCH V7 6/7] xl: add usb-assignable-list command Chunyan Liu
2015-10-06 16:55   ` George Dunlap
2015-10-07  8:40     ` Ian Campbell
2015-10-07  9:55       ` Juergen Gross
2015-10-07 10:08         ` Ian Campbell
2015-10-07 10:10       ` George Dunlap
2015-10-07 10:15         ` George Dunlap
2015-10-07 10:35         ` Christiane Groß
2015-10-07 11:09         ` Ian Campbell
2015-10-07 11:20           ` George Dunlap
2015-10-07 11:25             ` Juergen Gross
2015-10-07 11:32               ` George Dunlap
2015-10-07 11:37               ` Ian Campbell
2015-10-07 11:39                 ` Juergen Gross
2015-10-07 11:43                 ` Ian Campbell
2015-10-07 11:39               ` Ian Campbell
2015-10-07 11:49                 ` Juergen Gross
2015-10-07 11:55                   ` Ian Campbell
2015-10-07 12:05                     ` Juergen Gross
2015-10-07 12:51                       ` Ian Campbell
2015-10-07 13:21                       ` George Dunlap
2015-10-07 13:54                         ` Juergen Gross
2015-10-07 14:05                           ` Ian Campbell
2015-10-07 14:26                             ` Juergen Gross
2015-10-07 14:35                               ` George Dunlap
2015-10-07 14:47                                 ` Juergen Gross
2015-10-07 15:03                                   ` George Dunlap
2015-10-07 15:13                                     ` Juergen Gross
2015-10-07 14:10                           ` George Dunlap [this message]
2015-09-25  2:11 ` [PATCH V7 7/7] domcreate: support pvusb in configuration file Chunyan Liu
2015-10-07 15:06   ` George Dunlap

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=561527C6.3040308@citrix.com \
    --to=george.dunlap@citrix.com \
    --cc=George.Dunlap@eu.citrix.com \
    --cc=Ian.Jackson@eu.citrix.com \
    --cc=cyliu@suse.com \
    --cc=ian.campbell@citrix.com \
    --cc=jfehlig@suse.com \
    --cc=jgross@suse.com \
    --cc=wei.liu2@citrix.com \
    --cc=xen-devel@lists.xen.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.