All of lore.kernel.org
 help / color / mirror / Atom feed
From: Anthony Liguori <anthony@codemonkey.ws>
To: Avi Kivity <avi@qumranet.com>
Cc: kvm-devel@lists.sourceforge.net, allen.m.kay@intel.com,
	andrea@qumranet.com, benami@il.ibm.com
Subject: Re: [PATCH 1/1] direct mmio for passthrough - kernel part
Date: Tue, 01 Apr 2008 10:20:49 -0500	[thread overview]
Message-ID: <47F252D1.6060905@codemonkey.ws> (raw)
In-Reply-To: <47F249C3.6000300@codemonkey.ws>

Anthony Liguori wrote:
> Avi Kivity wrote:
>> benami@il.ibm.com wrote:
>>  
>>> From: Ben-Ami Yassour <benami@il.ibm.com>
>>>
>>> Enable a guest to access a device's memory mapped I/O regions directly.
>>> Userspace sends the mmio regions that the guest can access. On the 
>>> first
>>> page fault for an access to an mmio address the host translates the 
>>> gva to hpa,
>>> and updates the sptes.
>>>
>>>       
>>
>> Can you explain why you're not using the regular memory slot 
>> mechanism? i.e. have userspace mmap(/dev/mem) and create a memslot 
>> containing that at the appropriate guest physical address?
>>   
>
> /dev/mem is often restricted in what memory can be mapped.  However, 
> we can't add something like this to KVM that allows arbitrary HPA's to 
> be mapped into a guest from userspace.  This is just as bad as 
> /dev/mem and is going to upset a lot of people.
>
> Regardless of whether we can use /dev/mem, I think we should introduce 
> a new char device anyway.  We only need to mmap() MMIO regions which 
> are mapped by the PCI bus, presumably, the kernel should know about 
> these mappings.  The driver should only allow mappings that are valid 
> for a particular PCI device such that it cannot be abused to map 
> arbitrary regions of memory into a guest.  Bonus points if it can 
> validate that there isn't a valid Linux driver loaded for the given 
> PCI device.

Which is apparently entirely unnecessary as we already have 
/sys/bus/pci/.../region.  It's just a matter of checking if a vma is 
VM_IO and then dealing with the subsequent reference counting issues as 
Avi points out.

Regards,

Anthony Liguori

> Regards,
>
> Anthony Liguori
>
>> There are some issues with refcounting, but Andrea has some tricks to 
>> deal with that.
>>
>>   
>


-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace

  reply	other threads:[~2008-04-01 15:20 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-04-01 11:52 [RFC] direct mmio for passthrough benami
2008-04-01 11:52 ` [PATCH 1/1] direct mmio for passthrough - kernel part benami
2008-04-01 13:30   ` Avi Kivity
2008-04-01 14:42     ` Anthony Liguori
2008-04-01 15:20       ` Anthony Liguori [this message]
2008-04-01 17:05         ` Avi Kivity
2008-04-01 18:18         ` Andrea Arcangeli
2008-04-01 18:28           ` Anthony Liguori
2008-04-01 17:03       ` Avi Kivity
2008-04-01 17:18         ` Daniel P. Berrange
2008-04-01 18:10           ` Andrea Arcangeli
2008-04-01 18:18             ` Daniel P. Berrange
2008-04-01 18:23             ` Anthony Liguori
2008-04-01 18:21         ` Anthony Liguori
2008-04-01 19:22           ` Avi Kivity
2008-04-01 22:38             ` Andrea Arcangeli
2008-04-01 22:22           ` Andrea Arcangeli
2008-04-01 22:29             ` Anthony Liguori
2008-04-02  4:00               ` Avi Kivity
2008-04-01 19:28     ` Ben-Ami Yassour1
2008-04-01 19:43       ` Avi Kivity
2008-04-01 20:04         ` Anthony Liguori
2008-04-02  4:32           ` Avi Kivity
2008-04-02  7:03             ` Andrea Arcangeli
2008-04-02  9:50               ` Avi Kivity
2008-04-02 10:28                 ` Andrea Arcangeli
2008-04-02 10:59                   ` Avi Kivity
2008-04-02 11:16                     ` Avi Kivity
2008-04-02 11:50                       ` Andrea Arcangeli
2008-04-02 11:53                         ` Andrea Arcangeli
2008-04-03  8:51                         ` Avi Kivity
2008-04-02 14:59                     ` Anthony Liguori

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=47F252D1.6060905@codemonkey.ws \
    --to=anthony@codemonkey.ws \
    --cc=allen.m.kay@intel.com \
    --cc=andrea@qumranet.com \
    --cc=avi@qumranet.com \
    --cc=benami@il.ibm.com \
    --cc=kvm-devel@lists.sourceforge.net \
    /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.