All of lore.kernel.org
 help / color / mirror / Atom feed
From: hrg <hrgstephen@gmail.com>
To: stefano.stabellini@eu.citrix.com, anthony.perard@citrix.com,
	xen-devel@lists.xensource.com, qemu-devel@nongnu.org
Cc: xen-devel@lists.xenproject.org, xen-devel@lists.xen.org,
	"Herongguang (Stephen)" <herongguang.he@huawei.com>,
	wangxinxin.wang@huawei.com
Subject: [Qemu-devel]  [RFC/BUG] xen-mapcache: buggy invalidate map cache?
Date: Sun, 9 Apr 2017 23:52:34 +0800	[thread overview]
Message-ID: <CADZi59zNxk-Yr4rpN1E49s=AdsdW-W9YfhpCvpJeBz7oigyFSg@mail.gmail.com> (raw)

Hi,

In xen_map_cache_unlocked(), map to guest memory maybe in entry->next
instead of first level entry (if map to rom other than guest memory
comes first), while in xen_invalidate_map_cache(), when VM ballooned
out memory, qemu did not invalidate cache entries in linked
list(entry->next), so when VM balloon back in memory, gfns probably
mapped to different mfns, thus if guest asks device to DMA to these
GPA, qemu may DMA to stale MFNs.

So I think in xen_invalidate_map_cache() linked lists should also be
checked and invalidated.

What’s your opinion? Is this a bug? Is my analyze correct?

WARNING: multiple messages have this Message-ID (diff)
From: hrg <hrgstephen@gmail.com>
To: stefano.stabellini@eu.citrix.com, anthony.perard@citrix.com,
	xen-devel@lists.xensource.com, qemu-devel@nongnu.org
Cc: xen-devel@lists.xenproject.org, wangxinxin.wang@huawei.com,
	"Herongguang (Stephen)" <herongguang.he@huawei.com>,
	xen-devel@lists.xen.org
Subject: [Qemu-devel] [RFC/BUG] xen-mapcache: buggy invalidate map cache?
Date: Sun, 9 Apr 2017 23:52:34 +0800	[thread overview]
Message-ID: <CADZi59zNxk-Yr4rpN1E49s=AdsdW-W9YfhpCvpJeBz7oigyFSg@mail.gmail.com> (raw)

Hi,

In xen_map_cache_unlocked(), map to guest memory maybe in entry->next
instead of first level entry (if map to rom other than guest memory
comes first), while in xen_invalidate_map_cache(), when VM ballooned
out memory, qemu did not invalidate cache entries in linked
list(entry->next), so when VM balloon back in memory, gfns probably
mapped to different mfns, thus if guest asks device to DMA to these
GPA, qemu may DMA to stale MFNs.

So I think in xen_invalidate_map_cache() linked lists should also be
checked and invalidated.

What’s your opinion? Is this a bug? Is my analyze correct?

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

             reply	other threads:[~2017-04-09 15:52 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-09 15:52 hrg [this message]
2017-04-09 15:52 ` [Qemu-devel] [RFC/BUG] xen-mapcache: buggy invalidate map cache? hrg
2017-04-09 15:55 ` hrg
2017-04-09 15:55   ` hrg
2017-04-09 16:36   ` hrg
2017-04-09 16:36     ` hrg
2017-04-09 17:52     ` [Qemu-devel] [Xen-devel] " Alexey G
2017-04-09 17:52       ` [Qemu-devel] " Alexey G
2017-04-10 19:04     ` Stefano Stabellini
2017-04-10 19:04       ` Stefano Stabellini
2017-04-10 19:50       ` Stefano Stabellini
2017-04-10 19:50         ` Stefano Stabellini
2017-04-11  4:47         ` hrg
2017-04-11  4:47           ` hrg
2017-04-11 22:32           ` Stefano Stabellini
2017-04-11 22:32             ` Stefano Stabellini
2017-04-12  6:17             ` [Qemu-devel] [Xen-devel] " Alexey G
2017-04-12  6:17               ` [Qemu-devel] " Alexey G
2017-04-12  8:28               ` [Qemu-devel] [Xen-devel] " Herongguang (Stephen)
2017-04-12  8:28                 ` [Qemu-devel] " Herongguang (Stephen)
2017-04-12 23:51               ` [Qemu-devel] [Xen-devel] " Stefano Stabellini
2017-04-12 23:51                 ` [Qemu-devel] " Stefano Stabellini
2017-04-12  8:14             ` Herongguang (Stephen)
2017-04-12  8:14               ` Herongguang (Stephen)
2017-04-12 23:51               ` Stefano Stabellini
2017-04-12 23:51                 ` Stefano Stabellini
2017-04-13  5:47                 ` Herongguang (Stephen)
2017-04-13  5:47                   ` Herongguang (Stephen)
2017-04-28 23:51                   ` Stefano Stabellini
2017-04-28 23:51                     ` Stefano Stabellini

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='CADZi59zNxk-Yr4rpN1E49s=AdsdW-W9YfhpCvpJeBz7oigyFSg@mail.gmail.com' \
    --to=hrgstephen@gmail.com \
    --cc=anthony.perard@citrix.com \
    --cc=herongguang.he@huawei.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefano.stabellini@eu.citrix.com \
    --cc=wangxinxin.wang@huawei.com \
    --cc=xen-devel@lists.xen.org \
    --cc=xen-devel@lists.xenproject.org \
    --cc=xen-devel@lists.xensource.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.