Xen devel - Sep 2006 - [Patch] Fix blktap to work with a bootloader

If a Xen guest has a bootloader configured, then it will fail to start
on a blktap image.  The problem is blkdev_uname_to_file, which cannot
parse the "tap:aio:$filename" image strings: it tries to split the
string apart at ":" and assign the result to a 2-tuple, and this
results in a python error if the split results in three or more
strings.

The fix is to split only at the first ":", and then to split again
if we detect "tap:" as the image type.

Signed-off-by: Stephen Tweedie <sct@redhat.com>

diff -r 94df5bd84195 -r aeba3dd98fbb tools/python/xen/util/blkif.py
--- a/tools/python/xen/util/blkif.py	Thu Sep 28 17:09:09 2006 +0100
+++ b/tools/python/xen/util/blkif.py	Thu Sep 28 18:05:08 2006 +0100
@@ -64,9 +64,11 @@ def blkdev_uname_to_file(uname):
     """Take a blkdev uname and return the corresponding
filename."""
     fn = None
     if uname.find(":") != -1:
-        (typ, fn) = uname.split(":")
+        (typ, fn) = uname.split(":", 1)
         if typ == "phy" and not fn.startswith("/"):
             fn = "/dev/%s" %(fn,)
+        if typ == "tap":
+            (typ, fn) = fn.split(":", 1)
     return fn
 
 def mount_mode(name):



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Applied, thanks!

a.

On 9/28/06, Stephen C. Tweedie <sct@redhat.com>
wrote:
> If a Xen guest has a bootloader configured, then it will fail to start
> on a blktap image.  The problem is blkdev_uname_to_file, which cannot
> parse the "tap:aio:$filename" image strings: it tries to split
the
> string apart at ":" and assign the result to a 2-tuple, and this
> results in a python error if the split results in three or more
> strings.
>
> The fix is to split only at the first ":", and then to split
again
> if we detect "tap:" as the image type.
>
> Signed-off-by: Stephen Tweedie <sct@redhat.com>
>
> diff -r 94df5bd84195 -r aeba3dd98fbb tools/python/xen/util/blkif.py
> --- a/tools/python/xen/util/blkif.py    Thu Sep 28 17:09:09 2006 +0100
> +++ b/tools/python/xen/util/blkif.py    Thu Sep 28 18:05:08 2006 +0100
> @@ -64,9 +64,11 @@ def blkdev_uname_to_file(uname):
>      """Take a blkdev uname and return the corresponding
filename."""
>      fn = None
>      if uname.find(":") != -1:
> -        (typ, fn) = uname.split(":")
> +        (typ, fn) = uname.split(":", 1)
>          if typ == "phy" and not fn.startswith("/"):
>              fn = "/dev/%s" %(fn,)
> +        if typ == "tap":
> +            (typ, fn) = fn.split(":", 1)
>      return fn
>
>  def mount_mode(name):
>
>
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Hi all,

On Thu, 2006-09-28 at 18:36 +0100, Stephen C. Tweedie wrote:
> The fix is to split only at the first ":", and then to split
again
> if we detect "tap:" as the image type.
Whoops, xen is still broken ---
> Signed-off-by: Stephen Tweedie <sct@redhat.com>
> --- a/tools/python/xen/util/blkif.py	Thu Sep 28 17:09:09 2006 +0100
> +++ b/tools/python/xen/util/blkif.py	Thu Sep 28 18:05:08 2006 +0100
> @@ -64,9 +64,11 @@ def blkdev_uname_to_file(uname):
>      if uname.find(":") != -1:
> -        (typ, fn) = uname.split(":")
> +        (typ, fn) = uname.split(":", 1)
>          if typ == "phy" and not fn.startswith("/"):
>              fn = "/dev/%s" %(fn,)
> +        if typ == "tap":
> +            (typ, fn) = fn.split(":", 1)
>      return fn
was what I posted, but only the second two new lines actually got
committed to the xenbits repository.

Without the first split(":", 1) change, we''ll split at _all_
the ":"s in
the uname, and if there are more than one, then the assignment to the
(typ,fn) 2-tuple will fail.

Can this please be fixed properly?  It''s just a 3-character commit
that''s left. :-)

btw, is the process of merging patches from email being done completely
manually?  git certainly has really good pull-from-email scripts that
avoid this sort of oversight, but I''m not sure hg is quite so well
equipped.

Cheers,
 Stephen



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On 10/10/06 15:56, "Stephen C. Tweedie" <sct@redhat.com> wrote:
> btw, is the process of merging patches from email being done completely
> manually?  git certainly has really good pull-from-email scripts that
> avoid this sort of oversight, but I''m not sure hg is quite so well
> equipped.
Usually patches are applied as a whole unit using ''patch'',
even if they have
to be manually extracted from the email. Looks like Andy might have done
this one by hand and got it wrong!

 -- Keir



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On Tue, Oct 10, 2006 at 04:01:42PM +0100, Keir Fraser wrote:
> On 10/10/06 15:56, "Stephen C. Tweedie" <sct@redhat.com>
wrote:
> 
> > btw, is the process of merging patches from email being done
completely
> > manually?  git certainly has really good pull-from-email scripts that
> > avoid this sort of oversight, but I''m not sure hg is quite so
well
> > equipped.
> 
> Usually patches are applied as a whole unit using
''patch'', even if they have
> to be manually extracted from the email. Looks like Andy might have done
> this one by hand and got it wrong!
On that note, _please_ can people submit patches

  1. as attachments,
  2. with a text/plain MIME type, and preferably
  3. using hg export with a decent comment and a Signed-off-by: line in place.

It makes life a lot easier!

If your mailer doesn''t do text/plain attachments properly, and for some
reason
you still want to use such a broken heap of four-ex, then the next best thing
is plain-text inline in the message (because most people can''t review
patches
very easily if your mailer has sent it as an octet-stream).

Thanks,

Ewan.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Ewan Mellor wrote:  [Tue Oct 10 2006, 11:26:53AM EDT]
> On that note, _please_ can people submit patches
> 
>   1. as attachments,
>   2. with a text/plain MIME type, and preferably
>   3. using hg export with a decent comment and a Signed-off-by: line in
place.
> 
> It makes life a lot easier!
if this is what you''d like, it might be helpful if you''d
provide
a patch to mercurial''s patchbomb extension (i.e. hg email) to work in
this manner.  Presently it does inline only.

Aron

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On 17/10/06 8:45 pm, "Aron Griffis" <aron@hp.com> wrote:
>> On that note, _please_ can people submit patches
>> 
>>   1. as attachments,
>>   2. with a text/plain MIME type, and preferably
>>   3. using hg export with a decent comment and a Signed-off-by: line in
>> place.
>> 
>> It makes life a lot easier!
> 
> if this is what you''d like, it might be helpful if you''d
provide
> a patch to mercurial''s patchbomb extension (i.e. hg email) to work
in
> this manner.  Presently it does inline only.
Inline is fine so long as your email client doesn''t mangle the patch.

 -- Keir



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel