3:11 a.m.
Commit 47b5f245bec908f803f0a89c3b1e3166cfe33aad originally introduced
the filtering of files by extension. For the QXL driver at least,
there is a qxldd.dll file which is part of the driver, so that must be
copied to the guest as well.
This patch will cause 'WdfCoInstaller*.dll' files to be copied too.
While I'm not sure if this is correct, it seems as if it will be safe
as although there are multiple copies of this file, they are all the
same (per architecture).
Thanks: Vadim Rozenfeld, Tingting Zheng.
---
v2v/windows_virtio.ml | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/v2v/windows_virtio.ml b/v2v/windows_virtio.ml
index d78bb0c..df1df5a 100644
--- a/v2v/windows_virtio.ml
+++ b/v2v/windows_virtio.ml
@@ -426,7 +426,7 @@ and virtio_iso_path_matches_guest_os path inspect =
in
(* Skip files without specific extensions. *)
- let extensions = ["cat"; "inf"; "pdb"; "sys"]
in
+ let extensions = ["cat"; "dll"; "inf"; "pdb";
"sys"] in
if not (List.mem extension extensions) then raise Not_found;
(* Using the full path, work out what version of Windows
--
2.5.0
4:13 a.m.
On Fri, Mar 04, 2016 at 09:11:00AM +0000, Richard W.M. Jones wrote:
Commit 47b5f245bec908f803f0a89c3b1e3166cfe33aad originally
introduced
the filtering of files by extension. For the QXL driver at least,
there is a qxldd.dll file which is part of the driver, so that must be
copied to the guest as well.
This patch will cause 'WdfCoInstaller*.dll' files to be copied too.
While I'm not sure if this is correct, it seems as if it will be safe
as although there are multiple copies of this file, they are all the
same (per architecture).
Thanks: Vadim Rozenfeld, Tingting Zheng.
---
v2v/windows_virtio.ml | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/v2v/windows_virtio.ml b/v2v/windows_virtio.ml
index d78bb0c..df1df5a 100644
--- a/v2v/windows_virtio.ml
+++ b/v2v/windows_virtio.ml
@@ -426,7 +426,7 @@ and virtio_iso_path_matches_guest_os path inspect =
in
(* Skip files without specific extensions. *)
- let extensions = ["cat"; "inf"; "pdb";
"sys"] in
+ let extensions = ["cat"; "dll"; "inf";
"pdb"; "sys"] in
if not (List.mem extension extensions) then raise Not_found;
(* Using the full path, work out what version of Windows
--
This patch is bogus - please ignore.
v2 coming up soon.
Rich.
--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-builder quickly builds VMs from scratch
http://libguestfs.org/virt-builder.1.html
4:58 a.m.
On Fri, Mar 04, 2016 at 09:11:00AM +0000, Richard W.M. Jones wrote:
Commit 47b5f245bec908f803f0a89c3b1e3166cfe33aad originally
introduced
the filtering of files by extension. For the QXL driver at least,
there is a qxldd.dll file which is part of the driver, so that must be
copied to the guest as well.
This patch will cause 'WdfCoInstaller*.dll' files to be copied too.
While I'm not sure if this is correct,
It is. They are coinstaller dlls and are mentioned in the corresponding
.inf and .cat files, so the driver wouldn't install without them.
There are also other files that sometimes are and sometimes aren't
referred to in the corresponding .inf. I think we need to stop
filtering by extension at all and copy everything from the directory
where .inf resides.
it seems as if it will be safe
as although there are multiple copies of this file, they are all the
same (per architecture).
I wonder if we can stay with the flat structure at all...
Roman.
5:06 a.m.
On Fri, Mar 04, 2016 at 01:58:12PM +0300, Roman Kagan wrote:
On Fri, Mar 04, 2016 at 09:11:00AM +0000, Richard W.M. Jones wrote:
> Commit 47b5f245bec908f803f0a89c3b1e3166cfe33aad originally introduced
> the filtering of files by extension. For the QXL driver at least,
> there is a qxldd.dll file which is part of the driver, so that must be
> copied to the guest as well.
>
> This patch will cause 'WdfCoInstaller*.dll' files to be copied too.
> While I'm not sure if this is correct,
It is. They are coinstaller dlls and are mentioned in the corresponding
.inf and .cat files, so the driver wouldn't install without them.
There are also other files that sometimes are and sometimes aren't
referred to in the corresponding .inf. I think we need to stop
filtering by extension at all and copy everything from the directory
where .inf resides.
> it seems as if it will be safe
> as although there are multiple copies of this file, they are all the
> same (per architecture).
I wonder if we can stay with the flat structure at all...
See also v2 of this patch, and Vadim's comment here:
https://bugzilla.redhat.com/show_bug.cgi?id=1311373#c6
Rich.
--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-df lists disk usage of guests without needing to install any
software inside the virtual machine. Supports Linux and Windows.
http://people.redhat.com/~rjones/virt-df/
5:53 a.m.
On Fri, Mar 04, 2016 at 11:06:06AM +0000, Richard W.M. Jones wrote:
On Fri, Mar 04, 2016 at 01:58:12PM +0300, Roman Kagan wrote:
> On Fri, Mar 04, 2016 at 09:11:00AM +0000, Richard W.M. Jones wrote:
> > Commit 47b5f245bec908f803f0a89c3b1e3166cfe33aad originally introduced
> > the filtering of files by extension. For the QXL driver at least,
> > there is a qxldd.dll file which is part of the driver, so that must be
> > copied to the guest as well.
> >
> > This patch will cause 'WdfCoInstaller*.dll' files to be copied too.
> > While I'm not sure if this is correct,
>
> It is. They are coinstaller dlls and are mentioned in the corresponding
> .inf and .cat files, so the driver wouldn't install without them.
>
> There are also other files that sometimes are and sometimes aren't
> referred to in the corresponding .inf. I think we need to stop
> filtering by extension at all and copy everything from the directory
> where .inf resides.
>
> > it seems as if it will be safe
> > as although there are multiple copies of this file, they are all the
> > same (per architecture).
>
> I wonder if we can stay with the flat structure at all...
See also v2 of this patch, and Vadim's comment here:
https://bugzilla.redhat.com/show_bug.cgi?id=1311373#c6
I clarified this with Vadim, and he meant to say "harmless" :-)
So I'm going to do a v3 of this patch now ...
Rich.
--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-df lists disk usage of guests without needing to install any
software inside the virtual machine. Supports Linux and Windows.
http://people.redhat.com/~rjones/virt-df/
3185
days inactive
3185
days old
4 comments
2 participants
participants (2)
-
Richard W.M. Jones -
Roman Kagan