[PATCH] linux-custom: Install config and map file to /boot

46 views
Skip to first unread message

Vijai Kumar K

unread,
Mar 17, 2020, 1:15:13 PM3/17/20
to isar-...@googlegroups.com, Vijai Kumar K
Align with upstream debian linux-image package. Install the config and
System.map file to /boot directory. Skip installation for User Mode Linux.

Signed-off-by: Vijai Kumar K <Vijaikumar_...@mentor.com>
---
.../linux/files/debian/isar/install.tmpl | 14 ++++++++++++++
1 file changed, 14 insertions(+)

diff --git a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
index 12d592d..4e1f850 100644
--- a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
+++ b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
@@ -37,6 +37,8 @@ do_install() {
set -x

install_image
+ install_config
+ install_map
install_hooks
install_dtbs
install_kmods
@@ -69,6 +71,18 @@ install_image() {
install_image_debug
}

+install_config() {
+ if [ "${ARCH}" != "um" ]; then
+ cp ${O}/.config ${deb_img_dir}/$(dirname ${kimage_path})/config-${krel}
+ fi
+}
+
+install_map() {
+ if [ "${ARCH}" != "um" ]; then
+ cp ${O}/System.map ${deb_img_dir}/$(dirname ${kimage_path})/System.map-${krel}
+ fi
+}
+
install_image_debug() {
# Different tools want the image in different locations
# perf
--
2.17.1

vijai kumar

unread,
Mar 17, 2020, 1:19:20 PM3/17/20
to isar-users
I am not sure how to test the um use case in kernel in ISAR. Are there any specific targets in ISAR that I am missing?

Thanks,
Vijai Kumar K

On Tuesday, March 17, 2020 at 10:45:13 PM UTC+5:30, vijai kumar wrote:
Align with upstream debian linux-image package. Install the config and
System.map file to /boot directory. Skip installation for User Mode Linux.

Signed-off-by: Vijai Kumar K <Vijaikumar_Kanagarajan@mentor.com>

Jan Kiszka

unread,
Mar 17, 2020, 1:32:44 PM3/17/20
to Vijai Kumar K, isar-...@googlegroups.com, Vijai Kumar K
I think this should be made configurable. Not every embedded system may
need or even want these. Can be on be default nevertheless, specifically
if that is Debian behavior.

That said, we may also look into shrinking the dtb installations, only
packaging what the target actually needs. Unrelated to this patch, for sure.

Jan

--
Siemens AG, Corporate Technology, CT RDA IOT SES-DE
Corporate Competence Center Embedded Linux

Jan Kiszka

unread,
Mar 17, 2020, 1:40:01 PM3/17/20
to vijai kumar, isar-users, Cedric Hombourger
On 17.03.20 18:19, vijai kumar wrote:
> I am not sure how to test the um use case in kernel in ISAR. Are there
> any specific targets in ISAR that I am missing?
>

I don't think we had a um use case so far, so there might be things
silently broken. Not sure if Cedric tested the um special handling, I
surely didn't.

Jan

> Thanks,
> Vijai Kumar K
>
> On Tuesday, March 17, 2020 at 10:45:13 PM UTC+5:30, vijai kumar wrote:
>
> Align with upstream debian linux-image package. Install the config and
> System.map file to /boot directory. Skip installation for User Mode
> Linux.
>
> Signed-off-by: Vijai Kumar K <Vijaikumar_...@mentor.com
> <mailto:Vijaikumar_...@mentor.com>>
> --
> You received this message because you are subscribed to the Google
> Groups "isar-users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to isar-users+...@googlegroups.com
> <mailto:isar-users+...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/isar-users/e1c7b955-185e-4f10-83f5-349ad6847274%40googlegroups.com
> <https://groups.google.com/d/msgid/isar-users/e1c7b955-185e-4f10-83f5-349ad6847274%40googlegroups.com?utm_medium=email&utm_source=footer>.

vijai kumar

unread,
Mar 18, 2020, 1:53:42 AM3/18/20
to Jan Kiszka, isar-users, Vijai Kumar K
Sure. I believe that can be investigated and done as a separate patch.
We are now trying to match the contents wrt Debian upstream/ the previous
Linux build procedure as some of our downstream test cases rely on the config
file in /boot.

Thanks,
Vijai Kumar K

Henning Schild

unread,
Mar 18, 2020, 3:34:58 AM3/18/20
to vijai kumar, Jan Kiszka, isar-users, Vijai Kumar K
On Wed, 18 Mar 2020 11:23:24 +0530
vijai kumar <vijaikumar....@gmail.com> wrote:

> On Tue, Mar 17, 2020 at 11:02 PM Jan Kiszka <jan.k...@siemens.com>
> wrote:
> >
> > On 17.03.20 18:14, Vijai Kumar K wrote:
> > > Align with upstream debian linux-image package. Install the
> > > config and System.map file to /boot directory. Skip installation
> > > for User Mode Linux.
> > >
> > > Signed-off-by: Vijai Kumar K <Vijaikumar_...@mentor.com>
> > > ---
> > > .../linux/files/debian/isar/install.tmpl | 14
> > > ++++++++++++++ 1 file changed, 14 insertions(+)
> > >
> > > diff --git
> > > a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
> > > b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl index
> > > 12d592d..4e1f850 100644 ---
> > > a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl +++
> > > b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl @@
> > > -37,6 +37,8 @@ do_install() { set -x
> > >
> > > install_image

Looks like the "um" guard could move here to only have it once.
If the system under test has CONFIG_IKCONFIG_PROC and is running the
kernel, /proc/config.gz is probably the better source.
For the config i really do not see a point to not ship it, its not too
big. Some security gurus might argue that they do not want to ship it
for obfuscation ... but i would not listen to them.

I think your test should try /proc /boot and finally skip the test
instead of fail it ;).

Henning

vijai kumar

unread,
Mar 18, 2020, 9:03:14 AM3/18/20
to Henning Schild, Jan Kiszka, isar-users, Vijai Kumar K
On Wed, Mar 18, 2020 at 1:04 PM Henning Schild
<henning...@siemens.com> wrote:
>
> On Wed, 18 Mar 2020 11:23:24 +0530
> vijai kumar <vijaikumar....@gmail.com> wrote:
>
> > On Tue, Mar 17, 2020 at 11:02 PM Jan Kiszka <jan.k...@siemens.com>
> > wrote:
> > >
> > > On 17.03.20 18:14, Vijai Kumar K wrote:
> > > > Align with upstream debian linux-image package. Install the
> > > > config and System.map file to /boot directory. Skip installation
> > > > for User Mode Linux.
> > > >
> > > > Signed-off-by: Vijai Kumar K <Vijaikumar_...@mentor.com>
> > > > ---
> > > > .../linux/files/debian/isar/install.tmpl | 14
> > > > ++++++++++++++ 1 file changed, 14 insertions(+)
> > > >
> > > > diff --git
> > > > a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
> > > > b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl index
> > > > 12d592d..4e1f850 100644 ---
> > > > a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl +++
> > > > b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl @@
> > > > -37,6 +37,8 @@ do_install() { set -x
> > > >
> > > > install_image
>
> Looks like the "um" guard could move here to only have it once.

Yes. It could. I will send a v2.
This is something we are considering. For now, we are relying on /boot/config*.

I believe boards supported in CIP already have these configs enabled.
So it should
not be a problem for us to enable it for our boards too.

Thanks,
Vijai Kumar K

Vijai Kumar K

unread,
Mar 18, 2020, 9:21:48 AM3/18/20
to isar-...@googlegroups.com, Vijai Kumar K
Align with upstream debian linux-image package. Install the config and
System.map file to /boot directory. Skip installation for User Mode Linux.

Signed-off-by: Vijai Kumar K <Vijaikumar_...@mentor.com>
---
Changes in v2:
- Moved um condition check as per Henning's comment

.../linux/files/debian/isar/install.tmpl | 12 ++++++++++++
1 file changed, 12 insertions(+)

diff --git a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
index 12d592d..849da5e 100644
--- a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
+++ b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
@@ -37,6 +37,10 @@ do_install() {
set -x

install_image
+ if [ "${ARCH}" != "um" ]; then
+ install_config
+ install_map
+ fi
install_hooks
install_dtbs
install_kmods
@@ -69,6 +73,14 @@ install_image() {
install_image_debug
}

+install_config() {
+ cp ${O}/.config ${deb_img_dir}/$(dirname ${kimage_path})/config-${krel}
+}
+
+install_map() {
+ cp ${O}/System.map ${deb_img_dir}/$(dirname ${kimage_path})/System.map-${krel}
+}
+
install_image_debug() {
# Different tools want the image in different locations
# perf
--
2.17.1

Jan Kiszka

unread,
Mar 23, 2020, 2:12:06 AM3/23/20
to Vijai Kumar K, isar-...@googlegroups.com, Vijai Kumar K
I think those cp calls should be install with proper modes, no?

Jan

> +}
> +
> install_image_debug() {
> # Different tools want the image in different locations
> # perf
>

--

Henning Schild

unread,
Mar 23, 2020, 2:53:55 AM3/23/20
to [ext] Jan Kiszka, Vijai Kumar K, isar-...@googlegroups.com, Vijai Kumar K
I wanted to say that as well, but in fact the whole scripts seems to
use a mix of cp and install and this one does not make things worse.

Henning

Jan Kiszka

unread,
Mar 23, 2020, 2:56:04 AM3/23/20
to Henning Schild, Vijai Kumar K, isar-...@googlegroups.com, Vijai Kumar K
Yes, there are 2 other cases that should be fixed, separately.

Jan

vijai kumar

unread,
Mar 23, 2020, 3:49:28 AM3/23/20
to Jan Kiszka, Henning Schild, isar-users, Vijai Kumar K
I believe the permissions should be 644. Please correct me if I am
wrong. Will send v3 with the install command.

Thanks,
Vijai Kumar K

Henning Schild

unread,
Mar 23, 2020, 5:13:33 AM3/23/20
to Jan Kiszka, Vijai Kumar K, isar-...@googlegroups.com, Vijai Kumar K
On Mon, 23 Mar 2020 07:55:58 +0100
I would suggest we ask Vijai to do all on top, instead of changing here
and forgetting the 2 cases?

Or maybe just the two on top as part of the series? Vijai what do you
think?

Henning

> Jan
>

vijai kumar

unread,
Mar 23, 2020, 5:24:29 AM3/23/20
to Henning Schild, Jan Kiszka, isar-users, Vijai Kumar K
Hi Henning,

We could address the two cases here and rest two cases as a separate
patch. I could update this patch(v3) and can send a separate patch for
the remaining two cases.

Is this okay for all?

Thanks,
Vijai Kumar K

>
> Henning
>
> > Jan
> >
>

Henning Schild

unread,
Mar 23, 2020, 6:40:40 AM3/23/20
to vijai kumar, Jan Kiszka, isar-users, Vijai Kumar K
If you are willing to do the fix i would suggest to include it into
your series instead of sending it alone.

Henning

Vijai Kumar K

unread,
Mar 23, 2020, 9:40:41 AM3/23/20
to isar-...@googlegroups.com, Vijai Kumar K
Align with upstream debian linux-image package. Install the config and
System.map file to /boot directory. Skip installation for User Mode Linux.

Signed-off-by: Vijai Kumar K <Vijaikumar_...@mentor.com>
---
Changes in v3:
- Used install instead of cp in p1
- Introduced p2 which replaces cp with install for kernel image copy

Changes in v2:
- Moved um condition check as per Henning's comment

.../linux/files/debian/isar/install.tmpl | 12 ++++++++++++
1 file changed, 12 insertions(+)

diff --git a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
index 12d592d..eae3f6b 100644
--- a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
+++ b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
@@ -37,6 +37,10 @@ do_install() {
set -x

install_image
+ if [ "${ARCH}" != "um" ]; then
+ install_config
+ install_map
+ fi
install_hooks
install_dtbs
install_kmods
@@ -69,6 +73,14 @@ install_image() {
install_image_debug
}

+install_config() {
+ install -m 644 ${O}/.config ${deb_img_dir}/$(dirname ${kimage_path})/config-${krel}
+}
+
+install_map() {
+ install -m 644 ${O}/System.map ${deb_img_dir}/$(dirname ${kimage_path})/System.map-${krel}
+}
+
install_image_debug() {
# Different tools want the image in different locations
# perf
--
2.17.1

Vijai Kumar K

unread,
Mar 23, 2020, 9:40:46 AM3/23/20
to isar-...@googlegroups.com, Vijai Kumar K
Use install to copy the kernel image with necessary permissions
instead of using plain cp.

Signed-off-by: Vijai Kumar K <Vijaikumar_...@mentor.com>
---
meta/recipes-kernel/linux/files/debian/isar/install.tmpl | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
index eae3f6b..8eca4c7 100644
--- a/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
+++ b/meta/recipes-kernel/linux/files/debian/isar/install.tmpl
@@ -62,7 +62,7 @@ EOF

install_image() {
install -m 755 -d ${deb_img_dir}/$(dirname ${kimage_path})
- cp ${O}/${kimage} ${deb_img_dir}/${kimage_path}
+ install -m 644 ${O}/${kimage} ${deb_img_dir}/${kimage_path}

# Make sure arm64 and riscv kernels are decompressed
if [ "${ARCH}" = "arm64" -o "${ARCH}" = "riscv" ]; then
@@ -85,7 +85,7 @@ install_image_debug() {
# Different tools want the image in different locations
# perf
mkdir -p ${deb_dbg_dir}/usr/lib/debug/lib/modules/${krel}/
- cp ${O}/vmlinux ${deb_dbg_dir}/usr/lib/debug/lib/modules/${krel}/
+ install -m 644 ${O}/vmlinux ${deb_dbg_dir}/usr/lib/debug/lib/modules/${krel}/
# systemtap
mkdir -p ${deb_dbg_dir}/usr/lib/debug/boot/
ln -s ../lib/modules/$version/vmlinux ${deb_dbg_dir}/usr/lib/debug/boot/vmlinux-${krel}
--
2.17.1

cedric_h...@mentor.com

unread,
Mar 25, 2020, 3:07:12 AM3/25/20
to isar-users


On Tuesday, March 17, 2020 at 6:40:01 PM UTC+1, Jan Kiszka wrote:
On 17.03.20 18:19, vijai kumar wrote:
> I am not sure how to test the um use case in kernel in ISAR. Are there
> any specific targets in ISAR that I am missing?
>

I don't think we had a um use case so far, so there might be things
silently broken. Not sure if Cedric tested the um special handling, I
surely didn't.

used "um" for another project (and without Isar).
I would recommend we don't bother with it as far as Isar is concerned until we get a request for it


Jan

> Thanks,
> Vijai Kumar K
>
> On Tuesday, March 17, 2020 at 10:45:13 PM UTC+5:30, vijai kumar wrote:
>
>     Align with upstream debian linux-image package. Install the config and
>     System.map file to /boot directory. Skip installation for User Mode
>     Linux.
>
>     Signed-off-by: Vijai Kumar K <Vijaikumar_...@mentor.com

vijai kumar

unread,
Apr 3, 2020, 3:28:46 AM4/3/20
to isar-users


On Wednesday, March 25, 2020 at 12:37:12 PM UTC+5:30, cedric_h...@mentor.com wrote:


On Tuesday, March 17, 2020 at 6:40:01 PM UTC+1, Jan Kiszka wrote:
On 17.03.20 18:19, vijai kumar wrote:
> I am not sure how to test the um use case in kernel in ISAR. Are there
> any specific targets in ISAR that I am missing?
>

I don't think we had a um use case so far, so there might be things
silently broken. Not sure if Cedric tested the um special handling, I
surely didn't.

used "um" for another project (and without Isar).
I would recommend we don't bother with it as far as Isar is concerned until we get a request for it

Thank you for your input Cedric.

vijai kumar

unread,
Apr 3, 2020, 3:29:26 AM4/3/20
to isar-users


On Friday, April 3, 2020 at 12:58:46 PM UTC+5:30, vijai kumar wrote:


On Wednesday, March 25, 2020 at 12:37:12 PM UTC+5:30, cedric_h...@mentor.com wrote:


On Tuesday, March 17, 2020 at 6:40:01 PM UTC+1, Jan Kiszka wrote:
On 17.03.20 18:19, vijai kumar wrote:
> I am not sure how to test the um use case in kernel in ISAR. Are there
> any specific targets in ISAR that I am missing?
>

I don't think we had a um use case so far, so there might be things
silently broken. Not sure if Cedric tested the um special handling, I
surely didn't.

used "um" for another project (and without Isar).
I would recommend we don't bother with it as far as Isar is concerned until we get a request for it

Thank you for your input Cedric.


If there are no more review comments, can this be merged?

Thanks,
Vijai Kumar K

Baurzhan Ismagulov

unread,
Apr 17, 2020, 11:05:33 AM4/17/20
to isar-...@googlegroups.com
On Mon, Mar 23, 2020 at 07:10:31PM +0530, Vijai Kumar K wrote:
> Align with upstream debian linux-image package. Install the config and
> System.map file to /boot directory. Skip installation for User Mode Linux.

Applied to next, thanks.

With kind regards,
Baurzhan.

vijai kumar

unread,
Apr 17, 2020, 2:44:56 PM4/17/20
to isar-users, Baurzhan Ismagulov
On Fri, Apr 17, 2020 at 8:35 PM Baurzhan Ismagulov <i...@radix50.net> wrote:
>
> On Mon, Mar 23, 2020 at 07:10:31PM +0530, Vijai Kumar K wrote:
> > Align with upstream debian linux-image package. Install the config and
> > System.map file to /boot directory. Skip installation for User Mode Linux.
>
> Applied to next, thanks.

Hi Baurzhan,

Just checked the next branch, these commit's author name has my
personal email id instead of official one.Signed-off & author email
don't match now.
I do send the patches via my gmail(easier) but the author has always
been my official email address.

Ah, one more older commit of mine too has the same issue. Not sure why.

https://github.com/ilbers/isar/commit/b3e7b4eae3a0e41abd525b01ceb9484d199b3cc5


Thanks,
Vijai Kumar K

>
> With kind regards,
> Baurzhan.
>
> --
> You received this message because you are subscribed to the Google Groups "isar-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to isar-users+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/isar-users/20200417150529.fdli36qgg5mb6qw5%40yssyq.m.ilbers.de.

vijai kumar

unread,
Apr 17, 2020, 3:00:02 PM4/17/20
to isar-users, Baurzhan Ismagulov
On Sat, Apr 18, 2020 at 12:14 AM vijai kumar
<vijaikumar....@gmail.com> wrote:
>
> On Fri, Apr 17, 2020 at 8:35 PM Baurzhan Ismagulov <i...@radix50.net> wrote:
> >
> > On Mon, Mar 23, 2020 at 07:10:31PM +0530, Vijai Kumar K wrote:
> > > Align with upstream debian linux-image package. Install the config and
> > > System.map file to /boot directory. Skip installation for User Mode Linux.
> >
> > Applied to next, thanks.
>
> Hi Baurzhan,
>
> Just checked the next branch, these commit's author name has my
> personal email id instead of official one.Signed-off & author email
> don't match now.
> I do send the patches via my gmail(easier) but the author has always
> been my official email address.
>
> Ah, one more older commit of mine too has the same issue. Not sure why.
>
> https://github.com/ilbers/isar/commit/b3e7b4eae3a0e41abd525b01ceb9484d199b3cc5
>

My gitconfig is missing the from entry I had earlier. That should be
the root-cause for this issue I guess.

Baurzhan Ismagulov

unread,
Apr 20, 2020, 5:28:18 AM4/20/20
to isar-users
Hello Vijai Kumar,

On Sat, Apr 18, 2020 at 12:29:50AM +0530, vijai kumar wrote:
> My gitconfig is missing the from entry I had earlier. That should be
> the root-cause for this issue I guess.

Yes, e.g. https://groups.google.com/d/msg/isar-users/knMKY-9b5Nw/aDIpgN1MCQAJ
has From: Vijai Kumar K <Vijaikumar_...@mentor.com> as the first line,
and the latter patches don't.

With kind regards,
Baurzhan.

vijai kumar

unread,
Apr 20, 2020, 6:07:29 AM4/20/20
to isar-users
Yes. Sorry about that. I have fixed my configuration. Here after it should come with a proper from.

Baurzhan Ismagulov

unread,
Apr 20, 2020, 12:03:28 PM4/20/20
to isar-users
On Mon, Apr 20, 2020 at 03:07:29AM -0700, vijai kumar wrote:
> > Yes, e.g.
> > https://groups.google.com/d/msg/isar-users/knMKY-9b5Nw/aDIpgN1MCQAJ
> > has From: Vijai Kumar K <Vijaikumar_...@mentor.com <javascript:>> as the
> > first line,
> > and the latter patches don't.
>
> Yes. Sorry about that. I have fixed my configuration. Here after it should
> come with a proper from.

No problem! Just an explanation how one could verify that by looking at the
patch on the list.

With kind regards,
Baurzhan.

Henning Schild

unread,
Apr 20, 2020, 3:46:57 PM4/20/20
to Baurzhan Ismagulov, isar-users
Am Mon, 20 Apr 2020 18:03:16 +0200
schrieb Baurzhan Ismagulov <i...@radix50.net>:
That is not the duty of the maintainer. If contributors get their email
wrong its their problem. If weird IT of contributors prevents them from
sending raw emails, that is a problem of those companies. Because their
contributions will not show up in i.e. "git shortlog -sne | grep
company" ... so they might not get the "credits" ... whatever that
means to those companies should not be the maintainers business.

It might be a nice finding and the maintainer could comment on it, but
you should not waste your time.

On the list you will see the "From" header of the mail, an override
would be an addition "From" in the body. If the "From" in the mail is
not the "Signed-off" that would be an indication. Once you "git am" you
can see the same "problem" in your git.

Henning

> With kind regards,
> Baurzhan.
>

Henning Schild

unread,
Apr 20, 2020, 3:51:22 PM4/20/20
to vijai kumar, isar-users
Am Mon, 20 Apr 2020 03:07:29 -0700
schrieb vijai kumar <vijaikumar....@gmail.com>:

> On Monday, April 20, 2020 at 2:58:18 PM UTC+5:30, Baurzhan Ismagulov
> wrote:
> >
> > Hello Vijai Kumar,
> >
> > On Sat, Apr 18, 2020 at 12:29:50AM +0530, vijai kumar wrote:
> > > My gitconfig is missing the from entry I had earlier. That should
> > > be the root-cause for this issue I guess.
> >
> > Yes, e.g.
> > https://groups.google.com/d/msg/isar-users/knMKY-9b5Nw/aDIpgN1MCQAJ
> > has From: Vijai Kumar K <Vijaikumar_...@mentor.com <javascript:>>
> > as the first line,
> > and the latter patches don't.
> >
>
> Yes. Sorry about that. I have fixed my configuration. Here after it
> should come with a proper from.

I configured my git client to always include the additional "From" and
even patched it to even include it if it was the original "From".

https://github.com/henning-schild/git/commit/ca44f026a31537477e66559f0bf0f3acd2deecdf

Henning
Reply all
Reply to author
Forward
0 new messages