Re: fbcon non-desktop display use

To try to address both concerns, it feels easiest to do not in-line.

1) Just for background: The H3, and many other ARM systems use the
framebuffer for all video access, 3D accelerated as well as X11. If we
want to permit HMD (headmount display) or other non-desktop displays
on these devices are going to be used at all, it seems DRM must be set
up for them.  I don't think of this as a "feature".

2) Although I "wish" there was a way to permit non-desktop use with
multiple displays, I am having difficulty finding a specific need to
be able to turn it on/off.  All applications I can imagine with the
HMD currently would involve the HMD being the only connected device.
I am still "submitting" the patch with the parameter, however, if you
folks decide not to accept it, I intend to re-submit with just the &&
strict fix (which I just tested and it's good!) -- unless you (Keith)
are willing to put forward the && strict as a patch.

3) I am trying "plain text mode" for my patch, so hopefully it doesn't
truncate the lines.  Also, I misunderstood "Signed-off-by" Thanks!

Signed-off-by: Charles Lohr <redacted>

diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 035784ddd..e14624d0f 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c

@@ -55,6 +55,11 @@ MODULE_PARM_DESC(drm_fbdev_overalloc,
                 "Overallocation of the fbdev buffer (%) [default="
                 __MODULE_STRING(CONFIG_DRM_FBDEV_OVERALLOC) "]");

+static bool drm_fbdev_permit_non_desktop;
+module_param(drm_fbdev_permit_non_desktop, bool, 0644);
+MODULE_PARM_DESC(drm_fbdev_permit_non_desktop,
+               "Allow the framebuffer to use non-desktop displays

[default=off]");
+
 static LIST_HEAD(kernel_fb_helper_list);
 static DEFINE_MUTEX(kernel_fb_helper_lock);

@@ -2109,7 +2114,7 @@ static bool drm_connector_enabled(struct

drm_connector *connector, bool strict)
 {
        bool enable;

-       if (connector->display_info.non_desktop)
+       if (connector->display_info.non_desktop &&
!drm_fbdev_permit_non_desktop && strict)
                return false;

        if (strict)

On Thu, Mar 15, 2018 at 2:30 PM, Keith Packard [off-list ref] wrote:

Charles Lohr [off-list ref] writes:

quoted

Even if the vive is the only device connected, it will still not permit it
to be operated.  See https://github.com/linux-sunxi/linux-sunxi/issues/291
for dri with a lot of debugging turned on.

Oh, it's not supposed to do that. I had intended to write the code so
that if the only device available was a non-desktop device that it would
go ahead and use it. The X server patches did that, but the kernel ones
did not. It looks like that would be an easy patch -- just skip the
non_desktop check in the !strict case for drm_connector_enabled.

However, your patch is a good addition as it will allow you to also
enable the HMD when other monitors are connected.

quoted

I can understand that most users would probably prefer that the vive isn't
automatically used if no other displays are available, however, the current
behavior prevents use of the vive on all devices that use fbdev for their
primary output.

That was definitely not my intention, and thanks for discovering this!

quoted

I've never sent an email to the kernel devel list, so I'm still a little
nervous.  Especially because this is against a different branch, and I'm
starting to think that I should be messaging there, but this is something
that really needs to go upstream.

We'll get it sorted out; I'm not sure what Dave's preference is these
days anyways.

Aside from some minor formatting issues, this patch looks good to me.

quoted

Signed-off-by:

You'll need to add your name and email address here.

quoted

diff --git a/drivers/gpu/drm/drm_fb_helper.c

b/drivers/gpu/drm/drm_fb_helper.c
index 035784ddd..8bfaf79ff 100644

--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c

@@ -55,6 +55,11 @@ MODULE_PARM_DESC(drm_fbdev_overalloc,
                 "Overallocation of the fbdev buffer (%) [default="
                 __MODULE_STRING(CONFIG_DRM_FBDEV_OVERALLOC) "]");

+static bool drm_fbdev_permit_non_desktop;
+module_param(drm_fbdev_permit_non_desktop, bool, 0644);
+MODULE_PARM_DESC(drm_fbdev_permit_non_desktop,
+               "Allow the framebuffer to use non-desktop displays

[default=off]");
+

Your email client appears to be wrapping long lines, which breaks the patch.

quoted

 static LIST_HEAD(kernel_fb_helper_list);
 static DEFINE_MUTEX(kernel_fb_helper_lock);

@@ -2109,7 +2114,7 @@ static bool drm_connector_enabled(struct

drm_connector *connector, bool strict)
 {
        bool enable;

-       if (connector->display_info.non_desktop)
+       if (connector->display_info.non_desktop &&
!drm_fbdev_permit_non_desktop)

If you added '&& strict' here, it will use the HMD if there aren't any
desktop monitors connected.

--
-keith

On Thu, Mar 15, 2018 at 9:03 PM, Charles Lohr [off-list ref] wrote:

To try to address both concerns, it feels easiest to do not in-line.

1) Just for background: The H3, and many other ARM systems use the
framebuffer for all video access, 3D accelerated as well as X11. If we
want to permit HMD (headmount display) or other non-desktop displays
on these devices are going to be used at all, it seems DRM must be set
up for them.  I don't think of this as a "feature".

There's no open source 3D stack using fbdev. And I really don't care
much about the others from an upstream pov.

2) Although I "wish" there was a way to permit non-desktop use with
multiple displays, I am having difficulty finding a specific need to
be able to turn it on/off.  All applications I can imagine with the
HMD currently would involve the HMD being the only connected device.
I am still "submitting" the patch with the parameter, however, if you
folks decide not to accept it, I intend to re-submit with just the &&
strict fix (which I just tested and it's good!) -- unless you (Keith)
are willing to put forward the && strict as a patch.

I guess we can do that, but I'll defer to Keith/Dave on this since I'm
firmly meh for trying to get HMDs to show up on fbcon. Doesn't make
sense to me.
-Daniel

-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch