diff mbox series

[RFC] gpu: no need to compile drm/ if CONFIG_DRM=n

Message ID 20190911173453.958-1-cai@lca.pw (mailing list archive)
State New, archived
Headers show
Series [RFC] gpu: no need to compile drm/ if CONFIG_DRM=n | expand

Commit Message

Qian Cai Sept. 11, 2019, 5:34 p.m. UTC
The commit c0e09200dc08 ("drm: reorganise drm tree to be more future
proof.") changed the behavior from only compiling drm/ if CONFIG_DRM=y
to always compiling drm/. This restores the behavior, so people don't
need to waste time compiling stuff they don't need.

Fixes: c0e09200dc08 ("drm: reorganise drm tree to be more future proof.")
---
 drivers/gpu/Makefile | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

Comments

Qian Cai Sept. 11, 2019, 5:41 p.m. UTC | #1
> On Sep 11, 2019, at 1:37 PM, Maxime Ripard <maxime.ripard@anandra.org> wrote:
> 
> Hi,
> 
> Le mer. 11 sept. 2019 à 19:35, Qian Cai <cai@lca.pw> a écrit :
> The commit c0e09200dc08 ("drm: reorganise drm tree to be more future
> proof.") changed the behavior from only compiling drm/ if CONFIG_DRM=y
> to always compiling drm/. This restores the behavior, so people don't
> need to waste time compiling stuff they don't need.
> 
> Fixes: c0e09200dc08 ("drm: reorganise drm tree to be more future proof.")
> 
> You are missing your signed-off-by

That is intentional because this is a RFC patch to gather the feedback as I am no
expert in DRM and may miss something important.

Once people are happy with it, I plan to send a formal patch with Signed-off-by.
Masahiro Yamada Sept. 12, 2019, 12:52 a.m. UTC | #2
On Thu, Sep 12, 2019 at 2:41 AM Qian Cai <cai@lca.pw> wrote:
>
>
>
> > On Sep 11, 2019, at 1:37 PM, Maxime Ripard <maxime.ripard@anandra.org> wrote:
> >
> > Hi,
> >
> > Le mer. 11 sept. 2019 à 19:35, Qian Cai <cai@lca.pw> a écrit :
> > The commit c0e09200dc08 ("drm: reorganise drm tree to be more future
> > proof.") changed the behavior from only compiling drm/ if CONFIG_DRM=y
> > to always compiling drm/. This restores the behavior, so people don't
> > need to waste time compiling stuff they don't need.
> >
> > Fixes: c0e09200dc08 ("drm: reorganise drm tree to be more future proof.")
> >
> > You are missing your signed-off-by
>
> That is intentional because this is a RFC patch to gather the feedback as I am no
> expert in DRM and may miss something important.
>
> Once people are happy with it, I plan to send a formal patch with Signed-off-by.
>


CONFIG_DRM is tristate.

Your patch requires everything under drivers/gpu/drm/
modular when CONFIG_DRM=m.

At least, drivers/gpu/drm/drm_mipi_dsi.c
is built-in only.

I guess that is the reason why
they use 'obj-y += drm/'.


In other words, if you are able to
make all the drm code modular,
your patch is applicable.
diff mbox series

Patch

diff --git a/drivers/gpu/Makefile b/drivers/gpu/Makefile
index f17d01f076c7..a793b9ace34b 100644
--- a/drivers/gpu/Makefile
+++ b/drivers/gpu/Makefile
@@ -3,5 +3,6 @@ 
 # taken to initialize them in the correct order. Link order is the only way
 # to ensure this currently.
 obj-$(CONFIG_TEGRA_HOST1X)	+= host1x/
-obj-y			+= drm/ vga/
+obj-$(CONFIG_DRM)		+= drm/
+obj-y				+= vga/
 obj-$(CONFIG_IMX_IPUV3_CORE)	+= ipu-v3/