From: Stefan Agner <stefan@agner.ch>
To: Daniel Vetter <daniel@ffwll.ch>
Cc: marex@denx.de, airlied@linux.ie, tomi.valkeinen@ti.com,
laurent.pinchart@ideasonboard.com,
dri-devel@lists.freedesktop.org
Subject: Re: [RFC PATCH 2/3] drm/atomic-helper: add REQUIRE_MATCHING_FB flag
Date: Wed, 16 Sep 2020 12:44:18 +0200 [thread overview]
Message-ID: <9dc5bcf24c85f5e463453577b01dee03@agner.ch> (raw)
In-Reply-To: <20200911085226.GP438822@phenom.ffwll.local>
On 2020-09-11 10:52, Daniel Vetter wrote:
> On Thu, Sep 10, 2020 at 11:24:24AM +0200, Stefan Agner wrote:
>> Add flag which checks that the framebuffer size matches the plane size
>> exactly. This is useful for display controller which can't handle
>> framebuffers other than the plane/CRTC size.
>>
>> Signed-off-by: Stefan Agner <stefan@agner.ch>
>> ---
>> drivers/gpu/drm/drm_atomic_helper.c | 7 +++++++
>> drivers/gpu/drm/selftests/test-drm_plane_helper.c | 9 +++++++++
>> include/drm/drm_atomic_helper.h | 1 +
>> 3 files changed, 17 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
>> index 755572a37f3f..8bc7f8c2e566 100644
>> --- a/drivers/gpu/drm/drm_atomic_helper.c
>> +++ b/drivers/gpu/drm/drm_atomic_helper.c
>> @@ -802,6 +802,7 @@ int drm_atomic_helper_check_plane_state(struct drm_plane_state *plane_state,
>> int hscale, vscale;
>> bool can_position = flags & DRM_PLANE_CAN_POSITION;
>> bool can_update_disabled = flags & DRM_PLANE_CAN_UPDATE_DISABLED;
>> + bool require_matching_fb = flags & DRM_PLANE_REQUIRE_MATCHING_FB;
>>
>> WARN_ON(plane_state->crtc && plane_state->crtc != crtc_state->crtc);
>>
>> @@ -860,6 +861,12 @@ int drm_atomic_helper_check_plane_state(struct drm_plane_state *plane_state,
>> return -EINVAL;
>> }
>>
>> + if (require_matching_fb && (drm_rect_width(src) != fb->width ||
>> + drm_rect_height(src) != fb->height)) {
>> + DRM_DEBUG_KMS("Framebuffer size must match plane size.\n");
>> + return -EINVAL;
>> + }
>
> I think you also want to check that the x,y src are at (0, 0).
Hm, I was thinking that is handled by CAN_POSITION, but that really is
all about src vs. dst in plane coordinates.
I guess at that point its almost nicer to create a drm_rect for fb and
use drm_rect_equals.
--
Stefan
>
> Plus needs kerneldoc.
>
> But aside from the details, I like.
>
>> +
>> return 0;
>> }
>> EXPORT_SYMBOL(drm_atomic_helper_check_plane_state);
>> diff --git a/drivers/gpu/drm/selftests/test-drm_plane_helper.c b/drivers/gpu/drm/selftests/test-drm_plane_helper.c
>> index 01e95b2d572f..2c81bbef830e 100644
>> --- a/drivers/gpu/drm/selftests/test-drm_plane_helper.c
>> +++ b/drivers/gpu/drm/selftests/test-drm_plane_helper.c
>> @@ -139,6 +139,15 @@ int igt_check_plane_state(void *ignored)
>> FAIL_ON(!check_src_eq(&plane_state, 0, 0, 1023 << 16, 767 << 16));
>> FAIL_ON(!check_crtc_eq(&plane_state, 0, 0, 1023, 767));
>>
>> + /* Check whether requiring same size framebuffer works correctly. */
>> + set_src(&plane_state, 0, 0, 1024 << 16, 768 << 16);
>> + set_crtc(&plane_state, 0, 0, 1024, 768);
>> + ret = drm_atomic_helper_check_plane_state(&plane_state, &crtc_state,
>> + DRM_PLANE_HELPER_NO_SCALING,
>> + DRM_PLANE_HELPER_NO_SCALING,
>> + DRM_PLANE_REQUIRE_MATCHING_FB);
>> + FAIL(!ret, "Should not be able to use different size framebuffer with REQUIRE_MATCHING_FB\n");
>
> I think also needs a negative test with src_x/y != 0, plus maybe one that
> succeeds.
>
> Cheers, Daniel
>> +
>> /* Simple scaling tests. */
>> set_src(&plane_state, 0, 0, 512 << 16, 384 << 16);
>> set_crtc(&plane_state, 0, 0, 1024, 768);
>> diff --git a/include/drm/drm_atomic_helper.h b/include/drm/drm_atomic_helper.h
>> index bb9957b4f91b..244b730e84d3 100644
>> --- a/include/drm/drm_atomic_helper.h
>> +++ b/include/drm/drm_atomic_helper.h
>> @@ -43,6 +43,7 @@ int drm_atomic_helper_check_modeset(struct drm_device *dev,
>>
>> #define DRM_PLANE_CAN_POSITION BIT(0)
>> #define DRM_PLANE_CAN_UPDATE_DISABLED BIT(1)
>> +#define DRM_PLANE_REQUIRE_MATCHING_FB BIT(2)
>>
>> int drm_atomic_helper_check_plane_state(struct drm_plane_state *plane_state,
>> const struct drm_crtc_state *crtc_state,
>> --
>> 2.28.0
>>
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
next prev parent reply other threads:[~2020-09-16 10:44 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-10 9:24 [RFC PATCH 1/3] drm: use flags instead of boolean in plane check Stefan Agner
2020-09-10 9:24 ` [RFC PATCH 2/3] drm/atomic-helper: add REQUIRE_MATCHING_FB flag Stefan Agner
2020-09-11 8:52 ` Daniel Vetter
2020-09-16 10:44 ` Stefan Agner [this message]
2020-09-11 11:59 ` Ville Syrjälä
2020-09-16 10:49 ` Stefan Agner
2020-09-10 9:24 ` [RFC PATCH 3/3] drm: mxsfb: enforce framebuffer size matches plane size Stefan Agner
2020-09-11 8:50 ` [RFC PATCH 1/3] drm: use flags instead of boolean in plane check Daniel Vetter
2020-09-16 10:31 ` Stefan Agner
2020-09-16 11:26 ` Daniel Vetter
2020-09-15 22:58 ` Laurent Pinchart
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=9dc5bcf24c85f5e463453577b01dee03@agner.ch \
--to=stefan@agner.ch \
--cc=airlied@linux.ie \
--cc=daniel@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=laurent.pinchart@ideasonboard.com \
--cc=marex@denx.de \
--cc=tomi.valkeinen@ti.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.