linux-hyperv.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Thomas Zimmermann <tzimmermann@suse.de>
To: "Noralf Trønnes" <noralf@tronnes.org>,
	daniel@ffwll.ch, airlied@linux.ie, mripard@kernel.org,
	maarten.lankhorst@linux.intel.com, drawat.floss@gmail.com,
	airlied@redhat.com, kraxel@redhat.com, david@lechnology.com,
	sam@ravnborg.org, javierm@redhat.com, kernel@amanoeldawod.com,
	dirty.ice.hu@gmail.com, michael+lkml@stapelberg.ch, aros@gmx.com,
	joshua@stroblindustries.com, arnd@arndb.de
Cc: dri-devel@lists.freedesktop.org, linux-hyperv@vger.kernel.org,
	virtualization@lists.linux-foundation.org
Subject: Re: [PATCH 7/9] drm/simpledrm: Enable FB_DAMAGE_CLIPS property
Date: Mon, 1 Nov 2021 09:56:19 +0100	[thread overview]
Message-ID: <a9da0350-9bfb-7c03-6689-2aec2d81b9df@suse.de> (raw)
In-Reply-To: <14b1c21f-25e5-d862-40ea-dda1e076ef63@tronnes.org>


[-- Attachment #1.1: Type: text/plain, Size: 4732 bytes --]

Hi

Am 24.10.21 um 17:20 schrieb Noralf Trønnes:
> 
> 
> Den 22.10.2021 15.28, skrev Thomas Zimmermann:
>> Enable the FB_DAMAGE_CLIPS property to reduce display-update
>> overhead. Also fixes a warning in the kernel log.
>>
>>    simple-framebuffer simple-framebuffer.0: [drm] drm_plane_enable_fb_damage_clips() not called
>>
>> Fix the computation of the blit rectangle. This wasn't an issue so
>> far, as simpledrm always blitted the full framebuffer. The code now
>> supports damage clipping and virtual screen sizes.
>>
>> Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
>> ---
>>   drivers/gpu/drm/tiny/simpledrm.c | 30 ++++++++++++++++++++++--------
>>   1 file changed, 22 insertions(+), 8 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/tiny/simpledrm.c b/drivers/gpu/drm/tiny/simpledrm.c
>> index 571f716ff427..e872121e9fb0 100644
>> --- a/drivers/gpu/drm/tiny/simpledrm.c
>> +++ b/drivers/gpu/drm/tiny/simpledrm.c
>> @@ -642,7 +642,7 @@ simpledrm_simple_display_pipe_enable(struct drm_simple_display_pipe *pipe,
>>   	void *vmap = shadow_plane_state->data[0].vaddr; /* TODO: Use mapping abstraction */
>>   	struct drm_device *dev = &sdev->dev;
>>   	void __iomem *dst = sdev->screen_base;
>> -	struct drm_rect clip;
>> +	struct drm_rect src_clip, dst_clip;
>>   	int idx;
>>   
>>   	if (!fb)
>> @@ -651,10 +651,14 @@ simpledrm_simple_display_pipe_enable(struct drm_simple_display_pipe *pipe,
>>   	if (!drm_dev_enter(dev, &idx))
>>   		return;
>>   
>> -	drm_rect_init(&clip, 0, 0, fb->width, fb->height);
>> +	drm_rect_fp_to_int(&src_clip, &plane_state->src);
>>   
>> -	dst += drm_fb_clip_offset(sdev->pitch, sdev->format, &clip);
>> -	drm_fb_blit_toio(dst, sdev->pitch, sdev->format->format, vmap, fb, &clip);
>> +	dst_clip = plane_state->dst;
>> +	if (!drm_rect_intersect(&dst_clip, &src_clip))
>> +		return;
>> +
>> +	dst += drm_fb_clip_offset(sdev->pitch, sdev->format, &dst_clip);
>> +	drm_fb_blit_toio(dst, sdev->pitch, sdev->format->format, vmap, fb, &src_clip);
>>   
>>   	drm_dev_exit(idx);
>>   }
>> @@ -686,20 +690,28 @@ simpledrm_simple_display_pipe_update(struct drm_simple_display_pipe *pipe,
>>   	struct drm_framebuffer *fb = plane_state->fb;
>>   	struct drm_device *dev = &sdev->dev;
>>   	void __iomem *dst = sdev->screen_base;
>> -	struct drm_rect clip;
>> +	struct drm_rect damage_clip, src_clip, dst_clip;
>>   	int idx;
>>   
>>   	if (!fb)
>>   		return;
>>   
>> -	if (!drm_atomic_helper_damage_merged(old_plane_state, plane_state, &clip))
>> +	if (!drm_atomic_helper_damage_merged(old_plane_state, plane_state, &damage_clip))
>> +		return;
>> +
> 
> I'm afraid I don't understand what's going on here, but isn't it
> possible to put this logic into drm_atomic_helper_damage_merged()?

The code above gets the damage rectangle (i.e., the plane's area that 
needs to be updated). The code below get's the framebuffer area. If they 
don't overlap, return. (I think this can really only fail with the next 
patch in the series, which adds virtual screens.)

> 
> Noralf.
> 
>> +	drm_rect_fp_to_int(&src_clip, &plane_state->src);
>> +	if (!drm_rect_intersect(&src_clip, &damage_clip))
>> +		return;
>> +
>> +	dst_clip = plane_state->dst;
>> +	if (!drm_rect_intersect(&dst_clip, &src_clip))
>>   		return;

And here we check if the updated plane/framebuffer area is actually 
visible on screen; otherwise return. It could be in an area that is 
off-screen. (Again, this probably only fails with the virtual-screen patch.)

I don't think this is all generic enough to be within 
drm_atomic_helper_damage_merged(). But once we have multiple SHMEM-based 
drivers with virtual screens, we can move it into a helper for 
shadow-buffered planes. Your gud driver would be a candidate for this 
feature as well.

Best regards
Thomas

>>   
>>   	if (!drm_dev_enter(dev, &idx))
>>   		return;
>>   
>> -	dst += drm_fb_clip_offset(sdev->pitch, sdev->format, &clip);
>> -	drm_fb_blit_toio(dst, sdev->pitch, sdev->format->format, vmap, fb, &clip);
>> +	dst += drm_fb_clip_offset(sdev->pitch, sdev->format, &dst_clip);
>> +	drm_fb_blit_toio(dst, sdev->pitch, sdev->format->format, vmap, fb, &src_clip);
>>   
>>   	drm_dev_exit(idx);
>>   }
>> @@ -794,6 +806,8 @@ static int simpledrm_device_init_modeset(struct simpledrm_device *sdev)
>>   	if (ret)
>>   		return ret;
>>   
>> +	drm_plane_enable_fb_damage_clips(&pipe->plane);
>> +
>>   	drm_mode_config_reset(dev);
>>   
>>   	return 0;
>>

-- 
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5, 90409 Nürnberg, Germany
(HRB 36809, AG Nürnberg)
Geschäftsführer: Ivo Totev

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 840 bytes --]

  reply	other threads:[~2021-11-01  8:56 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-22 13:28 [PATCH 0/9] drm/simpledrm: Enable damage clips and virtual screens Thomas Zimmermann
2021-10-22 13:28 ` [PATCH 1/9] drm/format-helper: Export drm_fb_clip_offset() Thomas Zimmermann
2021-10-24  8:25   ` Noralf Trønnes
     [not found]   ` <YXO+bGIW9sn9Rcfw@ravnborg.org>
2021-11-01  8:43     ` Thomas Zimmermann
2021-10-22 13:28 ` [PATCH 2/9] drm/format-helper: Rework format-helper memcpy functions Thomas Zimmermann
2021-10-24  8:25   ` Noralf Trønnes
2021-10-22 13:28 ` [PATCH 3/9] drm/format-helper: Add destination-buffer pitch to drm_fb_swab() Thomas Zimmermann
2021-10-24  8:33   ` Noralf Trønnes
2021-10-22 13:28 ` [PATCH 4/9] drm/format-helper: Rework format-helper conversion functions Thomas Zimmermann
2021-10-24 11:32   ` Noralf Trønnes
2021-11-01 13:38     ` Thomas Zimmermann
2021-10-22 13:28 ` [PATCH 5/9] drm/format-helper: Streamline blit-helper interface Thomas Zimmermann
2021-10-24 14:59   ` Noralf Trønnes
2021-10-22 13:28 ` [PATCH 6/9] drm/fb-helper: Allocate shadow buffer of surface height Thomas Zimmermann
2021-10-24 15:10   ` Noralf Trønnes
2021-10-22 13:28 ` [PATCH 7/9] drm/simpledrm: Enable FB_DAMAGE_CLIPS property Thomas Zimmermann
2021-10-24 15:20   ` Noralf Trønnes
2021-11-01  8:56     ` Thomas Zimmermann [this message]
2021-10-22 13:28 ` [PATCH 8/9] drm/simpledrm: Support virtual screen sizes Thomas Zimmermann
2021-10-22 13:28 ` [PATCH 9/9] drm: Clarify semantics of struct drm_mode_config.{min,max}_{width,height} Thomas Zimmermann

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=a9da0350-9bfb-7c03-6689-2aec2d81b9df@suse.de \
    --to=tzimmermann@suse.de \
    --cc=airlied@linux.ie \
    --cc=airlied@redhat.com \
    --cc=arnd@arndb.de \
    --cc=aros@gmx.com \
    --cc=daniel@ffwll.ch \
    --cc=david@lechnology.com \
    --cc=dirty.ice.hu@gmail.com \
    --cc=drawat.floss@gmail.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=javierm@redhat.com \
    --cc=joshua@stroblindustries.com \
    --cc=kernel@amanoeldawod.com \
    --cc=kraxel@redhat.com \
    --cc=linux-hyperv@vger.kernel.org \
    --cc=maarten.lankhorst@linux.intel.com \
    --cc=michael+lkml@stapelberg.ch \
    --cc=mripard@kernel.org \
    --cc=noralf@tronnes.org \
    --cc=sam@ravnborg.org \
    --cc=virtualization@lists.linux-foundation.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).