<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Thu, Apr 26, 2018 at 2:40 PM, Gurchetan Singh <span dir="ltr"><<a href="mailto:gurchetansingh@chromium.org" target="_blank">gurchetansingh@chromium.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">On Thu, Apr 26, 2018 at 10:23 AM, Marek Olšák <<a href="mailto:maraeo@gmail.com">maraeo@gmail.com</a>> wrote:<br>
> Gurchetan, can you confirm that the stride returned at alloc() is only used<br>
> for CPU access and not for display programming?<br>
<br>
</span>It really depends on how you couple your HWComposer and gralloc HALs<br>
together. For example, drm_hwcomposer calls drmModeAddFB2 from with<br>
the allocation time stride:<br>
<br>
<a href="https://anongit.freedesktop.org/git/drm_hwcomposer.git" rel="noreferrer" target="_blank">https://anongit.freedesktop.<wbr>org/git/drm_hwcomposer.git</a><br>
<br>
For the minigbm + wayland_hwcomposer use case, we have a<br>
GRALLOC_DRM_GET_STRIDE call that returns the allocation time pixel<br>
stride to the Wayland client.<br>
<br>
We could:<br>
- Return the stride that works for display with GRALLOC_DRM_GET_STRIDE<br>
- Return the stride that works for CPU access with the alloc() call.<br>
<br>
Just curious -- which Android app and/or test case is running into this issue?<br></blockquote><div><br></div><div>There is some cros_gralloc test case that is failing, but nobody told me which one. Can you have a look at the attached minigbm patch?<br></div><div><br></div><div>Marek<br></div></div><br></div></div>