[Wayland-bugs] [Bug 744932] Wrong (ultra tiny/small) cursor size on HiDPI screen

[mutter (GNOME Bugzilla)]

https://bugzilla.gnome.org/show_bug.cgi?id=744932

Owen Taylor <otaylor at redhat.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #309854|none                        |needs-work
             status|                            |

--- Comment #95 from Owen Taylor <otaylor at redhat.com> ---
Review of attachment 309854:

I attached this patch for review as part of this branch... but looking at it,
I'd definitely like it if we can figure out a way to separate it - it looks
entirely unrelated to the idea of the branch, and I'm not convinced by it as a
patch yet. Is there some way we can land this branch without this>

 * It's not clear to me why Xwayland windows are a separate role... they seem
to behave a lot like shell surfaces, and in a brief look it appears Xwayland
calls wl_shell_get_shell_surface() on them - which by the protocol makes them
have the shell surface role. I don't particularly like introducing the idea
that we have internal pseudo-roles that are different from the roles of the
Wayland protocol. If we want to special-case Xwayland windows, shouldn't we do
it by some other method?

 * The commit message here doesn't reflect the idea of the patch at all,
because looking at the huge comment in meta-wayland-surface.c, it looks like
the *idea* of the patch is that we want to somehow treat frame callbacks
differently on Xwayland windows to try and fix visual artifacts during map?

::: src/wayland/meta-xwayland.c
@@ +49,3 @@
+                                     META_WAYLAND_SURFACE_ROLE_XWAYLAND,
+                                     surface->resource,
+                                     WL_DISPLAY_ERROR_INVALID_OBJECT))

Hmmm ... I'm quite surprised that meta_wayland_surface_set_role() doesn't
follow Mutter standards for error handling, but returns 0/-1... I'd definitely
write this as:

 != 0

as other callers do to make it clearer what is going on.

(But please don't add more functions like this, and probably this one would be
better changed.)

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/wayland-bugs/attachments/20150824/fc5598c1/attachment.html>