[PATCH v3] display/gtk: get proper refreshrate

[PATCH v3] display/gtk: get proper refreshrate

[Nikola Pavlica]

From 2d85b4a0edba0bf3fa3c221dcbde7e82a6a771fa Mon Sep 17 00:00:00 2001
From: Nikola Pavlica <pavlica.nikola@gmail.com>
Date: Tue, 31 Dec 2019 18:16:28 +0100
Subject: [PATCH v3] display/gtk: get proper refreshrate

Hi,
QEMU mailing list.

This is the third revision of the patch involving the GTK display
output getting the proper refresh rate for your particular display
output.

Because some VMs in QEMU can get GPU virtualization (using technologies
such as iGVT-g, as mentioned previously), they could produce a video
output that had a higher display refresh rate than of what the GTK
display was displaying. (fxp. Playing a video game inside of a Windows
VM at 60 Hz, while the output stood locked at 33 Hz because of defaults
set in include/ui/console.h)

Since QEMU does indeed have internal systems for determining frame
times as defined in ui/console.c.
The code checks for a variable called update_interval that it later
uses for time calculation. This variable, however, isn't defined
anywhere in ui/gtk.c and instead ui/console.c just sets it to
GUI_REFRESH_INTERVAL_DEFAULT which is 30

update_interval represents the number of milliseconds per display
refresh, and by doing some math we get that 1000/30 = 33.33... Hz

This creates the mentioned problem and what this patch does is that it
checks for the display refresh rate reported by GTK itself (we can take
this as a safe value) and just converts it back to a number of
milliseconds per display refresh.

Explinations of each revision:
v1: initial patch
v2: moved code to a more appropriate place and reduced overhead and
memory usage (by not creating a new GdkDisplay)
v3: removed magic value and turned it into something more readable and
added checks for an error state where GTK was unable to get the refresh
rate, in which case we use the already existing default

P.S. Big thanks to Phil (philmd@redhat.com) for mentioning these issues
and for teaching me on how to work with mailing lists.
P.S.S. Happy new years :)

Signed-off-by: Nikola Pavlica <pavlica.nikola@gmail.com>
---
 include/ui/gtk.h |  2 ++
 ui/gtk.c         | 12 ++++++++++++
 2 files changed, 14 insertions(+)

diff --git a/include/ui/gtk.h b/include/ui/gtk.h
index d9eedad976..d1b230848a 100644
--- a/include/ui/gtk.h
+++ b/include/ui/gtk.h
@@ -28,6 +28,8 @@
 #include "ui/egl-context.h"
 #endif
 
+#define MILLISEC_PER_SEC 1000000
+
 typedef struct GtkDisplayState GtkDisplayState;
 
 typedef struct VirtualGfxConsole {
diff --git a/ui/gtk.c b/ui/gtk.c
index 692ccc7bbb..197d962e3e 100644
--- a/ui/gtk.c
+++ b/ui/gtk.c
@@ -1967,6 +1967,12 @@ static GSList *gd_vc_gfx_init(GtkDisplayState
*s, VirtualConsole *vc,
 {
     bool zoom_to_fit = false;
 
+    int refresh_rate_millihz = 0;
+
+    GdkDisplay *dpy = gtk_widget_get_display(s->window);
+    GdkWindow *win = gtk_widget_get_window(s->window);
+    GdkMonitor *monitor = gdk_display_get_monitor_at_window(dpy, win);
+
     vc->label = qemu_console_get_label(con);
     vc->s = s;
     vc->gfx.scale_x = 1.0;
@@ -2026,6 +2032,12 @@ static GSList *gd_vc_gfx_init(GtkDisplayState
*s, VirtualConsole *vc,
 
     vc->gfx.kbd = qkbd_state_init(con);
     vc->gfx.dcl.con = con;
+
+    refresh_rate_millihz = gdk_monitor_get_refresh_rate(monitor);
+    if (refresh_rate_millihz) {
+        vc->gfx.dcl.update_interval = MILLISEC_PER_SEC /
refresh_rate_millihz;
+    }
+
     register_displaychangelistener(&vc->gfx.dcl);
 
     gd_connect_vc_gfx_signals(vc);
-- 
2.24.0

Re: [PATCH v3] display/gtk: get proper refreshrate

[Gerd Hoffmann]

  Hi,

>  {
>      bool zoom_to_fit = false;
>  
> +    int refresh_rate_millihz = 0;

Please drop the blank line between the variable declarations.
Also there is no need to zero-initialize the variable ...

> +    refresh_rate_millihz = gdk_monitor_get_refresh_rate(monitor);

... as it is set unconditionally here.

> +    if (refresh_rate_millihz) {
> +        vc->gfx.dcl.update_interval = MILLISEC_PER_SEC /
> refresh_rate_millihz;

Patch is corrupted here, your mailer wrapped the line.
Best way to avoid this is to send patches using "git send-email".

cheers,
  Gerd