Hi
----- Original Message -----
More worryingly, I tried xvidtune with that patch applied and that
led to a crash in Xwayland - I'll investigate further to see if this
is something in my tree or with the patch as I suspect.
Quick follow up, the crash occurs in VidModeGetMonitor() from xf86VidMode.c
because the output is NULL:
Program received signal SIGSEGV, Segmentation fault.
VidModeGetMonitor (scrnIndex=<optimized out>,
monitor=monitor@entry=0x7fff4b38a828) at xf86VidMode.c:178
178 mon->heightmm = output->mmHeight;
(gdb) bt
#0 VidModeGetMonitor (scrnIndex=<optimized out>,
monitor=monitor@entry=0x7fff4b38a828) at xf86VidMode.c:178
#1 0x0000000000427077 in ProcXF86VidModeGetMonitor (client=0x2cfce10) at
xf86vmode.c:1080
#2 0x000000000043b717 in Dispatch () at dispatch.c:430
#3 0x000000000043f603 in dix_main (argc=10, argv=0x7fff4b38aa38, envp=<optimized
out>) at main.c:300
#4 0x00007fe17c66f580 in __libc_start_main () from /lib64/libc.so.6
#5 0x0000000000422ad9 in _start ()
(gdb) p *mon
$1 = {id = 0x2cdef10 "Wayland", vendor = 0x2cf3e40 "X.org", model = 0x2d218e0
"Xwayland", nHsync = 1, hsync = {{hi = 0,
lo = 0}, {hi = 0, lo = 0}, {hi = 0, lo = 0}, {hi = 0, lo = 0}, {hi = 0,
lo = 0}, {hi = 0, lo = 0}, {hi = 0,
lo = 0}, {hi = 0, lo = 0}}, nVrefresh = 1, vrefresh = {{hi = 0, lo = 0},
{hi = 0, lo = 0}, {hi = 0, lo = 0}, {
hi = 0, lo = 0}, {hi = 0, lo = 0}, {hi = 0, lo = 0}, {hi = 0, lo = 0},
{hi = 0, lo = 0}}, Modes = 0x0, Last = 0x0,
gamma = {red = 0, green = 0, blue = 0}, widthmm = 0, heightmm = 0, options =
0x0, DDC = 0x0, reducedblanking = 0,
maxPixClock = 0}
(gdb) p output
$2 = (RROutputPtr) 0x0
The fix is pretty obvious, and xvidtune will work with this:
Bool
VidModeGetCurrentModeline(int scrnIndex, void **mode, int *dotClock)
{
@@ -188,8 +175,11 @@ VidModeGetMonitor(int scrnIndex, void **monitor)
mon->nVrefresh = 1;
mon->Modes = NULL;
mon->Last = NULL;
- mon->heightmm = output->mmHeight;
- mon->widthmm = output->mmWidth;
+ if (output)
+ {
+ mon->heightmm = output->mmHeight;
+ mon->widthmm = output->mmWidth;
+ }
*monitor = mon;
return TRUE;
(not saying this is the right solution, but that suffice to avoid the crash)
Yet, all reported values reported for HSynv/VSync/pixel clock/etc. are wrong:
Vendor: X.org, Model: Xwayland
Num hsync: 1, Num vsync: 1
hsync range 0: 0.00 - 0.00
vsync range 0: 0.00 - 0.00
So this implementation is really just to please apps that check for the
existence of the extension, no more - Any app that would actually try to do
something with it would be most likely disappointed :-)
I wonder, can't we reuse more of the xrandr data to get some of these values
correct (even if it's just for querying ant not changing/setting values)?
xrandr --verbose gives sensible data on this exact same hardware:
Screen 0: minimum 320 x 200, current 3840 x 1200, maximum 8192 x 8192
XWAYLAND0 connected 1920x1200+1920+0 (0x24) normal (normal) 520mm x 320mm
Identifier: 0x21
Timestamp: 23275741
Subpixel: unknown
Gamma: 1.0:1.0:1.0
Brightness: 0.0
Clones:
CRTC: 0
CRTCs: 0
Transform: 1.000000 0.000000 0.000000
0.000000 1.000000 0.000000
0.000000 0.000000 1.000000
filter:
1920x1200@0.1Hz (0x24) 0.000MHz -HSync +VSync *current +preferred
h: width 1920 start 1976 end 2160 total 2400 skew 0 clock 0.00KHz
v: height 1200 start 1203 end 1209 total 1212 clock 0.00Hz
XWAYLAND1 connected 1920x1080+0+0 (0x25) normal (normal) 310mm x 170mm
Identifier: 0x23
Timestamp: 23275741
Subpixel: unknown
Gamma: 1.0:1.0:1.0
Brightness: 0.0
Clones:
CRTC: 1
CRTCs: 1
Transform: 1.000000 0.000000 0.000000
0.000000 1.000000 0.000000
0.000000 0.000000 1.000000
filter:
1920x1080@0.1Hz (0x25) 0.000MHz -HSync +VSync *current +preferred
h: width 1920 start 1976 end 2160 total 2400 skew 0 clock 0.00KHz
v: height 1080 start 1083 end 1088 total 1091 clock 0.00Hz
So there should be a way to report some of these via Xfree86-VidModeExtension.
Cheers,
Olivier
_______________________________________________
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel
