1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
|
- monitor-get-edid-using-vbe:
o remove all x86emu and vga softbootloader code, and always use the
LRMI interface which was previously the backup one on 32-bit x86;
libx86 is now used instead of liblrmi, allowing the use of LRMI
interface on non-x86 hosts as well (fixes Mandriva bug #53866, which
was caused by a bug in the removed code)
o fix the retrieval of VBE vendor strings when using the LRMI interface
o retrieve up to 4 EDID extension blocks
o add --skip-vbe-check for skipping call for VBE info; useful if
calling the program multiple times for different ports
o on failure, return exit status 2 if successful VBE calls were made
o never retry in console mode if the card reports that the port does
not support DDC (usually this means that the port has no display
devices connected)
o add basic 15 sec timeout (exit status 3), with --no-timeout for
disabling it
- monitor-parse-edid:
o print EDID version and the number of EDID extension blocks
o accept EDID data with multiple extension blocks
o add support for Detailed Timing Descriptors in CEA EDID Timing
Extension
o add support for Short Video Descriptor formats 1-64 (as defined by
EIA/CEA-861) in Video Data Blocks of CEA EDID Timing Extension
o set Interlace flag for ModeLines created for interlaced Detailed
Timings (and note it in the ModeLine comment)
o correctly parse some interlaced video timings if the Detailed Timing
Descriptor specifies field vertical parameters instead of the frame
parameters (at least Fujitsu-Siemens Myrica VQ32-1T testcase)
o ignore 1x1 modes (at least Nokia Valuegraph 447w testcase)
o add support for standard timing descriptors in monitor descriptors
o add comment "Monitor preferred modeline" only for the modeline the
monitor reports as preferred, and add "Monitor supported modeline"
comment for the other modelines
o allow null manufacturer name, instead check that EDID version is 1.x
or 2.x (see Mandriva bug #28857)
o output an error message when --monitorsdb fails (see Mandriva bugs
#28857 and #40609)
- monitor-edid, monitor-get-edid:
o do not get duplicate EDIDs if the same EDID is available via
multiple methods
o allow retrieval of multiple EDIDs via VBE
o get VBE info only once even when probing multiple ports
- monitor-edid:
o add --first support for stopping processing after one EDID has been
found
- monitor-probe-using-X:
o accept EDID data that contains extension blocks
- monitor-probe:
o do not probe using X if DMI probe was successful
Version 2.5 - 18 October 2009, by Anssi Hannula
- monitor-probe-using-X:
o disable glx module to speed up X startup, especially if a proprietary
glx module is in use (fixes a timeout with slow filesystems)
o use -sharevts for X server when plymouth is active, as VT switching
does not work at that point (fixes Mandriva bug #53736)
Version 2.4 - 1 October 2009, by Anssi Hannula
- monitor-parse-edid:
o ignore the Manufacturer Specified Range Timing descriptor if the
first detailed timing descriptor appears to violate it (this probably
means that the descriptor is actually something else or in a
different format; this fixes Lenovo W500 detection, reported by Udo
Rader)
Version 2.3 - 6 September 2009, by Anssi Hannula
- monitor-get-edid-using-vbe:
o update x86emu from xserver git (fixes issues at least with newish
NVIDIA cards)
o check if the port supports DDC before trying to read EDID data (this
should prevent problems when probing nonexistent ports on old cards)
- monitor-edid, monitor-get-edid:
o probe ports 0-2 by default instead of just 0-1 (many NVIDIA cards
seem to be using port 2 and the changes in monitor-get-edid-using-vbe
should make it safe for old problematic NVIDIA cards as well)
Version 2.2 - 16 August 2009, by Anssi Hannula
- monitor-edid, monitor-get-edid:
o support getting EDID from kernel DRM when kernel mode-setting is
enabled
- monitor-probe-using-X:
o use resolution from LVDS initial mode when X reports that it is exact,
making it work with nouveau driver
Version 2.1 - 7 July 2008, by Pixel
- monitor-parse-edid:
o compute and display the "dpi" of the preferred modelines
o handle parsing of EDIDs found in "xrandr --prop" or Xorg.log
- monitor-get-edid-using-vbe:
o fix checking the current vt
o instead of checking vt >= 7 to know if X is running,
check wether $DISPLAY is set
Version 2.0 - 8 April 2008, by Pixel
- change license from LGPLv2+ to GPLv3+
- drop cvt (xorg now bundles a similar version)
- remove internal lrmi.c, build with external liblrmi instead
(Remi Collet and Ville Skyttä)
Version 1.16 - 10 March 2008, by Pixel
- do not install monitor-get-edid-using-vbe on archs where VBE is not
available (Remi Collet)
Version 1.15 - 23 January 2008, by Pixel
- monitor-probe:
o probe "using DMI" before "using X"
- monitor-probe-using-X:
o in last resort, get Intel BIOS mode when "BIOS panel mode is bigger than
probed programmed mode"
Version 1.14 - 10 January 2007, by Pixel
- monitor-edid, monitor-get-edid:
o call monitor-get-edid-using-vbe with a range of ports, it stops on first
success (by default it tries port 0 then port 1) (fixes mdvbz#38799)
Version 1.13 - 8 January 2007, by Pixel
- monitor-get-edid:
o skip /proc/acpi/video/**/EDID files which can't be valid (mdvbz#34417)
- monitor-get-edid-using-vbe:
o minimal support for getting EDID from different DDC port
(experimental, need testing before using it in monitor-edid)
|
