[Libdlo] [PATCH] libdlo: fix EDID parsing

[Martin Decky]

From: Martin Decky <martin at decky.cz>

The operations for getting the values of vertical sync offset and width
in libdlo do not respect the EDID specification. These values are 6
bits wide with no extra zero bits in them according to [1].

(Patch sent on 2011-02-13 directly to Bernie Thompson, but he's
probably too busy to deal with it.)

[1] VESA Enhanced Extended Display Identification Data Standard,
     Release A, Revision 2, page 34, VESA, Sep 2006

Signed-off-by: Martin Decky <martin at decky.cz>
----
diff -Naurp libdlo.orig/src/dlo_mode.c libdlo/src/dlo_mode.c
--- libdlo.orig/src/dlo_mode.c	2011-02-13 22:28:51.264317328 +0100
+++ libdlo/src/dlo_mode.c	2011-02-13 22:32:05.264625561 +0100
@@ -838,8 +838,8 @@ static dlo_retcode_t unpack_edid_detaile
    monitor->vBlanking       = ptr[0x06] + ((ptr[0x07] & 0x0F) << 8);
    monitor->hSyncOffset     = ptr[0x08] + ((ptr[0x0B] & 0xC0) << 2);
    monitor->hSyncPulseWidth = ptr[0x09] + ((ptr[0x0B] & 0x30) << 4);
-  monitor->vSyncOffset     = (ptr[0x0A] >> 4) + ((ptr[0x0B] & 0x0C) << 6);
-  monitor->vSyncPulseWidth = (ptr[0x0A] & 0x0F) + ((ptr[0x0B] & 0x03) << 8);
+  monitor->vSyncOffset     = (ptr[0x0A] >> 4) + ((ptr[0x0B] & 0x0C) << 2);
+  monitor->vSyncPulseWidth = (ptr[0x0A] & 0x0F) + ((ptr[0x0B] & 0x03) << 4);
    monitor->hImageSizeMm    = ptr[0x0C] + ((ptr[0x0E] & 0xF0) << 4);
    monitor->vImageSizeMm    = ptr[0x0D] + ((ptr[0x0E] & 0x0F) << 8);
    monitor->hBorder         = ptr[0x0F];