I’m not positive, but I think that patch was under R28.1. It may be that if you use R28.2 the patch is already there…I don’t know.
As for the rest of the steps “it depends”. Is your display currently sending an EDID? If there is no EDID, and you have another monitor which does have EDID, and if that monitor is put in a mode common and acceptable to both monitors and to the driver, then upon switching from one monitor with that EDID to another monitor which does not send EDID, then the mode will probably remain what it was…as a result of the two monitors having that mode in common, then the display should operate correctly until reboot or some other mode change.
You really must start by saying whether you have a monitor without EDID, or if you have a monitor with an EDID which is being rejected. The approach will differ. Also, there may be differences depending on which L4T release you use…I think R28.2 has improved some video aspects and so if your monitor simply has an EDID which isn’t liked by R28.1, then R28.2 might fix it.
One way to see EDID content:
cat `find /sys -name edid`
…if there is hex data, then its nature can be seen by pasting it into http://www.edidreader.com.
If there is hex data, then the note I made in a previous post about adding Option “ModeDebug” to Section “Device” would give you a way to see what the actual driver thinks about the modes presented in EDID. This may also be useful when exploring a different monitor if looking at modes which might work for you.