I followed the procedure as described in the post referenced by everyone:
https://www.eevblog.com/forum/testgear/siglent-sdg2042x-hack-door-closed/msg2409867/#msg2409867Before applying the hack my firmware version was: 2.01.01.23R8.
The model was upgraded and 120 MHz are available successfully, with the same version showing.
Later I updated to the latest firmware from
https://siglentna.com/service-and-support/firmware-software/waveform-generators/#sdg2000x-serieswhich was 2.01.01.35R3B2 (Release Date 12.16.19 ) at the time of this writing.
Again, it worked successfully, although it restarted twice, and appeared to hang for a while, doing nothing. So don't be afraid if it does that, it's not a failure.
Small caveat:
Above and close to 40 MHz the voltage (peak to peak) measured on the 100MHz scope drops as the frequency increases. This was with proper 50 Ohm termination on the scope (suitable for DC-1GHz) and 50 Ohm output on the signal generator. Before doing the test, the scope was calibrated again.
For a 2Vpp output sine wave, it will drop roughly 100 mV (varies a bit) until it reaches 100 MHz, above that it slowly starts to drop a bit more, peaking at a loss of roughly 180 mV.