I will say, based on the posts, there seem to be a lot of people who want the main channel knobs to control gain (pre-fade gain). As noted, this can be done using Basic mode or by using Custom mode with gain set to basic.
Sound devices seems to get a lot of heat for their position on gain both here and on other discussion boards and have pulled the manual or tech note section they had on A Note About Gain. But I think they are right and for many people here it's worth considering what they were saying.
For perhaps the majority of the board who are recording concerts with the more usual mic techniques (i.e., not using a Soundfield mic or not using M-S recording), messing with your gain while recording the show probably doesn't help with the quality of the recording and just makes your life more difficult. Assuming of course, you are "mastering" the recording in post with some sound editing software and normalizing your levels. For instance, I spent many an hour during post-production trying to smooth out very noticeable and very jarring level changes I made while recording so the result on playback is much more smooth.
When I started DAT recording, recorders like a Sony D7 were very noisy, making it best to record near 0dbFS. On top of that, for myself, I transferred my DATs to CDR using a standalone Marantz CDR burner. There was no raising my levels in post, so if I recorded with peaks at -16dbFS, my CDR recording was very low. The situation is very different today, both with the use of post-production with sound editing software and with much, much improved recorders.
The Mixpre-6 has a noise level of -130dbV and has 120db of dynamic range -- very quiet, very low noise, very wide dynamic range. If we're recording concerts in large arenas or bars, the noise level of the environment without any music playing would be something like 60db being very generous, probably more like 65-70db (or more). And ear-splittingly loud, the concert with the music playing is 120db. So we only need 50-60db of dynamic range.
If the noise level of the club is say 65db and the internal noise of the MP6 is -130dbV, you pretty much cannot possibly set your levels so low that the internal noise of the MP6 is louder than the noise level of the recording environment. So take an educated guess and set your levels so that you expect peaks no louder than -40dbFS. Plenty of headroom so you don't clip, and you can boost/normalize your recording in post by 40db and it will sound no different than if you rode your levels and recorded at -3dbFS and only boosted 3db in post.
That is set your gain so that you expect your levels to be quite low with plenty of headroom and no chance of going over, and you have a very easy job of simply normalizing in post. Compared to finding each time you decided to increase gain using the front panel gain knobs (if you set up in basic gain mode) and somehow massaging that recorded passage to make any level changes unnoticeable for the listener.
Long story short, given the noise level and dynamic range of the mixpre-6 and especially given the noise level of a concert environment, I really don't see what is the expected benefit of making any adjustments to gain once the show has started. What are people trying to accomplish? Maybe there are valid reasons and for some taping situations it might make sense, but for most of us doing the recording we do and mastering/normalizing in post, I don't see the benefit of needing to set levels at all during the show. Set the MP6 at a very, very conservative amount of gain before the show starts and never change levels again. Or set the MP6 to advanced mode with very, very conservative pre-fade gain and use the front faders to make a LR post-fade recording if it makes you feel more comfortable. I'd bet for many people once they get used to the new way of doing things would just archive the LR mix recording and would use the ISO tracks with the very low levels for mastering and final distribution.