To recap, I have four rails showing signs of trouble which are currently disabled in firmware on the supervisor while I focus on bringing up other parts of the design.
1) 1V2: powers MGTXAVTT for the FPGA transceivers plus the 1000baseT PHY. It powers up but only to ~500 mV. Unsure if problem with the feedback network or shorted so far.
2) GTX_1V0: powers MGTXAVCC for the FPGA transceivers, acting unstable and jumping all over the place.
3) GTX_1V8: powers MGTXAVCCAUX for the FPGA transceivers, not coming up at all
4) 3V0_N: regulated from +3V3, powers a single comparator for high speed CDR trigger. Not coming up at all.
Supplies 1-3 are Murata MYMGK00504ERSR, 4 is a LMR70503.
Suspecting solder defects as the root cause on all of them.