This is a good point. I think this won't happen though, because smaller nodes are that much more expensive, and the market for energy-efficient 32bit CPUs isn't nearly large enough to fund development. There likely won't be another ARMv7.

What I think will happen is that the huge amount of money that goes into developing mobile phone SoCs, which are clearly trending towards 64Bit while still requiring low power consumption, will cause that technology to trickle down and be used in ever lower cost markets as time passes and those initial costs have been recouped. The Apple M1, as a system, is clearly crazy power efficient, even at 5nm. And again, if you need an ultra-low power sleep mode, because you only wake up once a day to send some sensor data out, its probably not running Linux in the first place.

I think you get into practical problems at some point. WikiChip lists a 28nm Cortex-A7 core at 0.48mm², with every node shrink (20nm, 14nm, 10nm, 7nm, ...) you can halve that, but adding 64-bit support to a core might only add about 10%.

You can't make chips arbitrarily small, because you end up wasting more of your wafer for the area between the dies and for the wire bond pads. If the chip size stays the same but the CPU gets smaller, most of the chip is for off-core components and the CPU has less impact on the total cost and power consumption.

You could use more complex CPU cores, or simply more cores, for better performance, but then you also need a faster memory interface (wider buses, LD-DDR4+ instead of DDR3, ...) to make actually use of the performance. These lead to higher memory capacity as well, but then you can't actually use the available memory as you run into 32-bit addressing limits.

I wouldn't rule out shrinks of existing 32-bit SoC families to 22nm or below to lower cost once those processes get cheap enough and there is still demand for compatibility, but there is a good chance that 28nm SoCs is where 32-bit ends. (note that I'm not talking about the dozen or so additional ARC/Xtensa/RV32/Cortex-M/... microcontroller cores on high-end SoCs, as those are not the ones running Linux).

Why do you think it's always more energy efficient?

Because less transistors need to switch.

I'm honestly curious how that plays out in practice. 64b Intel CPUs can run in 32b mode. If we count actual switching then we would end up with ~same usage. But in 64b mode it gives us both longer pointers and offsets it with fewer instructions on vector operations.

I could not find any actual paper with real use case comparison. Have you got one?

I haven't got a reference, and would be interested in one too.

Sure, if you use any kind of 32bit mode then you'd use less energy in theory. But if we leave the responsibility to the software developer, then we all know what happens :)

How much energy would this save? Of all the energy usage on the planet, this seems like it would be really really tiny.