Request for clarification on prefetch.global.tensormap docs

Accelerated Computing CUDA CUDA Programming and Performance
1

The PTX docs give the following syntax:

prefetch{.tensormap_space}.tensormap [a];  // prefetch the tensormap
.tensormap_space =          { .const, .param };

This indicates that only .const and .param spaces can be used. In the documentation text, it says

If the .tensormap qualifier is specified then the prefetch instruction brings the cache line containing the specified address in the .const or .param memory state space for subsequent use by the cp.async.bulk.tensor instruction.

Again, it sounds like only .const or .param can be used.

However, under the Examples section we have:

prefetch.global.tensormap      [ptr];

I would appreciate clarification – is tensormap prefetch from global space allowed or not?