Question: multiply takes a different number of cycles based on input

[Emoun]

While looking at which instruction selection optimizations we do, I fell on the following sentence:

// following mul by constant patterns only pay off if mul takes > 2 cycles
// and offers no immediate variant

This implies that the mul instruction requires a different number of cycles depending on the input.
Therefore, I have the following questions:

1. Does the mul actually take a different number of cycles based on input?
2. Does WCET analysis know about how many cycles the mul will take depending on input?

[Emoun]

After talking with @schoeberl, it seems this message was probably copied from a different target when the Patmos target was first implemented. Therefore, it is not relevant for the Patmos target.

The Patmos mul instruction always takes a constant number of cycles (@schoeberl thinks 3 or 4).