This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

MULSHIFT32 in 14 cycles

I have asked just everyone if their is a fast way to find the top 32-bits of a 32-bit x 32-bit multiply? There were multiply instructions that returned all 64 bits but they take 17 or 18 cycles doing something not used so:

MULSHIFT32:
                      lsrs r3, r0, #16 //Factor0 hi [16:31]
                      uxth r0, r0 //Factor0 lo [0:15]
                      uxth r2, r1 //Factor1 lo [0:15]
                      lsrs r1, r1, #16 //Factor1 hi [16:31]

                      muls r0, r1 //Factor0 lo * Factor1 hi
                      muls r2, r3 //Factor1 lo * Factor0 hi
                      muls r1, r3 //Factor1 hi * Factor0 hi

                      adds r0, r2 //(Factor0 lo * Factor1 hi) + (Factor1 lo * Factor0 hi)

                      movs r2, #0 //
                      adcs r2, r2 //C --> bit 16 (r2 contains $00000000 or $00010000)
                      lsls r2, r2, #16 //

                      lsrs r3, r0, #16 //Extract partial result [bits 16-31]

                      adds r2, r3 //Partial [bits 16-47]
                      adds r1, r2 //Results [bit 32-63]

Now the problem I have is that I cannot find my copy of the red book (Joseph Yiu's book on programming the M0 & M0+).The fact that it currently takes 4 instructions to move C into bit 16 of a register looks like it MAY be possible to speed up so that rather than two ADDS at the end, a single ADDS Rd, Rn, Rm since all registers are low.

So, now we are getting somewhere. I should add that my good friend Sarah Avory wrote the logic in C and simply tested it with every possible value to check it was correct. She was also able to save a cycle which seems tiny by todays standards, but in certain applications, the MULSHIFT32 is used millions of times a second.