optimization - gcc optimisation with LEA -

this question has answer here:

i'm fiddling gcc's optimisation options , found these lines:

int bla(int moo) {   return moo * 384; }

are translated these:

0:   8d 04 7f                lea    (%rdi,%rdi,2),%eax 3:   c1 e0 07                shl    $0x7,%eax 6:   c3                      retq

i understand shifting represents multiplication 2^7. , first line must multiplication 3.

so utterly perplexed "lea" line. isn't lea supposed load address?

lea (%ebx, %esi, 2), %edi nothing more computing ebx + esi*2 , storing result in edi.

even if lea designed compute , store effective address, can , used optimization trick perform calculation on not memory address.

lea    (%rdi,%rdi,2),%eax shl    $0x7,%eax

is equivalent :

eax = rdi + rdi*2; eax = eax * 128;

and since moo in rdi, stores moo*384 in eax

Brande