https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117722
--- Comment #1 from JuzheZhong <juzhe.zhong at rivai dot ai> --- OK. I see we are lacking ssadd/usad pattern (SAD_EXPR): Compute the sum of absolute differences of two signed/unsigned elements. Operand 1 and operand 2 are of the same mode. Their absolute difference, which is of a wider mode, is computed and added to operand 3. Operand 3 is of a mode equal or wider than the mode of the absolute difference. The result is placed in operand 0, which is of the same mode as operand 3. m is the mode of operand 1 and operand 2. I think we implement this pattern by learning from Clang: https://godbolt.org/z/bvEKf4h1z vminu.vv v10, v8, v9 vmaxu.vv v8, v8, v9 vsub.vv v8, v8, v10 vsetvli zero, zero, e32, m4, ta, ma vzext.vf4 v12, v8 Since operand 3 is {0, 0, 0, 0}. So we don't need Addition here. For optimization, we should allow operand3 to be const vector. min + max + sub + extend + addition.
