Lines Matching full:instruction
30 // If the "sub" instruction all ready sets (or could be modified to set) the
31 // same flag that the "cmp" instruction sets and that "bz" uses, then we can
32 // eliminate the "cmp" instruction.
40 // If the branch instruction can use flag from "sub", then we can replace
41 // "sub" with "subs" and eliminate the "cmp" instruction.
54 // Loads that can be folded into a later instruction. A load is foldable
138 /// optimizeExtInstr - If instruction is a copy-like instruction, i.e. it reads
185 // Uses that are in the same BB of uses of the result of the instruction.
188 // Uses that the result of the instruction can reach.
223 // implicit zext occurs. It doesn't insert a zext instruction. If we allow
301 /// optimizeBitcastInstr - If the instruction is a bitcast instruction A that
304 /// bitcast instruction B. And if the register class of source of B matches
305 /// the register class of instruction A, then it is legal to replace all uses
336 assert(Def && Src && "Malformed bitcast instruction!");
373 /// optimizeCmpInstr - If the instruction is a compare and the previous
374 /// instruction it's comparing against all ready sets (or could be modified to
376 /// the flag from the previous instruction.
379 // If this instruction is a comparison against zero and isn't comparing a
388 // Attempt to optimize the comparison instruction.
397 /// Optimize a select instruction.
415 /// instruction. We only fold loads to virtual registers and the virtual
515 // If there exists an instruction which belongs to the following
549 // instruction. If MI is not a candidate, check whether we can fold an
