> Wouldn't it be better to use `pushl 4(%esp)' instead of loading it to %eax > and then pushing it (which probably causes pipeline stall as well)? cycles/ 386 486 pentium pentiumpro mov mem to reg 4 1 1 1 uop push reg 5 1 1 1 uop push mem 5 4 2 2 uop The 486 will stall in this case but you still come out ahead. Pentium and Pentiumpro won't stall.