http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58450
H.J. Lu <hjl.tools at gmail dot com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |WAITING Last reconfirmed| |2013-09-17 CC| |hjl.tools at gmail dot com Ever confirmed|0 |1 --- Comment #2 from H.J. Lu <hjl.tools at gmail dot com> --- This is related to PR 57988. Without -fno-trapping-math, there are 0x00000000004009b6 <+54>: cvtsi2sd %rdx,%xmm2 0x00000000004009bb <+59>: je 0x400998 <_Z6pi_apxv+24> 0x00000000004009bd <+61>: movapd %xmm1,%xmm4 0x00000000004009c1 <+65>: add $0x1,%rax 0x00000000004009c5 <+69>: add $0x2,%rdx 0x00000000004009c9 <+73>: cmp $0x5f5e100,%rax 0x00000000004009cf <+79>: divsd %xmm2,%xmm4 It hides the latency of cvtsi2sd %rdx,%xmm2 due to partial SSE register stall. Please try the current trunk.