GPU_PTX 3.LB+membar.cta+ctrl+addr-p0:p1:p2-xsyszg "Mcta-d-Rsl2-Wsl2 Rf-w-sl2l2 Ctrl-d-sl2-Wgl2 Rf-w-gl2l2 Addr-d-gl2-Wsl2 Rf-w-sl2l2" {} P0 | P1 | P2 ; ld.cg.s32 r0, [r1] | ld.cg.s32 r0, [r1] | ld.cg.s32 r0, [r1] ; membar.cta | and.b32 r2, r0 2147483648 | and.b32 r2, r0 2147483648 ; mov.s32 r2, 1 | setp.eq.b32 p1, r2, 0 | cvt.u64.u32 r3, r2 ; st.cg.s32 [r3], r2 | @p1 bra LC00 | mov.s32 r4, 1 ; | mov.s64 r4, 3494 | add.u64 r5, r5 r3 ; | LC00: | st.cg.s32 [r5], r4 ; | mov.s32 r3, 1 | ; | st.cg.s32 [r4], r3 | ; exists (0:r0=1 /\ 1:r0=1 /\ 2:r0=1) (device (kernel (cta (warp P0) (warp P1) (warp P2)))) x:local; y:local; z:global Generated assembler //START _litmus_P2 ld.cg.s32 sr02, [sr12] ; and.b32 sr22, sr02, 2147483648 ; cvt.u64.u32 sr32, sr22 ; mov.s32 sr42, 1 ; add.u64 sr52, sr52, sr32 ; st.cg.s32 [sr52], sr42 ; //START _litmus_P1 ld.cg.s32 sr01, [sr11] ; and.b32 sr21, sr01, 2147483648 ; setp.eq.b32 sp11, sr21, 0 ; @sp11 bra LC00 ; mov.s64 sr41, 3494 ; LC00: mov.s32 sr31, 1 ; st.cg.s32 [sr41], sr31 ; //START _litmus_P0 ld.cg.s32 sr00, [sr10] ; membar.cta ; mov.s32 sr20, 1 ; st.cg.s32 [sr30], sr20 ; Binary '/tmp/dir2fdd5d.tmp/3.LB+membar.cta+ctrl+addr-p0:p1:p2-xsyszg-0.exe' successfully loaded Specification clusters: 3 Specification: [[(644, 3, 'R4'), (646, 2, 'R8')], [(704, 3, 'R4'), (703, 2, 'R12')], [(766, 3, 'R4'), (765, 2, 'R12')]] Cluster 0: OK Cluster 1: OK Cluster 2: OK !!SUCCESS!! Test 3.LB+membar.cta+ctrl+addr-p0:p1:p2-xsyszg Allowed Histogram (7 states) 51082 :>0:r0=0; 1:r0=0; 2:r0=0; 2735 :>0:r0=1; 1:r0=0; 2:r0=0; 18338 :>0:r0=0; 1:r0=1; 2:r0=0; 6 :>0:r0=1; 1:r0=1; 2:r0=0; 24847 :>0:r0=0; 1:r0=0; 2:r0=1; 298 :>0:r0=1; 1:r0=0; 2:r0=1; 2694 :>0:r0=0; 1:r0=1; 2:r0=1; No Witnesses Positive: 0, Negative: 100000 Condition exists (0:r0=1 /\ 1:r0=1 /\ 2:r0=1) is NOT validated Observation 3.LB+membar.cta+ctrl+addr-p0:p1:p2-xsyszg Never 0 100000 Time 3.LB+membar.cta+ctrl+addr-p0:p1:p2-xsyszg 38.72 -- device info -- Name: GeForce GTX 660 Major revision number: 3 Minor revision number: 0 Total global memory: 2147155968 Total shared memory per block: 49152 Number of multiprocessors: 5 Machine: csu8975 Revision 967, version 5.01 GPU_PTX experimental backend Command line: ./litmus -gpu -ccopts -arch=sm_30 -gpu-max-ctas 8 -gpu-min-ctas 4 tests-sm30/3.LB+membar.cta+ctrl+addr-p0:p1:p2-xsyszg-0.litmus Parameters #ifndef SIZE_OF_TEST #define SIZE_OF_TEST 100000 #endif #ifndef NUMBER_OF_RUN #define NUMBER_OF_RUN 1 #endif /* cc options: -arch=sm_30 */ /* barrier: user */ /* speedcheck: no */ /* gpu_general_bc: true */ /* gpu_mem_stress: true */ /* gpu-rand-threads: true */ /* gpu_cuda: true */ /* gpu_gen_addr: true */ /* gpu_cvt_global_to_gen: true */ /* gpu_cvt_shared_to_gen: true */ /* gpu_max_ctas: 8 */ /* gpu_min_ctas: 4 */ /* gpu_max_threads: 512 */ /* gpu_min_threads: 128 */