GPU_PTX 3.LB+addr+po+data-p0:p1:p2-xgygzs "Addr-d-g-Wg Rf-w-g Po-d-Rg-Ws Rf-w-s Data-d-s-Wg Rf-w-g" {} P0 | P1 | P2 ; ld.cg.s32 r0, [r1] | ld.cg.s32 r0, [r1] | ld.cg.s32 r0, [r1] ; and.b32 r2, r0 2147483648 | mov.s32 r2, 1 | and.b32 r2, r0 2147483648 ; cvt.u64.u32 r3, r2 | st.cg.s32 [r3], r2 | add.s32 r2, r2 1 ; mov.s32 r4, 1 | | st.cg.s32 [r3], r2 ; add.u64 r5, r5 r3 | | ; st.cg.s32 [r5], r4 | | ; exists (0:r0=1 /\ 1:r0=1 /\ 2:r0=1) (device (kernel (cta (warp P0) (warp P1) (warp P2)))) x:global; y:global; z:local Generated assembler //START _litmus_P2 ld.cg.s32 sr02, [sr12] ; and.b32 sr22, sr02, 2147483648 ; add.s32 sr22, sr22, 1 ; st.cg.s32 [sr32], sr22 ; //START _litmus_P1 ld.cg.s32 sr01, [sr11] ; mov.s32 sr21, 1 ; st.cg.s32 [sr31], sr21 ; //START _litmus_P0 ld.cg.s32 sr00, [sr10] ; and.b32 sr20, sr00, 2147483648 ; cvt.u64.u32 sr30, sr20 ; mov.s32 sr40, 1 ; add.u64 sr50, sr50, sr30 ; st.cg.s32 [sr50], sr40 ; Binary '/tmp/dir35e328.tmp/3.LB+addr+po+data-p0:p1:p2-xgygzs-0.exe' successfully loaded Specification clusters: 3 Specification: [[(798, 3, 'R0'), (796, 2, 'R12')], [(870, 3, 'R0'), (871, 2, 'R12')], [(939, 3, 'R0'), (937, 2, 'R4')]] Cluster 0: OK Cluster 1: OK Cluster 2: OK !!SUCCESS!! Test 3.LB+addr+po+data-p0:p1:p2-xgygzs Allowed Histogram (8 states) 15112 :>0:r0=0; 1:r0=0; 2:r0=0; 12273 :>0:r0=1; 1:r0=0; 2:r0=0; 677 :>0:r0=0; 1:r0=1; 2:r0=0; 62 :>0:r0=1; 1:r0=1; 2:r0=0; 16448 :>0:r0=0; 1:r0=0; 2:r0=1; 4093 :>0:r0=1; 1:r0=0; 2:r0=1; 1334 :>0:r0=0; 1:r0=1; 2:r0=1; 1 *>0:r0=1; 1:r0=1; 2:r0=1; Ok Witnesses Positive: 1, Negative: 49999 Condition exists (0:r0=1 /\ 1:r0=1 /\ 2:r0=1) is validated Observation 3.LB+addr+po+data-p0:p1:p2-xgygzs Sometimes 1 49999 Time 3.LB+addr+po+data-p0:p1:p2-xgygzs 5.73 -- device info -- Name: GeForce GTX 750 Ti Major revision number: 5 Minor revision number: 0 Total global memory: 2147287040 Total shared memory per block: 49152 Number of multiprocessors: 5 Machine: formal Revision 1383, version 5.01 GPU_PTX experimental backend Command line: ./litmus -mach GTX750.cfg -a 8 -gpu-verf-comp true big_bucket/3.LB+addr+po+data-p0:p1:p2-xgygzs-0.litmus -s 50000 Parameters #ifndef SIZE_OF_TEST #define SIZE_OF_TEST 50000 #endif #ifndef NUMBER_OF_RUN #define NUMBER_OF_RUN 1 #endif /* cc options: -arch=sm_50 */ /* 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: 5 */ /* gpu_min_ctas: 4 */ /* gpu_max_threads: 256 */ /* gpu_min_threads: 128 */