GPU_PTX Z6.3+membar.gl+membar.cta+addr-p0:p1:p2-xgyszs "Mgl-d-Wg-Ws Ws-w-s Mcta-d-Ws-Ws Rf-w-s Addr-d-s-Rg Fr-w-g" {} P0 | P1 | P2 ; mov.s32 r0, 1 | mov.s32 r0, 2 | ld.cg.s32 r0, [r1] ; st.cg.s32 [r1], r0 | st.cg.s32 [r1], r0 | and.b32 r2, r0 2147483648 ; membar.gl | membar.cta | cvt.u64.u32 r3, r2 ; mov.s32 r2, 1 | mov.s32 r2, 1 | add.u64 r5, r5 r3 ; st.cg.s32 [r3], r2 | st.cg.s32 [r3], r2 | ld.cg.s32 r4, [r5] ; exists (0::y=2 /\ 2:r0=1 /\ 2:r4=0) (device (kernel (cta (warp P0) (warp P1) (warp P2)))) x:global; y:local; z:local Generated assembler //START _litmus_P2 ld.cg.s32 sr02, [sr12] ; and.b32 sr22, sr02, 2147483648 ; cvt.u64.u32 sr32, sr22 ; add.u64 sr52, sr52, sr32 ; ld.cg.s32 sr42, [sr52] ; //START _litmus_P1 mov.s32 sr01, 2 ; st.cg.s32 [sr11], sr01 ; membar.cta ; mov.s32 sr21, 1 ; st.cg.s32 [sr31], sr21 ; //START _litmus_P0 mov.s32 sr00, 1 ; st.cg.s32 [sr10], sr00 ; membar.gl ; mov.s32 sr20, 1 ; st.cg.s32 [sr30], sr20 ; Binary '/tmp/dir62c5d3.tmp/Z6.3+membar.gl+membar.cta+addr-p0:p1:p2-xgyszs-0.exe' successfully loaded Specification clusters: 3 Specification: [[(690, 3, 'R5'), (691, 3, 'R4')], [(757, 2, 'R10'), (759, 2, 'R12')], [(813, 2, 'R10'), (816, 2, 'R12')]] Cluster 0: OK Cluster 1: OK Cluster 2: OK !!SUCCESS!! Test Z6.3+membar.gl+membar.cta+addr-p0:p1:p2-xgyszs Allowed Histogram (4 states) 16299 :>2:r0=0; 2:r4=0; 0::y=1; 1095 :>2:r0=1; 2:r4=0; 0::y=1; 27520 :>2:r0=0; 2:r4=1; 0::y=1; 5086 :>2:r0=1; 2:r4=1; 0::y=1; No Witnesses Positive: 0, Negative: 50000 Condition exists (0::y=2 /\ 2:r0=1 /\ 2:r4=0) is NOT validated Observation Z6.3+membar.gl+membar.cta+addr-p0:p1:p2-xgyszs Never 0 50000 Time Z6.3+membar.gl+membar.cta+addr-p0:p1:p2-xgyszs 13.46 -- device info -- Name: GeForce GTX TITAN Major revision number: 3 Minor revision number: 5 Total global memory: 6442123264 Total shared memory per block: 49152 Number of multiprocessors: 14 Machine: formal Revision 1383, version 5.01 GPU_PTX experimental backend Command line: ./litmus -mach GTXTitan.cfg -a 8 -gpu-verf-comp true big_bucket/Z6.3+membar.gl+membar.cta+addr-p0:p1:p2-xgyszs-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_35 */ /* 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: 14 */ /* gpu_min_ctas: 10 */ /* gpu_max_threads: 512 */ /* gpu_min_threads: 128 */