GPU_PTX WRC+po+membar.cta-p0:p1::p2-xgyg "Rf-w-g Po-d-Rg-Wg Rf-c-g Mcta-d-Rg-Rg Fr-c-g" {} P0 | P1 | P2 ; mov.s32 r0, 1 | ld.cg.s32 r0, [r1] | ld.cg.s32 r0, [r1] ; st.cg.s32 [r1], r0 | mov.s32 r2, 1 | membar.cta ; | st.cg.s32 [r3], r2 | ld.cg.s32 r2, [r3] ; exists (1:r0=1 /\ 2:r0=1 /\ 2:r2=0) (device (kernel (cta (warp P0) (warp P1)) (cta (warp P2)))) x:global; y:global Generated assembler //START _litmus_P2 ld.cg.s32 sr02, [sr12] ; membar.cta ; ld.cg.s32 sr22, [sr32] ; //START _litmus_P1 ld.cg.s32 sr01, [sr11] ; mov.s32 sr21, 1 ; st.cg.s32 [sr31], sr21 ; //START _litmus_P0 mov.s32 sr00, 1 ; st.cg.s32 [sr10], sr00 ; Binary '/tmp/dir5287f0.tmp/WRC+po+membar.cta-p0:p1::p2-xgyg-0.exe' successfully loaded Specification clusters: 3 Specification: [[(524, 3, 'R4'), (523, 3, 'R5')], [(580, 3, 'R4'), (579, 2, 'R10')], [(625, 2, 'R10')]] Cluster 0: OK Cluster 1: OK Cluster 2: OK !!SUCCESS!! Test WRC+po+membar.cta-p0:p1::p2-xgyg Allowed Histogram (8 states) 9419 :>1:r0=0; 2:r0=0; 2:r2=0; 3490 :>1:r0=1; 2:r0=0; 2:r2=0; 3530 :>1:r0=0; 2:r0=1; 2:r2=0; 366 *>1:r0=1; 2:r0=1; 2:r2=0; 8370 :>1:r0=0; 2:r0=0; 2:r2=1; 10545 :>1:r0=1; 2:r0=0; 2:r2=1; 10007 :>1:r0=0; 2:r0=1; 2:r2=1; 4273 :>1:r0=1; 2:r0=1; 2:r2=1; Ok Witnesses Positive: 366, Negative: 49634 Condition exists (1:r0=1 /\ 2:r0=1 /\ 2:r2=0) is validated Observation WRC+po+membar.cta-p0:p1::p2-xgyg Sometimes 366 49634 Time WRC+po+membar.cta-p0:p1::p2-xgyg 16.86 -- 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/WRC+po+membar.cta-p0:p1::p2-xgyg-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 */