GPU_PTX ISA2+membar.cta+po+po-p0:p1:p2-xsygzg "Mcta-d-Ws-Wg Rf-w-g Po-d-Rg-Wg Rf-w-g Po-d-Rg-Rs Fr-w-s" {} 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 | ld.cg.s32 r2, [r3] ; membar.cta | st.cg.s32 [r3], r2 | ; mov.s32 r2, 1 | | ; st.cg.s32 [r3], r2 | | ; exists (1:r0=1 /\ 2:r0=1 /\ 2:r2=0) (device (kernel (cta (warp P0) (warp P1) (warp P2)))) x:local; y:global; z:global Generated assembler //START _litmus_P2 ld.cg.s32 sr02, [sr12] ; 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 ; membar.cta ; mov.s32 sr20, 1 ; st.cg.s32 [sr30], sr20 ; Binary '/tmp/dir47bd0a.tmp/ISA2+membar.cta+po+po-p0:p1:p2-xsygzg-0.exe' successfully loaded Specification clusters: 3 Specification: [[(671, 3, 'R4'), (669, 3, 'R0')], [(728, 3, 'R0'), (725, 2, 'R12')], [(786, 2, 'R12'), (788, 2, 'R14')]] Cluster 0: OK Cluster 1: OK Cluster 2: OK !!SUCCESS!! Test ISA2+membar.cta+po+po-p0:p1:p2-xsygzg Allowed Histogram (8 states) 17806 :>1:r0=0; 2:r0=0; 2:r2=0; 3397 :>1:r0=1; 2:r0=0; 2:r2=0; 3998 :>1:r0=0; 2:r0=1; 2:r2=0; 66 *>1:r0=1; 2:r0=1; 2:r2=0; 12334 :>1:r0=0; 2:r0=0; 2:r2=1; 6308 :>1:r0=1; 2:r0=0; 2:r2=1; 5434 :>1:r0=0; 2:r0=1; 2:r2=1; 657 :>1:r0=1; 2:r0=1; 2:r2=1; Ok Witnesses Positive: 66, Negative: 49934 Condition exists (1:r0=1 /\ 2:r0=1 /\ 2:r2=0) is validated Observation ISA2+membar.cta+po+po-p0:p1:p2-xsygzg Sometimes 66 49934 Time ISA2+membar.cta+po+po-p0:p1:p2-xsygzg 18.02 -- 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/ISA2+membar.cta+po+po-p0:p1:p2-xsygzg-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 */