GPU_PTX ISA2+po+addr+membar.cta-p0::p1:p2-xgygzs "Po-d-Wg-Wg Rf-c-g Addr-d-g-Ws Rf-w-s Mcta-d-Rs-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 | and.b32 r2, r0 2147483648 | membar.cta ; mov.s32 r2, 1 | cvt.u64.u32 r3, r2 | ld.cg.s32 r2, [r3] ; st.cg.s32 [r3], r2 | mov.s32 r4, 1 | ; | add.u64 r5, r5 r3 | ; | st.cg.s32 [r5], r4 | ; exists (1:r0=1 /\ 2:r0=1 /\ 2:r2=0) (device (kernel (cta (warp P0)) (cta (warp P1) (warp P2)))) x:global; y:global; z:local 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] ; and.b32 sr21, sr01, 2147483648 ; cvt.u64.u32 sr31, sr21 ; mov.s32 sr41, 1 ; add.u64 sr51, sr51, sr31 ; st.cg.s32 [sr51], sr41 ; //START _litmus_P0 mov.s32 sr00, 1 ; st.cg.s32 [sr10], sr00 ; mov.s32 sr20, 1 ; st.cg.s32 [sr30], sr20 ; Binary '/tmp/dira49ec5.tmp/ISA2+po+addr+membar.cta-p0::p1:p2-xgygzs-0.exe' successfully loaded Specification clusters: 3 Specification: [[(798, 3, 'R0'), (799, 3, 'R4')], [(880, 3, 'R0'), (878, 2, 'R4')], [(945, 2, 'R10'), (947, 2, 'R8')]] Cluster 0: OK Cluster 1: OK Cluster 2: OK !!SUCCESS!! Test ISA2+po+addr+membar.cta-p0::p1:p2-xgygzs Allowed Histogram (7 states) 25246 :>1:r0=0; 2:r0=0; 2:r2=0; 543 :>1:r0=1; 2:r0=0; 2:r2=0; 1271 :>1:r0=0; 2:r0=1; 2:r2=0; 10144 :>1:r0=0; 2:r0=0; 2:r2=1; 10681 :>1:r0=1; 2:r0=0; 2:r2=1; 1890 :>1:r0=0; 2:r0=1; 2:r2=1; 225 :>1:r0=1; 2:r0=1; 2:r2=1; No Witnesses Positive: 0, Negative: 50000 Condition exists (1:r0=1 /\ 2:r0=1 /\ 2:r2=0) is NOT validated Observation ISA2+po+addr+membar.cta-p0::p1:p2-xgygzs Never 0 50000 Time ISA2+po+addr+membar.cta-p0::p1:p2-xgygzs 5.88 -- 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/ISA2+po+addr+membar.cta-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 */