GPU_PTX WRR+2W+ctrl+membar.gl-p0:p1:p2-xsys "Rf-w-sl2l2 Ctrl-d-sl2-Rsl2 Fr-w-sl2l2 Mgl-d-Wsl2-Wsl2 Ws-w-sl2l2" {} P0 | P1 | P2 ; mov.s32 r0, 2 | ld.cg.s32 r0, [r1] | mov.s32 r0, 1 ; st.cg.s32 [r1], r0 | and.b32 r2, r0 2147483648 | st.cg.s32 [r1], r0 ; | setp.eq.b32 p2, r2, 0 | membar.gl ; | @p2 bra LC00 | mov.s32 r2, 1 ; | mov.s64 r4, 3494 | st.cg.s32 [r3], r2 ; | LC00: | ; | ld.cg.s32 r3, [r4] | ; exists (0::x=2 /\ 1:r0=2 /\ 1:r3=0) (device (kernel (cta (warp P0) (warp P1) (warp P2)))) x:local; y:local Generated assembler //START _litmus_P2 mov.s32 sr02, 1 ; st.cg.s32 [sr12], sr02 ; membar.gl ; mov.s32 sr22, 1 ; st.cg.s32 [sr32], sr22 ; //START _litmus_P1 ld.cg.s32 sr01, [sr11] ; and.b32 sr21, sr01, 2147483648 ; setp.eq.b32 sp21, sr21, 0 ; @sp21 bra LC00 ; mov.s64 sr41, 3494 ; LC00: ld.cg.s32 sr31, [sr41] ; //START _litmus_P0 mov.s32 sr00, 2 ; st.cg.s32 [sr10], sr00 ; Binary '/tmp/dir3fa1c2.tmp/WRR+2W+ctrl+membar.gl-p0:p1:p2-xsys-0.exe' successfully loaded Specification clusters: 3 Specification: [[(532, 2, 'R8'), (534, 2, 'R10')], [(592, 3, 'R4'), (593, 3, 'R5')], [(645, 2, 'R8')]] Cluster 0: OK Cluster 1: OK Cluster 2: OK !!SUCCESS!! Test WRR+2W+ctrl+membar.gl-p0:p1:p2-xsys Allowed Histogram (4 states) 15787 :>1:r0=0; 1:r3=0; 0::x=1; 5887 :>1:r0=2; 1:r3=0; 0::x=1; 37847 :>1:r0=0; 1:r3=1; 0::x=1; 40479 :>1:r0=2; 1:r3=1; 0::x=1; No Witnesses Positive: 0, Negative: 100000 Condition exists (0::x=2 /\ 1:r0=2 /\ 1:r3=0) is NOT validated Observation WRR+2W+ctrl+membar.gl-p0:p1:p2-xsys Never 0 100000 Time WRR+2W+ctrl+membar.gl-p0:p1:p2-xsys 27.68 -- device info -- Name: GeForce GTX 660 Major revision number: 3 Minor revision number: 0 Total global memory: 2147155968 Total shared memory per block: 49152 Number of multiprocessors: 5 Machine: csu8975 Revision 967, version 5.01 GPU_PTX experimental backend Command line: ./litmus -gpu -ccopts -arch=sm_30 -gpu-max-ctas 8 -gpu-min-ctas 4 tests-sm30/WRR+2W+ctrl+membar.gl-p0:p1:p2-xsys-0.litmus Parameters #ifndef SIZE_OF_TEST #define SIZE_OF_TEST 100000 #endif #ifndef NUMBER_OF_RUN #define NUMBER_OF_RUN 1 #endif /* cc options: -arch=sm_30 */ /* 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: 8 */ /* gpu_min_ctas: 4 */ /* gpu_max_threads: 512 */ /* gpu_min_threads: 128 */