Hello Everyone, I am trying to run the AMD APU on the public branch. I am using polybench-gpu 1.0 (OpenCL) as benchmark. When I use a dram size (mem-size) greater than 8GB, the GPU output all become 0, while any smaller value generates the correct output.
Is there an inherent limit on the dram size that I can simulate ? If not the what could be the reason ? Any pointers would be appreciated. Here's the command line config that I am using: $gem5/build/HSAIL_X86/gem5.fast $gem5/configs/example/apu_se.py --network=garnet2.0 --topology=Mesh_XY --num-cpus=28 --num-dirs=16 --cmd=polybench-gpu-1.0/OpenCL/3DCONV/3DConvolution.exe --num-compute-units=28 --num-tccs=8 --mem-size=32GB Thanks and Regards, Sampad Mohapatra [https://ipmcdn.avast.com/images/icons/icon-envelope-tick-green-avg-v1.png]<http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail> Virus-free. www.avg.com<http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
_______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
