You’d have to add some support to gem5 to segment off a piece of memory for the AMP and load some code for it to run. Right now all cores jump into a boot loader where they all spin except cpu0 which boots linux. When linux is ready to bring up the other cores it sends them an interrupt and they read a predefined location to find their starting address. You could modify the boot loader to alway jump to a known location and tell linux there is only 1 cpu, but you’d still need to segment off a piece of memory.
Ali On Aug 20, 2014, at 3:09 PM, Namitha Krishna via gem5-users <gem5-users@gem5.org> wrote: > Hi, > > Can anybody tell me how I can use Gem5 for a full system simulation of a > dual-core cortex-A9 in Asymmmetric multiprocessing mode when one core is > running Linux and the other core is running a bare-metal application. On > hardware, the linux core which is the master has to kick the second core with > a SEV (Send event) . Should this be done on Gem5 as well? Where can we > specify the .elf for the bare metal core if the linux image needs to be > passed to --kernel on the command line ? If the number of cores are > increased, where should the .elf files for these cores be specified? > Can somebody please elaborate the steps and the python scripts to be modified > please? > > regards, > Namitha. > _______________________________________________ > gem5-users mailing list > gem5-users@gem5.org > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
