I did programm differen apps. in c++ and assembler (NASM) and the problem is all the same: While a function uses a system or RTC delay an other core computes the code after that delay, there could also be a second delay after some instructions and that delay will break the execution of that app. e.g. in FIFO scheluling, too. I did read now the specifications of initialisation from AMD multicore chips for BIOS. There is a bit but not for cores 0 that can disable code execution of that cores. That bit is disabled on power on or reset, too. Now I thought if it would be an advantage for RT app. that run in FIFO scheduling, if linux does disable all cores. On the other hand it would be better to have a syscall for switching to mono core mode. But then the app. has to look if it runs on a multicore machine. bastl -- To unsubscribe from this list: send the line "unsubscribe linux-x86_64" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
