On 09/13/2015 01:34 AM, Bastian Koppelmann wrote: > Looking at the article, user mode seems to be optional, so I'm not against it, > but it does look weird. How does ork1sim do it?
It's haphazard. There are checks for supervisor in the l_mtspr and l_mfspr insns, but not other uses of the sprs, such as l_rfe. In the meantime, testing of the multiply-accumulate instructions is hindered by the fact that linux-user can't read the MACHI and MACLO special registers. r~
