> I'm interested in the results of the following test.
DS10: /su : 1/3 -i--- /sui : 1/3 -i--- /su : min*min -i--u /sui : min*min -i--u / : (long)4.5 -i--- /sv : (long)4.5 -i--- /svi : (long)4.5 -i--- / : (long)max -i--- /sv : (long)max -iI-- /svi : (long)max -iI-- AS200: /su : 1/3 ----- /sui : 1/3 ----- /su : min*min -i--u /sui : min*min -i--u / : (long)4.5 ----- /sv : (long)4.5 ----- /svi : (long)4.5 ----- / : (long)max ----- /sv : (long)max --I-- /svi : (long)max --I-- qemu: /su : 1/3 -i--- /sui : 1/3 -i--- /su : min*min -i--u /sui : min*min -i--u / : (long)4.5 -i--- /sv : (long)4.5 -i--- /svi : (long)4.5 -i--- / : (long)max -i--- /sv : (long)max -iI-- /svi : (long)max -iI-- IOW, same as EV6. The difference is due to the kernel trusting FPCR.INE as source for FE_INEXACT when it sees implver() returning 2. See a bit upthread for analysis...