Dose mstatus.mprv affect the trigger tdata1 fire mode？

[AlexGJL]

Trigger register tdata1(e.g. mcontrol6.s) make trigger must fire in S/HS-mode. but mstatus.mprv could affect the “privilege mode”，1) dose mstatus.mprv affect the mode that trigger fires in? 2)do dcsr.mprven and mnstatus.nmie affect the mode that trigger fires in?

[pdonahue-ventana]

The mode filters in tdata1 refer to the actual mode, not the effective mode. See https://lists.riscv.org/g/tech-debug/topic/111192085#msg1815 and various other messages and github links that are mentioned in that thread.

dcsr.mprven affects the effective mode. Because of the answer to 1, the value of mprven has no effect on the behavior of something like mcontrol6.s because only the actual mode matters.

I don't understand why mnstatus.NMIE would affect modes at all. Regardless of the value of NMIE, everything will behave exactly the same except that the RNMI will not trap if NMIE=0. The only debug thing that's indirectly affected is that itrigger won't fire if itrigger.nmi=1 and mnstatus.NMIE=0 because there will be no trap and itrigger is defined to fire only when an interrupt trap is taken. This doesn't have anything to do with privilege modes.

[AlexGJL]

Thank you for your replay.

form nmie, There is such a description in The RISC-V Instruction Set Manual: Volume II： "When NMIE=0, the hart behaves as though mstatus.MPRV were clear, regardless of the current setting of mstatus.MPRV." ，which means it could affect mstatus.MPRV work。as the mode filters in tdata1 refer to the actual mode, nmie also not affect triggers fire.

[pdonahue-ventana]

Now your question makes sense to me. But the answer to question 1 makes the dcsr.mprven and mnstatus.NMIE questions moot.