On Tue, 25 Feb 2025 12:04:40 -0800 Richard Henderson <richard.hender...@linaro.org> wrote:
> On 2/25/25 12:02, Richard Henderson wrote: > > Not checked here are any of the other reasons a flush might be ok: > > > > (2) The system as a whole is stopped, on the way in from migration/vmload. > > (3) The cpu is offline, on the way in from poweroff/reset. > (4) Running in round-robin mode, so there is *never* a race between cpus. > > Anything else I've forgotten? looking at x86 reset path * we have 2 resets per vcpu from main loop, when vcpu is created (at realize time and at system_reset time). this probably are nop for tcg as you said. (it likely applies to all targets) * And also a reset triggered by IPI (run in vcpu thread), which likely should do flush to clear whatever context vcpu had before reset.
