Dear Peter Maydell, Thank you for your detailed response.
We will take a closer look at why, when using WFI, the affected cores still appear to be at 100% utilization in htop. Additionally, we will investigate whether implementing WFE in QEMU would be necessary to achieve proper CPU core shutdown. Currently, we are not using PSCI, but this does seem like a promising approach that we will explore further. Regarding the use of the "virt" machine, our supervisor, M. Poquet, requires the Raspberry Pi 3B specifically for his courses due to its hardware specifications. Therefore, switching to "virt" would only be considered as a last resort. Thank you again for your insights. Any further recommendations would be greatly appreciated. Best regards, Clément Aldebert & Laurent Polzin ----- Mail original ----- De: "Peter Maydell" <peter.mayd...@linaro.org> À: "clement aldebert" <clement.aldeb...@univ-tlse3.fr> Cc: "Alex Bennée" <alex.ben...@linaro.org>, "qemu-devel" <qemu-devel@nongnu.org>, "millian.poquet" <millian.poq...@univ-tlse3.fr>, "laurent polzin" <laurent.pol...@univ-tlse3.fr> Envoyé: Jeudi 20 Mars 2025 21:11:52 Objet: Re: Raspberry Pi 3B energy consumption On Thu, 20 Mar 2025 at 20:09, Peter Maydell <peter.mayd...@linaro.org> wrote: > If you need WFE to work, that's certainly feasible and something it would > be nice to see, but potentially quite a bit of work in the guts of QEMU's > arm emulation. (Basically going to sleep on WFE is easy but then making > sure that all the events and situations that need to wake up a WFE is > tedious. We implement sleep-on-WFI but not sleep-on-WFI because the set should read "sleep-on-WFI but not sleep-on-WFE", of course. Oops... > of WFI-wakeup events is rather smaller than the WFE-wakeup events.) It's > been in the "we really should implement this but since the only downside > is the host CPUs spinning, we've never got round to it" bucket for years. -- PMM
