On 05/10/2020 16:45, Thierry Reding wrote: ... >>> Let Jon and myself do a bit of testing with this to verify that the wake >>> up paths are still working. >> >> Sure. Let me know what you find. > > The results are in and it's a bit of a mixed bag. I was able to confirm > that Tegra194 also boots again after this series and I'm also able to > resume from sleep using either rtcwake or the power-key as wakeup > source, so the wake-events mechanism is still functional after the > series. I do see a bit of breakage on resume, but none of that seems > related to your patches and is likely something that crept in while we > were looking into the current issue. > > Jon had started a job in our test farm in parallel and that came back > with a failing suspend/resume test on Tegra186 (Jetson TX2), but that > seems to have been a pre-existing issue. This was already in linux-next > around next-20200910 and Jon had been investigating it when the boot > failures due to the IPI changes started happening. So I then hooked up > my Jetson TX2 and verified locally that I can properly suspend/resume > using either rtcwake or the power-key as wakeup source, just like I > previously did on Tegra194 (Jetson AGX Xavier). Tegra186 seems to be a > little more unstable because it didn't boot every time for me, but that > is probably not related to this. Yes my feeling is that those are other issues too that we need to look at next. > So, I'm tempted to say: > > Tested-by: Thierry Reding <treding@xxxxxxxxxx> Yes and you can have my ... Tested-by: Jon Hunter <jonathanh@xxxxxxxxxx> Thanks again Marc for tracking this down! Cheers Jon -- nvpublic
