Some time ago I tried to migrate my laptop to FreeBSD. I've managed to install 13.0 and everything seemed to be fine, but then I started having these complete system freezes. Since I was running X, I wasn't seeing anything on the console. I didn't had too much time to investigate (and learn!), so I had to abort the migration. Today I tried to repeat the experiment with 13.1, on a Lenovo 13.1. The system freeze happened again while doing the installation: * I was configuring networking * iwm(4) scan didn't returned any wireless networks * plugged an ethernet cable to try to use the em(4) driver... * the system stopped responding, the laptop started to get hotter and the fans started I had to hard-reboot, of course. My question: I do I debug this? If I had to guess, it it's a network driver bug, either iwm(4) or em(4), but how can I get more info when this happens? I plan to try to re-install again without using the iwm(4) driver. Another detail (probably irrelevant): I'm installing with encrypted root-on-zfs.
(In reply to henrix from comment #0) > encrypted root-on-zfs. GELI encrypted, i.e. the method of encryption that's offered by the installer itself?
You can try latest CURRENT or disable hwpstate by setting hint.hwpstate_intel.0.disabled=1 in /boot/loader.conf See another bug #253288
(In reply to Graham Perrin from comment #1) > GELI encrypted, i.e. the method of encryption that's offered by the installer itself? Yes, that's correct. I've selected "Root-on-ZFS" and the changed the "Encrypt Disks?" to "YES".
(In reply to Oleksandr Kryvulia from comment #2) >You can try latest CURRENT or disable hwpstate by setting hint.hwpstate_intel.0.disabled=1 in /boot/loader.conf > See another bug #253288 Ah! Awesome! That you for your reply. I'll read through that bug, it could definitely be it. Disabling hwpstate doesn't sound great, but I'll need to go through all the comments. I guess the fix for that bug is already in CURRENT? I'm not really experienced in FreeBSD, so running CURRENT sounds scary for a laptop I want to be stable (it's been running OpenBSD for a long time).
(In reply to henrix from comment #4) > I guess the fix for that bug is already in CURRENT? I'm not really experienced > in FreeBSD, so running CURRENT sounds scary for a laptop I want to be stable Yes, it is already in CURRENT (development branch). I'm using CURRENT on my T490, but sometimes it is not stable, yes :)
(In reply to Oleksandr Kryvulia from comment #5) Yes, it is already in CURRENT (development branch). I'm using CURRENT on my T490, but sometimes it is not stable, yes :) Yeah, that's annoying. Thanks for the feedback. One last question, since you have the same laptop model: how long does it take for you to build CURRENT? Or did you install it from a snapshot? I guess that doing the initial full build will be a pain, as it'll require hint.hwpstate_intel.0.disabled=1 so that the system doesn't freeze, right?
(In reply to henrix from comment #6) > One last question, since you have the same laptop model: how long does it take for you to build CURRENT? Or did you install it from a snapshot? It is mostly depended on hardware. Since I'm build with WITH_CCACHE_BUILD=yes world+kernel is about one hour. When toolchain needs to be rebuilded it can take up to four hours. > I guess that doing the initial full build will be a pain, as it'll require hint.hwpstate_intel.0.disabled=1 so that the system doesn't freeze, right? Exactly.
(In reply to Oleksandr Kryvulia from comment #7) Great, thanks for your help. I'll just go ahead and go down the CURRENT route. I'll probably try to install from a recent snapshot (looks like the latest ones should have already the commit that fixes the issue) and then build everything myself.