Date: Wed, 18 May 2005 07:34:11 -0500 From: Sean Welch <Sean_Welch@alum.wofford.org> To: freebsd-mobile@freebsd.org, freebsd-acpi@freebsd.org Subject: S3 suspend on a Dell I8k laptop (ACPI) Message-ID: <20050518123411.GA1141@NitroPhys.welchsmnet.net>
next in thread | raw e-mail | index | archive | help
I thought that before I put in a pr I would check here to see if anyone else has experienced this issue and if there is a fix/work-around for it. Apologies in advance for crossposting. It has been some time since I tried to use S3 (I think the last time was about 4.6-RELEASE or so) and I've never tried it with ACPI before now. I had swapped out the graphics card and that confused the BIOS on resume causing a shutdown, but for various reasons I have now replaced the original card and want to get suspend working again. I've seen the typical "it comes back from suspend and then immediately reboots" but found that this behavior goes away if I set: hw.acpi.reset_video: 0 Likewise, on the console I get the screen showing all white during suspend and on resume it turns the backlight off. I can toggle it back on using acpi_video sysctls for the lcd and the problem is nonexistent if I suspend under X and issue a dpms command beforehand to force the lcd off. On the whole I'm quite pleased with the resume capability (I'm mostly in X anyway) -- X, usb, mouse, pcmcia wireless, sound, keyboard, and everything else come up flawlessly. I discovered a potentially fatal issue, though. It appears the processor is running full throttle and in suspension there is no cooling. I'm lucky I noticed this last night before I went to bed -- I might have ended up with a smoking pile of molten electronics and quite possibly a fire. By the time I noticed it was very hot indeed and the electronics/plastics were beginning to outgas... I quad boot the machine (WinME, FBSD 5.3R, FBSD 5.4R, FBSD 5.4R) and the behavior is the same under both 5.3R and 5.4R (other than fxp0 dying under 5.3R). WinME isn't giving me an option for anything other than StandBy at the moment so I can't test it there though it seemed to work fine when I triggered it from DOS (booted from floppy). I have a custom kernel under 5.3 and just GENERIC under 5.4 (haven't configured it yet). My first thought was that the hlt instruction wasn't being issued but a quick check brings up: machdep.cpu_idle_hlt: 1 Can anyone offer advice? I'm running the latest revision of the BIOS (A23) and will be happy to post the asl on request. Sean
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050518123411.GA1141>