Hi, I do a lot of Linux hardware testing as a hobby, and I own 30+ mainboards for this purpose. I have noticed of ACPI S3 State resume bug of certain NVIDIA and SiS chipset mainboards on Ubuntu 12.04 (uses Linux 3.2 kernel). Here is the links for my analysis.
(SiS USB) https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1068393 https://bugzilla.kernel.org/show_bug.cgi?id=47991 (NVIDIA USB) https://bugzilla.kernel.org/show_bug.cgi?id=48101 Lan Tianyu has helped me out on this issue a lot. Anyway if you read Comment #50 of Linux kernel bug 47991, I believe this bug might be specific to ASUS mainboards with OHCI USB-based chipset (i.e., USB implementation other than Intel and VIA Technologies) and stable ACPI S3 State implementation based on experiments I have done. https://bugzilla.kernel.org/show_bug.cgi?id=47991#c50 Reading the posting here, people are thinking of doing PCI Vendor ID/Device ID matching to workaroud this bug, but perhaps if people can consider my hypothesis (ASUS specific bug) before really doing that. Regarding my collection of 30+ mainboards, I believe I own these ASUS mainboards with OHCI USB and ACPI S3 State support. - ASUS A7N266-VM/AA - ASUS A7N8X-E Deluxe (If I recall it correctly. Some of the caps are bulging though.) - ASUS A8AE-LE (Special HP OEM mainboard) (http://h10025.www1.hp.com/ewfrf/wc/document?cc=us&dlc=en&docname=c00496280&lc=en) I will have access to these mainboard in about 2 weeks (currently traveling). Regards, fpgahardwareengineer
