Latest working kernel version: None Distribution: Gentoo Hardware Environment: Sony Z590 Problem Description: When powering off or rebooting, the regular init messages display without any errors popping up. Then, the screen turns off, but the power to the computer stays on, requiring me to hold down the power button for several seconds. I have tried fixing my DSDT, but the problem still occurs. I have tried waiting a long time, but it never shuts down. Steps to reproduce: Get a Sony Z590 and try rebooting or turning it off.
Created attachment 17785 [details] acpidump output
Created attachment 17786 [details] dmesg output before rebooting
Created attachment 17787 [details] DSDT source before "fix"
Created attachment 17788 [details] DSDT source error/warning fix (does not affect this bug) Note: There appears to be some windows only functionality (search _OS and OSYS)
Created attachment 17789 [details] /proc/cpuinfo
is this a regression? i.e. do you know if there is any earlier kernel releases than reboot/shutdown works well?
I don't know of any version where it works. When I installed linux, I moved through 2.6.25, 2.6.26, 2.6.27-rc2-iwlwifi, 2.6.27-rc5, and 2.6.27-rc6. Every version had this problem.
Hi, Colin Will you please try whether the suspend/resume can work on your laptops? From the acpidump it seems that there exist the following errors: PowerResource (MGPR, 0x00, 0x0002) { Method (_ON, 0, NotSerialized) { Return (Zero) // should have no return value. } Method (_OFF, 0, NotSerialized) { Return (Zero) // should have no return value } Method (_STA, 0, NotSerialized) { Return (One) // One indicates that the power resource is on } } Will you please confirm whether the shutdown can work on windows? Will you please try the patches http://bugzilla.kernel.org/show_bug.cgi?id=11000#C30,31,#32 on the latest kernel and see whether it can work well ? (Please add the boot option of "acpi.power_nocheck=1" Thanks.
Hi, Colin From the acpidump it seems that the acpi reboot mechanism is supported on your laptop.There exists the definition of RESET_REG in FADT table. And the bit of ACPI_FADT_RESET_REG is also supported. Will you please add the boot option of "reboot=a" on the latest kernel(2.6.27-rc6) and see whether the reboot can work on you laptop? thanks.
#s2ram Machine is unknown This machine can be identified by: sys_vendor = "Sony Corporation" sys_product = "VGN-Z590E" sys_version = "A222C3RK" bios_version = "R1280M3" See http://suspend.sf.net/s2ram-support.html for details If you report a problem, please include the complete output above. #s2ram -f Switching from vt2 to vt1 fbcon fb0 state 1 s2ram_do: No such device or address fbcon fb0 state 0 switching back to vt2 There is nothing added to dmesg. I will try the patches right after trying reboot=a. When I first got the laptop, it shut down fine in windows (The laptop only runs linux now). Thanks for your help.
I just tried reboot=a, and it seriously messed things up. All I get is a blinking cursor line at the very top of my screen. The cursor moves right to where the login prompt usually is, and I can log in (without seeing anything). I was able to pipe dmesg into a file and ssh it out. I tried rebooting, but the problem remains. Should I try without a dsdt override? I just removed the reboot=a option, and it's still doing the strange cursor thing. It booted fine before. startx give a black screen and ctrl+alt+backspace doesn't do anything.
Thanks for the confirm that shutdown can work well on windows. From the ACPI FADT we know that the RESET_REG is supported on this machine and the RESET REG address is 0xB2, which is the SMI I/O port. After 0xFB is written to 0xB2 I/O port, the system will cause SMI. Whether the system can be rebooted is related with the BIOS. >[074h 116 12] Reset Register : <Generic Address Structure> >[074h 116 1] Space ID : 01 (SystemIO) >[075h 117 1] Bit Width : 08 >[076h 118 1] Bit Offset : 00 >[077h 119 1] Access Width : 01 >[078h 120 8] Address : 00000000000000B2 >[080h 128 1] Value to cause reset : FB Please use the original DSDT and add the boot option of "acpi.power_nocheck=1" to see whether the shutdown can work. Thanks.
Created attachment 17797 [details] dmesg after reboot=a added
I'm rebuilding my kernel after a make clean to see if I can't get something to display on my screen again. I can't see anything after 1.5 seconds or so after booting the kernel (even after removing reboot=a). I'm not sure what caused the graphics problem. I rebuilt my kernel and am still having the graphics problems. I added acpi.power_nocheck=1, but am still having the problem with rebooting/power off.
The graphics problem wasn't related. My v86d was having issues for some reason.
Woops. Forgot to patch. Now lets see if acpi.power_nocheck=1 works. :D
Now that the kernel is patched and acpi.power_nocheck=1 is set, rebooting/shutting down still hangs.
This bug was fixed in 2.6.27-rc7 with no kernel parameters needed.