On (08/18/09 15:18), Lin Ming wrote: > On Tue, 2009-08-18 at 15:13 +0800, Sergey Senozhatsky wrote: > > On (08/17/09 21:22), Lin Ming wrote: > > > > rc6-git2 > > > > 25 successful reboots. > > > > > > > > I'll need some more time to test against rc5. > > > > > > > > /*should I also test DSDT in case region-debug.patch fixed the problem?*/ > > > > > > Yes, please also help to test the custome DSDT. > > > region-debug.patch is just a debug patch. > > > We need to root cause the problem clearly and then work out a clean solution. > > > > > > > Hi, Lin > > Looks like DSDT didn't solve the problem (oops on 4th reboot). > > > > So: > > region-debug.patch + > > DSDT - > > > > /* Is that possible? */ > > Yes, that is possible. > The same aml method is re-entered when interpreter lock is released. > So you still see AE_ALREADY_EXISTS error with the custom DSDT, right? > No, here it is: [ 0.000000] ACPI: Override [DSDT-F3J00001], this is unsafe: tainting kernel Problem with AE_ALREADY_EXISTS is that that is quite rare. I'll continue watching. At the moment: [ 0.165424] ACPI: EC: Look up EC in DSDT [ 0.181759] ACPI: Interpreter enabled [ 0.181780] ACPI: (supports S0 S1 S3 S4 S5) [ 0.181831] ACPI: Using IOAPIC for interrupt routing [ 0.182022] PCI: MCFG configuration 0: base e0000000 segment 0 buses 0 - 255 [ 0.193803] PCI: MCFG area at e0000000 reserved in ACPI motherboard resources [ 0.193822] PCI: updated MCFG configuration 0: base e0000000 segment 0 buses 0 - 63 [ 0.193840] PCI: Using MMCONFIG for extended config space [ 0.203749] ACPI: EC: GPE = 0x1c, I/O: command/status = 0x66, data = 0x62 [ 0.203766] ACPI: EC: driver started in poll mode [ 0.206839] ACPI: No dock devices found. [ 0.207026] ACPI: PCI Root Bridge [PCI0] (0000:00) Sergey
Attachment:
signature.asc
Description: Digital signature
