Most recent kernel where this bug did not occur: <unknown> Distribution: Debian (kernel is vanilla from ftp.kernel.org, though) Hardware Environment: Thinkpad T43 2687 Software Environment: Debian Problem Description: During *resume* from S3, on a kernel with CONFIG_DEBUG_MUTEXES, CONFIG_DEBUG_SPINLOCK, CONFIG_DEBUG_SPINLOCK_SLEEP and a few other debug uptions, I get: Jul 5 11:25:44 thorin kernel: PM: Preparing system for mem sleep Jul 5 11:26:13 thorin kernel: Stopping tasks: ===================================================================================== ==========| Jul 5 11:26:13 thorin kernel: pnp: Device 00:0a disabled. Jul 5 11:26:13 thorin kernel: ACPI: PCI interrupt for device 0000:0b:00.0 disabled Jul 5 11:26:13 thorin kernel: acpi_bus-0201 [71] bus_set_power : Device is not power manageable Jul 5 11:26:13 thorin kernel: ACPI: PCI interrupt for device 0000:00:1e.2 disabled Jul 5 11:26:13 thorin kernel: ACPI: PCI interrupt for device 0000:00:1d.7 disabled Jul 5 11:26:13 thorin kernel: ACPI: PCI interrupt for device 0000:00:1d.3 disabled Jul 5 11:26:13 thorin kernel: ACPI: PCI interrupt for device 0000:00:1d.2 disabled Jul 5 11:26:13 thorin kernel: ACPI: PCI interrupt for device 0000:00:1d.1 disabled Jul 5 11:26:13 thorin kernel: ACPI: PCI interrupt for device 0000:00:1d.0 disabled Jul 5 11:26:13 thorin kernel: PM: Entering mem sleep Jul 5 11:26:13 thorin kernel: hwsleep-0283 [72] enter_sleep_state : Entering sleep state [S3] Jul 5 11:26:13 thorin kernel: Intel machine check architecture supported. Jul 5 11:26:13 thorin kernel: Intel machine check reporting enabled on CPU#0. Jul 5 11:26:13 thorin kernel: Back to C! Jul 5 11:26:13 thorin kernel: BUG: sleeping function called from invalid context at include/linux/rwsem.h:43 Jul 5 11:26:13 thorin kernel: in_atomic():0, irqs_disabled():1 Jul 5 11:26:13 thorin kernel: [<c0129462>] blocking_notifier_call_chain+0x18/0x4b Jul 5 11:26:13 thorin kernel: [<c02d0f4f>] cpufreq_resume+0xdf/0xfc Jul 5 11:26:13 thorin kernel: [<c027767d>] __sysdev_resume+0x28/0x6b Jul 5 11:26:13 thorin kernel: [<c02776f4>] sysdev_resume+0x34/0x52 Jul 5 11:26:13 thorin kernel: [<c027bd5c>] device_power_up+0x5/0xa Jul 5 11:26:13 thorin kernel: [<c0136b94>] suspend_enter+0x44/0x46 Jul 5 11:26:13 thorin kernel: [<c011dee3>] printk+0x1b/0x1f Jul 5 11:26:13 thorin kernel: [<c0136cac>] enter_state+0xdf/0x1ba Jul 5 11:26:13 thorin kernel: [<c0136e10>] state_store+0x89/0x92 Jul 5 11:26:13 thorin kernel: [<c0136d87>] state_store+0x0/0x92 Jul 5 11:26:13 thorin kernel: [<c018fa5a>] subsys_attr_store+0x1e/0x22 Jul 5 11:26:13 thorin kernel: [<c019007b>] sysfs_write_file+0x95/0xd9 Jul 5 11:26:13 thorin kernel: [<c0154fc9>] vfs_write+0x8b/0x13f Jul 5 11:26:13 thorin kernel: [<c018ffe6>] sysfs_write_file+0x0/0xd9 Jul 5 11:26:13 thorin kernel: [<c01558ba>] sys_write+0x41/0x6a Jul 5 11:26:13 thorin kernel: [<c0102bbf>] syscall_call+0x7/0xb Jul 5 11:26:13 thorin kernel: PM: Finishing wakeup. Steps to reproduce: Suspend to RAM (S3), then resume the machine using the lid switch to generate the wakeup event.
This could be related to: http://article.gmane.org/gmane.linux.acpi.devel/18148
This bug was reported against the wrong subsystem. I will close it as INVALID and open a new one against the correct subsystem. Sorry about this.
Marking as closed (it IS still rejected and invalid, though :) ).