Distribution: Fedora Core 3 Hardware Environment: Sony VAIO B100B (Centrino/Dothan 2GHz, 855GME) Software Environment: Linux 2.6.10-1.741_FC3 Problem Description: Following resume from an S3 suspend, the hard drive appears to fail to spin up. The following errors eventually appear via dmesg: hda: lost interrupt ACPI-0286: *** Error: No installed handler for fixed event [00000002] hda: no DRQ after issuing MULTWRITE Steps to reproduce: * Boot with "acpi_sleep=s3_bios acpi_sleep=s3_mode" to correct video problem on resume (otherwise, screen remains blank) * Bring up system in single-user mode. The set of modules loaded doesn't appear to matter (I hacked rc.sysinit to avoid loading any modules in some tests) * First access of hard drive hangs system In some tests, I was able to catch errors related to device 00:1f.1 having IRQ problems, but I haven't managed to keep them on screen long enough to copy them down (I'll try to catch them with a photo if it would be helpful). This device on my system is (ala "lspci -vvv"): - - - - - 00:1f.1 IDE interface: Intel Corp. 82801DBM (ICH4-M) IDE Controller (rev 03) (prog-if 8a [Master SecP PriP]) Subsystem: Sony Corporation: Unknown device 8188 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap- 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0 Interrupt: pin A routed to IRQ 255 Region 0: I/O ports at <unassigned> Region 1: I/O ports at <unassigned> Region 2: I/O ports at <unassigned> Region 3: I/O ports at <unassigned> Region 4: I/O ports at 1810 [size=16] Region 5: Memory at 20000000 (32-bit, non-prefetchable) [size=1K] - - - - - Intel's documentation on the 855GME chipset has a note that might be relevant to the problem. Their documentation is at: ftp://download.intel.com/design/mobile/desguide/25261604.pdf Section 11.1.4.1 item 2 (page 198 in my copy) indicates: B0:D31:F0 Offset D0-D3h (GEN_CNTL
Created attachment 4485 [details] Output from "lspci -vvv"
I left out an obvious step in the "steps to reproduce" section. Before the last item: * echo 3 > /proc/acpi/sleep, then resume with keyboard press
I figured out how to catch more of the post-suspend output (by trying to put it to sleep again, which doesn't require disk interaction). Here's the output. I didn't write down all of the addresses and arguments, so let me know if there are any of particular interest. FYI, IRQ 14 on my system is allocated to ide0. I'll append the full interrupt list as an attachment to this bug. Also, the second attempt to switch back to S3 utterly fails and hangs the system. - - - - - Back to C! Warning: CPU frequency is 6000000, cpufreq assumed 2000000 kHz Debug: sleeping function called from invalid context at mm/slab.c:2055 in_atomic():0, irqs_disabled():1 [<c011b5d1>] __mgmtd_sleep+0xd1/0xf0 [<xxxxxxxx>] acpi_ut_trace ... [<xxxxxxxx>] kmem_cache_alloc ... [<xxxxxxxx>] acpi_pci_link_set ... [<xxxxxxxx>] acpi_pci_link_resume ... [<xxxxxxxx>] sysdev_resume ... [<xxxxxxxx>] device_power_up ... [<xxxxxxxx>] suspend_enter ... [<xxxxxxxx>] enter_state ... [<xxxxxxxx>] acpi_suspend ... [<xxxxxxxx>] copy_from_user ... [<xxxxxxxx>] acpi_system_write_sleep ... [<xxxxxxxx>] vfs_write ... [<xxxxxxxx>] sys_write ... [<xxxxxxxx>] sysenter_post_exp ... PM: Finishing up. pci_irq-0357 [14] acpi_pci_irq_derive : Unable to derive IRQ for device 0000:00:1f.1 ACPI: PCI interrupt 0000:00:1f.1[A]: no GSI hda: lost interrupt evevent-0286: *** Error: No installed handler for fixed event [00000002] Restarting tasks... done
Created attachment 4488 [details] Contents of /proc/interrupts
The last test was performed with a plain 2.6.10 kernel with the "acpi-20050125-2.6.10.diff.gz" patch applied. The kernel was passed "init=/bin/sh" to prevent module loading (beyond ext3 and jbd via initrd).
still an issue with 2.6.13?
Could you please try the patch at bug 2039 and report it back? Let's see if the ide failure is caused by lack of invoking ACPI metthods.
*** This bug has been marked as a duplicate of 2039 ***