Bug 212970
| Summary: | SWAP space lost due to ipr 0002:c8:01.0: pci_map_sg failed AND iommu_alloc failed | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Alexey Bozrikov <a> |
| Component: | kernel | Assignee: | Kernel Maintainer List <kernel-maint> |
| Status: | CLOSED CURRENTRELEASE | QA Contact: | Brian Brock <bbrock> |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | 5 | CC: | dwmw2, nobody+pnasrat, wtogami |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | ppc64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | 2.6.19-1.2911 | Doc Type: | Bug Fix |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2007-03-06 08:59:48 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
Please check /etc/yaboot.conf and see whether it's listing your swap partition as one of the boot drives. If so, that would explain why it got reformatted. The problem does not seem to re-appear with latest FC6-PPC64 kernels. Considering it fixed? # uname -a Linux lnxsrv.company.com 2.6.19-1.2911.fc6 #1 SMP Sat Feb 10 15:04:12 EST 2007 ppc64 ppc64 ppc64 GNU/Linux |
Description of problem: After system restart with new kernel, SWAP space was missing. After examining output of 'dmesg' several kernel messages found related to the problem. Version-Release number of selected component (if applicable): Linux lnxsrv.local 2.6.18-1.2200.fc5 #1 SMP Sat Oct 14 17:08:02 EDT 2006 ppc64 ppc64 ppc64 GNU/Linux How reproducible: Just started the new kernel and noted swap space absense Steps to Reproduce: 1. in Fedora Core 5-install kernel 2.6.18-1.2200.fc5 and reboot. Swap space must be defined 2. 3. Actual results: swap space missing and 'swapon -a' command reports '/dev/md0: invalid argument' Expected results: normal swap space activation Additional info: (a) Device /dev/md0 reported as operational and OK (from /proc/mdstat): md0 : active raid1 sdb3[1] sda3[0] 3148672 blocks [2/2] [UU] (b) After execting `mkswap -c -L swap /dev/md0' everything reverted back to normal operation. (c) Excerpt from 'dmesg': [quote] iommu_alloc failed, tbl c000000001014c00 vaddr c0000000249b9000 npages 1 ipr 0002:c8:01.0: pci_map_sg failed! iommu_alloc failed, tbl c000000001014c00 vaddr c00000002e48b000 npages 1 ipr 0002:c8:01.0: pci_map_sg failed! iommu_alloc failed, tbl c000000001014c00 vaddr c000000049c03000 npages 1 ipr 0002:c8:01.0: pci_map_sg failed! iommu_alloc failed, tbl c000000001014c00 vaddr c00000001f39f000 npages 1 ipr 0002:c8:01.0: pci_map_sg failed! iommu_alloc failed, tbl c000000001014c00 vaddr c0000000234be000 npages 1 ipr 0002:c8:01.0: pci_map_sg failed! iommu_alloc failed, tbl c000000001014c00 vaddr c00000001f34c000 npages 1 ipr 0002:c8:01.0: pci_map_sg failed! iommu_alloc failed, tbl c000000001014c00 vaddr c000000021403000 npages 1 ipr 0002:c8:01.0: pci_map_sg failed! Unable to find swap-space signature [unquote] (d) Device /dev/md0 was created on two SCSI disks attached to: 0002:c8:01.0 SCSI storage controller: Mylex Corporation AcceleRAID 600/500/400/Sapphire support Device (rev 04) Subsystem: IBM Dual Channel PCI-X U320 SCSI Adapter Flags: bus master, 66MHz, slow devsel, latency 144, IRQ 117 BIST result: 00 Memory at 500f8900000 (64-bit, non-prefetchable) [size=256K] Memory at 500f8000000 (64-bit, prefetchable) [size=8M] Expansion ROM at 500f8800000 [disabled] [size=1M] Capabilities: [40] PCI-X non-bridge device Capabilities: [50] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable- Capabilities: [78] Power Management version 2 (e) Linux is running on an IBM 9117-P570 LPAR, 3GB RAM, 0.5 physical CPU (POWER5, 1.65GHz) (f) No other errors or unusual messages spotted in 'dmeag' output.