Michael Olson
2008-Jan-29 04:07 UTC
[CentOS-virt] Kernel panic in DomU when using swiotlb=force option
I'm trying to get a PVR-150 PCI card imported into a Xen DomU, however the driver blows up the kernel when loaded. The work around for this mentioned in various places is to use swiotlb=force on the DomU kernel. That leads to a fast and furious kernel panic. I've tried this with all Dom0 & DomU running kernel-xen-2.6.18-53.el5 kernel-xen-2.6.18-53.1.6.el5, bot x86 and 86_64 versions, also tried iommu=soft and various sizing options with swiotbl. Any ideas on how to fix or work-around this? Thanks -- Michael Kernel splat: ------------------------------------------------------- [root at hydra mythic]# xm create mythic -c Using config file "./mythic". Started domain mythic Bootdata ok (command line is ro root=LABEL=/1 swiotlb=force) Linux version 2.6.18-53.1.6.el5xen (mockbuild at builder6.centos.org) (gcc version 4.1.2 20070626 (Red Hat 4.1.2-14)) #1 SMP Wed Jan 23 11:59:21 EST 2008 BIOS-provided physical RAM map: Xen: 0000000000000000 - 0000000030800000 (usable) No mptable found. Built 1 zonelists. Total pages: 198656 Kernel command line: ro root=LABEL=/1 swiotlb=force Initializing CPU#0 PID hash table entries: 4096 (order: 12, 32768 bytes) Xen reported: 2400.082 MHz processor. Console: colour dummy device 80x25 Dentry cache hash table entries: 131072 (order: 8, 1048576 bytes) Inode-cache hash table entries: 65536 (order: 7, 524288 bytes) ----------- [cut here ] --------- [please bite here ] --------- Kernel BUG at lib/../arch/i386/kernel/swiotlb.c:160 invalid opcode: 0000 [1] SMP last sysfs file: CPU 0 Modules linked in: Pid: 0, comm: swapper Not tainted 2.6.18-53.1.6.el5xen #1 RIP: e030:[<ffffffff80334da7>] [<ffffffff80334da7>] swiotlb_init_with_default_size+0xa0/0x19e RSP: e02b:ffffffff805d9f40 EFLAGS: 00010282 RAX: 00000000fffffff4 RBX: 0000000000000000 RCX: 0000000000063906 RDX: ffffffffff578000 RSI: 0000000000000000 RDI: 0000000000000000 RBP: 0000000004000000 R08: 0000000000001000 R09: ffffffff806399e0 R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffffffff80599000(0000) knlGS:0000000000000000 CS: e033 DS: 0000 ES: 0000 Process swapper (pid: 0, threadinfo ffffffff805d8000, task ffffffff804c4a00) Stack: 0000000000000295 0000000000020800 0000000000000000 ffffffff80273167 0000000000000000 ffffffff805ec50a 0000000000000000 0000000000020800 0000000000000000 0000000000000000 Call Trace: [<ffffffff80273167>] pci_swiotlb_init+0x9/0x2d [<ffffffff805ec50a>] mem_init+0x5e/0x233 [<ffffffff805e2a7a>] start_kernel+0x189/0x224 [<ffffffff805e21ed>] _sinittext+0x1ed/0x1f3 Code: 0f 0b 68 1b e1 47 80 c2 a0 00 48 83 eb 80 48 8b 05 4c ca 33 RIP [<ffffffff80334da7>] swiotlb_init_with_default_size+0xa0/0x19e RSP <ffffffff805d9f40> <0>Kernel panic - not syncing: Fatal exception -------------------------------------------------------
Ross S. W. Walker
2008-Jan-29 04:16 UTC
[CentOS-virt] Kernel panic in DomU when using swiotlb=force option
Michael Olson wrote:> > > I'm trying to get a PVR-150 PCI card imported into a Xen > DomU, however > the driver blows up the kernel when loaded. The work around for this > mentioned in various places is to use swiotlb=force on the > DomU kernel. > That leads to a fast and furious kernel panic. I've tried > this with all > Dom0 & DomU running kernel-xen-2.6.18-53.el5 > kernel-xen-2.6.18-53.1.6.el5, bot x86 and 86_64 versions, also tried > iommu=soft and various sizing options with swiotbl. Any ideas > on how to > fix or work-around this? > > ThanksMaybe you should try the Xen 3.2 for CentOS 5.1 that is posted on xen.org. I have had good luck with this. -Ross ______________________________________________________________________ This e-mail, and any attachments thereto, is intended only for use by the addressee(s) named herein and may contain legally privileged and/or confidential information. If you are not the intended recipient of this e-mail, you are hereby notified that any dissemination, distribution or copying of this e-mail, and any attachments thereto, is strictly prohibited. If you have received this e-mail in error, please immediately notify the sender and permanently delete the original and any copy or printout thereof.
Possibly Parallel Threads
- [PATCH 35/35] x86: Don't panic if can not alloc buffer for swiotlb
- [PATCH 35/35] x86: Don't panic if can not alloc buffer for swiotlb
- Still hitting odd problems trying to install R 3.5.2 on Debian amd64 machine
- CESA-2008:0089 Important CentOS 5 i386 kernel Update
- CentOS 4.4, libstdc++ 3.4.6 and NPTL