From mboxrd@z Thu Jan 1 00:00:00 1970 From: David F Barrera Subject: Re: Daily Xen Builds Date: Thu, 18 Aug 2005 17:12:08 -0500 Message-ID: <430507B8.1060302@us.ibm.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Ian Pratt Cc: Li Ge , xen-devel@lists.xensource.com, xen-devel-bounces@lists.xensource.com List-Id: xen-devel@lists.xenproject.org Ian Pratt wrote: > > > > >>>I can't see why creating a domU would be helped by a swiotlb. >>> >>> >>> >>Without it, dom0 crashes. I reported on Bugzilla #168 a >>couple of days ago the following crash when creating a domU >>on machines built with PAE: >> >>Fatal DMA error! Please use 'swiotlb=force' >>kernel BUG at arch/xen/i386/kernel/pci-dma.c:30 (iommu_bug)! >> [] iommu_bug+0x3b/0x50 >> [] dma_map_sg+0x14a/0x170 >> [] ide_map_sg+0xa3/0xb0 >> [] ide_build_sglist+0x7c/0xd0 [] >>ide_build_dmatable+0x3f/0x190 [] >>ide_dma_setup+0x3a/0xe0 [] >>__ide_do_rw_disk+0x3d0/0x5a0 [] >>ide_wait_stat+0+30/0x230 [] >>as_move_to_dispatch+0xec/0x220 [] >>start_request+0x270/0x290 [] >>ide_do_request+0x27c/0x470 [] >>elv_next_request+0x3c/0x1b0 [] >>do_ide_request+0x24/0x30 [] >>generic_unplug_device+0x2a/0x70 [] >>generic_unplug_device+0x0/0x70 [] >>blkio_schedule+0x186/0x190 [] >>default_wake_function+0x0/0x20 [] >>ret_from_fork+0x6/0x14 [] >>default_wake_function+0x0/0x20 [] >>blkio_schedule+0x0/0x190 [] >>kernel_thread_helper+0x5/0x10 Kernel panic - not syncing: BUG! >> >> >> >>>Precisely which adaptec card is it, what driver? >>> >>> >>> >>SCSI storage controller: Adaptec AIC-7901 U320 (rev 10) >> >> > >This backtrace looks like its doing ide transfers. Have you got active >ide devices in this system? > > Actually, the trace was obtained from a machine that has an IDE drive. It was one of two that exhibited this behavior (the other has the SCSI drive), and the only one where I could capture the console log. >Also, please can you rerun this test as the debugging output has been >improved. > > > I'll do that as soon as I can get the machine back; it is being used to look into another issue at the moment. >Thanks, >Ian > > >