On 06/04/2009 09:33 AM, FUJITA Tomonori wrote: > On Thu, 4 Jun 2009 08:12:34 +0200 > Torsten Kaiser <just.for.lkml@xxxxxxxxxxxxxx> wrote: > >> On Thu, Jun 4, 2009 at 2:02 AM, FUJITA Tomonori >> <fujita.tomonori@xxxxxxxxxxxxx> wrote: >>> On Wed, 3 Jun 2009 21:30:32 +0200 >>> Torsten Kaiser <just.for.lkml@xxxxxxxxxxxxxx> wrote: >>>> Still happens with 2.6.30-rc8 (see trace at the end of the email) >>>> >>>> As orig_n_elem is only used two times in libata-core.c I suspected a >>>> corruption of the qc->sg, but adding checks for this did not trigger. >>>> So I looked into lib/dma-debug.c. >>>> It seems add_dma_entry() does not protect against adding the same >>>> entry twice. >>> Do you mean that add_dma_entry() doesn't protect against adding a new >>> entry identical to the existing entry, right? >> Yes, as I read the hash bucket code in lib/dma-debug.c a second entry >> from the same device and the same address will just be added to the >> list and on unmap it will always return the first entry. > > It means that two different DMA operations will be performed against > the same dma addresss on the same device at the same time. It doesn't > happen unless there is a bug in a driver, an IOMMU or somewhere, as I > wrote in the previous mail. > What about the draining buffers used by libata. Are they not the same buffer for all devices for all requests? > >>> Then it's not a >>> dma-debug bug (it might be better for dma-debug to check it though), >>> that is, such situation should not happen. >> At least the warning about the wrong unmap count is a bug in the >> dma-debug, as that is not what happens on my system. >> >>> Probably, it's an IOMMU bug >>> or a driver bug. >> Could it be just a forgotten unmap? >> That would leave the old entry in the dma-debug list, but from the >> driver side it would be valid to map the same place again without >> corrupting any data transfer to the harddisk. > > Yeah, I thought about this possibility. However, you use GART IOMMU, > right (you can see "PCI-DMA: using GART IOMMU." in a boot message if > so)? If you use GART IOMMU, unmapped addresses are not reused. > > >> What also would point in this direction, sometime I have seen this in my log: >> [ 1004.061989] DMA-API: debugging out of memory - disabling > > Sounds like there is a leak... Boaz -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
