Message ID | 1489676848-103256-1-git-send-email-john.garry@huawei.com (mailing list archive) |
---|---|
State | Accepted, archived |
Headers | show |
John Garry <john.garry@huawei.com> writes: John, > The total ata xfer length may not be calculated properly, > in that we do not use the proper method to get an sg element > dma length. > > According to the code comment, sg_dma_len() should be used > after dma_map_sg() is called. > > This issue was found by turning on the SMMUv3 in front of > the hisi_sas controller in hip07. Multiple sg elements > were being combined into a single element, but the original > first element length was being use as the total xfer length. Applied to 4.11/scsi-fixes.
On 19/03/2017 17:21, Martin K. Petersen wrote: > John Garry <john.garry@huawei.com> writes: > > John, > >> The total ata xfer length may not be calculated properly, >> in that we do not use the proper method to get an sg element >> dma length. >> >> According to the code comment, sg_dma_len() should be used >> after dma_map_sg() is called. >> >> This issue was found by turning on the SMMUv3 in front of >> the hisi_sas controller in hip07. Multiple sg elements >> were being combined into a single element, but the original >> first element length was being use as the total xfer length. > I should have added this originally to the changelog: Fixes: ff2aeb1eb64c8a4770a6 ("libata: convert to chained sg") BTW, I am surprised this issue has not been seen in almost 10 years, but we cannot attach a SATA disk when SMMU enabled without it. Cheers, John > Applied to 4.11/scsi-fixes. >
John Garry <john.garry@huawei.com> writes: > I should have added this originally to the changelog: > Fixes: ff2aeb1eb64c8a4770a6 ("libata: convert to chained sg") Added.
diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c index 763f012..87f5e694 100644 --- a/drivers/scsi/libsas/sas_ata.c +++ b/drivers/scsi/libsas/sas_ata.c @@ -221,7 +221,7 @@ static unsigned int sas_ata_qc_issue(struct ata_queued_cmd *qc) task->num_scatter = qc->n_elem; } else { for_each_sg(qc->sg, sg, qc->n_elem, si) - xfer += sg->length; + xfer += sg_dma_len(sg); task->total_xfer_len = xfer; task->num_scatter = si;
The total ata xfer length may not be calculated properly, in that we do not use the proper method to get an sg element dma length. According to the code comment, sg_dma_len() should be used after dma_map_sg() is called. This issue was found by turning on the SMMUv3 in front of the hisi_sas controller in hip07. Multiple sg elements were being combined into a single element, but the original first element length was being use as the total xfer length. Signed-off-by: John Garry <john.garry@huawei.com>