2.6.24-rc8-mm1 is corrupting. smartd does some sg_ioctl into its stack, and depending on how its stack randomization worked out, this is liable to end up writing into the adjacent physical page too. If you're lucky you have highmem, and ioread16_rep oopses on the virtual address beyond what ata_pio_sector's kmap_atomic set up; when you're unlucky, it'll go ahead and corrupt the next page more obscurely. Bisect led to scsi-misc-2.6.git's 465ff3185e0cb76d46137335a4d21d0d9d3ac8a2 [SCSI] relax scsi dma alignment, and the patch below is good for a hot-fix. Though I doubt it'll be the final fix, since it appears to undo the whole point of blk_queue_update_dma_alignment: perhaps libata is at fault to be going through sectory code for unsectory I/O - I wouldn't know. Signed-off-by: Hugh Dickins <hugh@xxxxxxxxxxx> --- drivers/ata/libata-scsi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- 2.6.24-rc8-mm1/drivers/ata/libata-scsi.c 2008-01-17 16:49:47.000000000 +0000 +++ linux/drivers/ata/libata-scsi.c 2008-01-22 15:45:40.000000000 +0000 @@ -826,7 +826,7 @@ static void ata_scsi_sdev_config(struct sdev->max_device_blocked = 1; /* set the min alignment */ - blk_queue_update_dma_alignment(sdev->request_queue, ATA_DMA_PAD_SZ - 1); + blk_queue_update_dma_alignment(sdev->request_queue, ATA_SECT_SIZE - 1); } static void ata_scsi_dev_config(struct scsi_device *sdev, - To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
