Hi Tejun, I haven't placed the drive under stress test, but it does work OK under normal conditions without errors or timeouts in the log. >From the cache error, I know that the physical region iomapped in 0xf0000000 - 0xf0000200 is the PCI memory space, so the exception is being caused by trying to access this area. I haven't yet looked deeply into it, I was assuming the action of stopping the bmdma engine might have triggered the exception. If not, why not receive an immediate exception without the 10secs timeout? Regards, Alex On Mon, Sep 15, 2008 at 11:10 AM, Tejun Heo <tj@xxxxxxxxxx> wrote: > Hello, > > Alex Gonzalez wrote: >> :<6> sda:0:msdos_partition: sector_size=1 >> 0:read_dev_sector: mapping = 87946cac >> 0:__read_cache_page:cached_page=83a15580 >> 0:read_cache_page: page=83a15580 >> 0:lock_page >> 0:sync_page on mm/filemap.c >> 0:block_sync_page >> 0:blk_run_backing_dev() >> 0:blk_backing_dev_unplug() >> 0:scsi_request_fn >> 0:scsi_dispatch_cmd >> 0:ata_scsi_queuecmd >> 0:ata_scsi_translate >> 0:***ata_qc_prep*** >> 0:***ata_check_status*** >> 0:***ata_std_dev_select*** >> 0:***ata_check_status*** >> 0:***ata_check_status*** >> 0:***ata_tf_load >> 0:***ata_bmdma_setup*** >> 0:***ata_exec_command*** >> 0:***ata_bmdma_start*** >> 0:Dispatched cmd 8b64eba0 with rth 0 >> >> <PAUSE> >> >> 0:***sil_freeze()*** >> 0:***ata_bmdma_error_handler*** > > This is libata error handling kicking in after detecting command > timeout. > >> 0:***ata_bmdma_status*** >> 0:***ata_bmdma_stop*** > > And EH tries to stop the bmdma engine. > >> ********************************************* >> cpu_0 received a bus/cache error >> ********************************************* >> Bridge: Phys Addr = 0x0000000000, Device_AERR = 0x00000000 >> Bridge: The devices reporting AERR are: >> CPU: (XLR specific) Cache Error log = 0x0000007800004601, Phy Addr = >> 0x00f0000088 >> CPU: epc = 0x83435dfc, errorepc = 0x835bf054, cacheerr = 0x00000000 >> Can not handle bus/cache error - Halting cpu > > Which triggers bus/cache error. Hmmm... When the booting works, does > the drive keep working after boot? ie. if you put the drive under > stress test, does the system survive. Also, error handling shouldn't > trigger bus/cache error. Do you know what the bus/cache error means? > > Thanks. > > -- > tejun > -- 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