Re: boot failure with 4.13.0-rc6 due to ATA errors

[Tejun Heo <tj@kernel.org>]

(cc'ing Christoph)

On Mon, Aug 28, 2017 at 12:40:39PM -0600, David Ahern wrote:
> Not sure why mailing list to direct this bug report to, so starting with
> libata based on the error messages.
> 
> Some where between v4.12 and 4.13.0-rc6 a Celestica redstone switch
> fails to boot due to ATA errors:
> 
> [    9.185203] ata1.00: failed to set xfermode (err_mask=0x40)
> [    9.500825] ata1.00: revalidation failed (errno=-5)
> [   20.449205] ata1.00: failed to set xfermode (err_mask=0x40)
> 
> I just tried Linus' top of tree (cc4a41fe5541) and it still fails. With
> v4.12 the same switch boots and 'dmesg | grep ata' shows:
> 
> [    0.129080] libata version 3.00 loaded.
> [    1.016520] ata1: SATA max UDMA/133 abar m2048@0xdffce000 port
> 0xdffce100 irq 27
> [    1.016524] ata2: SATA max UDMA/133 abar m2048@0xdffce000 port
> 0xdffce180 irq 27
> [    1.016528] ata3: SATA max UDMA/133 abar m2048@0xdffce000 port
> 0xdffce200 irq 27
> [    1.016531] ata4: SATA max UDMA/133 abar m2048@0xdffce000 port
> 0xdffce280 irq 27
> [    1.028623] ata5: SATA max UDMA/133 abar m2048@0xdffcd000 port
> 0xdffcd100 irq 28
> [    1.028627] ata6: SATA max UDMA/133 abar m2048@0xdffcd000 port
> 0xdffcd180 irq 28
> [    1.326767] ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
> [    1.328646] ata2: SATA link down (SStatus 0 SControl 300)
> [    1.330519] ata4: SATA link down (SStatus 0 SControl 300)
> [    1.330554] ata3: SATA link down (SStatus 0 SControl 300)
> [    1.330575] ata1.00: ATA-9: InnoDisk Corp. - mSATA 3ME, S130604, max
> UDMA/133
> [    1.330581] ata1.00: 31277232 sectors, multi 16: LBA48 NCQ (depth
> 31/32), AA
> [    1.332433] ata1.00: failed to get Identify Device Data, Emask 0x1
> [    1.332709] ata1.00: failed to get Identify Device Data, Emask 0x1
> [    1.332717] ata1.00: configured for UDMA/133
> [    1.335813] ata6: SATA link down (SStatus 0 SControl 300)
> [    1.339829] ata5: SATA link down (SStatus 0 SControl 300)
> 
> Given the overhead of building, installing, booting and recovering from
> a failed boot, 'git bisect' is not a realistic option for this switch
> option unless some one can cut the span to a few iterations.
> 
> If it helps, lspci and lsscsi output from an older kernel:
> 
> # lspci
> 00:00.0 Host bridge: Intel Corporation Atom processor C2000 SoC
> Transaction Router (rev 02)
> 00:01.0 PCI bridge: Intel Corporation Atom processor C2000 PCIe Root
> Port 1 (rev 02)
> 00:02.0 PCI bridge: Intel Corporation Atom processor C2000 PCIe Root
> Port 2 (rev 02)
> 00:03.0 PCI bridge: Intel Corporation Atom processor C2000 PCIe Root
> Port 3 (rev 02)
> 00:0e.0 Host bridge: Intel Corporation Atom processor C2000 RAS (rev 02)
> 00:0f.0 IOMMU: Intel Corporation Atom processor C2000 RCEC (rev 02)
> 00:13.0 System peripheral: Intel Corporation Atom processor C2000 SMBus
> 2.0 (rev 02)
> 00:14.0 Ethernet controller: Intel Corporation Ethernet Connection I354
> (rev 03)
> 00:14.1 Ethernet controller: Intel Corporation Ethernet Connection I354
> (rev 03)
> 00:14.2 Ethernet controller: Intel Corporation Ethernet Connection I354
> (rev 03)
> 00:16.0 USB controller: Intel Corporation Atom processor C2000 USB
> Enhanced Host Controller (rev 02)
> 00:17.0 SATA controller: Intel Corporation Atom processor C2000 AHCI
> SATA2 Controller (rev 02)
> 00:18.0 SATA controller: Intel Corporation Atom processor C2000 AHCI
> SATA3 Controller (rev 02)
> 00:1f.0 ISA bridge: Intel Corporation Atom processor C2000 PCU (rev 02)
> 00:1f.3 SMBus: Intel Corporation Atom processor C2000 PCU SMBus (rev 02)
> 01:00.0 Ethernet controller: Broadcom Corporation Device b854 (rev 03)
> 
> 
> # lsscsi
> [0:0:0:0]    disk    ATA      InnoDisk Corp. - 604   /dev/sda

Can you please verify whether 818831c8b22f ("libata: implement
SECURITY PROTOCOL IN/OUT") is the culprit?  ie. try to boot the commit
to verify that the problem is there, and try the one prior?

Thanks.

-- 
tejun