QNAP TS-419: sata_mv fails to initialise with "can't set global reset"

[Ian Campbell]

On Mon, 2015-06-01 at 04:50 +0200, Andrew Lunn wrote:
> On Sun, May 31, 2015 at 11:35:34AM +0100, Ian Campbell wrote:
> > Hi,
> > 
> > After updating to 4.0.4 (from the Debian kernel package) my TS-419 can
> > no longer see its SATA disks. In dmesg I see:
> > 
> > [    2.599060] SCSI subsystem initialized
> > [    2.662882] libata version 3.00 loaded.
> > [    2.664776] sata_mv 0000:01:00.0: version 1.28
> > [    2.664888] pci 0000:00:01.0: enabling device (0140 -> 0143)
> > [    2.687579] sata_mv: can't set global reset
> > [    2.691825] sata_mv 0000:01:00.0: Driver probe function unexpectedly returned 1
> > [    2.734954] sata_mv f1080000.sata: version 1.28
> > [    2.735222] sata_mv f1080000.sata: slots 32 ports 2
> > [    2.773358] scsi host0: sata_mv
> > [    2.785712] scsi host1: sata_mv
> > [    2.796204] ata1: SATA max UDMA/133 irq 35
> > [    2.800359] ata2: SATA max UDMA/133 irq 35
> > [    3.120986] ata1: SATA link down (SStatus 0 SControl F300)
> > [    3.444974] ata2: SATA link down (SStatus 0 SControl F300)
> > 
> > The last version I tried which worked was the 3.18.5 package.
> > 
> > This is using kirkwood-ts419-6281.dtb, which given this I think is the
> > correct one:
> > 
> > [    0.000000] Linux version 4.0.0-2-kirkwood (debian-kernel at lists.debian.org) (gcc version 4.9.2 (Debian 4.9.2-18) ) #1 Debian 4.0.4-1 (2015-05-26)
> > [    0.000000] CPU: Feroceon 88FR131 [56251311] revision 1 (ARMv5TE), cr=0005397f
> > [    0.000000] CPU: VIVT data cache, VIVT instruction cache
> > [    0.000000] Machine model: QNAP TS419 family
> > [...]
> > [    0.044889] mvebu-soc-id: MVEBU SoC ID=0x6281, Rev=0x2
> > 
> > I can also see both PCI devices which IIRC is the main difference
> > between the 6281 and 6282.
> > 
> > I don't see anything in the git logs after 4.0 which looks like an
> > obviously related fix, but I'm currently building 4.1-rc5 to have a
> > check.
>  
> Hi Ian
> 
> You might also want to try 
> 
> http://comments.gmane.org/gmane.linux.ports.arm.kernel/416589
> 
> In particular the first patch.

Thanks, I'll try those before I bother with a bisect.

Ian.