Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.kernel > #55834 > unrolled thread

Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL)

Back to article view | Back to linux.debian.kernel

Contents

  Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL) Ryan Tandy <ryan@nardis.ca> - 2016-11-25 08:40 +0100
    Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL) Ian Campbell <ijc@debian.org> - 2016-11-25 08:50 +0100
      Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL) Ryan Tandy <ryan@nardis.ca> - 2016-11-25 17:20 +0100
        Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL) Ian Campbell <ijc@debian.org> - 2016-11-25 17:30 +0100
          Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL) Ryan Tandy <ryan@nardis.ca> - 2016-11-25 18:00 +0100
            Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL) Ryan Tandy <ryan@nardis.ca> - 2016-11-25 18:10 +0100
            Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL) Ian Campbell <ijc@debian.org> - 2016-11-25 18:10 +0100
        Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL) Roger Shimizu <rogershimizu@gmail.com> - 2016-11-26 17:30 +0100
          Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL) Ryan Tandy <ryan@nardis.ca> - 2016-11-26 22:10 +0100
            Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL) Ryan Tandy <ryan@nardis.ca> - 2016-11-27 00:00 +0100
              Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL) Roger Shimizu <rogershimizu@gmail.com> - 2016-11-27 01:20 +0100
    Processed: Re: Bug#845611: linux-image-4.8.0-1-marvell: hard  drive not detected on LinkStation Pro (LS-GL) owner@bugs.debian.org (Debian Bug Tracking System) - 2016-11-27 00:00 +0100

#55834 — Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL)

Package: linux-image-4.8.0-1-marvell
Version: 4.8.7-1
Severity: normal
Tags: upstream

I found that on recent d-i images, the kernel does not detect the hard 
drive on my LinkStation. I tried some older images and found that kernel 
4.3.3-5 worked and kernel 4.6.2-2 did not.

I suspect the problem is with the device tree file, therefore CCing 
Roger Shimizu who contributed that file.

The old kurobox_pro-setup.c contains this code:

static struct mv_sata_platform_data kurobox_pro_sata_data = {
        .n_ports        = 2,
};

whereas the new orion5x-linkstation.dtsi contains this code:

&sata {
        status = "okay";
        nr-ports = <1>;
};

The dmesg output seems to reinforce this conclusion.

dmesg from 4.3.0-1:

[  611.914912] sata_mv sata_mv.0: version 1.28
[  611.915107] sata_mv sata_mv.0: cannot get optional clkdev
[  611.916086] sata_mv sata_mv.0: slots 32 ports 2
[  611.971107] scsi host0: sata_mv
[  611.984111] scsi host1: sata_mv
[  611.984881] ata1: SATA max UDMA/133 irq 30
[  611.984916] ata2: SATA max UDMA/133 irq 30
[  612.302763] ata1: SATA link down (SStatus 0 SControl 300)
[  612.778697] ata2: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
[  612.787074] ata2.00: ATA-7: SAMSUNG SP2504C, VT100-41, max UDMA7
[  612.787118] ata2.00: 488397168 sectors, multi 0: LBA48 NCQ (depth 31/32)
[  612.819082] ata2.00: configured for UDMA/133
[  612.820339] scsi 1:0:0:0: Direct-Access     ATA      SAMSUNG SP2504C  0-41 PQ: 0 ANSI: 5
[  612.899766] sd 1:0:0:0: [sda] 488397168 512-byte logical blocks: (250 GB/232 GiB)
[  612.900971] sd 1:0:0:0: [sda] Write Protect is off
[  612.901031] sd 1:0:0:0: [sda] Mode Sense: 00 3a 00 00
[  612.901533] sd 1:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[  612.924886] sd 1:0:0:0: [sda] Attached SCSI disk

dmesg from 4.8.7-1:

[   11.140091] sata_mv sata_mv.0: version 1.28
[   11.140252] sata_mv sata_mv.0: cannot get optional clkdev
[   11.170717] sata_mv sata_mv.0: slots 32 ports 1
[   11.226581] scsi host0: sata_mv
[   11.236985] ata1: SATA max UDMA/133 irq 28
[   11.554565] ata1: SATA link down (SStatus 0 SControl 300)

In this extract, ata2 is apparently never scanned.

I expect changing nr-ports to 2 will fix this. I have not tested the 
change yet, but will try to do so this weekend.

thanks
Ryan

[toc] | [next] | [standalone]

#55836

On Thu, 2016-11-24 at 23:24 -0800, Ryan Tandy wrote:
> whereas the new orion5x-linkstation.dtsi contains this code:
> 
> &sata {
>         status = "okay";
>         nr-ports = <1>;
> };

A .dtsi is an include file, not the final thing for any given device,
for that you need a .dts (which is compiled into a .dtb).

Assuming orion5x-linkstation.dtsi is correctly relating to your
platform you would appear to want one of:
    $ git grep orion5x-linkstation.dtsi arch/arm/boot/dts/*.dts
    arch/arm/boot/dts/orion5x-kuroboxpro.dts:#include "orion5x-linkstation.dtsi"
    arch/arm/boot/dts/orion5x-linkstation-lsgl.dts:#include "orion5x-linkstation.dtsi"
    arch/arm/boot/dts/orion5x-linkstation-lswtgl.dts:#include "orion5x-linkstation.dtsi"

(Or something else not yet present in the kernel tree.)

Of the three above orion5x-kuroboxpro.dts and orion5x-linkstation-
lswtgl both set ports to 2 using:

    &sata {
            nr-ports = <2>;
    };

which overrides the defaults from the dtsi. You mentioned kurobox_pro-
setup.c so perhaps orion5x-kuroboxpro.dts is the one you want?

Ian.

[toc] | [prev] | [next] | [standalone]

#55840

On Fri, Nov 25, 2016 at 07:42:37AM +0000, Ian Campbell wrote:
>Assuming orion5x-linkstation.dtsi is correctly relating to your
>platform you would appear to want one of:
>    $ git grep orion5x-linkstation.dtsi arch/arm/boot/dts/*.dts
>    arch/arm/boot/dts/orion5x-kuroboxpro.dts:#include "orion5x-linkstation.dtsi"
>    arch/arm/boot/dts/orion5x-linkstation-lsgl.dts:#include "orion5x-linkstation.dtsi"
>    arch/arm/boot/dts/orion5x-linkstation-lswtgl.dts:#include "orion5x-linkstation.dtsi"
>
>(Or something else not yet present in the kernel tree.)

Mine is an LS-GL, so orion5x-linkstation-lsgl.dts should be the one. 
(The model string in there is the same one I see at runtime, too.)

>Of the three above orion5x-kuroboxpro.dts and orion5x-linkstation-
>lswtgl both set ports to 2 using:
>
>    &sata {
>            nr-ports = <2>;
>    };
>
>which overrides the defaults from the dtsi. You mentioned kurobox_pro-
>setup.c so perhaps orion5x-kuroboxpro.dts is the one you want?

Exactly - orion5x-linkstation-lsgl.dts is missing this chunk. So my 
suspicion is it needs the same override (or, if they all end up using 
the same value, maybe the dtsi can change - Roger would know better than 
I would). Sorry for the lack of clarity.

kurobox_pro-setup.c, AFAIK, used to be used on several related 
platforms. The kurobox pro and LS-GL are nearly identical.

thanks
Ryan

[toc] | [prev] | [next] | [standalone]

#55841

On Fri, 2016-11-25 at 08:10 -0800, Ryan Tandy wrote:
> On Fri, Nov 25, 2016 at 07:42:37AM +0000, Ian Campbell wrote:
> > 
> > Assuming orion5x-linkstation.dtsi is correctly relating to your
> > platform you would appear to want one of:
> >    $ git grep orion5x-linkstation.dtsi arch/arm/boot/dts/*.dts
> >    arch/arm/boot/dts/orion5x-kuroboxpro.dts:#include "orion5x-
> > linkstation.dtsi"
> >    arch/arm/boot/dts/orion5x-linkstation-lsgl.dts:#include
> > "orion5x-linkstation.dtsi"
> >    arch/arm/boot/dts/orion5x-linkstation-lswtgl.dts:#include
> > "orion5x-linkstation.dtsi"
> > 
> > (Or something else not yet present in the kernel tree.)
> 
> Mine is an LS-GL, so orion5x-linkstation-lsgl.dts should be the one. 

Is it possible that there are multiple variants of this one with
differing numbers of disks?

It's a little tricky to google for but all the LS-GL's I can see _look_
like they are single disk (see [*] below). Or maybe the naming is just
confusing?

Ian.

[0] http://www.gkspk.com/view/techie/upgrade-hdd-buffalo-linkstation-ls
-gl-nas/
[1] http://buffalo.nas-central.org/wiki/Category:LSPro reached from th
"LS-GL v1" or "v2" link at  http://buffalo.nas-central.org/wiki/Main_Pa
ge
[2] http://buffalo.nas-central.org/wiki/Information/LSPROOverview which
is the result of searching for "LS-GL" on that site
[3] http://buffalo.nas-central.org/wiki/Disassemble_the_LS_Pro_v1/LS_Li
ve_v1

[toc] | [prev] | [next] | [standalone]

#55842

On Fri, Nov 25, 2016 at 04:24:31PM +0000, Ian Campbell wrote:
>Is it possible that there are multiple variants of this one with
>differing numbers of disks?
>
>It's a little tricky to google for but all the LS-GL's I can see _look_
>like they are single disk (see [*] below). Or maybe the naming is just
>confusing?

I don't believe there is anything called LS-GL that supports more than 
one disk. Now that you mention it though, there are two hardware 
revisions out there:

http://buffalo.nas-central.org/wiki/LinkstationProLiveDifferences

I have the older (v1) hardware.

I will crack the box open this weekend and confirm, but as far as I 
remember there was only one SATA connector.

Aha, there was a guide for wiring up the second port, though:

http://buffalo.nas-central.org/wiki/Add_a_Second_Hard_Drive_to_Your_LS_Pro_v1/LS_Live_v1

Mine has zero mods though, totally stock hardware-wise.

[toc] | [prev] | [next] | [standalone]

#55843

On Fri, Nov 25, 2016 at 05:02:06PM +0000, Ian Campbell wrote:
>Oh, so the issue is not that you are missing a second disk, but rather
>that the one disk you do have seems to not be detected? You theory is
>that it is actually only the second port which is connected to anything
>in this hw?

Yes, exactly.

[toc] | [prev] | [next] | [standalone]

#55844

On Fri, 2016-11-25 at 08:56 -0800, Ryan Tandy wrote:
> On Fri, Nov 25, 2016 at 04:24:31PM +0000, Ian Campbell wrote:
> > 
> > Is it possible that there are multiple variants of this one with
> > differing numbers of disks?
> > 
> > It's a little tricky to google for but all the LS-GL's I can see
> > _look_
> > like they are single disk (see [*] below). Or maybe the naming is
> > just
> > confusing?
> 
> 
> [...snip...]

Thanks for all that.

> Mine has zero mods though, totally stock hardware-wise.

Oh, so the issue is not that you are missing a second disk, but rather
that the one disk you do have seems to not be detected? You theory is
that it is actually only the second port which is connected to anything
in this hw?

Sorry, took me a while to grok what you were saying in your original
mail!

Ian.

[toc] | [prev] | [next] | [standalone]

#55849

Dear Ryan,

On Fri, Nov 25, 2016 at 4:24 PM, Ryan Tandy <ryan@nardis.ca> wrote:
>
> [  611.971107] scsi host0: sata_mv
> [  611.984111] scsi host1: sata_mv
> [  611.984881] ata1: SATA max UDMA/133 irq 30
> [  611.984916] ata2: SATA max UDMA/133 irq 30
> [  612.302763] ata1: SATA link down (SStatus 0 SControl 300)
> [  612.778697] ata2: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
> [  612.787074] ata2.00: ATA-7: SAMSUNG SP2504C, VT100-41, max UDMA7
> [  612.787118] ata2.00: 488397168 sectors, multi 0: LBA48 NCQ (depth 31/32)
> [  612.819082] ata2.00: configured for UDMA/133

Looks like your LS-GL really uses the 2nd SATA port.
Though my LS-GL works fine with current DTS.
So it need the following patch to be applied.

diff --git a/arch/arm/boot/dts/orion5x-linkstation-lsgl.dts
b/arch/arm/boot/dts/orion5x-linkstation-lsgl.dts
index 1cf644b..51dc734 100644
--- a/arch/arm/boot/dts/orion5x-linkstation-lsgl.dts
+++ b/arch/arm/boot/dts/orion5x-linkstation-lsgl.dts
@@ -82,6 +82,10 @@
     gpios = <&gpio0 9 GPIO_ACTIVE_HIGH>;
 };

+&sata {
+    nr-ports = <2>;
+};
+
 &ehci1 {
     status = "okay";
 };


On Sat, Nov 26, 2016 at 1:10 AM, Ryan Tandy <ryan@nardis.ca> wrote:
>
> kurobox_pro-setup.c, AFAIK, used to be used on several related platforms.
> The kurobox pro and LS-GL are nearly identical.

Yes. So maybe the easiest way for you is:
- start your device by kernel 4.3 first.
- confirm your have flash-kernel and linux-image-4.8.0-1-marvell installed.
- cp /usr/lib/linux-image-4.8.0-1-marvell/orion5x-kuroboxpro.dtb
/etc/flash-kernel/dtbs/orion5x-linkstation-lsgl.dtb
- flash-kernel --force 4.8.0-1-marvell
- check the log of above command, it should use
/etc/flash-kernel/dtbs/orion5x-linkstation-lsgl.dtb for building
uImage.buffalo, the kernel image for booting
- reboot and see the result

If the above steps works for you, I can submit the above patch to
kernel upstream, and include it into Stretch release.
So I'm looking forward to your result. Thank you!

Cheers,
-- 
Roger Shimizu, GMT +9 Tokyo
PGP/GPG: 4096R/6C6ACD6417B3ACB1

[toc] | [prev] | [next] | [standalone]

#55853

On Sun, Nov 27, 2016 at 01:20:10AM +0900, Roger Shimizu wrote:
>- start your device by kernel 4.3 first.
>- confirm your have flash-kernel and linux-image-4.8.0-1-marvell installed.
>- cp /usr/lib/linux-image-4.8.0-1-marvell/orion5x-kuroboxpro.dtb
>/etc/flash-kernel/dtbs/orion5x-linkstation-lsgl.dtb
>- flash-kernel --force 4.8.0-1-marvell

~ # chroot /target flash-kernel --force 4.8.0-1-marvell
DTB: orion5x-linkstation-lsgl.dtb
Installing /etc/flash-kernel/dtbs/orion5x-linkstation-lsgl.dtb into /boot/dtbs/4.8.0-1-marvell/orion5x-linkstation-lsgl.dtb
Taking backup of orion5x-linkstation-lsgl.dtb.
Installing new orion5x-linkstation-lsgl.dtb.
Installing /etc/flash-kernel/dtbs/orion5x-linkstation-lsgl.dtb into /boot/dtbs/4.8.0-1-marvell/orion5x-linkstation-lsgl.dtb
Taking backup of orion5x-linkstation-lsgl.dtb.
Installing new orion5x-linkstation-lsgl.dtb.
flash-kernel: installing version 4.8.0-1-marvell
flash-kernel: appending /etc/flash-kernel/dtbs/orion5x-linkstation-lsgl.dtb to kernel
Generating kernel u-boot image... done.
Taking backup of uImage.buffalo.
Installing new uImage.buffalo.
Generating initramfs u-boot image... done.
Taking backup of initrd.buffalo.
Installing new initrd.buffalo.

>- check the log of above command, it should use
>/etc/flash-kernel/dtbs/orion5x-linkstation-lsgl.dtb for building
>uImage.buffalo, the kernel image for booting
>- reboot and see the result
>
>If the above steps works for you, I can submit the above patch to
>kernel upstream, and include it into Stretch release.
>So I'm looking forward to your result. Thank you!

It booted! :)

Linux xenon 4.8.0-1-marvell #1 Debian 4.8.5-1 (2016-10-28) armv5tel GNU/Linux

Unfortunately dmesg and kern.log are flooded with messages like:

[  161.781360] Bad eraseblock 32764 at 0x0001fff00000
[  161.786261] Bad eraseblock 32765 at 0x0001fff40000
[  161.791159] Bad eraseblock 32766 at 0x0001fff80000
[  161.796058] Bad eraseblock 32767 at 0x0001fffc0000

and boot takes a very long time due to logging all these messages. I 
suppose that's a side-effect of using the kurobox-pro dtb - guessing it 
has a different flash layout?

Anyway, the change to nr-ports=2 looks good!

Tested-by: Ryan Tandy <ryan@nardis.ca>

Thanks!

[toc] | [prev] | [next] | [standalone]

#55855

Control: tag -1 patch

On Sat, Nov 26, 2016 at 12:59:21PM -0800, Ryan Tandy wrote:
>Unfortunately dmesg and kern.log are flooded with messages like:
>
>[  161.781360] Bad eraseblock 32764 at 0x0001fff00000
>[  161.786261] Bad eraseblock 32765 at 0x0001fff40000
>[  161.791159] Bad eraseblock 32766 at 0x0001fff80000
>[  161.796058] Bad eraseblock 32767 at 0x0001fffc0000
>
>and boot takes a very long time due to logging all these messages. I 
>suppose that's a side-effect of using the kurobox-pro dtb - guessing 
>it has a different flash layout?

Looks like that was the case. I applied your patch to 
orion5x-linkstation-lsgl.dts, rebuilt the dtb, dropped it into 
/etc/flash-kernel/dtbs, and ran flash-kernel again. Much better:

[    0.000000] Booting Linux on physical CPU 0x0
[    0.000000] Linux version 4.8.0-1-marvell (debian-kernel@lists.debian.org) (gcc version 5.4.1 20161019 (Debian 5.4.1-3) ) #1 Debian 4.8.5-1 (2016-10-28)
[    0.000000] CPU: Feroceon [41069260] revision 0 (ARMv5TEJ), cr=a005317f
[    0.000000] CPU: VIVT data cache, VIVT instruction cache
[    0.000000] OF: fdt:Machine model: Buffalo Linkstation Pro/Live
 
[...]

[    3.297636] sata_mv sata_mv.0: version 1.28
[    3.297844] sata_mv sata_mv.0: cannot get optional clkdev
[    3.320803] sata_mv sata_mv.0: slots 32 ports 2
[    3.386495] scsi host0: sata_mv
[    3.403996] scsi host1: sata_mv
[    3.411527] ata1: SATA max UDMA/133 irq 28
[    3.415723] ata2: SATA max UDMA/133 irq 28
[    3.733937] ata1: SATA link down (SStatus 0 SControl 300)
[    4.212767] ata2: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
[    4.241078] ata2.00: ATA-7: SAMSUNG SP2504C, VT100-41, max UDMA7
[    4.247154] ata2.00: 488397168 sectors, multi 0: LBA48 NCQ (depth 31/32)
[    4.293102] ata2.00: configured for UDMA/133
[    4.298609] scsi 1:0:0:0: Direct-Access     ATA      SAMSUNG SP2504C  0-41 PQ: 0 ANSI: 5
[    4.402529] sd 1:0:0:0: [sda] 488397168 512-byte logical blocks: (250 GB/233 GiB)
[    4.416368] sd 1:0:0:0: [sda] Write Protect is off
[    4.421270] sd 1:0:0:0: [sda] Mode Sense: 00 3a 00 00
[    4.421705] sd 1:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[    4.474159]  sda: sda1 sda2 sda3 < sda5 >
[    4.493519] sd 1:0:0:0: [sda] Attached SCSI disk

[toc] | [prev] | [next] | [standalone]

#55856

On Sun, Nov 27, 2016 at 7:50 AM, Ryan Tandy <ryan@nardis.ca> wrote:
> Control: tag -1 patch
>
> On Sat, Nov 26, 2016 at 12:59:21PM -0800, Ryan Tandy wrote:
>>
>> Unfortunately dmesg and kern.log are flooded with messages like:
>>
>> [  161.781360] Bad eraseblock 32764 at 0x0001fff00000
>> [  161.786261] Bad eraseblock 32765 at 0x0001fff40000
>> [  161.791159] Bad eraseblock 32766 at 0x0001fff80000
>> [  161.796058] Bad eraseblock 32767 at 0x0001fffc0000
>>
>> and boot takes a very long time due to logging all these messages. I
>> suppose that's a side-effect of using the kurobox-pro dtb - guessing it has
>> a different flash layout?

Glad to hear it works for you.

Actually there's less flash on Linkstation GL or Pro/Live than on KuroBox Pro.
So using kurobox-pro dtb is just a tentative solution for test purpose.

> On Sun, Nov 27, 2016 at 01:20:10AM +0900, Roger Shimizu wrote:
>>
>> - start your device by kernel 4.3 first.
>> - confirm your have flash-kernel and linux-image-4.8.0-1-marvell
>> installed.
>> - cp /usr/lib/linux-image-4.8.0-1-marvell/orion5x-kuroboxpro.dtb
>> /etc/flash-kernel/dtbs/orion5x-linkstation-lsgl.dtb
>> - flash-kernel --force 4.8.0-1-marvell

Sorry, I forgot to mention after steps above, when you want to switch
to another kernel version, and run like
- flash-kernel --force <debian-kernel-version>

The flash-kernel will treat the device as Kurobox Pro. So you need the
following command to tell flash-kernel that your device is Linkstation
Pro/Live:

# echo -n "Buffalo Linkstation Pro/Live" > /etc/flash-kernel/machine

After that, you can switch back to kernel 4.3 or 4.4, which use the
device file instead of device tree, if you still installed on your
system:
# flash-kernel --force 4.3.0-1-orion5x
- or -
# flash-kernel --force 4.4.0-1-marvell

Cheers,
-- 
Roger Shimizu, GMT +9 Tokyo
PGP/GPG: 4096R/6C6ACD6417B3ACB1

[toc] | [prev] | [next] | [standalone]

#55854 — Processed: Re: Bug#845611: linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL)

Processing control commands:

> tag -1 patch
Bug #845611 [linux-image-4.8.0-1-marvell] linux-image-4.8.0-1-marvell: hard drive not detected on LinkStation Pro (LS-GL)
Added tag(s) patch.

-- 
845611: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=845611
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.kernel

csiph-web