5.2 hppa merge seems to work on my machines

5.2 hppa merge seems to work on my machines

[Meelis Roos]

Just tested v5.1-5445-g80f232121b69 git on A180c, rp2450, rp2470, rp3410 and rp3440 and
it seems to work everywhere for me.

-- 
Meelis Roos

Re: 5.2 hppa merge seems to work on my machines

[Helge Deller]

On 08.05.19 16:23, Meelis Roos wrote:
> Just tested v5.1-5445-g80f232121b69 git on A180c, rp2450, rp2470,
> rp3410 and rp3440 and it seems to work everywhere for me.

Great!
Thanks for testing, Meelis!

Helge

Re: 5.2 hppa merge seems to work on my machines

[Carlo Pisani]

hi
guys, unfortunately, it panics with Marvell MV88SX6081 sata controller.
I will publish more details later

I am testing a few sata controllers, see here
http://www.downthebunker.com/reloaded/space/viewtopic.php?f=50&p=1563

Il giorno mer 8 mag 2019 alle ore 21:15 Helge Deller <deller@gmx.de> ha scritto:
>
> On 08.05.19 16:23, Meelis Roos wrote:
> > Just tested v5.1-5445-g80f232121b69 git on A180c, rp2450, rp2470,
> > rp3410 and rp3440 and it seems to work everywhere for me.
>
> Great!
> Thanks for testing, Meelis!
>
> Helge

Re: 5.2 hppa merge seems to work on my machines

[Helge Deller]

On 10.05.19 13:43, Carlo Pisani wrote:
> hi
> guys, unfortunately, it panics with Marvell MV88SX6081 sata controller.
> I will publish more details later

syslog ?

> I am testing a few sata controllers, see here
> http://www.downthebunker.com/reloaded/space/viewtopic.php?f=50&p=1563

In one of our buildds we use a
Silicon Image, Inc. SiI 3112 Serial ATA Controller (rev 02)
PCI card.
I'm not sure if it's fast, but at least it seems somewhat stable.

Helge


>
> Il giorno mer 8 mag 2019 alle ore 21:15 Helge Deller <deller@gmx.de> ha scritto:
>>
>> On 08.05.19 16:23, Meelis Roos wrote:
>>> Just tested v5.1-5445-g80f232121b69 git on A180c, rp2450, rp2470,
>>> rp3410 and rp3440 and it seems to work everywhere for me.
>>
>> Great!
>> Thanks for testing, Meelis!
>>
>> Helge

Re: 5.2 hppa merge seems to work on my machines

[John David Anglin]

On 2019-05-11 6:47 a.m., Helge Deller wrote:
>> I am testing a few sata controllers, see here
>> http://www.downthebunker.com/reloaded/space/viewtopic.php?f=50&p=1563
> In one of our buildds we use a
> Silicon Image, Inc. SiI 3112 Serial ATA Controller (rev 02)
> PCI card.
This Syba card also works:
00:01.0 RAID bus controller: Silicon Image, Inc. SiI 3124 PCI-X Serial ATA Controller (rev 02)

Dave

-- 
John David Anglin  dave.anglin@bell.net

Re: 5.2 hppa merge seems to work on my machines

[Helge Deller]

On 11.05.19 17:27, John David Anglin wrote:
> On 2019-05-11 6:47 a.m., Helge Deller wrote:
>>> I am testing a few sata controllers, see here
>>> http://www.downthebunker.com/reloaded/space/viewtopic.php?f=50&p=1563
>> In one of our buildds we use a
>> Silicon Image, Inc. SiI 3112 Serial ATA Controller (rev 02)
>> PCI card.
> This Syba card also works:
> 00:01.0 RAID bus controller: Silicon Image, Inc. SiI 3124 PCI-X Serial ATA Controller (rev 02)

I've added both to a new Wiki section "SATA Controllers":
https://parisc.wiki.kernel.org/index.php/Hardware_support

Helge

Re: 5.2 hppa merge seems to work on my machines

[Rolf Eike Beer]

[-- Attachment #1: Type: text/plain, Size: 853 bytes --]

Am Samstag, 11. Mai 2019, 17:43:23 CEST schrieb Helge Deller:
> On 11.05.19 17:27, John David Anglin wrote:
> > On 2019-05-11 6:47 a.m., Helge Deller wrote:
> >>> I am testing a few sata controllers, see here
> >>> http://www.downthebunker.com/reloaded/space/viewtopic.php?f=50&p=1563
> >> 
> >> In one of our buildds we use a
> >> Silicon Image, Inc. SiI 3112 Serial ATA Controller (rev 02)
> >> PCI card.
> > 
> > This Syba card also works:
> > 00:01.0 RAID bus controller: Silicon Image, Inc. SiI 3124 PCI-X Serial ATA
> > Controller (rev 02)
> I've added both to a new Wiki section "SATA Controllers":
> https://parisc.wiki.kernel.org/index.php/Hardware_support

You can add this one:

40:03.0 RAID bus controller: VIA Technologies, Inc. VT6421 IDE/SATA Controller 
(rev 50)

I have been using it for years, first in my C3600, now in my C8000.

Eike

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

Re: 5.2 hppa merge seems to work on my machines

[Helge Deller]

[-- Attachment #1.1: Type: text/plain, Size: 957 bytes --]

On 11.05.19 19:43, Rolf Eike Beer wrote:
> Am Samstag, 11. Mai 2019, 17:43:23 CEST schrieb Helge Deller:
>> On 11.05.19 17:27, John David Anglin wrote:
>>> On 2019-05-11 6:47 a.m., Helge Deller wrote:
>>>>> I am testing a few sata controllers, see here
>>>>> http://www.downthebunker.com/reloaded/space/viewtopic.php?f=50&p=1563
>>>>
>>>> In one of our buildds we use a
>>>> Silicon Image, Inc. SiI 3112 Serial ATA Controller (rev 02)
>>>> PCI card.
>>>
>>> This Syba card also works:
>>> 00:01.0 RAID bus controller: Silicon Image, Inc. SiI 3124 PCI-X Serial ATA
>>> Controller (rev 02)
>> I've added both to a new Wiki section "SATA Controllers":
>> https://parisc.wiki.kernel.org/index.php/Hardware_support
> 
> You can add this one:
> 
> 40:03.0 RAID bus controller: VIA Technologies, Inc. VT6421 IDE/SATA Controller 
> (rev 50)
> 
> I have been using it for years, first in my C3600, now in my C8000.

Added.
Thanks!
Helge


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

Re: 5.2 hppa merge seems to work on my machines

[Carlo Pisani]

guys, bad news!
as described here (1) we are testing several sata controllers by a script that
1) copy a very big file cp data_16GB.bin copy.bin
2) compare the md5sum for checking the integrity

we have a couple of SIL24 controllers, and they both work but only
with small files <8Gbyte
trying to copy a big file, 16GB, results in the following HPMC error

A Data I/O Fetch Timeout occurred while CPU 0 was requesting
information from a device at the path 10/6/2/0 (PCI slot 2).

I have uploaded a ser-pim-log there for more details, it's related to
the Marvell MV88SX6081's sata controller, but I have the same behavior
with a Silicon Image Sil3124 controller

I am on the last kernel v5.1 from the git repository.

Performances are, in both cases, very low: something about
10Mbyte/sec, which is ... really too low

Tomorrow I will try the old 4.16, which I remember to be more stable
with the SIL24 controller.

(1) http://www.downthebunker.com/reloaded/space/viewtopic.php?f=50&p=1618

p.s.
has anyone ever tried an FC adapter on an FC-bridge?

Il giorno mer 8 mag 2019 alle ore 21:15 Helge Deller <deller@gmx.de> ha scritto:
>
> On 08.05.19 16:23, Meelis Roos wrote:
> > Just tested v5.1-5445-g80f232121b69 git on A180c, rp2450, rp2470,
> > rp3410 and rp3440 and it seems to work everywhere for me.
>
> Great!
> Thanks for testing, Meelis!
>
> Helge

Re: 5.2 hppa merge seems to work on my machines

[John David Anglin]

On 2019-05-11 5:41 p.m., Carlo Pisani wrote:
> guys, bad news!
> as described here (1) we are testing several sata controllers by a script that
> 1) copy a very big file cp data_16GB.bin copy.bin
> 2) compare the md5sum for checking the integrity
>
> we have a couple of SIL24 controllers, and they both work but only
> with small files <8Gbyte
> trying to copy a big file, 16GB, results in the following HPMC error
>
> A Data I/O Fetch Timeout occurred while CPU 0 was requesting
> information from a device at the path 10/6/2/0 (PCI slot 2).
I don't see problem on my c8000:

dave@atlas:~/big$ cp x x2
dave@atlas:~/big$ md5sum *
9e6b3d79358077360a55861a5054f878  x
9e6b3d79358077360a55861a5054f878  x2
dave@atlas:~/big$ ls -l
total 49152
-rw-r--r-- 1 dave dave 25165824 May 11 18:31 x
-rw-r--r-- 1 dave dave 25165824 May 11 18:31 x2

Note card is PCIX slot.

Dave

-- 
John David Anglin  dave.anglin@bell.net

Re: 5.2 hppa merge seems to work on my machines

[Carlo Pisani]

> Note card is PCIX slot.

tested on a C3600
with a Silicon Image Sil3124
the first card's brand is "SYBA-II" (from Amazon dot com)
it's 4 SATA channels, PCI 32bit

the second card's brand is "Adaptec 1210SA"
it's a 2 SATA channels, PCI 32bit

we do not have a PCI 64bit Silicon Image Sil3124

which is your card's brand?

are you using PCI-hardfail ? or softfail?
do you have CONFIG_MLONGCALLS=y ?

unfortunately, we do not have a C8000 to test

Re: 5.2 hppa merge seems to work on my machines

[Meelis Roos]

> 1) copy a very big file cp data_16GB.bin copy.bin
> 2) compare the md5sum for checking the integrity

Problems from test like this suggest a problem outside the libata drivers themselves.

To me, it sounds rather something like dma mapping/sync errors, or memory
management errors (possibly dependent on hppa bus(bridge types).

-- 
Meelis Roos

Re: 5.2 hppa merge seems to work on my machines

[Carlo Pisani]

> Problems from test like this suggest a problem outside the libata drivers themselves.
> To me, it sounds rather something like dma mapping/sync errors, or memory
> management errors (possibly dependent on hppa bus(bridge types).

It's not installed in our C3600 at the moment,
but it seems what we observed 2 days ago with Highpoint RocketRAID
1640HTP374 card
it's PCI 32bit, 4 SATA channels, it doesn't crash, but it randomly
offers file corruptions.
The dmseg shows a lot of DMA errors.

How to investigate? I am not a kernel developer, but rather an HDL-guy (FPGA)

Re: 5.2 hppa merge seems to work on my machines

[John David Anglin]

On 2019-05-12 6:51 a.m., Carlo Pisani wrote:
>> Problems from test like this suggest a problem outside the libata drivers themselves.
>> To me, it sounds rather something like dma mapping/sync errors, or memory
>> management errors (possibly dependent on hppa bus(bridge types).
> It's not installed in our C3600 at the moment,
> but it seems what we observed 2 days ago with Highpoint RocketRAID
> 1640HTP374 card
> it's PCI 32bit, 4 SATA channels, it doesn't crash, but it randomly
> offers file corruptions.
> The dmseg shows a lot of DMA errors.
I agree.  The memory errors in the PIM dump are recorded by the PDC firmware and they don't
directly involve the linux drivers.  Might be hardware problem.

-- 
John David Anglin  dave.anglin@bell.net

Re: 5.2 hppa merge seems to work on my machines

[Carlo Pisani]

> I agree.  The memory errors in the PIM dump are recorded by the PDC firmware and they don't
> directly involve the linux drivers.  Might be hardware problem.

which kind of error problem?

those two cards have no problem when tested on PowerMac-G4
and we have a second C3600, reporting the same behavior

so, supposing those cards are OK, and C3600s are OK
what might be wrong about the hardware?

Re: 5.2 hppa merge seems to work on my machines

[John David Anglin]

On 2019-05-12 9:45 a.m., Carlo Pisani wrote:
>> I agree.  The memory errors in the PIM dump are recorded by the PDC firmware and they don't
>> directly involve the linux drivers.  Might be hardware problem.
> which kind of error problem?
|Memory Error Log Information:

Timestamp =
  Sat May  11 11:37:53 GMT 2019    (20:19:05:11:11:37:53)


'9000/785 B,C,J Workstation Memory Error Log', rev 0, 64 bytes:

This log displays the contents of memory specific registers when the
HPMC occurred.  If there are multiple memory errors, the order they are
listed is not indicative of the order they occurred.

                                   Trans  Addr
   Memory Error Type(s)  OV  MID    ID    par  CP   DIMM       Runway Address
   --------------------  --  ---  -----  ----  --  -------  -------------------
1) Correctable Mem       1   0x6  0x a   na    na  05       0x       0036c76b40

                                                Syndrome
                                           ------------------
                                        1) 0x94
<Press any key to continue (q to quit)>

I/O Module Error Log Information:

Timestamp =
  Sat May  11 11:37:53 GMT 2019    (20:19:05:11:11:37:53)


'9000/785 B,C,J Workstation IO Error Log', rev 0, 228 bytes:

Rope     Word1        Word2            Word3
------ ------------ ------------
   0    0x00000000   0x0e0cc009   0x00000000fed30048
   1    0x00000000   0x1e0cc009   0x00000000fed32048
   2    ----------   0x2e0cc009   ------------------
   3    ----------   0x3e0cc009   ------------------
   4    0x00000000   0x4e0cc009   0x00000000fed38048
   5    ----------   0x5e0cc009   ------------------
   6    0x00000000   0x6e0cc2a9   0x00000000fed3c048
   7    ----------   0x7e0cc009   ------------------
|
>
> those two cards have no problem when tested on PowerMac-G4
> and we have a second C3600, reporting the same behavior
>
> so, supposing those cards are OK, and C3600s are OK
> what might be wrong about the hardware?
>
-- 
John David Anglin  dave.anglin@bell.net

Re: 5.2 hppa merge seems to work on my machines

[John David Anglin]

On 2019-05-12 2:28 a.m., Carlo Pisani wrote:
> tested on a C3600
> with a Silicon Image Sil3124
> the first card's brand is "SYBA-II" (from Amazon dot com)
> it's 4 SATA channels, PCI 32bit
There is also this card with Silicon Image Sil3512:
https://www.sybausa.com/index.php?route=product/product&product_id=171&search=SD-SATA150R
https://www.amazon.ca/Syba-Controller-Silicon-Software-Components/dp/B000BU7XNG?th=1
Keying looks universal.

-- 
John David Anglin  dave.anglin@bell.net