Promise TX4200 support?

Promise TX4200 support?

[Daniel Drake]

Hi,

I recieved an email from someone claiming to be stuck with Linux 2.4, due to 
relying on a Promise TX4200 disk controller (using the fdsata driver from 
promise's website, which is 2.4-only):

0000:01:09.0 RAID bus controller: Promise Technology, Inc.: Unknown device 
3519 (rev 02)
         Subsystem: Promise Technology, Inc.: Unknown device 3519
         Flags: bus master, 66Mhz, medium devsel, latency 64, IRQ 9
         I/O ports at dc00 [size=128]
         I/O ports at d800 [size=256]
         Memory at ff8ff000 (32-bit, non-prefetchable) [size=4K]
         Memory at ff8c0000 (32-bit, non-prefetchable) [size=128K]
         Expansion ROM at ff8e0000 [disabled] [size=64K]
         Capabilities: [60] Power Management version 2

What is the status of this on 2.6? I found a blank changeset (??) in the mail 
below, from 24th May:

Jeff Garzik wrote:
 > Please pull the 'new-ids' branch from
 >
 > rsync://rsync.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev.git
 >
 > This add new PCI ids to some SATA drivers.
<snip>
 > commit 37c15447c565ab458ee3778e198d08f4041caa99
 > tree 2eda289903e3bf19eebce7d5f9aaed2240a02479
 > parent 9422e59ddf6cae68e46d7a2c3afe1ce4e739d3eb
 > author Martin Povolny <martin.povolny@solnet.cz> Mon, 16 May 2005 02:41:00 
-0400
 > committer Jeff Garzik <jgarzik@pobox.com> Mon, 16 May 2005 02:41:00 -0400
 >
 > [PATCH] sata_promise: new PCI ID for TX4200
 >
 > [note - blank changeset]
 >

Was this accidently removed, or is the sata_promise driver actually 
incompatible with this hardware?

Thanks,
Daniel

Re: Promise TX4200 support?

[Martin Povolný]

Daniel Drake wrote:
> Hi,
> 
> I recieved an email from someone claiming to be stuck with Linux 2.4,
> due to relying on a Promise TX4200 disk controller (using the fdsata
> driver from promise's website, which is 2.4-only):
> 
> 0000:01:09.0 RAID bus controller: Promise Technology, Inc.: Unknown
> device 3519 (rev 02)
>         Subsystem: Promise Technology, Inc.: Unknown device 3519
>         Flags: bus master, 66Mhz, medium devsel, latency 64, IRQ 9
>         I/O ports at dc00 [size=128]
>         I/O ports at d800 [size=256]
>         Memory at ff8ff000 (32-bit, non-prefetchable) [size=4K]
>         Memory at ff8c0000 (32-bit, non-prefetchable) [size=128K]
>         Expansion ROM at ff8e0000 [disabled] [size=64K]
>         Capabilities: [60] Power Management version 2
> 
> What is the status of this on 2.6? I found a blank changeset (??) in the
> mail below, from 24th May:
> 
> Jeff Garzik wrote:
>> Please pull the 'new-ids' branch from
>>
>> rsync://rsync.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev.git
>>
>> This add new PCI ids to some SATA drivers.
> <snip>
>> commit 37c15447c565ab458ee3778e198d08f4041caa99
>> tree 2eda289903e3bf19eebce7d5f9aaed2240a02479
>> parent 9422e59ddf6cae68e46d7a2c3afe1ce4e739d3eb
>> author Martin Povolny <martin.povolny@solnet.cz> Mon, 16 May 2005
> 02:41:00 -0400
>> committer Jeff Garzik <jgarzik@pobox.com> Mon, 16 May 2005 02:41:00 -0400
>>
>> [PATCH] sata_promise: new PCI ID for TX4200
>>
>> [note - blank changeset]
>>
> 
> Was this accidently removed, or is the sata_promise driver actually
> incompatible with this hardware?
> 

We are succesfully running patched sata_promise with 3 disks in a
raid5/raid1 setup. (Patched against ubuntu linux-image 2.6.11-1-686
package.)

# check_partitions
disk: [8.0] => '/dev/sda', 279.4 GB
         1 : /dev/sda1 : Linux raid autodetect     (      55 MB)
         2 : /dev/sda2 : Linux raid autodetect     (  285640 MB)
         3 : /dev/sda3 : Linux swap                (     486 MB)
disk: [8.16] => '/dev/sdb', 279.4 GB
         1 : /dev/sdb1 : Linux raid autodetect     (      55 MB)
         2 : /dev/sdb2 : Linux raid autodetect     (  285640 MB)
         3 : /dev/sdb3 : Linux swap                (     486 MB)
disk: [8.32] => '/dev/sdc', 279.4 GB
         1 : /dev/sdc1 : Linux raid autodetect     (      55 MB)
         2 : /dev/sdc2 : Linux raid autodetect     (  285640 MB)
         3 : /dev/sdc3 : Linux swap                (     486 MB)
disk: [9.0] => '/dev/md',
         0 : /dev/md0  : raid1                     (      55 MB)
                uuid: db3bdcf4:3e4774c7:b2541959:eeef67e0
                0 active sync   /dev/sda1
                1 active sync   /dev/sdb1
                2 active sync   /dev/sdc1
         1 : /dev/md1  : raid5                     (  571280 MB)
                uuid: b9562e24:8a095e31:40702712:19009a40
                0 active sync   /dev/sda2
                1 active sync   /dev/sdb2
                2 active sync   /dev/sdc2

'lspci -v' says:

02:02.0 RAID bus controller: Promise Technology, Inc.: Unknown device
3519 (rev 02)
        Subsystem: Promise Technology, Inc.: Unknown device 3519
        Flags: bus master, 66Mhz, medium devsel, latency 64, IRQ 18
        I/O ports at dc00 [size=128]
        I/O ports at d800 [size=256]
        Memory at feaff000 (32-bit, non-prefetchable) [size=4K]
        Memory at feac0000 (32-bit, non-prefetchable) [size=128K]
        Expansion ROM at feae0000 [disabled] [size=64K]
        Capabilities: [60] Power Management version 2

No problems at all.

Regards,

-- 
Mgr. Martin Povolný, soLNet, s.r.o.,
+420777714458, martin.povolny@solnet.cz

Re: Promise TX4200 support?

[Daniel Drake]

Hi Martin,

Martin Povolný wrote:
> We are succesfully running patched sata_promise with 3 disks in a
> raid5/raid1 setup. (Patched against ubuntu linux-image 2.6.11-1-686
> package.)

Could you please either send in your patch, or tell me which board_ setting 
(2037x/20319/20619) the device ID table should include so I can write submit 
one myself.

> 'lspci -v' says:
> 
> 02:02.0 RAID bus controller: Promise Technology, Inc.: Unknown device
> 3519 (rev 02)
>         Subsystem: Promise Technology, Inc.: Unknown device 3519
>         Flags: bus master, 66Mhz, medium devsel, latency 64, IRQ 18
>         I/O ports at dc00 [size=128]
>         I/O ports at d800 [size=256]
>         Memory at feaff000 (32-bit, non-prefetchable) [size=4K]
>         Memory at feac0000 (32-bit, non-prefetchable) [size=128K]
>         Expansion ROM at feae0000 [disabled] [size=64K]
>         Capabilities: [60] Power Management version 2

Thanks,
Daniel

Re: Promise TX4200 support?

[Martin Povolný]

Daniel Drake wrote:
> Hi Martin,
> 
> Martin Povolný wrote:
> 
>> We are succesfully running patched sata_promise with 3 disks in a
>> raid5/raid1 setup. (Patched against ubuntu linux-image 2.6.11-1-686
>> package.)
> 
> 
> Could you please either send in your patch, or tell me which board_
> setting (2037x/20319/20619) the device ID table should include so I can
> write submit one myself.
> 

For me it works with 20319, but I don't understand the difference
between different settings.

*** sata_promise.c	2005-05-11 21:22:20.000000000 +0200
--- sata_promise_new.c	2005-05-11 21:22:02.000000000 +0200
***************
*** 164,171 ****
--- 164,173 ----
  	{ PCI_VENDOR_ID_PROMISE, 0x3318, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
  	  board_20319 },
  	{ PCI_VENDOR_ID_PROMISE, 0x3319, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
  	  board_20319 },
+ 	{ PCI_VENDOR_ID_PROMISE, 0x3519, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
+ 	  board_20319 },
  	{ PCI_VENDOR_ID_PROMISE, 0x3d18, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
  	  board_20319 },

  	{ }	/* terminate list */

Regards,

-- 
Mgr. Martin Povolný, soLNet, s.r.o.,
+420777714458, martin.povolny@solnet.cz

Re: Promise TX4200 support?

[Daniel Drake]

Martin Povolný wrote:
> For me it works with 20319, but I don't understand the difference
> between different settings.

20319 is 4 port SATA.
2037x is 2 port SATA, optionally with 1 PATA port
20619 is 4 port PATA

So I believe 20319 is the correct option.

Jeff, the chip on the TX4200 is actually a PDC40519 but it meets the 
description of the 20319. Is something like the patch below ok, or should we 
add a new board_ entry?

> *** sata_promise.c	2005-05-11 21:22:20.000000000 +0200
> --- sata_promise_new.c	2005-05-11 21:22:02.000000000 +0200
> ***************
> *** 164,171 ****
> --- 164,173 ----
>   	{ PCI_VENDOR_ID_PROMISE, 0x3318, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
>   	  board_20319 },
>   	{ PCI_VENDOR_ID_PROMISE, 0x3319, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
>   	  board_20319 },
> + 	{ PCI_VENDOR_ID_PROMISE, 0x3519, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
> + 	  board_20319 },
>   	{ PCI_VENDOR_ID_PROMISE, 0x3d18, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
>   	  board_20319 },
> 
>   	{ }	/* terminate list */


Thanks,
Daniel