Driver for sata adapter promise sata300 tx4

Driver for sata adapter promise sata300 tx4

[Otto Meier]

Hi,

This card use the sata chip pdc 40718 (as of my card)
the lastest sata_promise kernel with sata promise patch driver doesn't 
recognise
this card.

I added the following line to static struct pci_device_id 
pdc_ata_pci_tbl[]  in sata_promise.c:

       { PCI_VENDOR_ID_PROMISE, 0x3d17, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
         board_20319 },

and the card was recognised and seam to work without errors so far.

bilding a soft raid5 on it and moving data doesn't broke it.

It would be interesting to hear from the guru's if this is ok to do?.

Otto

Re: Driver for sata adapter promise sata300 tx4

[Daniel Drake]

Otto Meier wrote:
> This card use the sata chip pdc 40718 (as of my card)
> the lastest sata_promise kernel with sata promise patch driver doesn't 
> recognise
> this card.
> 
> I added the following line to static struct pci_device_id 
> pdc_ata_pci_tbl[]  in sata_promise.c:
> 
>        { PCI_VENDOR_ID_PROMISE, 0x3d17, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
>          board_20319 },
> 
> and the card was recognised and seam to work without errors so far.

Yes, this should be fine (this is a 4-port SATA card right?)

Are you happy to produce and submit a patch yourself (read 
Documentation/SubmittingPatches) or should I submit one for you?

Thanks,
Daniel

Re: Driver for sata adapter promise sata300 tx4

[Otto Meier]

Daniel Drake wrote:
> Otto Meier wrote:
>> This card use the sata chip pdc 40718 (as of my card)
>> the lastest sata_promise kernel with sata promise patch driver 
>> doesn't recognise
>> this card.
>>
>> I added the following line to static struct pci_device_id 
>> pdc_ata_pci_tbl[] in sata_promise.c:
>>
>> { PCI_VENDOR_ID_PROMISE, 0x3d17, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
>> board_20319 },
>>
>> and the card was recognised and seam to work without errors so far.
>
> Yes, this should be fine (this is a 4-port SATA card right?)
>
> Are you happy to produce and submit a patch yourself (read 
> Documentation/SubmittingPatches) or should I submit one for you?
>
> Thanks,
> Daniel
>
>
Yes you are right it is a 4-port sata-II 300 card (PDC40718 ). According to
the promise feature list it should support :

SATA300™ TX4 Highlights

    * Native Command Queuing (NCQ)
    * SATA Tagged Command Queuing (TCQ)
    * Large LBA support for drives above 137GB
    * Supports Serial ATAPI devices
    * Disk Activity LED Headers
    * Flexible future-proof upgrade for users with motherboards that
      only have a PCI interface


My question is also are these features (NCQ/TCQ) and the heigher 
datarate be supported by this
modification? or is only the basic feature set of sata 150 TX4 supported?

Here is the patch:

--- linux/drivers/scsi/sata_promise.c.orig 2005-08-01 17:09:48.474824778 
+0200
+++ linux/drivers/scsi/sata_promise.c 2005-07-31 12:57:06.415979512 +0200
@@ -183,6 +183,8 @@ static struct pci_device_id pdc_ata_pci_
board_20319 },
{ PCI_VENDOR_ID_PROMISE, 0x3319, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
board_20319 },
+ { PCI_VENDOR_ID_PROMISE, 0x3d17, 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 },



Thanks
Otto

Re: Driver for sata adapter promise sata300 tx4

[Otto Meier]

Daniel Drake wrote:
> Otto Meier wrote:
>> This card use the sata chip pdc 40718 (as of my card)
>> the lastest sata_promise kernel with sata promise patch driver 
>> doesn't recognise
>> this card.
>>
>> I added the following line to static struct pci_device_id 
>> pdc_ata_pci_tbl[] in sata_promise.c:
>>
>> { PCI_VENDOR_ID_PROMISE, 0x3d17, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
>> board_20319 },
>>
>> and the card was recognised and seam to work without errors so far.
>
> Yes, this should be fine (this is a 4-port SATA card right?)
>
> Are you happy to produce and submit a patch yourself (read 
> Documentation/SubmittingPatches) or should I submit one for you?
>
> Thanks,
> Daniel
>
>
Yes you are right it is a 4-port sata-II 300 card (PDC40718 ). According to
the promise feature list it should support :

SATA300™ TX4 Highlights

   * Native Command Queuing (NCQ)
   * SATA Tagged Command Queuing (TCQ)
   * Large LBA support for drives above 137GB
   * Supports Serial ATAPI devices
   * Disk Activity LED Headers
   * Flexible future-proof upgrade for users with motherboards that
     only have a PCI interface


My question is also are these features (NCQ/TCQ) and the heigher 
datarate be supported by this
modification? or is only the basic feature set of sata 150 TX4 supported?

Here is the patch:

--- linux/drivers/scsi/sata_promise.c.orig 2005-08-01 17:09:48.474824778 +0200
+++ linux/drivers/scsi/sata_promise.c 2005-07-31 12:57:06.415979512 +0200
@@ -183,6 +183,8 @@ static struct pci_device_id pdc_ata_pci_board_20319 },
{ PCI_VENDOR_ID_PROMISE, 0x3319, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 
board_20319 },
+ { PCI_VENDOR_ID_PROMISE, 0x3d17, 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 },



Thanks
Otto


I just saw the patches of  Luke Kosewski regarding the SATA150 TX4 antipating
them it might be right to modify the patch to 


--- linux/drivers/scsi/sata_promise.c.orig 2005-08-01 17:09:48.474824778 +0200
+++ linux/drivers/scsi/sata_promise.c 2005-07-31 12:57:06.415979512 +0200
@@ -183,6 +183,8 @@ static struct pci_device_id pdc_ata_pci_board_20319 },
{ PCI_VENDOR_ID_PROMISE, 0x3319, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 
board_20319 },
+ { PCI_VENDOR_ID_PROMISE, 0x3d17, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
+ board_40518 },
{ PCI_VENDOR_ID_PROMISE, 0x3d18, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
board_20319 },

bye 
Otto

Re: Driver for sata adapter promise sata300 tx4

[Daniel Drake]

Otto Meier wrote:
> My question is also are these features (NCQ/TCQ) and the heigher 
> datarate be supported by this
> modification? or is only the basic feature set of sata 150 TX4 supported?

NCQ support is under development. Search the archives for Jens Axboe's recent 
patches to support this. I don't know about TCQ.

> Here is the patch:
> 
> --- linux/drivers/scsi/sata_promise.c.orig 2005-08-01 17:09:48.474824778 
> +0200
> +++ linux/drivers/scsi/sata_promise.c 2005-07-31 12:57:06.415979512 +0200

Your patch will not apply because it is linewrapped. You also need to submit 
it in a mail of its own to the relevent lists and maintainer, with your 
sign-off  (see Documentation/SubmittingPatches)

> I just saw the patches of  Luke Kosewski regarding the SATA150 TX4 
> antipating
> them it might be right to modify the patch to
> 
> + { PCI_VENDOR_ID_PROMISE, 0x3d17, PCI_ANY_ID, PCI_ANY_ID, 0, 0,
> + board_40518 },

It won't compile like this because board_40518 is not a valid identifier. I 
also think it doesn't really matter as it looks like these identifier codes 
have lost their numerical meanings, and now just signify:

board_2037x - 2 port SATA, maybe with an extra PATA port
board_20319 - 4 port SATA
board_20619 - 4 port PATA

Daniel

Re: Driver for sata adapter promise sata300 tx4

[Otto Meier]

Jens Axboe wrote:
> On Mon, Aug 01 2005, Jeff Garzik wrote:
>   
>> Jens Axboe wrote:
>>     
>>> On Mon, Aug 01 2005, Daniel Drake wrote:
>>>
>>>       
>>>> Otto Meier wrote:
>>>>
>>>>         
>>>>> My question is also are these features (NCQ/TCQ) and the heigher 
>>>>> datarate be supported by this
>>>>> modification? or is only the basic feature set of sata 150 TX4 supported?
>>>>>           
>>>> NCQ support is under development. Search the archives for Jens Axboe's 
>>>> recent patches to support this. I don't know about TCQ.
>>>>         
>>> It's done for ahci, because we have documentation. I have no intention
>>> on working on NCQ for chipset where full documentation is not available.
>>> But the bulk of the code is the libata core support, adding NCQ support
>>> to a sata_* driver should now be fairly trivial (with docs).
>>>       
>> I have docs for the Promise NCQ stuff.  Once NCQ is fully fleshed out (I 
>> haven't wrapped my brain around it in a couple weeks), it shouldn't be 
>> difficult to add NCQ support to sata_promise.
>>     
>
> Excellent!
>
>   
Sounds great. If you have implemented NCQ  for sata_promise it would be 
nice if you
could forward me the patch, because i'm not  subscribed to the ML's

best regards
Otto Meier