I m working on this driver as of now and I have found out a site which =
says that tsb43ab22 disables the physical dma during bus reset. Here is =
a link=20

http://www.cs.washington.edu/homes/samr7/upl/

Nebody having any suggestions in this regard is welcomed.



-----Original Message-----
From: Perry E.Metzger [mailto:perry@piermont.com]=20
Sent: Tuesday, November 04, 2003 8:10 PM
To: Gerald C. Simmons
Cc: Rishabh Kumar Goel; netbsd-users@NetBSD.org; tech-misc@NetBSD.org
Subject: Re: AnyBody worked with TSB43AB22 FireWire OHCI chip


"Gerald C. Simmons" <simmons@darykon.cet.com> writes:
> This is a Texas Instruments 400Mb Firewire controller.
>
> Was our OHCI driver checked out on the TSB43AB22?
>
> Maybe there is some subtle difference in the way NEC implemented
> IEEE1394a-2000 that is different than the TI chip.
>
> I don't see any PCI Firewire controllers listed in supported H/W on
> the website. OHCI or otherwise.

My experience has been that our Firewire stuff is highly
unreliable. On some machines it works, and on some it totally fails,
and no one appears to be working on fixing it.

Perry


> On Mon, 3 Nov 2003 Rishabh Kumar Goel wrote:
>>=20
>> Hi,
>>=20
>> I have a TSB43AB22 FireWire PCI controller card. I m trying get my =
FireWire=20
>> HDD workable on NetBSD. Though the ohci driver is fine with NEC OHCI=20
>> chip(72873) but it stucks with TSB43AB22 Card.=20
>>=20
>> ________Dmesg________
>> sbpscsi0 at fwnode0fwohci_handler_set: ctx 2, tcode 1, key 0x0, =
0xf0010004,=20
>> 0x0
>>=20
>> scsibus0 at sbpscsi0: 1 target, 1 lun per target
>> Called sbpscsi_scsipi_request
>> Got req_run_xfer
>> xs control: 0x000001c4, timeout: 10000
>> opcode: 0x00
>> 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 =
0x00
>> Alloc'ing more orbs
>> fwohci_handler_set: ctx 2, tcode 5, key 0x0, 0xf0010008, 0x0
>> fwohci_handler_set: ctx 2, tcode 1, key 0x0, 0xf001000c, 0x0
>> fwohci_at_output: tcode 0x1, tlabel 0x37 hlen 16, dlen 8
>>     0002dd10 ffc0ffff f0030000 00080000$
>>     00000000 f0010008$
>> fwohci_at_output: desc 24
>>     02000010 00000000 00000000 00000000 0002dd10 ffc0ffff f0030000 =
00080000
>>     103c0008 00535b30 00000000 00000000
>> fwohci1: intmask=3D0x00000001: ReqTxComplete(0x8052)
>> fwohci1: intmask=3D0x00000030: RQPkt(0x8452) RSPkt(0x9451)
>> fwohci_handler_set: ctx 2, tcode 5, key 0x0, 0xf0010010, 0x0
>> fwohci_handler_set: ctx 2, tcode 5, key 0x0, 0xf0010014, 0x0
>> fwohci_event_thread: intmask 00000031 ReqTxComp RQPkt RSPkt
>> fwohci_at_done: desc 24 (3)
>>     02000010 00000000 00000000 00000000 0002dd10 ffc0ffff f0030000 =
00080000
>>     103c0008 00535b30 00000000 8452c799
>> fwohci_buf_pktget: desc 0, off 0, req 4096, res 4060, len 4, avail 36
>> fwohci_buf_pktget: desc 0, off 4, req 4096, res 4060, len 8, avail 32
>> fwohci_buf_input: tcode=3D0xe, tlabel=3D0x0, hlen=3D12, dlen=3D0
>> fwohci_buf_pktget: desc 0, off 12, req 4096, res 4060, len 4, avail =
24
>> fwohci_phy_input: BusReset: 0x00010000
>> fwohci_buf_pktget: desc 0, off 16, req 4096, res 4060, len 4, avail =
20
>> fwohci_buf_pktget: desc 0, off 20, req 4096, res 4060, len 12, avail =
16
>> fwohci_buf_input: tcode=3D0x5, tlabel=3D0x16, hlen=3D16, dlen=3D0
>> fwohci_buf_pktget: desc 0, off 32, req 4096, res 4060, len 4, avail 4
>>=20
>> sbp2_login_ans: (addr): (0x80053d4c)
>> fwohci_at_output: tcode 0x7, tlabel 0x16 hlen 16, dlen 32
>>     00025970 ffc00020 00000000 00200000$
>>     00000000 00000000 00000000 f001000c 90400000 00000010 00000000 =
f0010004$
>> fwohci_at_output: desc 24
>>     02000010 00000000 00000000 000031d6 00025970 ffc00020 00000000 =
00200000
>>     103c0020 00cb3c00 00000000 00000000
>> fwohci_buf_pktget: desc 0, off 36, req 4096, res 4060, len 4, avail 0
>> fwohci1: intmask=3D0x00000002: RespTxComplete(0x8051)
>> fwohci_buf_input: no input for 2
>> fwohci_buf_pktget: desc 16, off 1068, req 4096, res 3012, len 4, =
avail 16
>> fwohci_buf_pktget: desc 16, off 1072, req 4096, res 3012, len 8, =
avail 12
>> fwohci_buf_input: tcode=3D0x2, tlabel=3D0x37, hlen=3D12, dlen=3D0
>> fwohci_buf_pktget: desc 16, off 1080, req 4096, res 3012, len 4, =
avail 4
>> fwohci_arrs_input: tcode 0x2, from 0xffc0, tlabel 0x37, rcode 0x0, =
hlen 12,=20
>> dle0
>> fwohci_arrs_input: no listner
>> fwohci_buf_pktget: desc 16, off 1084, req 4096, res 3012, len 4, =
avail 0
>> fwohci_buf_input: no input for 3
>> fwohci_event_thread: intmask 00000002 RespTxComp
>> fwohci_at_done: desc 24 (3)
>>     02000010 00000000 00000000 000031d6 00025970 ffc00020 00000000 =
00200000
>>     103c0020 00cb3c00 00000000 84510000
>>=20
>> sbp2_login_ans: (addr): (0x80053d4c)
>>=20
>> sbp2_free_orb:(addr):(0x8005575c)
>> Freeing orb at addr: 0x00000000f0010008 status_rec: 0x0
>> sbpscsi_timeout
>> Called sbp2_abort
>> ________Dmesg________
>>=20
>> I have seen some mails regarding this chip on linux mailing lists, =
but i m=20
>> unable to apply it with netbsd code?
>>=20
>> http://sourceforge.net/mailarchive/message.php?msg_id=3D1458191
>> --=20
>> Regards,
>> Rishabh Kumar Goel
>> Software Engineer
>> SoCrates Software India Pvt. Ltd.
>> Bangalore
>>=20
>> Scanned by SecureSynergy VirusScreen Service.=20
>> For more information log on to : http://www.securesynergyonline.com =
or http://www.securesynergy.com
>>=20
>

--=20
Perry E. Metzger		perry@piermont.com
Scanned by SecureSynergy VirusScreen Service. 
For more information log on to : http://www.securesynergyonline.com or http://www.securesynergy.com