1. This forum section is a read-only archive which contains old newsgroup posts. If you wish to post a query, please do so in one of our main forum sections (here). This way you will get a faster, better response from the members on Motherboard Point.

U5 and U60: PCI slot weirdness

Discussion in 'Sun Hardware' started by Gert Doering, Jan 7, 2009.

  1. Gert Doering

    Gert Doering Guest

    Hi,

    (there are some NetBSD parts in this problem, but I don't think they
    are actually relevant - but still included for completeness)

    I'm mystified... I'm trying to get SATA to my Ultra5, and for that purpose
    got a number of "cheap off-the shelf" PCI-to-SATA controllers, to see which
    one works with NetBSD/Sparc64.

    The one that looks most promising is SATAlink based, and it works perfectly
    well (!!) in my Ultra60, with NetBSD-5.0:

    satalink0 at pci0 dev 5 function 0
    satalink0: Silicon Image SATALink 3512 (rev. 0x01)
    satalink0: SATALink BA5 register space disabled
    satalink0: bus-master DMA support present
    satalink0: primary channel wired to native-PCI mode
    satalink0: using ivec 1c for native-PCI interrupt
    atabus0 at satalink0 channel 0
    satalink0: secondary channel wired to native-PCI mode
    atabus1 at satalink0 channel 1

    (no devices attached right now, due to the Ultra60 not being suited very
    well to power non-SCA drives - but I've tested it with a WD 750G drive, and
    everything behaved quite normally).


    Now the funny part. The "target" machine for this controller is an Ultra5,
    which direly needs a larger and faster hard disk. PCI being PCI, I assumed
    it would "just work". But it doesn't.

    - OBP does not recognize the card ("setenv diag-switch? true" shows the
    PCI slot in question as "empty" - the slot is OK, though, as a PCI
    RTL8019 network card is recognized just fine).

    - the kernel does not give any indication of the card being found

    - "pcictl /dev/pci[012] list" does not display anything either.


    Looking at Wikipedia and at my machines, both machines have 5V PCI slots
    (the U60 has 3.3V and 5V slots, but the SATA card is in one of the 5V
    slots - and of course the U60 has 64bit, but that shouldn't matter for a
    32 bit-only card).

    The SATA card itself is coded as "universal 3.3V and 5V card".

    Maybe there is some other difference, like "PCI 2.1" vs. "PCI 2.2" or
    something? Should it matter? How can I find out?

    confused regards,

    gert


    PS: for reference, the "pcictl dump" output for the SATAlink card (moved
    back into the U60) is:

    PCI configuration registers:
    Common header:
    0x00: 0x35121095 0x02b00007 0x01040001 0x00000038

    Vendor Name: CMD Technology (0x1095)
    Device Name: SiI3512 SATALink (0x3512)
    Command register: 0x0007
    I/O space accesses: on
    Memory space accesses: on
    Bus mastering: on
    Special cycles: off
    MWI transactions: off
    Palette snooping: off
    Parity error checking: off
    Address/data stepping: off
    System error (SERR): off
    Fast back-to-back transactions: off
    Interrupt disable: off
    Status register: 0x02b0
    Capability List support: on
    66 MHz capable: on
    User Definable Features (UDF) support: off
    Fast back-to-back capable: on
    Data parity error detected: off
    DEVSEL timing: medium (0x1)
    Slave signaled Target Abort: off
    Master received Target Abort: off
    Master received Master Abort: off
    Asserted System Error (SERR): off
    Parity error detected: off
    Class Name: mass storage (0x01)
    Subclass Name: RAID (0x04)
    Interface: 0x00
    Revision ID: 0x01
    BIST: 0x00
    Header Type: 0x00 (0x00)
    Latency Timer: 0x00
    Cache Line Size: 0x38

    Type 0 ("normal" device) header:
    0x10: 0x00000c01 0x00000c09 0x00000c11 0x00000c19
    0x20: 0x00000c21 0x00018000 0x00000000 0x65121095
    0x30: 0x00080000 0x00000060 0x00000000 0x00000100

    Base address register at 0x10
    type: i/o
    base: 0x00000c00, not sized
    Base address register at 0x14
    type: i/o
    base: 0x00000c08, not sized
    Base address register at 0x18
    type: i/o
    base: 0x00000c10, not sized
    Base address register at 0x1c
    type: i/o
    base: 0x00000c18, not sized
    Base address register at 0x20
    type: i/o
    base: 0x00000c20, not sized
    Base address register at 0x24
    type: 32-bit nonprefetchable memory
    base: 0x00018000, not sized
    Cardbus CIS Pointer: 0x00000000
    Subsystem vendor ID: 0x1095
    Subsystem ID: 0x6512
    Expansion ROM Base Address: 0x00080000
    Capability list pointer: 0x60
    Reserved @ 0x38: 0x00000000
    Maximum Latency: 0x00
    Minimum Grant: 0x00
    Interrupt pin: 0x01 (pin A)
    Interrupt line: 0x00

    Capability register at 0x60
    type: 0x01 (Power Management, rev. 1.0)

    Device-dependent header:
    0x40: 0x00000002 0xa0102204 0x00000000 0x00000000
    0x50: 0x00000000 0x00000000 0x00000000 0x00000000
    0x60: 0x06220001 0x64004000 0x00000000 0x00000000
    0x70: 0x00000000 0x00000000 0x00000000 0x00000000
    0x80: 0x00000022 0x00000022 0x00000000 0xbafebde7
    0x90: 0x0c000000 0x80003512 0x18000000 0x00000000
    0xa0: 0x65150101 0x62dd62dd 0x43924392 0x40094009
    0xb0: 0x65150101 0x62dd62dd 0x43924392 0x40094009
    0xc0: 0x00000184 0x00000000 0x00000000 0x00000000
    0xd0: 0x00000000 0x00000000 0x00000000 0x00000000
    0xe0: 0x00000000 0x00000000 0x00000000 0x00000000
    0xf0: 0x00000000 0x00000000 0x00000000 0x00000000


    --
    Yield to temptation ... it may not pass your way again! -- Lazarus Long
    //www.muc.de/~gert
    Gert Doering - Munich, Germany
    fax: +49-89-3243328 -muenchen.de
    Gert Doering, Jan 7, 2009
    #1
    1. Advertising

  2. Gert Doering

    DoN. Nichols Guest

    On 2009-01-07, Gert Doering <> wrote:
    > Hi,
    >
    > (there are some NetBSD parts in this problem, but I don't think they
    > are actually relevant - but still included for completeness)
    >
    > I'm mystified... I'm trying to get SATA to my Ultra5, and for that purpose
    > got a number of "cheap off-the shelf" PCI-to-SATA controllers, to see which
    > one works with NetBSD/Sparc64.
    >
    > The one that looks most promising is SATAlink based, and it works perfectly
    > well (!!) in my Ultra60, with NetBSD-5.0:
    >
    > satalink0 at pci0 dev 5 function 0
    > satalink0: Silicon Image SATALink 3512 (rev. 0x01)
    > satalink0: SATALink BA5 register space disabled
    > satalink0: bus-master DMA support present
    > satalink0: primary channel wired to native-PCI mode
    > satalink0: using ivec 1c for native-PCI interrupt
    > atabus0 at satalink0 channel 0
    > satalink0: secondary channel wired to native-PCI mode
    > atabus1 at satalink0 channel 1
    >
    > (no devices attached right now, due to the Ultra60 not being suited very
    > well to power non-SCA drives - but I've tested it with a WD 750G drive, and
    > everything behaved quite normally).
    >
    >
    > Now the funny part. The "target" machine for this controller is an Ultra5,
    > which direly needs a larger and faster hard disk. PCI being PCI, I assumed
    > it would "just work". But it doesn't.
    >
    > - OBP does not recognize the card ("setenv diag-switch? true" shows the
    > PCI slot in question as "empty" - the slot is OK, though, as a PCI
    > RTL8019 network card is recognized just fine).


    Your OBP version probably is not recent enough to know the SATA
    cards. I'm not surprised that it does not find anything. I believe
    that it was EOL'ed before SATA drives came on the market.

    > - the kernel does not give any indication of the card being found
    >
    > - "pcictl /dev/pci[012] list" does not display anything either.


    Just for the fun of it -- have you tried putting Solaris 10 in
    the system to see what it does?

    I don't have access to NetBSD (any version), and there is no
    "pcictl" utility documented in the OpenBSD systems which I have running.

    Maybe others will come up with something more helpful, this is
    all that I can do.

    Good Luck,
    DoN.

    --
    Email: <> | Voice (all times): (703) 938-4564
    (too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
    --- Black Holes are where God is dividing by zero ---
    DoN. Nichols, Jan 8, 2009
    #2
    1. Advertising

  3. Gert Doering

    Gert Doering Guest

    "DoN. Nichols" <> writes:

    >On 2009-01-07, Gert Doering <> wrote:
    >> Now the funny part. The "target" machine for this controller is an Ultra5,
    >> which direly needs a larger and faster hard disk. PCI being PCI, I assumed
    >> it would "just work". But it doesn't.
    >>
    >> - OBP does not recognize the card ("setenv diag-switch? true" shows the
    >> PCI slot in question as "empty" - the slot is OK, though, as a PCI
    >> RTL8019 network card is recognized just fine).


    > Your OBP version probably is not recent enough to know the SATA
    >cards. I'm not surprised that it does not find anything. I believe
    >that it was EOL'ed before SATA drives came on the market.


    Maybe. But shouldn't it display something like "unknown" then? Or more
    generic device class ("storage")?

    The U60 displays "raid", which is not very precise either :)

    OTOH, the *card* seems to be a bit weird. After posting this, I've tested
    the card in a PC of about the same age as the U5 (P3-600, Intel 440BX
    chipset) and the effect is the same - the card isn't recognized, Linux'
    "lspci" displays nothing at all.


    >> - the kernel does not give any indication of the card being found
    >>
    >> - "pcictl /dev/pci[012] list" does not display anything either.


    > Just for the fun of it -- have you tried putting Solaris 10 in
    >the system to see what it does?


    I've tested Sol10 on the U60 - and it doesn't know how to talk to that
    card at all, so I didn't test Sol10 on the U5 (which would be a somewhat
    painful process, given the slow IDE controllers...)

    gert
    --
    Yield to temptation ... it may not pass your way again! -- Lazarus Long
    //www.muc.de/~gert
    Gert Doering - Munich, Germany
    fax: +49-89-3243328 -muenchen.de
    Gert Doering, Jan 8, 2009
    #3
  4. On 2009-01-08, Gert Doering <> wrote:

    > "DoN. Nichols" <> writes:
    >
    >>On 2009-01-07, Gert Doering <> wrote:
    >>> PCI being PCI, I assumed it would "just work".


    That's what I used to think.

    >>>
    >>> - OBP does not recognize the card ("setenv diag-switch? true" shows the
    >>> PCI slot in question as "empty"

    >
    > OTOH, the *card* seems to be a bit weird. After posting this, I've tested
    > the card in a PC of about the same age as the U5 (P3-600, Intel 440BX
    > chipset) and the effect is the same - the card isn't recognized, Linux'
    > "lspci" displays nothing at all.


    It may be the case that the card in question is PCI 2.2 and
    the motherboard only supports PCI 2.1

    I have a couple of PCI 2.2 cards and they work in some old
    motherboards but not in others (Asus P2B for one).

    I never tried them in my Ultra5 or Ultra60 since I didn't have
    a use for them (a D-Link wireless card and an ATI graphics card)
    in these machines, but I figured they probably wouldn't work.

    >>> - the kernel does not give any indication of the card being found
    >>>
    >>> - "pcictl /dev/pci[012] list" does not display anything either.

    >
    >> Just for the fun of it -- have you tried putting Solaris 10 in
    >>the system to see what it does?

    >
    > I've tested Sol10 on the U60 - and it doesn't know how to talk to that
    > card at all,


    Exactly. It's like it's not there.

    Howard E.
    Ottawa
    Howard Eisenberger, Jan 20, 2009
    #4
  5. Gert Doering

    Gert Doering Guest

    Howard Eisenberger <> writes:

    >> OTOH, the *card* seems to be a bit weird. After posting this, I've tested
    >> the card in a PC of about the same age as the U5 (P3-600, Intel 440BX
    >> chipset) and the effect is the same - the card isn't recognized, Linux'
    >> "lspci" displays nothing at all.


    >It may be the case that the card in question is PCI 2.2 and
    >the motherboard only supports PCI 2.1


    >I have a couple of PCI 2.2 cards and they work in some old
    >motherboards but not in others (Asus P2B for one).


    Mmmh. The fact that it doesn't work in your P2B makes this an even more
    likely theory - the "440BX" board mentioned above is a P2B-LS.

    Is there a way to see whether a card is PCI 2.1 or 2.2? Does it show up
    somewhere where linux "lspci" or its NetBSD friends could display it?

    >I never tried them in my Ultra5 or Ultra60 since I didn't have
    >a use for them (a D-Link wireless card and an ATI graphics card)
    >in these machines, but I figured they probably wouldn't work.


    Since my SATA card works in the U60, I'd bet the cards would work there :)

    (Well, at least the OBP would indicate that "something is there").

    gert
    --
    Yield to temptation ... it may not pass your way again! -- Lazarus Long
    //www.muc.de/~gert
    Gert Doering - Munich, Germany
    fax: +49-89-3243328 -muenchen.de
    Gert Doering, Jan 21, 2009
    #5
  6. On 2009-01-21, Gert Doering <> wrote:

    > Howard Eisenberger <> writes:
    >
    >>It may be the case that the card in question is PCI 2.2 and
    >>the motherboard only supports PCI 2.1

    >
    >>I have a couple of PCI 2.2 cards and they work in some old
    >>motherboards but not in others (Asus P2B for one).

    >
    > Mmmh. The fact that it doesn't work in your P2B makes this an even more
    > likely theory - the "440BX" board mentioned above is a P2B-LS.


    The PCI 2.2 cards I have are not recognized in any of my old Asus
    motherboards, but they do show up in a couple of old Intel boards.

    > Is there a way to see whether a card is PCI 2.1 or 2.2? Does it show up
    > somewhere where linux "lspci" or its NetBSD friends could display it?


    For the PCI 2.2 graphics card that I do have working, I don't see
    anything in lspci that indicates whether it is 2.2 or 2.1. I don't
    see anything on the card, so unless you can find the actual specs,
    I guess you just have to try it.

    >>I never tried them in my Ultra5 or Ultra60 since I didn't have
    >>a use for them (a D-Link wireless card and an ATI graphics card)
    >>in these machines, but I figured they probably wouldn't work.

    >
    > Since my SATA card works in the U60, I'd bet the cards would work there :)
    >
    > (Well, at least the OBP would indicate that "something is there").


    You're probably right.

    Howard
    Howard Eisenberger, Jan 21, 2009
    #6
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. bigman
    Replies:
    1
    Views:
    349
  2. Christof Kreiner

    3rd party GigaBit Ethernet (copper) in U60?

    Christof Kreiner, Sep 27, 2003, in forum: Sun Hardware
    Replies:
    4
    Views:
    409
    Dean Strik
    Oct 1, 2003
  3. Andrew Tyson

    450MHz CPU from U60 to U2

    Andrew Tyson, Feb 9, 2004, in forum: Sun Hardware
    Replies:
    4
    Views:
    237
    Yh Chen
    Feb 10, 2004
  4. Robert Mazur

    U60 vs. E220R and Netra line

    Robert Mazur, Apr 10, 2004, in forum: Sun Hardware
    Replies:
    7
    Views:
    245
    Robert Mazur
    Apr 13, 2004
  5. Andrew Tyson
    Replies:
    34
    Views:
    557
    Juhan Leemet
    Aug 25, 2004
Loading...

Share This Page