1. Forum
  2. Usenet
  3. COMP.SYS.IBM.PS2.HARDWARE

  • IBM SCSI SCB Commands

    From Tomas Slavotinek@21:1/5 to All on Sun Jan 8 14:31:58 2023
    https://ardent-tool.com/IBM_SCSI/SCB_Commands.html

    Will have to do more digging to figure out the Read/Write Sector Buffer structure (and yes the commands are in a reversed order compared to the
    other read/write ops - the RD bit confirms this.)

    Not sure what the cmd. 12h is, but it's definitely implemented. The
    ESDI-era Int 13 AH=1Ch iface may reveal more context.

    Will have to czech the refdisk/diag utils too, as these are more likely
    to use undocumented SCSI SCB commands.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From IBMMuseum@21:1/5 to All on Sun Jan 8 10:57:23 2023
    https://ardent-tool.com/IBM_SCSI/SCB_Commands.html

    Will have to do more digging to figure out the Read/Write Sector Buffer structure (and yes the commands are in a reversed order compared to the
    other read/write ops - the RD bit confirms this.)

    Not sure what the cmd. 12h is, but it's definitely implemented. The
    ESDI-era Int 13 AH=1Ch iface may reveal more context.

    Will have to czech the refdisk/diag utils too, as these are more likely
    to use undocumented SCSI SCB commands.

    Color me interested - looks like it docks right into the ESDI and DBA-ESDI level.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Tomas Slavotinek@21:1/5 to IBMMuseum on Sun Jan 8 20:41:27 2023
    On 08.01.2023 19:57, IBMMuseum wrote:
    Color me interested - looks like it docks right into the ESDI and DBA-ESDI level.

    Yea, SCSI SCB was most likely derived from ESDI's command iface.

    IBM SCSI also implements a subset of the poorly documented Int 13 AH=1C
    "ESDI FIXED DISK" functions - specifically subfunctions 0D, 10, 11, 16,
    and 17 (hex). I'll probably start by reversing the ESDI BIOS and then
    compare it with the SCSI implementation.

    http://www.ctyme.com/intr/int-13.htm
    Not great, but better than nothing. Not sure if The Undocumented PC has anything on this.

    https://ibmmuseum.com/Interrupts/INT13h/
    Dave, where did this come from? A20 control is not something I would
    expect to fine among HD functions...

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From IBMMuseum@21:1/5 to All on Sun Jan 8 13:42:08 2023
    https://ibmmuseum.com/Interrupts/INT13h/
    Dave, where did this come from? A20 control is not something I would
    expect to fine among HD functions...

    Oh, that was me copying a base template for a page never populated - Yes, I really need to get into an overhaul of my site on these days off.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Tomas Slavotinek@21:1/5 to IBMMuseum on Sun Jan 8 23:07:08 2023
    On 08.01.2023 22:42, IBMMuseum wrote:
    Oh, that was me copying a base template for a page never populated - Yes, I really need to get into an overhaul of my site on these days off.

    I have just emailed you some details regarding the index page "issue".

    Other than that, most of it is hidden, so it's not really a problem. I
    don't think the Int 13h page is linked from anywhere anyway (apart from
    my previous post... oops).

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Louis Ohland@21:1/5 to Tomas Slavotinek on Mon Jan 9 08:08:33 2023
    https://ardent-tool.com/IBM_SCSI/SCB_Commands.html#Cmd_1F

    "This SCB command is used to send any SCSI command (SCSI CDB) not
    supported by the adapter directly to a SCSI device."

    "This command should be used only when other commands cannot perform the operation; "

    Leave it up to IBM to write something nicht klar. Again, for those
    immersed in IBM denkt, this sez it all. But to mortals such as me, it
    leaves out one important fact. If you wandt to speak to the device
    alone, this is it.

    I am blissfully unaware if you can sendt a command to an adapter, or use
    SOS to sendt that same command to a device. Are there any commands that
    can be sendt to the adapter, or by using SOS, sendt directly to the device?

    https://ardent-tool.com/IBM_SCSI/SCB_Commands.html#Cmd_1A

    " Any SCB issued requesting an I/O to the DASD at an address over the
    set limit will not be processed by the SCSI adapter."

    Again, konfuzed and kontuzed. Is that you, Helga?

    Because I wallowed about in the patent, and related pages in the Tool, I
    know enough about the function.

    "Requesting an I/O"... then "address over the set limit" address of
    what? I/O address, page address, or perhaps something related to a hard
    drive?

    Tomas Slavotinek wrote:
    https://ardent-tool.com/IBM_SCSI/SCB_Commands.html

    Will have to do more digging to figure out the Read/Write Sector Buffer structure (and yes the commands are in a reversed order compared to the
    other read/wrte ops - the RD bit confirms this.)

    Not sure what the cmd. 12h is, but it's definitely implemented. The
    ESDI-era Int 13 AH=1Ch iface may reveal more context.

    Will have to czech the refdisk/diag utils too, as these are more likely
    to use undocumented SCSI SCB commands.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Tomas Slavotinek@21:1/5 to Louis Ohland on Mon Jan 9 15:43:33 2023
    On 09.01.2023 15:08, Louis Ohland wrote:
    Leave it up to IBM to write something nicht klar. Again, for those
    immersed in IBM denkt, this sez it all. But to mortals such as me, it
    leaves out one important fact. If you wandt to speak to the device
    alone, this is it.

    Because I wallowed about in the patent, and related pages in the Tool, I
    know enough about the function.

    Oh you dingle arm! You were too lazy to eliminate the IBM speech, and so
    was I...

    (https://youtu.be/Ac7G7xOG2Ag)

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)