home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.aix
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!cs.utexas.edu!geraldo.cc.utexas.edu!portal.austin.ibm.com!awdprime.austin.ibm.com!swan
- From: swan@austin.ibm.com (Randal C. Swanberg)
- Subject: Re: Adding SCSI OSDISK
- Originator: swan@swan2.austin.ibm.com
- Sender: news@austin.ibm.com (News id)
- Message-ID: <C0JIv9.Gx9@austin.ibm.com>
- Date: Fri, 8 Jan 1993 14:57:09 GMT
- References: <1993Jan6.105119.1266@yvax.byu.edu> <19930107.062511.795@almaden.ibm.com> <C0I05H.1A5I@austin.ibm.com> <uk19il4@zuni.esd.sgi.com>
- Organization: IBM
- Lines: 42
-
-
- In article <uk19il4@zuni.esd.sgi.com>, olson@anchor.esd.sgi.com (Dave Olson) writes:
- > In <C0I05H.1A5I@austin.ibm.com> swan@austin.ibm.com (Randal C. Swanberg) writes:
- >
- >
- > | In article <19930107.062511.795@almaden.ibm.com>, russotto@vnet.ibm.com (Matthew T. Russotto) writes:
- > | >
- > | > Many disks have a jumper which set whether they spin-up on power-on
- > | > or only when they receive a SCSI start. Would an request to spin up
- > | > when the disk was already spun up give that result code?
- > | >
- > | > Some disks always spin up on power up, and probably wouldn't like SCSI
- > | > start at all... will such disks work with AIX?
- > |
- > | All of the SCSI disks that we have experience with (i.e. IBM Japan 160MB, IBM
- > | Havant 857MB, IBM Rochester 320MB, 400MB, 1 GB, MAXTOR 355MB, 670MB, Seagate 1GB,
- > | ...and I'm sure there are others) basically no-op a SCSI Start-Unit if the disk
- > | is already spun-up.
- > |
- > | If a disk will not allow a Start-Unit when already spun-up, then the answer
- > | is NO, it won't work with AIX. The SCSI disk driver issues a Start-Unit command
- > | on the first open() to the disk. Basically I guess the assumption that the AIX
- > | SCSI Driver makes is that a)the disk may be spun down and b)if it isn't the
- > | Start-Unit command will be a no-op, not an error.
- >
- > But why fail at that point? The obvious thing to do is to follow
- > a startunit (failed or successful!) with a testunitready, and only
- > cause a failure if the testunitready says the drive isn't ready...
-
- I agree, that sounds like a more robust method. Actually, you could issue the
- testunitready first to see if the startunit was even needed.
-
- You could also ask the same question of the disk implementation: Why fail at that
- point? The obvious thing to do is to see that you are already spun-up and act as
- though you successfully completed the startunit.
-
-
- --
- _____________________________________________________________________________
-
- Randy Swanberg swan@austin.ibm.com
- AIX Kernel Development 512-838-3363
-