Re: libata in 2.4.24?

From: Greg Stark (gsstark_at_mit.edu)
Date: 12/02/03

  • Next message: Jeff Garzik: "Re: libata in 2.4.24?" 
    To: Jeff Garzik <jgarzik@pobox.com>
Date:	02 Dec 2003 13:51:17 -0500

    Jeff Garzik <jgarzik@pobox.com> writes:

    > If true, this is an IDE driver bug... assuming the drive itself
    > doesn't lie about FLUSH CACHE results (a few do).

    I don't think the IDE drivers issue FLUSH CACHE after every write on O_SYNC,
    or after fsync calls. The "lying" discussed on the database lists is when a
    normal write is issued, IDE disks report immediate success even before the
    write hits disk. As far as I know from the lists it seems *all* IDE disks
    behave this way unless write caching is disabled.

    This doesn't happen with SCSI disks where multiple requests can be pending so
    there's no urgency to reporting a false success. The request doesn't complete
    until the write hits disk. As a result SCSI disks are reliable for database
    operation and IDE disks aren't unless write caching is disabled.

    > If the drive lies, there isn't a darned thing we can do about it...
    >
    > If the drive lies, there isn't a darned thing the controller can do
    > about it, either ;-)

    Of course not, but the assumption is that they're only lying because of the
    lack of a TCQ interface. If they could have multiple requests pending there
    would be no need to lie any more at all.

    I'm unclear on which of your #2 or #3 will be the solution though. Do either
    or both of them require that writes actually hit disk before the drive reports
    success? Do either of them allow that semantic without destroying concurrent
    performance? 

    -- 
greg
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/
  • Next message: Jeff Garzik: "Re: libata in 2.4.24?"

    Relevant Pages

    • Re: DS10L hardware configuration question
      ... > regarding IDE/ATA/ATAPI on OpenVMS.. ... simple controller-level disk interface for IDE ... > this IDE interface is used for disks ... > any support for 48-bit addressing. ...
      (comp.os.vms)
    • Re: RAID5 performance
      ... >I just finished setting up a RAID5 array of four new Seagate IDE ... How do you have the disks attached? ... So, when an access request is made to one drive on the bus, ... stripe size or multiple of that), ...
      (comp.os.linux.hardware)
    • Re: to scsi or not to scsi
      ... generation SCSI disk and compare it to a modern IDE drive. ... Choosing between SCSI or IDE or SAS or SATA or FC is mostly a question ... Generally speaking, SCSI, SAS and FC disks are Enterprise class disks ...
      (freebsd-questions)
    • Re: Maximum IDE disk size for DS10L ?
      ... I started running DNEWS on it). ... My guess is that the IDE disks are more sensitive to heat, so if you do not have good ventilation on them, or are in a warmer room, you can expect a higher failure rate. ... accessing the IDE devices past the 127 GB limit requires putting the device in UDMA mode. ...
      (comp.os.vms)
    • Re: Removing Unix for linux
      ... you're working on IDE disks... ... I'm quite sure the boot kernel already has ... but common sense tells not to use IDE disks on PWS ... Seems I eventually clicked on "install from net" rather than ...
      (comp.os.linux.alpha)