lkml.org 
[lkml]   [1999]   [Nov]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    From
    SubjectRe: disk corruptions on "tuned" disks Was: APM killing low-latency performance on BX mainboard
    Date
    On Thu, 11 Nov 1999, Alan Cox wrote:
    >
    > > It would be nice if we could collect some data about problematic
    > > drives / mainboards.
    > > Alan any know drives ?
    >
    > The problematic stuff is almost all ISA. That makes life hard. With PCI we
    > can probe and pick up problem stuff (RZ1000 etc) The VIA VP1/VP2 are others
    > with interesting issues (our DMA autotune code fails on them)
    >
    > > Is this more a harddisk issue or an mainboard issue ?
    >
    > IDE controller issue.

    thanks.
    >
    > > On Pentium+ boxes, with not too old disks, the problem should be
    > > pratically inexistent.
    >
    > Correct.

    Is this true even for CDROMs in DMA mode
    (DMAed CDROM drives on Pentium+ boxes bugfree or not ?)

    Alan, how do you stand to this IDE CDROM DMA BUG ?
    (pointed out by Dan Hollis)

    >
    > > But such an automatic tuning/check has soon to be implemented into distros,
    > > in order to get the maximum performance out of the HW.
    > > ( Or are you not willing to try out the 5th gear of your Ferrari ? :-) )
    >
    > Distributions have to work with as much hardware as possible.

    Agreed, and that is the reason because you have to be careful
    when doing the tuning.
    But users with non-flawed hardware should be able to use the 5th gear.

    It would be a bit disappointing to say:
    "do not use the CDROM drive while usign your harddisk recording app,
    because it might ruin your realtime performance"

    >It's more a case
    > of ensuring that the car delivered goes and the wheels don't fall off than
    > the hand tuning it. The latter is _very_ hard.

    Of course I agree 100% : the system should be reliabe and THEN efficient.
    >
    > Alan

    Benno.

    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.rutgers.edu
    Please read the FAQ at http://www.tux.org/lkml/

    \
     
     \ /
      Last update: 2005-03-22 13:54    [W:0.022 / U:0.160 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site