Re: [FLASH-BUGS] Running flash on a SUNblade.

From: Andrew Siegel (siegela@flash.uchicago.edu)
Date: Fri Jan 24 2003 - 09:33:33 CST

  • Next message: Sean Matt: "[FLASH-BUGS] more FLASH issues"

    regarding the second point, we have compilers that flag variables which
    are read before being written. I've never seen any warning for these
    variables, and the few I tracked down just now by hand didn't have this
    problem. I'll do a more thorough search in a few minutes.

    Does Sun have a compiler flag that initializes all local variables to zero
    (most compilers do)? Did you compile with it and verify this was the
    problem? If not we can do it -- just wanted to
    avoid duplicating work.

    thanks for the feedback. -andrew siegel

    On Fri, 24 Jan 2003, Jonathan Dursi wrote:

    > Hi:
    >
    > Every time FLASH gets compiled on a new machine (or compiler), we end
    > up finding new `gotcha's. We had seen your number (1) before, and it's
    > fixed in the new release. We hadn't seen (2) or corresponding problems,
    > before; we'll start looking at that.
    >
    > Once you've got everything running succesfully, please do submit your
    > Makefile.h; we'll include it in the distribution so that others using
    > Sun machines don't have to re-do the work you've done.
    >
    > - Jonathan
    >



    This archive was generated by hypermail 2b30 : Fri Jan 24 2003 - 09:33:38 CST