This is the mail archive of the ecos-devel@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: problem with gcc3.2 compiler


>>>>> "Gary" == Gary Thomas <gary@mlbassoc.com> writes:

    Gary> On Fri, 2003-10-24 at 06:04, Mark Salter wrote:
    >> >>>>> Nick Garnett writes:
    >> 
    >> > Ottawa Guy <ottawaguy81@yahoo.com> writes: >> Now in gcc3.2
    >> that option "-finit-priority" doesnt >> exist any more and I
    >> didnt find any replacement yet.
    >> 
    >> > This is not true. The recommended version of GCC to use with
    >> eCos is > 3.2.1. This certainly supports -finit-priority, I use
    >> it every day.
    >> 
    >> I recently learned that -finit-priority actually went away some
    >> time ago. The compiler ignores it entirely, because the
    >> init_priority attribute is now a standard gcc extension. In
    >> fact, a recent tightening of option handling in gcc causes very
    >> recent versions of the compiler to -finit-priority as an error.

    Gary> That's really too bad - we can't take it out globally
    Gary> because that could possibly break things for folks using
    Gary> older compilers and we may have troubles using newer ones..

    Gary> The proverbial rock and hard place :-(

I don't actually understand why -finit-priority should go away. AFAIK
it is only supported on ELF-based targets. It would be useful if
somebody trying to use COFF or PE got a compiler error because
-finit-priority was specified but cannot be supported for the target
architecture.

Question for Mark - do you happen to know if -finit-priority was
already the default by gcc 2.95.2? Alternatively, does anybody still
have a vintage compiler of that era around and can run some tests? I
think 2.95.2 is the oldest version of the compiler we care about. If
-finit-priority was already ignored by then we could just extend the
pkgconf/rules.mak hack to remove it completely. That would be
preferable to trying to figuring out the compiler version at
configure-time or build-time.

Bart


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]