xfs
[Top] [All Lists]

gcc 2.95.4-prerelease RPMS

To: linux-xfs@xxxxxxxxxxx
Subject: gcc 2.95.4-prerelease RPMS
From: Joseph Fannin <jhf@xxxxxxxxxxxxxx>
Date: Sat, 13 Oct 2001 03:05:51 -0400
Sender: owner-linux-xfs@xxxxxxxxxxx
    Now that I've thrown my two bits in, I'll contribute what I can.

    I've been running an -xfs kernel built with the 2.95.4-prerelease 
compiler in Debian unstable for about a month now with no problems.  Others 
have reported pretty much the same, but the compiler needs further testing 
than we few.

    So I've rolled some 2.95.4-prerelease rpms -- well, actually, I just 
converted them from .debs with alien.  I've tested them on RedHat 7.1 and 
they seem to work fine, both in install and for building an -xfs kernel.  I 
don't see any benefit in making new ones from scratch.

    Please, anyone with the opportunity, build an -xfs kernel with this 
instead of kgcc/egcs/gcc-2.91.66 and use it, then report how things are doing 
in a few weeks.  There's two rpms you'll need -- 
gcc-2.95-2.95.4-1.011006.i386.rpm and cpp-2.95-2.95.4-1.011006.i386.rpm .  
Both are up at http://home.columbus.rr.com/jfannin/gcc/ .  After installing 
it, you invoke the new compiler as gcc-2.95 .

    If this compiler works where 2.95.2 doesn't, it may help track down just 
where the problem lies with building the XFS code.  At the least, it'll 
provide an interesting datapoint (and AMD users will get k6 and/or athlon 
optimizations!)

    It would also be interesting to see if the 2.95.3 release works.

    Anyone who wants to build their own compiler should just pull the source 
from gcc's 2.95 CVS -- this version, specifically, is dated 20010522.

-- 
Joseph Fannin
jhf@xxxxxxxxxxxxxx

"Bull in pure form is rare; there is usually some contamination by data."
    -- William Graves Perry Jr.


<Prev in Thread] Current Thread [Next in Thread>