Re: fortran/893: Preliminary loop exit compiled wrongly

["Tim Prince" <tprince at computer dot org>]

The following reply was made to PR fortran/893; it has been noted by GNATS.

From: "Tim Prince" <tprince@computer.org>
To: "List subscription" <list@pceet030.cern.ch>,
        "Richard Henderson" <rth@redhat.com>
Cc: <toon@moene.indiv.nluug.nl>, <gcc-gnats@gcc.gnu.org>,
        <gcc-bugs@gcc.gnu.org>, "Jakub Jelinek" <jakub@redhat.com>
Subject: Re: fortran/893: Preliminary loop exit compiled wrongly
Date: Wed, 29 Nov 2000 06:23:53 -0800

 It's difficult to compare speeds of different versions; you must search for
 the best combination of options for each version, undo any code
 optimizations which are peculiar to one version, and so on.  I've found the
 current versions more accepting of standard source code arrangments and more
 likely to want -O2 rather than -Os.  Processor specific options
 [-march=pentiumpro [-ffast-math]] are useful in different circumstances in
 different versions. -O3 is more likely to be useful in later versions; I
 believe it has always dependent on the sequence of subroutines in a source
 file.
 
 ----- Original Message -----
 From: "List subscription" <list@pceet030.cern.ch>
 To: "Richard Henderson" <rth@redhat.com>
 Cc: <toon@moene.indiv.nluug.nl>; <gcc-gnats@gcc.gnu.org>;
 <gcc-bugs@gcc.gnu.org>; "Jakub Jelinek" <jakub@redhat.com>
 Sent: Wednesday, November 29, 2000 12:12 AM
 Subject: Re: fortran/893: Preliminary loop exit compiled wrongly
 
 
 > Yes
 >
 > that patch seems to fix that bug and similar ones I found and reported.
 > The patch was issued after I made a report to Bugzilla (the RH bug
 tracking
 > system). They reacted at the speed of light.... many thanks to Jakub
 > Jelinek from RedHat.
 >
 > For Toon: indeed the 7+ miscompiled routines I had seem to be all ok after
 > this patch (the bug was indeed very "popular"), I am still in the process
 > to check carefully whether there are other problems floating around or not
 > for what concerns our codes, I'll let you know. A naive question: is
 > there any reason why the present g77 is significantly slower than the
 > one in egcs-1.1.2 (20-30% slower on all the tests I made up to now, and
 > now the results are consistent so the comparison should be sound)?
 >
 >                             Ciao
 >                         Alfredo Ferrari
 >
 > On Tue, 28 Nov 2000, Richard Henderson wrote:
 >
 > > On Sun, Nov 26, 2000 at 04:30:48PM -0000, toon@moene.indiv.nluug.nl
 wrote:
 > > > The following code:
 > > >
 > > >       DOUBLE PRECISION VALUE(2), TOLD, BK
 > > >       DATA VALUE /0D0, 1D0/
 > > >       DATA TOLD /0D0/
 > > >       DO I=1, 2
 > > >          BK = VALUE(I)
 > > >          IF(BK .GT. TOLD) GOTO 10
 > > >       ENDDO
 > > >       WRITE(*,*)'Error: BK = ', BK
 > > >       CALL ABORT
 > > >  10   CONTINUE
 > > >       WRITE(*,*)'No Error: BK = ', BK
 > > >       END
 > > >
 > > > which is g77 (execute) testsuite item 20001111.f compiles
 > > > wrongly on i?86-linux with any optimisation on.
 > >
 > > See if http://gcc.gnu.org/ml/gcc-patches/2000-11/msg01594.html
 > > helps.  The general form of the bug appears to be identical.
 > >
 > >
 > > r~
 > >
 >
 > --
 >
 >
 +---------------------------------------------------------------------------
 -+
 > |  Alfredo Ferrari                         ||  Tel.: +41.22.767.6119
 |
 > |  C.E.R.N.                                ||  Fax.: +41.22.767.7555
 |
 > |  European Laboratory for Particle Physics||
 |
 > |  SL Division / EET Project               ||  e-mail:
 |
 > |  1211 Geneva 23                          ||     Alfredo.Ferrari@cern.ch
 |
 > |  Switzerland                             ||
 Alfredo.Ferrari@mi.infn.it |
 >
 +---------------------------------------------------------------------------
 -+
 >