This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


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

Re: BUG with libio/dbz/rdbz test case on m88k-motorola-sysv3


On Fri, 17 July 1998, 14:05:11, manfred@s-direktnet.de wrote:

 > Running the various test cases in the libio/dbz subdir fails for me on 
 > this particular platform since several weeks. This is with the r1.1
 > branch last updated from CVS at 1998/07/15 08:19:04 +0200:
 > 
 > $ pwd
 > /u/b60/manfred/work/GNU/build/slsvybt/egcs-1.1-delta88/delta88/libio/dbz
 > $ gmake check CFLAGS="-O2 -g"
 > ...
 > : test various perversions of byte ordering
 > awk -f ../../../../egcs-1.1/libio/dbz/revbytes dbase.dir >dbase2.dir
 > ./byteflip `../../../../egcs-1.1/libio/dbz/getmap dbase.dir` `../../../../egcs-1.1/libio/dbz/getmap dbase2.dir` <dbase.pag >dbase2.pag
 > cp dbase dbase2
 > ./rdbz -E 1000 -0 -c dbase2
 > ./rdbz: (file `dbase2', line 1) can't find `<m....VO1.9q.@s95e1zKsIj7LrIwa1>    6000000...'
 > gmake: *** [stamp-r7] Error 1
 > 
 > Reducing the optimization level to 1 lets the tests succeed.
 > 
 > AFAIR, the last snapshot that worked for me was around 19980525.
 > 
 > manfred

Running the test with

  $ gmake check CFLAGS="-O9 -fno-gcse -g"

makes  the error go   away, too!  If this is   actually a GCSE related
problem, does anybody of you GCSE experts know, where to look at?

Thanks,

manfred


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