[Bug libstdc++/45841] [4.6 Regression]: r164529 cris-elf libstdc++ 27_io/basic_filebuf/seekoff/char/2-io.cc

["paolo.carlini at oracle dot com" <gcc-bugzilla at gcc dot gnu dot org>]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45841

Paolo Carlini <paolo.carlini at oracle dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|paolo.carlini at oracle dot |
                   |com                         |

--- Comment #3 from Paolo Carlini <paolo.carlini at oracle dot com> 2010-09-30 09:10:09 UTC ---
Please David and Hans-Peter, collaborate a bit on this, I can imagine David
needs details because the problem can't be reproduced on many widespread
targets, like Darwin or Linux. Indeed, the filebuf code has been changed, but
only in the generic parts, thus I would really try to sort out first possible
issues like code generation (what happens if the library is built -O0 instead
of -O2?), undefined behavior (eg, uninitialized data) in the testcase itself,
hidden dependency on target details in the testcase (eg, BUFSIZ), etc, because
it seems highly unlikely to me that the generic code per se is incorrect,
basing on this specific fail on this specific target only. By the way, is the
first time one of the filebuf testcases fails unexpectedly on cris-axis-elf? I
remember something puzzling in the past...