[Patch, Fortran] (4.4/4.5 regression) PR 43551 fix buffered writing with direct access

[Toon Moene]

Tobias Burnus wrote:

> The bug is rather nasty. Due to the bug, the result for phonon
> calculations with Quantum Espresso are wrong - however, the numbers are
> only slightly off; some for instance by "just" 10%. Too small to see the
> error immediately - but still the error is significant.

Phew, we just escaped that one by pure luck: Although we do write (3000 
"integer") unformatted direct access records in our output files, we do 
the writing sequentially (we want direct access files for reading 
selectively).

On the other hand, if this had been wrong throughout, the bug would have 
been caught much earlier, because we do not write floating point data, 
but counted bit streams :-)

-- 
Toon Moene - e-mail: toon@moene.org - phone: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
At home: http://moene.org/~toon/
Progress of GNU Fortran: http://gcc.gnu.org/gcc-4.5/changes.html