This is the mail archive of the
mailing list for the GCC project.
Re: Gcc certification
Robert Dewar wrote:
Joe Buck wrote:
The famous Pentium bug, which reportedly cost Intel 470 million dollars,
was due to an incorrectly generated table (the table was originally
correct, but the script to embed the table into a programmable logic
had a bug). Five positions in a lookup table used by the microcode for
the division instruction had the wrong value.
And most certainly applications could be certified on that flawed
Sure. I tried runs with (simulated) FDIV bugs of our Numerical Weather
Prediction software at the calculated occurence rate, and it didn't make
a difference, except for those phenomena (convective precipitation) that
are unpredictable anyway ...
Toon Moene - e-mail: email@example.com - phone: +31 346 214290
Saturnushof 14, 3738 XG Maartensdijk, The Netherlands
Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
A maintainer of GNU Fortran 95: http://gcc.gnu.org/fortran/