On 7 May 2013 11:27, Torbjorn Granlund <t...@gmplib.org> wrote: > Alexander Graf <ag...@suse.de> writes: > > There's a certain chance that happens, yes. We don't have instruction > test suites for the PPC target. > > There certainly are more bugs. GMP still crashes all over the place.
If you want to more seriously test the PPC instructions it might be worth extending risu to cope with more than just the ARM architecture... https://wiki.linaro.org/PeterMaydell/Risu That would be a moderate chunk of work -- about 300 lines of C code in the client, plus refactoring the generator perl script to separate out the architecture-dependent bits and add ppc support. The advantage is that once you've done it it's very easy to add support for testing a single instruction, provided you have a known-good hardware implementation to be the reference. (one day soon I will have to make it cope with aarch64) -- PMM
