On Sun, Jan 13, 2008 at 05:45:44PM -0500, Zack Weinberg wrote:

> It may be that it's all botan's fault, but I don't immediately see how
> botan could break the cset_invalid_csets unit test.

Probably is. :)

IIRC there was one file in Botan that was miscompiled on Alpha with
GCC 2.95 with any -O (I never bothered tracing it down to the
particular option, given how ancient that release is/was).  And
unfortunately I cannot remember which file it was (why yes, I *do*
need to keep better working notes!)

GCC 3.0.1 (on Debian/Alpha) was fine (though produced very slow code
compared to Compaq's compiler). Due to lack of access I have not been
able to check Botan on any more recent GCC on Alpha, and I don't know
of anyone else who has either. I know Debian is including 1.6.3 in
testing and unstable on all archs, but I don't know if their build
process runs the test suite.

-Jack


_______________________________________________
Monotone-devel mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/monotone-devel

Reply via email to