On Tue, Jun 8, 2010 at 2:10 PM, Jason Evans <[email protected]> wrote:
> I just re-ran the master branch's libunwind tests on a similar machine that > has a 2.6.27-derived kernel, and only 7/26 tests failed. That looks a lot closer to the more recent kernels that I test with. > > So, my problems are almost certainly due to running an old kernel. Maybe > 2.6.20 is sufficiently ancient that it doesn't make sense to support it in > libunwind. Thoughts? > We could add an autoconf test to check for old kernels and disable HAVE_MINCORE and use msync() instead. The only downside I'm aware of with using msync is the disk traffic that would not be there otherwise. This is somewhat mitigated by MS_ASYNC flag. It also looks like FreeBSD and Linux (they share this code) have minor differences with respect to function prototypes and rounding arguments to the page boundary. I'll go ahead and apply the part of your patch that fixes the page boundary crossing bug on Linux. -Arun _______________________________________________ Libunwind-devel mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/libunwind-devel
