On Wed, Jan 29, 2014 at 3:48 PM, Belcourt, Kenneth <kbelco@sandia.gov> wrote:

I reran clang on develop with -m2, the log and xml file are now reasonably sized, hopefully the results will appear on the web page shortly (Sandia-Linux).

-rw------- 1 kbelco kbelco 35687373 Jan 29 13:45 bjam.log
-rw------- 1 kbelco kbelco 23085286 Jan 29 13:47 Sandia-Linux.xml

IIUC, this is a problem that could hit any compiler if a widely used library happened to kick off a voluminous error message. So should -m# (for some appropriate #) be the default for regression testing?

Perhaps for develop yes, let's constrain it to, what, 1 or 2 kb message size maximum?  But for master (release), hopefully Boost code will be much cleaner and we'd probably want to know if a Boost release is throwing off 100 kb messages when you compile.

Are you planning to make such a change?

I can make the change, 2 kb (-m2) for develop and for master ...?

Thanks for figuring this out!

Sure.