On 11/09/10 09:37, nathan binkert wrote: >> My change seems to work, but in the process of wedging that old a >> version of scons onto my system (no portage support, no download, had to >> get it from svn, had to manually "install" it). It looks like there are >> a few places where >> >> main.Append(CCFLAGS='-pipe') >> >> is incorrectly interpreted as adding an array of flags "-", "p", "i", >> "p", "e", to CCFLAGS. This makes scons choke because it can't compile >> it's test programs to configure itself and gets confused trying to find >> Python.h. This is apparently not a problem with new versions of scons, >> and we inconsistently deal with the problem by sometimes putting []s >> around strings and sometimes not. > Seems like a reasonable patch to me. If it works in the old and the > new, we should probably go for it.
Ok. I noticed one spot I missed in an if that wasn't getting hit, so I'll add that too, make sure it still works, and then push it. There's a reasonable chance I'll still have missed some spots. >> I don't know for sure if this is an issue with my "installation" method, >> aka a big hammer and a callous disregard for how it's -supposed- to >> work, but if nothing else we should be consistent with how we stick []s >> in there. > I doubt it. What was so strange about your installation method? I > haven't had trouble manually installing scons in the past. (I usually > even manually set --prefix to something like $HOME/local/scons-0.98.1, > and either symlink scons or change my path.) > I pulled down the "ARCHIVE" version, extracted the tarball, moved the system installation of the way and put the contents of the tarball where it was. I didn't know how to screw around with the various paths for python to get it to work, and it didn't seem worthwhile for a quick sanity check. Gabe _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
