Macports, doing only x64 build (haven't even tried Universal or i386 alone, and not planning to).
Sent from my iPad > On Jul 6, 2015, at 10:12, Jeffrey Walton <[email protected]> wrote: > >> On Mon, Jul 6, 2015 at 9:35 AM, Mobile Mouse <[email protected]> wrote: >> FWIW, I've had problems with GCC above 4.8.* on Mac OS X - everything builds >> fine, but the cryptest.exe dies with SEGV. Same with 4.9.*, same with 5.1. >> So I'm building with clang, which doesn't seem to exhibit such problems. > > OK, so it sounds like I need to do more testing. > > Where did you get you GCC for OS X? MacPorts or Brew? Or did you build > it yourself? (I'm suspect of GCC on OS X because Apple abandoned it. > Even when it was being developed, they were not actively supplying > patches upstream). > > I have not been able to duplicate the SEGV on Debian, Ubuntu, Mint, > Oracle or Fedora. And I have i386 and x86_64 test systems set up for > them just for testing compilers. > > I was able to duplicate it on Cygwin. But they don't have a bug > reporter, so I think there's little hope it will be fixed. I think its > easier just to keep them at -O2 until they fix their processes. > >> On the thread subject - I think it's OK. Whoever for whatever reasons needs >> lower optimization level, can accomplish it with vi. :-) > > Its even easier than vi: `make OPTIMIZE = -Ox` will work just fine. > That's the benefit of using GNU Make or any non-anemic Make (Posix > Make is pathetically anemic). > > Maybe breaking out the OPTIMIZE and setting it to -O2 is all we need. > (Though it may not be apparent, this will help with the Crypto++ > engineering process when CI is cut-in) > > Jeff > >> On Jul 6, 2015, at 06:39, Jeffrey Walton <[email protected]> wrote: >> >> If there are no objections, I am going to commit this today. >> >>> On Friday, July 3, 2015 at 10:45:40 AM UTC-4, Jeffrey Walton wrote: >>> >>> And yet another change after more testing... It appears Cygwin's GCC still >>> has problems in 4.9. >>> >>> ... -- -- You received this message because you are subscribed to the "Crypto++ Users" Google Group. To unsubscribe, send an email to [email protected]. More information about Crypto++ and this group is available at http://www.cryptopp.com. --- You received this message because you are subscribed to the Google Groups "Crypto++ Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
