Re: [MacPorts] #9053: BUG gmp: crash in config.guess due to mfpvr instruction
#9053: BUG gmp: crash in config.guess due to mfpvr instruction ----------------------------------------+----------------------------------- Reporter: vincent-opdarw@… | Owner: gwright@… Type: defect | Status: assigned Priority: Low | Milestone: Port Bugs Component: ports | Version: 1.0 Resolution: | Keywords: Port: gmp | ----------------------------------------+----------------------------------- Comment(by vinc17@…): Replying to [comment:18 mcalhoun@…]:
Replying to [comment:16 vinc17@…]:
GMP's config.guess does something wrong (in addition to the crash) in MacPorts. Could you please clarify what else the config.guess doe wrong (other than the crash)?
Together with the configure.in, it provides a -mcpu option to gcc (though MacPorts seems to override this setting), going against the default target; MPN_PATH (see configure output) is also affected, and I don't know whether MacPorts does something about it. Such a configuration should be done at a higher level, e.g. at the system level (Apple has done some choice in the configuration of the gcc provided by Xcode) or, here, at the MacPorts level (there's no reason why ''only'' GMP would use a different target). Moreover the -force_cpusubtype_ALL option (that is used in the Portfile) is probably contradictory. Disabling GMP's CPU subtype detection is probably the best thing to do. This is a bit like the ABI chosen by GMP, which can be different from the default ABI (chosen by Apple and MacPorts). Concerning this problem (much more critical), the Portfiles already fixes it: {{{ if {![variant_isset universal]} { configure.env ABI=32 } }}} -- Ticket URL: <http://trac.macports.org/ticket/9053#comment:19> MacPorts <http://www.macports.org/> Ports system for Mac OS
participants (1)
-
MacPorts