On Sun, Oct 13, 2019 at 1:30 PM Chris Jones <jon...@hep.phy.cam.ac.uk> wrote:
> > > On 13 Oct 2019, at 6:20 pm, Jack Howarth <howarth.at.macpo...@gmail.com> > wrote: > > > Shouldn't that also strip out... > > if {${configure.sdkroot} ne ""} { > # We should be using --with-build-sysroot here. Using > --with-sysroot > # changes the behavior of the installed gcc to look in that sysroot > # by default instead of /. Using --with-build-sysroot is supposed > # to be used during the build but not impact the installed product. > # Unfortunately, the build fails because the value doesn't get > # plumbed everywhere it is supposed to. > # > # https://trac.macports.org/ticket/53726 > # https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79885 > configure.args-append --with-sysroot="${configure.sdkroot}" > } > > so that if path to an Xcode-beta.app is selected in xcode-select it is > honored? > Jack > > > > I did not want to touch that at this point. The gcc bug report is still > open, as far as I can see. > > If someone wants try a build using —with-build-sysroot instead, they are > very welcome to try... > > Cheers chris > Chris, I'll try your changes with that hunk removed. I don't see why it shouldn't bootstrap with only that patch applied as the xgcc and final gcc should now all look via xcrun for the SDK location. Jack > > On Sun, Oct 13, 2019 at 1:00 PM Christopher Jones < > jon...@hep.phy.cam.ac.uk> wrote: > >> >> >> https://github.com/macports/macports-ports/commit/1850136d289019f3b29a5b24d3ec8ef9b23913ee >> >> Chris >> >> On 13 Oct 2019, at 5:49 pm, Kenneth F. Cunningham < >> ken.cunningham.web...@gmail.com> wrote: >> >> Chris, >> FYI, this has already been committed upstream. >> https://gcc.gnu.org/ml/gcc-cvs/2019-10/msg00148.html >> >> Jack >> >> >> Yes, this plan is coming along nicely -- some of us know how this came to >> pass :> I sent that link to Josh a while back when it came out, as support >> for passing the SYSROOT in base, so it's been floating around a while. >> >> >> After this, a couple of older gccs maybe, and then that covers most of our >> toolchain. >> >> >> Then we begin stripping all the baked-in -isysroot stuff out of installed >> ports, which is just plain bad. (And prevents, for example, reliably >> building against an alternate SDK, amongst other heartburn). >> >> >> Ken >> >> >>