Ryo ONODERA <ryo%tetera.org@localhost> writes: > Hi, > > Greg Troxel <gdt%lexort.com@localhost> writes: > >> "Ryo ONODERA" <ryoon%netbsd.org@localhost> writes: >> >>> Module Name: pkgsrc >>> Committed By: ryoon >>> Date: Mon Jun 6 20:59:40 UTC 2022 >>> >>> Modified Files: >>> pkgsrc/www/firefox: Makefile distinfo mozilla-common.mk >>> pkgsrc/www/firefox/patches: patch-libwebrtc.diff >>> patch-toolkit_modules_subprocess_subprocess__shared__unix.js >>> >>> Log Message: >>> firefox: Update to 101.0 >>> >>> * Under NetBSD/i386 9, rustc consumes all RAM and swap and failed >>> to build this package. >> >> So how is this getting fixed before the branch? >> >> (firefox updates after 6/1 require approval, which is about not breaking >> it no platforms where it worked. I admit to not knowing the previous >> status.) > > I am sorry. I have forgotten June is freeze month. > > I am testing potential build fix for NetBSD/i386 with > https://bugzilla.mozilla.org/show_bug.cgi?id=1644409 now. Thanks - would be great to make this build. This being close to freeze makes the situation harder, but the real issue is that huge things like firefox and rust are becoming non-viable on i386. (I am very unclear on rust and i586/i686 (in terms of cpu target), but it seems reasonable to configure things for 686, if that makes rust/i386 work better on most computers and only leaves out actual i586 hardware. That however should have a separate thread, and in an ideal world would happen post-branch.)
Attachment:
signature.asc
Description: PGP signature