[MacPorts] #64913: vlang fails for PowerPC: -latomic flag needed, Header file <sys/errno.h>, needed for module `sync` was not found

[MacPorts]

#64913: vlang fails for PowerPC: -latomic flag needed, Header file <sys/errno.h>,
needed for module `sync` was not found
---------------------------+-------------------------------------------
  Reporter:  barracuda156  |      Owner:  harens
      Type:  defect        |     Status:  accepted
  Priority:  Normal        |  Milestone:
 Component:  ports         |    Version:  2.7.2
Resolution:                |   Keywords:  powerpc, leopard, snowleopard
      Port:  vlang         |
---------------------------+-------------------------------------------

Comment (by harens):

 Why don't we try the following:

 {{{#!bash
 > sudo port extract vlang
 > # Remove errno header
 > sudo sed -i "" "10d" "$(port work
 vlang)/v-weekly.2022.13/vlib/sync/sync_darwin.c.v"
 > # Remove -no-pie flag
 > sudo sed -i "" "252d" "$(port work
 vlang)/v-weekly.2022.13/vlib/v/builder/cc.v"
 > sudo sed -i "" "91227d" "$(port work vlang)/vc-
 de63146da0a0f93628f8a65e70267e2c8d4c7ce1/v.c"
 > # Remove Objective-C Flag
 > sudo sed -i "" "335d" "$(port work
 vlang)/v-weekly.2022.13/vlib/v/builder/cc.v"
 > sudo sed -i "" "91301d" "$(port work vlang)/vc-
 de63146da0a0f93628f8a65e70267e2c8d4c7ce1/v.c"
 > sudo port -o install vlang
 }}}

 I'm not sure if this Objective-C patch will work. We are also slowly
 removing the more complex functionality provided by the V programming
 language, which isn't ideal. However, I guess there's no point complaining
 if we can't build V in the first place.

 What I'm more curious about is that V had no issues building without any
 of these patches on the 10.6 i386 and x86_64 buildbots. It's surprising
 that this much work is needed for 10.6 ppc and 10.5.8.

-- 
Ticket URL: <https://trac.macports.org/ticket/64913#comment:25>
MacPorts <https://www.macports.org/>
Ports system for macOS