(260087) log of changes to /usr/ports/UPDATING (was: Unable to build Python38 ports like devel/py38-pycparser and even py38-setuptools)

On 28/11/2021 02:52, Kevin Oberman wrote: … like /usr/ports/UPDATING had been abandoned. The last update was over a year ago. … Either Kevin's ports tree is more than a year out of date, or (more likely) we're disoriented by

Re: Unable to build Python38 ports like devel/py38-pycparser and even py38-setuptools

On Sat, Nov 27, 2021 at 12:04 PM Dr. Peter Voigt wrote: > I am currently facing a strange problem when tryping to upgrade > following ports: > > # pkg version -vIL= > py38-pycparser-2.20< needs updating (index has 2.21) > py38-pyparsing-3.0.4 < needs updating (in

Re: armv7 main [so: 14] targeting via poudriere bulk builds: iperf3 fails to build: iperf_api.c:4460:5: error: unknown type name 'cpu_set_t'; did you mean 'cpuset_t'?

On 2021-Nov-27, at 06:04, Mark Millard wrote: > Notably amd64 targeting did not have this problem. I've not > tried aarch64 targeting yet. I've now built for aarch64 based on the same /usr/ports as used for the other builds and iperf3 built fione. I do not know about 32-bit or 64-bit powerpc bu

Unable to build Python38 ports like devel/py38-pycparser and even py38-setuptools

I am currently facing a strange problem when tryping to upgrade following ports: # pkg version -vIL= py38-pycparser-2.20< needs updating (index has 2.21) py38-pyparsing-3.0.4 < needs updating (index has 3.0.6) py38-snowballstemmer-2.1.0 < need

armv7 main [so: 14] targeting via poudriere bulk builds: iperf3 fails to build: iperf_api.c:4460:5: error: unknown type name 'cpu_set_t'; did you mean 'cpuset_t'?

Notably amd64 targeting did not have this problem. I've not tried aarch64 targeting yet. Still, it looks like iperf3 needs the #define _WITH_CPU_SET_T opt-in someplace: --- iperf_api.lo --- iperf_api.c:4460:5: error: unknown type name 'cpu_set_t'; did you mean 'cpuset_t'? cpu_set_t cpu_set;

amd64 main [so: 14]: recent bulk builds are reporting: pid ??? comm conftest has trashed its stack, killing

I do not remember the output from poudriere builds having reports like the following before: pid 35298 comm conftest has trashed its stack, killing . . . pid 36773 comm conftest has trashed its stack, killing (There are 16 builders active in the bulk run. The output does not make it clear which b