[OE-core] [PATCH 1/2] rng-tools: 5 -> 6.6

[Richard Purdie]

On Fri, 2018-11-09 at 22:50 +0000, richard.purdie at linuxfoundation.org
wrote:
> On Fri, 2018-11-09 at 17:12 +0800, Hongxu Jia wrote:
> > -PACKAGECONFIG = "libgcrypt"
> > +PACKAGECONFIG ??= "libgcrypt libjitterentropy"
> >  PACKAGECONFIG_libc-musl = "libargp"
> >  PACKAGECONFIG[libargp] = "--with-libargp,--without-libargp,argp-
> > standalone,"
> >  PACKAGECONFIG[libgcrypt] = "--with-libgcrypt,--without-
> > libgcrypt,libgcrypt,"
> > +PACKAGECONFIG[libjitterentropy] = "--enable-jitterentropy,
> > --disable-jitterntropy,libjitterentropy"
> > +
> > +# Refer autogen.sh in rng-tools
> > +do_configure_prepend() {
> > +    cp ${S}/README.md ${S}/README
> > +}
> >  
> >  do_install_append() {
> >      # Only install the init script when 'sysvinit' is in
> > DISTRO_FEATURES.
> 
> This fails on musl, likely as the libc-musl override wasn't tweaked.
> It
> also means that the no libjitterentropy case wasn't tested as there
> is
> a typo likely causing this:
> 
> 
https://autobuilder.yoctoproject.org/typhoon/#/builders/11/builds/190/steps/7/logs/step1b

The above is definitely from the typo and musl misconfiguration in this
patch.

> There are also other weird hangs being observed on the autobuilder
> after these changes. I can't prove its this patchset but I am
> suspicious and its something in master-next as things were working
> just
> fine until we started poking around all this again :(
> 
> 
https://autobuilder.yoctoproject.org/typhoon/#/builders/34/builds/182/steps/7/logs/stdio
> 
https://autobuilder.yoctoproject.org/typhoon/#/builders/2/builds/188/steps/7/logs/stdio

Looking further into the above shows:

NOTE:
======================================================================
NOTE: ERROR: tearDownClass (opkg.OpkgRepoTest)
NOTE: ----------------------------------------------------------------------
NOTE: Traceback (most recent call last):
  File "/home/pokybuild/yocto-worker/nightly-x32/build/meta/lib/oeqa/core/case.py", line 36, in _oeTearDownClass
    clss.tearDownClassMethod()
  File "/home/pokybuild/yocto-worker/nightly-x32/build/meta/lib/oeqa/runtime/cases/opkg.py", line 29, in tearDownClass
    cls.repo_server.stop()
  File "/home/pokybuild/yocto-worker/nightly-x32/build/meta/lib/oeqa/utils/httpserver.py", line 36, in stop
    self.process.join()
  File "/usr/lib/python3.6/multiprocessing/process.py", line 124, in join
    res = self._popen.wait(timeout)
  File "/usr/lib/python3.6/multiprocessing/popen_fork.py", line 50, in wait
    return self.poll(os.WNOHANG if timeout == 0.0 else 0)
  File "/usr/lib/python3.6/multiprocessing/popen_fork.py", line 28, in poll
    pid, sts = os.waitpid(self.pid, flag)
  File "/home/pokybuild/yocto-worker/nightly-x32/build/meta/classes/testimage.bbclass", line 166, in sigterm_exception
    raise RuntimeError
RuntimeError

which is probably other patches in -next but raises some questions
about why an error like this would just hang.

Cheers,

Richard