On Thu, Feb 23, 2017 at 05:28:39PM +0000, Ken Moffat wrote: > > 1. I've built current svn (sysv) with 4.9.11 headers on two > machines. One was running 4.10-rc7 for the build, and that was all > fine. The other was running 4.10.0, and there the gettext tests > stalled at the start using all cores and I ended up not running > those tests.
Finally had time to look at this, and discovered the same problem on kernel 4.9.0 on the affected machine (AMD A10 Kaveri). It's a gnulib problem, there was a fix in December. Gettext has _two_ copies of the affected file, not yet sure if both need to be fixed or whether only one is implicated. With both fixed, the tests run ok. So now I need to see if both files need to be fixed, and once I've got something which works for me run it a few more times (there have been sporadic hangs in that test over the years!), and then test non-affected machines to make sure they aren't worsened. My real reason for posting now is that I eventually remembered I'd turned off the non-toolchain tests to get my script to continue. So, there is a *possibility* that aomething else using gnulib might also have a similar problem. At the moment I don't have any idea why that one machine is affected. Hopefully I'll have a repeatably working patch for gettext in a day or so, in case anybody sees a similar problem elsewhere - the hang is at make check-TESTS make[4]: Entering directory '/scratch/ken/gettext-0.19.8.1/gettext-runtime/tests' make[5]: Entering directory '/scratch/ken/gettext-0.19.8.1/gettext-runtime/tests' ĸen -- `I shall take my mountains', said Lu-Tze. `The climate will be good for them.' -- Small Gods -- http://lists.linuxfromscratch.org/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
