On 7 October 2013 13:08, Anil Singhar anil.singhar@linaro.org wrote:
Hi Neil,
Thanks for your response. I prepared the mksh test and ran them. There were 5 failures (out of the 436 tests) as follows.
Well done! 4 of those 5 are listed in the output as ignored.
Right now, I don't know if these are failures due to klibc implementation or due to problems with the way I have build mksh itself. To some extent I suspect mksh as well because (i) I could not get mksh_44 (from this link: dget http://ftp.de.debian.org/debian/pool/main/m/mksh/mksh_44-1.dsc , the link seems to be broken)
Since I started on this (and got that link), mksh has been updated. You can get 44-1.dsc from snapshot.debian.org:
http://snapshot.debian.org/archive/debian/20130305T212453Z/pool/main/m/mksh/...
but I used whatever I got ( mksh_44.orig.tar.gz , this may be missing some patches, how to get that.. ).(ii) While building mksh, encountered some compiler errors (e.g. conftest.c) but the building seemed to go fine despite these errors, so I went ahead with testing and ended up with 5 failures out of which mksh ignores 4 and didn't expect 1 as it reports.
So, do you have any idea if I am doing the right thing..?
I think you've done the right thing.
Should I worry about getting mksh build without any compiler error or warning first before attempting to use the tests..?
Compare with the Debian build log for 44-1: https://buildd.debian.org/status/fetch.php?pkg=mksh&arch=amd64&ver=4...
There's a different failure there.
I think this is ready to take to klibc upstream, let them decide on the mksh test results.