On Tue, 2017-11-14 at 19:29 +0100, Florian Weimer wrote:
So the only thing that's missing is the __set_errno (ENODEV); call, I
think. (It's what the ioctl should fail with for an unknown interface
name.)
Thanks,
Florian
OK, here is a new version of the patch that sets errno to ENODEV. I
tested it (Thanks to Joseph for fixing the tests that would not compile
due to the new GCC warning) and got three failures that I think are all
unrelated to this change:
FAIL: crypt/badsalttest
FAIL: nptl/tst-thread_local1
FAIL: nss/tst-nss-files-hosts-multi
tst-thread_local1 is a failure I have seen before, tst-nss-files-hosts-
multi is a new test, and I am not sure what is happening with
badsalttest but it does not seem to be related to this change. I also
have the not-checked-in __NONSTRING changes in utmp.h in my tree in
order to get this to build.
OK to checkin?