From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Petazzoni Date: Wed, 9 May 2018 09:22:54 +0200 Subject: [Buildroot] [PATCH] Revert "zeromq: needs NPTL" In-Reply-To: <20180508173631.GP31635@waldemar-brodkorb.de> References: <20180505160530.GA1774@waldemar-brodkorb.de> <20180508153420.0c334f27@windsurf.home> <20180508173631.GP31635@waldemar-brodkorb.de> Message-ID: <20180509092254.24a64713@windsurf.home> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Hello, On Tue, 8 May 2018 19:36:31 +0200, Waldemar Brodkorb wrote: > > Thanks Waldemar for fixing this issue. Some questions below, though. > > > > On Sat, 5 May 2018 18:05:31 +0200, Waldemar Brodkorb wrote: > > > This reverts commit 1e2a8d4111f57e79c1848c7a70c6501e2bdacd58. > > > > > > Latest uClibc-ng release supports the missing functions even for > > > > What is "Latest" ? 1.0.30 ? > > Yes, 1.0.30. Or do you know of any newer release? ;) No, but the problem with saying "Latest uClibc-ng release" in a commit log is that it's not going to be very informative 6 months or 2 years from now, when someone reads the commit log :-) Saying "Since uClibc-ng 1.0.30, the missing functions are supported" makes it very clear, and will remain useful in the future. > > In fact, it's not even in an uClibc-ng release I believe, it's fixed by > > 0002-librt-declare-clock_nanosleep-independent-of-thread-.patch, which > > will be part of 1.0.31. Is this correct ? > > No. ZeroMQ introduced the usage of pthread_condattr_{s,g}etclock(), > which was added here 04a676f3c8d2443499f27612f69ee88e12089e61 for > Linuxthreads. > It compiled very well without the hotfix for clock_nanosleep > declaration added via > 0002-librt-declare-clock_nanosleep-independent-of-thread-.patch. OK, thanks for the explanation! Thomas -- Thomas Petazzoni, CTO, Bootlin (formerly Free Electrons) Embedded Linux and Kernel engineering https://bootlin.com