Created attachment 163826 [details] Proposed patch The attached patch imports an upstream commit that fixes the build with -pedantic. I've tested it to fix at least devel/ccrtp's build on 9.3, which is currently broken (http://portsmon.freebsd.org/portoverview.py?category=devel&portname=ccrtp): In file included from ccrtp/crypto/gcrypt/gcrypthmac.cpp:23: /usr/local/include/gcrypt.h:509: error: comma at end of enumerator list /usr/local/include/gcrypt.h:1346: error: comma at end of enumerator list Makefile:571: recipe for target 'gcrypthmac.lo' failed
Reporter is committer, assign accordingly. Appears to be a great candidate for quarterly (MFH)
Resetting "merge-quarterly?" flag. Looking at portsmon, the regression seems to be present only in ports-head (libgcrypt 1.6.4): http://portsmon.freebsd.org/portoverview.py?category=devel&portname=ccrtp http://portsmon.freebsd.org/portoverview.py?category=security&portname=libgcrypt
Set - for 'not necessary' that way the request/response is tracked :) Thanks Raphael!
A commit references this bug: Author: rakuco Date: Fri Dec 18 12:25:01 UTC 2015 New revision: 403961 URL: https://svnweb.freebsd.org/changeset/ports/403961 Log: Add upstream commit to fix build errors with -pedantic. This fixes at least devel/ccrtp's build on 9.3, which is currently broken: In file included from ccrtp/crypto/gcrypt/gcrypthmac.cpp:23: /usr/local/include/gcrypt.h:509: error: comma at end of enumerator list /usr/local/include/gcrypt.h:1346: error: comma at end of enumerator list Makefile:571: recipe for target 'gcrypthmac.lo' failed MFH'ing this is not necessary, this bug is only present in libgcrypt 1.6.4. PR: 205000 Approved by: maintainer timeout (15 days) Changes: head/security/libgcrypt/Makefile head/security/libgcrypt/files/patch-git_f5832285