Bug 242810

Summary: [exp-run] with Clang where pointer-to-bool is narrowing
Product: Ports & Packages Reporter: lichray
Component: Individual Port(s)Assignee: Port Management Team <portmgr>
Status: New ---    
Severity: Affects Only Me CC: vishwin
Priority: --- Flags: lichray: exp-run?
Version: Latest   
Hardware: Any   
OS: Any   
Attachments:
Description Flags
git patch
none
ports patch none

Description lichray 2019-12-22 17:48:25 UTC
I would like to get some help on testing

https://reviews.llvm.org/D64034

on FreeBSD ports (you should be able to patch it on base Clang).  The change
may be a part of the coming C++20, and we want to know its impact on real-world
code.  The change has been ran on Google's code base, and the result is generally
positive.
Comment 1 lichray 2019-12-22 20:46:27 UTC
Created attachment 210160 [details]
git patch
Comment 3 lichray 2019-12-28 03:05:45 UTC
(In reply to Antoine Brodin from comment #2)

Thanks.  Finished analyzing these ports.  Before I digging deeper (by patching a few ports to unblock the skipped ones), I want to ask:

1. How many ports in the ports tree are expected to be built with base clang++?
2. How many ports depends on java/openjfx8-devel, www/webkit2-gtk3, and www/node10?
Comment 4 lichray 2020-01-06 09:05:30 UTC
Created attachment 210477 [details]
ports patch

Uploaded patch to fix the 7 ports in the ports tree.  See if they can unblock the 300 ports that are skipped.  Can @ports exp-run the two patches?  Thanks.
Comment 5 Antoine Brodin freebsd_committer 2020-01-09 21:02:41 UTC
Exp-run looks fine with the ports patch applied.
Comment 6 Charlie Li freebsd_committer 2020-08-24 04:05:27 UTC
With www/webkit2-gtk3 2.28.4 the file patched has been moved to Source/WebCore/style/StyleResolver.cpp; the hunk remains the same. The port builds successfully after making this adjustment.