NetBSD Problem Report #52997

From gson@gson.org  Fri Feb  9 21:11:47 2018
Return-Path: <gson@gson.org>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.NetBSD.org", Issuer "mail.NetBSD.org CA" (not verified))
	by mollari.NetBSD.org (Postfix) with ESMTPS id C5E607A1B5
	for <gnats-bugs@gnats.NetBSD.org>; Fri,  9 Feb 2018 21:11:47 +0000 (UTC)
Message-Id: <20180209211140.DDC0A98B852@guava.gson.org>
Date: Fri,  9 Feb 2018 23:11:40 +0200 (EET)
From: gson@gson.org (Andreas Gustafsson)
Reply-To: gson@gson.org (Andreas Gustafsson)
To: gnats-bugs@NetBSD.org
Subject: Build logs from successful builds contain "fatal error"
X-Send-Pr-Version: 3.95

>Number:         52997
>Category:       misc
>Synopsis:       Build logs from successful builds contain "fatal error"
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    misc-bug-people
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Feb 09 21:15:00 +0000 2018
>Closed-Date:    Fri Apr 05 13:40:03 +0000 2019
>Last-Modified:  Fri Apr 05 13:40:03 +0000 2019
>Originator:     Andreas Gustafsson
>Release:        NetBSD-current,source date 2018.02.09.17.14.26
>Organization:

>Environment:
System: NetBSD
Architecture: i386
Machine: i386
>Description:

When searching for the compiler error message in the log output from a
failed build, I found several error messages like the following:

  /tmp/bracket/build/2018.02.09.17.14.26-i386/src/external/cddl/osnet/dist/uts/common/sys/dtrace.h:55:24: fatal error: sys/modctl.h: No such file or directory
  compilation terminated.

However, this turned out not to be the cause of the build failure -
messages like the above also occur in every successful build log.
I don't know if the errors being reported are causing actual harm or
not, but regardless of that, they are confusing and make it harder to
find other errors that actually cause the build to fail.

The amd64 and evbarm-earmv7hf ports are also affected.

>How-To-Repeat:

Build NetBSD-current successfully.  Grep the build log for "fatal error".

>Fix:

>Release-Note:

>Audit-Trail:
From: Valery Ushakov <uwe@stderr.spb.ru>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: misc/52997: Build logs from successful builds contain "fatal
 error"
Date: Sat, 10 Feb 2018 18:16:53 +0300

 This happens when we generate kdump-ioctl.c - which we do four times,
 { kdump, ktruss } x { normal, rescue }.  src/usr.bin/kdump/mkioctls
 runs CC with only DESTDIR includes and we don't install osnet compat
 stuff.

 I'm not sure what's the right thing to do here.  Adding osnet includes
 to the path is quite intrusive as we will run into namespace issues.

 Perhaps the easiest way is to provide fakes for the mkioctl run.  We
 only need to provide empty

   ctf_api.h
   linker.h
   modctl.h
   processor.h

 With these added the CC invocation in mkioctl completes without errors
 and the generated kdump-ioctl.c file also contains the dtrace ioctls.

 -uwe

From: Andreas Gustafsson <gson@gson.org>
To: gnats-bugs@NetBSD.org
Cc: christos@NetBSD.org
Subject: Re: misc/52997: Build logs from successful builds contain "fatal error"
Date: Fri, 5 Apr 2019 16:30:21 +0300

 The "fatal error" messages are gone since this commit:

   2018.06.12.15.40.39 christos src/usr.bin/kdump/Makefile.ioctl-c 1.36
   2018.06.12.15.40.39 christos src/usr.bin/kdump/mkioctls 1.52

 -- 
 Andreas Gustafsson, gson@gson.org

State-Changed-From-To: open->closed
State-Changed-By: gson@NetBSD.org
State-Changed-When: Fri, 05 Apr 2019 13:40:03 +0000
State-Changed-Why:
Fixed by christos almost a year ago, thanks!


>Unformatted:

Home	PR Database Search