1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 <2023> 2024 2025 | Index | 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 <2023> 2024 2025 |
<== Date ==> | <== Thread ==> |
---|
Subject: | Re: EPICS 3.15.9 build system weird behaviour upon missing g++ |
From: | "Dmitry Yu. Bolkhovityanov via Tech-talk" <tech-talk at aps.anl.gov> |
To: | "Johnson, Andrew N." <anj at anl.gov> |
Cc: | tech-talk at aps.anl.gov |
Date: | Thu, 27 Apr 2023 22:33:11 +0700 (+07) |
Andrew, thanks for your reply! It does shed some light on what's going on, but not everything is clear. On Wed, 26 Apr 2023, Johnson, Andrew N. wrote:
If you look carefully at the compiler commands there that fail they include flags for generating dependency files (-M?), they are not actually compiling the code yet. I don?t have the rules in front of me right now but our build rules evidently don?t want to fail a build if the compiler isn?t able to generate some of the .d files, for various reasons. Those .d files are actually Makefiles and get included so are created in the first pass of the build. GNU-make finds that it has rules to rebuild makefiles so it runs them before attempting to run the actual compile, and that?s where the build actually fails. This is fairly complex, but there is documentation about what GNU make does which is worth reading if you really want to understand it.
Unfortunately, the build system ignores errors from not only g++ runs for depencency generation (those with -MM -MF), but ALL of them, including .cc to .o compilation (those with "-c").
It is visible in my build logs excerpts(https://epics.anl.gov/tech-talk/2023/msg00595.php): the last command, before freeList.html, is *compilation* of epicsTempFile.cpp. But instead of immediately stopping, the build process continues and stops a bit later because of absent epicsTempFile.o, which should have been made by g++.
Is there any document describing how EPICS build system is configured to ignore *some* errors (curiously, without a use of ".IGNORE:", "--ignore" or "-" before commands)?
P.S. I'm trying to investigate this problem not because of "missing g++" (which is obviously a non-issue), but to understand how the EPICS build system works. Which, in turn, is needed to solve another class of problems: when something goes wrong during build, e.g. "genApps" SIGSEGVs on some architectures (due to a buggy gddUtils.h), the error isn't shown "as is" immediately but the build process continues and only the "induced" error due to a missing file is shown later.