linux-kbuild.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Masahiro Yamada <yamada.masahiro@socionext.com>
To: Jonas Gorski <jonas.gorski@gmail.com>
Cc: Linux Kbuild mailing list <linux-kbuild@vger.kernel.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Michal Marek <michal.lkml@markovi.net>,
	Sam Ravnborg <sam@ravnborg.org>
Subject: Re: [PATCH] kbuild: allow making undefined symbols fatal for external modules
Date: Tue, 10 Oct 2017 01:25:38 +0900	[thread overview]
Message-ID: <CAK7LNASvdFT8uttzNmiEUVk6rc=M0RC4Z=iOd9XjNnmqaYv7bw@mail.gmail.com> (raw)
In-Reply-To: <20171002105004.32426-1-jonas.gorski@gmail.com>

2017-10-02 19:50 GMT+09:00 Jonas Gorski <jonas.gorski@gmail.com>:
> By passing appropriate values in KBUILD_EXTRA_SYMBOLS it is possible to
> make modpost be able to resolve all symbols for external modules, even
> between to other external modules.
>
> Because of this, it might be desirable to make unresolved symbols an
> error on external module builds as well, to catch missing exports there
> as well.
>
> So add a new flag KBUILD_MODPOST_ERROR which can be set to enable this
> behaviour. It has no effect on non external module builds.
>
> Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
> ---
> The main target is to allow distributions (like LEDE) that make use of
> kernel backports and other out of tree drivers to be able to catch
> errors like missing EXPORT_SYMBOLs in the kernel early.
>
>  scripts/Makefile.modpost | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/scripts/Makefile.modpost b/scripts/Makefile.modpost
> index 16923ba4b5b1..29f787bf8c3a 100644
> --- a/scripts/Makefile.modpost
> +++ b/scripts/Makefile.modpost
> @@ -34,6 +34,8 @@
>
>  # KBUILD_MODPOST_WARN can be set to avoid error out in case of undefined
>  # symbols in the final module linking stage
> +# KBUILD_MODPOST_ERROR can be set to error out in case of undefined
> +# symbols in the final module linking stage for external modules
>  # KBUILD_MODPOST_NOFINAL can be set to skip the final link of modules.
>  # This is solely useful to speed up test compiles
>  PHONY := _modpost
> @@ -49,6 +51,8 @@ ifneq ($(KBUILD_EXTMOD),)
>  obj := $(KBUILD_EXTMOD)
>  src := $(obj)
>
> +KBUILD_MODPOST_WARN = $(if $(KBUILD_MODPOST_ERROR),,1)
> +
>  # Include the module's Makefile to find KBUILD_EXTRA_SYMBOLS
>  include $(if $(wildcard $(KBUILD_EXTMOD)/Kbuild), \
>               $(KBUILD_EXTMOD)/Kbuild, $(KBUILD_EXTMOD)/Makefile)
> @@ -78,7 +82,7 @@ modpost = scripts/mod/modpost                    \
>   $(if $(KBUILD_EXTMOD),-o $(modulesymfile))      \
>   $(if $(CONFIG_DEBUG_SECTION_MISMATCH),,-S)      \
>   $(if $(CONFIG_SECTION_MISMATCH_WARN_ONLY),,-E)  \
> - $(if $(KBUILD_EXTMOD)$(KBUILD_MODPOST_WARN),-w)
> + $(if $(KBUILD_MODPOST_WARN),-w)
>
>  MODPOST_OPT=$(subst -i,-n,$(filter -i,$(MAKEFLAGS)))
>


The motivation looks reasonable to me,
but having both KBUILD_MODPOST_WARN and KBUILD_MODPOST_ERROR
is a bit confusing, I think.

(Some people may wonder what would happen if both are set...)


So, here is my suggestion.


We have only KBUILD_MODPOST_WARN.

The default value is 1 for external module building.
                     0 for kernel building.


If you want to change the default behavior,
you can specify KBUILD_MODPOST_WARN=1 when building kernel,
or KBUILD_MODPOST_WARN=0 when building external module
from the command line.


Also, you need to update Documentation/kbuild/kbuild.txt





-- 
Best Regards
Masahiro Yamada

  reply	other threads:[~2017-10-09 16:26 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-02 10:50 [PATCH] kbuild: allow making undefined symbols fatal for external modules Jonas Gorski
2017-10-09 16:25 ` Masahiro Yamada [this message]
2017-10-10  9:04   ` Jonas Gorski

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAK7LNASvdFT8uttzNmiEUVk6rc=M0RC4Z=iOd9XjNnmqaYv7bw@mail.gmail.com' \
    --to=yamada.masahiro@socionext.com \
    --cc=jonas.gorski@gmail.com \
    --cc=linux-kbuild@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=michal.lkml@markovi.net \
    --cc=sam@ravnborg.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).