Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20201128004505.27619-1-scott.branden@broadcom.com>
 <CAK7LNATD0J3C_mFrXAju8-WmdCmrPmRFn7Um0yebnfL-_zcu8w@mail.gmail.com> <a5ea40a5-17ef-f037-16b3-177f01db6447@broadcom.com>
In-Reply-To: <a5ea40a5-17ef-f037-16b3-177f01db6447@broadcom.com>
From:   Bruce Ashfield <bruce.ashfield@gmail.com>
Date:   Tue, 1 Dec 2020 14:01:19 -0500
Message-ID: <CADkTA4NTPJrvoFqumNjkU-o-ZTt8iVUzQec+=Qvm4J_59FFQqg@mail.gmail.com>
Subject: Re: [PATCH] menuconfig,mconf-cfg: Allow specification of ncurses location
To:     Scott Branden <scott.branden@broadcom.com>
Cc:     Masahiro Yamada <masahiroy@kernel.org>,
        Linux Kbuild mailing list <linux-kbuild@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Broadcom Kernel Feedback List 
        <bcm-kernel-feedback-list@broadcom.com>,
        OE-core <openembedded-core@lists.openembedded.org>,
        openembeded-devel <openembedded-devel@lists.openembedded.org>,
        Michal Marek <mmarek@suse.cz>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, Dec 1, 2020 at 12:19 PM Scott Branden
<scott.branden@broadcom.com> wrote:
>
> Hi Masahiro,
>
> On 2020-12-01 4:25 a.m., Masahiro Yamada wrote:
> > On Sat, Nov 28, 2020 at 9:45 AM Scott Branden
> > <scott.branden@broadcom.com> wrote:
> >> From: Bruce Ashfield <bruce.ashfield@windriver.com>
> >>
> >> In some cross build environments such as the Yocto Project build
> >> environment it provides an ncurses library that is compiled
> >> differently than the host's version.  This causes display corruption
> >> problems when the host's curses includes are used instead of the
> >> includes from the provided compiler are overridden.  There is a second
> >> case where there is no curses libraries at all on the host system and
> >> menuconfig will just fail entirely.
> >>
> >> The solution is simply to allow an override variable in
> >> check-lxdialog.sh for environments such as the Yocto Project.  Adding
> >> a CROSS_CURSES_LIB and CROSS_CURSES_INC solves the issue and allowing
> >> compiling and linking against the right headers and libraries.
> >>
> >> Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
> >> cc: Michal Marek <mmarek@suse.cz>
> >> cc: linux-kbuild@vger.kernel.org
> >> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
> >> Signed-off-by: Scott Branden <scott.branden@broadcom.com>
> >> ---
> >
> > Some people solve the cross-compiling in Yocto
> > by using pkg-config.
> >
> >
> > For example,
> >
> > commit 067c650c456e758f933aaf87a202f841d34be269
> > Author: Pavel Modilaynen <pavel.modilaynen@axis.com>
> > Date:   Fri Jul 12 13:52:19 2019 +0200
> >
> >     dtc: Use pkg-config to locate libyaml
> >
> >     Using Makefile's wildcard with absolute path to detect
> >     the presence of libyaml results in false-positive
> >     detection when cross-compiling e.g. in yocto environment.
> >
> >
> >
> > mconf-cfg.sh already allows the path flexibility with pkg-config.
> > Why do you want yet another hook?
> I hope the yocto community can provide more details on this patch.
> The yocto environment isolates the build environment from the host tools.
> Running menuconfig with the upstream kernel does not work on the latest yocto without this patch.

Sorry for not commenting on the origin patch, gmail buried it within
some other threads, but luckily this one popped up.

It is true that we've been carrying this for several years to deal with
the fact that the native sysroot is not searched by the pkg-config
called by mconf-cfg.sh (since it is separate from host and target
pkg-config).

As it turns out, in the past few weeks, we have come up with a way
to inject those native sysroot components into pkg-config without
the need for any changes to the scripts.

Scott: if you try again the the latest oe-core, and are still seeing
the problem with the mainline kernel, ping me, and we can see if
the pkg-config fix isn't holding for you, at that point, yes, we may
still need a hook like this to solve the problem.

Cheers,

Bruce


> >>  scripts/kconfig/mconf-cfg.sh | 8 ++++++++
> >>  1 file changed, 8 insertions(+)
> >>  mode change 100755 => 100644 scripts/kconfig/mconf-cfg.sh
> >>
> >> diff --git a/scripts/kconfig/mconf-cfg.sh b/scripts/kconfig/mconf-cfg.sh
> >> old mode 100755
> >> new mode 100644
> >> index aa68ec95620d..32448bc198a5
> >> --- a/scripts/kconfig/mconf-cfg.sh
> >> +++ b/scripts/kconfig/mconf-cfg.sh
> >> @@ -4,6 +4,14 @@
> >>  PKG="ncursesw"
> >>  PKG2="ncurses"
> >>
> >> +if [ "$CROSS_CURSES_LIB" != "" ]; then
> >> +    echo libs=\'$CROSS_CURSES_LIB\'
> >> +    if [ x"$CROSS_CURSES_INC" != x ]; then
> >> +       echo cflags=\'$CROSS_CURSES_INC\'
> >> +    fi
> >> +    exit 0
> >> +fi
> >> +
> >>  if [ -n "$(command -v pkg-config)" ]; then
> >>         if pkg-config --exists $PKG; then
> >>                 echo cflags=\"$(pkg-config --cflags $PKG)\"
> >> --
> >> 2.17.1
> >>
> >
>


-- 
- Thou shalt not follow the NULL pointer, for chaos and madness await
thee at its end
- "Use the force Harry" - Gandalf, Star Trek II