All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sam Ravnborg <sam@ravnborg.org>
To: Sascha Hauer <s.hauer@pengutronix.de>
Cc: barebox@lists.infradead.org, Oleksij Rempel <o.rempel@pengutronix.de>
Subject: Re: [PATCH v2] treewide: remove CONFIG_DEBUG_INFO from defconfigs
Date: Thu, 20 Jun 2019 22:41:01 +0200	[thread overview]
Message-ID: <20190620204101.GB12106@ravnborg.org> (raw)
In-Reply-To: <20190620142945.mxvvrslfz6vp2wkn@pengutronix.de>

Hi Sascha.

On Thu, Jun 20, 2019 at 04:29:45PM +0200, Sascha Hauer wrote:
> On Thu, Jun 20, 2019 at 10:54:21AM +0200, Sam Ravnborg wrote:
> > Hi Antony
> > 
> > On Thu, Jun 20, 2019 at 11:25:09AM +0300, Antony Pavlov wrote:
> > > The commit b917f7864115a35 ("remove CONFIG_DEBUG_INFO")
> > > has dropped Kconfig DEBUG_INFO option however we
> > > still have very many DEBUG_INFO mentions in defconfig
> > > files. Drop them using sed:
> > > 
> > >   find -iname '*defconfig' -type f -exec \
> > >       sed -i "/CONFIG_DEBUG_INFO=y/d" {} ';'
> > Another approach could be to regenerate all defconfigs.
> > A little scripting around savedefconfig should do it.
> > 
> > Then we would get rid of all obsolete symbols in one go - for all
> > defconfigs.
> > 
> > The generated defconfig would be the one that people would be using
> > anyway, so there should be only a little risk to introduce new issues
> > doing it this way.
> > 
> > Willing to try this approcah - maybe across all architectures?
> 
> I once thought the same and came up with this little thing:
> 
> #!/bin/bash
> 
> for a in arch/*; do
>         arch=$(basename $a)
>         for c in $a/configs/*; do
>                 config=$(basename $c)
> 		export ARCH=$arch
> 		make $config && make savedefconfig && mv defconfig $c
>         done
> done
> 
> Just tested, still works. Maybe I should just run this from time to time
> and commit the result. What do you think?
Me like.
Much better than removing individual CONFIG symbols.
The diff is not scary:

84 files changed, 38 insertions(+), 129 deletions(-)

Maybe part of your release scripts - then you could evaluate
if the changes are worth it?

Or maybe add the script to barebox so anyone could do it, when
they want to get rid of a symbol.
that would be better than sed scripts.
Actually I like the latter idea the best, then we could better
distribute the task and let anyone do it.
I'm a bit occupied, so hope someone else jumps to do it.

	Sam

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

  parent reply	other threads:[~2019-06-20 20:41 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-20  8:25 [PATCH v2] treewide: remove CONFIG_DEBUG_INFO from defconfigs Antony Pavlov
2019-06-20  8:54 ` Sam Ravnborg
2019-06-20 14:29   ` Sascha Hauer
2019-06-20 20:21     ` Antony Pavlov
2019-06-20 20:41     ` Sam Ravnborg [this message]
2019-07-01 12:41     ` Masahiro Yamada

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=20190620204101.GB12106@ravnborg.org \
    --to=sam@ravnborg.org \
    --cc=barebox@lists.infradead.org \
    --cc=o.rempel@pengutronix.de \
    --cc=s.hauer@pengutronix.de \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.