[U-Boot] [PATCH 31/39] env: Drop _ENTRY

From: Simon Glass <sjg@chromium.org>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH 31/39] env: Drop _ENTRY
Date: Fri, 2 Aug 2019 08:07:03 -0600	[thread overview]
Message-ID: <CAPnjgZ3PZi3CxbDNRvBEm3PdWd_4+f0U6dU1nkS49E4giZj6ww@mail.gmail.com> (raw)
In-Reply-To: <CANr=Z=Z-nYDiTEAq5zJez84S0zu_=SQQbhrvN_U73mmsjnk0_Q@mail.gmail.com>

Hi Joe,

On Wed, 31 Jul 2019 at 16:03, Joe Hershberger <joe.hershberger@ni.com> wrote:
>
> On Wed, Jul 31, 2019 at 4:56 PM Simon Glass <sjg@chromium.org> wrote:
> >
> > Hi,
> >
> > On Wed, 31 Jul 2019 at 15:07, Joe Hershberger <joe.hershberger@ni.com> wrote:
> > >
> > > On Wed, Jul 31, 2019 at 3:57 PM Simon Glass <sjg@chromium.org> wrote:
> > > >
> > > > Hi Joe,
> > > >
> > > > On Tue, 30 Jul 2019 at 15:35, Joe Hershberger <joe.hershberger@ni.com> wrote:
> > > > >
> > > > > On Sun, Jul 28, 2019 at 9:28 AM Simon Glass <sjg@chromium.org> wrote:
> > > > > >
> > > > > > This typedef does not need to be defined in the search.h header since it
> > > > > > is only used in one file (hashtable.c). Remove it from the header and
> > > > > > change it to a struct.
> > > > > >
> > > > > > Signed-off-by: Simon Glass <sjg@chromium.org>
> > > > > > ---
> > > > > >
> > > > > >  include/search.h | 2 +-
> > > > > >  lib/hashtable.c  | 7 ++++---
> > > > > >  2 files changed, 5 insertions(+), 4 deletions(-)
> > > > > >
> > > > > > diff --git a/include/search.h b/include/search.h
> > > > > > index efa8bcbef6..c99648f80b 100644
> > > > > > --- a/include/search.h
> > > > > > +++ b/include/search.h
> > > > > > @@ -42,7 +42,7 @@ struct env_entry {
> > > > > >
> > > > > >  /* Data type for reentrant functions.  */
> > > > > >  struct hsearch_data {
> > > > > > -       struct _ENTRY *table;
> > > > > > +       struct env_entry_node *table;
> > > > >
> > > > > Don't you need an opaque definition of this?
> > > >
> > > > I don't see why. We can just use struct env_entry_node which is opaque
> > > > if the definition is not available.
> > >
> > > I agree, but doesn't it need to be defined? Maybe the misunderstanding
> > > is happening because of the intermediate state of things through out
> > > this series.
> >
> > (yes, I had trouble figuring out how to split this series up so people
> > could actually review it!)
>
> Well, I think you did a fantastic job at it.
>
> >
> > I don't think it needs to be defined separate here, since it is not in
> > a function scope, so mentioning it inside a struct seems to work OK.
>
> Sounds good to me, if it works!
>
> > >
> > > >
> > > > >
> > > > > Also, there is an opaque definition of _ENTRY in this file that needs
> > > > > to go away.
> > > >
> > > > Where is that? I can't see it.
> > >
> > > I'm looking at master... "include/search.h" line 42 of 123
> >
> > OK, I see. That is removed in the previous patch "env: Drop the ENTRY typdef"
>
> Ah, ok... It would be better to group it with this patch since _ENTRY
> shouldn't be related to ENTRY.

Ah yes of course. Fixed. I'll just send a few updated v3 patches.

Regards,
Simon