[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200818141940.GA357037@tsnow>
Date: Tue, 18 Aug 2020 17:19:40 +0300
From: Tomer Samara <tomersamara98@...il.com>
To: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc: devel@...verdev.osuosl.org, Todd Kjos <tkjos@...roid.com>,
Suren Baghdasaryan <surenb@...gle.com>,
Riley Andrews <riandrews@...roid.com>,
dri-devel@...ts.freedesktop.org, linux-kernel@...r.kernel.org,
Hridya Valsaraju <hridya@...gle.com>,
Arve Hjønnevåg <arve@...roid.com>,
Joel Fernandes <joel@...lfernandes.org>,
Laura Abbott <labbott@...hat.com>,
Martijn Coenen <maco@...roid.com>,
Sumit Semwal <sumit.semwal@...aro.org>,
Christian Brauner <christian@...uner.io>
Subject: Re: [PATCH v2 1/4] staging: android: Replace BUG_ON with WARN_ON
On Tue, Aug 18, 2020 at 04:11:06PM +0200, Greg Kroah-Hartman wrote:
> On Sun, Aug 16, 2020 at 10:23:25PM +0300, Tomer Samara wrote:
> > BUG_ON() is replaced with WARN_ON at ion_page_pool.c
>
> Why?
>
> > Fixes the following issue:
> > Avoid crashing the kernel - try using WARN_ON & recovery code ratherthan BUG() or BUG_ON().
>
> Ideally you can get rid of WARN_ON() too, right?
>
> Many systems run in panic-on-warn mode, so this really does not change
> anything. Try fixing this up properly to not crash at all.
>
You mean by that to just remove the WARN_ON and leave the condition the
same?
> >
> > Signed-off-by: Tomer Samara <tomersamara98@...il.com>
> > ---
> > drivers/staging/android/ion/ion_page_pool.c | 12 ++++++++----
> > 1 file changed, 8 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/staging/android/ion/ion_page_pool.c b/drivers/staging/android/ion/ion_page_pool.c
> > index 0198b886d906..c1b9eda35c96 100644
> > --- a/drivers/staging/android/ion/ion_page_pool.c
> > +++ b/drivers/staging/android/ion/ion_page_pool.c
> > @@ -46,11 +46,13 @@ static struct page *ion_page_pool_remove(struct ion_page_pool *pool, bool high)
> > struct page *page;
> >
> > if (high) {
> > - BUG_ON(!pool->high_count);
> > + if (WARN_ON(!pool->high_count))
> > + return NULL;
>
> And can you test this that it works properly?
>
> thanks,
>
> greg k-h
Will do :)
Powered by blists - more mailing lists