Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20190830035716.GA190684@LGEARND20B15> <20190830135007.8b5949bd57975d687ff0a3f8@linux-foundation.org>
In-Reply-To: <20190830135007.8b5949bd57975d687ff0a3f8@linux-foundation.org>
From:   Austin Kim <austindh.kim@gmail.com>
Date:   Tue, 3 Sep 2019 06:28:08 +0900
Message-ID: <CADLLry497WBX0y+y5UuKgSLRjCd+5vbL1qAfUW-U4qsJ8zR6Vg@mail.gmail.com>
Subject: Re: [PATCH] mm/vmalloc: move 'area->pages' after if statement
To:     Andrew Morton <akpm@linux-foundation.org>
Cc:     urezki@gmail.com, guro@fb.com, rpenyaev@suse.de, mhocko@suse.com,
        rick.p.edgecombe@intel.com, rppt@linux.ibm.com,
        aryabinin@virtuozzo.com, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

2019=EB=85=84 8=EC=9B=94 31=EC=9D=BC (=ED=86=A0) =EC=98=A4=EC=A0=84 5:50, A=
ndrew Morton <akpm@linux-foundation.org>=EB=8B=98=EC=9D=B4 =EC=9E=91=EC=84=
=B1:
>
> On Fri, 30 Aug 2019 12:57:16 +0900 Austin Kim <austindh.kim@gmail.com> wr=
ote:
>
> > If !area->pages statement is true where memory allocation fails,
> > area is freed.
> >
> > In this case 'area->pages =3D pages' should not executed.
> > So move 'area->pages =3D pages' after if statement.
> >
> > ...
> >
> > --- a/mm/vmalloc.c
> > +++ b/mm/vmalloc.c
> > @@ -2416,13 +2416,15 @@ static void *__vmalloc_area_node(struct vm_stru=
ct *area, gfp_t gfp_mask,
> >       } else {
> >               pages =3D kmalloc_node(array_size, nested_gfp, node);
> >       }
> > -     area->pages =3D pages;
> > -     if (!area->pages) {
> > +
> > +     if (!pages) {
> >               remove_vm_area(area->addr);
> >               kfree(area);
> >               return NULL;
> >       }
> >
> > +     area->pages =3D pages;
> > +
> >       for (i =3D 0; i < area->nr_pages; i++) {
> >               struct page *page;
> >
>
> Fair enough.  But we can/should also do this?

I agreed since it is the same treatment.
Thanks for feedback.

>
> --- a/mm/vmalloc.c~mm-vmalloc-move-area-pages-after-if-statement-fix
> +++ a/mm/vmalloc.c
> @@ -2409,7 +2409,6 @@ static void *__vmalloc_area_node(struct
>         nr_pages =3D get_vm_area_size(area) >> PAGE_SHIFT;
>         array_size =3D (nr_pages * sizeof(struct page *));
>
> -       area->nr_pages =3D nr_pages;
>         /* Please note that the recursion is strictly bounded. */
>         if (array_size > PAGE_SIZE) {
>                 pages =3D __vmalloc_node(array_size, 1, nested_gfp|highme=
m_mask,
> @@ -2425,6 +2424,7 @@ static void *__vmalloc_area_node(struct
>         }
>
>         area->pages =3D pages;
> +       area->nr_pages =3D nr_pages;
>
>         for (i =3D 0; i < area->nr_pages; i++) {
>                 struct page *page;
> _
>