Received-SPF: pass (google.com: domain of linux-kernel+bounces-160594-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Precedence: bulk
MIME-Version: 1.0
References: <20240424165646.1625690-2-dtatulea@nvidia.com> <4ba023709249e11d97c78a98ac7db3b37f419960.camel@nvidia.com>
 <CAHS8izMbAJHatnM6SvsZVLPY+N7LgGJg03pSdNfSRFCufGh9Zg@mail.gmail.com>
 <4c20b500c2ed615aba424c0f3c7a79f5f5a04171.camel@nvidia.com>
 <CAHS8izPkRJyLctmyj+Ppc5j3Qq5O1u3aPe5h9mnFNHDU2OxA=A@mail.gmail.com>
 <63222bf6a298ae38e77b0c0f49d13581dd9d3a74.camel@nvidia.com>
 <CAHS8izNyJWhcRro8OFPTqsh9J4LEbm7Le6-CiW_oxi2NopAqeQ@mail.gmail.com> <88222bf4d3b4d142ee227eb56d24fa7a7ca43fe1.camel@nvidia.com>
In-Reply-To: <88222bf4d3b4d142ee227eb56d24fa7a7ca43fe1.camel@nvidia.com>
From: Mina Almasry <almasrymina@google.com>
Date: Fri, 26 Apr 2024 12:03:41 -0700
Message-ID: <CAHS8izOhDknjJFZNstXLg=UvbtKTwjbw+0w+4Xx1p+cnqOKRxw@mail.gmail.com>
Subject: Re: [RFC PATCH] net: Fix one page_pool page leak from skb_frag_unref
To: Dragos Tatulea <dtatulea@nvidia.com>
Cc: "davem@davemloft.net" <davem@davemloft.net>, "pabeni@redhat.com" <pabeni@redhat.com>, 
	"ilias.apalodimas@linaro.org" <ilias.apalodimas@linaro.org>, 
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>, 
	"jacob.e.keller@intel.com" <jacob.e.keller@intel.com>, 
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>, Jianbo Liu <jianbol@nvidia.com>, 
	"edumazet@google.com" <edumazet@google.com>, "kuba@kernel.org" <kuba@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Fri, Apr 26, 2024 at 7:59=E2=80=AFAM Dragos Tatulea <dtatulea@nvidia.com=
> wrote:
>
> On Thu, 2024-04-25 at 13:42 -0700, Mina Almasry wrote:
> > On Thu, Apr 25, 2024 at 12:48=E2=80=AFPM Dragos Tatulea <dtatulea@nvidi=
a.com> wrote:
> > >
> > > On Thu, 2024-04-25 at 12:20 -0700, Mina Almasry wrote:
> > > > diff --git a/include/linux/skbuff_ref.h b/include/linux/skbuff_ref.=
h
> > > > index 4dcdbe9fbc5f..4c72227dce1b 100644
> > > > --- a/include/linux/skbuff_ref.h
> > > > +++ b/include/linux/skbuff_ref.h
> > > > @@ -31,7 +31,7 @@ static inline bool napi_pp_get_page(struct page *=
page)
> > > >  static inline void skb_page_ref(struct page *page, bool recycle)
> > > >  {
> > > >  #ifdef CONFIG_PAGE_POOL
> > > > -       if (recycle && napi_pp_get_page(page))
> > > > +       if (napi_pp_get_page(page))
> > > >                 return;
> > > >  #endif
> > > >         get_page(page);
> > > > @@ -69,7 +69,7 @@ static inline void
> > > >  skb_page_unref(struct page *page, bool recycle)
> > > >  {
> > > >  #ifdef CONFIG_PAGE_POOL
> > > > -       if (recycle && napi_pp_put_page(page))
> > > > +       if (napi_pp_put_page(page))
> > > >                 return;
> > > >  #endif
> > > >         put_page(page);
> > > >
> > > >
> > > This is option 2. I thought this would fix everything. But I just tes=
ted and
> > > it's not the case: we are now reaching a negative pp_ref_count.
> > >
> I was tired and botched the revert of the code in this RFC when testing.
> Dropping the recycle flag works as expected. Do we need an RFC v2 or is t=
his non
> RFC material?
>
> Thanks,
> Dragos
>

IMO, it needs to be cleaned up to remove the bool recycle flag dead
code, but other than that I think it's good as a non RFC.

To save you some time I did the said clean up and here is a patch
passing make allmodconfig build + checkpatch/kdoc checks + tests on my
end:

https://pastebin.com/bX5KcHTb

I could submit it to the list if you prefer.

FWIW, if this approach shows no regressions, I think we may be able to
deprecate skb->pp_recycle flag entirely, but I can look into that as a
separate change.

--
Thanks,
Mina