netdev - Re: [PATCH RFC net-next v2 7/7] net: skbuff: always try to recycle PP pages directly when in softirq

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Fri, 21 Jul 2023 09:01:29 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Alexander Lobakin <aleksander.lobakin@...el.com>, Yunsheng Lin
 <linyunsheng@...wei.com>
Cc: "David S. Miller" <davem@...emloft.net>, Eric Dumazet
 <edumazet@...gle.com>, Paolo Abeni <pabeni@...hat.com>, Maciej Fijalkowski
 <maciej.fijalkowski@...el.com>, Larysa Zaremba <larysa.zaremba@...el.com>,
 Alexander Duyck <alexanderduyck@...com>, Jesper Dangaard Brouer
 <hawk@...nel.org>, Ilias Apalodimas <ilias.apalodimas@...aro.org>,
 <netdev@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH RFC net-next v2 7/7] net: skbuff: always try to recycle
 PP pages directly when in softirq

On Fri, 21 Jul 2023 17:37:57 +0200 Alexander Lobakin wrote:
> > Does it mean ptr_ring_produce_any() is needed in
> > page_pool_recycle_in_ring() too?
> > 
> > As it is assumed that page pool API can be called in the context with
> > irqs_disabled() || in_hardirq(), and force recylcling happens in the
> > prt_ring.
> > 
> > Isn't it conflit with the below patch? as the below patch assume page
> > pool API can not be called in the context with irqs_disabled() || in_hardirq():
> > [PATCH net-next] page_pool: add a lockdep check for recycling in hardirq
> > 
> > Or am I missing something obvious here?  
> 
> No, Page Pool is *not* intended to be called when IRQs are disabled,
> hence the fix Jakub's posted in the separate thread.

Yeah, it's just a bandaid / compromise, since Olek really wants his
optimization and I really don't want to spend a month debugging
potential production crashes :)

On the ptr ring the use may still be incorrect but there is a spin lock
so things will explode in much more obvious way, if they do.