netdev - Re: [Intel-wired-lan] [iwl-next v4 2/2] igc: Link queues to NAPI instances

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]

Message-ID: <529d08d7-94ee-43da-904e-cf89823a59fb@intel.com>
Date: Mon, 28 Oct 2024 11:53:55 -0700
From: Jacob Keller <jacob.e.keller@...el.com>
To: Joe Damato <jdamato@...tly.com>, "Lifshits, Vitaly"
	<vitaly.lifshits@...el.com>, "netdev@...r.kernel.org"
	<netdev@...r.kernel.org>, "kurt@...utronix.de" <kurt@...utronix.de>, "Gomes,
 Vinicius" <vinicius.gomes@...el.com>, "Nguyen, Anthony L"
	<anthony.l.nguyen@...el.com>, "Kitszel, Przemyslaw"
	<przemyslaw.kitszel@...el.com>, Andrew Lunn <andrew+netdev@...n.ch>, "David
 S. Miller" <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, "Jakub
 Kicinski" <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>, "Alexei
 Starovoitov" <ast@...nel.org>, Daniel Borkmann <daniel@...earbox.net>,
	"Jesper Dangaard Brouer" <hawk@...nel.org>, John Fastabend
	<john.fastabend@...il.com>, "moderated list:INTEL ETHERNET DRIVERS"
	<intel-wired-lan@...ts.osuosl.org>, open list <linux-kernel@...r.kernel.org>,
	"open list:XDP (eXpress Data Path)" <bpf@...r.kernel.org>,
	<stanislaw.gruszka@...ux.intel.com>
Subject: Re: [Intel-wired-lan] [iwl-next v4 2/2] igc: Link queues to NAPI
 instances



On 10/28/2024 9:00 AM, Joe Damato wrote:
> 
> I see, so it looks like there is:
>    - resume
>    - runtime_resume
> 
> The bug I am reintroducing is runtime_resume already holding RTNL
> before my added call to rtnl_lock.
> 
> OK.
> 
> Does resume also hold rtnl before the driver's igc_resume is called?
> I am asking because I don't know much about how PM works.
> 
> If resume does not hold RTNL (but runtime resume does, as the bug
> you pointed out shows), it seems like a wrapper can be added to tell
> the code whether rtnl should be held or not based on which resume is
> happening.
> 
> Does anyone know if: resume (not runtime_resume) already holds RTNL?
> I'll try to take a look and see, but I am not very familiar with PM.

I believe the resume doesn't hold RTNL, as its part of the core device
code, which is not networking specific. It shouldn't be acquiring RTNL
since that is a network specific lock.

I believe the code you posted as v5 should resolve this, and makes sense
to me.

Thanks for digging into this :)

-Jake