netdev - Re: [PATCH net-next 05/15] idpf: add create vport and netdev configuration

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 for Android: free password hash cracker in your pocket

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

Date: Wed, 7 Jun 2023 12:10:06 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Willem de Bruijn <willemdebruijn.kernel@...il.com>
Cc: "Linga, Pavan Kumar" <pavan.kumar.linga@...el.com>, Tony Nguyen
 <anthony.l.nguyen@...el.com>, davem@...emloft.net, pabeni@...hat.com,
 edumazet@...gle.com, netdev@...r.kernel.org, emil.s.tantilov@...el.com,
 jesse.brandeburg@...el.com, sridhar.samudrala@...el.com,
 shiraz.saleem@...el.com, sindhu.devale@...el.com, willemb@...gle.com,
 decot@...gle.com, andrew@...n.ch, leon@...nel.org, mst@...hat.com,
 simon.horman@...igine.com, shannon.nelson@....com,
 stephen@...workplumber.org, Alan Brady <alan.brady@...el.com>, Joshua Hay
 <joshua.a.hay@...el.com>, Madhu Chittim <madhu.chittim@...el.com>, Phani
 Burra <phani.r.burra@...el.com>, Shailendra Bhatnagar
 <shailendra.bhatnagar@...el.com>, Krishneil Singh
 <krishneil.k.singh@...el.com>
Subject: Re: [PATCH net-next 05/15] idpf: add create vport and netdev
 configuration

On Wed, 7 Jun 2023 20:20:57 +0200 Willem de Bruijn wrote:
> > > Please use locks. Every single Intel driver comes with gazillion flags
> > > and endless bugs when the flags go out of sync.  
> >
> > Thanks for the feedback. Will use mutex lock instead of 'VC_MSG_PENDING'
> > flag.  
> 
> Was that the intent of the comment?
> 
> Or is it to replace these individual atomic test_and_set bit
> operations with a single spinlock-protected critical section around
> all the flag operations?

No, no. Intel drivers have a history of adding flags to work around
locking problems. Whatever this bit is protecting should be protected
by a normal synchronization primitive instead.

I don't understand why.

Replacing an atomic bitop with a spin lock is a non-change.

> That's how I read the suggestion.