Re: [PATCH] can: gw: prefer kfree_rcu() over call_rcu() with cgw_job_free_rcu()

To: Eric Dumazet <edumazet@xxxxxxxxxx>
Subject: Re: [PATCH] can: gw: prefer kfree_rcu() over call_rcu() with cgw_job_free_rcu()
From: Dmitry Antipov <dmantipov@xxxxxxxxx>
Date: Wed, 13 Mar 2024 13:28:12 +0300
Cc: Oliver Hartkopp <socketcan@xxxxxxxxxxxx>, Marc Kleine-Budde <mkl@xxxxxxxxxxxxxx>, linux-can@xxxxxxxxxxxxxxx, netdev@xxxxxxxxxxxxxxx
In-reply-to: <CANn89iLCK10J_6=1xSDquYpToZ-YNG3TzjS60L-g5Cyng92gFw@mail.gmail.com>
References: <20240313094207.70334-1-dmantipov@yandex.ru> <CANn89iLCK10J_6=1xSDquYpToZ-YNG3TzjS60L-g5Cyng92gFw@mail.gmail.com>
User-agent: Mozilla Thunderbird

On 3/13/24 13:18, Eric Dumazet wrote:

kmem_cache_free() is not the same than kfree()

Unless I have missed something in mm territory , your patch is not
going to work.


Hm... it seems that you're better to check include/linux/rcupdate.h
and the comment before kfree_rcu() definition in particular.

Dmitry

Follow-Ups:
- Re: [PATCH] can: gw: prefer kfree_rcu() over call_rcu() with cgw_job_free_rcu()
  - From: Eric Dumazet

References:
- [PATCH] can: gw: prefer kfree_rcu() over call_rcu() with cgw_job_free_rcu()
  - From: Dmitry Antipov
- Re: [PATCH] can: gw: prefer kfree_rcu() over call_rcu() with cgw_job_free_rcu()
  - From: Eric Dumazet