linux-kernel - Re: [PATCH v2 1/4] mm/mempolicy: Fix memory leaks in mempolicy_sysfs

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: <20250317082406.818-1-rakie.kim@sk.com>
Date: Mon, 17 Mar 2025 17:24:00 +0900
From: Rakie Kim <rakie.kim@...com>
To: Gregory Price <gourry@...rry.net>
Cc: Rakie Kim <rakie.kim@...com>,
	akpm@...ux-foundation.org,
	linux-mm@...ck.org,
	linux-kernel@...r.kernel.org,
	linux-cxl@...r.kernel.org,
	joshua.hahnjy@...il.com,
	dan.j.williams@...el.com,
	ying.huang@...ux.alibaba.com,
	kernel_team@...ynix.com,
	honggyu.kim@...com,
	yunjeong.mun@...com,
	Jonathan Cameron <Jonathan.Cameron@...wei.com>
Subject: Re: [PATCH v2 1/4] mm/mempolicy: Fix memory leaks in mempolicy_sysfs_init()

On Fri, 14 Mar 2025 09:42:31 -0400 Gregory Price <gourry@...rry.net> wrote:
> On Fri, Mar 14, 2025 at 10:55:00AM +0000, Jonathan Cameron wrote:
> > > 
> > > 1) allocate node_attrs and mempolicy_kobj up front and keep your
> > >    reordering, this lets us clean up allocations on failure before
> > >    kobject_init is called
> > > 
> > > 2) after this remove all the other code and just let
> > >    mempolicy_kobj_release clean up node_attrs
> > > 
> > > 3) Add a (%d) to the error message to differentiate failures
> > 
> > Given how unlikely (and noisy) a memory allocation failure is,
> > maybe just drop the printing at all in those paths - allowing
> > early returns.
> > 
> > The lifetime rules around node_attrs in here are making readability
> > poor. It is implicitly owned by the mempolicy_kobj, but no direct association.
> > Maybe just encapsulating the kobject in a structure that contains
> > this as a [] array at the end.  Then we end up with single allocation of
> > stuff that is effectively one thing.
> >
> 
> Even better recommendation, lets do as Jonathan suggests. <3
> 
> ~Gregory

Hi Gregory

I will revise the next version based on Jonathan's feedback.
Moreover, I'll separate this patch from the hotplug series and make it an
independent patch.

Rakie