[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20190523203523.8f638f378b54df0ef7a18f9d@linux-foundation.org>
Date: Thu, 23 May 2019 20:35:23 -0700
From: Andrew Morton <akpm@...ux-foundation.org>
To: Gen Zhang <blackgod016574@...il.com>
Cc: rppt@...ux.ibm.com, david.engraf@...go.com, steven.price@....com,
osandov@...com, luc.vanoostenryck@...il.com, axboe@...nel.dk,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] initramfs: Fix a missing-chek bug in dir_add()
On Fri, 24 May 2019 11:30:45 +0800 Gen Zhang <blackgod016574@...il.com> wrote:
> In dir_add() and do_name(), de->name and vcollected are allocated by
> kstrdup(). And de->name and vcollected are dereferenced in the following
> codes. However, memory allocation functions such as kstrdup() may fail.
> Dereferencing this null pointer may cause the kernel go wrong. Thus we
> should check these two kstrdup() operations.
> Further, if kstrdup() returns NULL, we should free de in dir_add().
We generally assume that memory allocations within __init code cannot
fail. If one does fail, something quite horrid has happened. The
resulting oops will provide the same information as the proposed panic()
anyway.
Powered by blists - more mailing lists