From mboxrd@z Thu Jan 1 00:00:00 1970 From: Heinrich Schuchardt Date: Wed, 23 Sep 2020 22:36:38 +0200 Subject: [PATCH 1/1] fs/squashfs: parameter check sqfs_read_metablock() In-Reply-To: <20200923214941.5a39daf1@windsurf> References: <20200923191351.33474-1-xypron.glpk@gmx.de> <20200923214941.5a39daf1@windsurf> Message-ID: List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de On 9/23/20 9:49 PM, Thomas Petazzoni wrote: > On Wed, 23 Sep 2020 21:13:51 +0200 > Heinrich Schuchardt wrote: > >> We should check if the incoming parameter file_mapping is not NULL instead >> of checking after adding an offset. >> >> Reported-by: Coverity CID 307210 >> Signed-off-by: Heinrich Schuchardt > > Acked-by: Thomas Petazzoni > > However, I wonder if this check is really useful. sqfs_read_metablock() > is an internal function, so it should be up to the callers to make sure > that they don't pass a NULL file_mapping argument. > > Thomas > This is a question of programming style. Eliminating the check may be justified if the input variable is checked by every caller. The real problems are elsewhere, e.g. sqfs_search_dir() and sqfs_readdir() do not check the return value of sqfs_find_inode() which may be NULL. sqfs_opendir() leaks allocated memory if an error occurs. Best regards Heinrich