Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Thu, 3 Dec 2020 20:01:45 -0800
From:   Eric Biggers <ebiggers@kernel.org>
To:     Daeho Jeong <daeho43@gmail.com>
Cc:     linux-kernel@vger.kernel.org,
        linux-f2fs-devel@lists.sourceforge.net, kernel-team@android.com,
        Daeho Jeong <daehojeong@google.com>
Subject: Re: [f2fs-dev] [PATCH] f2fs: fix race of pending_pages in
 decompression
Message-ID: <X8m0qbd7rvQyiwOt@sol.localdomain>
References: <20201204005847.654074-1-daeho43@gmail.com>
 <X8msy1T8uqZ4Z/iR@sol.localdomain>
 <CACOAw_wTFsfgLfrWKRoM1o_HQorJE-=2ztZftQTn+comcpmHxQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CACOAw_wTFsfgLfrWKRoM1o_HQorJE-=2ztZftQTn+comcpmHxQ@mail.gmail.com>
Precedence: bulk

On Fri, Dec 04, 2020 at 12:43:23PM +0900, Daeho Jeong wrote:
> STEP_VERITY is enabled by f2fs_need_verity() and the function is like below.
> We already know the second condition (idx < DIV_ROUND_UP...) is
> satisfied when invoking f2fs_alloc_dic().
> 
> static inline bool f2fs_need_verity(const struct inode *inode, pgoff_t idx)
> {
>         return fsverity_active(inode) &&
>                idx < DIV_ROUND_UP(inode->i_size, PAGE_SIZE);
> }

Are you sure?  I thought that compression (and encryption) apply to the whole
file, including any Merkle tree blocks past i_size.

Also, even if you include the i_size check, it's still wrong to check
fsverity_active() in the middle of the I/O because FS_IOC_ENABLE_VERITY can
execute concurrently, causing fsverity_active() to return false at the beginning
of the I/O and true later in the I/O.  It needs to be checked only once, at the
beginning...

- Eric