Received-SPF: pass (google.com: domain of linux-crypto+bounces-458-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Message-ID: <tencent_B2D2C5864C49BDAEE0AEC1CC9627E3C6CF06@qq.com>
From: Edward Adam Davis <eadavis@qq.com>
To: herbert@gondor.apana.org.au
Cc: davem@davemloft.net,
	eadavis@qq.com,
	linux-crypto@vger.kernel.org,
	linux-kernel@vger.kernel.org,
	olivia@selenic.com,
	syzbot+c52ab18308964d248092@syzkaller.appspotmail.com,
	syzkaller-bugs@googlegroups.com
Subject: Re: [PATCH] hwrng: core - fix task hung in hwrng_fillfn
Date: Fri,  1 Dec 2023 20:34:12 +0800
In-Reply-To: <ZWnGV39HJr9uUB2/@gondor.apana.org.au>
References: <ZWnGV39HJr9uUB2/@gondor.apana.org.au>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

On Fri, 1 Dec 2023 19:41:11 +0800, Herbert Xu wrote:
> >
> > Reduce the scope of critical zone protection.
> > The original critical zone contains a too large range, especially like
> > copy_to_user() should not be included in the critical zone.
> 
> Which part in particular is taking 143 seconds? The buffer is
> only 128 bytes long.  Why is a 128-byte copy taking 143 seconds,
> even with a page fault?
According to splat, after a page fault occurred, the attempt to retrieve 
rcu_read_lock() failed, resulting in a timeout of 143s. This is my speculation.

Edward