Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Jiaxun Yang <jiaxun.yang@flygoat.com>
To:     linux-mips@linux-mips.org
Cc:     Peter Zijlstra <peterz@infradead.org>,
        =?utf-8?B?6ZmI5Y2O5omN?= <chenhc@lemote.com>,
        Paul Burton <paul.burton@mips.com>,
        Ralf Baechle <ralf@linux-mips.org>,
        James Hogan <jhogan@kernel.org>,
        Fuxin Zhang <zhangfx@lemote.com>,
        wuzhangjin <wuzhangjin@gmail.com>,
        stable <stable@vger.kernel.org>,
        Alan Stern <stern@rowland.harvard.edu>,
        Andrea Parri <andrea.parri@amarulasolutions.com>,
        Will Deacon <will.deacon@arm.com>,
        Boqun Feng <boqun.feng@gmail.com>,
        Nicholas Piggin <npiggin@gmail.com>,
        David Howells <dhowells@redhat.com>,
        Jade Alglave <j.alglave@ucl.ac.uk>,
        Luc Maranget <luc.maranget@inria.fr>,
        "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
        Akira Yokosawa <akiyks@gmail.com>,
        LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH V2] MIPS: implement smp_cond_load_acquire() for Loongson-3
Date:   Wed, 11 Jul 2018 18:05:51 +0800
Message-ID: <5471216.FKXZRxKFUI@flygoat-ry>
In-Reply-To: <20180710121727.GK2476@hirez.programming.kicks-ass.net>
References: <1531103198-16764-1-git-send-email-chenhc@lemote.com> <tencent_26F8B9E004D4512B2225FCE1@qq.com> <20180710121727.GK2476@hirez.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 2018-7-10 Tue at 20:17:27=EF=BC=8CPeter Zijlstra Wrote=EF=BC=9A

Hi Peter
Since Huacai unable to send email via client, I'm going to reply for him=20
=20
> Sure.. we all got that far. And no, this isn't the _real_ problem. This
> is a manifestation of the problem.
>=20
> The problem is that your SFB is broken (per the Linux requirements). We
> require that stores will become visible. That is, they must not
> indefinitely (for whatever reason) stay in the store buffer.
>=20
> > I don't think this is a hardware bug, in design, SFB will flushed to
> > L1 cache in three cases:
> >=20
> > 1, data in SFB is full (be a complete cache line);
> > 2, there is a subsequent read access in the same cache line;
> > 3, a 'sync' instruction is executed.
>=20
> And I think this _is_ a hardware bug. You just designed the bug instead
> of it being by accident.
Yes, we understood that this hardware feature is not supported by LKML,
so it should be a hardware bug for LKML.
>=20
> It doesn't happen an _any_ other architecture except that dodgy
> ARM11MPCore part. Linux hard relies on stores to become available
> _eventually_.
>=20
> Still, even with the rules above, the best work-around is still the very
> same cpu_relax() hack.

As you say, SFB makes Loongson not fully SMP-coherent.
However, modify cpu_relax can solve the current problem,
but not so straight forward. On the other hand, providing a Loongson-specif=
ic=20
WRITE_ONCE looks more reasonable, because it the eliminate the "non-cohrenc=
y".
So we can solve the bug from the root.

Thanks.
=2D-
Jiaxun Yang