Received-SPF: pass (google.com: domain of linux-kernel+bounces-43877-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240125062044.63344-1-cuiyunhui@bytedance.com>
 <ZbdYijWK1PnHXn47@xhacker> <CAEEQ3wnSN8dLh3FcmHq5yPwQRdjLQa_VjcuTH+7YYZLOqCzaCQ@mail.gmail.com>
 <ZbdyLJ24I2fa6oNb@xhacker> <CAEEQ3w=gpSTxzxTzh-zHbP37P8s5eVCzG6RZndBXF-U+mNrziw@mail.gmail.com>
 <ZbeRgRBhvNnnTd8h@xhacker>
In-Reply-To: <ZbeRgRBhvNnnTd8h@xhacker>
From: yunhui cui <cuiyunhui@bytedance.com>
Date: Tue, 30 Jan 2024 11:18:31 +0800
Message-ID: <CAEEQ3wmoQ=f9OWu5FsuinFDRUtGE6kFk6xac-diYD664Mt_VFg@mail.gmail.com>
Subject: Re: [External] Re: [PATCH] RISC-V: add uniprocessor flush_tlb_range() support
To: Jisheng Zhang <jszhang@kernel.org>
Cc: paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, 
	alexghiti@rivosinc.com, samuel.holland@sifive.com, ajones@ventanamicro.com, 
	mchitale@ventanamicro.com, dylan@andestech.com, 
	sergey.matyukevich@syntacore.com, prabhakar.mahadev-lad.rj@bp.renesas.com, 
	apatel@ventanamicro.com, linux-riscv@lists.infradead.org, 
	linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi Jisheng,


On Mon, Jan 29, 2024 at 8:05=E2=80=AFPM Jisheng Zhang <jszhang@kernel.org> =
wrote:
>
> On Mon, Jan 29, 2024 at 07:02:10PM +0800, yunhui cui wrote:
> > Hi Jisheng,
> >
> > On Mon, Jan 29, 2024 at 5:51=E2=80=AFPM Jisheng Zhang <jszhang@kernel.o=
rg> wrote:
> > >
> > > On Mon, Jan 29, 2024 at 04:26:57PM +0800, yunhui cui wrote:
> > > > Hi Jisheng,
> > > >
> > > > On Mon, Jan 29, 2024 at 4:02=E2=80=AFPM Jisheng Zhang <jszhang@kern=
el.org> wrote:
> > > > >
> > > > > On Thu, Jan 25, 2024 at 02:20:44PM +0800, Yunhui Cui wrote:
> > > > > > Add support for flush_tlb_range() to improve TLB performance fo=
r
> > > > > > UP systems. In order to avoid the mutual inclusion of tlbflush.=
h
> > > > > > and hugetlb.h, the UP part is also implemented in tlbflush.c.
> > > > >
> > > > > Hi Yunhui,
> > > > >
> > > > > IIRC, Samuel sent similar patch series a few weeks ago.
> > > > >
> > > > > https://lore.kernel.org/linux-riscv/20240102220134.3229156-1-samu=
el.holland@sifive.com/
> > > > >
> > > > > After that series, do you still need this patch?
> > > >
> > > > Thank you for your reminder. I didn't find it before I mailed my
> > > > patch. I just looked at the content of this patch. I understand tha=
t
> > > > my patch is needed. For a single core, a more concise TLB flush log=
ic
> > > > is needed, and it is helpful to improve performance.
> > >
> > > Currently, riscv UP flush_tlb_range still use flush all TLB entries,
> > > obviously it's is a big hammer, this is what your patch is trying to
> > > optimize. I'm not sure whether I understand your code correctly or no=
t.
> > > Let me know if I misunderstand your code.
> > >
> > > After patch5 of the Samuel's series, __flush_tlb_range is unified for
> > > SMP and UP, so that UP can also benefit from recent improvements, suc=
h
> > > as range flush rather than all.
> >
> > In my opinion, UP does not need to combine some SMP if... else,
> > on_each_cpu(...) logic, which is also a manifestation of performance
>
> Hi Yunhui,
>
> IIRC, the compiler will optimise out the unnecessary logic under UP, I
> may misread the code. But if no, indeed, there's improvement room.
> However, even in this case, IMHO, it's better if you can base on
> Samuel's series.
> Anyway, the optimization(range tlb entries rather than *all* entries unde=
r
> UP case) you want to do has been implemented. While I'm not sure whether
> we can rely on the compiler to optimize out all unnecessary logics.

Okay, let's see if there are any necessary optimizations based on
Samuel's series.

Thanks,
Yunhui