Received-SPF: pass (google.com: domain of linux-kernel+bounces-55968-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Precedence: bulk
MIME-Version: 1.0
References: <20240206202511.4799-1-21cnbao@gmail.com> <20240207014442.GI69174@google.com>
 <41226c84-e780-4408-b7d2-bd105f4834f5@kernel.dk> <20240207031447.GA489524@google.com>
 <79926b88-c1ef-4042-a871-61752d29c838@kernel.dk> <20240207032148.GB489524@google.com>
In-Reply-To: <20240207032148.GB489524@google.com>
From: Barry Song <21cnbao@gmail.com>
Date: Wed, 7 Feb 2024 18:00:55 +1300
Message-ID: <CAGsJ_4yPChtForHNqqT_C7Yge6-0KvDDs=oNqqDuDXNwOeuXoA@mail.gmail.com>
Subject: Re: [PATCH v2] zram: easy the allocation of zcomp_strm's buffers
 through vmalloc
To: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: Jens Axboe <axboe@kernel.dk>, Minchan Kim <minchan@kernel.org>, 
	Andrew Morton <akpm@linux-foundation.org>, linux-block@vger.kernel.org, 
	linux-kernel@vger.kernel.org, zhengtangquan@oppo.com, 
	Barry Song <v-songbaohua@oppo.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Wed, Feb 7, 2024 at 4:21=E2=80=AFPM Sergey Senozhatsky
<senozhatsky@chromium.org> wrote:
>
> On (24/02/06 20:17), Jens Axboe wrote:
> [..]
> > >>> Minchan, if you have any objections, please chime in.
> > >>
> > >> Not Minchan, but I do have an issue with the title of the commit, it
> > >> doesn't make any sense. Can the maintainer please re-write that to b=
e
> > >> something that is appropriate and actually describes what the patch
> > >> does?
> > >
> > > Thanks Jens. I fully agree, I requested a better commit message in
> > > v1 feedback, we probably still can improve on this.
> > >
> > >
> > > Something like this?
> > >
> > > ---
> > >
> > > zram: do not allocate physically contiguous strm buffers
> > >
> > > Currently zram allocates 2 physically contigous pages per-CPU's
> > > compression stream (we may have up to 3 streams per-CPU). Since
> > > those buffers are per-CPU we allocate them from CPU hotplug path,
> > > which may have higher risks of failed allocations on devices with
> > > fragmented memory.
> > >
> > > Switch to virtually contiguos allocations - crypto comp does not
> > > seem impose requirements on compression working buffers to be
> > > physically contiguous.
> >
> > Yep, this is much better! Thanks.
>
> Thanks.
>

Hi Sergey, Jens,

> Barry, can you please send v3 with the suggested subject and commit
> message?

Thanks for your comments and improvements. will send v3 accordingly.

Best regard
Barry