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;
Date:   Thu, 29 Aug 2019 15:30:38 -0400
User-Agent: K-9 Mail for Android
In-Reply-To: <20190829185901.GA4680@mark-All-Series>
References: <20190829054953.GA18328@mark-All-Series> <20190829064229.GA30423@kroah.com> <20190829135359.GB63638@google.com> <20190829152721.ttsyfwaeygmwmcu7@wittgenstein> <20190829185901.GA4680@mark-All-Series>
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable
Subject: Re: [PATCH] binder: Use kmem_cache for binder_thread
To:     Peikan Tsai <peikantsai@gmail.com>,
        Christian Brauner <christian.brauner@ubuntu.com>
CC:     Greg KH <gregkh@linuxfoundation.org>, arve@android.com,
        tkjos@android.com, maco@android.com, devel@driverdev.osuosl.org,
        linux-kernel@vger.kernel.org
From:   joel@joelfernandes.org
Message-ID: <009FE4FC-E1C2-43A4-AE75-A1FFE9A57356@joelfernandes.org>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On August 29, 2019 2:59:01 PM EDT, Peikan Tsai <peikantsai@gmail=2Ecom> wr=
ote:
>On Thu, Aug 29, 2019 at 05:27:22PM +0200, Christian Brauner wrote:
>> On Thu, Aug 29, 2019 at 09:53:59AM -0400, Joel Fernandes wrote:
>> > On Thu, Aug 29, 2019 at 08:42:29AM +0200, Greg KH wrote:
>> > > On Thu, Aug 29, 2019 at 01:49:53PM +0800, Peikan Tsai wrote:
>> > [snip]=20
>> > > > The allocated size for each binder_thread is 512 bytes by
>kzalloc=2E
>> > > > Because the size of binder_thread is fixed and it's only 304
>bytes=2E
>> > > > It will save 208 bytes per binder_thread when use create a
>kmem_cache
>> > > > for the binder_thread=2E
>> > >=20
>> > > Are you _sure_ it really will save that much memory?  You want to
>do
>> > > allocations based on a nice alignment for lots of good reasons,
>> > > especially for something that needs quick accesses=2E
>> >=20
>> > Alignment can be done for slab allocations, kmem_cache_create()
>takes an
>> > align argument=2E I am not sure what the default alignment of objects
>is
>> > though (probably no default alignment)=2E What is an optimal
>alignment in your
>> > view?
>>=20
>> Probably SLAB_HWCACHE_ALIGN would make most sense=2E
>>=20
>
>Agree=2E Thanks for yours comments and suggestions=2E
>I'll put SLAB_HWCACHE_ALIGN it in patch v2=2E
>
>> >=20
>> > > Did you test your change on a system that relies on binder and
>find any
>> > > speed improvement or decrease, and any actual memory savings?
>> > >=20
>> > > If so, can you post your results?
>> >=20
>> > That's certainly worth it and I thought of asking for the same, but
>spoke too
>> > soon!
>>=20
>> Yeah, it'd be interesting to see what difference this actually makes=2E
>
>>=20
>> Christian
>
>I tested this change on an Android device(arm) with AOSP kernel 4=2E19
>and
>observed
>memory usage of binder_thread=2E But I didn't do binder benchmark yet=2E
>
>On my platform the memory usage of binder_thread reduce about 90 KB as
>the
>following result=2E
>        nr obj          obj size        total
>	before: 624             512             319488 bytes
>	after:  728             312             227136 bytes

And add this to the changelog as well=2E Curious- why is nrobj higher with=
 the patch?

Please don't use my reviewed-by tag yet and I will review the new patch an=
d provide tag separately=2E

Thank you=2E

--=20
Sent from my Android device with K-9 Mail=2E Please excuse my brevity=2E