From mboxrd@z Thu Jan  1 00:00:00 1970
X-Msuck: nntp://news.gmane.org/gmane.linux.lib.musl.general/2875
Path: news.gmane.org!not-for-mail
From: Andre Renaud <andre@bluewatersys.com>
Newsgroups: gmane.linux.lib.musl.general
Subject: Re: Re: [PATCH] Added ARM optimised memcpy implementation
Date: Sun, 3 Mar 2013 09:28:23 +1300
Message-ID: <CAPfzE3ZA5i1Ag75CaZRTCa_pvf9-=7kPmLhE6RZt+ZucD4N46w@mail.gmail.com>
References: <1362101090-8525-1-git-send-email-andre@bluewatersys.com>
	<CAPfzE3ZbhCaGNgYuQjd4Yt-TpmdJD3YZXOGm_j67aiZp0eOdXw@mail.gmail.com>
	<A879C957-AA17-4737-8268-55B1FB7BB840@gmail.com>
	<20130301072617.GP6181@port70.net>
Reply-To: musl@lists.openwall.com
NNTP-Posting-Host: plane.gmane.org
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
X-Trace: ger.gmane.org 1362256115 858 80.91.229.3 (2 Mar 2013 20:28:35 GMT)
X-Complaints-To: usenet@ger.gmane.org
NNTP-Posting-Date: Sat, 2 Mar 2013 20:28:35 +0000 (UTC)
To: musl@lists.openwall.com
Original-X-From: musl-return-2876-gllmg-musl=m.gmane.org@lists.openwall.com Sat Mar 02 21:28:59 2013
Return-path: <musl-return-2876-gllmg-musl=m.gmane.org@lists.openwall.com>
Envelope-to: gllmg-musl@plane.gmane.org
Original-Received: from mother.openwall.net ([195.42.179.200])
	by plane.gmane.org with smtp (Exim 4.69)
	(envelope-from <musl-return-2876-gllmg-musl=m.gmane.org@lists.openwall.com>)
	id 1UBt3C-0007UN-BE
	for gllmg-musl@plane.gmane.org; Sat, 02 Mar 2013 21:28:58 +0100
Original-Received: (qmail 11861 invoked by uid 550); 2 Mar 2013 20:28:36 -0000
Mailing-List: contact musl-help@lists.openwall.com; run by ezmlm
Precedence: bulk
List-Post: <mailto:musl@lists.openwall.com>
List-Help: <mailto:musl-help@lists.openwall.com>
List-Unsubscribe: <mailto:musl-unsubscribe@lists.openwall.com>
List-Subscribe: <mailto:musl-subscribe@lists.openwall.com>
Original-Received: (qmail 11853 invoked from network); 2 Mar 2013 20:28:36 -0000
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
        d=google.com; s=20120113;
        h=mime-version:x-received:in-reply-to:references:date:message-id
         :subject:from:to:content-type:x-gm-message-state;
        bh=OYN7ZuQz4L0habh9Tk1AWdGIAvcjtGZKDPn4w/W55Kw=;
        b=G/UILwXfALlw4Lz4NYVJUpuuOuZvfklwJPyaTDQTBEIQ38eCNyQGm10JEFYJdiS8H2
         brYLfb4W2jOgbGHTI0jq+cuQcQO+0TnmnPkcw17dtpuTzP/R+v+/Pm3pEEOgLwRFxLdd
         7WoPI5x9cW/qYhnl5txfS1a2Kd3E3HrF7eakuOwFr2LCHrsWmfEDOKjGOXCHntGotFVb
         riFx4XZdEPE8H2ZW1u6/KphoX9hp3zB4SSSU6bxfDtNNSNHOQPzw3AGhDnStTjBN/Uu5
         EG4dsI3PQ9BTGRvX1VU5W3ygztvZrH7jaLD114Dml/clZCXC6meT65I6UKM5eLkQy58b
         YNuw==
X-Received: by 10.60.5.165 with SMTP id t5mr12316150oet.117.1362256103998;
 Sat, 02 Mar 2013 12:28:23 -0800 (PST)
In-Reply-To: <20130301072617.GP6181@port70.net>
X-Gm-Message-State: ALoCoQkVVaOVMJna4/6TwE5rg3Dh+Qqmt7uFBXyacMvbnOg+uVyCv2Y04duCkNFv4OwSLgJtc89+
Xref: news.gmane.org gmane.linux.lib.musl.general:2875
Archived-At: <http://permalink.gmane.org/gmane.linux.lib.musl.general/2875>

On 1 March 2013 20:26, Szabolcs Nagy <nsz@port70.net> wrote:
> * nwmcsween@gmail.com <nwmcsween@gmail.com> [2013-02-28 19:14:28 -0800]:
>> Hmm what does this do that builtins cannot? What I'm asking is why is this more preformant than preload + word-at-a-time.
>>
>
> musl uses naive memcpy if src and dst are not congruent (src%4 != dst%4)
>
> the android asm takes care of that by fetching a 32bytes
> from src into registers and dumping it into dst with
> apropriate shifts
>
> and in the congruent case 32byte alignment is used (cacheline aligned)

It also takes advantage of the arm load/store multiple instructions,
allowing it to manipulate up to 8 32-bit words with a single
instruction.

Regards,
Andre