Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:80244
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain ajf.me designates 192.64.116.207 as permitted sender)
Content-Type: text/plain; charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 8.1 \(1993\))
In-Reply-To: <CA+9eiLv6MQjvqe2XXQjpUF1PjaNgPMYOoUmVDR3qZLkcjTd+SA@mail.gmail.com>
Date: Tue, 6 Jan 2015 23:54:29 +0000
Cc: PHP Internals <internals@lists.php.net>
Content-Transfer-Encoding: quoted-printable
Message-ID: <93376D88-A704-4F89-AB90-82EEDAE82081@ajf.me>
References: <8C47FA53-0964-49C0-963C-332A936348A5@ajf.me> <CA+9eiLuOcd-u_kzu+MZM71Ns_-9vsSbifFF5x9Q1b4k0NV12Wg@mail.gmail.com> <F41D165E-A2FB-4423-BF8F-6F6B5F4E201E@ajf.me> <68229C26-4EEC-49DC-BA05-D5AC9728D1E8@ajf.me> <CA+9eiLvh9RFVmYD1_y0AU0NVkkxYTfjv_C6H4tN8RH-w7X7M1g@mail.gmail.com> <EBBCF759-F622-4903-9821-DE3C63CD142A@ajf.me> <CA+9eiLv6MQjvqe2XXQjpUF1PjaNgPMYOoUmVDR3qZLkcjTd+SA@mail.gmail.com>
To: Dmitry Stogov <dmitry@zend.com>
Subject: Re: [PHP-DEV] [RFC] Big Integer Support
From: ajf@ajf.me (Andrea Faulds)

Hey Dmitry,

> On 23 Oct 2014, at 16:22, Dmitry Stogov <dmitry@zend.com> wrote:
>=20
>> I=E2=80=99ve so far been scared to touch the asm=E2=80=A6 but =
actually, I don=E2=80=99t think it
>> could be *that* hard. It=E2=80=99s not doing something especially =
complex. The
>> bigint API looks fairly stable now and I=E2=80=99m unlikely to change =
it much
>> further, so there=E2=80=99s little worry about having to change the =
asm a second
>> time. The main problem with asm, I suppose, is testing it. I do have =
a
>> 32-bit Ubuntu VM set up, but I=E2=80=99d also need to set up Windows =
VMs, and
>> possibly others (don=E2=80=99t we have PowerPC in the source just =
now?).
>>=20
>=20
> change asm for 32-bit Linux and add TODO marks for others. I don't =
test PHP
> on PPC as well.

After procrastinating about this for a long time, I finally went and =
updated the overflow checks today and ran bench.php.

I still haven=E2=80=99t touched the inline asm, I=E2=80=99ve just =
removed it, since clang and GCC (only in GCC 5.0, sadly) have checked =
arithmetic intrinsics. If someone wants to, they can rewrite the inline =
asm for compilers that have no overflow-checking intrinsics, but this is =
good enough for now, at least for the purposes of performance checking =
on my machine. I=E2=80=99m using clang, by the way. If you want to =
replicate these results, you=E2=80=99ll probably also need it, since GCC =
5.0 isn=E2=80=99t out yet, unfortunately.

I compiled the bigint-libtommath branch (theoretically this was just a =
branch, but actually all the new changes have gone there, I=E2=80=99ll =
merge it into the bigint branch once LibTomMath port is done), and the =
current master branch.

For bigint-libtommath, I used ./configure --enable-debug --enable-phpdbg =
=E2=80=94-disable-all =E2=80=94-enable-bigint-gmp

Because of the =E2=80=94-enable-bigint-gmp flag, it=E2=80=99s using the =
GMP backend, not the LibTomMath one. I=E2=80=99m doing this since =
there=E2=80=99s still one or two small things I haven=E2=80=99t finished =
implementing for LibTomMath, e.g. the binary bitwise ops have the wrong =
behaviour just now.

For master, I used ./configure --enable-debug --enable-phpdbg =
=E2=80=94-disable-all

Then, I ran bench.php four times, and each time I ran it first on =
./php-bigint-gmp, then on ./php-bigint-master.

On each run, the bigint branch turned out faster, as well as overall:

bigint	master
6.593	6.659
6.424	6.661
6.414	6.588
6.381	6.673
AVERAGE=09
6.453	6.64525
DIFFERENCE=09
-0.19225	0.19225
RATIO=09
0.971069561	1.0297923446

So master is 2.9% slower! Full output here: =
https://gist.github.com/TazeTSchnitzel/759c1513b442571f5e26

I can=E2=80=99t actually explain why bigints would be faster. It might =
just be because I got rid of fast_increment_function in favour of just =
checking of op1 =3D=3D ZEND_LONG_MAX in zend_vm_execute.h, ditto for =
fast_decrement_function. Maybe using overflow intrinsics is faster than =
inline asm. Maybe it=E2=80=99s something completely different. I =
honestly don=E2=80=99t know.

The result surprised me as I expect bigints would be slower, so I redid =
it. Again, bigints came out on top:

bigint	master
6.55	6.779
6.353	6.738
6.326	6.674
6.144	6.177
AVERAGE=09
6.34325	6.592
DIFFERENCE=09
-0.24875	0.24875
RATIO=09
0.9622648665	1.0392149135

This time master was around 3.9% slower. Full log here: =
https://gist.github.com/TazeTSchnitzel/59c190b86c9dd5b20570

If we combine the two runs:

bigint	master
6.593	6.659
6.424	6.661
6.414	6.588
6.381	6.673
6.55	6.779
6.353	6.738
6.326	6.674
6.144	6.177
AVERAGE=09
6.398125	6.618625
DIFFERENCE=09
-0.2205	0.2205
RATIO=09
0.9666849232	1.0344632216

master=E2=80=99s 3.4% slower.

Just to check I named the files correctly:

oa-res-26-240:php-src ajf$ ./php-master -r 'var_dump(PHP_INT_MAX * 2);'
float(1.844674407371E+19)
oa-res-26-240:php-src ajf$ ./php-bigint-gmp -r 'var_dump(PHP_INT_MAX * =
2);'
int(18446744073709551614)

Yes, it=E2=80=99s definitely the bigint branch.

So, at least by these preliminary results, the bigint branch would =
appear to be faster than master. This is merely bench.php, but it=E2=80=99=
s still a good sign. :)

Thanks!
--
Andrea Faulds
http://ajf.me/