qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Sergey Fedorov <serge.fdrv@gmail.com>
To: "Emilio G. Cota" <cota@braap.org>,
	QEMU Developers <qemu-devel@nongnu.org>,
	MTTCG Devel <mttcg@listserver.greensocs.com>
Cc: "Alex Bennée" <alex.bennee@linaro.org>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Peter Crosthwaite" <crosthwaite.peter@gmail.com>,
	"Richard Henderson" <rth@twiddle.net>
Subject: Re: [Qemu-devel] [PATCH v5 08/18] exec: add tb_hash_func5, derived from xxhash
Date: Tue, 17 May 2016 20:22:52 +0300	[thread overview]
Message-ID: <573B536C.90804@gmail.com> (raw)
In-Reply-To: <1463196873-17737-9-git-send-email-cota@braap.org>

On 14/05/16 06:34, Emilio G. Cota wrote:
> This will be used by upcoming changes for hashing the tb hash.
>
> Add this into a separate file to include the copyright notice from
> xxhash.
>
> Reviewed-by: Richard Henderson <rth@twiddle.net>
> Signed-off-by: Emilio G. Cota <cota@braap.org>
> ---
>  include/exec/tb-hash-xx.h | 94 +++++++++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 94 insertions(+)
>  create mode 100644 include/exec/tb-hash-xx.h
>
> diff --git a/include/exec/tb-hash-xx.h b/include/exec/tb-hash-xx.h
> new file mode 100644
> index 0000000..67f4e6f
> --- /dev/null
> +++ b/include/exec/tb-hash-xx.h
> @@ -0,0 +1,94 @@
> +/*
> + * xxHash - Fast Hash algorithm
> + * Copyright (C) 2012-2016, Yann Collet
> + *
> + * BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
> + *
> + * Redistribution and use in source and binary forms, with or without
> + * modification, are permitted provided that the following conditions are
> + * met:
> + *
> + * + Redistributions of source code must retain the above copyright
> + * notice, this list of conditions and the following disclaimer.
> + * + Redistributions in binary form must reproduce the above
> + * copyright notice, this list of conditions and the following disclaimer
> + * in the documentation and/or other materials provided with the
> + * distribution.
> + *
> + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
> + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
> + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
> + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
> + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
> + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
> + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
> + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
> + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
> + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
> + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
> + *
> + * You can contact the author at :
> + * - xxHash source repository : https://github.com/Cyan4973/xxHash
> + */
> +#ifndef EXEC_TB_HASH_XX
> +#define EXEC_TB_HASH_XX
> +
> +#include <qemu/bitops.h>
> +
> +#define PRIME32_1   2654435761U
> +#define PRIME32_2   2246822519U
> +#define PRIME32_3   3266489917U
> +#define PRIME32_4    668265263U
> +#define PRIME32_5    374761393U
> +
> +#define TB_HASH_XX_SEED 1
> +
> +/*
> + * xxhash32, customized for input variables that are not guaranteed to be
> + * contiguous in memory.
> + */
> +static inline
> +uint32_t tb_hash_func5(uint64_t a0, uint64_t b0, uint32_t e)
> +{
> +    uint32_t v1 = TB_HASH_XX_SEED + PRIME32_1 + PRIME32_2;
> +    uint32_t v2 = TB_HASH_XX_SEED + PRIME32_2;
> +    uint32_t v3 = TB_HASH_XX_SEED + 0;
> +    uint32_t v4 = TB_HASH_XX_SEED - PRIME32_1;
> +    uint32_t a = a0 >> 31 >> 1;

I'm wondering if there's something special forcing us to make ">> 31
>>1" instead of just ">> 32" on uint64_t?

Kind regards,
Sergey

> +    uint32_t b = a0;
> +    uint32_t c = b0 >> 31 >> 1;
> +    uint32_t d = b0;
> +    uint32_t h32;
> +
> +    v1 += a * PRIME32_2;
> +    v1 = rol32(v1, 13);
> +    v1 *= PRIME32_1;
> +
> +    v2 += b * PRIME32_2;
> +    v2 = rol32(v2, 13);
> +    v2 *= PRIME32_1;
> +
> +    v3 += c * PRIME32_2;
> +    v3 = rol32(v3, 13);
> +    v3 *= PRIME32_1;
> +
> +    v4 += d * PRIME32_2;
> +    v4 = rol32(v4, 13);
> +    v4 *= PRIME32_1;
> +
> +    h32 = rol32(v1, 1) + rol32(v2, 7) + rol32(v3, 12) + rol32(v4, 18);
> +    h32 += 20;
> +
> +    h32 += e * PRIME32_3;
> +    h32  = rol32(h32, 17) * PRIME32_4;
> +
> +    h32 ^= h32 >> 15;
> +    h32 *= PRIME32_2;
> +    h32 ^= h32 >> 13;
> +    h32 *= PRIME32_3;
> +    h32 ^= h32 >> 16;
> +
> +    return h32;
> +}
> +
> +#endif /* EXEC_TB_HASH_XX */

  reply	other threads:[~2016-05-17 17:23 UTC|newest]

Thread overview: 79+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-05-14  3:34 [Qemu-devel] [PATCH v5 00/18] tb hash improvements Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 01/18] compiler.h: add QEMU_ALIGNED() to enforce struct alignment Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 02/18] seqlock: remove optional mutex Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 03/18] seqlock: rename write_lock/unlock to write_begin/end Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 04/18] include/processor.h: define cpu_relax() Emilio G. Cota
2016-05-18 17:47   ` Sergey Fedorov
2016-05-18 18:29     ` Emilio G. Cota
2016-05-18 18:37       ` Sergey Fedorov
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 05/18] atomics: add atomic_test_and_set_acquire Emilio G. Cota
2016-05-16 10:05   ` Paolo Bonzini
2016-05-17 16:15   ` Sergey Fedorov
2016-05-17 16:23     ` Paolo Bonzini
2016-05-17 16:47       ` Sergey Fedorov
2016-05-17 17:08         ` Paolo Bonzini
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 06/18] atomics: add atomic_read_acquire and atomic_set_release Emilio G. Cota
2016-05-15 10:22   ` Pranith Kumar
2016-05-16 18:27     ` Emilio G. Cota
2016-05-17 16:53   ` Sergey Fedorov
2016-05-17 17:08     ` Paolo Bonzini
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 07/18] qemu-thread: add simple test-and-set spinlock Emilio G. Cota
     [not found]   ` <573B5134.8060104@gmail.com>
2016-05-17 19:19     ` Richard Henderson
2016-05-17 19:57       ` Sergey Fedorov
2016-05-17 20:01         ` Sergey Fedorov
2016-05-17 22:12           ` Richard Henderson
2016-05-17 22:22             ` Richard Henderson
2016-05-17 20:04       ` Emilio G. Cota
2016-05-17 20:20         ` Sergey Fedorov
2016-05-18  0:28           ` Emilio G. Cota
2016-05-18 14:18             ` Sergey Fedorov
2016-05-18 14:47               ` Sergey Fedorov
2016-05-18 14:59                 ` Paolo Bonzini
2016-05-18 15:05                   ` Sergey Fedorov
2016-05-18 15:09                     ` Paolo Bonzini
2016-05-18 16:59                       ` Emilio G. Cota
2016-05-18 17:00                         ` Paolo Bonzini
2016-05-18 15:35                     ` Peter Maydell
2016-05-18 15:36                       ` Paolo Bonzini
2016-05-18 15:44                         ` Peter Maydell
2016-05-18 15:59                           ` Sergey Fedorov
2016-05-18 16:02                       ` Richard Henderson
2016-05-17 19:38     ` Emilio G. Cota
2016-05-17 20:35       ` Sergey Fedorov
2016-05-17 23:18         ` Emilio G. Cota
2016-05-18 13:59           ` Sergey Fedorov
2016-05-18 14:05             ` Paolo Bonzini
2016-05-18 14:10               ` Sergey Fedorov
2016-05-18 14:40                 ` Paolo Bonzini
2016-05-18 18:21   ` Sergey Fedorov
2016-05-18 19:04     ` Emilio G. Cota
2016-05-18 19:51   ` Sergey Fedorov
2016-05-18 20:52     ` Emilio G. Cota
2016-05-18 20:57       ` Sergey Fedorov
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 08/18] exec: add tb_hash_func5, derived from xxhash Emilio G. Cota
2016-05-17 17:22   ` Sergey Fedorov [this message]
2016-05-17 19:48     ` Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 09/18] tb hash: hash phys_pc, pc, and flags with xxhash Emilio G. Cota
2016-05-17 17:47   ` Sergey Fedorov
2016-05-17 19:09     ` Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 10/18] qdist: add module to represent frequency distributions of data Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 11/18] qdist: add test program Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 12/18] qht: QEMU's fast, resizable and scalable Hash Table Emilio G. Cota
2016-05-20 22:13   ` Sergey Fedorov
2016-05-21  2:48     ` Emilio G. Cota
2016-05-21 17:41       ` Emilio G. Cota
2016-05-22  8:01         ` Alex Bennée
2016-05-23  5:35           ` Emilio G. Cota
2016-05-21 20:07       ` Sergey Fedorov
2016-05-23 19:29       ` Sergey Fedorov
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 13/18] qht: support parallel writes Emilio G. Cota
2016-05-23 20:28   ` Sergey Fedorov
2016-05-24 22:07     ` Emilio G. Cota
2016-05-24 22:17       ` Sergey Fedorov
2016-05-25  0:10         ` Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 14/18] qht: add test program Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 15/18] qht: add qht-bench, a performance benchmark Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 16/18] qht: add test-qht-par to invoke qht-bench from 'check' target Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 17/18] tb hash: track translated blocks with qht Emilio G. Cota
2016-05-14  3:34 ` [Qemu-devel] [PATCH v5 18/18] translate-all: add tb hash bucket info to 'info jit' dump Emilio G. Cota
2016-05-23 22:26 ` [Qemu-devel] [PATCH v5 00/18] tb hash improvements Sergey Fedorov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=573B536C.90804@gmail.com \
    --to=serge.fdrv@gmail.com \
    --cc=alex.bennee@linaro.org \
    --cc=cota@braap.org \
    --cc=crosthwaite.peter@gmail.com \
    --cc=mttcg@listserver.greensocs.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).