From: David Laight <David.Laight@ACULAB.COM>
To: 'Jeff Lien' <jeff.lien@wdc.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-crypto@vger.kernel.org" <linux-crypto@vger.kernel.org>,
"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
"linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>
Cc: "herbert@gondor.apana.org.au" <herbert@gondor.apana.org.au>,
"tim.c.chen@linux.intel.com" <tim.c.chen@linux.intel.com>,
"martin.petersen@oracle.com" <martin.petersen@oracle.com>,
"david.darrington@wdc.com" <david.darrington@wdc.com>,
"jeff.furlong@wdc.com" <jeff.furlong@wdc.com>
Subject: RE: [PATCH] Performance Improvement in CRC16 Calculations.
Date: Mon, 13 Aug 2018 11:45:21 +0000 [thread overview]
Message-ID: <8508c92d70dc41ffb66557e13c988dca@AcuMS.aculab.com> (raw)
In-Reply-To: <1533928331-21303-1-git-send-email-jeff.lien@wdc.com>
From: Jeff Lien
> Sent: 10 August 2018 20:12
>
> This patch provides a performance improvement for the CRC16 calculations done in read/write
> workloads using the T10 Type 1/2/3 guard field. For example, today with sequential write
> workloads (one thread/CPU of IO) we consume 100% of the CPU because of the CRC16 computation
> bottleneck. Today's block devices are considerably faster, but the CRC16 calculation prevents
> folks from utilizing the throughput of such devices. To speed up this calculation and expose
> the block device throughput, we slice the old single byte for loop into a 16 byte for loop,
> with a larger CRC table to match. The result has shown 5x performance improvements on various
> big endian and little endian systems running the 4.18.0 kernel version.
>
> FIO Sequential Write, 64K Block Size, Queue Depth 64
> BE Base Kernel: bw=201.5 MiB/s
> BE Modified CRC Calc: bw=968.1 MiB/s
> 4.80x performance improvement
>
> LE Base Kernel: bw=357 MiB/s
> LE Modified CRC Calc: bw=1964 MiB/s
> 5.51x performance improvement
>
> FIO Sequential Read, 64K Block Size, Queue Depth 64
> BE Base Kernel: bw=611.2 MiB/s
> BE Modified CRC calc: bw=684.9 MiB/s
> 1.12x performance improvement
>
> LE Base Kernel: bw=797 MiB/s
> LE Modified CRC Calc: bw=2730 MiB/s
> 3.42x performance improvement
>
> Reviewed-by: Dave Darrington <david.darrington@wdc.com>
> Reviewed-by: Jeff Furlong <jeff.furlong@wdc.com>
> Signed-off-by: Jeff Lien <jeff.lien@wdc.com>
> ---
> crypto/crct10dif_common.c | 605 +++++++++++++++++++++++++++++++++++++++++++---
> 1 file changed, 569 insertions(+), 36 deletions(-)
>
> diff --git a/crypto/crct10dif_common.c b/crypto/crct10dif_common.c
> index b2fab36..40e1d6c 100644
> --- a/crypto/crct10dif_common.c
> +++ b/crypto/crct10dif_common.c
> @@ -32,47 +32,580 @@
> * x^16 + x^15 + x^11 + x^9 + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1
> * gt: 0x8bb7
> */
> -static const __u16 t10_dif_crc_table[256] = {
...
> + }
> };
>
> __u16 crc_t10dif_generic(__u16 crc, const unsigned char *buffer, size_t len)
> {
> - unsigned int i;
> + const __u8 *i = (const __u8 *)buffer;
> + const __u8 *i_end = i + len;
> + const __u8 *i_last16 = i + (len / 16 * 16);
>
> - for (i = 0 ; i < len ; i++)
> - crc = (crc << 8) ^ t10_dif_crc_table[((crc >> 8) ^ buffer[i]) & 0xff];
> + for (; i < i_last16; i += 16) {
> + crc = t10_dif_crc_table[15][i[0] ^ (__u8)(crc >> 8)] ^
> + t10_dif_crc_table[14][i[1] ^ (__u8)(crc >> 0)] ^
> + t10_dif_crc_table[13][i[2]] ^
> + t10_dif_crc_table[12][i[3]] ^
> + t10_dif_crc_table[11][i[4]] ^
> + t10_dif_crc_table[10][i[5]] ^
> + t10_dif_crc_table[9][i[6]] ^
> + t10_dif_crc_table[8][i[7]] ^
> + t10_dif_crc_table[7][i[8]] ^
> + t10_dif_crc_table[6][i[9]] ^
> + t10_dif_crc_table[5][i[10]] ^
> + t10_dif_crc_table[4][i[11]] ^
> + t10_dif_crc_table[3][i[12]] ^
> + t10_dif_crc_table[2][i[13]] ^
> + t10_dif_crc_table[1][i[14]] ^
> + t10_dif_crc_table[0][i[15]];
...
I suspect that all the gain comes from a slight relaxation of the
register dependency chain and the loop unrolling.
A more interesting version would be to generate the lookup table
for a byte followed by 3 zero bytes.
You could then run four separate register dependency chains using the
same 256 entry lookup table.
A little bit of work at the end of the buffer should sort it all out.
There is also the lookup table free version:
uint32_t
crc_step(uint32_t crc, uint32_t byte_val)
{
uint32_t t = crc ^ (byte_val & 0xff);
t = (t ^ t << 4) & 0xff;
return crc >> 8 ^ t << 8 ^ t << 3 ^ t >> 4;
}
David
-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)
next prev parent reply other threads:[~2018-08-13 11:43 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-08-10 19:12 [PATCH] Performance Improvement in CRC16 Calculations Jeff Lien
2018-08-10 19:23 ` Joe Perches
2018-08-10 20:02 ` Nicolas Pitre
2018-08-11 0:11 ` Joe Perches
2018-08-11 0:34 ` Nicolas Pitre
2018-08-11 2:39 ` Douglas Gilbert
2018-08-11 9:04 ` Joe Perches
2018-08-11 15:06 ` Joe Perches
2018-08-13 18:41 ` Jeffrey Lien
2018-08-13 3:36 ` Douglas Gilbert
2018-08-13 4:29 ` Joe Perches
2018-08-10 20:00 ` Nicolas Pitre
2018-08-10 20:16 ` Eric Biggers
2018-08-16 14:02 ` Jeffrey Lien
2018-08-16 14:22 ` Douglas Gilbert
2018-08-16 15:41 ` Christophe LEROY
2018-08-16 17:38 ` Douglas Gilbert
2018-08-17 3:20 ` Martin K. Petersen
2018-08-16 15:47 ` Christophe LEROY
2018-08-10 20:56 ` Douglas Gilbert
2018-08-11 15:36 ` Martin K. Petersen
2018-08-11 16:35 ` Joe Perches
2018-08-22 1:40 ` Martin K. Petersen
2018-08-22 6:20 ` Christoph Hellwig
2018-08-24 15:32 ` Jeffrey Lien
2018-08-24 15:39 ` Ard Biesheuvel
2018-08-24 16:29 ` Martin K. Petersen
2018-08-24 17:38 ` Ard Biesheuvel
2018-08-24 21:46 ` Martin K. Petersen
2018-08-24 21:54 ` Ard Biesheuvel
2018-08-24 22:12 ` Martin K. Petersen
2018-08-25 6:12 ` Herbert Xu
2018-08-26 2:35 ` Martin K. Petersen
2018-08-26 2:40 ` [PATCH 1/4] crypto: Introduce notifier for new crypto algorithms Martin K. Petersen
2018-08-26 2:40 ` [PATCH 2/4] crc-t10dif: Pick better transform if one becomes available Martin K. Petersen
2018-08-27 6:13 ` Herbert Xu
2018-08-26 2:40 ` [PATCH 3/4] crc-t10dif: Allow current transform to be inspected in sysfs Martin K. Petersen
2018-08-26 2:40 ` [PATCH 4/4] block: Integrity profile init function to trigger module loads Martin K. Petersen
2018-08-26 8:22 ` Ard Biesheuvel
2018-08-26 13:30 ` Martin K. Petersen
2018-08-26 13:44 ` Ard Biesheuvel
2018-08-26 13:48 ` Martin K. Petersen
2018-08-27 6:09 ` [PATCH 1/4] crypto: Introduce notifier for new crypto algorithms Herbert Xu
2018-08-30 14:57 ` Martin K. Petersen
2018-08-30 15:00 ` [PATCH v2 1/3] " Martin K. Petersen
2018-08-30 15:00 ` [PATCH v2 2/3] crc-t10dif: Pick better transform if one becomes available Martin K. Petersen
2018-08-30 15:00 ` [PATCH v2 3/3] crc-t10dif: Allow current transform to be inspected in sysfs Martin K. Petersen
2018-08-31 17:17 ` [PATCH v2 1/3] crypto: Introduce notifier for new crypto algorithms Jeffrey Lien
2018-09-04 5:21 ` Herbert Xu
2018-09-04 13:30 ` Torsten Duwe
2018-08-24 16:30 ` [PATCH] Performance Improvement in CRC16 Calculations Martin K. Petersen
2018-08-13 4:44 ` Chaitanya Kulkarni
2018-08-13 11:45 ` David Laight [this message]
2018-08-13 13:50 ` David Laight
2018-08-13 22:44 ` Tim Chen
2018-08-15 12:51 ` Jeffrey Lien
2018-08-15 18:31 ` Pavel Machek
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=8508c92d70dc41ffb66557e13c988dca@AcuMS.aculab.com \
--to=david.laight@aculab.com \
--cc=david.darrington@wdc.com \
--cc=herbert@gondor.apana.org.au \
--cc=jeff.furlong@wdc.com \
--cc=jeff.lien@wdc.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-crypto@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=tim.c.chen@linux.intel.com \
/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).