lkml.org 
[lkml]   [2018]   [Aug]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH] Performance Improvement in CRC16 Calculations.
From
Date
On 2018-08-16 11:41 AM, Christophe LEROY wrote:
> Hi,
>
> Please include your new patch as plain text inside the mail, not as a MIME
> attachment. Otherwise it is not downloadable from
> https://patchwork.kernel.org/patch/10563093/

It should be downloadable from:
http://sg.danny.cz/sg/p/0001-T10-CRC16-function-build-time-sized-table.patch

With regard to your comment about slice (table ?) size, that is partially
addressed by a kernel build time option shown in the above patch. That
could be taken a bit further with a sysfs knob (where ?) to reduce the
effective table size from that which the kernel is built with. To increase
the size of the table would imply fetching some more heap and having an
algorithm that could generate the extra part of that table required.

Doug Gilbert

> Christophe
>
> Le 16/08/2018 à 16:22, Douglas Gilbert a écrit :
>> Hi,
>> Rather than present this formerly as an alternate patch, attached is a
>> clean-up of my patch which uses the variable size table proposed by
>> Joe Perches <joe@perches.com> and is based on the original patch that
>> started this thread.
>>
>> Doug Gilbert
>>
>> On 2018-08-16 10:02 AM, Jeffrey Lien wrote:
>>> Eric,
>>> We did not test the slice by 4 or 8 tables.  I'm not sure of  the value of
>>> doing that since the slice by 16 will provide the best performance gain.   If
>>> I'm missing anything here, please let me know.
>>>
>>> I'm working on a new version of the patch based on the feedback from others
>>> and will also change the pointer variables to start with p and fix the
>>> indenting you mentioned below in the new version of the patch.
>>>
>>> Thanks
>>>
>>> Jeff Lien
>>>
>>> -----Original Message-----
>>> From: Eric Biggers [mailto:ebiggers@kernel.org]
>>> Sent: Friday, August 10, 2018 3:16 PM
>>> To: Jeffrey Lien <Jeff.Lien@wdc.com>
>>> Cc: linux-kernel@vger.kernel.org; linux-crypto@vger.kernel.org;
>>> linux-block@vger.kernel.org; linux-scsi@vger.kernel.org;
>>> herbert@gondor.apana.org.au; tim.c.chen@linux.intel.com;
>>> martin.petersen@oracle.com; David Darrington <david.darrington@wdc.com>; Jeff
>>> Furlong <jeff.furlong@wdc.com>
>>> Subject: Re: [PATCH] Performance Improvement in CRC16 Calculations.
>>>
>>> On Fri, Aug 10, 2018 at 02:12:11PM -0500, Jeff Lien wrote:
>>>> 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
>>>
>>> Did you also test the slice-by-4 (requires 2048-byte table) and slice-by-8
>>> (requires 4096-byte table) methods?  Your proposal is slice-by-16 (requires
>>> 8192-byte table); the original was slice-by-1 (requires 512-byte table).
>>>
>>>>   __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);
>>>
>>> 'i' is normally a loop counter, not a pointer.
>>> Use 'p', 'p_end', and 'p_last16'.
>>>
>>>> -    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]];
>>>> +    }
>>>
>>> Please indent this properly.
>>>
>>>         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]] ^
>>>               ...
>>>
>>> - Eric
>>>
>>
>

\
 
 \ /
  Last update: 2018-08-16 19:39    [W:0.129 / U:0.500 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site