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@xxxxxxxxxxx> 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@xxxxxxxxxx]
Sent: Friday, August 10, 2018 3:16 PM
To: Jeffrey Lien <Jeff.Lien@xxxxxxx>
Cc: linux-kernel@xxxxxxxxxxxxxxx; linux-crypto@xxxxxxxxxxxxxxx;
linux-block@xxxxxxxxxxxxxxx; linux-scsi@xxxxxxxxxxxxxxx;
herbert@xxxxxxxxxxxxxxxxxxx; tim.c.chen@xxxxxxxxxxxxxxx;
martin.petersen@xxxxxxxxxx; David Darrington <david.darrington@xxxxxxx>; Jeff
Furlong <jeff.furlong@xxxxxxx>
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