linux/crypto/cryptd.c

659 lines
17 KiB
C
Raw Normal View History

/*
* Software async crypto daemon.
*
* Copyright (c) 2006 Herbert Xu <herbert@gondor.apana.org.au>
*
* This program is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License as published by the Free
* Software Foundation; either version 2 of the License, or (at your option)
* any later version.
*
*/
#include <crypto/algapi.h>
#include <crypto/internal/hash.h>
#include <crypto/cryptd.h>
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
#include <crypto/crypto_wq.h>
#include <linux/err.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/list.h>
#include <linux/module.h>
#include <linux/scatterlist.h>
#include <linux/sched.h>
#include <linux/slab.h>
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
#define CRYPTD_MAX_CPU_QLEN 100
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
struct cryptd_cpu_queue {
struct crypto_queue queue;
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
struct work_struct work;
};
struct cryptd_queue {
struct cryptd_cpu_queue *cpu_queue;
};
struct cryptd_instance_ctx {
struct crypto_spawn spawn;
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
struct cryptd_queue *queue;
};
struct hashd_instance_ctx {
struct crypto_shash_spawn spawn;
struct cryptd_queue *queue;
};
struct cryptd_blkcipher_ctx {
struct crypto_blkcipher *child;
};
struct cryptd_blkcipher_request_ctx {
crypto_completion_t complete;
};
struct cryptd_hash_ctx {
struct crypto_shash *child;
};
struct cryptd_hash_request_ctx {
crypto_completion_t complete;
struct shash_desc desc;
};
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
static void cryptd_queue_worker(struct work_struct *work);
static int cryptd_init_queue(struct cryptd_queue *queue,
unsigned int max_cpu_qlen)
{
int cpu;
struct cryptd_cpu_queue *cpu_queue;
queue->cpu_queue = alloc_percpu(struct cryptd_cpu_queue);
if (!queue->cpu_queue)
return -ENOMEM;
for_each_possible_cpu(cpu) {
cpu_queue = per_cpu_ptr(queue->cpu_queue, cpu);
crypto_init_queue(&cpu_queue->queue, max_cpu_qlen);
INIT_WORK(&cpu_queue->work, cryptd_queue_worker);
}
return 0;
}
static void cryptd_fini_queue(struct cryptd_queue *queue)
{
int cpu;
struct cryptd_cpu_queue *cpu_queue;
for_each_possible_cpu(cpu) {
cpu_queue = per_cpu_ptr(queue->cpu_queue, cpu);
BUG_ON(cpu_queue->queue.qlen);
}
free_percpu(queue->cpu_queue);
}
static int cryptd_enqueue_request(struct cryptd_queue *queue,
struct crypto_async_request *request)
{
int cpu, err;
struct cryptd_cpu_queue *cpu_queue;
cpu = get_cpu();
cpu_queue = per_cpu_ptr(queue->cpu_queue, cpu);
err = crypto_enqueue_request(&cpu_queue->queue, request);
queue_work_on(cpu, kcrypto_wq, &cpu_queue->work);
put_cpu();
return err;
}
/* Called in workqueue context, do one real cryption work (via
* req->complete) and reschedule itself if there are more work to
* do. */
static void cryptd_queue_worker(struct work_struct *work)
{
struct cryptd_cpu_queue *cpu_queue;
struct crypto_async_request *req, *backlog;
cpu_queue = container_of(work, struct cryptd_cpu_queue, work);
/* Only handle one request at a time to avoid hogging crypto
* workqueue. preempt_disable/enable is used to prevent
* being preempted by cryptd_enqueue_request() */
preempt_disable();
backlog = crypto_get_backlog(&cpu_queue->queue);
req = crypto_dequeue_request(&cpu_queue->queue);
preempt_enable();
if (!req)
return;
if (backlog)
backlog->complete(backlog, -EINPROGRESS);
req->complete(req, 0);
if (cpu_queue->queue.qlen)
queue_work(kcrypto_wq, &cpu_queue->work);
}
static inline struct cryptd_queue *cryptd_get_queue(struct crypto_tfm *tfm)
{
struct crypto_instance *inst = crypto_tfm_alg_instance(tfm);
struct cryptd_instance_ctx *ictx = crypto_instance_ctx(inst);
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
return ictx->queue;
}
static int cryptd_blkcipher_setkey(struct crypto_ablkcipher *parent,
const u8 *key, unsigned int keylen)
{
struct cryptd_blkcipher_ctx *ctx = crypto_ablkcipher_ctx(parent);
struct crypto_blkcipher *child = ctx->child;
int err;
crypto_blkcipher_clear_flags(child, CRYPTO_TFM_REQ_MASK);
crypto_blkcipher_set_flags(child, crypto_ablkcipher_get_flags(parent) &
CRYPTO_TFM_REQ_MASK);
err = crypto_blkcipher_setkey(child, key, keylen);
crypto_ablkcipher_set_flags(parent, crypto_blkcipher_get_flags(child) &
CRYPTO_TFM_RES_MASK);
return err;
}
static void cryptd_blkcipher_crypt(struct ablkcipher_request *req,
struct crypto_blkcipher *child,
int err,
int (*crypt)(struct blkcipher_desc *desc,
struct scatterlist *dst,
struct scatterlist *src,
unsigned int len))
{
struct cryptd_blkcipher_request_ctx *rctx;
struct blkcipher_desc desc;
rctx = ablkcipher_request_ctx(req);
if (unlikely(err == -EINPROGRESS))
goto out;
desc.tfm = child;
desc.info = req->info;
desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP;
err = crypt(&desc, req->dst, req->src, req->nbytes);
req->base.complete = rctx->complete;
out:
local_bh_disable();
rctx->complete(&req->base, err);
local_bh_enable();
}
static void cryptd_blkcipher_encrypt(struct crypto_async_request *req, int err)
{
struct cryptd_blkcipher_ctx *ctx = crypto_tfm_ctx(req->tfm);
struct crypto_blkcipher *child = ctx->child;
cryptd_blkcipher_crypt(ablkcipher_request_cast(req), child, err,
crypto_blkcipher_crt(child)->encrypt);
}
static void cryptd_blkcipher_decrypt(struct crypto_async_request *req, int err)
{
struct cryptd_blkcipher_ctx *ctx = crypto_tfm_ctx(req->tfm);
struct crypto_blkcipher *child = ctx->child;
cryptd_blkcipher_crypt(ablkcipher_request_cast(req), child, err,
crypto_blkcipher_crt(child)->decrypt);
}
static int cryptd_blkcipher_enqueue(struct ablkcipher_request *req,
crypto_completion_t complete)
{
struct cryptd_blkcipher_request_ctx *rctx = ablkcipher_request_ctx(req);
struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
struct cryptd_queue *queue;
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
queue = cryptd_get_queue(crypto_ablkcipher_tfm(tfm));
rctx->complete = req->base.complete;
req->base.complete = complete;
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
return cryptd_enqueue_request(queue, &req->base);
}
static int cryptd_blkcipher_encrypt_enqueue(struct ablkcipher_request *req)
{
return cryptd_blkcipher_enqueue(req, cryptd_blkcipher_encrypt);
}
static int cryptd_blkcipher_decrypt_enqueue(struct ablkcipher_request *req)
{
return cryptd_blkcipher_enqueue(req, cryptd_blkcipher_decrypt);
}
static int cryptd_blkcipher_init_tfm(struct crypto_tfm *tfm)
{
struct crypto_instance *inst = crypto_tfm_alg_instance(tfm);
struct cryptd_instance_ctx *ictx = crypto_instance_ctx(inst);
struct crypto_spawn *spawn = &ictx->spawn;
struct cryptd_blkcipher_ctx *ctx = crypto_tfm_ctx(tfm);
struct crypto_blkcipher *cipher;
cipher = crypto_spawn_blkcipher(spawn);
if (IS_ERR(cipher))
return PTR_ERR(cipher);
ctx->child = cipher;
tfm->crt_ablkcipher.reqsize =
sizeof(struct cryptd_blkcipher_request_ctx);
return 0;
}
static void cryptd_blkcipher_exit_tfm(struct crypto_tfm *tfm)
{
struct cryptd_blkcipher_ctx *ctx = crypto_tfm_ctx(tfm);
crypto_free_blkcipher(ctx->child);
}
static struct crypto_instance *cryptd_alloc_instance(struct crypto_alg *alg,
unsigned int tail)
{
struct crypto_instance *inst;
int err;
inst = kzalloc(sizeof(*inst) + tail, GFP_KERNEL);
if (!inst) {
inst = ERR_PTR(-ENOMEM);
goto out;
}
err = -ENAMETOOLONG;
if (snprintf(inst->alg.cra_driver_name, CRYPTO_MAX_ALG_NAME,
"cryptd(%s)", alg->cra_driver_name) >= CRYPTO_MAX_ALG_NAME)
goto out_free_inst;
memcpy(inst->alg.cra_name, alg->cra_name, CRYPTO_MAX_ALG_NAME);
inst->alg.cra_priority = alg->cra_priority + 50;
inst->alg.cra_blocksize = alg->cra_blocksize;
inst->alg.cra_alignmask = alg->cra_alignmask;
out:
return inst;
out_free_inst:
kfree(inst);
inst = ERR_PTR(err);
goto out;
}
static int cryptd_create_blkcipher(struct crypto_template *tmpl,
struct rtattr **tb,
struct cryptd_queue *queue)
{
struct cryptd_instance_ctx *ctx;
struct crypto_instance *inst;
struct crypto_alg *alg;
int err;
alg = crypto_get_attr_alg(tb, CRYPTO_ALG_TYPE_BLKCIPHER,
CRYPTO_ALG_TYPE_MASK);
if (IS_ERR(alg))
return PTR_ERR(alg);
inst = cryptd_alloc_instance(alg, sizeof(*ctx));
if (IS_ERR(inst))
goto out_put_alg;
ctx = crypto_instance_ctx(inst);
ctx->queue = queue;
err = crypto_init_spawn(&ctx->spawn, alg, inst,
CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_ASYNC);
if (err)
goto out_free_inst;
inst->alg.cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC;
inst->alg.cra_type = &crypto_ablkcipher_type;
inst->alg.cra_ablkcipher.ivsize = alg->cra_blkcipher.ivsize;
inst->alg.cra_ablkcipher.min_keysize = alg->cra_blkcipher.min_keysize;
inst->alg.cra_ablkcipher.max_keysize = alg->cra_blkcipher.max_keysize;
inst->alg.cra_ablkcipher.geniv = alg->cra_blkcipher.geniv;
inst->alg.cra_ctxsize = sizeof(struct cryptd_blkcipher_ctx);
inst->alg.cra_init = cryptd_blkcipher_init_tfm;
inst->alg.cra_exit = cryptd_blkcipher_exit_tfm;
inst->alg.cra_ablkcipher.setkey = cryptd_blkcipher_setkey;
inst->alg.cra_ablkcipher.encrypt = cryptd_blkcipher_encrypt_enqueue;
inst->alg.cra_ablkcipher.decrypt = cryptd_blkcipher_decrypt_enqueue;
err = crypto_register_instance(tmpl, inst);
if (err) {
crypto_drop_spawn(&ctx->spawn);
out_free_inst:
kfree(inst);
}
out_put_alg:
crypto_mod_put(alg);
return err;
}
static int cryptd_hash_init_tfm(struct crypto_tfm *tfm)
{
struct crypto_instance *inst = crypto_tfm_alg_instance(tfm);
struct hashd_instance_ctx *ictx = crypto_instance_ctx(inst);
struct crypto_shash_spawn *spawn = &ictx->spawn;
struct cryptd_hash_ctx *ctx = crypto_tfm_ctx(tfm);
struct crypto_shash *hash;
hash = crypto_spawn_shash(spawn);
if (IS_ERR(hash))
return PTR_ERR(hash);
ctx->child = hash;
crypto_ahash_set_reqsize(__crypto_ahash_cast(tfm),
sizeof(struct cryptd_hash_request_ctx) +
crypto_shash_descsize(hash));
return 0;
}
static void cryptd_hash_exit_tfm(struct crypto_tfm *tfm)
{
struct cryptd_hash_ctx *ctx = crypto_tfm_ctx(tfm);
crypto_free_shash(ctx->child);
}
static int cryptd_hash_setkey(struct crypto_ahash *parent,
const u8 *key, unsigned int keylen)
{
struct cryptd_hash_ctx *ctx = crypto_ahash_ctx(parent);
struct crypto_shash *child = ctx->child;
int err;
crypto_shash_clear_flags(child, CRYPTO_TFM_REQ_MASK);
crypto_shash_set_flags(child, crypto_ahash_get_flags(parent) &
CRYPTO_TFM_REQ_MASK);
err = crypto_shash_setkey(child, key, keylen);
crypto_ahash_set_flags(parent, crypto_shash_get_flags(child) &
CRYPTO_TFM_RES_MASK);
return err;
}
static int cryptd_hash_enqueue(struct ahash_request *req,
crypto_completion_t complete)
{
struct cryptd_hash_request_ctx *rctx = ahash_request_ctx(req);
struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
struct cryptd_queue *queue =
cryptd_get_queue(crypto_ahash_tfm(tfm));
rctx->complete = req->base.complete;
req->base.complete = complete;
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
return cryptd_enqueue_request(queue, &req->base);
}
static void cryptd_hash_init(struct crypto_async_request *req_async, int err)
{
struct cryptd_hash_ctx *ctx = crypto_tfm_ctx(req_async->tfm);
struct crypto_shash *child = ctx->child;
struct ahash_request *req = ahash_request_cast(req_async);
struct cryptd_hash_request_ctx *rctx = ahash_request_ctx(req);
struct shash_desc *desc = &rctx->desc;
if (unlikely(err == -EINPROGRESS))
goto out;
desc->tfm = child;
desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP;
err = crypto_shash_init(desc);
req->base.complete = rctx->complete;
out:
local_bh_disable();
rctx->complete(&req->base, err);
local_bh_enable();
}
static int cryptd_hash_init_enqueue(struct ahash_request *req)
{
return cryptd_hash_enqueue(req, cryptd_hash_init);
}
static void cryptd_hash_update(struct crypto_async_request *req_async, int err)
{
struct ahash_request *req = ahash_request_cast(req_async);
struct cryptd_hash_request_ctx *rctx;
rctx = ahash_request_ctx(req);
if (unlikely(err == -EINPROGRESS))
goto out;
err = shash_ahash_update(req, &rctx->desc);
req->base.complete = rctx->complete;
out:
local_bh_disable();
rctx->complete(&req->base, err);
local_bh_enable();
}
static int cryptd_hash_update_enqueue(struct ahash_request *req)
{
return cryptd_hash_enqueue(req, cryptd_hash_update);
}
static void cryptd_hash_final(struct crypto_async_request *req_async, int err)
{
struct ahash_request *req = ahash_request_cast(req_async);
struct cryptd_hash_request_ctx *rctx = ahash_request_ctx(req);
if (unlikely(err == -EINPROGRESS))
goto out;
err = crypto_shash_final(&rctx->desc, req->result);
req->base.complete = rctx->complete;
out:
local_bh_disable();
rctx->complete(&req->base, err);
local_bh_enable();
}
static int cryptd_hash_final_enqueue(struct ahash_request *req)
{
return cryptd_hash_enqueue(req, cryptd_hash_final);
}
static void cryptd_hash_digest(struct crypto_async_request *req_async, int err)
{
struct cryptd_hash_ctx *ctx = crypto_tfm_ctx(req_async->tfm);
struct crypto_shash *child = ctx->child;
struct ahash_request *req = ahash_request_cast(req_async);
struct cryptd_hash_request_ctx *rctx = ahash_request_ctx(req);
struct shash_desc *desc = &rctx->desc;
if (unlikely(err == -EINPROGRESS))
goto out;
desc->tfm = child;
desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP;
err = shash_ahash_digest(req, desc);
req->base.complete = rctx->complete;
out:
local_bh_disable();
rctx->complete(&req->base, err);
local_bh_enable();
}
static int cryptd_hash_digest_enqueue(struct ahash_request *req)
{
return cryptd_hash_enqueue(req, cryptd_hash_digest);
}
static int cryptd_create_hash(struct crypto_template *tmpl, struct rtattr **tb,
struct cryptd_queue *queue)
{
struct hashd_instance_ctx *ctx;
struct crypto_instance *inst;
struct shash_alg *salg;
struct crypto_alg *alg;
int err;
salg = shash_attr_alg(tb[1], 0, 0);
if (IS_ERR(salg))
return PTR_ERR(salg);
alg = &salg->base;
inst = cryptd_alloc_instance(alg, sizeof(*ctx));
if (IS_ERR(inst))
goto out_put_alg;
ctx = crypto_instance_ctx(inst);
ctx->queue = queue;
err = crypto_init_shash_spawn(&ctx->spawn, salg, inst);
if (err)
goto out_free_inst;
inst->alg.cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_ASYNC;
inst->alg.cra_type = &crypto_ahash_type;
inst->alg.cra_ahash.digestsize = salg->digestsize;
inst->alg.cra_ctxsize = sizeof(struct cryptd_hash_ctx);
inst->alg.cra_init = cryptd_hash_init_tfm;
inst->alg.cra_exit = cryptd_hash_exit_tfm;
inst->alg.cra_ahash.init = cryptd_hash_init_enqueue;
inst->alg.cra_ahash.update = cryptd_hash_update_enqueue;
inst->alg.cra_ahash.final = cryptd_hash_final_enqueue;
inst->alg.cra_ahash.setkey = cryptd_hash_setkey;
inst->alg.cra_ahash.digest = cryptd_hash_digest_enqueue;
err = crypto_register_instance(tmpl, inst);
if (err) {
crypto_drop_shash(&ctx->spawn);
out_free_inst:
kfree(inst);
}
out_put_alg:
crypto_mod_put(alg);
return err;
}
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
static struct cryptd_queue queue;
static int cryptd_create(struct crypto_template *tmpl, struct rtattr **tb)
{
struct crypto_attr_type *algt;
algt = crypto_get_attr_type(tb);
if (IS_ERR(algt))
return PTR_ERR(algt);
switch (algt->type & algt->mask & CRYPTO_ALG_TYPE_MASK) {
case CRYPTO_ALG_TYPE_BLKCIPHER:
return cryptd_create_blkcipher(tmpl, tb, &queue);
case CRYPTO_ALG_TYPE_DIGEST:
return cryptd_create_hash(tmpl, tb, &queue);
}
return -EINVAL;
}
static void cryptd_free(struct crypto_instance *inst)
{
struct cryptd_instance_ctx *ctx = crypto_instance_ctx(inst);
crypto_drop_spawn(&ctx->spawn);
kfree(inst);
}
static struct crypto_template cryptd_tmpl = {
.name = "cryptd",
.create = cryptd_create,
.free = cryptd_free,
.module = THIS_MODULE,
};
struct cryptd_ablkcipher *cryptd_alloc_ablkcipher(const char *alg_name,
u32 type, u32 mask)
{
char cryptd_alg_name[CRYPTO_MAX_ALG_NAME];
struct crypto_tfm *tfm;
if (snprintf(cryptd_alg_name, CRYPTO_MAX_ALG_NAME,
"cryptd(%s)", alg_name) >= CRYPTO_MAX_ALG_NAME)
return ERR_PTR(-EINVAL);
type &= ~(CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_GENIV);
type |= CRYPTO_ALG_TYPE_BLKCIPHER;
mask &= ~CRYPTO_ALG_TYPE_MASK;
mask |= (CRYPTO_ALG_GENIV | CRYPTO_ALG_TYPE_BLKCIPHER_MASK);
tfm = crypto_alloc_base(cryptd_alg_name, type, mask);
if (IS_ERR(tfm))
return ERR_CAST(tfm);
if (tfm->__crt_alg->cra_module != THIS_MODULE) {
crypto_free_tfm(tfm);
return ERR_PTR(-EINVAL);
}
return __cryptd_ablkcipher_cast(__crypto_ablkcipher_cast(tfm));
}
EXPORT_SYMBOL_GPL(cryptd_alloc_ablkcipher);
struct crypto_blkcipher *cryptd_ablkcipher_child(struct cryptd_ablkcipher *tfm)
{
struct cryptd_blkcipher_ctx *ctx = crypto_ablkcipher_ctx(&tfm->base);
return ctx->child;
}
EXPORT_SYMBOL_GPL(cryptd_ablkcipher_child);
void cryptd_free_ablkcipher(struct cryptd_ablkcipher *tfm)
{
crypto_free_ablkcipher(&tfm->base);
}
EXPORT_SYMBOL_GPL(cryptd_free_ablkcipher);
static int __init cryptd_init(void)
{
int err;
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
err = cryptd_init_queue(&queue, CRYPTD_MAX_CPU_QLEN);
if (err)
return err;
err = crypto_register_template(&cryptd_tmpl);
if (err)
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
cryptd_fini_queue(&queue);
return err;
}
static void __exit cryptd_exit(void)
{
crypto: cryptd - Per-CPU thread implementation based on kcrypto_wq Original cryptd thread implementation has scalability issue, this patch solve the issue with a per-CPU thread implementation. struct cryptd_queue is defined to be a per-CPU queue, which holds one struct cryptd_cpu_queue for each CPU. In struct cryptd_cpu_queue, a struct crypto_queue holds all requests for the CPU, a struct work_struct is used to run all requests for the CPU. Testing based on dm-crypt on an Intel Core 2 E6400 (two cores) machine shows 19.2% performance gain. The testing script is as follow: -------------------- script begin --------------------------- #!/bin/sh dmc_create() { # Create a crypt device using dmsetup dmsetup create $2 --table "0 `blockdev --getsize $1` crypt cbc(aes-asm)?cryptd?plain:plain babebabebabebabebabebabebabebabe 0 $1 0" } dmsetup remove crypt0 dmsetup remove crypt1 dd if=/dev/zero of=/dev/ram0 bs=1M count=4 >& /dev/null dd if=/dev/zero of=/dev/ram1 bs=1M count=4 >& /dev/null dmc_create /dev/ram0 crypt0 dmc_create /dev/ram1 crypt1 cat >tr.sh <<EOF #!/bin/sh for n in \$(seq 10); do dd if=/dev/dm-0 of=/dev/null >& /dev/null & dd if=/dev/dm-1 of=/dev/null >& /dev/null & done wait EOF for n in $(seq 10); do /usr/bin/time sh tr.sh done rm tr.sh -------------------- script end --------------------------- The separator of dm-crypt parameter is changed from "-" to "?", because "-" is used in some cipher driver name too, and cryptds need to specify cipher driver name instead of cipher name. The test result on an Intel Core2 E6400 (two cores) is as follow: without patch: -----------------wo begin -------------------------- 0.04user 0.38system 0:00.39elapsed 107%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6566minor)pagefaults 0swaps 0.07user 0.35system 0:00.35elapsed 121%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6567minor)pagefaults 0swaps 0.06user 0.34system 0:00.30elapsed 135%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.36elapsed 119%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6607minor)pagefaults 0swaps 0.06user 0.36system 0:00.35elapsed 120%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.37system 0:00.31elapsed 136%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.34system 0:00.30elapsed 126%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6597minor)pagefaults 0swaps 0.06user 0.32system 0:00.31elapsed 125%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6571minor)pagefaults 0swaps 0.06user 0.34system 0:00.31elapsed 134%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6581minor)pagefaults 0swaps 0.05user 0.38system 0:00.31elapsed 138%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6600minor)pagefaults 0swaps -----------------wo end -------------------------- with patch: ------------------w begin -------------------------- 0.02user 0.31system 0:00.24elapsed 141%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6554minor)pagefaults 0swaps 0.05user 0.34system 0:00.31elapsed 127%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6606minor)pagefaults 0swaps 0.07user 0.33system 0:00.26elapsed 155%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6559minor)pagefaults 0swaps 0.07user 0.32system 0:00.26elapsed 151%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.05user 0.34system 0:00.26elapsed 150%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6603minor)pagefaults 0swaps 0.03user 0.36system 0:00.31elapsed 124%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6586minor)pagefaults 0swaps 0.03user 0.37system 0:00.27elapsed 146%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6562minor)pagefaults 0swaps 0.04user 0.36system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6594minor)pagefaults 0swaps 0.04user 0.35system 0:00.26elapsed 154%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+6557minor)pagefaults 0swaps ------------------w end -------------------------- The middle value of elapsed time is: wo cryptwq: 0.31 w cryptwq: 0.26 The performance gain is about (0.31-0.26)/0.26 = 0.192. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2009-02-19 06:42:19 +00:00
cryptd_fini_queue(&queue);
crypto_unregister_template(&cryptd_tmpl);
}
module_init(cryptd_init);
module_exit(cryptd_exit);
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("Software async crypto daemon");