diff options
Diffstat (limited to 'drivers')
-rw-r--r-- | drivers/char/hw_random/Kconfig | 14 | ||||
-rw-r--r-- | drivers/char/hw_random/Makefile | 1 | ||||
-rw-r--r-- | drivers/char/hw_random/timeriomem-rng.c | 151 | ||||
-rw-r--r-- | drivers/char/tpm/tpm.c | 530 | ||||
-rw-r--r-- | drivers/char/tpm/tpm.h | 142 | ||||
-rw-r--r-- | drivers/cpufreq/cpufreq.c | 55 | ||||
-rw-r--r-- | drivers/cpufreq/cpufreq_conservative.c | 404 | ||||
-rw-r--r-- | drivers/cpufreq/cpufreq_ondemand.c | 74 | ||||
-rw-r--r-- | drivers/cpufreq/cpufreq_stats.c | 74 | ||||
-rw-r--r-- | drivers/cpufreq/cpufreq_userspace.c | 27 | ||||
-rw-r--r-- | drivers/cpufreq/freq_table.c | 18 | ||||
-rw-r--r-- | drivers/crypto/Kconfig | 15 | ||||
-rw-r--r-- | drivers/crypto/Makefile | 1 | ||||
-rw-r--r-- | drivers/crypto/amcc/Makefile | 2 | ||||
-rw-r--r-- | drivers/crypto/amcc/crypto4xx_alg.c | 293 | ||||
-rw-r--r-- | drivers/crypto/amcc/crypto4xx_core.c | 1310 | ||||
-rw-r--r-- | drivers/crypto/amcc/crypto4xx_core.h | 177 | ||||
-rw-r--r-- | drivers/crypto/amcc/crypto4xx_reg_def.h | 284 | ||||
-rw-r--r-- | drivers/crypto/amcc/crypto4xx_sa.c | 108 | ||||
-rw-r--r-- | drivers/crypto/amcc/crypto4xx_sa.h | 243 |
20 files changed, 3365 insertions, 558 deletions
diff --git a/drivers/char/hw_random/Kconfig b/drivers/char/hw_random/Kconfig index 8822eca58ff..5fab6470f4b 100644 --- a/drivers/char/hw_random/Kconfig +++ b/drivers/char/hw_random/Kconfig @@ -20,6 +20,20 @@ config HW_RANDOM If unsure, say Y. +config HW_RANDOM_TIMERIOMEM + tristate "Timer IOMEM HW Random Number Generator support" + depends on HW_RANDOM && HAS_IOMEM + ---help--- + This driver provides kernel-side support for a generic Random + Number Generator used by reading a 'dumb' iomem address that + is to be read no faster than, for example, once a second; + the default FPGA bitstream on the TS-7800 has such functionality. + + To compile this driver as a module, choose M here: the + module will be called timeriomem-rng. + + If unsure, say Y. + config HW_RANDOM_INTEL tristate "Intel HW Random Number Generator support" depends on HW_RANDOM && (X86 || IA64) && PCI diff --git a/drivers/char/hw_random/Makefile b/drivers/char/hw_random/Makefile index b6effb7522c..e81d21a5f28 100644 --- a/drivers/char/hw_random/Makefile +++ b/drivers/char/hw_random/Makefile @@ -4,6 +4,7 @@ obj-$(CONFIG_HW_RANDOM) += rng-core.o rng-core-y := core.o +obj-$(CONFIG_HW_RANDOM_TIMERIOMEM) += timeriomem-rng.o obj-$(CONFIG_HW_RANDOM_INTEL) += intel-rng.o obj-$(CONFIG_HW_RANDOM_AMD) += amd-rng.o obj-$(CONFIG_HW_RANDOM_GEODE) += geode-rng.o diff --git a/drivers/char/hw_random/timeriomem-rng.c b/drivers/char/hw_random/timeriomem-rng.c new file mode 100644 index 00000000000..10ad41be589 --- /dev/null +++ b/drivers/char/hw_random/timeriomem-rng.c @@ -0,0 +1,151 @@ +/* + * drivers/char/hw_random/timeriomem-rng.c + * + * Copyright (C) 2009 Alexander Clouter <alex@digriz.org.uk> + * + * Derived from drivers/char/hw_random/omap-rng.c + * Copyright 2005 (c) MontaVista Software, Inc. + * Author: Deepak Saxena <dsaxena@plexity.net> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Overview: + * This driver is useful for platforms that have an IO range that provides + * periodic random data from a single IO memory address. All the platform + * has to do is provide the address and 'wait time' that new data becomes + * available. + * + * TODO: add support for reading sizes other than 32bits and masking + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/platform_device.h> +#include <linux/hw_random.h> +#include <linux/io.h> +#include <linux/timeriomem-rng.h> +#include <linux/jiffies.h> +#include <linux/sched.h> +#include <linux/timer.h> +#include <linux/completion.h> + +static struct timeriomem_rng_data *timeriomem_rng_data; + +static void timeriomem_rng_trigger(unsigned long); +static DEFINE_TIMER(timeriomem_rng_timer, timeriomem_rng_trigger, 0, 0); + +/* + * have data return 1, however return 0 if we have nothing + */ +static int timeriomem_rng_data_present(struct hwrng *rng, int wait) +{ + if (rng->priv == 0) + return 1; + + if (!wait || timeriomem_rng_data->present) + return timeriomem_rng_data->present; + + wait_for_completion(&timeriomem_rng_data->completion); + + return 1; +} + +static int timeriomem_rng_data_read(struct hwrng *rng, u32 *data) +{ + unsigned long cur; + s32 delay; + + *data = readl(timeriomem_rng_data->address); + + if (rng->priv != 0) { + cur = jiffies; + + delay = cur - timeriomem_rng_timer.expires; + delay = rng->priv - (delay % rng->priv); + + timeriomem_rng_timer.expires = cur + delay; + timeriomem_rng_data->present = 0; + + init_completion(&timeriomem_rng_data->completion); + add_timer(&timeriomem_rng_timer); + } + + return 4; +} + +static void timeriomem_rng_trigger(unsigned long dummy) +{ + timeriomem_rng_data->present = 1; + complete(&timeriomem_rng_data->completion); +} + +static struct hwrng timeriomem_rng_ops = { + .name = "timeriomem", + .data_present = timeriomem_rng_data_present, + .data_read = timeriomem_rng_data_read, + .priv = 0, +}; + +static int __init timeriomem_rng_probe(struct platform_device *pdev) +{ + int ret; + + timeriomem_rng_data = pdev->dev.platform_data; + + if (timeriomem_rng_data->period != 0 + && usecs_to_jiffies(timeriomem_rng_data->period) > 0) { + timeriomem_rng_timer.expires = jiffies; + + timeriomem_rng_ops.priv = usecs_to_jiffies( + timeriomem_rng_data->period); + } + timeriomem_rng_data->present = 1; + + ret = hwrng_register(&timeriomem_rng_ops); + if (ret) { + dev_err(&pdev->dev, "problem registering\n"); + return ret; + } + + dev_info(&pdev->dev, "32bits from 0x%p @ %dus\n", + timeriomem_rng_data->address, + timeriomem_rng_data->period); + + return 0; +} + +static int __devexit timeriomem_rng_remove(struct platform_device *pdev) +{ + del_timer_sync(&timeriomem_rng_timer); + hwrng_unregister(&timeriomem_rng_ops); + + return 0; +} + +static struct platform_driver timeriomem_rng_driver = { + .driver = { + .name = "timeriomem_rng", + .owner = THIS_MODULE, + }, + .probe = timeriomem_rng_probe, + .remove = __devexit_p(timeriomem_rng_remove), +}; + +static int __init timeriomem_rng_init(void) +{ + return platform_driver_register(&timeriomem_rng_driver); +} + +static void __exit timeriomem_rng_exit(void) +{ + platform_driver_unregister(&timeriomem_rng_driver); +} + +module_init(timeriomem_rng_init); +module_exit(timeriomem_rng_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Alexander Clouter <alex@digriz.org.uk>"); +MODULE_DESCRIPTION("Timer IOMEM H/W RNG driver"); diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c index 9c47dc48c9f..ccdd828adce 100644 --- a/drivers/char/tpm/tpm.c +++ b/drivers/char/tpm/tpm.c @@ -429,134 +429,148 @@ out: #define TPM_DIGEST_SIZE 20 #define TPM_ERROR_SIZE 10 #define TPM_RET_CODE_IDX 6 -#define TPM_GET_CAP_RET_SIZE_IDX 10 -#define TPM_GET_CAP_RET_UINT32_1_IDX 14 -#define TPM_GET_CAP_RET_UINT32_2_IDX 18 -#define TPM_GET_CAP_RET_UINT32_3_IDX 22 -#define TPM_GET_CAP_RET_UINT32_4_IDX 26 -#define TPM_GET_CAP_PERM_DISABLE_IDX 16 -#define TPM_GET_CAP_PERM_INACTIVE_IDX 18 -#define TPM_GET_CAP_RET_BOOL_1_IDX 14 -#define TPM_GET_CAP_TEMP_INACTIVE_IDX 16 - -#define TPM_CAP_IDX 13 -#define TPM_CAP_SUBCAP_IDX 21 enum tpm_capabilities { - TPM_CAP_FLAG = 4, - TPM_CAP_PROP = 5, + TPM_CAP_FLAG = cpu_to_be32(4), + TPM_CAP_PROP = cpu_to_be32(5), + CAP_VERSION_1_1 = cpu_to_be32(0x06), + CAP_VERSION_1_2 = cpu_to_be32(0x1A) }; enum tpm_sub_capabilities { - TPM_CAP_PROP_PCR = 0x1, - TPM_CAP_PROP_MANUFACTURER = 0x3, - TPM_CAP_FLAG_PERM = 0x8, - TPM_CAP_FLAG_VOL = 0x9, - TPM_CAP_PROP_OWNER = 0x11, - TPM_CAP_PROP_TIS_TIMEOUT = 0x15, - TPM_CAP_PROP_TIS_DURATION = 0x20, -}; + TPM_CAP_PROP_PCR = cpu_to_be32(0x101), + TPM_CAP_PROP_MANUFACTURER = cpu_to_be32(0x103), + TPM_CAP_FLAG_PERM = cpu_to_be32(0x108), + TPM_CAP_FLAG_VOL = cpu_to_be32(0x109), + TPM_CAP_PROP_OWNER = cpu_to_be32(0x111), + TPM_CAP_PROP_TIS_TIMEOUT = cpu_to_be32(0x115), + TPM_CAP_PROP_TIS_DURATION = cpu_to_be32(0x120), -/* - * This is a semi generic GetCapability command for use - * with the capability type TPM_CAP_PROP or TPM_CAP_FLAG - * and their associated sub_capabilities. - */ - -static const u8 tpm_cap[] = { - 0, 193, /* TPM_TAG_RQU_COMMAND */ - 0, 0, 0, 22, /* length */ - 0, 0, 0, 101, /* TPM_ORD_GetCapability */ - 0, 0, 0, 0, /* TPM_CAP_<TYPE> */ - 0, 0, 0, 4, /* TPM_CAP_SUB_<TYPE> size */ - 0, 0, 1, 0 /* TPM_CAP_SUB_<TYPE> */ }; -static ssize_t transmit_cmd(struct tpm_chip *chip, u8 *data, int len, - char *desc) +static ssize_t transmit_cmd(struct tpm_chip *chip, struct tpm_cmd_t *cmd, + int len, const char *desc) { int err; - len = tpm_transmit(chip, data, len); + len = tpm_transmit(chip,(u8 *) cmd, len); if (len < 0) return len; if (len == TPM_ERROR_SIZE) { - err = be32_to_cpu(*((__be32 *) (data + TPM_RET_CODE_IDX))); + err = be32_to_cpu(cmd->header.out.return_code); dev_dbg(chip->dev, "A TPM error (%d) occurred %s\n", err, desc); return err; } return 0; } +#define TPM_INTERNAL_RESULT_SIZE 200 +#define TPM_TAG_RQU_COMMAND cpu_to_be16(193) +#define TPM_ORD_GET_CAP cpu_to_be32(101) + +static const struct tpm_input_header tpm_getcap_header = { + .tag = TPM_TAG_RQU_COMMAND, + .length = cpu_to_be32(22), + .ordinal = TPM_ORD_GET_CAP +}; + +ssize_t tpm_getcap(struct device *dev, __be32 subcap_id, cap_t *cap, + const char *desc) +{ + struct tpm_cmd_t tpm_cmd; + int rc; + struct tpm_chip *chip = dev_get_drvdata(dev); + + tpm_cmd.header.in = tpm_getcap_header; + if (subcap_id == CAP_VERSION_1_1 || subcap_id == CAP_VERSION_1_2) { + tpm_cmd.params.getcap_in.cap = subcap_id; + /*subcap field not necessary */ + tpm_cmd.params.getcap_in.subcap_size = cpu_to_be32(0); + tpm_cmd.header.in.length -= cpu_to_be32(sizeof(__be32)); + } else { + if (subcap_id == TPM_CAP_FLAG_PERM || + subcap_id == TPM_CAP_FLAG_VOL) + tpm_cmd.params.getcap_in.cap = TPM_CAP_FLAG; + else + tpm_cmd.params.getcap_in.cap = TPM_CAP_PROP; + tpm_cmd.params.getcap_in.subcap_size = cpu_to_be32(4); + tpm_cmd.params.getcap_in.subcap = subcap_id; + } + rc = transmit_cmd(chip, &tpm_cmd, TPM_INTERNAL_RESULT_SIZE, desc); + if (!rc) + *cap = tpm_cmd.params.getcap_out.cap; + return rc; +} + void tpm_gen_interrupt(struct tpm_chip *chip) { - u8 data[max_t(int, ARRAY_SIZE(tpm_cap), 30)]; + struct tpm_cmd_t tpm_cmd; ssize_t rc; - memcpy(data, tpm_cap, sizeof(tpm_cap)); - data[TPM_CAP_IDX] = TPM_CAP_PROP; - data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_PROP_TIS_TIMEOUT; + tpm_cmd.header.in = tpm_getcap_header; + tpm_cmd.params.getcap_in.cap = TPM_CAP_PROP; + tpm_cmd.params.getcap_in.subcap_size = cpu_to_be32(4); + tpm_cmd.params.getcap_in.subcap = TPM_CAP_PROP_TIS_TIMEOUT; - rc = transmit_cmd(chip, data, sizeof(data), + rc = transmit_cmd(chip, &tpm_cmd, TPM_INTERNAL_RESULT_SIZE, "attempting to determine the timeouts"); } EXPORT_SYMBOL_GPL(tpm_gen_interrupt); void tpm_get_timeouts(struct tpm_chip *chip) { - u8 data[max_t(int, ARRAY_SIZE(tpm_cap), 30)]; + struct tpm_cmd_t tpm_cmd; + struct timeout_t *timeout_cap; + struct duration_t *duration_cap; ssize_t rc; u32 timeout; - memcpy(data, tpm_cap, sizeof(tpm_cap)); - data[TPM_CAP_IDX] = TPM_CAP_PROP; - data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_PROP_TIS_TIMEOUT; + tpm_cmd.header.in = tpm_getcap_header; + tpm_cmd.params.getcap_in.cap = TPM_CAP_PROP; + tpm_cmd.params.getcap_in.subcap_size = cpu_to_be32(4); + tpm_cmd.params.getcap_in.subcap = TPM_CAP_PROP_TIS_TIMEOUT; - rc = transmit_cmd(chip, data, sizeof(data), + rc = transmit_cmd(chip, &tpm_cmd, TPM_INTERNAL_RESULT_SIZE, "attempting to determine the timeouts"); if (rc) goto duration; - if (be32_to_cpu(*((__be32 *) (data + TPM_GET_CAP_RET_SIZE_IDX))) + if (be32_to_cpu(tpm_cmd.header.out.length) != 4 * sizeof(u32)) goto duration; + timeout_cap = &tpm_cmd.params.getcap_out.cap.timeout; /* Don't overwrite default if value is 0 */ - timeout = - be32_to_cpu(*((__be32 *) (data + TPM_GET_CAP_RET_UINT32_1_IDX))); + timeout = be32_to_cpu(timeout_cap->a); if (timeout) chip->vendor.timeout_a = usecs_to_jiffies(timeout); - timeout = - be32_to_cpu(*((__be32 *) (data + TPM_GET_CAP_RET_UINT32_2_IDX))); + timeout = be32_to_cpu(timeout_cap->b); if (timeout) chip->vendor.timeout_b = usecs_to_jiffies(timeout); - timeout = - be32_to_cpu(*((__be32 *) (data + TPM_GET_CAP_RET_UINT32_3_IDX))); + timeout = be32_to_cpu(timeout_cap->c); if (timeout) chip->vendor.timeout_c = usecs_to_jiffies(timeout); - timeout = - be32_to_cpu(*((__be32 *) (data + TPM_GET_CAP_RET_UINT32_4_IDX))); + timeout = be32_to_cpu(timeout_cap->d); if (timeout) chip->vendor.timeout_d = usecs_to_jiffies(timeout); duration: - memcpy(data, tpm_cap, sizeof(tpm_cap)); - data[TPM_CAP_IDX] = TPM_CAP_PROP; - data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_PROP_TIS_DURATION; + tpm_cmd.header.in = tpm_getcap_header; + tpm_cmd.params.getcap_in.cap = TPM_CAP_PROP; + tpm_cmd.params.getcap_in.subcap_size = cpu_to_be32(4); + tpm_cmd.params.getcap_in.subcap = TPM_CAP_PROP_TIS_DURATION; - rc = transmit_cmd(chip, data, sizeof(data), + rc = transmit_cmd(chip, &tpm_cmd, TPM_INTERNAL_RESULT_SIZE, "attempting to determine the durations"); if (rc) return; - if (be32_to_cpu(*((__be32 *) (data + TPM_GET_CAP_RET_SIZE_IDX))) + if (be32_to_cpu(tpm_cmd.header.out.return_code) != 3 * sizeof(u32)) return; - + duration_cap = &tpm_cmd.params.getcap_out.cap.duration; chip->vendor.duration[TPM_SHORT] = - usecs_to_jiffies(be32_to_cpu - (*((__be32 *) (data + - TPM_GET_CAP_RET_UINT32_1_IDX)))); + usecs_to_jiffies(be32_to_cpu(duration_cap->tpm_short)); /* The Broadcom BCM0102 chipset in a Dell Latitude D820 gets the above * value wrong and apparently reports msecs rather than usecs. So we * fix up the resulting too-small TPM_SHORT value to make things work. @@ -565,13 +579,9 @@ duration: chip->vendor.duration[TPM_SHORT] = HZ; chip->vendor.duration[TPM_MEDIUM] = - usecs_to_jiffies(be32_to_cpu - (*((__be32 *) (data + - TPM_GET_CAP_RET_UINT32_2_IDX)))); + usecs_to_jiffies(be32_to_cpu(duration_cap->tpm_medium)); chip->vendor.duration[TPM_LONG] = - usecs_to_jiffies(be32_to_cpu - (*((__be32 *) (data + - TPM_GET_CAP_RET_UINT32_3_IDX)))); + usecs_to_jiffies(be32_to_cpu(duration_cap->tpm_long)); } EXPORT_SYMBOL_GPL(tpm_get_timeouts); @@ -587,36 +597,18 @@ void tpm_continue_selftest(struct tpm_chip *chip) } EXPORT_SYMBOL_GPL(tpm_continue_selftest); -#define TPM_INTERNAL_RESULT_SIZE 200 - ssize_t tpm_show_enabled(struct device * dev, struct device_attribute * attr, char *buf) { - u8 *data; + cap_t cap; ssize_t rc; - struct tpm_chip *chip = dev_get_drvdata(dev); - if (chip == NULL) - return -ENODEV; - - data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL); - if (!data) - return -ENOMEM; - - memcpy(data, tpm_cap, sizeof(tpm_cap)); - data[TPM_CAP_IDX] = TPM_CAP_FLAG; - data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_FLAG_PERM; - - rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE, - "attemtping to determine the permanent enabled state"); - if (rc) { - kfree(data); + rc = tpm_getcap(dev, TPM_CAP_FLAG_PERM, &cap, + "attempting to determine the permanent enabled state"); + if (rc) return 0; - } - - rc = sprintf(buf, "%d\n", !data[TPM_GET_CAP_PERM_DISABLE_IDX]); - kfree(data); + rc = sprintf(buf, "%d\n", !cap.perm_flags.disable); return rc; } EXPORT_SYMBOL_GPL(tpm_show_enabled); @@ -624,31 +616,15 @@ EXPORT_SYMBOL_GPL(tpm_show_enabled); ssize_t tpm_show_active(struct device * dev, struct device_attribute * attr, char *buf) { - u8 *data; + cap_t cap; ssize_t rc; - struct tpm_chip *chip = dev_get_drvdata(dev); - if (chip == NULL) - return -ENODEV; - - data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL); - if (!data) - return -ENOMEM; - - memcpy(data, tpm_cap, sizeof(tpm_cap)); - data[TPM_CAP_IDX] = TPM_CAP_FLAG; - data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_FLAG_PERM; - - rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE, - "attemtping to determine the permanent active state"); - if (rc) { - kfree(data); + rc = tpm_getcap(dev, TPM_CAP_FLAG_PERM, &cap, + "attempting to determine the permanent active state"); + if (rc) return 0; - } - rc = sprintf(buf, "%d\n", !data[TPM_GET_CAP_PERM_INACTIVE_IDX]); - - kfree(data); + rc = sprintf(buf, "%d\n", !cap.perm_flags.deactivated); return rc; } EXPORT_SYMBOL_GPL(tpm_show_active); @@ -656,31 +632,15 @@ EXPORT_SYMBOL_GPL(tpm_show_active); ssize_t tpm_show_owned(struct device * dev, struct device_attribute * attr, char *buf) { - u8 *data; + cap_t cap; ssize_t rc; - struct tpm_chip *chip = dev_get_drvdata(dev); - if (chip == NULL) - return -ENODEV; - - data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL); - if (!data) - return -ENOMEM; - - memcpy(data, tpm_cap, sizeof(tpm_cap)); - data[TPM_CAP_IDX] = TPM_CAP_PROP; - data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_PROP_OWNER; - - rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE, - "attempting to determine the owner state"); - if (rc) { - kfree(data); + rc = tpm_getcap(dev, TPM_CAP_PROP_OWNER, &cap, + "attempting to determine the owner state"); + if (rc) return 0; - } - - rc = sprintf(buf, "%d\n", data[TPM_GET_CAP_RET_BOOL_1_IDX]); - kfree(data); + rc = sprintf(buf, "%d\n", cap.owned); return rc; } EXPORT_SYMBOL_GPL(tpm_show_owned); @@ -688,116 +648,180 @@ EXPORT_SYMBOL_GPL(tpm_show_owned); ssize_t tpm_show_temp_deactivated(struct device * dev, struct device_attribute * attr, char *buf) { - u8 *data; + cap_t cap; ssize_t rc; - struct tpm_chip *chip = dev_get_drvdata(dev); - if (chip == NULL) - return -ENODEV; + rc = tpm_getcap(dev, TPM_CAP_FLAG_VOL, &cap, + "attempting to determine the temporary state"); + if (rc) + return 0; - data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL); - if (!data) - return -ENOMEM; + rc = sprintf(buf, "%d\n", cap.stclear_flags.deactivated); + return rc; +} +EXPORT_SYMBOL_GPL(tpm_show_temp_deactivated); - memcpy(data, tpm_cap, sizeof(tpm_cap)); - data[TPM_CAP_IDX] = TPM_CAP_FLAG; - data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_FLAG_VOL; +/* + * tpm_chip_find_get - return tpm_chip for given chip number + */ +static struct tpm_chip *tpm_chip_find_get(int chip_num) +{ + struct tpm_chip *pos, *chip = NULL; - rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE, - "attempting to determine the temporary state"); - if (rc) { - kfree(data); - return 0; + rcu_read_lock(); + list_for_each_entry_rcu(pos, &tpm_chip_list, list) { + if (chip_num != TPM_ANY_NUM && chip_num != pos->dev_num) + continue; + + if (try_module_get(pos->dev->driver->owner)) { + chip = pos; + break; + } } + rcu_read_unlock(); + return chip; +} - rc = sprintf(buf, "%d\n", data[TPM_GET_CAP_TEMP_INACTIVE_IDX]); +#define TPM_ORDINAL_PCRREAD cpu_to_be32(21) +#define READ_PCR_RESULT_SIZE 30 +static struct tpm_input_header pcrread_header = { + .tag = TPM_TAG_RQU_COMMAND, + .length = cpu_to_be32(14), + .ordinal = TPM_ORDINAL_PCRREAD +}; - kfree(data); +int __tpm_pcr_read(struct tpm_chip *chip, int pcr_idx, u8 *res_buf) +{ + int rc; + struct tpm_cmd_t cmd; + + cmd.header.in = pcrread_header; + cmd.params.pcrread_in.pcr_idx = cpu_to_be32(pcr_idx); + BUILD_BUG_ON(cmd.header.in.length > READ_PCR_RESULT_SIZE); + rc = transmit_cmd(chip, &cmd, cmd.header.in.length, + "attempting to read a pcr value"); + + if (rc == 0) + memcpy(res_buf, cmd.params.pcrread_out.pcr_result, + TPM_DIGEST_SIZE); return rc; } -EXPORT_SYMBOL_GPL(tpm_show_temp_deactivated); -static const u8 pcrread[] = { - 0, 193, /* TPM_TAG_RQU_COMMAND */ - 0, 0, 0, 14, /* length */ - 0, 0, 0, 21, /* TPM_ORD_PcrRead */ - 0, 0, 0, 0 /* PCR index */ +/** + * tpm_pcr_read - read a pcr value + * @chip_num: tpm idx # or ANY + * @pcr_idx: pcr idx to retrieve + * @res_buf: TPM_PCR value + * size of res_buf is 20 bytes (or NULL if you don't care) + * + * The TPM driver should be built-in, but for whatever reason it + * isn't, protect against the chip disappearing, by incrementing + * the module usage count. + */ +int tpm_pcr_read(u32 chip_num, int pcr_idx, u8 *res_buf) +{ + struct tpm_chip *chip; + int rc; + + chip = tpm_chip_find_get(chip_num); + if (chip == NULL) + return -ENODEV; + rc = __tpm_pcr_read(chip, pcr_idx, res_buf); + module_put(chip->dev->driver->owner); + return rc; +} +EXPORT_SYMBOL_GPL(tpm_pcr_read); + +/** + * tpm_pcr_extend - extend pcr value with hash + * @chip_num: tpm idx # or AN& + * @pcr_idx: pcr idx to extend + * @hash: hash value used to extend pcr value + * + * The TPM driver should be built-in, but for whatever reason it + * isn't, protect against the chip disappearing, by incrementing + * the module usage count. + */ +#define TPM_ORD_PCR_EXTEND cpu_to_be32(20) +#define EXTEND_PCR_SIZE 34 +static struct tpm_input_header pcrextend_header = { + .tag = TPM_TAG_RQU_COMMAND, + .length = cpu_to_be32(34), + .ordinal = TPM_ORD_PCR_EXTEND }; +int tpm_pcr_extend(u32 chip_num, int pcr_idx, const u8 *hash) +{ + struct tpm_cmd_t cmd; + int rc; + struct tpm_chip *chip; + + chip = tpm_chip_find_get(chip_num); + if (chip == NULL) + return -ENODEV; + + cmd.header.in = pcrextend_header; + BUILD_BUG_ON(be32_to_cpu(cmd.header.in.length) > EXTEND_PCR_SIZE); + cmd.params.pcrextend_in.pcr_idx = cpu_to_be32(pcr_idx); + memcpy(cmd.params.pcrextend_in.hash, hash, TPM_DIGEST_SIZE); + rc = transmit_cmd(chip, &cmd, cmd.header.in.length, + "attempting extend a PCR value"); + + module_put(chip->dev->driver->owner); + return rc; +} +EXPORT_SYMBOL_GPL(tpm_pcr_extend); + ssize_t tpm_show_pcrs(struct device *dev, struct device_attribute *attr, char *buf) { - u8 *data; + cap_t cap; + u8 digest[TPM_DIGEST_SIZE]; ssize_t rc; int i, j, num_pcrs; - __be32 index; char *str = buf; - struct tpm_chip *chip = dev_get_drvdata(dev); - if (chip == NULL) - return -ENODEV; - data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL); - if (!data) - return -ENOMEM; - - memcpy(data, tpm_cap, sizeof(tpm_cap)); - data[TPM_CAP_IDX] = TPM_CAP_PROP; - data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_PROP_PCR; - - rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE, + rc = tpm_getcap(dev, TPM_CAP_PROP_PCR, &cap, "attempting to determine the number of PCRS"); - if (rc) { - kfree(data); + if (rc) return 0; - } - num_pcrs = be32_to_cpu(*((__be32 *) (data + 14))); + num_pcrs = be32_to_cpu(cap.num_pcrs); for (i = 0; i < num_pcrs; i++) { - memcpy(data, pcrread, sizeof(pcrread)); - index = cpu_to_be32(i); - memcpy(data + 10, &index, 4); - rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE, - "attempting to read a PCR"); + rc = __tpm_pcr_read(chip, i, digest); if (rc) - goto out; + break; str += sprintf(str, "PCR-%02d: ", i); for (j = 0; j < TPM_DIGEST_SIZE; j++) - str += sprintf(str, "%02X ", *(data + 10 + j)); + str += sprintf(str, "%02X ", digest[j]); str += sprintf(str, "\n"); } -out: - kfree(data); return str - buf; } EXPORT_SYMBOL_GPL(tpm_show_pcrs); #define READ_PUBEK_RESULT_SIZE 314 -static const u8 readpubek[] = { - 0, 193, /* TPM_TAG_RQU_COMMAND */ - 0, 0, 0, 30, /* length */ - 0, 0, 0, 124, /* TPM_ORD_ReadPubek */ +#define TPM_ORD_READPUBEK cpu_to_be32(124) +struct tpm_input_header tpm_readpubek_header = { + .tag = TPM_TAG_RQU_COMMAND, + .length = cpu_to_be32(30), + .ordinal = TPM_ORD_READPUBEK }; ssize_t tpm_show_pubek(struct device *dev, struct device_attribute *attr, char *buf) { u8 *data; + struct tpm_cmd_t tpm_cmd; ssize_t err; int i, rc; char *str = buf; struct tpm_chip *chip = dev_get_drvdata(dev); - if (chip == NULL) - return -ENODEV; - data = kzalloc(READ_PUBEK_RESULT_SIZE, GFP_KERNEL); - if (!data) - return -ENOMEM; - - memcpy(data, readpubek, sizeof(readpubek)); - - err = transmit_cmd(chip, data, READ_PUBEK_RESULT_SIZE, + tpm_cmd.header.in = tpm_readpubek_header; + err = transmit_cmd(chip, &tpm_cmd, READ_PUBEK_RESULT_SIZE, "attempting to read the PUBEK"); if (err) goto out; @@ -812,7 +836,7 @@ ssize_t tpm_show_pubek(struct device *dev, struct device_attribute *attr, 256 byte modulus ignore checksum 20 bytes */ - + data = tpm_cmd.params.readpubek_out_buffer; str += sprintf(str, "Algorithm: %02X %02X %02X %02X\nEncscheme: %02X %02X\n" @@ -832,65 +856,33 @@ ssize_t tpm_show_pubek(struct device *dev, struct device_attribute *attr, } out: rc = str - buf; - kfree(data); return rc; } EXPORT_SYMBOL_GPL(tpm_show_pubek); -#define CAP_VERSION_1_1 6 -#define CAP_VERSION_1_2 0x1A -#define CAP_VERSION_IDX 13 -static const u8 cap_version[] = { - 0, 193, /* TPM_TAG_RQU_COMMAND */ - 0, 0, 0, 18, /* length */ - 0, 0, 0, 101, /* TPM_ORD_GetCapability */ - 0, 0, 0, 0, - 0, 0, 0, 0 -}; ssize_t tpm_show_caps(struct device *dev, struct device_attribute *attr, char *buf) { - u8 *data; + cap_t cap; ssize_t rc; char *str = buf; - struct tpm_chip *chip = dev_get_drvdata(dev); - if (chip == NULL) - return -ENODEV; - - data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL); - if (!data) - return -ENOMEM; - - memcpy(data, tpm_cap, sizeof(tpm_cap)); - data[TPM_CAP_IDX] = TPM_CAP_PROP; - data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_PROP_MANUFACTURER; - - rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE, + rc = tpm_getcap(dev, TPM_CAP_PROP_MANUFACTURER, &cap, "attempting to determine the manufacturer"); - if (rc) { - kfree(data); + if (rc) return 0; - } - str += sprintf(str, "Manufacturer: 0x%x\n", - be32_to_cpu(*((__be32 *) (data + TPM_GET_CAP_RET_UINT32_1_IDX)))); + be32_to_cpu(cap.manufacturer_id)); - memcpy(data, cap_version, sizeof(cap_version)); - data[CAP_VERSION_IDX] = CAP_VERSION_1_1; - rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE, - "attempting to determine the 1.1 version"); + rc = tpm_getcap(dev, CAP_VERSION_1_1, &cap, + "attempting to determine the 1.1 version"); if (rc) - goto out; - + return 0; str += sprintf(str, "TCG version: %d.%d\nFirmware version: %d.%d\n", - (int) data[14], (int) data[15], (int) data[16], - (int) data[17]); - -out: - kfree(data); + cap.tpm_version.Major, cap.tpm_version.Minor, + cap.tpm_version.revMajor, cap.tpm_version.revMinor); return str - buf; } EXPORT_SYMBOL_GPL(tpm_show_caps); @@ -898,51 +890,25 @@ EXPORT_SYMBOL_GPL(tpm_show_caps); ssize_t tpm_show_caps_1_2(struct device * dev, struct device_attribute * attr, char *buf) { - u8 *data; - ssize_t len; + cap_t cap; + ssize_t rc; char *str = buf; - struct tpm_chip *chip = dev_get_drvdata(dev); - if (chip == NULL) - return -ENODEV; - - data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL); - if (!data) - return -ENOMEM; - - memcpy(data, tpm_cap, sizeof(tpm_cap)); - data[TPM_CAP_IDX] = TPM_CAP_PROP; - data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_PROP_MANUFACTURER; - - len = tpm_transmit(chip, data, TPM_INTERNAL_RESULT_SIZE); - if (len <= TPM_ERROR_SIZE) { - dev_dbg(chip->dev, "A TPM error (%d) occurred " - "attempting to determine the manufacturer\n", - be32_to_cpu(*((__be32 *) (data + TPM_RET_CODE_IDX)))); - kfree(data); + rc = tpm_getcap(dev, TPM_CAP_PROP_MANUFACTURER, &cap, + "attempting to determine the manufacturer"); + if (rc) return 0; - } - str += sprintf(str, "Manufacturer: 0x%x\n", - be32_to_cpu(*((__be32 *) (data + TPM_GET_CAP_RET_UINT32_1_IDX)))); - - memcpy(data, cap_version, sizeof(cap_version)); - data[CAP_VERSION_IDX] = CAP_VERSION_1_2; - - len = tpm_transmit(chip, data, TPM_INTERNAL_RESULT_SIZE); - if (len <= TPM_ERROR_SIZE) { - dev_err(chip->dev, "A TPM error (%d) occurred " - "attempting to determine the 1.2 version\n", - be32_to_cpu(*((__be32 *) (data + TPM_RET_CODE_IDX)))); - goto out; - } + be32_to_cpu(cap.manufacturer_id)); + rc = tpm_getcap(dev, CAP_VERSION_1_2, &cap, + "attempting to determine the 1.2 version"); + if (rc) + return 0; str += sprintf(str, "TCG version: %d.%d\nFirmware version: %d.%d\n", - (int) data[16], (int) data[17], (int) data[18], - (int) data[19]); - -out: - kfree(data); + cap.tpm_version_1_2.Major, cap.tpm_version_1_2.Minor, + cap.tpm_version_1_2.revMajor, + cap.tpm_version_1_2.revMinor); return str - buf; } EXPORT_SYMBOL_GPL(tpm_show_caps_1_2); diff --git a/drivers/char/tpm/tpm.h b/drivers/char/tpm/tpm.h index 8e30df4a438..8e00b4ddd08 100644 --- a/drivers/char/tpm/tpm.h +++ b/drivers/char/tpm/tpm.h @@ -26,6 +26,7 @@ #include <linux/miscdevice.h> #include <linux/platform_device.h> #include <linux/io.h> +#include <linux/tpm.h> enum tpm_timeout { TPM_TIMEOUT = 5, /* msecs */ @@ -123,6 +124,147 @@ static inline void tpm_write_index(int base, int index, int value) outb(index, base); outb(value & 0xFF, base+1); } +struct tpm_input_header { + __be16 tag; + __be32 length; + __be32 ordinal; +}__attribute__((packed)); + +struct tpm_output_header { + __be16 tag; + __be32 length; + __be32 return_code; +}__attribute__((packed)); + +struct stclear_flags_t { + __be16 tag; + u8 deactivated; + u8 disableForceClear; + u8 physicalPresence; + u8 physicalPresenceLock; + u8 bGlobalLock; +}__attribute__((packed)); + +struct tpm_version_t { + u8 Major; + u8 Minor; + u8 revMajor; + u8 revMinor; +}__attribute__((packed)); + +struct tpm_version_1_2_t { + __be16 tag; + u8 Major; + u8 Minor; + u8 revMajor; + u8 revMinor; +}__attribute__((packed)); + +struct timeout_t { + __be32 a; + __be32 b; + __be32 c; + __be32 d; +}__attribute__((packed)); + +struct duration_t { + __be32 tpm_short; + __be32 tpm_medium; + __be32 tpm_long; +}__attribute__((packed)); + +struct permanent_flags_t { + __be16 tag; + u8 disable; + u8 ownership; + u8 deactivated; + u8 readPubek; + u8 disableOwnerClear; + u8 allowMaintenance; + u8 physicalPresenceLifetimeLock; + u8 physicalPresenceHWEnable; + u8 physicalPresenceCMDEnable; + u8 CEKPUsed; + u8 TPMpost; + u8 TPMpostLock; + u8 FIPS; + u8 operator; + u8 enableRevokeEK; + u8 nvLocked; + u8 readSRKPub; + u8 tpmEstablished; + u8 maintenanceDone; + u8 disableFullDALogicInfo; +}__attribute__((packed)); + +typedef union { + struct permanent_flags_t perm_flags; + struct stclear_flags_t stclear_flags; + bool owned; + __be32 num_pcrs; + struct tpm_version_t tpm_version; + struct tpm_version_1_2_t tpm_version_1_2; + __be32 manufacturer_id; + struct timeout_t timeout; + struct duration_t duration; +} cap_t; + +struct tpm_getcap_params_in { + __be32 cap; + __be32 subcap_size; + __be32 subcap; +}__attribute__((packed)); + +struct tpm_getcap_params_out { + __be32 cap_size; + cap_t cap; +}__attribute__((packed)); + +struct tpm_readpubek_params_out { + u8 algorithm[4]; + u8 encscheme[2]; + u8 sigscheme[2]; + u8 parameters[12]; /*assuming RSA*/ + __be32 keysize; + u8 modulus[256]; + u8 checksum[20]; +}__attribute__((packed)); + +typedef union { + struct tpm_input_header in; + struct tpm_output_header out; +} tpm_cmd_header; + +#define TPM_DIGEST_SIZE 20 +struct tpm_pcrread_out { + u8 pcr_result[TPM_DIGEST_SIZE]; +}__attribute__((packed)); + +struct tpm_pcrread_in { + __be32 pcr_idx; +}__attribute__((packed)); + +struct tpm_pcrextend_in { + __be32 pcr_idx; + u8 hash[TPM_DIGEST_SIZE]; +}__attribute__((packed)); + +typedef union { + struct tpm_getcap_params_out getcap_out; + struct tpm_readpubek_params_out readpubek_out; + u8 readpubek_out_buffer[sizeof(struct tpm_readpubek_params_out)]; + struct tpm_getcap_params_in getcap_in; + struct tpm_pcrread_in pcrread_in; + struct tpm_pcrread_out pcrread_out; + struct tpm_pcrextend_in pcrextend_in; +} tpm_cmd_params; + +struct tpm_cmd_t { + tpm_cmd_header header; + tpm_cmd_params params; +}__attribute__((packed)); + +ssize_t tpm_getcap(struct device *, __be32, cap_t *, const char *); extern void tpm_get_timeouts(struct tpm_chip *); extern void tpm_gen_interrupt(struct tpm_chip *); diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index d6daf3c507d..d270e8eb3e6 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -104,7 +104,8 @@ EXPORT_SYMBOL_GPL(unlock_policy_rwsem_write); /* internal prototypes */ -static int __cpufreq_governor(struct cpufreq_policy *policy, unsigned int event); +static int __cpufreq_governor(struct cpufreq_policy *policy, + unsigned int event); static unsigned int __cpufreq_get(unsigned int cpu); static void handle_update(struct work_struct *work); @@ -128,7 +129,7 @@ static int __init init_cpufreq_transition_notifier_list(void) pure_initcall(init_cpufreq_transition_notifier_list); static LIST_HEAD(cpufreq_governor_list); -static DEFINE_MUTEX (cpufreq_governor_mutex); +static DEFINE_MUTEX(cpufreq_governor_mutex); struct cpufreq_policy *cpufreq_cpu_get(unsigned int cpu) { @@ -371,7 +372,7 @@ static struct cpufreq_governor *__find_governor(const char *str_governor) struct cpufreq_governor *t; list_for_each_entry(t, &cpufreq_governor_list, governor_list) - if (!strnicmp(str_governor,t->name,CPUFREQ_NAME_LEN)) + if (!strnicmp(str_governor, t->name, CPUFREQ_NAME_LEN)) return t; return NULL; @@ -429,15 +430,11 @@ static int cpufreq_parse_governor(char *str_governor, unsigned int *policy, mutex_unlock(&cpufreq_governor_mutex); } - out: +out: return err; } -/* drivers/base/cpu.c */ -extern struct sysdev_class cpu_sysdev_class; - - /** * cpufreq_per_cpu_attr_read() / show_##file_name() - * print out cpufreq information @@ -450,11 +447,12 @@ extern struct sysdev_class cpu_sysdev_class; static ssize_t show_##file_name \ (struct cpufreq_policy *policy, char *buf) \ { \ - return sprintf (buf, "%u\n", policy->object); \ + return sprintf(buf, "%u\n", policy->object); \ } show_one(cpuinfo_min_freq, cpuinfo.min_freq); show_one(cpuinfo_max_freq, cpuinfo.max_freq); +show_one(cpuinfo_transition_latency, cpuinfo.transition_latency); show_one(scaling_min_freq, min); show_one(scaling_max_freq, max); show_one(scaling_cur_freq, cur); @@ -476,7 +474,7 @@ static ssize_t store_##file_name \ if (ret) \ return -EINVAL; \ \ - ret = sscanf (buf, "%u", &new_policy.object); \ + ret = sscanf(buf, "%u", &new_policy.object); \ if (ret != 1) \ return -EINVAL; \ \ @@ -486,8 +484,8 @@ static ssize_t store_##file_name \ return ret ? ret : count; \ } -store_one(scaling_min_freq,min); -store_one(scaling_max_freq,max); +store_one(scaling_min_freq, min); +store_one(scaling_max_freq, max); /** * show_cpuinfo_cur_freq - current CPU frequency as detected by hardware @@ -507,12 +505,13 @@ static ssize_t show_cpuinfo_cur_freq(struct cpufreq_policy *policy, */ static ssize_t show_scaling_governor(struct cpufreq_policy *policy, char *buf) { - if(policy->policy == CPUFREQ_POLICY_POWERSAVE) + if (policy->policy == CPUFREQ_POLICY_POWERSAVE) return sprintf(buf, "powersave\n"); else if (policy->policy == CPUFREQ_POLICY_PERFORMANCE) return sprintf(buf, "performance\n"); else if (policy->governor) - return scnprintf(buf, CPUFREQ_NAME_LEN, "%s\n", policy->governor->name); + return scnprintf(buf, CPUFREQ_NAME_LEN, "%s\n", + policy->governor->name); return -EINVAL; } @@ -531,7 +530,7 @@ static ssize_t store_scaling_governor(struct cpufreq_policy *policy, if (ret) return ret; - ret = sscanf (buf, "%15s", str_governor); + ret = sscanf(buf, "%15s", str_governor); if (ret != 1) return -EINVAL; @@ -575,7 +574,8 @@ static ssize_t show_scaling_available_governors(struct cpufreq_policy *policy, } list_for_each_entry(t, &cpufreq_governor_list, governor_list) { - if (i >= (ssize_t) ((PAGE_SIZE / sizeof(char)) - (CPUFREQ_NAME_LEN + 2))) + if (i >= (ssize_t) ((PAGE_SIZE / sizeof(char)) + - (CPUFREQ_NAME_LEN + 2))) goto out; i += scnprintf(&buf[i], CPUFREQ_NAME_LEN, "%s ", t->name); } @@ -594,7 +594,7 @@ static ssize_t show_cpus(const struct cpumask *mask, char *buf) i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), " "); i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), "%u", cpu); if (i >= (PAGE_SIZE - 5)) - break; + break; } i += sprintf(&buf[i], "\n"); return i; @@ -660,6 +660,7 @@ __ATTR(_name, 0644, show_##_name, store_##_name) define_one_ro0400(cpuinfo_cur_freq); define_one_ro(cpuinfo_min_freq); define_one_ro(cpuinfo_max_freq); +define_one_ro(cpuinfo_transition_latency); define_one_ro(scaling_available_governors); define_one_ro(scaling_driver); define_one_ro(scaling_cur_freq); @@ -673,6 +674,7 @@ define_one_rw(scaling_setspeed); static struct attribute *default_attrs[] = { &cpuinfo_min_freq.attr, &cpuinfo_max_freq.attr, + &cpuinfo_transition_latency.attr, &scaling_min_freq.attr, &scaling_max_freq.attr, &affected_cpus.attr, @@ -684,10 +686,10 @@ static struct attribute *default_attrs[] = { NULL }; -#define to_policy(k) container_of(k,struct cpufreq_policy,kobj) -#define to_attr(a) container_of(a,struct freq_attr,attr) +#define to_policy(k) container_of(k, struct cpufreq_policy, kobj) +#define to_attr(a) container_of(a, struct freq_attr, attr) -static ssize_t show(struct kobject *kobj, struct attribute *attr ,char *buf) +static ssize_t show(struct kobject *kobj, struct attribute *attr, char *buf) { struct cpufreq_policy *policy = to_policy(kobj); struct freq_attr *fattr = to_attr(attr); @@ -853,10 +855,10 @@ static int cpufreq_add_dev(struct sys_device *sys_dev) if (cpu == j) continue; - /* check for existing affected CPUs. They may not be aware - * of it due to CPU Hotplug. + /* Check for existing affected CPUs. + * They may not be aware of it due to CPU Hotplug. */ - managed_policy = cpufreq_cpu_get(j); // FIXME: Where is this released? What about error paths? + managed_policy = cpufreq_cpu_get(j); /* FIXME: Where is this released? What about error paths? */ if (unlikely(managed_policy)) { /* Set proper policy_cpu */ @@ -1127,8 +1129,8 @@ static void handle_update(struct work_struct *work) * @old_freq: CPU frequency the kernel thinks the CPU runs at * @new_freq: CPU frequency the CPU actually runs at * - * We adjust to current frequency first, and need to clean up later. So either call - * to cpufreq_update_policy() or schedule handle_update()). + * We adjust to current frequency first, and need to clean up later. + * So either call to cpufreq_update_policy() or schedule handle_update()). */ static void cpufreq_out_of_sync(unsigned int cpu, unsigned int old_freq, unsigned int new_freq) @@ -1610,7 +1612,8 @@ EXPORT_SYMBOL_GPL(cpufreq_unregister_governor); /** * cpufreq_get_policy - get the current cpufreq_policy - * @policy: struct cpufreq_policy into which the current cpufreq_policy is written + * @policy: struct cpufreq_policy into which the current cpufreq_policy + * is written * * Reads the current cpufreq policy. */ diff --git a/drivers/cpufreq/cpufreq_conservative.c b/drivers/cpufreq/cpufreq_conservative.c index 0320962c4ec..2ecd95e4ab1 100644 --- a/drivers/cpufreq/cpufreq_conservative.c +++ b/drivers/cpufreq/cpufreq_conservative.c @@ -4,7 +4,7 @@ * Copyright (C) 2001 Russell King * (C) 2003 Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>. * Jun Nakajima <jun.nakajima@intel.com> - * (C) 2004 Alexander Clouter <alex-kernel@digriz.org.uk> + * (C) 2009 Alexander Clouter <alex@digriz.org.uk> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -13,22 +13,17 @@ #include <linux/kernel.h> #include <linux/module.h> -#include <linux/smp.h> #include <linux/init.h> -#include <linux/interrupt.h> -#include <linux/ctype.h> #include <linux/cpufreq.h> -#include <linux/sysctl.h> -#include <linux/types.h> -#include <linux/fs.h> -#include <linux/sysfs.h> #include <linux/cpu.h> -#include <linux/kmod.h> -#include <linux/workqueue.h> #include <linux/jiffies.h> #include <linux/kernel_stat.h> -#include <linux/percpu.h> #include <linux/mutex.h> +#include <linux/hrtimer.h> +#include <linux/tick.h> +#include <linux/ktime.h> +#include <linux/sched.h> + /* * dbs is used in this file as a shortform for demandbased switching * It helps to keep variable names smaller, simpler @@ -43,19 +38,31 @@ * latency of the processor. The governor will work on any processor with * transition latency <= 10mS, using appropriate sampling * rate. - * For CPUs with transition latency > 10mS (mostly drivers - * with CPUFREQ_ETERNAL), this governor will not work. + * For CPUs with transition latency > 10mS (mostly drivers with CPUFREQ_ETERNAL) + * this governor will not work. * All times here are in uS. */ static unsigned int def_sampling_rate; #define MIN_SAMPLING_RATE_RATIO (2) /* for correct statistics, we need at least 10 ticks between each measure */ -#define MIN_STAT_SAMPLING_RATE \ +#define MIN_STAT_SAMPLING_RATE \ (MIN_SAMPLING_RATE_RATIO * jiffies_to_usecs(10)) #define MIN_SAMPLING_RATE \ (def_sampling_rate / MIN_SAMPLING_RATE_RATIO) +/* Above MIN_SAMPLING_RATE will vanish with its sysfs file soon + * Define the minimal settable sampling rate to the greater of: + * - "HW transition latency" * 100 (same as default sampling / 10) + * - MIN_STAT_SAMPLING_RATE + * To avoid that userspace shoots itself. +*/ +static unsigned int minimum_sampling_rate(void) +{ + return max(def_sampling_rate / 10, MIN_STAT_SAMPLING_RATE); +} + +/* This will also vanish soon with removing sampling_rate_max */ #define MAX_SAMPLING_RATE (500 * def_sampling_rate) -#define DEF_SAMPLING_RATE_LATENCY_MULTIPLIER (1000) +#define LATENCY_MULTIPLIER (1000) #define DEF_SAMPLING_DOWN_FACTOR (1) #define MAX_SAMPLING_DOWN_FACTOR (10) #define TRANSITION_LATENCY_LIMIT (10 * 1000 * 1000) @@ -63,12 +70,15 @@ static unsigned int def_sampling_rate; static void do_dbs_timer(struct work_struct *work); struct cpu_dbs_info_s { + cputime64_t prev_cpu_idle; + cputime64_t prev_cpu_wall; + cputime64_t prev_cpu_nice; struct cpufreq_policy *cur_policy; - unsigned int prev_cpu_idle_up; - unsigned int prev_cpu_idle_down; - unsigned int enable; + struct delayed_work work; unsigned int down_skip; unsigned int requested_freq; + int cpu; + unsigned int enable:1; }; static DEFINE_PER_CPU(struct cpu_dbs_info_s, cpu_dbs_info); @@ -82,19 +92,18 @@ static unsigned int dbs_enable; /* number of CPUs using this policy */ * cpu_hotplug lock should be taken before that. Note that cpu_hotplug lock * is recursive for the same process. -Venki */ -static DEFINE_MUTEX (dbs_mutex); -static DECLARE_DELAYED_WORK(dbs_work, do_dbs_timer); +static DEFINE_MUTEX(dbs_mutex); -struct dbs_tuners { +static struct workqueue_struct *kconservative_wq; + +static struct dbs_tuners { unsigned int sampling_rate; unsigned int sampling_down_factor; unsigned int up_threshold; unsigned int down_threshold; unsigned int ignore_nice; unsigned int freq_step; -}; - -static struct dbs_tuners dbs_tuners_ins = { +} dbs_tuners_ins = { .up_threshold = DEF_FREQUENCY_UP_THRESHOLD, .down_threshold = DEF_FREQUENCY_DOWN_THRESHOLD, .sampling_down_factor = DEF_SAMPLING_DOWN_FACTOR, @@ -102,18 +111,37 @@ static struct dbs_tuners dbs_tuners_ins = { .freq_step = 5, }; -static inline unsigned int get_cpu_idle_time(unsigned int cpu) +static inline cputime64_t get_cpu_idle_time_jiffy(unsigned int cpu, + cputime64_t *wall) { - unsigned int add_nice = 0, ret; + cputime64_t idle_time; + cputime64_t cur_wall_time; + cputime64_t busy_time; - if (dbs_tuners_ins.ignore_nice) - add_nice = kstat_cpu(cpu).cpustat.nice; + cur_wall_time = jiffies64_to_cputime64(get_jiffies_64()); + busy_time = cputime64_add(kstat_cpu(cpu).cpustat.user, + kstat_cpu(cpu).cpustat.system); + + busy_time = cputime64_add(busy_time, kstat_cpu(cpu).cpustat.irq); + busy_time = cputime64_add(busy_time, kstat_cpu(cpu).cpustat.softirq); + busy_time = cputime64_add(busy_time, kstat_cpu(cpu).cpustat.steal); + busy_time = cputime64_add(busy_time, kstat_cpu(cpu).cpustat.nice); + + idle_time = cputime64_sub(cur_wall_time, busy_time); + if (wall) + *wall = cur_wall_time; + + return idle_time; +} + +static inline cputime64_t get_cpu_idle_time(unsigned int cpu, cputime64_t *wall) +{ + u64 idle_time = get_cpu_idle_time_us(cpu, wall); - ret = kstat_cpu(cpu).cpustat.idle + - kstat_cpu(cpu).cpustat.iowait + - add_nice; + if (idle_time == -1ULL) + return get_cpu_idle_time_jiffy(cpu, wall); - return ret; + return idle_time; } /* keep track of frequency transitions */ @@ -125,10 +153,21 @@ dbs_cpufreq_notifier(struct notifier_block *nb, unsigned long val, struct cpu_dbs_info_s *this_dbs_info = &per_cpu(cpu_dbs_info, freq->cpu); + struct cpufreq_policy *policy; + if (!this_dbs_info->enable) return 0; - this_dbs_info->requested_freq = freq->new; + policy = this_dbs_info->cur_policy; + + /* + * we only care if our internally tracked freq moves outside + * the 'valid' ranges of freqency available to us otherwise + * we do not change it + */ + if (this_dbs_info->requested_freq > policy->max + || this_dbs_info->requested_freq < policy->min) + this_dbs_info->requested_freq = freq->new; return 0; } @@ -140,16 +179,31 @@ static struct notifier_block dbs_cpufreq_notifier_block = { /************************** sysfs interface ************************/ static ssize_t show_sampling_rate_max(struct cpufreq_policy *policy, char *buf) { - return sprintf (buf, "%u\n", MAX_SAMPLING_RATE); + static int print_once; + + if (!print_once) { + printk(KERN_INFO "CPUFREQ: conservative sampling_rate_max " + "sysfs file is deprecated - used by: %s\n", + current->comm); + print_once = 1; + } + return sprintf(buf, "%u\n", MAX_SAMPLING_RATE); } static ssize_t show_sampling_rate_min(struct cpufreq_policy *policy, char *buf) { - return sprintf (buf, "%u\n", MIN_SAMPLING_RATE); + static int print_once; + + if (!print_once) { + printk(KERN_INFO "CPUFREQ: conservative sampling_rate_max " + "sysfs file is deprecated - used by: %s\n", current->comm); + print_once = 1; + } + return sprintf(buf, "%u\n", MIN_SAMPLING_RATE); } -#define define_one_ro(_name) \ -static struct freq_attr _name = \ +#define define_one_ro(_name) \ +static struct freq_attr _name = \ __ATTR(_name, 0444, show_##_name, NULL) define_one_ro(sampling_rate_max); @@ -174,7 +228,8 @@ static ssize_t store_sampling_down_factor(struct cpufreq_policy *unused, { unsigned int input; int ret; - ret = sscanf (buf, "%u", &input); + ret = sscanf(buf, "%u", &input); + if (ret != 1 || input > MAX_SAMPLING_DOWN_FACTOR || input < 1) return -EINVAL; @@ -190,15 +245,13 @@ static ssize_t store_sampling_rate(struct cpufreq_policy *unused, { unsigned int input; int ret; - ret = sscanf (buf, "%u", &input); + ret = sscanf(buf, "%u", &input); - mutex_lock(&dbs_mutex); - if (ret != 1 || input > MAX_SAMPLING_RATE || input < MIN_SAMPLING_RATE) { - mutex_unlock(&dbs_mutex); + if (ret != 1) return -EINVAL; - } - dbs_tuners_ins.sampling_rate = input; + mutex_lock(&dbs_mutex); + dbs_tuners_ins.sampling_rate = max(input, minimum_sampling_rate()); mutex_unlock(&dbs_mutex); return count; @@ -209,10 +262,11 @@ static ssize_t store_up_threshold(struct cpufreq_policy *unused, { unsigned int input; int ret; - ret = sscanf (buf, "%u", &input); + ret = sscanf(buf, "%u", &input); mutex_lock(&dbs_mutex); - if (ret != 1 || input > 100 || input <= dbs_tuners_ins.down_threshold) { + if (ret != 1 || input > 100 || + input <= dbs_tuners_ins.down_threshold) { mutex_unlock(&dbs_mutex); return -EINVAL; } @@ -228,10 +282,12 @@ static ssize_t store_down_threshold(struct cpufreq_policy *unused, { unsigned int input; int ret; - ret = sscanf (buf, "%u", &input); + ret = sscanf(buf, "%u", &input); mutex_lock(&dbs_mutex); - if (ret != 1 || input > 100 || input >= dbs_tuners_ins.up_threshold) { + /* cannot be lower than 11 otherwise freq will not fall */ + if (ret != 1 || input < 11 || input > 100 || + input >= dbs_tuners_ins.up_threshold) { mutex_unlock(&dbs_mutex); return -EINVAL; } @@ -264,12 +320,14 @@ static ssize_t store_ignore_nice_load(struct cpufreq_policy *policy, } dbs_tuners_ins.ignore_nice = input; - /* we need to re-evaluate prev_cpu_idle_up and prev_cpu_idle_down */ + /* we need to re-evaluate prev_cpu_idle */ for_each_online_cpu(j) { - struct cpu_dbs_info_s *j_dbs_info; - j_dbs_info = &per_cpu(cpu_dbs_info, j); - j_dbs_info->prev_cpu_idle_up = get_cpu_idle_time(j); - j_dbs_info->prev_cpu_idle_down = j_dbs_info->prev_cpu_idle_up; + struct cpu_dbs_info_s *dbs_info; + dbs_info = &per_cpu(cpu_dbs_info, j); + dbs_info->prev_cpu_idle = get_cpu_idle_time(j, + &dbs_info->prev_cpu_wall); + if (dbs_tuners_ins.ignore_nice) + dbs_info->prev_cpu_nice = kstat_cpu(j).cpustat.nice; } mutex_unlock(&dbs_mutex); @@ -281,7 +339,6 @@ static ssize_t store_freq_step(struct cpufreq_policy *policy, { unsigned int input; int ret; - ret = sscanf(buf, "%u", &input); if (ret != 1) @@ -310,7 +367,7 @@ define_one_rw(down_threshold); define_one_rw(ignore_nice_load); define_one_rw(freq_step); -static struct attribute * dbs_attributes[] = { +static struct attribute *dbs_attributes[] = { &sampling_rate_max.attr, &sampling_rate_min.attr, &sampling_rate.attr, @@ -329,55 +386,78 @@ static struct attribute_group dbs_attr_group = { /************************** sysfs end ************************/ -static void dbs_check_cpu(int cpu) +static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info) { - unsigned int idle_ticks, up_idle_ticks, down_idle_ticks; - unsigned int tmp_idle_ticks, total_idle_ticks; + unsigned int load = 0; unsigned int freq_target; - unsigned int freq_down_sampling_rate; - struct cpu_dbs_info_s *this_dbs_info = &per_cpu(cpu_dbs_info, cpu); - struct cpufreq_policy *policy; - if (!this_dbs_info->enable) - return; + struct cpufreq_policy *policy; + unsigned int j; policy = this_dbs_info->cur_policy; /* - * The default safe range is 20% to 80% - * Every sampling_rate, we check - * - If current idle time is less than 20%, then we try to - * increase frequency - * Every sampling_rate*sampling_down_factor, we check - * - If current idle time is more than 80%, then we try to - * decrease frequency + * Every sampling_rate, we check, if current idle time is less + * than 20% (default), then we try to increase frequency + * Every sampling_rate*sampling_down_factor, we check, if current + * idle time is more than 80%, then we try to decrease frequency * * Any frequency increase takes it to the maximum frequency. * Frequency reduction happens at minimum steps of - * 5% (default) of max_frequency + * 5% (default) of maximum frequency */ - /* Check for frequency increase */ - idle_ticks = UINT_MAX; + /* Get Absolute Load */ + for_each_cpu(j, policy->cpus) { + struct cpu_dbs_info_s *j_dbs_info; + cputime64_t cur_wall_time, cur_idle_time; + unsigned int idle_time, wall_time; - /* Check for frequency increase */ - total_idle_ticks = get_cpu_idle_time(cpu); - tmp_idle_ticks = total_idle_ticks - - this_dbs_info->prev_cpu_idle_up; - this_dbs_info->prev_cpu_idle_up = total_idle_ticks; + j_dbs_info = &per_cpu(cpu_dbs_info, j); + + cur_idle_time = get_cpu_idle_time(j, &cur_wall_time); + + wall_time = (unsigned int) cputime64_sub(cur_wall_time, + j_dbs_info->prev_cpu_wall); + j_dbs_info->prev_cpu_wall = cur_wall_time; + + idle_time = (unsigned int) cputime64_sub(cur_idle_time, + j_dbs_info->prev_cpu_idle); + j_dbs_info->prev_cpu_idle = cur_idle_time; + + if (dbs_tuners_ins.ignore_nice) { + cputime64_t cur_nice; + unsigned long cur_nice_jiffies; + + cur_nice = cputime64_sub(kstat_cpu(j).cpustat.nice, + j_dbs_info->prev_cpu_nice); + /* + * Assumption: nice time between sampling periods will + * be less than 2^32 jiffies for 32 bit sys + */ + cur_nice_jiffies = (unsigned long) + cputime64_to_jiffies64(cur_nice); - if (tmp_idle_ticks < idle_ticks) - idle_ticks = tmp_idle_ticks; + j_dbs_info->prev_cpu_nice = kstat_cpu(j).cpustat.nice; + idle_time += jiffies_to_usecs(cur_nice_jiffies); + } + + if (unlikely(!wall_time || wall_time < idle_time)) + continue; + + load = 100 * (wall_time - idle_time) / wall_time; + } - /* Scale idle ticks by 100 and compare with up and down ticks */ - idle_ticks *= 100; - up_idle_ticks = (100 - dbs_tuners_ins.up_threshold) * - usecs_to_jiffies(dbs_tuners_ins.sampling_rate); + /* + * break out if we 'cannot' reduce the speed as the user might + * want freq_step to be zero + */ + if (dbs_tuners_ins.freq_step == 0) + return; - if (idle_ticks < up_idle_ticks) { + /* Check for frequency increase */ + if (load > dbs_tuners_ins.up_threshold) { this_dbs_info->down_skip = 0; - this_dbs_info->prev_cpu_idle_down = - this_dbs_info->prev_cpu_idle_up; /* if we are already at full speed then break out early */ if (this_dbs_info->requested_freq == policy->max) @@ -398,49 +478,24 @@ static void dbs_check_cpu(int cpu) return; } - /* Check for frequency decrease */ - this_dbs_info->down_skip++; - if (this_dbs_info->down_skip < dbs_tuners_ins.sampling_down_factor) - return; - - /* Check for frequency decrease */ - total_idle_ticks = this_dbs_info->prev_cpu_idle_up; - tmp_idle_ticks = total_idle_ticks - - this_dbs_info->prev_cpu_idle_down; - this_dbs_info->prev_cpu_idle_down = total_idle_ticks; - - if (tmp_idle_ticks < idle_ticks) - idle_ticks = tmp_idle_ticks; - - /* Scale idle ticks by 100 and compare with up and down ticks */ - idle_ticks *= 100; - this_dbs_info->down_skip = 0; - - freq_down_sampling_rate = dbs_tuners_ins.sampling_rate * - dbs_tuners_ins.sampling_down_factor; - down_idle_ticks = (100 - dbs_tuners_ins.down_threshold) * - usecs_to_jiffies(freq_down_sampling_rate); - - if (idle_ticks > down_idle_ticks) { - /* - * if we are already at the lowest speed then break out early - * or if we 'cannot' reduce the speed as the user might want - * freq_target to be zero - */ - if (this_dbs_info->requested_freq == policy->min - || dbs_tuners_ins.freq_step == 0) - return; - + /* + * The optimal frequency is the frequency that is the lowest that + * can support the current CPU usage without triggering the up + * policy. To be safe, we focus 10 points under the threshold. + */ + if (load < (dbs_tuners_ins.down_threshold - 10)) { freq_target = (dbs_tuners_ins.freq_step * policy->max) / 100; - /* max freq cannot be less than 100. But who knows.... */ - if (unlikely(freq_target == 0)) - freq_target = 5; - this_dbs_info->requested_freq -= freq_target; if (this_dbs_info->requested_freq < policy->min) this_dbs_info->requested_freq = policy->min; + /* + * if we cannot reduce the frequency anymore, break out early + */ + if (policy->cur == policy->min) + return; + __cpufreq_driver_target(policy, this_dbs_info->requested_freq, CPUFREQ_RELATION_H); return; @@ -449,27 +504,45 @@ static void dbs_check_cpu(int cpu) static void do_dbs_timer(struct work_struct *work) { - int i; - mutex_lock(&dbs_mutex); - for_each_online_cpu(i) - dbs_check_cpu(i); - schedule_delayed_work(&dbs_work, - usecs_to_jiffies(dbs_tuners_ins.sampling_rate)); - mutex_unlock(&dbs_mutex); + struct cpu_dbs_info_s *dbs_info = + container_of(work, struct cpu_dbs_info_s, work.work); + unsigned int cpu = dbs_info->cpu; + + /* We want all CPUs to do sampling nearly on same jiffy */ + int delay = usecs_to_jiffies(dbs_tuners_ins.sampling_rate); + + delay -= jiffies % delay; + + if (lock_policy_rwsem_write(cpu) < 0) + return; + + if (!dbs_info->enable) { + unlock_policy_rwsem_write(cpu); + return; + } + + dbs_check_cpu(dbs_info); + + queue_delayed_work_on(cpu, kconservative_wq, &dbs_info->work, delay); + unlock_policy_rwsem_write(cpu); } -static inline void dbs_timer_init(void) +static inline void dbs_timer_init(struct cpu_dbs_info_s *dbs_info) { - init_timer_deferrable(&dbs_work.timer); - schedule_delayed_work(&dbs_work, - usecs_to_jiffies(dbs_tuners_ins.sampling_rate)); - return; + /* We want all CPUs to do sampling nearly on same jiffy */ + int delay = usecs_to_jiffies(dbs_tuners_ins.sampling_rate); + delay -= jiffies % delay; + + dbs_info->enable = 1; + INIT_DELAYED_WORK_DEFERRABLE(&dbs_info->work, do_dbs_timer); + queue_delayed_work_on(dbs_info->cpu, kconservative_wq, &dbs_info->work, + delay); } -static inline void dbs_timer_exit(void) +static inline void dbs_timer_exit(struct cpu_dbs_info_s *dbs_info) { - cancel_delayed_work(&dbs_work); - return; + dbs_info->enable = 0; + cancel_delayed_work(&dbs_info->work); } static int cpufreq_governor_dbs(struct cpufreq_policy *policy, @@ -503,11 +576,13 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy, j_dbs_info = &per_cpu(cpu_dbs_info, j); j_dbs_info->cur_policy = policy; - j_dbs_info->prev_cpu_idle_up = get_cpu_idle_time(cpu); - j_dbs_info->prev_cpu_idle_down - = j_dbs_info->prev_cpu_idle_up; + j_dbs_info->prev_cpu_idle = get_cpu_idle_time(j, + &j_dbs_info->prev_cpu_wall); + if (dbs_tuners_ins.ignore_nice) { + j_dbs_info->prev_cpu_nice = + kstat_cpu(j).cpustat.nice; + } } - this_dbs_info->enable = 1; this_dbs_info->down_skip = 0; this_dbs_info->requested_freq = policy->cur; @@ -523,38 +598,36 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy, if (latency == 0) latency = 1; - def_sampling_rate = 10 * latency * - DEF_SAMPLING_RATE_LATENCY_MULTIPLIER; - - if (def_sampling_rate < MIN_STAT_SAMPLING_RATE) - def_sampling_rate = MIN_STAT_SAMPLING_RATE; + def_sampling_rate = + max(latency * LATENCY_MULTIPLIER, + MIN_STAT_SAMPLING_RATE); dbs_tuners_ins.sampling_rate = def_sampling_rate; - dbs_timer_init(); cpufreq_register_notifier( &dbs_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER); } + dbs_timer_init(this_dbs_info); mutex_unlock(&dbs_mutex); + break; case CPUFREQ_GOV_STOP: mutex_lock(&dbs_mutex); - this_dbs_info->enable = 0; + dbs_timer_exit(this_dbs_info); sysfs_remove_group(&policy->kobj, &dbs_attr_group); dbs_enable--; + /* * Stop the timerschedule work, when this governor * is used for first time */ - if (dbs_enable == 0) { - dbs_timer_exit(); + if (dbs_enable == 0) cpufreq_unregister_notifier( &dbs_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER); - } mutex_unlock(&dbs_mutex); @@ -571,6 +644,7 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy, this_dbs_info->cur_policy, policy->min, CPUFREQ_RELATION_L); mutex_unlock(&dbs_mutex); + break; } return 0; @@ -588,23 +662,33 @@ struct cpufreq_governor cpufreq_gov_conservative = { static int __init cpufreq_gov_dbs_init(void) { - return cpufreq_register_governor(&cpufreq_gov_conservative); + int err; + + kconservative_wq = create_workqueue("kconservative"); + if (!kconservative_wq) { + printk(KERN_ERR "Creation of kconservative failed\n"); + return -EFAULT; + } + + err = cpufreq_register_governor(&cpufreq_gov_conservative); + if (err) + destroy_workqueue(kconservative_wq); + + return err; } static void __exit cpufreq_gov_dbs_exit(void) { - /* Make sure that the scheduled work is indeed not running */ - flush_scheduled_work(); - cpufreq_unregister_governor(&cpufreq_gov_conservative); + destroy_workqueue(kconservative_wq); } -MODULE_AUTHOR ("Alexander Clouter <alex-kernel@digriz.org.uk>"); -MODULE_DESCRIPTION ("'cpufreq_conservative' - A dynamic cpufreq governor for " +MODULE_AUTHOR("Alexander Clouter <alex@digriz.org.uk>"); +MODULE_DESCRIPTION("'cpufreq_conservative' - A dynamic cpufreq governor for " "Low Latency Frequency Transition capable processors " "optimised for use in a battery environment"); -MODULE_LICENSE ("GPL"); +MODULE_LICENSE("GPL"); #ifdef CONFIG_CPU_FREQ_DEFAULT_GOV_CONSERVATIVE fs_initcall(cpufreq_gov_dbs_init); diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c index 6f45b1658a6..338f428a15b 100644 --- a/drivers/cpufreq/cpufreq_ondemand.c +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -21,6 +21,7 @@ #include <linux/hrtimer.h> #include <linux/tick.h> #include <linux/ktime.h> +#include <linux/sched.h> /* * dbs is used in this file as a shortform for demandbased switching @@ -51,8 +52,20 @@ static unsigned int def_sampling_rate; (MIN_SAMPLING_RATE_RATIO * jiffies_to_usecs(10)) #define MIN_SAMPLING_RATE \ (def_sampling_rate / MIN_SAMPLING_RATE_RATIO) +/* Above MIN_SAMPLING_RATE will vanish with its sysfs file soon + * Define the minimal settable sampling rate to the greater of: + * - "HW transition latency" * 100 (same as default sampling / 10) + * - MIN_STAT_SAMPLING_RATE + * To avoid that userspace shoots itself. +*/ +static unsigned int minimum_sampling_rate(void) +{ + return max(def_sampling_rate / 10, MIN_STAT_SAMPLING_RATE); +} + +/* This will also vanish soon with removing sampling_rate_max */ #define MAX_SAMPLING_RATE (500 * def_sampling_rate) -#define DEF_SAMPLING_RATE_LATENCY_MULTIPLIER (1000) +#define LATENCY_MULTIPLIER (1000) #define TRANSITION_LATENCY_LIMIT (10 * 1000 * 1000) static void do_dbs_timer(struct work_struct *work); @@ -65,14 +78,14 @@ struct cpu_dbs_info_s { cputime64_t prev_cpu_wall; cputime64_t prev_cpu_nice; struct cpufreq_policy *cur_policy; - struct delayed_work work; + struct delayed_work work; struct cpufreq_frequency_table *freq_table; unsigned int freq_lo; unsigned int freq_lo_jiffies; unsigned int freq_hi_jiffies; int cpu; unsigned int enable:1, - sample_type:1; + sample_type:1; }; static DEFINE_PER_CPU(struct cpu_dbs_info_s, cpu_dbs_info); @@ -203,12 +216,28 @@ static void ondemand_powersave_bias_init(void) /************************** sysfs interface ************************/ static ssize_t show_sampling_rate_max(struct cpufreq_policy *policy, char *buf) { - return sprintf (buf, "%u\n", MAX_SAMPLING_RATE); + static int print_once; + + if (!print_once) { + printk(KERN_INFO "CPUFREQ: ondemand sampling_rate_max " + "sysfs file is deprecated - used by: %s\n", + current->comm); + print_once = 1; + } + return sprintf(buf, "%u\n", MAX_SAMPLING_RATE); } static ssize_t show_sampling_rate_min(struct cpufreq_policy *policy, char *buf) { - return sprintf (buf, "%u\n", MIN_SAMPLING_RATE); + static int print_once; + + if (!print_once) { + printk(KERN_INFO "CPUFREQ: ondemand sampling_rate_min " + "sysfs file is deprecated - used by: %s\n", + current->comm); + print_once = 1; + } + return sprintf(buf, "%u\n", MIN_SAMPLING_RATE); } #define define_one_ro(_name) \ @@ -238,13 +267,11 @@ static ssize_t store_sampling_rate(struct cpufreq_policy *unused, ret = sscanf(buf, "%u", &input); mutex_lock(&dbs_mutex); - if (ret != 1 || input > MAX_SAMPLING_RATE - || input < MIN_SAMPLING_RATE) { + if (ret != 1) { mutex_unlock(&dbs_mutex); return -EINVAL; } - - dbs_tuners_ins.sampling_rate = input; + dbs_tuners_ins.sampling_rate = max(input, minimum_sampling_rate()); mutex_unlock(&dbs_mutex); return count; @@ -279,14 +306,14 @@ static ssize_t store_ignore_nice_load(struct cpufreq_policy *policy, unsigned int j; ret = sscanf(buf, "%u", &input); - if ( ret != 1 ) + if (ret != 1) return -EINVAL; - if ( input > 1 ) + if (input > 1) input = 1; mutex_lock(&dbs_mutex); - if ( input == dbs_tuners_ins.ignore_nice ) { /* nothing to do */ + if (input == dbs_tuners_ins.ignore_nice) { /* nothing to do */ mutex_unlock(&dbs_mutex); return count; } @@ -337,7 +364,7 @@ define_one_rw(up_threshold); define_one_rw(ignore_nice_load); define_one_rw(powersave_bias); -static struct attribute * dbs_attributes[] = { +static struct attribute *dbs_attributes[] = { &sampling_rate_max.attr, &sampling_rate_min.attr, &sampling_rate.attr, @@ -512,8 +539,7 @@ static void do_dbs_timer(struct work_struct *work) } } else { __cpufreq_driver_target(dbs_info->cur_policy, - dbs_info->freq_lo, - CPUFREQ_RELATION_H); + dbs_info->freq_lo, CPUFREQ_RELATION_H); } queue_delayed_work_on(cpu, kondemand_wq, &dbs_info->work, delay); unlock_policy_rwsem_write(cpu); @@ -530,7 +556,7 @@ static inline void dbs_timer_init(struct cpu_dbs_info_s *dbs_info) dbs_info->sample_type = DBS_NORMAL_SAMPLE; INIT_DELAYED_WORK_DEFERRABLE(&dbs_info->work, do_dbs_timer); queue_delayed_work_on(dbs_info->cpu, kondemand_wq, &dbs_info->work, - delay); + delay); } static inline void dbs_timer_exit(struct cpu_dbs_info_s *dbs_info) @@ -591,11 +617,9 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy, if (latency == 0) latency = 1; - def_sampling_rate = latency * - DEF_SAMPLING_RATE_LATENCY_MULTIPLIER; - - if (def_sampling_rate < MIN_STAT_SAMPLING_RATE) - def_sampling_rate = MIN_STAT_SAMPLING_RATE; + def_sampling_rate = + max(latency * LATENCY_MULTIPLIER, + MIN_STAT_SAMPLING_RATE); dbs_tuners_ins.sampling_rate = def_sampling_rate; } @@ -617,12 +641,10 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy, mutex_lock(&dbs_mutex); if (policy->max < this_dbs_info->cur_policy->cur) __cpufreq_driver_target(this_dbs_info->cur_policy, - policy->max, - CPUFREQ_RELATION_H); + policy->max, CPUFREQ_RELATION_H); else if (policy->min > this_dbs_info->cur_policy->cur) __cpufreq_driver_target(this_dbs_info->cur_policy, - policy->min, - CPUFREQ_RELATION_L); + policy->min, CPUFREQ_RELATION_L); mutex_unlock(&dbs_mutex); break; } @@ -677,7 +699,7 @@ static void __exit cpufreq_gov_dbs_exit(void) MODULE_AUTHOR("Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>"); MODULE_AUTHOR("Alexey Starikovskiy <alexey.y.starikovskiy@intel.com>"); MODULE_DESCRIPTION("'cpufreq_ondemand' - A dynamic cpufreq governor for " - "Low Latency Frequency Transition capable processors"); + "Low Latency Frequency Transition capable processors"); MODULE_LICENSE("GPL"); #ifdef CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND diff --git a/drivers/cpufreq/cpufreq_stats.c b/drivers/cpufreq/cpufreq_stats.c index c0ff97d375d..5a62d678dd1 100644 --- a/drivers/cpufreq/cpufreq_stats.c +++ b/drivers/cpufreq/cpufreq_stats.c @@ -2,7 +2,7 @@ * drivers/cpufreq/cpufreq_stats.c * * Copyright (C) 2003-2004 Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>. - * (C) 2004 Zou Nan hai <nanhai.zou@intel.com>. + * (C) 2004 Zou Nan hai <nanhai.zou@intel.com>. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -23,7 +23,7 @@ static spinlock_t cpufreq_stats_lock; -#define CPUFREQ_STATDEVICE_ATTR(_name,_mode,_show) \ +#define CPUFREQ_STATDEVICE_ATTR(_name, _mode, _show) \ static struct freq_attr _attr_##_name = {\ .attr = {.name = __stringify(_name), .mode = _mode, }, \ .show = _show,\ @@ -50,8 +50,7 @@ struct cpufreq_stats_attribute { ssize_t(*show) (struct cpufreq_stats *, char *); }; -static int -cpufreq_stats_update (unsigned int cpu) +static int cpufreq_stats_update(unsigned int cpu) { struct cpufreq_stats *stat; unsigned long long cur_time; @@ -68,8 +67,7 @@ cpufreq_stats_update (unsigned int cpu) return 0; } -static ssize_t -show_total_trans(struct cpufreq_policy *policy, char *buf) +static ssize_t show_total_trans(struct cpufreq_policy *policy, char *buf) { struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->cpu); if (!stat) @@ -78,8 +76,7 @@ show_total_trans(struct cpufreq_policy *policy, char *buf) per_cpu(cpufreq_stats_table, stat->cpu)->total_trans); } -static ssize_t -show_time_in_state(struct cpufreq_policy *policy, char *buf) +static ssize_t show_time_in_state(struct cpufreq_policy *policy, char *buf) { ssize_t len = 0; int i; @@ -89,14 +86,14 @@ show_time_in_state(struct cpufreq_policy *policy, char *buf) cpufreq_stats_update(stat->cpu); for (i = 0; i < stat->state_num; i++) { len += sprintf(buf + len, "%u %llu\n", stat->freq_table[i], - (unsigned long long)cputime64_to_clock_t(stat->time_in_state[i])); + (unsigned long long) + cputime64_to_clock_t(stat->time_in_state[i])); } return len; } #ifdef CONFIG_CPU_FREQ_STAT_DETAILS -static ssize_t -show_trans_table(struct cpufreq_policy *policy, char *buf) +static ssize_t show_trans_table(struct cpufreq_policy *policy, char *buf) { ssize_t len = 0; int i, j; @@ -139,11 +136,11 @@ show_trans_table(struct cpufreq_policy *policy, char *buf) return PAGE_SIZE; return len; } -CPUFREQ_STATDEVICE_ATTR(trans_table,0444,show_trans_table); +CPUFREQ_STATDEVICE_ATTR(trans_table, 0444, show_trans_table); #endif -CPUFREQ_STATDEVICE_ATTR(total_trans,0444,show_total_trans); -CPUFREQ_STATDEVICE_ATTR(time_in_state,0444,show_time_in_state); +CPUFREQ_STATDEVICE_ATTR(total_trans, 0444, show_total_trans); +CPUFREQ_STATDEVICE_ATTR(time_in_state, 0444, show_time_in_state); static struct attribute *default_attrs[] = { &_attr_total_trans.attr, @@ -158,8 +155,7 @@ static struct attribute_group stats_attr_group = { .name = "stats" }; -static int -freq_table_get_index(struct cpufreq_stats *stat, unsigned int freq) +static int freq_table_get_index(struct cpufreq_stats *stat, unsigned int freq) { int index; for (index = 0; index < stat->max_state; index++) @@ -183,8 +179,7 @@ static void cpufreq_stats_free_table(unsigned int cpu) cpufreq_cpu_put(policy); } -static int -cpufreq_stats_create_table (struct cpufreq_policy *policy, +static int cpufreq_stats_create_table(struct cpufreq_policy *policy, struct cpufreq_frequency_table *table) { unsigned int i, j, count = 0, ret = 0; @@ -194,7 +189,8 @@ cpufreq_stats_create_table (struct cpufreq_policy *policy, unsigned int cpu = policy->cpu; if (per_cpu(cpufreq_stats_table, cpu)) return -EBUSY; - if ((stat = kzalloc(sizeof(struct cpufreq_stats), GFP_KERNEL)) == NULL) + stat = kzalloc(sizeof(struct cpufreq_stats), GFP_KERNEL); + if ((stat) == NULL) return -ENOMEM; data = cpufreq_cpu_get(cpu); @@ -203,13 +199,14 @@ cpufreq_stats_create_table (struct cpufreq_policy *policy, goto error_get_fail; } - if ((ret = sysfs_create_group(&data->kobj, &stats_attr_group))) + ret = sysfs_create_group(&data->kobj, &stats_attr_group); + if (ret) goto error_out; stat->cpu = cpu; per_cpu(cpufreq_stats_table, cpu) = stat; - for (i=0; table[i].frequency != CPUFREQ_TABLE_END; i++) { + for (i = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) { unsigned int freq = table[i].frequency; if (freq == CPUFREQ_ENTRY_INVALID) continue; @@ -255,9 +252,8 @@ error_get_fail: return ret; } -static int -cpufreq_stat_notifier_policy (struct notifier_block *nb, unsigned long val, - void *data) +static int cpufreq_stat_notifier_policy(struct notifier_block *nb, + unsigned long val, void *data) { int ret; struct cpufreq_policy *policy = data; @@ -268,14 +264,14 @@ cpufreq_stat_notifier_policy (struct notifier_block *nb, unsigned long val, table = cpufreq_frequency_get_table(cpu); if (!table) return 0; - if ((ret = cpufreq_stats_create_table(policy, table))) + ret = cpufreq_stats_create_table(policy, table); + if (ret) return ret; return 0; } -static int -cpufreq_stat_notifier_trans (struct notifier_block *nb, unsigned long val, - void *data) +static int cpufreq_stat_notifier_trans(struct notifier_block *nb, + unsigned long val, void *data) { struct cpufreq_freqs *freq = data; struct cpufreq_stats *stat; @@ -340,19 +336,20 @@ static struct notifier_block notifier_trans_block = { .notifier_call = cpufreq_stat_notifier_trans }; -static int -__init cpufreq_stats_init(void) +static int __init cpufreq_stats_init(void) { int ret; unsigned int cpu; spin_lock_init(&cpufreq_stats_lock); - if ((ret = cpufreq_register_notifier(¬ifier_policy_block, - CPUFREQ_POLICY_NOTIFIER))) + ret = cpufreq_register_notifier(¬ifier_policy_block, + CPUFREQ_POLICY_NOTIFIER); + if (ret) return ret; - if ((ret = cpufreq_register_notifier(¬ifier_trans_block, - CPUFREQ_TRANSITION_NOTIFIER))) { + ret = cpufreq_register_notifier(¬ifier_trans_block, + CPUFREQ_TRANSITION_NOTIFIER); + if (ret) { cpufreq_unregister_notifier(¬ifier_policy_block, CPUFREQ_POLICY_NOTIFIER); return ret; @@ -364,8 +361,7 @@ __init cpufreq_stats_init(void) } return 0; } -static void -__exit cpufreq_stats_exit(void) +static void __exit cpufreq_stats_exit(void) { unsigned int cpu; @@ -379,10 +375,10 @@ __exit cpufreq_stats_exit(void) } } -MODULE_AUTHOR ("Zou Nan hai <nanhai.zou@intel.com>"); -MODULE_DESCRIPTION ("'cpufreq_stats' - A driver to export cpufreq stats " +MODULE_AUTHOR("Zou Nan hai <nanhai.zou@intel.com>"); +MODULE_DESCRIPTION("'cpufreq_stats' - A driver to export cpufreq stats " "through sysfs filesystem"); -MODULE_LICENSE ("GPL"); +MODULE_LICENSE("GPL"); module_init(cpufreq_stats_init); module_exit(cpufreq_stats_exit); diff --git a/drivers/cpufreq/cpufreq_userspace.c b/drivers/cpufreq/cpufreq_userspace.c index 1442bbada05..66d2d1d6c80 100644 --- a/drivers/cpufreq/cpufreq_userspace.c +++ b/drivers/cpufreq/cpufreq_userspace.c @@ -24,9 +24,6 @@ #include <linux/sysfs.h> #include <linux/mutex.h> -#include <asm/uaccess.h> - - /** * A few values needed by the userspace governor */ @@ -37,7 +34,7 @@ static DEFINE_PER_CPU(unsigned int, cpu_set_freq); /* CPU freq desired by userspace */ static DEFINE_PER_CPU(unsigned int, cpu_is_managed); -static DEFINE_MUTEX (userspace_mutex); +static DEFINE_MUTEX(userspace_mutex); static int cpus_using_userspace_governor; #define dprintk(msg...) \ @@ -46,9 +43,9 @@ static int cpus_using_userspace_governor; /* keep track of frequency transitions */ static int userspace_cpufreq_notifier(struct notifier_block *nb, unsigned long val, - void *data) + void *data) { - struct cpufreq_freqs *freq = data; + struct cpufreq_freqs *freq = data; if (!per_cpu(cpu_is_managed, freq->cpu)) return 0; @@ -57,11 +54,11 @@ userspace_cpufreq_notifier(struct notifier_block *nb, unsigned long val, freq->cpu, freq->new); per_cpu(cpu_cur_freq, freq->cpu) = freq->new; - return 0; + return 0; } static struct notifier_block userspace_cpufreq_notifier_block = { - .notifier_call = userspace_cpufreq_notifier + .notifier_call = userspace_cpufreq_notifier }; @@ -93,8 +90,11 @@ static int cpufreq_set(struct cpufreq_policy *policy, unsigned int freq) * We're safe from concurrent calls to ->target() here * as we hold the userspace_mutex lock. If we were calling * cpufreq_driver_target, a deadlock situation might occur: - * A: cpufreq_set (lock userspace_mutex) -> cpufreq_driver_target(lock policy->lock) - * B: cpufreq_set_policy(lock policy->lock) -> __cpufreq_governor -> cpufreq_governor_userspace (lock userspace_mutex) + * A: cpufreq_set (lock userspace_mutex) -> + * cpufreq_driver_target(lock policy->lock) + * B: cpufreq_set_policy(lock policy->lock) -> + * __cpufreq_governor -> + * cpufreq_governor_userspace (lock userspace_mutex) */ ret = __cpufreq_driver_target(policy, freq, CPUFREQ_RELATION_L); @@ -210,9 +210,10 @@ static void __exit cpufreq_gov_userspace_exit(void) } -MODULE_AUTHOR ("Dominik Brodowski <linux@brodo.de>, Russell King <rmk@arm.linux.org.uk>"); -MODULE_DESCRIPTION ("CPUfreq policy governor 'userspace'"); -MODULE_LICENSE ("GPL"); +MODULE_AUTHOR("Dominik Brodowski <linux@brodo.de>, " + "Russell King <rmk@arm.linux.org.uk>"); +MODULE_DESCRIPTION("CPUfreq policy governor 'userspace'"); +MODULE_LICENSE("GPL"); #ifdef CONFIG_CPU_FREQ_DEFAULT_GOV_USERSPACE fs_initcall(cpufreq_gov_userspace_init); diff --git a/drivers/cpufreq/freq_table.c b/drivers/cpufreq/freq_table.c index 9071d80fbba..a9bd3a05a68 100644 --- a/drivers/cpufreq/freq_table.c +++ b/drivers/cpufreq/freq_table.c @@ -28,7 +28,7 @@ int cpufreq_frequency_table_cpuinfo(struct cpufreq_policy *policy, unsigned int max_freq = 0; unsigned int i; - for (i=0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { + for (i = 0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { unsigned int freq = table[i].frequency; if (freq == CPUFREQ_ENTRY_INVALID) { dprintk("table entry %u is invalid, skipping\n", i); @@ -70,7 +70,7 @@ int cpufreq_frequency_table_verify(struct cpufreq_policy *policy, cpufreq_verify_within_limits(policy, policy->cpuinfo.min_freq, policy->cpuinfo.max_freq); - for (i=0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { + for (i = 0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { unsigned int freq = table[i].frequency; if (freq == CPUFREQ_ENTRY_INVALID) continue; @@ -125,13 +125,13 @@ int cpufreq_frequency_table_target(struct cpufreq_policy *policy, if (!cpu_online(policy->cpu)) return -EINVAL; - for (i=0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { + for (i = 0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { unsigned int freq = table[i].frequency; if (freq == CPUFREQ_ENTRY_INVALID) continue; if ((freq < policy->min) || (freq > policy->max)) continue; - switch(relation) { + switch (relation) { case CPUFREQ_RELATION_H: if (freq <= target_freq) { if (freq >= optimal.frequency) { @@ -178,7 +178,7 @@ static DEFINE_PER_CPU(struct cpufreq_frequency_table *, show_table); /** * show_available_freqs - show available frequencies for the specified CPU */ -static ssize_t show_available_freqs (struct cpufreq_policy *policy, char *buf) +static ssize_t show_available_freqs(struct cpufreq_policy *policy, char *buf) { unsigned int i = 0; unsigned int cpu = policy->cpu; @@ -190,7 +190,7 @@ static ssize_t show_available_freqs (struct cpufreq_policy *policy, char *buf) table = per_cpu(show_table, cpu); - for (i=0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { + for (i = 0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { if (table[i].frequency == CPUFREQ_ENTRY_INVALID) continue; count += sprintf(&buf[count], "%d ", table[i].frequency); @@ -234,6 +234,6 @@ struct cpufreq_frequency_table *cpufreq_frequency_get_table(unsigned int cpu) } EXPORT_SYMBOL_GPL(cpufreq_frequency_get_table); -MODULE_AUTHOR ("Dominik Brodowski <linux@brodo.de>"); -MODULE_DESCRIPTION ("CPUfreq frequency table helpers"); -MODULE_LICENSE ("GPL"); +MODULE_AUTHOR("Dominik Brodowski <linux@brodo.de>"); +MODULE_DESCRIPTION("CPUfreq frequency table helpers"); +MODULE_LICENSE("GPL"); diff --git a/drivers/crypto/Kconfig b/drivers/crypto/Kconfig index e522144cba3..01afd758072 100644 --- a/drivers/crypto/Kconfig +++ b/drivers/crypto/Kconfig @@ -86,7 +86,7 @@ config ZCRYPT_MONOLITHIC config CRYPTO_SHA1_S390 tristate "SHA1 digest algorithm" depends on S390 - select CRYPTO_ALGAPI + select CRYPTO_HASH help This is the s390 hardware accelerated implementation of the SHA-1 secure hash standard (FIPS 180-1/DFIPS 180-2). @@ -94,7 +94,7 @@ config CRYPTO_SHA1_S390 config CRYPTO_SHA256_S390 tristate "SHA256 digest algorithm" depends on S390 - select CRYPTO_ALGAPI + select CRYPTO_HASH help This is the s390 hardware accelerated implementation of the SHA256 secure hash standard (DFIPS 180-2). @@ -105,7 +105,7 @@ config CRYPTO_SHA256_S390 config CRYPTO_SHA512_S390 tristate "SHA384 and SHA512 digest algorithm" depends on S390 - select CRYPTO_ALGAPI + select CRYPTO_HASH help This is the s390 hardware accelerated implementation of the SHA512 secure hash standard. @@ -200,4 +200,13 @@ config CRYPTO_DEV_IXP4XX help Driver for the IXP4xx NPE crypto engine. +config CRYPTO_DEV_PPC4XX + tristate "Driver AMCC PPC4xx crypto accelerator" + depends on PPC && 4xx + select CRYPTO_HASH + select CRYPTO_ALGAPI + select CRYPTO_BLKCIPHER + help + This option allows you to have support for AMCC crypto acceleration. + endif # CRYPTO_HW diff --git a/drivers/crypto/Makefile b/drivers/crypto/Makefile index 73557b2968d..9bf4a2bc884 100644 --- a/drivers/crypto/Makefile +++ b/drivers/crypto/Makefile @@ -4,3 +4,4 @@ obj-$(CONFIG_CRYPTO_DEV_GEODE) += geode-aes.o obj-$(CONFIG_CRYPTO_DEV_HIFN_795X) += hifn_795x.o obj-$(CONFIG_CRYPTO_DEV_TALITOS) += talitos.o obj-$(CONFIG_CRYPTO_DEV_IXP4XX) += ixp4xx_crypto.o +obj-$(CONFIG_CRYPTO_DEV_PPC4XX) += amcc/ diff --git a/drivers/crypto/amcc/Makefile b/drivers/crypto/amcc/Makefile new file mode 100644 index 00000000000..aa376e8d5ed --- /dev/null +++ b/drivers/crypto/amcc/Makefile @@ -0,0 +1,2 @@ +obj-$(CONFIG_CRYPTO_DEV_PPC4XX) += crypto4xx.o +crypto4xx-objs := crypto4xx_core.o crypto4xx_alg.o crypto4xx_sa.o diff --git a/drivers/crypto/amcc/crypto4xx_alg.c b/drivers/crypto/amcc/crypto4xx_alg.c new file mode 100644 index 00000000000..61b6e1bec8c --- /dev/null +++ b/drivers/crypto/amcc/crypto4xx_alg.c @@ -0,0 +1,293 @@ +/** + * AMCC SoC PPC4xx Crypto Driver + * + * Copyright (c) 2008 Applied Micro Circuits Corporation. + * All rights reserved. James Hsiao <jhsiao@amcc.com> + * + * 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. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * This file implements the Linux crypto algorithms. + */ + +#include <linux/kernel.h> +#include <linux/interrupt.h> +#include <linux/spinlock_types.h> +#include <linux/scatterlist.h> +#include <linux/crypto.h> +#include <linux/hash.h> +#include <crypto/internal/hash.h> +#include <linux/dma-mapping.h> +#include <crypto/algapi.h> +#include <crypto/aes.h> +#include <crypto/sha.h> +#include "crypto4xx_reg_def.h" +#include "crypto4xx_sa.h" +#include "crypto4xx_core.h" + +void set_dynamic_sa_command_0(struct dynamic_sa_ctl *sa, u32 save_h, + u32 save_iv, u32 ld_h, u32 ld_iv, u32 hdr_proc, + u32 h, u32 c, u32 pad_type, u32 op_grp, u32 op, + u32 dir) +{ + sa->sa_command_0.w = 0; + sa->sa_command_0.bf.save_hash_state = save_h; + sa->sa_command_0.bf.save_iv = save_iv; + sa->sa_command_0.bf.load_hash_state = ld_h; + sa->sa_command_0.bf.load_iv = ld_iv; + sa->sa_command_0.bf.hdr_proc = hdr_proc; + sa->sa_command_0.bf.hash_alg = h; + sa->sa_command_0.bf.cipher_alg = c; + sa->sa_command_0.bf.pad_type = pad_type & 3; + sa->sa_command_0.bf.extend_pad = pad_type >> 2; + sa->sa_command_0.bf.op_group = op_grp; + sa->sa_command_0.bf.opcode = op; + sa->sa_command_0.bf.dir = dir; +} + +void set_dynamic_sa_command_1(struct dynamic_sa_ctl *sa, u32 cm, u32 hmac_mc, + u32 cfb, u32 esn, u32 sn_mask, u32 mute, + u32 cp_pad, u32 cp_pay, u32 cp_hdr) +{ + sa->sa_command_1.w = 0; + sa->sa_command_1.bf.crypto_mode31 = (cm & 4) >> 2; + sa->sa_command_1.bf.crypto_mode9_8 = cm & 3; + sa->sa_command_1.bf.feedback_mode = cfb, + sa->sa_command_1.bf.sa_rev = 1; + sa->sa_command_1.bf.extended_seq_num = esn; + sa->sa_command_1.bf.seq_num_mask = sn_mask; + sa->sa_command_1.bf.mutable_bit_proc = mute; + sa->sa_command_1.bf.copy_pad = cp_pad; + sa->sa_command_1.bf.copy_payload = cp_pay; + sa->sa_command_1.bf.copy_hdr = cp_hdr; +} + +int crypto4xx_encrypt(struct ablkcipher_request *req) +{ + struct crypto4xx_ctx *ctx = crypto_tfm_ctx(req->base.tfm); + + ctx->direction = DIR_OUTBOUND; + ctx->hash_final = 0; + ctx->is_hash = 0; + ctx->pd_ctl = 0x1; + + return crypto4xx_build_pd(&req->base, ctx, req->src, req->dst, + req->nbytes, req->info, + get_dynamic_sa_iv_size(ctx)); +} + +int crypto4xx_decrypt(struct ablkcipher_request *req) +{ + struct crypto4xx_ctx *ctx = crypto_tfm_ctx(req->base.tfm); + + ctx->direction = DIR_INBOUND; + ctx->hash_final = 0; + ctx->is_hash = 0; + ctx->pd_ctl = 1; + + return crypto4xx_build_pd(&req->base, ctx, req->src, req->dst, + req->nbytes, req->info, + get_dynamic_sa_iv_size(ctx)); +} + +/** + * AES Functions + */ +static int crypto4xx_setkey_aes(struct crypto_ablkcipher *cipher, + const u8 *key, + unsigned int keylen, + unsigned char cm, + u8 fb) +{ + struct crypto_tfm *tfm = crypto_ablkcipher_tfm(cipher); + struct crypto4xx_ctx *ctx = crypto_tfm_ctx(tfm); + struct dynamic_sa_ctl *sa; + int rc; + + if (keylen != AES_KEYSIZE_256 && + keylen != AES_KEYSIZE_192 && keylen != AES_KEYSIZE_128) { + crypto_ablkcipher_set_flags(cipher, + CRYPTO_TFM_RES_BAD_KEY_LEN); + return -EINVAL; + } + + /* Create SA */ + if (ctx->sa_in_dma_addr || ctx->sa_out_dma_addr) + crypto4xx_free_sa(ctx); + + rc = crypto4xx_alloc_sa(ctx, SA_AES128_LEN + (keylen-16) / 4); + if (rc) + return rc; + + if (ctx->state_record_dma_addr == 0) { + rc = crypto4xx_alloc_state_record(ctx); + if (rc) { + crypto4xx_free_sa(ctx); + return rc; + } + } + /* Setup SA */ + sa = (struct dynamic_sa_ctl *) ctx->sa_in; + ctx->hash_final = 0; + + set_dynamic_sa_command_0(sa, SA_NOT_SAVE_HASH, SA_NOT_SAVE_IV, + SA_LOAD_HASH_FROM_SA, SA_LOAD_IV_FROM_STATE, + SA_NO_HEADER_PROC, SA_HASH_ALG_NULL, + SA_CIPHER_ALG_AES, SA_PAD_TYPE_ZERO, + SA_OP_GROUP_BASIC, SA_OPCODE_DECRYPT, + DIR_INBOUND); + + set_dynamic_sa_command_1(sa, cm, SA_HASH_MODE_HASH, + fb, SA_EXTENDED_SN_OFF, + SA_SEQ_MASK_OFF, SA_MC_ENABLE, + SA_NOT_COPY_PAD, SA_NOT_COPY_PAYLOAD, + SA_NOT_COPY_HDR); + crypto4xx_memcpy_le(ctx->sa_in + get_dynamic_sa_offset_key_field(ctx), + key, keylen); + sa->sa_contents = SA_AES_CONTENTS | (keylen << 2); + sa->sa_command_1.bf.key_len = keylen >> 3; + ctx->is_hash = 0; + ctx->direction = DIR_INBOUND; + memcpy(ctx->sa_in + get_dynamic_sa_offset_state_ptr_field(ctx), + (void *)&ctx->state_record_dma_addr, 4); + ctx->offset_to_sr_ptr = get_dynamic_sa_offset_state_ptr_field(ctx); + + memcpy(ctx->sa_out, ctx->sa_in, ctx->sa_len * 4); + sa = (struct dynamic_sa_ctl *) ctx->sa_out; + sa->sa_command_0.bf.dir = DIR_OUTBOUND; + + return 0; +} + +int crypto4xx_setkey_aes_cbc(struct crypto_ablkcipher *cipher, + const u8 *key, unsigned int keylen) +{ + return crypto4xx_setkey_aes(cipher, key, keylen, CRYPTO_MODE_CBC, + CRYPTO_FEEDBACK_MODE_NO_FB); +} + +/** + * HASH SHA1 Functions + */ +static int crypto4xx_hash_alg_init(struct crypto_tfm *tfm, + unsigned int sa_len, + unsigned char ha, + unsigned char hm) +{ + struct crypto_alg *alg = tfm->__crt_alg; + struct crypto4xx_alg *my_alg = crypto_alg_to_crypto4xx_alg(alg); + struct crypto4xx_ctx *ctx = crypto_tfm_ctx(tfm); + struct dynamic_sa_ctl *sa; + struct dynamic_sa_hash160 *sa_in; + int rc; + + ctx->dev = my_alg->dev; + ctx->is_hash = 1; + ctx->hash_final = 0; + + /* Create SA */ + if (ctx->sa_in_dma_addr || ctx->sa_out_dma_addr) + crypto4xx_free_sa(ctx); + + rc = crypto4xx_alloc_sa(ctx, sa_len); + if (rc) + return rc; + + if (ctx->state_record_dma_addr == 0) { + crypto4xx_alloc_state_record(ctx); + if (!ctx->state_record_dma_addr) { + crypto4xx_free_sa(ctx); + return -ENOMEM; + } + } + + tfm->crt_ahash.reqsize = sizeof(struct crypto4xx_ctx); + sa = (struct dynamic_sa_ctl *) ctx->sa_in; + set_dynamic_sa_command_0(sa, SA_SAVE_HASH, SA_NOT_SAVE_IV, + SA_NOT_LOAD_HASH, SA_LOAD_IV_FROM_SA, + SA_NO_HEADER_PROC, ha, SA_CIPHER_ALG_NULL, + SA_PAD_TYPE_ZERO, SA_OP_GROUP_BASIC, + SA_OPCODE_HASH, DIR_INBOUND); + set_dynamic_sa_command_1(sa, 0, SA_HASH_MODE_HASH, + CRYPTO_FEEDBACK_MODE_NO_FB, SA_EXTENDED_SN_OFF, + SA_SEQ_MASK_OFF, SA_MC_ENABLE, + SA_NOT_COPY_PAD, SA_NOT_COPY_PAYLOAD, + SA_NOT_COPY_HDR); + ctx->direction = DIR_INBOUND; + sa->sa_contents = SA_HASH160_CONTENTS; + sa_in = (struct dynamic_sa_hash160 *) ctx->sa_in; + /* Need to zero hash digest in SA */ + memset(sa_in->inner_digest, 0, sizeof(sa_in->inner_digest)); + memset(sa_in->outer_digest, 0, sizeof(sa_in->outer_digest)); + sa_in->state_ptr = ctx->state_record_dma_addr; + ctx->offset_to_sr_ptr = get_dynamic_sa_offset_state_ptr_field(ctx); + + return 0; +} + +int crypto4xx_hash_init(struct ahash_request *req) +{ + struct crypto4xx_ctx *ctx = crypto_tfm_ctx(req->base.tfm); + int ds; + struct dynamic_sa_ctl *sa; + + sa = (struct dynamic_sa_ctl *) ctx->sa_in; + ds = crypto_ahash_digestsize( + __crypto_ahash_cast(req->base.tfm)); + sa->sa_command_0.bf.digest_len = ds >> 2; + sa->sa_command_0.bf.load_hash_state = SA_LOAD_HASH_FROM_SA; + ctx->is_hash = 1; + ctx->direction = DIR_INBOUND; + + return 0; +} + +int crypto4xx_hash_update(struct ahash_request *req) +{ + struct crypto4xx_ctx *ctx = crypto_tfm_ctx(req->base.tfm); + + ctx->is_hash = 1; + ctx->hash_final = 0; + ctx->pd_ctl = 0x11; + ctx->direction = DIR_INBOUND; + + return crypto4xx_build_pd(&req->base, ctx, req->src, + (struct scatterlist *) req->result, + req->nbytes, NULL, 0); +} + +int crypto4xx_hash_final(struct ahash_request *req) +{ + return 0; +} + +int crypto4xx_hash_digest(struct ahash_request *req) +{ + struct crypto4xx_ctx *ctx = crypto_tfm_ctx(req->base.tfm); + + ctx->hash_final = 1; + ctx->pd_ctl = 0x11; + ctx->direction = DIR_INBOUND; + + return crypto4xx_build_pd(&req->base, ctx, req->src, + (struct scatterlist *) req->result, + req->nbytes, NULL, 0); +} + +/** + * SHA1 Algorithm + */ +int crypto4xx_sha1_alg_init(struct crypto_tfm *tfm) +{ + return crypto4xx_hash_alg_init(tfm, SA_HASH160_LEN, SA_HASH_ALG_SHA1, + SA_HASH_MODE_HASH); +} + + diff --git a/drivers/crypto/amcc/crypto4xx_core.c b/drivers/crypto/amcc/crypto4xx_core.c new file mode 100644 index 00000000000..4c0dfb2b872 --- /dev/null +++ b/drivers/crypto/amcc/crypto4xx_core.c @@ -0,0 +1,1310 @@ +/** + * AMCC SoC PPC4xx Crypto Driver + * + * Copyright (c) 2008 Applied Micro Circuits Corporation. + * All rights reserved. James Hsiao <jhsiao@amcc.com> + * + * 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. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * This file implements AMCC crypto offload Linux device driver for use with + * Linux CryptoAPI. + */ + +#include <linux/kernel.h> +#include <linux/interrupt.h> +#include <linux/spinlock_types.h> +#include <linux/random.h> +#include <linux/scatterlist.h> +#include <linux/crypto.h> +#include <linux/dma-mapping.h> +#include <linux/platform_device.h> +#include <linux/init.h> +#include <linux/of_platform.h> +#include <asm/dcr.h> +#include <asm/dcr-regs.h> +#include <asm/cacheflush.h> +#include <crypto/internal/hash.h> +#include <crypto/algapi.h> +#include <crypto/aes.h> +#include <crypto/sha.h> +#include "crypto4xx_reg_def.h" +#include "crypto4xx_core.h" +#include "crypto4xx_sa.h" + +#define PPC4XX_SEC_VERSION_STR "0.5" + +/** + * PPC4xx Crypto Engine Initialization Routine + */ +static void crypto4xx_hw_init(struct crypto4xx_device *dev) +{ + union ce_ring_size ring_size; + union ce_ring_contol ring_ctrl; + union ce_part_ring_size part_ring_size; + union ce_io_threshold io_threshold; + u32 rand_num; + union ce_pe_dma_cfg pe_dma_cfg; + + writel(PPC4XX_BYTE_ORDER, dev->ce_base + CRYPTO4XX_BYTE_ORDER_CFG); + /* setup pe dma, include reset sg, pdr and pe, then release reset */ + pe_dma_cfg.w = 0; + pe_dma_cfg.bf.bo_sgpd_en = 1; + pe_dma_cfg.bf.bo_data_en = 0; + pe_dma_cfg.bf.bo_sa_en = 1; + pe_dma_cfg.bf.bo_pd_en = 1; + pe_dma_cfg.bf.dynamic_sa_en = 1; + pe_dma_cfg.bf.reset_sg = 1; + pe_dma_cfg.bf.reset_pdr = 1; + pe_dma_cfg.bf.reset_pe = 1; + writel(pe_dma_cfg.w, dev->ce_base + CRYPTO4XX_PE_DMA_CFG); + /* un reset pe,sg and pdr */ + pe_dma_cfg.bf.pe_mode = 0; + pe_dma_cfg.bf.reset_sg = 0; + pe_dma_cfg.bf.reset_pdr = 0; + pe_dma_cfg.bf.reset_pe = 0; + pe_dma_cfg.bf.bo_td_en = 0; + writel(pe_dma_cfg.w, dev->ce_base + CRYPTO4XX_PE_DMA_CFG); + writel(dev->pdr_pa, dev->ce_base + CRYPTO4XX_PDR_BASE); + writel(dev->pdr_pa, dev->ce_base + CRYPTO4XX_RDR_BASE); + writel(PPC4XX_PRNG_CTRL_AUTO_EN, dev->ce_base + CRYPTO4XX_PRNG_CTRL); + get_random_bytes(&rand_num, sizeof(rand_num)); + writel(rand_num, dev->ce_base + CRYPTO4XX_PRNG_SEED_L); + get_random_bytes(&rand_num, sizeof(rand_num)); + writel(rand_num, dev->ce_base + CRYPTO4XX_PRNG_SEED_H); + ring_size.w = 0; + ring_size.bf.ring_offset = PPC4XX_PD_SIZE; + ring_size.bf.ring_size = PPC4XX_NUM_PD; + writel(ring_size.w, dev->ce_base + CRYPTO4XX_RING_SIZE); + ring_ctrl.w = 0; + writel(ring_ctrl.w, dev->ce_base + CRYPTO4XX_RING_CTRL); + writel(PPC4XX_DC_3DES_EN, dev->ce_base + CRYPTO4XX_DEVICE_CTRL); + writel(dev->gdr_pa, dev->ce_base + CRYPTO4XX_GATH_RING_BASE); + writel(dev->sdr_pa, dev->ce_base + CRYPTO4XX_SCAT_RING_BASE); + part_ring_size.w = 0; + part_ring_size.bf.sdr_size = PPC4XX_SDR_SIZE; + part_ring_size.bf.gdr_size = PPC4XX_GDR_SIZE; + writel(part_ring_size.w, dev->ce_base + CRYPTO4XX_PART_RING_SIZE); + writel(PPC4XX_SD_BUFFER_SIZE, dev->ce_base + CRYPTO4XX_PART_RING_CFG); + io_threshold.w = 0; + io_threshold.bf.output_threshold = PPC4XX_OUTPUT_THRESHOLD; + io_threshold.bf.input_threshold = PPC4XX_INPUT_THRESHOLD; + writel(io_threshold.w, dev->ce_base + CRYPTO4XX_IO_THRESHOLD); + writel(0, dev->ce_base + CRYPTO4XX_PDR_BASE_UADDR); + writel(0, dev->ce_base + CRYPTO4XX_RDR_BASE_UADDR); + writel(0, dev->ce_base + CRYPTO4XX_PKT_SRC_UADDR); + writel(0, dev->ce_base + CRYPTO4XX_PKT_DEST_UADDR); + writel(0, dev->ce_base + CRYPTO4XX_SA_UADDR); + writel(0, dev->ce_base + CRYPTO4XX_GATH_RING_BASE_UADDR); + writel(0, dev->ce_base + CRYPTO4XX_SCAT_RING_BASE_UADDR); + /* un reset pe,sg and pdr */ + pe_dma_cfg.bf.pe_mode = 1; + pe_dma_cfg.bf.reset_sg = 0; + pe_dma_cfg.bf.reset_pdr = 0; + pe_dma_cfg.bf.reset_pe = 0; + pe_dma_cfg.bf.bo_td_en = 0; + writel(pe_dma_cfg.w, dev->ce_base + CRYPTO4XX_PE_DMA_CFG); + /*clear all pending interrupt*/ + writel(PPC4XX_INTERRUPT_CLR, dev->ce_base + CRYPTO4XX_INT_CLR); + writel(PPC4XX_INT_DESCR_CNT, dev->ce_base + CRYPTO4XX_INT_DESCR_CNT); + writel(PPC4XX_INT_DESCR_CNT, dev->ce_base + CRYPTO4XX_INT_DESCR_CNT); + writel(PPC4XX_INT_CFG, dev->ce_base + CRYPTO4XX_INT_CFG); + writel(PPC4XX_PD_DONE_INT, dev->ce_base + CRYPTO4XX_INT_EN); +} + +int crypto4xx_alloc_sa(struct crypto4xx_ctx *ctx, u32 size) +{ + ctx->sa_in = dma_alloc_coherent(ctx->dev->core_dev->device, size * 4, + &ctx->sa_in_dma_addr, GFP_ATOMIC); + if (ctx->sa_in == NULL) + return -ENOMEM; + + ctx->sa_out = dma_alloc_coherent(ctx->dev->core_dev->device, size * 4, + &ctx->sa_out_dma_addr, GFP_ATOMIC); + if (ctx->sa_out == NULL) { + dma_free_coherent(ctx->dev->core_dev->device, + ctx->sa_len * 4, + ctx->sa_in, ctx->sa_in_dma_addr); + return -ENOMEM; + } + + memset(ctx->sa_in, 0, size * 4); + memset(ctx->sa_out, 0, size * 4); + ctx->sa_len = size; + + return 0; +} + +void crypto4xx_free_sa(struct crypto4xx_ctx *ctx) +{ + if (ctx->sa_in != NULL) + dma_free_coherent(ctx->dev->core_dev->device, ctx->sa_len * 4, + ctx->sa_in, ctx->sa_in_dma_addr); + if (ctx->sa_out != NULL) + dma_free_coherent(ctx->dev->core_dev->device, ctx->sa_len * 4, + ctx->sa_out, ctx->sa_out_dma_addr); + + ctx->sa_in_dma_addr = 0; + ctx->sa_out_dma_addr = 0; + ctx->sa_len = 0; +} + +u32 crypto4xx_alloc_state_record(struct crypto4xx_ctx *ctx) +{ + ctx->state_record = dma_alloc_coherent(ctx->dev->core_dev->device, + sizeof(struct sa_state_record), + &ctx->state_record_dma_addr, GFP_ATOMIC); + if (!ctx->state_record_dma_addr) + return -ENOMEM; + memset(ctx->state_record, 0, sizeof(struct sa_state_record)); + + return 0; +} + +void crypto4xx_free_state_record(struct crypto4xx_ctx *ctx) +{ + if (ctx->state_record != NULL) + dma_free_coherent(ctx->dev->core_dev->device, + sizeof(struct sa_state_record), + ctx->state_record, + ctx->state_record_dma_addr); + ctx->state_record_dma_addr = 0; +} + +/** + * alloc memory for the gather ring + * no need to alloc buf for the ring + * gdr_tail, gdr_head and gdr_count are initialized by this function + */ +static u32 crypto4xx_build_pdr(struct crypto4xx_device *dev) +{ + int i; + struct pd_uinfo *pd_uinfo; + dev->pdr = dma_alloc_coherent(dev->core_dev->device, + sizeof(struct ce_pd) * PPC4XX_NUM_PD, + &dev->pdr_pa, GFP_ATOMIC); + if (!dev->pdr) + return -ENOMEM; + + dev->pdr_uinfo = kzalloc(sizeof(struct pd_uinfo) * PPC4XX_NUM_PD, + GFP_KERNEL); + if (!dev->pdr_uinfo) { + dma_free_coherent(dev->core_dev->device, + sizeof(struct ce_pd) * PPC4XX_NUM_PD, + dev->pdr, + dev->pdr_pa); + return -ENOMEM; + } + memset(dev->pdr, 0, sizeof(struct ce_pd) * PPC4XX_NUM_PD); + dev->shadow_sa_pool = dma_alloc_coherent(dev->core_dev->device, + 256 * PPC4XX_NUM_PD, + &dev->shadow_sa_pool_pa, + GFP_ATOMIC); + if (!dev->shadow_sa_pool) + return -ENOMEM; + + dev->shadow_sr_pool = dma_alloc_coherent(dev->core_dev->device, + sizeof(struct sa_state_record) * PPC4XX_NUM_PD, + &dev->shadow_sr_pool_pa, GFP_ATOMIC); + if (!dev->shadow_sr_pool) + return -ENOMEM; + for (i = 0; i < PPC4XX_NUM_PD; i++) { + pd_uinfo = (struct pd_uinfo *) (dev->pdr_uinfo + + sizeof(struct pd_uinfo) * i); + + /* alloc 256 bytes which is enough for any kind of dynamic sa */ + pd_uinfo->sa_va = dev->shadow_sa_pool + 256 * i; + pd_uinfo->sa_pa = dev->shadow_sa_pool_pa + 256 * i; + + /* alloc state record */ + pd_uinfo->sr_va = dev->shadow_sr_pool + + sizeof(struct sa_state_record) * i; + pd_uinfo->sr_pa = dev->shadow_sr_pool_pa + + sizeof(struct sa_state_record) * i; + } + + return 0; +} + +static void crypto4xx_destroy_pdr(struct crypto4xx_device *dev) +{ + if (dev->pdr != NULL) + dma_free_coherent(dev->core_dev->device, + sizeof(struct ce_pd) * PPC4XX_NUM_PD, + dev->pdr, dev->pdr_pa); + if (dev->shadow_sa_pool) + dma_free_coherent(dev->core_dev->device, 256 * PPC4XX_NUM_PD, + dev->shadow_sa_pool, dev->shadow_sa_pool_pa); + if (dev->shadow_sr_pool) + dma_free_coherent(dev->core_dev->device, + sizeof(struct sa_state_record) * PPC4XX_NUM_PD, + dev->shadow_sr_pool, dev->shadow_sr_pool_pa); + + kfree(dev->pdr_uinfo); +} + +static u32 crypto4xx_get_pd_from_pdr_nolock(struct crypto4xx_device *dev) +{ + u32 retval; + u32 tmp; + + retval = dev->pdr_head; + tmp = (dev->pdr_head + 1) % PPC4XX_NUM_PD; + + if (tmp == dev->pdr_tail) + return ERING_WAS_FULL; + + dev->pdr_head = tmp; + + return retval; +} + +static u32 crypto4xx_put_pd_to_pdr(struct crypto4xx_device *dev, u32 idx) +{ + struct pd_uinfo *pd_uinfo; + unsigned long flags; + + pd_uinfo = (struct pd_uinfo *)(dev->pdr_uinfo + + sizeof(struct pd_uinfo) * idx); + spin_lock_irqsave(&dev->core_dev->lock, flags); + if (dev->pdr_tail != PPC4XX_LAST_PD) + dev->pdr_tail++; + else + dev->pdr_tail = 0; + pd_uinfo->state = PD_ENTRY_FREE; + spin_unlock_irqrestore(&dev->core_dev->lock, flags); + + return 0; +} + +static struct ce_pd *crypto4xx_get_pdp(struct crypto4xx_device *dev, + dma_addr_t *pd_dma, u32 idx) +{ + *pd_dma = dev->pdr_pa + sizeof(struct ce_pd) * idx; + + return dev->pdr + sizeof(struct ce_pd) * idx; +} + +/** + * alloc memory for the gather ring + * no need to alloc buf for the ring + * gdr_tail, gdr_head and gdr_count are initialized by this function + */ +static u32 crypto4xx_build_gdr(struct crypto4xx_device *dev) +{ + dev->gdr = dma_alloc_coherent(dev->core_dev->device, + sizeof(struct ce_gd) * PPC4XX_NUM_GD, + &dev->gdr_pa, GFP_ATOMIC); + if (!dev->gdr) + return -ENOMEM; + + memset(dev->gdr, 0, sizeof(struct ce_gd) * PPC4XX_NUM_GD); + + return 0; +} + +static inline void crypto4xx_destroy_gdr(struct crypto4xx_device *dev) +{ + dma_free_coherent(dev->core_dev->device, + sizeof(struct ce_gd) * PPC4XX_NUM_GD, + dev->gdr, dev->gdr_pa); +} + +/* + * when this function is called. + * preemption or interrupt must be disabled + */ +u32 crypto4xx_get_n_gd(struct crypto4xx_device *dev, int n) +{ + u32 retval; + u32 tmp; + if (n >= PPC4XX_NUM_GD) + return ERING_WAS_FULL; + + retval = dev->gdr_head; + tmp = (dev->gdr_head + n) % PPC4XX_NUM_GD; + if (dev->gdr_head > dev->gdr_tail) { + if (tmp < dev->gdr_head && tmp >= dev->gdr_tail) + return ERING_WAS_FULL; + } else if (dev->gdr_head < dev->gdr_tail) { + if (tmp < dev->gdr_head || tmp >= dev->gdr_tail) + return ERING_WAS_FULL; + } + dev->gdr_head = tmp; + + return retval; +} + +static u32 crypto4xx_put_gd_to_gdr(struct crypto4xx_device *dev) +{ + unsigned long flags; + + spin_lock_irqsave(&dev->core_dev->lock, flags); + if (dev->gdr_tail == dev->gdr_head) { + spin_unlock_irqrestore(&dev->core_dev->lock, flags); + return 0; + } + + if (dev->gdr_tail != PPC4XX_LAST_GD) + dev->gdr_tail++; + else + dev->gdr_tail = 0; + + spin_unlock_irqrestore(&dev->core_dev->lock, flags); + + return 0; +} + +static inline struct ce_gd *crypto4xx_get_gdp(struct crypto4xx_device *dev, + dma_addr_t *gd_dma, u32 idx) +{ + *gd_dma = dev->gdr_pa + sizeof(struct ce_gd) * idx; + + return (struct ce_gd *) (dev->gdr + sizeof(struct ce_gd) * idx); +} + +/** + * alloc memory for the scatter ring + * need to alloc buf for the ring + * sdr_tail, sdr_head and sdr_count are initialized by this function + */ +static u32 crypto4xx_build_sdr(struct crypto4xx_device *dev) +{ + int i; + struct ce_sd *sd_array; + + /* alloc memory for scatter descriptor ring */ + dev->sdr = dma_alloc_coherent(dev->core_dev->device, + sizeof(struct ce_sd) * PPC4XX_NUM_SD, + &dev->sdr_pa, GFP_ATOMIC); + if (!dev->sdr) + return -ENOMEM; + + dev->scatter_buffer_size = PPC4XX_SD_BUFFER_SIZE; + dev->scatter_buffer_va = + dma_alloc_coherent(dev->core_dev->device, + dev->scatter_buffer_size * PPC4XX_NUM_SD, + &dev->scatter_buffer_pa, GFP_ATOMIC); + if (!dev->scatter_buffer_va) { + dma_free_coherent(dev->core_dev->device, + sizeof(struct ce_sd) * PPC4XX_NUM_SD, + dev->sdr, dev->sdr_pa); + return -ENOMEM; + } + + sd_array = dev->sdr; + + for (i = 0; i < PPC4XX_NUM_SD; i++) { + sd_array[i].ptr = dev->scatter_buffer_pa + + dev->scatter_buffer_size * i; + } + + return 0; +} + +static void crypto4xx_destroy_sdr(struct crypto4xx_device *dev) +{ + if (dev->sdr != NULL) + dma_free_coherent(dev->core_dev->device, + sizeof(struct ce_sd) * PPC4XX_NUM_SD, + dev->sdr, dev->sdr_pa); + + if (dev->scatter_buffer_va != NULL) + dma_free_coherent(dev->core_dev->device, + dev->scatter_buffer_size * PPC4XX_NUM_SD, + dev->scatter_buffer_va, + dev->scatter_buffer_pa); +} + +/* + * when this function is called. + * preemption or interrupt must be disabled + */ +static u32 crypto4xx_get_n_sd(struct crypto4xx_device *dev, int n) +{ + u32 retval; + u32 tmp; + + if (n >= PPC4XX_NUM_SD) + return ERING_WAS_FULL; + + retval = dev->sdr_head; + tmp = (dev->sdr_head + n) % PPC4XX_NUM_SD; + if (dev->sdr_head > dev->gdr_tail) { + if (tmp < dev->sdr_head && tmp >= dev->sdr_tail) + return ERING_WAS_FULL; + } else if (dev->sdr_head < dev->sdr_tail) { + if (tmp < dev->sdr_head || tmp >= dev->sdr_tail) + return ERING_WAS_FULL; + } /* the head = tail, or empty case is already take cared */ + dev->sdr_head = tmp; + + return retval; +} + +static u32 crypto4xx_put_sd_to_sdr(struct crypto4xx_device *dev) +{ + unsigned long flags; + + spin_lock_irqsave(&dev->core_dev->lock, flags); + if (dev->sdr_tail == dev->sdr_head) { + spin_unlock_irqrestore(&dev->core_dev->lock, flags); + return 0; + } + if (dev->sdr_tail != PPC4XX_LAST_SD) + dev->sdr_tail++; + else + dev->sdr_tail = 0; + spin_unlock_irqrestore(&dev->core_dev->lock, flags); + + return 0; +} + +static inline struct ce_sd *crypto4xx_get_sdp(struct crypto4xx_device *dev, + dma_addr_t *sd_dma, u32 idx) +{ + *sd_dma = dev->sdr_pa + sizeof(struct ce_sd) * idx; + + return (struct ce_sd *)(dev->sdr + sizeof(struct ce_sd) * idx); +} + +static u32 crypto4xx_fill_one_page(struct crypto4xx_device *dev, + dma_addr_t *addr, u32 *length, + u32 *idx, u32 *offset, u32 *nbytes) +{ + u32 len; + + if (*length > dev->scatter_buffer_size) { + memcpy(phys_to_virt(*addr), + dev->scatter_buffer_va + + *idx * dev->scatter_buffer_size + *offset, + dev->scatter_buffer_size); + *offset = 0; + *length -= dev->scatter_buffer_size; + *nbytes -= dev->scatter_buffer_size; + if (*idx == PPC4XX_LAST_SD) + *idx = 0; + else + (*idx)++; + *addr = *addr + dev->scatter_buffer_size; + return 1; + } else if (*length < dev->scatter_buffer_size) { + memcpy(phys_to_virt(*addr), + dev->scatter_buffer_va + + *idx * dev->scatter_buffer_size + *offset, *length); + if ((*offset + *length) == dev->scatter_buffer_size) { + if (*idx == PPC4XX_LAST_SD) + *idx = 0; + else + (*idx)++; + *nbytes -= *length; + *offset = 0; + } else { + *nbytes -= *length; + *offset += *length; + } + + return 0; + } else { + len = (*nbytes <= dev->scatter_buffer_size) ? + (*nbytes) : dev->scatter_buffer_size; + memcpy(phys_to_virt(*addr), + dev->scatter_buffer_va + + *idx * dev->scatter_buffer_size + *offset, + len); + *offset = 0; + *nbytes -= len; + + if (*idx == PPC4XX_LAST_SD) + *idx = 0; + else + (*idx)++; + + return 0; + } +} + +static void crypto4xx_copy_pkt_to_dst(struct crypto4xx_device *dev, + struct ce_pd *pd, + struct pd_uinfo *pd_uinfo, + u32 nbytes, + struct scatterlist *dst) +{ + dma_addr_t addr; + u32 this_sd; + u32 offset; + u32 len; + u32 i; + u32 sg_len; + struct scatterlist *sg; + + this_sd = pd_uinfo->first_sd; + offset = 0; + i = 0; + + while (nbytes) { + sg = &dst[i]; + sg_len = sg->length; + addr = dma_map_page(dev->core_dev->device, sg_page(sg), + sg->offset, sg->length, DMA_TO_DEVICE); + + if (offset == 0) { + len = (nbytes <= sg->length) ? nbytes : sg->length; + while (crypto4xx_fill_one_page(dev, &addr, &len, + &this_sd, &offset, &nbytes)) + ; + if (!nbytes) + return; + i++; + } else { + len = (nbytes <= (dev->scatter_buffer_size - offset)) ? + nbytes : (dev->scatter_buffer_size - offset); + len = (sg->length < len) ? sg->length : len; + while (crypto4xx_fill_one_page(dev, &addr, &len, + &this_sd, &offset, &nbytes)) + ; + if (!nbytes) + return; + sg_len -= len; + if (sg_len) { + addr += len; + while (crypto4xx_fill_one_page(dev, &addr, + &sg_len, &this_sd, &offset, &nbytes)) + ; + } + i++; + } + } +} + +static u32 crypto4xx_copy_digest_to_dst(struct pd_uinfo *pd_uinfo, + struct crypto4xx_ctx *ctx) +{ + struct dynamic_sa_ctl *sa = (struct dynamic_sa_ctl *) ctx->sa_in; + struct sa_state_record *state_record = + (struct sa_state_record *) pd_uinfo->sr_va; + + if (sa->sa_command_0.bf.hash_alg == SA_HASH_ALG_SHA1) { + memcpy((void *) pd_uinfo->dest_va, state_record->save_digest, + SA_HASH_ALG_SHA1_DIGEST_SIZE); + } + + return 0; +} + +static void crypto4xx_ret_sg_desc(struct crypto4xx_device *dev, + struct pd_uinfo *pd_uinfo) +{ + int i; + if (pd_uinfo->num_gd) { + for (i = 0; i < pd_uinfo->num_gd; i++) + crypto4xx_put_gd_to_gdr(dev); + pd_uinfo->first_gd = 0xffffffff; + pd_uinfo->num_gd = 0; + } + if (pd_uinfo->num_sd) { + for (i = 0; i < pd_uinfo->num_sd; i++) + crypto4xx_put_sd_to_sdr(dev); + + pd_uinfo->first_sd = 0xffffffff; + pd_uinfo->num_sd = 0; + } +} + +static u32 crypto4xx_ablkcipher_done(struct crypto4xx_device *dev, + struct pd_uinfo *pd_uinfo, + struct ce_pd *pd) +{ + struct crypto4xx_ctx *ctx; + struct ablkcipher_request *ablk_req; + struct scatterlist *dst; + dma_addr_t addr; + + ablk_req = ablkcipher_request_cast(pd_uinfo->async_req); + ctx = crypto_tfm_ctx(ablk_req->base.tfm); + + if (pd_uinfo->using_sd) { + crypto4xx_copy_pkt_to_dst(dev, pd, pd_uinfo, ablk_req->nbytes, + ablk_req->dst); + } else { + dst = pd_uinfo->dest_va; + addr = dma_map_page(dev->core_dev->device, sg_page(dst), + dst->offset, dst->length, DMA_FROM_DEVICE); + } + crypto4xx_ret_sg_desc(dev, pd_uinfo); + if (ablk_req->base.complete != NULL) + ablk_req->base.complete(&ablk_req->base, 0); + + return 0; +} + +static u32 crypto4xx_ahash_done(struct crypto4xx_device *dev, + struct pd_uinfo *pd_uinfo) +{ + struct crypto4xx_ctx *ctx; + struct ahash_request *ahash_req; + + ahash_req = ahash_request_cast(pd_uinfo->async_req); + ctx = crypto_tfm_ctx(ahash_req->base.tfm); + + crypto4xx_copy_digest_to_dst(pd_uinfo, + crypto_tfm_ctx(ahash_req->base.tfm)); + crypto4xx_ret_sg_desc(dev, pd_uinfo); + /* call user provided callback function x */ + if (ahash_req->base.complete != NULL) + ahash_req->base.complete(&ahash_req->base, 0); + + return 0; +} + +static u32 crypto4xx_pd_done(struct crypto4xx_device *dev, u32 idx) +{ + struct ce_pd *pd; + struct pd_uinfo *pd_uinfo; + + pd = dev->pdr + sizeof(struct ce_pd)*idx; + pd_uinfo = dev->pdr_uinfo + sizeof(struct pd_uinfo)*idx; + if (crypto_tfm_alg_type(pd_uinfo->async_req->tfm) == + CRYPTO_ALG_TYPE_ABLKCIPHER) + return crypto4xx_ablkcipher_done(dev, pd_uinfo, pd); + else + return crypto4xx_ahash_done(dev, pd_uinfo); +} + +/** + * Note: Only use this function to copy items that is word aligned. + */ +void crypto4xx_memcpy_le(unsigned int *dst, + const unsigned char *buf, + int len) +{ + u8 *tmp; + for (; len >= 4; buf += 4, len -= 4) + *dst++ = cpu_to_le32(*(unsigned int *) buf); + + tmp = (u8 *)dst; + switch (len) { + case 3: + *tmp++ = 0; + *tmp++ = *(buf+2); + *tmp++ = *(buf+1); + *tmp++ = *buf; + break; + case 2: + *tmp++ = 0; + *tmp++ = 0; + *tmp++ = *(buf+1); + *tmp++ = *buf; + break; + case 1: + *tmp++ = 0; + *tmp++ = 0; + *tmp++ = 0; + *tmp++ = *buf; + break; + default: + break; + } +} + +static void crypto4xx_stop_all(struct crypto4xx_core_device *core_dev) +{ + crypto4xx_destroy_pdr(core_dev->dev); + crypto4xx_destroy_gdr(core_dev->dev); + crypto4xx_destroy_sdr(core_dev->dev); + dev_set_drvdata(core_dev->device, NULL); + iounmap(core_dev->dev->ce_base); + kfree(core_dev->dev); + kfree(core_dev); +} + +void crypto4xx_return_pd(struct crypto4xx_device *dev, + u32 pd_entry, struct ce_pd *pd, + struct pd_uinfo *pd_uinfo) +{ + /* irq should be already disabled */ + dev->pdr_head = pd_entry; + pd->pd_ctl.w = 0; + pd->pd_ctl_len.w = 0; + pd_uinfo->state = PD_ENTRY_FREE; +} + +/* + * derive number of elements in scatterlist + * Shamlessly copy from talitos.c + */ +static int get_sg_count(struct scatterlist *sg_list, int nbytes) +{ + struct scatterlist *sg = sg_list; + int sg_nents = 0; + + while (nbytes) { + sg_nents++; + if (sg->length > nbytes) + break; + nbytes -= sg->length; + sg = sg_next(sg); + } + + return sg_nents; +} + +static u32 get_next_gd(u32 current) +{ + if (current != PPC4XX_LAST_GD) + return current + 1; + else + return 0; +} + +static u32 get_next_sd(u32 current) +{ + if (current != PPC4XX_LAST_SD) + return current + 1; + else + return 0; +} + +u32 crypto4xx_build_pd(struct crypto_async_request *req, + struct crypto4xx_ctx *ctx, + struct scatterlist *src, + struct scatterlist *dst, + unsigned int datalen, + void *iv, u32 iv_len) +{ + struct crypto4xx_device *dev = ctx->dev; + dma_addr_t addr, pd_dma, sd_dma, gd_dma; + struct dynamic_sa_ctl *sa; + struct scatterlist *sg; + struct ce_gd *gd; + struct ce_pd *pd; + u32 num_gd, num_sd; + u32 fst_gd = 0xffffffff; + u32 fst_sd = 0xffffffff; + u32 pd_entry; + unsigned long flags; + struct pd_uinfo *pd_uinfo = NULL; + unsigned int nbytes = datalen, idx; + unsigned int ivlen = 0; + u32 gd_idx = 0; + + /* figure how many gd is needed */ + num_gd = get_sg_count(src, datalen); + if (num_gd == 1) + num_gd = 0; + + /* figure how many sd is needed */ + if (sg_is_last(dst) || ctx->is_hash) { + num_sd = 0; + } else { + if (datalen > PPC4XX_SD_BUFFER_SIZE) { + num_sd = datalen / PPC4XX_SD_BUFFER_SIZE; + if (datalen % PPC4XX_SD_BUFFER_SIZE) + num_sd++; + } else { + num_sd = 1; + } + } + + /* + * The follow section of code needs to be protected + * The gather ring and scatter ring needs to be consecutive + * In case of run out of any kind of descriptor, the descriptor + * already got must be return the original place. + */ + spin_lock_irqsave(&dev->core_dev->lock, flags); + if (num_gd) { + fst_gd = crypto4xx_get_n_gd(dev, num_gd); + if (fst_gd == ERING_WAS_FULL) { + spin_unlock_irqrestore(&dev->core_dev->lock, flags); + return -EAGAIN; + } + } + if (num_sd) { + fst_sd = crypto4xx_get_n_sd(dev, num_sd); + if (fst_sd == ERING_WAS_FULL) { + if (num_gd) + dev->gdr_head = fst_gd; + spin_unlock_irqrestore(&dev->core_dev->lock, flags); + return -EAGAIN; + } + } + pd_entry = crypto4xx_get_pd_from_pdr_nolock(dev); + if (pd_entry == ERING_WAS_FULL) { + if (num_gd) + dev->gdr_head = fst_gd; + if (num_sd) + dev->sdr_head = fst_sd; + spin_unlock_irqrestore(&dev->core_dev->lock, flags); + return -EAGAIN; + } + spin_unlock_irqrestore(&dev->core_dev->lock, flags); + + pd_uinfo = (struct pd_uinfo *)(dev->pdr_uinfo + + sizeof(struct pd_uinfo) * pd_entry); + pd = crypto4xx_get_pdp(dev, &pd_dma, pd_entry); + pd_uinfo->async_req = req; + pd_uinfo->num_gd = num_gd; + pd_uinfo->num_sd = num_sd; + + if (iv_len || ctx->is_hash) { + ivlen = iv_len; + pd->sa = pd_uinfo->sa_pa; + sa = (struct dynamic_sa_ctl *) pd_uinfo->sa_va; + if (ctx->direction == DIR_INBOUND) + memcpy(sa, ctx->sa_in, ctx->sa_len * 4); + else + memcpy(sa, ctx->sa_out, ctx->sa_len * 4); + + memcpy((void *) sa + ctx->offset_to_sr_ptr, + &pd_uinfo->sr_pa, 4); + + if (iv_len) + crypto4xx_memcpy_le(pd_uinfo->sr_va, iv, iv_len); + } else { + if (ctx->direction == DIR_INBOUND) { + pd->sa = ctx->sa_in_dma_addr; + sa = (struct dynamic_sa_ctl *) ctx->sa_in; + } else { + pd->sa = ctx->sa_out_dma_addr; + sa = (struct dynamic_sa_ctl *) ctx->sa_out; + } + } + pd->sa_len = ctx->sa_len; + if (num_gd) { + /* get first gd we are going to use */ + gd_idx = fst_gd; + pd_uinfo->first_gd = fst_gd; + pd_uinfo->num_gd = num_gd; + gd = crypto4xx_get_gdp(dev, &gd_dma, gd_idx); + pd->src = gd_dma; + /* enable gather */ + sa->sa_command_0.bf.gather = 1; + idx = 0; + src = &src[0]; + /* walk the sg, and setup gather array */ + while (nbytes) { + sg = &src[idx]; + addr = dma_map_page(dev->core_dev->device, sg_page(sg), + sg->offset, sg->length, DMA_TO_DEVICE); + gd->ptr = addr; + gd->ctl_len.len = sg->length; + gd->ctl_len.done = 0; + gd->ctl_len.ready = 1; + if (sg->length >= nbytes) + break; + nbytes -= sg->length; + gd_idx = get_next_gd(gd_idx); + gd = crypto4xx_get_gdp(dev, &gd_dma, gd_idx); + idx++; + } + } else { + pd->src = (u32)dma_map_page(dev->core_dev->device, sg_page(src), + src->offset, src->length, DMA_TO_DEVICE); + /* + * Disable gather in sa command + */ + sa->sa_command_0.bf.gather = 0; + /* + * Indicate gather array is not used + */ + pd_uinfo->first_gd = 0xffffffff; + pd_uinfo->num_gd = 0; + } + if (ctx->is_hash || sg_is_last(dst)) { + /* + * we know application give us dst a whole piece of memory + * no need to use scatter ring. + * In case of is_hash, the icv is always at end of src data. + */ + pd_uinfo->using_sd = 0; + pd_uinfo->first_sd = 0xffffffff; + pd_uinfo->num_sd = 0; + pd_uinfo->dest_va = dst; + sa->sa_command_0.bf.scatter = 0; + if (ctx->is_hash) + pd->dest = virt_to_phys((void *)dst); + else + pd->dest = (u32)dma_map_page(dev->core_dev->device, + sg_page(dst), dst->offset, + dst->length, DMA_TO_DEVICE); + } else { + struct ce_sd *sd = NULL; + u32 sd_idx = fst_sd; + nbytes = datalen; + sa->sa_command_0.bf.scatter = 1; + pd_uinfo->using_sd = 1; + pd_uinfo->dest_va = dst; + pd_uinfo->first_sd = fst_sd; + pd_uinfo->num_sd = num_sd; + sd = crypto4xx_get_sdp(dev, &sd_dma, sd_idx); + pd->dest = sd_dma; + /* setup scatter descriptor */ + sd->ctl.done = 0; + sd->ctl.rdy = 1; + /* sd->ptr should be setup by sd_init routine*/ + idx = 0; + if (nbytes >= PPC4XX_SD_BUFFER_SIZE) + nbytes -= PPC4XX_SD_BUFFER_SIZE; + else + nbytes = 0; + while (nbytes) { + sd_idx = get_next_sd(sd_idx); + sd = crypto4xx_get_sdp(dev, &sd_dma, sd_idx); + /* setup scatter descriptor */ + sd->ctl.done = 0; + sd->ctl.rdy = 1; + if (nbytes >= PPC4XX_SD_BUFFER_SIZE) + nbytes -= PPC4XX_SD_BUFFER_SIZE; + else + /* + * SD entry can hold PPC4XX_SD_BUFFER_SIZE, + * which is more than nbytes, so done. + */ + nbytes = 0; + } + } + + sa->sa_command_1.bf.hash_crypto_offset = 0; + pd->pd_ctl.w = ctx->pd_ctl; + pd->pd_ctl_len.w = 0x00400000 | (ctx->bypass << 24) | datalen; + pd_uinfo->state = PD_ENTRY_INUSE; + wmb(); + /* write any value to push engine to read a pd */ + writel(1, dev->ce_base + CRYPTO4XX_INT_DESCR_RD); + return -EINPROGRESS; +} + +/** + * Algorithm Registration Functions + */ +static int crypto4xx_alg_init(struct crypto_tfm *tfm) +{ + struct crypto_alg *alg = tfm->__crt_alg; + struct crypto4xx_alg *amcc_alg = crypto_alg_to_crypto4xx_alg(alg); + struct crypto4xx_ctx *ctx = crypto_tfm_ctx(tfm); + + ctx->dev = amcc_alg->dev; + ctx->sa_in = NULL; + ctx->sa_out = NULL; + ctx->sa_in_dma_addr = 0; + ctx->sa_out_dma_addr = 0; + ctx->sa_len = 0; + + if (alg->cra_type == &crypto_ablkcipher_type) + tfm->crt_ablkcipher.reqsize = sizeof(struct crypto4xx_ctx); + else if (alg->cra_type == &crypto_ahash_type) + tfm->crt_ahash.reqsize = sizeof(struct crypto4xx_ctx); + + return 0; +} + +static void crypto4xx_alg_exit(struct crypto_tfm *tfm) +{ + struct crypto4xx_ctx *ctx = crypto_tfm_ctx(tfm); + + crypto4xx_free_sa(ctx); + crypto4xx_free_state_record(ctx); +} + +int crypto4xx_register_alg(struct crypto4xx_device *sec_dev, + struct crypto_alg *crypto_alg, int array_size) +{ + struct crypto4xx_alg *alg; + int i; + int rc = 0; + + for (i = 0; i < array_size; i++) { + alg = kzalloc(sizeof(struct crypto4xx_alg), GFP_KERNEL); + if (!alg) + return -ENOMEM; + + alg->alg = crypto_alg[i]; + INIT_LIST_HEAD(&alg->alg.cra_list); + if (alg->alg.cra_init == NULL) + alg->alg.cra_init = crypto4xx_alg_init; + if (alg->alg.cra_exit == NULL) + alg->alg.cra_exit = crypto4xx_alg_exit; + alg->dev = sec_dev; + rc = crypto_register_alg(&alg->alg); + if (rc) { + list_del(&alg->entry); + kfree(alg); + } else { + list_add_tail(&alg->entry, &sec_dev->alg_list); + } + } + + return 0; +} + +static void crypto4xx_unregister_alg(struct crypto4xx_device *sec_dev) +{ + struct crypto4xx_alg *alg, *tmp; + + list_for_each_entry_safe(alg, tmp, &sec_dev->alg_list, entry) { + list_del(&alg->entry); + crypto_unregister_alg(&alg->alg); + kfree(alg); + } +} + +static void crypto4xx_bh_tasklet_cb(unsigned long data) +{ + struct device *dev = (struct device *)data; + struct crypto4xx_core_device *core_dev = dev_get_drvdata(dev); + struct pd_uinfo *pd_uinfo; + struct ce_pd *pd; + u32 tail; + + while (core_dev->dev->pdr_head != core_dev->dev->pdr_tail) { + tail = core_dev->dev->pdr_tail; + pd_uinfo = core_dev->dev->pdr_uinfo + + sizeof(struct pd_uinfo)*tail; + pd = core_dev->dev->pdr + sizeof(struct ce_pd) * tail; + if ((pd_uinfo->state == PD_ENTRY_INUSE) && + pd->pd_ctl.bf.pe_done && + !pd->pd_ctl.bf.host_ready) { + pd->pd_ctl.bf.pe_done = 0; + crypto4xx_pd_done(core_dev->dev, tail); + crypto4xx_put_pd_to_pdr(core_dev->dev, tail); + pd_uinfo->state = PD_ENTRY_FREE; + } else { + /* if tail not done, break */ + break; + } + } +} + +/** + * Top Half of isr. + */ +static irqreturn_t crypto4xx_ce_interrupt_handler(int irq, void *data) +{ + struct device *dev = (struct device *)data; + struct crypto4xx_core_device *core_dev = dev_get_drvdata(dev); + + if (core_dev->dev->ce_base == 0) + return 0; + + writel(PPC4XX_INTERRUPT_CLR, + core_dev->dev->ce_base + CRYPTO4XX_INT_CLR); + tasklet_schedule(&core_dev->tasklet); + + return IRQ_HANDLED; +} + +/** + * Supported Crypto Algorithms + */ +struct crypto_alg crypto4xx_alg[] = { + /* Crypto AES modes */ + { + .cra_name = "cbc(aes)", + .cra_driver_name = "cbc-aes-ppc4xx", + .cra_priority = CRYPTO4XX_CRYPTO_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct crypto4xx_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .ablkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_IV_SIZE, + .setkey = crypto4xx_setkey_aes_cbc, + .encrypt = crypto4xx_encrypt, + .decrypt = crypto4xx_decrypt, + } + } + }, + /* Hash SHA1 */ + { + .cra_name = "sha1", + .cra_driver_name = "sha1-ppc4xx", + .cra_priority = CRYPTO4XX_CRYPTO_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_ASYNC, + .cra_blocksize = SHA1_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct crypto4xx_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ahash_type, + .cra_init = crypto4xx_sha1_alg_init, + .cra_module = THIS_MODULE, + .cra_u = { + .ahash = { + .digestsize = SHA1_DIGEST_SIZE, + .init = crypto4xx_hash_init, + .update = crypto4xx_hash_update, + .final = crypto4xx_hash_final, + .digest = crypto4xx_hash_digest, + } + } + }, +}; + +/** + * Module Initialization Routine + */ +static int __init crypto4xx_probe(struct of_device *ofdev, + const struct of_device_id *match) +{ + int rc; + struct resource res; + struct device *dev = &ofdev->dev; + struct crypto4xx_core_device *core_dev; + + rc = of_address_to_resource(ofdev->node, 0, &res); + if (rc) + return -ENODEV; + + if (of_find_compatible_node(NULL, NULL, "amcc,ppc460ex-crypto")) { + mtdcri(SDR0, PPC460EX_SDR0_SRST, + mfdcri(SDR0, PPC460EX_SDR0_SRST) | PPC460EX_CE_RESET); + mtdcri(SDR0, PPC460EX_SDR0_SRST, + mfdcri(SDR0, PPC460EX_SDR0_SRST) & ~PPC460EX_CE_RESET); + } else if (of_find_compatible_node(NULL, NULL, + "amcc,ppc405ex-crypto")) { + mtdcri(SDR0, PPC405EX_SDR0_SRST, + mfdcri(SDR0, PPC405EX_SDR0_SRST) | PPC405EX_CE_RESET); + mtdcri(SDR0, PPC405EX_SDR0_SRST, + mfdcri(SDR0, PPC405EX_SDR0_SRST) & ~PPC405EX_CE_RESET); + } else if (of_find_compatible_node(NULL, NULL, + "amcc,ppc460sx-crypto")) { + mtdcri(SDR0, PPC460SX_SDR0_SRST, + mfdcri(SDR0, PPC460SX_SDR0_SRST) | PPC460SX_CE_RESET); + mtdcri(SDR0, PPC460SX_SDR0_SRST, + mfdcri(SDR0, PPC460SX_SDR0_SRST) & ~PPC460SX_CE_RESET); + } else { + printk(KERN_ERR "Crypto Function Not supported!\n"); + return -EINVAL; + } + + core_dev = kzalloc(sizeof(struct crypto4xx_core_device), GFP_KERNEL); + if (!core_dev) + return -ENOMEM; + + dev_set_drvdata(dev, core_dev); + core_dev->ofdev = ofdev; + core_dev->dev = kzalloc(sizeof(struct crypto4xx_device), GFP_KERNEL); + if (!core_dev->dev) + goto err_alloc_dev; + + core_dev->dev->core_dev = core_dev; + core_dev->device = dev; + spin_lock_init(&core_dev->lock); + INIT_LIST_HEAD(&core_dev->dev->alg_list); + rc = crypto4xx_build_pdr(core_dev->dev); + if (rc) + goto err_build_pdr; + + rc = crypto4xx_build_gdr(core_dev->dev); + if (rc) + goto err_build_gdr; + + rc = crypto4xx_build_sdr(core_dev->dev); + if (rc) + goto err_build_sdr; + + /* Init tasklet for bottom half processing */ + tasklet_init(&core_dev->tasklet, crypto4xx_bh_tasklet_cb, + (unsigned long) dev); + + /* Register for Crypto isr, Crypto Engine IRQ */ + core_dev->irq = irq_of_parse_and_map(ofdev->node, 0); + rc = request_irq(core_dev->irq, crypto4xx_ce_interrupt_handler, 0, + core_dev->dev->name, dev); + if (rc) + goto err_request_irq; + + core_dev->dev->ce_base = of_iomap(ofdev->node, 0); + if (!core_dev->dev->ce_base) { + dev_err(dev, "failed to of_iomap\n"); + goto err_iomap; + } + + /* need to setup pdr, rdr, gdr and sdr before this */ + crypto4xx_hw_init(core_dev->dev); + + /* Register security algorithms with Linux CryptoAPI */ + rc = crypto4xx_register_alg(core_dev->dev, crypto4xx_alg, + ARRAY_SIZE(crypto4xx_alg)); + if (rc) + goto err_start_dev; + + return 0; + +err_start_dev: + iounmap(core_dev->dev->ce_base); +err_iomap: + free_irq(core_dev->irq, dev); + irq_dispose_mapping(core_dev->irq); + tasklet_kill(&core_dev->tasklet); +err_request_irq: + crypto4xx_destroy_sdr(core_dev->dev); +err_build_sdr: + crypto4xx_destroy_gdr(core_dev->dev); +err_build_gdr: + crypto4xx_destroy_pdr(core_dev->dev); +err_build_pdr: + kfree(core_dev->dev); +err_alloc_dev: + kfree(core_dev); + + return rc; +} + +static int __exit crypto4xx_remove(struct of_device *ofdev) +{ + struct device *dev = &ofdev->dev; + struct crypto4xx_core_device *core_dev = dev_get_drvdata(dev); + + free_irq(core_dev->irq, dev); + irq_dispose_mapping(core_dev->irq); + + tasklet_kill(&core_dev->tasklet); + /* Un-register with Linux CryptoAPI */ + crypto4xx_unregister_alg(core_dev->dev); + /* Free all allocated memory */ + crypto4xx_stop_all(core_dev); + + return 0; +} + +static struct of_device_id crypto4xx_match[] = { + { .compatible = "amcc,ppc4xx-crypto",}, + { }, +}; + +static struct of_platform_driver crypto4xx_driver = { + .name = "crypto4xx", + .match_table = crypto4xx_match, + .probe = crypto4xx_probe, + .remove = crypto4xx_remove, +}; + +static int __init crypto4xx_init(void) +{ + return of_register_platform_driver(&crypto4xx_driver); +} + +static void __exit crypto4xx_exit(void) +{ + of_unregister_platform_driver(&crypto4xx_driver); +} + +module_init(crypto4xx_init); +module_exit(crypto4xx_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("James Hsiao <jhsiao@amcc.com>"); +MODULE_DESCRIPTION("Driver for AMCC PPC4xx crypto accelerator"); + diff --git a/drivers/crypto/amcc/crypto4xx_core.h b/drivers/crypto/amcc/crypto4xx_core.h new file mode 100644 index 00000000000..1ef10344936 --- /dev/null +++ b/drivers/crypto/amcc/crypto4xx_core.h @@ -0,0 +1,177 @@ +/** + * AMCC SoC PPC4xx Crypto Driver + * + * Copyright (c) 2008 Applied Micro Circuits Corporation. + * All rights reserved. James Hsiao <jhsiao@amcc.com> + * + * 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. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * This is the header file for AMCC Crypto offload Linux device driver for + * use with Linux CryptoAPI. + + */ + +#ifndef __CRYPTO4XX_CORE_H__ +#define __CRYPTO4XX_CORE_H__ + +#define PPC460SX_SDR0_SRST 0x201 +#define PPC405EX_SDR0_SRST 0x200 +#define PPC460EX_SDR0_SRST 0x201 +#define PPC460EX_CE_RESET 0x08000000 +#define PPC460SX_CE_RESET 0x20000000 +#define PPC405EX_CE_RESET 0x00000008 + +#define CRYPTO4XX_CRYPTO_PRIORITY 300 +#define PPC4XX_LAST_PD 63 +#define PPC4XX_NUM_PD 64 +#define PPC4XX_LAST_GD 1023 +#define PPC4XX_NUM_GD 1024 +#define PPC4XX_LAST_SD 63 +#define PPC4XX_NUM_SD 64 +#define PPC4XX_SD_BUFFER_SIZE 2048 + +#define PD_ENTRY_INUSE 1 +#define PD_ENTRY_FREE 0 +#define ERING_WAS_FULL 0xffffffff + +struct crypto4xx_device; + +struct pd_uinfo { + struct crypto4xx_device *dev; + u32 state; + u32 using_sd; + u32 first_gd; /* first gather discriptor + used by this packet */ + u32 num_gd; /* number of gather discriptor + used by this packet */ + u32 first_sd; /* first scatter discriptor + used by this packet */ + u32 num_sd; /* number of scatter discriptors + used by this packet */ + void *sa_va; /* shadow sa, when using cp from ctx->sa */ + u32 sa_pa; + void *sr_va; /* state record for shadow sa */ + u32 sr_pa; + struct scatterlist *dest_va; + struct crypto_async_request *async_req; /* base crypto request + for this packet */ +}; + +struct crypto4xx_device { + struct crypto4xx_core_device *core_dev; + char *name; + u64 ce_phy_address; + void __iomem *ce_base; + + void *pdr; /* base address of packet + descriptor ring */ + dma_addr_t pdr_pa; /* physical address used to + program ce pdr_base_register */ + void *gdr; /* gather descriptor ring */ + dma_addr_t gdr_pa; /* physical address used to + program ce gdr_base_register */ + void *sdr; /* scatter descriptor ring */ + dma_addr_t sdr_pa; /* physical address used to + program ce sdr_base_register */ + void *scatter_buffer_va; + dma_addr_t scatter_buffer_pa; + u32 scatter_buffer_size; + + void *shadow_sa_pool; /* pool of memory for sa in pd_uinfo */ + dma_addr_t shadow_sa_pool_pa; + void *shadow_sr_pool; /* pool of memory for sr in pd_uinfo */ + dma_addr_t shadow_sr_pool_pa; + u32 pdr_tail; + u32 pdr_head; + u32 gdr_tail; + u32 gdr_head; + u32 sdr_tail; + u32 sdr_head; + void *pdr_uinfo; + struct list_head alg_list; /* List of algorithm supported + by this device */ +}; + +struct crypto4xx_core_device { + struct device *device; + struct of_device *ofdev; + struct crypto4xx_device *dev; + u32 int_status; + u32 irq; + struct tasklet_struct tasklet; + spinlock_t lock; +}; + +struct crypto4xx_ctx { + struct crypto4xx_device *dev; + void *sa_in; + dma_addr_t sa_in_dma_addr; + void *sa_out; + dma_addr_t sa_out_dma_addr; + void *state_record; + dma_addr_t state_record_dma_addr; + u32 sa_len; + u32 offset_to_sr_ptr; /* offset to state ptr, in dynamic sa */ + u32 direction; + u32 next_hdr; + u32 save_iv; + u32 pd_ctl_len; + u32 pd_ctl; + u32 bypass; + u32 is_hash; + u32 hash_final; +}; + +struct crypto4xx_req_ctx { + struct crypto4xx_device *dev; /* Device in which + operation to send to */ + void *sa; + u32 sa_dma_addr; + u16 sa_len; +}; + +struct crypto4xx_alg { + struct list_head entry; + struct crypto_alg alg; + struct crypto4xx_device *dev; +}; + +#define crypto_alg_to_crypto4xx_alg(x) \ + container_of(x, struct crypto4xx_alg, alg) + +extern int crypto4xx_alloc_sa(struct crypto4xx_ctx *ctx, u32 size); +extern void crypto4xx_free_sa(struct crypto4xx_ctx *ctx); +extern u32 crypto4xx_alloc_sa_rctx(struct crypto4xx_ctx *ctx, + struct crypto4xx_ctx *rctx); +extern void crypto4xx_free_sa_rctx(struct crypto4xx_ctx *rctx); +extern void crypto4xx_free_ctx(struct crypto4xx_ctx *ctx); +extern u32 crypto4xx_alloc_state_record(struct crypto4xx_ctx *ctx); +extern u32 get_dynamic_sa_offset_state_ptr_field(struct crypto4xx_ctx *ctx); +extern u32 get_dynamic_sa_offset_key_field(struct crypto4xx_ctx *ctx); +extern u32 get_dynamic_sa_iv_size(struct crypto4xx_ctx *ctx); +extern void crypto4xx_memcpy_le(unsigned int *dst, + const unsigned char *buf, int len); +extern u32 crypto4xx_build_pd(struct crypto_async_request *req, + struct crypto4xx_ctx *ctx, + struct scatterlist *src, + struct scatterlist *dst, + unsigned int datalen, + void *iv, u32 iv_len); +extern int crypto4xx_setkey_aes_cbc(struct crypto_ablkcipher *cipher, + const u8 *key, unsigned int keylen); +extern int crypto4xx_encrypt(struct ablkcipher_request *req); +extern int crypto4xx_decrypt(struct ablkcipher_request *req); +extern int crypto4xx_sha1_alg_init(struct crypto_tfm *tfm); +extern int crypto4xx_hash_digest(struct ahash_request *req); +extern int crypto4xx_hash_final(struct ahash_request *req); +extern int crypto4xx_hash_update(struct ahash_request *req); +extern int crypto4xx_hash_init(struct ahash_request *req); +#endif diff --git a/drivers/crypto/amcc/crypto4xx_reg_def.h b/drivers/crypto/amcc/crypto4xx_reg_def.h new file mode 100644 index 00000000000..7d4edb00261 --- /dev/null +++ b/drivers/crypto/amcc/crypto4xx_reg_def.h @@ -0,0 +1,284 @@ +/** + * AMCC SoC PPC4xx Crypto Driver + * + * Copyright (c) 2008 Applied Micro Circuits Corporation. + * All rights reserved. James Hsiao <jhsiao@amcc.com> + * + * 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. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * This filr defines the register set for Security Subsystem + */ + +#ifndef __CRYPTO4XX_REG_DEF_H__ +#define __CRYPTO4XX_REG_DEF_H__ + +/* CRYPTO4XX Register offset */ +#define CRYPTO4XX_DESCRIPTOR 0x00000000 +#define CRYPTO4XX_CTRL_STAT 0x00000000 +#define CRYPTO4XX_SOURCE 0x00000004 +#define CRYPTO4XX_DEST 0x00000008 +#define CRYPTO4XX_SA 0x0000000C +#define CRYPTO4XX_SA_LENGTH 0x00000010 +#define CRYPTO4XX_LENGTH 0x00000014 + +#define CRYPTO4XX_PE_DMA_CFG 0x00000040 +#define CRYPTO4XX_PE_DMA_STAT 0x00000044 +#define CRYPTO4XX_PDR_BASE 0x00000048 +#define CRYPTO4XX_RDR_BASE 0x0000004c +#define CRYPTO4XX_RING_SIZE 0x00000050 +#define CRYPTO4XX_RING_CTRL 0x00000054 +#define CRYPTO4XX_INT_RING_STAT 0x00000058 +#define CRYPTO4XX_EXT_RING_STAT 0x0000005c +#define CRYPTO4XX_IO_THRESHOLD 0x00000060 +#define CRYPTO4XX_GATH_RING_BASE 0x00000064 +#define CRYPTO4XX_SCAT_RING_BASE 0x00000068 +#define CRYPTO4XX_PART_RING_SIZE 0x0000006c +#define CRYPTO4XX_PART_RING_CFG 0x00000070 + +#define CRYPTO4XX_PDR_BASE_UADDR 0x00000080 +#define CRYPTO4XX_RDR_BASE_UADDR 0x00000084 +#define CRYPTO4XX_PKT_SRC_UADDR 0x00000088 +#define CRYPTO4XX_PKT_DEST_UADDR 0x0000008c +#define CRYPTO4XX_SA_UADDR 0x00000090 +#define CRYPTO4XX_GATH_RING_BASE_UADDR 0x000000A0 +#define CRYPTO4XX_SCAT_RING_BASE_UADDR 0x000000A4 + +#define CRYPTO4XX_SEQ_RD 0x00000408 +#define CRYPTO4XX_SEQ_MASK_RD 0x0000040C + +#define CRYPTO4XX_SA_CMD_0 0x00010600 +#define CRYPTO4XX_SA_CMD_1 0x00010604 + +#define CRYPTO4XX_STATE_PTR 0x000106dc +#define CRYPTO4XX_STATE_IV 0x00010700 +#define CRYPTO4XX_STATE_HASH_BYTE_CNT_0 0x00010710 +#define CRYPTO4XX_STATE_HASH_BYTE_CNT_1 0x00010714 + +#define CRYPTO4XX_STATE_IDIGEST_0 0x00010718 +#define CRYPTO4XX_STATE_IDIGEST_1 0x0001071c + +#define CRYPTO4XX_DATA_IN 0x00018000 +#define CRYPTO4XX_DATA_OUT 0x0001c000 + +#define CRYPTO4XX_INT_UNMASK_STAT 0x000500a0 +#define CRYPTO4XX_INT_MASK_STAT 0x000500a4 +#define CRYPTO4XX_INT_CLR 0x000500a4 +#define CRYPTO4XX_INT_EN 0x000500a8 + +#define CRYPTO4XX_INT_PKA 0x00000002 +#define CRYPTO4XX_INT_PDR_DONE 0x00008000 +#define CRYPTO4XX_INT_MA_WR_ERR 0x00020000 +#define CRYPTO4XX_INT_MA_RD_ERR 0x00010000 +#define CRYPTO4XX_INT_PE_ERR 0x00000200 +#define CRYPTO4XX_INT_USER_DMA_ERR 0x00000040 +#define CRYPTO4XX_INT_SLAVE_ERR 0x00000010 +#define CRYPTO4XX_INT_MASTER_ERR 0x00000008 +#define CRYPTO4XX_INT_ERROR 0x00030258 + +#define CRYPTO4XX_INT_CFG 0x000500ac +#define CRYPTO4XX_INT_DESCR_RD 0x000500b0 +#define CRYPTO4XX_INT_DESCR_CNT 0x000500b4 +#define CRYPTO4XX_INT_TIMEOUT_CNT 0x000500b8 + +#define CRYPTO4XX_DEVICE_CTRL 0x00060080 +#define CRYPTO4XX_DEVICE_ID 0x00060084 +#define CRYPTO4XX_DEVICE_INFO 0x00060088 +#define CRYPTO4XX_DMA_USER_SRC 0x00060094 +#define CRYPTO4XX_DMA_USER_DEST 0x00060098 +#define CRYPTO4XX_DMA_USER_CMD 0x0006009C + +#define CRYPTO4XX_DMA_CFG 0x000600d4 +#define CRYPTO4XX_BYTE_ORDER_CFG 0x000600d8 +#define CRYPTO4XX_ENDIAN_CFG 0x000600d8 + +#define CRYPTO4XX_PRNG_STAT 0x00070000 +#define CRYPTO4XX_PRNG_CTRL 0x00070004 +#define CRYPTO4XX_PRNG_SEED_L 0x00070008 +#define CRYPTO4XX_PRNG_SEED_H 0x0007000c + +#define CRYPTO4XX_PRNG_RES_0 0x00070020 +#define CRYPTO4XX_PRNG_RES_1 0x00070024 +#define CRYPTO4XX_PRNG_RES_2 0x00070028 +#define CRYPTO4XX_PRNG_RES_3 0x0007002C + +#define CRYPTO4XX_PRNG_LFSR_L 0x00070030 +#define CRYPTO4XX_PRNG_LFSR_H 0x00070034 + +/** + * Initilize CRYPTO ENGINE registers, and memory bases. + */ +#define PPC4XX_PDR_POLL 0x3ff +#define PPC4XX_OUTPUT_THRESHOLD 2 +#define PPC4XX_INPUT_THRESHOLD 2 +#define PPC4XX_PD_SIZE 6 +#define PPC4XX_CTX_DONE_INT 0x2000 +#define PPC4XX_PD_DONE_INT 0x8000 +#define PPC4XX_BYTE_ORDER 0x22222 +#define PPC4XX_INTERRUPT_CLR 0x3ffff +#define PPC4XX_PRNG_CTRL_AUTO_EN 0x3 +#define PPC4XX_DC_3DES_EN 1 +#define PPC4XX_INT_DESCR_CNT 4 +#define PPC4XX_INT_TIMEOUT_CNT 0 +#define PPC4XX_INT_CFG 1 +/** + * all follow define are ad hoc + */ +#define PPC4XX_RING_RETRY 100 +#define PPC4XX_RING_POLL 100 +#define PPC4XX_SDR_SIZE PPC4XX_NUM_SD +#define PPC4XX_GDR_SIZE PPC4XX_NUM_GD + +/** + * Generic Security Association (SA) with all possible fields. These will + * never likely used except for reference purpose. These structure format + * can be not changed as the hardware expects them to be layout as defined. + * Field can be removed or reduced but ordering can not be changed. + */ +#define CRYPTO4XX_DMA_CFG_OFFSET 0x40 +union ce_pe_dma_cfg { + struct { + u32 rsv:7; + u32 dir_host:1; + u32 rsv1:2; + u32 bo_td_en:1; + u32 dis_pdr_upd:1; + u32 bo_sgpd_en:1; + u32 bo_data_en:1; + u32 bo_sa_en:1; + u32 bo_pd_en:1; + u32 rsv2:4; + u32 dynamic_sa_en:1; + u32 pdr_mode:2; + u32 pe_mode:1; + u32 rsv3:5; + u32 reset_sg:1; + u32 reset_pdr:1; + u32 reset_pe:1; + } bf; + u32 w; +} __attribute__((packed)); + +#define CRYPTO4XX_PDR_BASE_OFFSET 0x48 +#define CRYPTO4XX_RDR_BASE_OFFSET 0x4c +#define CRYPTO4XX_RING_SIZE_OFFSET 0x50 +union ce_ring_size { + struct { + u32 ring_offset:16; + u32 rsv:6; + u32 ring_size:10; + } bf; + u32 w; +} __attribute__((packed)); + +#define CRYPTO4XX_RING_CONTROL_OFFSET 0x54 +union ce_ring_contol { + struct { + u32 continuous:1; + u32 rsv:5; + u32 ring_retry_divisor:10; + u32 rsv1:4; + u32 ring_poll_divisor:10; + } bf; + u32 w; +} __attribute__((packed)); + +#define CRYPTO4XX_IO_THRESHOLD_OFFSET 0x60 +union ce_io_threshold { + struct { + u32 rsv:6; + u32 output_threshold:10; + u32 rsv1:6; + u32 input_threshold:10; + } bf; + u32 w; +} __attribute__((packed)); + +#define CRYPTO4XX_GATHER_RING_BASE_OFFSET 0x64 +#define CRYPTO4XX_SCATTER_RING_BASE_OFFSET 0x68 + +union ce_part_ring_size { + struct { + u32 sdr_size:16; + u32 gdr_size:16; + } bf; + u32 w; +} __attribute__((packed)); + +#define MAX_BURST_SIZE_32 0 +#define MAX_BURST_SIZE_64 1 +#define MAX_BURST_SIZE_128 2 +#define MAX_BURST_SIZE_256 3 + +/* gather descriptor control length */ +struct gd_ctl_len { + u32 len:16; + u32 rsv:14; + u32 done:1; + u32 ready:1; +} __attribute__((packed)); + +struct ce_gd { + u32 ptr; + struct gd_ctl_len ctl_len; +} __attribute__((packed)); + +struct sd_ctl { + u32 ctl:30; + u32 done:1; + u32 rdy:1; +} __attribute__((packed)); + +struct ce_sd { + u32 ptr; + struct sd_ctl ctl; +} __attribute__((packed)); + +#define PD_PAD_CTL_32 0x10 +#define PD_PAD_CTL_64 0x20 +#define PD_PAD_CTL_128 0x40 +#define PD_PAD_CTL_256 0x80 +union ce_pd_ctl { + struct { + u32 pd_pad_ctl:8; + u32 status:8; + u32 next_hdr:8; + u32 rsv:2; + u32 cached_sa:1; + u32 hash_final:1; + u32 init_arc4:1; + u32 rsv1:1; + u32 pe_done:1; + u32 host_ready:1; + } bf; + u32 w; +} __attribute__((packed)); + +union ce_pd_ctl_len { + struct { + u32 bypass:8; + u32 pe_done:1; + u32 host_ready:1; + u32 rsv:2; + u32 pkt_len:20; + } bf; + u32 w; +} __attribute__((packed)); + +struct ce_pd { + union ce_pd_ctl pd_ctl; + u32 src; + u32 dest; + u32 sa; /* get from ctx->sa_dma_addr */ + u32 sa_len; /* only if dynamic sa is used */ + union ce_pd_ctl_len pd_ctl_len; + +} __attribute__((packed)); +#endif diff --git a/drivers/crypto/amcc/crypto4xx_sa.c b/drivers/crypto/amcc/crypto4xx_sa.c new file mode 100644 index 00000000000..466fd94cd4a --- /dev/null +++ b/drivers/crypto/amcc/crypto4xx_sa.c @@ -0,0 +1,108 @@ +/** + * AMCC SoC PPC4xx Crypto Driver + * + * Copyright (c) 2008 Applied Micro Circuits Corporation. + * All rights reserved. James Hsiao <jhsiao@amcc.com> + * + * 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. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * @file crypto4xx_sa.c + * + * This file implements the security context + * assoicate format. + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/mod_devicetable.h> +#include <linux/interrupt.h> +#include <linux/spinlock_types.h> +#include <linux/highmem.h> +#include <linux/scatterlist.h> +#include <linux/crypto.h> +#include <crypto/algapi.h> +#include <crypto/des.h> +#include "crypto4xx_reg_def.h" +#include "crypto4xx_sa.h" +#include "crypto4xx_core.h" + +u32 get_dynamic_sa_offset_iv_field(struct crypto4xx_ctx *ctx) +{ + u32 offset; + union dynamic_sa_contents cts; + + if (ctx->direction == DIR_INBOUND) + cts.w = ((struct dynamic_sa_ctl *)(ctx->sa_in))->sa_contents; + else + cts.w = ((struct dynamic_sa_ctl *)(ctx->sa_out))->sa_contents; + offset = cts.bf.key_size + + cts.bf.inner_size + + cts.bf.outer_size + + cts.bf.spi + + cts.bf.seq_num0 + + cts.bf.seq_num1 + + cts.bf.seq_num_mask0 + + cts.bf.seq_num_mask1 + + cts.bf.seq_num_mask2 + + cts.bf.seq_num_mask3; + + return sizeof(struct dynamic_sa_ctl) + offset * 4; +} + +u32 get_dynamic_sa_offset_state_ptr_field(struct crypto4xx_ctx *ctx) +{ + u32 offset; + union dynamic_sa_contents cts; + + if (ctx->direction == DIR_INBOUND) + cts.w = ((struct dynamic_sa_ctl *) ctx->sa_in)->sa_contents; + else + cts.w = ((struct dynamic_sa_ctl *) ctx->sa_out)->sa_contents; + offset = cts.bf.key_size + + cts.bf.inner_size + + cts.bf.outer_size + + cts.bf.spi + + cts.bf.seq_num0 + + cts.bf.seq_num1 + + cts.bf.seq_num_mask0 + + cts.bf.seq_num_mask1 + + cts.bf.seq_num_mask2 + + cts.bf.seq_num_mask3 + + cts.bf.iv0 + + cts.bf.iv1 + + cts.bf.iv2 + + cts.bf.iv3; + + return sizeof(struct dynamic_sa_ctl) + offset * 4; +} + +u32 get_dynamic_sa_iv_size(struct crypto4xx_ctx *ctx) +{ + union dynamic_sa_contents cts; + + if (ctx->direction == DIR_INBOUND) + cts.w = ((struct dynamic_sa_ctl *) ctx->sa_in)->sa_contents; + else + cts.w = ((struct dynamic_sa_ctl *) ctx->sa_out)->sa_contents; + return (cts.bf.iv0 + cts.bf.iv1 + cts.bf.iv2 + cts.bf.iv3) * 4; +} + +u32 get_dynamic_sa_offset_key_field(struct crypto4xx_ctx *ctx) +{ + union dynamic_sa_contents cts; + + if (ctx->direction == DIR_INBOUND) + cts.w = ((struct dynamic_sa_ctl *) ctx->sa_in)->sa_contents; + else + cts.w = ((struct dynamic_sa_ctl *) ctx->sa_out)->sa_contents; + + return sizeof(struct dynamic_sa_ctl); +} diff --git a/drivers/crypto/amcc/crypto4xx_sa.h b/drivers/crypto/amcc/crypto4xx_sa.h new file mode 100644 index 00000000000..4b83ed7e557 --- /dev/null +++ b/drivers/crypto/amcc/crypto4xx_sa.h @@ -0,0 +1,243 @@ +/** + * AMCC SoC PPC4xx Crypto Driver + * + * Copyright (c) 2008 Applied Micro Circuits Corporation. + * All rights reserved. James Hsiao <jhsiao@amcc.com> + * + * 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. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * This file defines the security context + * assoicate format. + */ + +#ifndef __CRYPTO4XX_SA_H__ +#define __CRYPTO4XX_SA_H__ + +#define AES_IV_SIZE 16 + +/** + * Contents of Dynamic Security Association (SA) with all possible fields + */ +union dynamic_sa_contents { + struct { + u32 arc4_state_ptr:1; + u32 arc4_ij_ptr:1; + u32 state_ptr:1; + u32 iv3:1; + u32 iv2:1; + u32 iv1:1; + u32 iv0:1; + u32 seq_num_mask3:1; + u32 seq_num_mask2:1; + u32 seq_num_mask1:1; + u32 seq_num_mask0:1; + u32 seq_num1:1; + u32 seq_num0:1; + u32 spi:1; + u32 outer_size:5; + u32 inner_size:5; + u32 key_size:4; + u32 cmd_size:4; + } bf; + u32 w; +} __attribute__((packed)); + +#define DIR_OUTBOUND 0 +#define DIR_INBOUND 1 +#define SA_OP_GROUP_BASIC 0 +#define SA_OPCODE_ENCRYPT 0 +#define SA_OPCODE_DECRYPT 0 +#define SA_OPCODE_HASH 3 +#define SA_CIPHER_ALG_DES 0 +#define SA_CIPHER_ALG_3DES 1 +#define SA_CIPHER_ALG_ARC4 2 +#define SA_CIPHER_ALG_AES 3 +#define SA_CIPHER_ALG_KASUMI 4 +#define SA_CIPHER_ALG_NULL 15 + +#define SA_HASH_ALG_MD5 0 +#define SA_HASH_ALG_SHA1 1 +#define SA_HASH_ALG_NULL 15 +#define SA_HASH_ALG_SHA1_DIGEST_SIZE 20 + +#define SA_LOAD_HASH_FROM_SA 0 +#define SA_LOAD_HASH_FROM_STATE 2 +#define SA_NOT_LOAD_HASH 3 +#define SA_LOAD_IV_FROM_SA 0 +#define SA_LOAD_IV_FROM_INPUT 1 +#define SA_LOAD_IV_FROM_STATE 2 +#define SA_LOAD_IV_GEN_IV 3 + +#define SA_PAD_TYPE_CONSTANT 2 +#define SA_PAD_TYPE_ZERO 3 +#define SA_PAD_TYPE_TLS 5 +#define SA_PAD_TYPE_DTLS 5 +#define SA_NOT_SAVE_HASH 0 +#define SA_SAVE_HASH 1 +#define SA_NOT_SAVE_IV 0 +#define SA_SAVE_IV 1 +#define SA_HEADER_PROC 1 +#define SA_NO_HEADER_PROC 0 + +union sa_command_0 { + struct { + u32 scatter:1; + u32 gather:1; + u32 save_hash_state:1; + u32 save_iv:1; + u32 load_hash_state:2; + u32 load_iv:2; + u32 digest_len:4; + u32 hdr_proc:1; + u32 extend_pad:1; + u32 stream_cipher_pad:1; + u32 rsv:1; + u32 hash_alg:4; + u32 cipher_alg:4; + u32 pad_type:2; + u32 op_group:2; + u32 dir:1; + u32 opcode:3; + } bf; + u32 w; +} __attribute__((packed)); + +#define CRYPTO_MODE_ECB 0 +#define CRYPTO_MODE_CBC 1 + +#define CRYPTO_FEEDBACK_MODE_NO_FB 0 +#define CRYPTO_FEEDBACK_MODE_64BIT_OFB 0 +#define CRYPTO_FEEDBACK_MODE_8BIT_CFB 1 +#define CRYPTO_FEEDBACK_MODE_1BIT_CFB 2 +#define CRYPTO_FEEDBACK_MODE_128BIT_CFB 3 + +#define SA_AES_KEY_LEN_128 2 +#define SA_AES_KEY_LEN_192 3 +#define SA_AES_KEY_LEN_256 4 + +#define SA_REV2 1 +/** + * The follow defines bits sa_command_1 + * In Basic hash mode this bit define simple hash or hmac. + * In IPsec mode, this bit define muting control. + */ +#define SA_HASH_MODE_HASH 0 +#define SA_HASH_MODE_HMAC 1 +#define SA_MC_ENABLE 0 +#define SA_MC_DISABLE 1 +#define SA_NOT_COPY_HDR 0 +#define SA_COPY_HDR 1 +#define SA_NOT_COPY_PAD 0 +#define SA_COPY_PAD 1 +#define SA_NOT_COPY_PAYLOAD 0 +#define SA_COPY_PAYLOAD 1 +#define SA_EXTENDED_SN_OFF 0 +#define SA_EXTENDED_SN_ON 1 +#define SA_SEQ_MASK_OFF 0 +#define SA_SEQ_MASK_ON 1 + +union sa_command_1 { + struct { + u32 crypto_mode31:1; + u32 save_arc4_state:1; + u32 arc4_stateful:1; + u32 key_len:5; + u32 hash_crypto_offset:8; + u32 sa_rev:2; + u32 byte_offset:1; + u32 hmac_muting:1; + u32 feedback_mode:2; + u32 crypto_mode9_8:2; + u32 extended_seq_num:1; + u32 seq_num_mask:1; + u32 mutable_bit_proc:1; + u32 ip_version:1; + u32 copy_pad:1; + u32 copy_payload:1; + u32 copy_hdr:1; + u32 rsv1:1; + } bf; + u32 w; +} __attribute__((packed)); + +struct dynamic_sa_ctl { + u32 sa_contents; + union sa_command_0 sa_command_0; + union sa_command_1 sa_command_1; +} __attribute__((packed)); + +/** + * State Record for Security Association (SA) + */ +struct sa_state_record { + u32 save_iv[4]; + u32 save_hash_byte_cnt[2]; + u32 save_digest[16]; +} __attribute__((packed)); + +/** + * Security Association (SA) for AES128 + * + */ +struct dynamic_sa_aes128 { + struct dynamic_sa_ctl ctrl; + u32 key[4]; + u32 iv[4]; /* for CBC, OFC, and CFB mode */ + u32 state_ptr; + u32 reserved; +} __attribute__((packed)); + +#define SA_AES128_LEN (sizeof(struct dynamic_sa_aes128)/4) +#define SA_AES128_CONTENTS 0x3e000042 + +/* + * Security Association (SA) for AES192 + */ +struct dynamic_sa_aes192 { + struct dynamic_sa_ctl ctrl; + u32 key[6]; + u32 iv[4]; /* for CBC, OFC, and CFB mode */ + u32 state_ptr; + u32 reserved; +} __attribute__((packed)); + +#define SA_AES192_LEN (sizeof(struct dynamic_sa_aes192)/4) +#define SA_AES192_CONTENTS 0x3e000062 + +/** + * Security Association (SA) for AES256 + */ +struct dynamic_sa_aes256 { + struct dynamic_sa_ctl ctrl; + u32 key[8]; + u32 iv[4]; /* for CBC, OFC, and CFB mode */ + u32 state_ptr; + u32 reserved; +} __attribute__((packed)); + +#define SA_AES256_LEN (sizeof(struct dynamic_sa_aes256)/4) +#define SA_AES256_CONTENTS 0x3e000082 +#define SA_AES_CONTENTS 0x3e000002 + +/** + * Security Association (SA) for HASH160: HMAC-SHA1 + */ +struct dynamic_sa_hash160 { + struct dynamic_sa_ctl ctrl; + u32 inner_digest[5]; + u32 outer_digest[5]; + u32 state_ptr; + u32 reserved; +} __attribute__((packed)); +#define SA_HASH160_LEN (sizeof(struct dynamic_sa_hash160)/4) +#define SA_HASH160_CONTENTS 0x2000a502 + +#endif |