7 files changed, 251 insertions, 47 deletions

diff --git a/arch/x86/crypto/Makefile b/arch/x86/crypto/Makefile
index 3874c2de5403..903de4aa5094 100644
--- a/arch/x86/crypto/Makefile
+++ b/arch/x86/crypto/Makefile
@@ -10,6 +10,8 @@ obj-$(CONFIG_CRYPTO_AES_X86_64) += aes-x86_64.o
 obj-$(CONFIG_CRYPTO_TWOFISH_X86_64) += twofish-x86_64.o
 obj-$(CONFIG_CRYPTO_SALSA20_X86_64) += salsa20-x86_64.o
 
+obj-$(CONFIG_CRYPTO_CRC32C_INTEL) += crc32c-intel.o
+
 aes-i586-y := aes-i586-asm_32.o aes_glue.o
 twofish-i586-y := twofish-i586-asm_32.o twofish_glue.o
 salsa20-i586-y := salsa20-i586-asm_32.o salsa20_glue.o
diff --git a/arch/x86/crypto/crc32c-intel.c b/arch/x86/crypto/crc32c-intel.c
new file mode 100644
index 000000000000..070afc5b6c94
--- /dev/null
+++ b/arch/x86/crypto/crc32c-intel.c
@@ -0,0 +1,197 @@
+/*
+ * Using hardware provided CRC32 instruction to accelerate the CRC32 disposal.
+ * CRC32C polynomial:0x1EDC6F41(BE)/0x82F63B78(LE)
+ * CRC32 is a new instruction in Intel SSE4.2, the reference can be found at:
+ * http://www.intel.com/products/processor/manuals/
+ * Intel(R) 64 and IA-32 Architectures Software Developer's Manual
+ * Volume 2A: Instruction Set Reference, A-M
+ *
+ * Copyright (c) 2008 Austin Zhang <austin_zhang@linux.intel.com>
+ * Copyright (c) 2008 Kent Liu <kent.liu@intel.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.
+ *
+ */
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <crypto/internal/hash.h>
+
+#include <asm/cpufeature.h>
+
+#define CHKSUM_BLOCK_SIZE       1
+#define CHKSUM_DIGEST_SIZE      4
+
+#define SCALE_F sizeof(unsigned long)
+
+#ifdef CONFIG_X86_64
+#define REX_PRE "0x48, "
+#else
+#define REX_PRE
+#endif
+
+static u32 crc32c_intel_le_hw_byte(u32 crc, unsigned char const *data, size_t length)
+{
+        while (length--) {
+                __asm__ __volatile__(
+                        ".byte 0xf2, 0xf, 0x38, 0xf0, 0xf1"
+                        :"=S"(crc)
+                        :"0"(crc), "c"(*data)
+                );
+                data++;
+        }
+
+        return crc;
+}
+
+static u32 __pure crc32c_intel_le_hw(u32 crc, unsigned char const *p, size_t len)
+{
+        unsigned int iquotient = len / SCALE_F;
+        unsigned int iremainder = len % SCALE_F;
+        unsigned long *ptmp = (unsigned long *)p;
+
+        while (iquotient--) {
+                __asm__ __volatile__(
+                        ".byte 0xf2, " REX_PRE "0xf, 0x38, 0xf1, 0xf1;"
+                        :"=S"(crc)
+                        :"0"(crc), "c"(*ptmp)
+                );
+                ptmp++;
+        }
+
+        if (iremainder)
+                crc = crc32c_intel_le_hw_byte(crc, (unsigned char *)ptmp,
+                                 iremainder);
+
+        return crc;
+}
+
+/*
+ * Setting the seed allows arbitrary accumulators and flexible XOR policy
+ * If your algorithm starts with ~0, then XOR with ~0 before you set
+ * the seed.
+ */
+static int crc32c_intel_setkey(struct crypto_ahash *hash, const u8 *key,
+                        unsigned int keylen)
+{
+        u32 *mctx = crypto_ahash_ctx(hash);
+
+        if (keylen != sizeof(u32)) {
+                crypto_ahash_set_flags(hash, CRYPTO_TFM_RES_BAD_KEY_LEN);
+                return -EINVAL;
+        }
+        *mctx = le32_to_cpup((__le32 *)key);
+        return 0;
+}
+
+static int crc32c_intel_init(struct ahash_request *req)
+{
+        u32 *mctx = crypto_ahash_ctx(crypto_ahash_reqtfm(req));
+        u32 *crcp = ahash_request_ctx(req);
+
+        *crcp = *mctx;
+
+        return 0;
+}
+
+static int crc32c_intel_update(struct ahash_request *req)
+{
+        struct crypto_hash_walk walk;
+        u32 *crcp = ahash_request_ctx(req);
+        u32 crc = *crcp;
+        int nbytes;
+
+        for (nbytes = crypto_hash_walk_first(req, &walk); nbytes;
+           nbytes = crypto_hash_walk_done(&walk, 0))
+        crc = crc32c_intel_le_hw(crc, walk.data, nbytes);
+
+        *crcp = crc;
+        return 0;
+}
+
+static int crc32c_intel_final(struct ahash_request *req)
+{
+        u32 *crcp = ahash_request_ctx(req);
+
+        *(__le32 *)req->result = ~cpu_to_le32p(crcp);
+        return 0;
+}
+
+static int crc32c_intel_digest(struct ahash_request *req)
+{
+        struct crypto_hash_walk walk;
+        u32 *mctx = crypto_ahash_ctx(crypto_ahash_reqtfm(req));
+        u32 crc = *mctx;
+        int nbytes;
+
+        for (nbytes = crypto_hash_walk_first(req, &walk); nbytes;
+           nbytes = crypto_hash_walk_done(&walk, 0))
+                crc = crc32c_intel_le_hw(crc, walk.data, nbytes);
+
+        *(__le32 *)req->result = ~cpu_to_le32(crc);
+        return 0;
+}
+
+static int crc32c_intel_cra_init(struct crypto_tfm *tfm)
+{
+        u32 *key = crypto_tfm_ctx(tfm);
+
+        *key = ~0;
+
+        tfm->crt_ahash.reqsize = sizeof(u32);
+
+        return 0;
+}
+
+static struct crypto_alg alg = {
+        .cra_name               =       "crc32c",
+        .cra_driver_name        =       "crc32c-intel",
+        .cra_priority           =       200,
+        .cra_flags              =       CRYPTO_ALG_TYPE_AHASH,
+        .cra_blocksize          =       CHKSUM_BLOCK_SIZE,
+        .cra_alignmask          =       3,
+        .cra_ctxsize            =       sizeof(u32),
+        .cra_module             =       THIS_MODULE,
+        .cra_list               =       LIST_HEAD_INIT(alg.cra_list),
+        .cra_init               =       crc32c_intel_cra_init,
+        .cra_type               =       &crypto_ahash_type,
+        .cra_u                  =       {
+                .ahash = {
+                        .digestsize    =       CHKSUM_DIGEST_SIZE,
+                        .setkey        =       crc32c_intel_setkey,
+                        .init          =       crc32c_intel_init,
+                        .update        =       crc32c_intel_update,
+                        .final         =       crc32c_intel_final,
+                        .digest        =       crc32c_intel_digest,
+                }
+        }
+};
+
+
+static int __init crc32c_intel_mod_init(void)
+{
+        if (cpu_has_xmm4_2)
+                return crypto_register_alg(&alg);
+        else
+                return -ENODEV;
+}
+
+static void __exit crc32c_intel_mod_fini(void)
+{
+        crypto_unregister_alg(&alg);
+}
+
+module_init(crc32c_intel_mod_init);
+module_exit(crc32c_intel_mod_fini);
+
+MODULE_AUTHOR("Austin Zhang <austin.zhang@intel.com>, Kent Liu <kent.liu@intel.com>");
+MODULE_DESCRIPTION("CRC32c (Castagnoli) optimization using Intel Hardware.");
+MODULE_LICENSE("GPL");
+
+MODULE_ALIAS("crc32c");
+MODULE_ALIAS("crc32c-intel");
+
diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index dd097b835839..c24c4a487b7c 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -256,7 +256,8 @@ static u32 get_cur_val(const cpumask_t *mask)
  * Only IA32_APERF/IA32_MPERF ratio is architecturally defined and
  * no meaning should be associated with absolute values of these MSRs.
  */
-static unsigned int get_measured_perf(unsigned int cpu)
+static unsigned int get_measured_perf(struct cpufreq_policy *policy,
+                                      unsigned int cpu)
 {
         union {
                 struct {
@@ -326,7 +327,7 @@ static unsigned int get_measured_perf(unsigned int cpu)
 
 #endif
 
-        retval = per_cpu(drv_data, cpu)->max_freq * perf_percent / 100;
+        retval = per_cpu(drv_data, policy->cpu)->max_freq * perf_percent / 100;
 
         put_cpu();
         set_cpus_allowed_ptr(current, &saved_mask);
@@ -785,7 +786,11 @@ static int __init acpi_cpufreq_init(void)
         if (ret)
                 return ret;
 
-        return cpufreq_register_driver(&acpi_cpufreq_driver);
+        ret = cpufreq_register_driver(&acpi_cpufreq_driver);
+        if (ret)
+                free_percpu(acpi_perf_data);
+
+        return ret;
 }
 
 static void __exit acpi_cpufreq_exit(void)
@@ -795,8 +800,6 @@ static void __exit acpi_cpufreq_exit(void)
         cpufreq_unregister_driver(&acpi_cpufreq_driver);
 
         free_percpu(acpi_perf_data);
-
-        return;
 }
 
 module_param(acpi_pstate_strict, uint, 0644);
diff --git a/arch/x86/kernel/cpu/cpufreq/elanfreq.c b/arch/x86/kernel/cpu/cpufreq/elanfreq.c
index e4a4bf870e94..fe613c93b366 100644
--- a/arch/x86/kernel/cpu/cpufreq/elanfreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/elanfreq.c
@@ -25,8 +25,8 @@
 #include <linux/cpufreq.h>
 
 #include <asm/msr.h>
-#include <asm/timex.h>
-#include <asm/io.h>
+#include <linux/timex.h>
+#include <linux/io.h>
 
 #define REG_CSCIR 0x22          /* Chip Setup and Control Index Register    */
 #define REG_CSCDR 0x23          /* Chip Setup and Control Data  Register    */
@@ -82,7 +82,7 @@ static unsigned int elanfreq_get_cpu_frequency(unsigned int cpu)
         u8 clockspeed_reg;    /* Clock Speed Register */
 
         local_irq_disable();
-        outb_p(0x80,REG_CSCIR);
+        outb_p(0x80, REG_CSCIR);
         clockspeed_reg = inb_p(REG_CSCDR);
         local_irq_enable();
 
@@ -98,10 +98,10 @@ static unsigned int elanfreq_get_cpu_frequency(unsigned int cpu)
         }
 
         /* 33 MHz is not 32 MHz... */
-        if ((clockspeed_reg & 0xE0)==0xA0)
+        if ((clockspeed_reg & 0xE0) == 0xA0)
                 return 33000;
 
-        return ((1<<((clockspeed_reg & 0xE0) >> 5)) * 1000);
+        return (1<<((clockspeed_reg & 0xE0) >> 5)) * 1000;
 }
 
 
@@ -117,7 +117,7 @@ static unsigned int elanfreq_get_cpu_frequency(unsigned int cpu)
  *      There is no return value.
  */
 
-static void elanfreq_set_cpu_state (unsigned int state)
+static void elanfreq_set_cpu_state(unsigned int state)
 {
         struct cpufreq_freqs    freqs;
 
@@ -144,20 +144,20 @@ static void elanfreq_set_cpu_state (unsigned int state)
          */
 
         local_irq_disable();
-        outb_p(0x40,REG_CSCIR);         /* Disable hyperspeed mode */
-        outb_p(0x00,REG_CSCDR);
+        outb_p(0x40, REG_CSCIR);                /* Disable hyperspeed mode */
+        outb_p(0x00, REG_CSCDR);
         local_irq_enable();             /* wait till internal pipelines and */
         udelay(1000);                   /* buffers have cleaned up          */
 
         local_irq_disable();
 
         /* now, set the CPU clock speed register (0x80) */
-        outb_p(0x80,REG_CSCIR);
-        outb_p(elan_multiplier[state].val80h,REG_CSCDR);
+        outb_p(0x80, REG_CSCIR);
+        outb_p(elan_multiplier[state].val80h, REG_CSCDR);
 
         /* now, the hyperspeed bit in PMU Force Mode Register (0x40) */
-        outb_p(0x40,REG_CSCIR);
-        outb_p(elan_multiplier[state].val40h,REG_CSCDR);
+        outb_p(0x40, REG_CSCIR);
+        outb_p(elan_multiplier[state].val40h, REG_CSCDR);
         udelay(10000);
         local_irq_enable();
 
@@ -173,12 +173,12 @@ static void elanfreq_set_cpu_state (unsigned int state)
  *      for the hardware supported by the driver.
  */
 
-static int elanfreq_verify (struct cpufreq_policy *policy)
+static int elanfreq_verify(struct cpufreq_policy *policy)
 {
         return cpufreq_frequency_table_verify(policy, &elanfreq_table[0]);
 }
 
-static int elanfreq_target (struct cpufreq_policy *policy,
+static int elanfreq_target(struct cpufreq_policy *policy,
                             unsigned int target_freq,
                             unsigned int relation)
 {
@@ -205,7 +205,7 @@ static int elanfreq_cpu_init(struct cpufreq_policy *policy)
 
         /* capability check */
         if ((c->x86_vendor != X86_VENDOR_AMD) ||
-            (c->x86 != 4) || (c->x86_model!=10))
+            (c->x86 != 4) || (c->x86_model != 10))
                 return -ENODEV;
 
         /* max freq */
@@ -213,7 +213,7 @@ static int elanfreq_cpu_init(struct cpufreq_policy *policy)
                 max_freq = elanfreq_get_cpu_frequency(0);
 
         /* table init */
-        for (i=0; (elanfreq_table[i].frequency != CPUFREQ_TABLE_END); i++) {
+        for (i = 0; (elanfreq_table[i].frequency != CPUFREQ_TABLE_END); i++) {
                 if (elanfreq_table[i].frequency > max_freq)
                         elanfreq_table[i].frequency = CPUFREQ_ENTRY_INVALID;
         }
@@ -224,7 +224,7 @@ static int elanfreq_cpu_init(struct cpufreq_policy *policy)
 
         result = cpufreq_frequency_table_cpuinfo(policy, elanfreq_table);
         if (result)
-                return (result);
+                return result;
 
         cpufreq_frequency_table_get_attr(elanfreq_table, policy->cpu);
         return 0;
@@ -260,7 +260,7 @@ __setup("elanfreq=", elanfreq_setup);
 #endif
 
 
-static struct freq_attr* elanfreq_attr[] = {
+static struct freq_attr *elanfreq_attr[] = {
         &cpufreq_freq_attr_scaling_available_freqs,
         NULL,
 };
@@ -284,9 +284,9 @@ static int __init elanfreq_init(void)
 
         /* Test if we have the right hardware */
         if ((c->x86_vendor != X86_VENDOR_AMD) ||
-                (c->x86 != 4) || (c->x86_model!=10)) {
+                (c->x86 != 4) || (c->x86_model != 10)) {
                 printk(KERN_INFO "elanfreq: error: no Elan processor found!\n");
-                return -ENODEV;
+                return -ENODEV;
         }
         return cpufreq_register_driver(&elanfreq_driver);
 }
@@ -298,7 +298,7 @@ static void __exit elanfreq_exit(void)
 }
 
 
-module_param (max_freq, int, 0444);
+module_param(max_freq, int, 0444);
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Robert Schwebel <r.schwebel@pengutronix.de>, Sven Geggus <sven@geggus.net>");
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c
index eb9b62b0830c..b5ced806a316 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c
@@ -15,12 +15,11 @@
 #include <linux/slab.h>
 
 #include <asm/msr.h>
-#include <asm/timex.h>
-#include <asm/io.h>
+#include <linux/timex.h>
+#include <linux/io.h>
 
-
-#define POWERNOW_IOPORT 0xfff0         /* it doesn't matter where, as long
-                                          as it is unused */
+#define POWERNOW_IOPORT 0xfff0          /* it doesn't matter where, as long
+                                           as it is unused */
 
 static unsigned int                     busfreq;   /* FSB, in 10 kHz */
 static unsigned int                     max_multiplier;
@@ -53,7 +52,7 @@ static int powernow_k6_get_cpu_multiplier(void)
 
         msrval = POWERNOW_IOPORT + 0x1;
         wrmsr(MSR_K6_EPMR, msrval, 0); /* enable the PowerNow port */
-        invalue=inl(POWERNOW_IOPORT + 0x8);
+        invalue = inl(POWERNOW_IOPORT + 0x8);
         msrval = POWERNOW_IOPORT + 0x0;
         wrmsr(MSR_K6_EPMR, msrval, 0); /* disable it again */
 
@@ -67,9 +66,9 @@ static int powernow_k6_get_cpu_multiplier(void)
  *
  *   Tries to change the PowerNow! multiplier
  */
-static void powernow_k6_set_state (unsigned int best_i)
+static void powernow_k6_set_state(unsigned int best_i)
 {
-        unsigned long           outvalue=0, invalue=0;
+        unsigned long           outvalue = 0, invalue = 0;
         unsigned long           msrval;
         struct cpufreq_freqs    freqs;
 
@@ -90,10 +89,10 @@ static void powernow_k6_set_state (unsigned int best_i)
 
         msrval = POWERNOW_IOPORT + 0x1;
         wrmsr(MSR_K6_EPMR, msrval, 0); /* enable the PowerNow port */
-        invalue=inl(POWERNOW_IOPORT + 0x8);
+        invalue = inl(POWERNOW_IOPORT + 0x8);
         invalue = invalue & 0xf;
         outvalue = outvalue | invalue;
-        outl(outvalue ,(POWERNOW_IOPORT + 0x8));
+        outl(outvalue , (POWERNOW_IOPORT + 0x8));
         msrval = POWERNOW_IOPORT + 0x0;
         wrmsr(MSR_K6_EPMR, msrval, 0); /* disable it again */
 
@@ -124,7 +123,7 @@ static int powernow_k6_verify(struct cpufreq_policy *policy)
  *
  * sets a new CPUFreq policy
  */
-static int powernow_k6_target (struct cpufreq_policy *policy,
+static int powernow_k6_target(struct cpufreq_policy *policy,
                                unsigned int target_freq,
                                unsigned int relation)
 {
@@ -152,7 +151,7 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy)
         busfreq = cpu_khz / max_multiplier;
 
         /* table init */
-        for (i=0; (clock_ratio[i].frequency != CPUFREQ_TABLE_END); i++) {
+        for (i = 0; (clock_ratio[i].frequency != CPUFREQ_TABLE_END); i++) {
                 if (clock_ratio[i].index > max_multiplier)
                         clock_ratio[i].frequency = CPUFREQ_ENTRY_INVALID;
                 else
@@ -165,7 +164,7 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy)
 
         result = cpufreq_frequency_table_cpuinfo(policy, clock_ratio);
         if (result)
-                return (result);
+                return result;
 
         cpufreq_frequency_table_get_attr(clock_ratio, policy->cpu);
 
@@ -176,8 +175,8 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy)
 static int powernow_k6_cpu_exit(struct cpufreq_policy *policy)
 {
         unsigned int i;
-        for (i=0; i<8; i++) {
-                if (i==max_multiplier)
+        for (i = 0; i < 8; i++) {
+                if (i == max_multiplier)
                         powernow_k6_set_state(i);
         }
         cpufreq_frequency_table_put_attr(policy->cpu);
@@ -189,7 +188,7 @@ static unsigned int powernow_k6_get(unsigned int cpu)
         return busfreq * powernow_k6_get_cpu_multiplier();
 }
 
-static struct freq_attr* powernow_k6_attr[] = {
+static struct freq_attr *powernow_k6_attr[] = {
         &cpufreq_freq_attr_scaling_available_freqs,
         NULL,
 };
@@ -227,7 +226,7 @@ static int __init powernow_k6_init(void)
         }
 
         if (cpufreq_register_driver(&powernow_k6_driver)) {
-                release_region (POWERNOW_IOPORT, 16);
+                release_region(POWERNOW_IOPORT, 16);
                 return -EINVAL;
         }
 
@@ -243,13 +242,13 @@ static int __init powernow_k6_init(void)
 static void __exit powernow_k6_exit(void)
 {
         cpufreq_unregister_driver(&powernow_k6_driver);
-        release_region (POWERNOW_IOPORT, 16);
+        release_region(POWERNOW_IOPORT, 16);
 }
 
 
-MODULE_AUTHOR ("Arjan van de Ven <arjanv@redhat.com>, Dave Jones <davej@codemonkey.org.uk>, Dominik Brodowski <linux@brodo.de>");
-MODULE_DESCRIPTION ("PowerNow! driver for AMD K6-2+ / K6-3+ processors.");
-MODULE_LICENSE ("GPL");
+MODULE_AUTHOR("Arjan van de Ven <arjanv@redhat.com>, Dave Jones <davej@codemonkey.org.uk>, Dominik Brodowski <linux@brodo.de>");
+MODULE_DESCRIPTION("PowerNow! driver for AMD K6-2+ / K6-3+ processors.");
+MODULE_LICENSE("GPL");
 
 module_init(powernow_k6_init);
 module_exit(powernow_k6_exit);
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 2ff0bbcd5bd1..9056f7e272c0 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -258,6 +258,7 @@ static void __cpuinit smp_callin(void)
         end_local_APIC_setup();
         map_cpu_to_logical_apicid();
 
+        notify_cpu_starting(cpuid);
         /*
          * Get our bogomips.
          *
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index ee0fba092157..199a5f4a873c 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -448,6 +448,8 @@ static void __init start_secondary(void *unused)
 
         VDEBUG(("VOYAGER SMP: CPU%d, stack at about %p\n", cpuid, &cpuid));
 
+        notify_cpu_starting(cpuid);
+
         /* enable interrupts */
         local_irq_enable();