summaryrefslogtreecommitdiffstats
path: root/arch/x86/kernel/cpu
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/kernel/cpu')
-rw-r--r--arch/x86/kernel/cpu/Makefile3
-rw-r--r--arch/x86/kernel/cpu/amd.c13
-rw-r--r--arch/x86/kernel/cpu/centaur.c1
-rw-r--r--arch/x86/kernel/cpu/common.c7
-rw-r--r--arch/x86/kernel/cpu/cyrix.c1
-rw-r--r--arch/x86/kernel/cpu/intel.c29
-rw-r--r--arch/x86/kernel/cpu/mcheck/mce_intel.c1
-rw-r--r--arch/x86/kernel/cpu/mcheck/p5.c1
-rw-r--r--arch/x86/kernel/cpu/mcheck/winchip.c1
-rw-r--r--arch/x86/kernel/cpu/microcode/Makefile7
-rw-r--r--arch/x86/kernel/cpu/microcode/amd.c492
-rw-r--r--arch/x86/kernel/cpu/microcode/amd_early.c380
-rw-r--r--arch/x86/kernel/cpu/microcode/core.c645
-rw-r--r--arch/x86/kernel/cpu/microcode/core_early.c141
-rw-r--r--arch/x86/kernel/cpu/microcode/intel.c333
-rw-r--r--arch/x86/kernel/cpu/microcode/intel_early.c787
-rw-r--r--arch/x86/kernel/cpu/microcode/intel_lib.c174
-rw-r--r--arch/x86/kernel/cpu/perf_event.c16
-rw-r--r--arch/x86/kernel/cpu/perf_event_amd_ibs.c53
-rw-r--r--arch/x86/kernel/cpu/perf_event_intel_rapl.c679
-rw-r--r--arch/x86/kernel/cpu/transmeta.c1
-rw-r--r--arch/x86/kernel/cpu/umc.c1
22 files changed, 3731 insertions, 35 deletions
diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
index 47b56a7e99cb..7fd54f09b011 100644
--- a/arch/x86/kernel/cpu/Makefile
+++ b/arch/x86/kernel/cpu/Makefile
@@ -36,12 +36,13 @@ obj-$(CONFIG_CPU_SUP_AMD) += perf_event_amd_iommu.o
endif
obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_p6.o perf_event_knc.o perf_event_p4.o
obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_intel_lbr.o perf_event_intel_ds.o perf_event_intel.o
-obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_intel_uncore.o
+obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_intel_uncore.o perf_event_intel_rapl.o
endif
obj-$(CONFIG_X86_MCE) += mcheck/
obj-$(CONFIG_MTRR) += mtrr/
+obj-$(CONFIG_MICROCODE) += microcode/
obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o perf_event_amd_ibs.o
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index bca023bdd6b2..525712c7ffa2 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -1,5 +1,4 @@
#include <linux/export.h>
-#include <linux/init.h>
#include <linux/bitops.h>
#include <linux/elf.h>
#include <linux/mm.h>
@@ -487,7 +486,7 @@ static void early_init_amd(struct cpuinfo_x86 *c)
set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC);
if (!check_tsc_unstable())
- sched_clock_stable = 1;
+ set_sched_clock_stable();
}
#ifdef CONFIG_X86_64
@@ -790,14 +789,10 @@ static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c)
}
/* Handle DTLB 2M and 4M sizes, fall back to L1 if L2 is disabled */
- if (!((eax >> 16) & mask)) {
- u32 a, b, c, d;
-
- cpuid(0x80000005, &a, &b, &c, &d);
- tlb_lld_2m[ENTRIES] = (a >> 16) & 0xff;
- } else {
+ if (!((eax >> 16) & mask))
+ tlb_lld_2m[ENTRIES] = (cpuid_eax(0x80000005) >> 16) & 0xff;
+ else
tlb_lld_2m[ENTRIES] = (eax >> 16) & mask;
- }
/* a 4M entry uses two 2M entries */
tlb_lld_4m[ENTRIES] = tlb_lld_2m[ENTRIES] >> 1;
diff --git a/arch/x86/kernel/cpu/centaur.c b/arch/x86/kernel/cpu/centaur.c
index 8d5652dc99dd..8779edab684e 100644
--- a/arch/x86/kernel/cpu/centaur.c
+++ b/arch/x86/kernel/cpu/centaur.c
@@ -1,6 +1,5 @@
#include <linux/bitops.h>
#include <linux/kernel.h>
-#include <linux/init.h>
#include <asm/processor.h>
#include <asm/e820.h>
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 6abc172b8258..24b6fd10625a 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -472,6 +472,7 @@ u16 __read_mostly tlb_lli_4m[NR_INFO];
u16 __read_mostly tlb_lld_4k[NR_INFO];
u16 __read_mostly tlb_lld_2m[NR_INFO];
u16 __read_mostly tlb_lld_4m[NR_INFO];
+u16 __read_mostly tlb_lld_1g[NR_INFO];
/*
* tlb_flushall_shift shows the balance point in replacing cr3 write
@@ -486,13 +487,13 @@ void cpu_detect_tlb(struct cpuinfo_x86 *c)
if (this_cpu->c_detect_tlb)
this_cpu->c_detect_tlb(c);
- printk(KERN_INFO "Last level iTLB entries: 4KB %d, 2MB %d, 4MB %d\n" \
- "Last level dTLB entries: 4KB %d, 2MB %d, 4MB %d\n" \
+ printk(KERN_INFO "Last level iTLB entries: 4KB %d, 2MB %d, 4MB %d\n"
+ "Last level dTLB entries: 4KB %d, 2MB %d, 4MB %d, 1GB %d\n"
"tlb_flushall_shift: %d\n",
tlb_lli_4k[ENTRIES], tlb_lli_2m[ENTRIES],
tlb_lli_4m[ENTRIES], tlb_lld_4k[ENTRIES],
tlb_lld_2m[ENTRIES], tlb_lld_4m[ENTRIES],
- tlb_flushall_shift);
+ tlb_lld_1g[ENTRIES], tlb_flushall_shift);
}
void detect_ht(struct cpuinfo_x86 *c)
diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c
index d0969c75ab54..aaf152e79637 100644
--- a/arch/x86/kernel/cpu/cyrix.c
+++ b/arch/x86/kernel/cpu/cyrix.c
@@ -1,4 +1,3 @@
-#include <linux/init.h>
#include <linux/bitops.h>
#include <linux/delay.h>
#include <linux/pci.h>
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index ea04b342c026..3db61c644e44 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -1,4 +1,3 @@
-#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/string.h>
@@ -93,7 +92,7 @@ static void early_init_intel(struct cpuinfo_x86 *c)
set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC);
if (!check_tsc_unstable())
- sched_clock_stable = 1;
+ set_sched_clock_stable();
}
/* Penwell and Cloverview have the TSC which doesn't sleep on S3 */
@@ -506,6 +505,7 @@ static unsigned int intel_size_cache(struct cpuinfo_x86 *c, unsigned int size)
#define TLB_DATA0_2M_4M 0x23
#define STLB_4K 0x41
+#define STLB_4K_2M 0x42
static const struct _tlb_table intel_tlb_table[] = {
{ 0x01, TLB_INST_4K, 32, " TLB_INST 4 KByte pages, 4-way set associative" },
@@ -526,13 +526,20 @@ static const struct _tlb_table intel_tlb_table[] = {
{ 0x5b, TLB_DATA_4K_4M, 64, " TLB_DATA 4 KByte and 4 MByte pages" },
{ 0x5c, TLB_DATA_4K_4M, 128, " TLB_DATA 4 KByte and 4 MByte pages" },
{ 0x5d, TLB_DATA_4K_4M, 256, " TLB_DATA 4 KByte and 4 MByte pages" },
+ { 0x61, TLB_INST_4K, 48, " TLB_INST 4 KByte pages, full associative" },
+ { 0x63, TLB_DATA_1G, 4, " TLB_DATA 1 GByte pages, 4-way set associative" },
+ { 0x76, TLB_INST_2M_4M, 8, " TLB_INST 2-MByte or 4-MByte pages, fully associative" },
{ 0xb0, TLB_INST_4K, 128, " TLB_INST 4 KByte pages, 4-way set associative" },
{ 0xb1, TLB_INST_2M_4M, 4, " TLB_INST 2M pages, 4-way, 8 entries or 4M pages, 4-way entries" },
{ 0xb2, TLB_INST_4K, 64, " TLB_INST 4KByte pages, 4-way set associative" },
{ 0xb3, TLB_DATA_4K, 128, " TLB_DATA 4 KByte pages, 4-way set associative" },
{ 0xb4, TLB_DATA_4K, 256, " TLB_DATA 4 KByte pages, 4-way associative" },
+ { 0xb5, TLB_INST_4K, 64, " TLB_INST 4 KByte pages, 8-way set ssociative" },
+ { 0xb6, TLB_INST_4K, 128, " TLB_INST 4 KByte pages, 8-way set ssociative" },
{ 0xba, TLB_DATA_4K, 64, " TLB_DATA 4 KByte pages, 4-way associative" },
{ 0xc0, TLB_DATA_4K_4M, 8, " TLB_DATA 4 KByte and 4 MByte pages, 4-way associative" },
+ { 0xc1, STLB_4K_2M, 1024, " STLB 4 KByte and 2 MByte pages, 8-way associative" },
+ { 0xc2, TLB_DATA_2M_4M, 16, " DTLB 2 MByte/4MByte pages, 4-way associative" },
{ 0xca, STLB_4K, 512, " STLB 4 KByte pages, 4-way associative" },
{ 0x00, 0, 0 }
};
@@ -558,6 +565,20 @@ static void intel_tlb_lookup(const unsigned char desc)
if (tlb_lld_4k[ENTRIES] < intel_tlb_table[k].entries)
tlb_lld_4k[ENTRIES] = intel_tlb_table[k].entries;
break;
+ case STLB_4K_2M:
+ if (tlb_lli_4k[ENTRIES] < intel_tlb_table[k].entries)
+ tlb_lli_4k[ENTRIES] = intel_tlb_table[k].entries;
+ if (tlb_lld_4k[ENTRIES] < intel_tlb_table[k].entries)
+ tlb_lld_4k[ENTRIES] = intel_tlb_table[k].entries;
+ if (tlb_lli_2m[ENTRIES] < intel_tlb_table[k].entries)
+ tlb_lli_2m[ENTRIES] = intel_tlb_table[k].entries;
+ if (tlb_lld_2m[ENTRIES] < intel_tlb_table[k].entries)
+ tlb_lld_2m[ENTRIES] = intel_tlb_table[k].entries;
+ if (tlb_lli_4m[ENTRIES] < intel_tlb_table[k].entries)
+ tlb_lli_4m[ENTRIES] = intel_tlb_table[k].entries;
+ if (tlb_lld_4m[ENTRIES] < intel_tlb_table[k].entries)
+ tlb_lld_4m[ENTRIES] = intel_tlb_table[k].entries;
+ break;
case TLB_INST_ALL:
if (tlb_lli_4k[ENTRIES] < intel_tlb_table[k].entries)
tlb_lli_4k[ENTRIES] = intel_tlb_table[k].entries;
@@ -603,6 +624,10 @@ static void intel_tlb_lookup(const unsigned char desc)
if (tlb_lld_4m[ENTRIES] < intel_tlb_table[k].entries)
tlb_lld_4m[ENTRIES] = intel_tlb_table[k].entries;
break;
+ case TLB_DATA_1G:
+ if (tlb_lld_1g[ENTRIES] < intel_tlb_table[k].entries)
+ tlb_lld_1g[ENTRIES] = intel_tlb_table[k].entries;
+ break;
}
}
diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c
index 4cfe0458ca66..fb6156fee6f7 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_intel.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c
@@ -6,7 +6,6 @@
*/
#include <linux/gfp.h>
-#include <linux/init.h>
#include <linux/interrupt.h>
#include <linux/percpu.h>
#include <linux/sched.h>
diff --git a/arch/x86/kernel/cpu/mcheck/p5.c b/arch/x86/kernel/cpu/mcheck/p5.c
index 1c044b1ccc59..a3042989398c 100644
--- a/arch/x86/kernel/cpu/mcheck/p5.c
+++ b/arch/x86/kernel/cpu/mcheck/p5.c
@@ -5,7 +5,6 @@
#include <linux/interrupt.h>
#include <linux/kernel.h>
#include <linux/types.h>
-#include <linux/init.h>
#include <linux/smp.h>
#include <asm/processor.h>
diff --git a/arch/x86/kernel/cpu/mcheck/winchip.c b/arch/x86/kernel/cpu/mcheck/winchip.c
index e9a701aecaa1..7dc5564d0cdf 100644
--- a/arch/x86/kernel/cpu/mcheck/winchip.c
+++ b/arch/x86/kernel/cpu/mcheck/winchip.c
@@ -5,7 +5,6 @@
#include <linux/interrupt.h>
#include <linux/kernel.h>
#include <linux/types.h>
-#include <linux/init.h>
#include <asm/processor.h>
#include <asm/mce.h>
diff --git a/arch/x86/kernel/cpu/microcode/Makefile b/arch/x86/kernel/cpu/microcode/Makefile
new file mode 100644
index 000000000000..285c85427c32
--- /dev/null
+++ b/arch/x86/kernel/cpu/microcode/Makefile
@@ -0,0 +1,7 @@
+microcode-y := core.o
+obj-$(CONFIG_MICROCODE) += microcode.o
+microcode-$(CONFIG_MICROCODE_INTEL) += intel.o intel_lib.o
+microcode-$(CONFIG_MICROCODE_AMD) += amd.o
+obj-$(CONFIG_MICROCODE_EARLY) += core_early.o
+obj-$(CONFIG_MICROCODE_INTEL_EARLY) += intel_early.o
+obj-$(CONFIG_MICROCODE_AMD_EARLY) += amd_early.o
diff --git a/arch/x86/kernel/cpu/microcode/amd.c b/arch/x86/kernel/cpu/microcode/amd.c
new file mode 100644
index 000000000000..4a6ff747aaad
--- /dev/null
+++ b/arch/x86/kernel/cpu/microcode/amd.c
@@ -0,0 +1,492 @@
+/*
+ * AMD CPU Microcode Update Driver for Linux
+ * Copyright (C) 2008-2011 Advanced Micro Devices Inc.
+ *
+ * Author: Peter Oruba <peter.oruba@amd.com>
+ *
+ * Based on work by:
+ * Tigran Aivazian <tigran@aivazian.fsnet.co.uk>
+ *
+ * Maintainers:
+ * Andreas Herrmann <herrmann.der.user@googlemail.com>
+ * Borislav Petkov <bp@alien8.de>
+ *
+ * This driver allows to upgrade microcode on F10h AMD
+ * CPUs and later.
+ *
+ * Licensed under the terms of the GNU General Public
+ * License version 2. See file COPYING for details.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/firmware.h>
+#include <linux/pci_ids.h>
+#include <linux/uaccess.h>
+#include <linux/vmalloc.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+
+#include <asm/microcode.h>
+#include <asm/processor.h>
+#include <asm/msr.h>
+#include <asm/microcode_amd.h>
+
+MODULE_DESCRIPTION("AMD Microcode Update Driver");
+MODULE_AUTHOR("Peter Oruba");
+MODULE_LICENSE("GPL v2");
+
+static struct equiv_cpu_entry *equiv_cpu_table;
+
+struct ucode_patch {
+ struct list_head plist;
+ void *data;
+ u32 patch_id;
+ u16 equiv_cpu;
+};
+
+static LIST_HEAD(pcache);
+
+static u16 __find_equiv_id(unsigned int cpu)
+{
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+ return find_equiv_id(equiv_cpu_table, uci->cpu_sig.sig);
+}
+
+static u32 find_cpu_family_by_equiv_cpu(u16 equiv_cpu)
+{
+ int i = 0;
+
+ BUG_ON(!equiv_cpu_table);
+
+ while (equiv_cpu_table[i].equiv_cpu != 0) {
+ if (equiv_cpu == equiv_cpu_table[i].equiv_cpu)
+ return equiv_cpu_table[i].installed_cpu;
+ i++;
+ }
+ return 0;
+}
+
+/*
+ * a small, trivial cache of per-family ucode patches
+ */
+static struct ucode_patch *cache_find_patch(u16 equiv_cpu)
+{
+ struct ucode_patch *p;
+
+ list_for_each_entry(p, &pcache, plist)
+ if (p->equiv_cpu == equiv_cpu)
+ return p;
+ return NULL;
+}
+
+static void update_cache(struct ucode_patch *new_patch)
+{
+ struct ucode_patch *p;
+
+ list_for_each_entry(p, &pcache, plist) {
+ if (p->equiv_cpu == new_patch->equiv_cpu) {
+ if (p->patch_id >= new_patch->patch_id)
+ /* we already have the latest patch */
+ return;
+
+ list_replace(&p->plist, &new_patch->plist);
+ kfree(p->data);
+ kfree(p);
+ return;
+ }
+ }
+ /* no patch found, add it */
+ list_add_tail(&new_patch->plist, &pcache);
+}
+
+static void free_cache(void)
+{
+ struct ucode_patch *p, *tmp;
+
+ list_for_each_entry_safe(p, tmp, &pcache, plist) {
+ __list_del(p->plist.prev, p->plist.next);
+ kfree(p->data);
+ kfree(p);
+ }
+}
+
+static struct ucode_patch *find_patch(unsigned int cpu)
+{
+ u16 equiv_id;
+
+ equiv_id = __find_equiv_id(cpu);
+ if (!equiv_id)
+ return NULL;
+
+ return cache_find_patch(equiv_id);
+}
+
+static int collect_cpu_info_amd(int cpu, struct cpu_signature *csig)
+{
+ struct cpuinfo_x86 *c = &cpu_data(cpu);
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+ struct ucode_patch *p;
+
+ csig->sig = cpuid_eax(0x00000001);
+ csig->rev = c->microcode;
+
+ /*
+ * a patch could have been loaded early, set uci->mc so that
+ * mc_bp_resume() can call apply_microcode()
+ */
+ p = find_patch(cpu);
+ if (p && (p->patch_id == csig->rev))
+ uci->mc = p->data;
+
+ pr_info("CPU%d: patch_level=0x%08x\n", cpu, csig->rev);
+
+ return 0;
+}
+
+static unsigned int verify_patch_size(u8 family, u32 patch_size,
+ unsigned int size)
+{
+ u32 max_size;
+
+#define F1XH_MPB_MAX_SIZE 2048
+#define F14H_MPB_MAX_SIZE 1824
+#define F15H_MPB_MAX_SIZE 4096
+#define F16H_MPB_MAX_SIZE 3458
+
+ switch (family) {
+ case 0x14:
+ max_size = F14H_MPB_MAX_SIZE;
+ break;
+ case 0x15:
+ max_size = F15H_MPB_MAX_SIZE;
+ break;
+ case 0x16:
+ max_size = F16H_MPB_MAX_SIZE;
+ break;
+ default:
+ max_size = F1XH_MPB_MAX_SIZE;
+ break;
+ }
+
+ if (patch_size > min_t(u32, size, max_size)) {
+ pr_err("patch size mismatch\n");
+ return 0;
+ }
+
+ return patch_size;
+}
+
+int __apply_microcode_amd(struct microcode_amd *mc_amd)
+{
+ u32 rev, dummy;
+
+ native_wrmsrl(MSR_AMD64_PATCH_LOADER, (u64)(long)&mc_amd->hdr.data_code);
+
+ /* verify patch application was successful */
+ native_rdmsr(MSR_AMD64_PATCH_LEVEL, rev, dummy);
+ if (rev != mc_amd->hdr.patch_id)
+ return -1;
+
+ return 0;
+}
+
+int apply_microcode_amd(int cpu)
+{
+ struct cpuinfo_x86 *c = &cpu_data(cpu);
+ struct microcode_amd *mc_amd;
+ struct ucode_cpu_info *uci;
+ struct ucode_patch *p;
+ u32 rev, dummy;
+
+ BUG_ON(raw_smp_processor_id() != cpu);
+
+ uci = ucode_cpu_info + cpu;
+
+ p = find_patch(cpu);
+ if (!p)
+ return 0;
+
+ mc_amd = p->data;
+ uci->mc = p->data;
+
+ rdmsr(MSR_AMD64_PATCH_LEVEL, rev, dummy);
+
+ /* need to apply patch? */
+ if (rev >= mc_amd->hdr.patch_id) {
+ c->microcode = rev;
+ uci->cpu_sig.rev = rev;
+ return 0;
+ }
+
+ if (__apply_microcode_amd(mc_amd)) {
+ pr_err("CPU%d: update failed for patch_level=0x%08x\n",
+ cpu, mc_amd->hdr.patch_id);
+ return -1;
+ }
+ pr_info("CPU%d: new patch_level=0x%08x\n", cpu,
+ mc_amd->hdr.patch_id);
+
+ uci->cpu_sig.rev = mc_amd->hdr.patch_id;
+ c->microcode = mc_amd->hdr.patch_id;
+
+ return 0;
+}
+
+static int install_equiv_cpu_table(const u8 *buf)
+{
+ unsigned int *ibuf = (unsigned int *)buf;
+ unsigned int type = ibuf[1];
+ unsigned int size = ibuf[2];
+
+ if (type != UCODE_EQUIV_CPU_TABLE_TYPE || !size) {
+ pr_err("empty section/"
+ "invalid type field in container file section header\n");
+ return -EINVAL;
+ }
+
+ equiv_cpu_table = vmalloc(size);
+ if (!equiv_cpu_table) {
+ pr_err("failed to allocate equivalent CPU table\n");
+ return -ENOMEM;
+ }
+
+ memcpy(equiv_cpu_table, buf + CONTAINER_HDR_SZ, size);
+
+ /* add header length */
+ return size + CONTAINER_HDR_SZ;
+}
+
+static void free_equiv_cpu_table(void)
+{
+ vfree(equiv_cpu_table);
+ equiv_cpu_table = NULL;
+}
+
+static void cleanup(void)
+{
+ free_equiv_cpu_table();
+ free_cache();
+}
+
+/*
+ * We return the current size even if some of the checks failed so that
+ * we can skip over the next patch. If we return a negative value, we
+ * signal a grave error like a memory allocation has failed and the
+ * driver cannot continue functioning normally. In such cases, we tear
+ * down everything we've used up so far and exit.
+ */
+static int verify_and_add_patch(u8 family, u8 *fw, unsigned int leftover)
+{
+ struct microcode_header_amd *mc_hdr;
+ struct ucode_patch *patch;
+ unsigned int patch_size, crnt_size, ret;
+ u32 proc_fam;
+ u16 proc_id;
+
+ patch_size = *(u32 *)(fw + 4);
+ crnt_size = patch_size + SECTION_HDR_SIZE;
+ mc_hdr = (struct microcode_header_amd *)(fw + SECTION_HDR_SIZE);
+ proc_id = mc_hdr->processor_rev_id;
+
+ proc_fam = find_cpu_family_by_equiv_cpu(proc_id);
+ if (!proc_fam) {
+ pr_err("No patch family for equiv ID: 0x%04x\n", proc_id);
+ return crnt_size;
+ }
+
+ /* check if patch is for the current family */
+ proc_fam = ((proc_fam >> 8) & 0xf) + ((proc_fam >> 20) & 0xff);
+ if (proc_fam != family)
+ return crnt_size;
+
+ if (mc_hdr->nb_dev_id || mc_hdr->sb_dev_id) {
+ pr_err("Patch-ID 0x%08x: chipset-specific code unsupported.\n",
+ mc_hdr->patch_id);
+ return crnt_size;
+ }
+
+ ret = verify_patch_size(family, patch_size, leftover);
+ if (!ret) {
+ pr_err("Patch-ID 0x%08x: size mismatch.\n", mc_hdr->patch_id);
+ return crnt_size;
+ }
+
+ patch = kzalloc(sizeof(*patch), GFP_KERNEL);
+ if (!patch) {
+ pr_err("Patch allocation failure.\n");
+ return -EINVAL;
+ }
+
+ patch->data = kzalloc(patch_size, GFP_KERNEL);
+ if (!patch->data) {
+ pr_err("Patch data allocation failure.\n");
+ kfree(patch);
+ return -EINVAL;
+ }
+
+ /* All looks ok, copy patch... */
+ memcpy(patch->data, fw + SECTION_HDR_SIZE, patch_size);
+ INIT_LIST_HEAD(&patch->plist);
+ patch->patch_id = mc_hdr->patch_id;
+ patch->equiv_cpu = proc_id;
+
+ pr_debug("%s: Added patch_id: 0x%08x, proc_id: 0x%04x\n",
+ __func__, patch->patch_id, proc_id);
+
+ /* ... and add to cache. */
+ update_cache(patch);
+
+ return crnt_size;
+}
+
+static enum ucode_state __load_microcode_amd(u8 family, const u8 *data,
+ size_t size)
+{
+ enum ucode_state ret = UCODE_ERROR;
+ unsigned int leftover;
+ u8 *fw = (u8 *)data;
+ int crnt_size = 0;
+ int offset;
+
+ offset = install_equiv_cpu_table(data);
+ if (offset < 0) {
+ pr_err("failed to create equivalent cpu table\n");
+ return ret;
+ }
+ fw += offset;
+ leftover = size - offset;
+
+ if (*(u32 *)fw != UCODE_UCODE_TYPE) {
+ pr_err("invalid type field in container file section header\n");
+ free_equiv_cpu_table();
+ return ret;
+ }
+
+ while (leftover) {
+ crnt_size = verify_and_add_patch(family, fw, leftover);
+ if (crnt_size < 0)
+ return ret;
+
+ fw += crnt_size;
+ leftover -= crnt_size;
+ }
+
+ return UCODE_OK;
+}
+
+enum ucode_state load_microcode_amd(u8 family, const u8 *data, size_t size)
+{
+ enum ucode_state ret;
+
+ /* free old equiv table */
+ free_equiv_cpu_table();
+
+ ret = __load_microcode_amd(family, data, size);
+
+ if (ret != UCODE_OK)
+ cleanup();
+
+#if defined(CONFIG_MICROCODE_AMD_EARLY) && defined(CONFIG_X86_32)
+ /* save BSP's matching patch for early load */
+ if (cpu_data(smp_processor_id()).cpu_index == boot_cpu_data.cpu_index) {
+ struct ucode_patch *p = find_patch(smp_processor_id());
+ if (p) {
+ memset(amd_ucode_patch, 0, PATCH_MAX_SIZE);
+ memcpy(amd_ucode_patch, p->data, min_t(u32, ksize(p->data),
+ PATCH_MAX_SIZE));
+ }
+ }
+#endif
+ return ret;
+}
+
+/*
+ * AMD microcode firmware naming convention, up to family 15h they are in
+ * the legacy file:
+ *
+ * amd-ucode/microcode_amd.bin
+ *
+ * This legacy file is always smaller than 2K in size.
+ *
+ * Beginning with family 15h, they are in family-specific firmware files:
+ *
+ * amd-ucode/microcode_amd_fam15h.bin
+ * amd-ucode/microcode_amd_fam16h.bin
+ * ...
+ *
+ * These might be larger than 2K.
+ */
+static enum ucode_state request_microcode_amd(int cpu, struct device *device,
+ bool refresh_fw)
+{
+ char fw_name[36] = "amd-ucode/microcode_amd.bin";
+ struct cpuinfo_x86 *c = &cpu_data(cpu);
+ enum ucode_state ret = UCODE_NFOUND;
+ const struct firmware *fw;
+
+ /* reload ucode container only on the boot cpu */
+ if (!refresh_fw || c->cpu_index != boot_cpu_data.cpu_index)
+ return UCODE_OK;
+
+ if (c->x86 >= 0x15)
+ snprintf(fw_name, sizeof(fw_name), "amd-ucode/microcode_amd_fam%.2xh.bin", c->x86);
+
+ if (request_firmware(&fw, (const char *)fw_name, device)) {
+ pr_debug("failed to load file %s\n", fw_name);
+ goto out;
+ }
+
+ ret = UCODE_ERROR;
+ if (*(u32 *)fw->data != UCODE_MAGIC) {
+ pr_err("invalid magic value (0x%08x)\n", *(u32 *)fw->data);
+ goto fw_release;
+ }
+
+ ret = load_microcode_amd(c->x86, fw->data, fw->size);
+
+ fw_release:
+ release_firmware(fw);
+
+ out:
+ return ret;
+}
+
+static enum ucode_state
+request_microcode_user(int cpu, const void __user *buf, size_t size)
+{
+ return UCODE_ERROR;
+}
+
+static void microcode_fini_cpu_amd(int cpu)
+{
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+ uci->mc = NULL;
+}
+
+static struct microcode_ops microcode_amd_ops = {
+ .request_microcode_user = request_microcode_user,
+ .request_microcode_fw = request_microcode_amd,
+ .collect_cpu_info = collect_cpu_info_amd,
+ .apply_microcode = apply_microcode_amd,
+ .microcode_fini_cpu = microcode_fini_cpu_amd,
+};
+
+struct microcode_ops * __init init_amd_microcode(void)
+{
+ struct cpuinfo_x86 *c = &cpu_data(0);
+
+ if (c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) {
+ pr_warning("AMD CPU family 0x%x not supported\n", c->x86);
+ return NULL;
+ }
+
+ return &microcode_amd_ops;
+}
+
+void __exit exit_amd_microcode(void)
+{
+ cleanup();
+}
diff --git a/arch/x86/kernel/cpu/microcode/amd_early.c b/arch/x86/kernel/cpu/microcode/amd_early.c
new file mode 100644
index 000000000000..8384c0fa206f
--- /dev/null
+++ b/arch/x86/kernel/cpu/microcode/amd_early.c
@@ -0,0 +1,380 @@
+/*
+ * Copyright (C) 2013 Advanced Micro Devices, Inc.
+ *
+ * Author: Jacob Shin <jacob.shin@amd.com>
+ * Fixes: Borislav Petkov <bp@suse.de>
+ *
+ * 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.
+ */
+
+#include <linux/earlycpio.h>
+#include <linux/initrd.h>
+
+#include <asm/cpu.h>
+#include <asm/setup.h>
+#include <asm/microcode_amd.h>
+
+/*
+ * This points to the current valid container of microcode patches which we will
+ * save from the initrd before jettisoning its contents.
+ */
+static u8 *container;
+static size_t container_size;
+
+static u32 ucode_new_rev;
+u8 amd_ucode_patch[PATCH_MAX_SIZE];
+static u16 this_equiv_id;
+
+struct cpio_data ucode_cpio;
+
+/*
+ * Microcode patch container file is prepended to the initrd in cpio format.
+ * See Documentation/x86/early-microcode.txt
+ */
+static __initdata char ucode_path[] = "kernel/x86/microcode/AuthenticAMD.bin";
+
+static struct cpio_data __init find_ucode_in_initrd(void)
+{
+ long offset = 0;
+ char *path;
+ void *start;
+ size_t size;
+
+#ifdef CONFIG_X86_32
+ struct boot_params *p;
+
+ /*
+ * On 32-bit, early load occurs before paging is turned on so we need
+ * to use physical addresses.
+ */
+ p = (struct boot_params *)__pa_nodebug(&boot_params);
+ path = (char *)__pa_nodebug(ucode_path);
+ start = (void *)p->hdr.ramdisk_image;
+ size = p->hdr.ramdisk_size;
+#else
+ path = ucode_path;
+ start = (void *)(boot_params.hdr.ramdisk_image + PAGE_OFFSET);
+ size = boot_params.hdr.ramdisk_size;
+#endif
+
+ return find_cpio_data(path, start, size, &offset);
+}
+
+static size_t compute_container_size(u8 *data, u32 total_size)
+{
+ size_t size = 0;
+ u32 *header = (u32 *)data;
+
+ if (header[0] != UCODE_MAGIC ||
+ header[1] != UCODE_EQUIV_CPU_TABLE_TYPE || /* type */
+ header[2] == 0) /* size */
+ return size;
+
+ size = header[2] + CONTAINER_HDR_SZ;
+ total_size -= size;
+ data += size;
+
+ while (total_size) {
+ u16 patch_size;
+
+ header = (u32 *)data;
+
+ if (header[0] != UCODE_UCODE_TYPE)
+ break;
+
+ /*
+ * Sanity-check patch size.
+ */
+ patch_size = header[1];
+ if (patch_size > PATCH_MAX_SIZE)
+ break;
+
+ size += patch_size + SECTION_HDR_SIZE;
+ data += patch_size + SECTION_HDR_SIZE;
+ total_size -= patch_size + SECTION_HDR_SIZE;
+ }
+
+ return size;
+}
+
+/*
+ * Early load occurs before we can vmalloc(). So we look for the microcode
+ * patch container file in initrd, traverse equivalent cpu table, look for a
+ * matching microcode patch, and update, all in initrd memory in place.
+ * When vmalloc() is available for use later -- on 64-bit during first AP load,
+ * and on 32-bit during save_microcode_in_initrd_amd() -- we can call
+ * load_microcode_amd() to save equivalent cpu table and microcode patches in
+ * kernel heap memory.
+ */
+static void apply_ucode_in_initrd(void *ucode, size_t size)
+{
+ struct equiv_cpu_entry *eq;
+ size_t *cont_sz;
+ u32 *header;
+ u8 *data, **cont;
+ u16 eq_id = 0;
+ int offset, left;
+ u32 rev, eax, ebx, ecx, edx;
+ u32 *new_rev;
+
+#ifdef CONFIG_X86_32
+ new_rev = (u32 *)__pa_nodebug(&ucode_new_rev);
+ cont_sz = (size_t *)__pa_nodebug(&container_size);
+ cont = (u8 **)__pa_nodebug(&container);
+#else
+ new_rev = &ucode_new_rev;
+ cont_sz = &container_size;
+ cont = &container;
+#endif
+
+ data = ucode;
+ left = size;
+ header = (u32 *)data;
+
+ /* find equiv cpu table */
+ if (header[0] != UCODE_MAGIC ||
+ header[1] != UCODE_EQUIV_CPU_TABLE_TYPE || /* type */
+ header[2] == 0) /* size */
+ return;
+
+ eax = 0x00000001;
+ ecx = 0;
+ native_cpuid(&eax, &ebx, &ecx, &edx);
+
+ while (left > 0) {
+ eq = (struct equiv_cpu_entry *)(data + CONTAINER_HDR_SZ);
+
+ *cont = data;
+
+ /* Advance past the container header */
+ offset = header[2] + CONTAINER_HDR_SZ;
+ data += offset;
+ left -= offset;
+
+ eq_id = find_equiv_id(eq, eax);
+ if (eq_id) {
+ this_equiv_id = eq_id;
+ *cont_sz = compute_container_size(*cont, left + offset);
+
+ /*
+ * truncate how much we need to iterate over in the
+ * ucode update loop below
+ */
+ left = *cont_sz - offset;
+ break;
+ }
+
+ /*
+ * support multiple container files appended together. if this
+ * one does not have a matching equivalent cpu entry, we fast
+ * forward to the next container file.
+ */
+ while (left > 0) {
+ header = (u32 *)data;
+ if (header[0] == UCODE_MAGIC &&
+ header[1] == UCODE_EQUIV_CPU_TABLE_TYPE)
+ break;
+
+ offset = header[1] + SECTION_HDR_SIZE;
+ data += offset;
+ left -= offset;
+ }
+
+ /* mark where the next microcode container file starts */
+ offset = data - (u8 *)ucode;
+ ucode = data;
+ }
+
+ if (!eq_id) {
+ *cont = NULL;
+ *cont_sz = 0;
+ return;
+ }
+
+ /* find ucode and update if needed */
+
+ native_rdmsr(MSR_AMD64_PATCH_LEVEL, rev, eax);
+
+ while (left > 0) {
+ struct microcode_amd *mc;
+
+ header = (u32 *)data;
+ if (header[0] != UCODE_UCODE_TYPE || /* type */
+ header[1] == 0) /* size */
+ break;
+
+ mc = (struct microcode_amd *)(data + SECTION_HDR_SIZE);
+
+ if (eq_id == mc->hdr.processor_rev_id && rev < mc->hdr.patch_id) {
+
+ if (!__apply_microcode_amd(mc)) {
+ rev = mc->hdr.patch_id;
+ *new_rev = rev;
+
+ /* save ucode patch */
+ memcpy(amd_ucode_patch, mc,
+ min_t(u32, header[1], PATCH_MAX_SIZE));
+ }
+ }
+
+ offset = header[1] + SECTION_HDR_SIZE;
+ data += offset;
+ left -= offset;
+ }
+}
+
+void __init load_ucode_amd_bsp(void)
+{
+ struct cpio_data cp;
+ void **data;
+ size_t *size;
+
+#ifdef CONFIG_X86_32
+ data = (void **)__pa_nodebug(&ucode_cpio.data);
+ size = (size_t *)__pa_nodebug(&ucode_cpio.size);
+#else
+ data = &ucode_cpio.data;
+ size = &ucode_cpio.size;
+#endif
+
+ cp = find_ucode_in_initrd();
+ if (!cp.data)
+ return;
+
+ *data = cp.data;
+ *size = cp.size;
+
+ apply_ucode_in_initrd(cp.data, cp.size);
+}
+
+#ifdef CONFIG_X86_32
+/*
+ * On 32-bit, since AP's early load occurs before paging is turned on, we
+ * cannot traverse cpu_equiv_table and pcache in kernel heap memory. So during
+ * cold boot, AP will apply_ucode_in_initrd() just like the BSP. During
+ * save_microcode_in_initrd_amd() BSP's patch is copied to amd_ucode_patch,
+ * which is used upon resume from suspend.
+ */
+void load_ucode_amd_ap(void)
+{
+ struct microcode_amd *mc;
+ size_t *usize;
+ void **ucode;
+
+ mc = (struct microcode_amd *)__pa(amd_ucode_patch);
+ if (mc->hdr.patch_id && mc->hdr.processor_rev_id) {
+ __apply_microcode_amd(mc);
+ return;
+ }
+
+ ucode = (void *)__pa_nodebug(&container);
+ usize = (size_t *)__pa_nodebug(&container_size);
+
+ if (!*ucode || !*usize)
+ return;
+
+ apply_ucode_in_initrd(*ucode, *usize);
+}
+
+static void __init collect_cpu_sig_on_bsp(void *arg)
+{
+ unsigned int cpu = smp_processor_id();
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+ uci->cpu_sig.sig = cpuid_eax(0x00000001);
+}
+#else
+void load_ucode_amd_ap(void)
+{
+ unsigned int cpu = smp_processor_id();
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+ struct equiv_cpu_entry *eq;
+ struct microcode_amd *mc;
+ u32 rev, eax;
+ u16 eq_id;
+
+ /* Exit if called on the BSP. */
+ if (!cpu)
+ return;
+
+ if (!container)
+ return;
+
+ rdmsr(MSR_AMD64_PATCH_LEVEL, rev, eax);
+
+ uci->cpu_sig.rev = rev;
+ uci->cpu_sig.sig = eax;
+
+ eax = cpuid_eax(0x00000001);
+ eq = (struct equiv_cpu_entry *)(container + CONTAINER_HDR_SZ);
+
+ eq_id = find_equiv_id(eq, eax);
+ if (!eq_id)
+ return;
+
+ if (eq_id == this_equiv_id) {
+ mc = (struct microcode_amd *)amd_ucode_patch;
+
+ if (mc && rev < mc->hdr.patch_id) {
+ if (!__apply_microcode_amd(mc))
+ ucode_new_rev = mc->hdr.patch_id;
+ }
+
+ } else {
+ if (!ucode_cpio.data)
+ return;
+
+ /*
+ * AP has a different equivalence ID than BSP, looks like
+ * mixed-steppings silicon so go through the ucode blob anew.
+ */
+ apply_ucode_in_initrd(ucode_cpio.data, ucode_cpio.size);
+ }
+}
+#endif
+
+int __init save_microcode_in_initrd_amd(void)
+{
+ enum ucode_state ret;
+ u32 eax;
+
+#ifdef CONFIG_X86_32
+ unsigned int bsp = boot_cpu_data.cpu_index;
+ struct ucode_cpu_info *uci = ucode_cpu_info + bsp;
+
+ if (!uci->cpu_sig.sig)
+ smp_call_function_single(bsp, collect_cpu_sig_on_bsp, NULL, 1);
+
+ /*
+ * Take into account the fact that the ramdisk might get relocated
+ * and therefore we need to recompute the container's position in
+ * virtual memory space.
+ */
+ container = (u8 *)(__va((u32)relocated_ramdisk) +
+ ((u32)container - boot_params.hdr.ramdisk_image));
+#endif
+ if (ucode_new_rev)
+ pr_info("microcode: updated early to new patch_level=0x%08x\n",
+ ucode_new_rev);
+
+ if (!container)
+ return -EINVAL;
+
+ eax = cpuid_eax(0x00000001);
+ eax = ((eax >> 8) & 0xf) + ((eax >> 20) & 0xff);
+
+ ret = load_microcode_amd(eax, container, container_size);
+ if (ret != UCODE_OK)
+ return -EINVAL;
+
+ /*
+ * This will be freed any msec now, stash patches for the current
+ * family and switch to patch cache for cpu hotplug, etc later.
+ */
+ container = NULL;
+ container_size = 0;
+
+ return 0;
+}
diff --git a/arch/x86/kernel/cpu/microcode/core.c b/arch/x86/kernel/cpu/microcode/core.c
new file mode 100644
index 000000000000..15c987698b0f
--- /dev/null
+++ b/arch/x86/kernel/cpu/microcode/core.c
@@ -0,0 +1,645 @@
+/*
+ * Intel CPU Microcode Update Driver for Linux
+ *
+ * Copyright (C) 2000-2006 Tigran Aivazian <tigran@aivazian.fsnet.co.uk>
+ * 2006 Shaohua Li <shaohua.li@intel.com>
+ *
+ * This driver allows to upgrade microcode on Intel processors
+ * belonging to IA-32 family - PentiumPro, Pentium II,
+ * Pentium III, Xeon, Pentium 4, etc.
+ *
+ * Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture
+ * Software Developer's Manual
+ * Order Number 253668 or free download from:
+ *
+ * http://developer.intel.com/Assets/PDF/manual/253668.pdf
+ *
+ * For more information, go to http://www.urbanmyth.org/microcode
+ *
+ * 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.
+ *
+ * 1.0 16 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ * Initial release.
+ * 1.01 18 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ * Added read() support + cleanups.
+ * 1.02 21 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ * Added 'device trimming' support. open(O_WRONLY) zeroes
+ * and frees the saved copy of applied microcode.
+ * 1.03 29 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ * Made to use devfs (/dev/cpu/microcode) + cleanups.
+ * 1.04 06 Jun 2000, Simon Trimmer <simon@veritas.com>
+ * Added misc device support (now uses both devfs and misc).
+ * Added MICROCODE_IOCFREE ioctl to clear memory.
+ * 1.05 09 Jun 2000, Simon Trimmer <simon@veritas.com>
+ * Messages for error cases (non Intel & no suitable microcode).
+ * 1.06 03 Aug 2000, Tigran Aivazian <tigran@veritas.com>
+ * Removed ->release(). Removed exclusive open and status bitmap.
+ * Added microcode_rwsem to serialize read()/write()/ioctl().
+ * Removed global kernel lock usage.
+ * 1.07 07 Sep 2000, Tigran Aivazian <tigran@veritas.com>
+ * Write 0 to 0x8B msr and then cpuid before reading revision,
+ * so that it works even if there were no update done by the
+ * BIOS. Otherwise, reading from 0x8B gives junk (which happened
+ * to be 0 on my machine which is why it worked even when I
+ * disabled update by the BIOS)
+ * Thanks to Eric W. Biederman <ebiederman@lnxi.com> for the fix.
+ * 1.08 11 Dec 2000, Richard Schaal <richard.schaal@intel.com> and
+ * Tigran Aivazian <tigran@veritas.com>
+ * Intel Pentium 4 processor support and bugfixes.
+ * 1.09 30 Oct 2001, Tigran Aivazian <tigran@veritas.com>
+ * Bugfix for HT (Hyper-Threading) enabled processors
+ * whereby processor resources are shared by all logical processors
+ * in a single CPU package.
+ * 1.10 28 Feb 2002 Asit K Mallick <asit.k.mallick@intel.com> and
+ * Tigran Aivazian <tigran@veritas.com>,
+ * Serialize updates as required on HT processors due to
+ * speculative nature of implementation.
+ * 1.11 22 Mar 2002 Tigran Aivazian <tigran@veritas.com>
+ * Fix the panic when writing zero-length microcode chunk.
+ * 1.12 29 Sep 2003 Nitin Kamble <nitin.a.kamble@intel.com>,
+ * Jun Nakajima <jun.nakajima@intel.com>
+ * Support for the microcode updates in the new format.
+ * 1.13 10 Oct 2003 Tigran Aivazian <tigran@veritas.com>
+ * Removed ->read() method and obsoleted MICROCODE_IOCFREE ioctl
+ * because we no longer hold a copy of applied microcode
+ * in kernel memory.
+ * 1.14 25 Jun 2004 Tigran Aivazian <tigran@veritas.com>
+ * Fix sigmatch() macro to handle old CPUs with pf == 0.
+ * Thanks to Stuart Swales for pointing out this bug.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/platform_device.h>
+#include <linux/miscdevice.h>
+#include <linux/capability.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/cpu.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include <linux/syscore_ops.h>
+
+#include <asm/microcode.h>
+#include <asm/processor.h>
+#include <asm/cpu_device_id.h>
+#include <asm/perf_event.h>
+
+MODULE_DESCRIPTION("Microcode Update Driver");
+MODULE_AUTHOR("Tigran Aivazian <tigran@aivazian.fsnet.co.uk>");
+MODULE_LICENSE("GPL");
+
+#define MICROCODE_VERSION "2.00"
+
+static struct microcode_ops *microcode_ops;
+
+/*
+ * Synchronization.
+ *
+ * All non cpu-hotplug-callback call sites use:
+ *
+ * - microcode_mutex to synchronize with each other;
+ * - get/put_online_cpus() to synchronize with
+ * the cpu-hotplug-callback call sites.
+ *
+ * We guarantee that only a single cpu is being
+ * updated at any particular moment of time.
+ */
+static DEFINE_MUTEX(microcode_mutex);
+
+struct ucode_cpu_info ucode_cpu_info[NR_CPUS];
+EXPORT_SYMBOL_GPL(ucode_cpu_info);
+
+/*
+ * Operations that are run on a target cpu:
+ */
+
+struct cpu_info_ctx {
+ struct cpu_signature *cpu_sig;
+ int err;
+};
+
+static void collect_cpu_info_local(void *arg)
+{
+ struct cpu_info_ctx *ctx = arg;
+
+ ctx->err = microcode_ops->collect_cpu_info(smp_processor_id(),
+ ctx->cpu_sig);
+}
+
+static int collect_cpu_info_on_target(int cpu, struct cpu_signature *cpu_sig)
+{
+ struct cpu_info_ctx ctx = { .cpu_sig = cpu_sig, .err = 0 };
+ int ret;
+
+ ret = smp_call_function_single(cpu, collect_cpu_info_local, &ctx, 1);
+ if (!ret)
+ ret = ctx.err;
+
+ return ret;
+}
+
+static int collect_cpu_info(int cpu)
+{
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+ int ret;
+
+ memset(uci, 0, sizeof(*uci));
+
+ ret = collect_cpu_info_on_target(cpu, &uci->cpu_sig);
+ if (!ret)
+ uci->valid = 1;
+
+ return ret;
+}
+
+struct apply_microcode_ctx {
+ int err;
+};
+
+static void apply_microcode_local(void *arg)
+{
+ struct apply_microcode_ctx *ctx = arg;
+
+ ctx->err = microcode_ops->apply_microcode(smp_processor_id());
+}
+
+static int apply_microcode_on_target(int cpu)
+{
+ struct apply_microcode_ctx ctx = { .err = 0 };
+ int ret;
+
+ ret = smp_call_function_single(cpu, apply_microcode_local, &ctx, 1);
+ if (!ret)
+ ret = ctx.err;
+
+ return ret;
+}
+
+#ifdef CONFIG_MICROCODE_OLD_INTERFACE
+static int do_microcode_update(const void __user *buf, size_t size)
+{
+ int error = 0;
+ int cpu;
+
+ for_each_online_cpu(cpu) {
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+ enum ucode_state ustate;
+
+ if (!uci->valid)
+ continue;
+
+ ustate = microcode_ops->request_microcode_user(cpu, buf, size);
+ if (ustate == UCODE_ERROR) {
+ error = -1;
+ break;
+ } else if (ustate == UCODE_OK)
+ apply_microcode_on_target(cpu);
+ }
+
+ return error;
+}
+
+static int microcode_open(struct inode *inode, struct file *file)
+{
+ return capable(CAP_SYS_RAWIO) ? nonseekable_open(inode, file) : -EPERM;
+}
+
+static ssize_t microcode_write(struct file *file, const char __user *buf,
+ size_t len, loff_t *ppos)
+{
+ ssize_t ret = -EINVAL;
+
+ if ((len >> PAGE_SHIFT) > totalram_pages) {
+ pr_err("too much data (max %ld pages)\n", totalram_pages);
+ return ret;
+ }
+
+ get_online_cpus();
+ mutex_lock(&microcode_mutex);
+
+ if (do_microcode_update(buf, len) == 0)
+ ret = (ssize_t)len;
+
+ if (ret > 0)
+ perf_check_microcode();
+
+ mutex_unlock(&microcode_mutex);
+ put_online_cpus();
+
+ return ret;
+}
+
+static const struct file_operations microcode_fops = {
+ .owner = THIS_MODULE,
+ .write = microcode_write,
+ .open = microcode_open,
+ .llseek = no_llseek,
+};
+
+static struct miscdevice microcode_dev = {
+ .minor = MICROCODE_MINOR,
+ .name = "microcode",
+ .nodename = "cpu/microcode",
+ .fops = &microcode_fops,
+};
+
+static int __init microcode_dev_init(void)
+{
+ int error;
+
+ error = misc_register(&microcode_dev);
+ if (error) {
+ pr_err("can't misc_register on minor=%d\n", MICROCODE_MINOR);
+ return error;
+ }
+
+ return 0;
+}
+
+static void __exit microcode_dev_exit(void)
+{
+ misc_deregister(&microcode_dev);
+}
+
+MODULE_ALIAS_MISCDEV(MICROCODE_MINOR);
+MODULE_ALIAS("devname:cpu/microcode");
+#else
+#define microcode_dev_init() 0
+#define microcode_dev_exit() do { } while (0)
+#endif
+
+/* fake device for request_firmware */
+static struct platform_device *microcode_pdev;
+
+static int reload_for_cpu(int cpu)
+{
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+ enum ucode_state ustate;
+ int err = 0;
+
+ if (!uci->valid)
+ return err;
+
+ ustate = microcode_ops->request_microcode_fw(cpu, &microcode_pdev->dev, true);
+ if (ustate == UCODE_OK)
+ apply_microcode_on_target(cpu);
+ else
+ if (ustate == UCODE_ERROR)
+ err = -EINVAL;
+ return err;
+}
+
+static ssize_t reload_store(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t size)
+{
+ unsigned long val;
+ int cpu;
+ ssize_t ret = 0, tmp_ret;
+
+ ret = kstrtoul(buf, 0, &val);
+ if (ret)
+ return ret;
+
+ if (val != 1)
+ return size;
+
+ get_online_cpus();
+ mutex_lock(&microcode_mutex);
+ for_each_online_cpu(cpu) {
+ tmp_ret = reload_for_cpu(cpu);
+ if (tmp_ret != 0)
+ pr_warn("Error reloading microcode on CPU %d\n", cpu);
+
+ /* save retval of the first encountered reload error */
+ if (!ret)
+ ret = tmp_ret;
+ }
+ if (!ret)
+ perf_check_microcode();
+ mutex_unlock(&microcode_mutex);
+ put_online_cpus();
+
+ if (!ret)
+ ret = size;
+
+ return ret;
+}
+
+static ssize_t version_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
+
+ return sprintf(buf, "0x%x\n", uci->cpu_sig.rev);
+}
+
+static ssize_t pf_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
+
+ return sprintf(buf, "0x%x\n", uci->cpu_sig.pf);
+}
+
+static DEVICE_ATTR(reload, 0200, NULL, reload_store);
+static DEVICE_ATTR(version, 0400, version_show, NULL);
+static DEVICE_ATTR(processor_flags, 0400, pf_show, NULL);
+
+static struct attribute *mc_default_attrs[] = {
+ &dev_attr_version.attr,
+ &dev_attr_processor_flags.attr,
+ NULL
+};
+
+static struct attribute_group mc_attr_group = {
+ .attrs = mc_default_attrs,
+ .name = "microcode",
+};
+
+static void microcode_fini_cpu(int cpu)
+{
+ microcode_ops->microcode_fini_cpu(cpu);
+}
+
+static enum ucode_state microcode_resume_cpu(int cpu)
+{
+ pr_debug("CPU%d updated upon resume\n", cpu);
+
+ if (apply_microcode_on_target(cpu))
+ return UCODE_ERROR;
+
+ return UCODE_OK;
+}
+
+static enum ucode_state microcode_init_cpu(int cpu, bool refresh_fw)
+{
+ enum ucode_state ustate;
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+ if (uci && uci->valid)
+ return UCODE_OK;
+
+ if (collect_cpu_info(cpu))
+ return UCODE_ERROR;
+
+ /* --dimm. Trigger a delayed update? */
+ if (system_state != SYSTEM_RUNNING)
+ return UCODE_NFOUND;
+
+ ustate = microcode_ops->request_microcode_fw(cpu, &microcode_pdev->dev,
+ refresh_fw);
+
+ if (ustate == UCODE_OK) {
+ pr_debug("CPU%d updated upon init\n", cpu);
+ apply_microcode_on_target(cpu);
+ }
+
+ return ustate;
+}
+
+static enum ucode_state microcode_update_cpu(int cpu)
+{
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+ if (uci->valid)
+ return microcode_resume_cpu(cpu);
+
+ return microcode_init_cpu(cpu, false);
+}
+
+static int mc_device_add(struct device *dev, struct subsys_interface *sif)
+{
+ int err, cpu = dev->id;
+
+ if (!cpu_online(cpu))
+ return 0;
+
+ pr_debug("CPU%d added\n", cpu);
+
+ err = sysfs_create_group(&dev->kobj, &mc_attr_group);
+ if (err)
+ return err;
+
+ if (microcode_init_cpu(cpu, true) == UCODE_ERROR)
+ return -EINVAL;
+
+ return err;
+}
+
+static int mc_device_remove(struct device *dev, struct subsys_interface *sif)
+{
+ int cpu = dev->id;
+
+ if (!cpu_online(cpu))
+ return 0;
+
+ pr_debug("CPU%d removed\n", cpu);
+ microcode_fini_cpu(cpu);
+ sysfs_remove_group(&dev->kobj, &mc_attr_group);
+ return 0;
+}
+
+static struct subsys_interface mc_cpu_interface = {
+ .name = "microcode",
+ .subsys = &cpu_subsys,
+ .add_dev = mc_device_add,
+ .remove_dev = mc_device_remove,
+};
+
+/**
+ * mc_bp_resume - Update boot CPU microcode during resume.
+ */
+static void mc_bp_resume(void)
+{
+ int cpu = smp_processor_id();
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+ if (uci->valid && uci->mc)
+ microcode_ops->apply_microcode(cpu);
+}
+
+static struct syscore_ops mc_syscore_ops = {
+ .resume = mc_bp_resume,
+};
+
+static int
+mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu)
+{
+ unsigned int cpu = (unsigned long)hcpu;
+ struct device *dev;
+
+ dev = get_cpu_device(cpu);
+
+ switch (action & ~CPU_TASKS_FROZEN) {
+ case CPU_ONLINE:
+ microcode_update_cpu(cpu);
+ pr_debug("CPU%d added\n", cpu);
+ /*
+ * "break" is missing on purpose here because we want to fall
+ * through in order to create the sysfs group.
+ */
+
+ case CPU_DOWN_FAILED:
+ if (sysfs_create_group(&dev->kobj, &mc_attr_group))
+ pr_err("Failed to create group for CPU%d\n", cpu);
+ break;
+
+ case CPU_DOWN_PREPARE:
+ /* Suspend is in progress, only remove the interface */
+ sysfs_remove_group(&dev->kobj, &mc_attr_group);
+ pr_debug("CPU%d removed\n", cpu);
+ break;
+
+ /*
+ * case CPU_DEAD:
+ *
+ * When a CPU goes offline, don't free up or invalidate the copy of
+ * the microcode in kernel memory, so that we can reuse it when the
+ * CPU comes back online without unnecessarily requesting the userspace
+ * for it again.
+ */
+ }
+
+ /* The CPU refused to come up during a system resume */
+ if (action == CPU_UP_CANCELED_FROZEN)
+ microcode_fini_cpu(cpu);
+
+ return NOTIFY_OK;
+}
+
+static struct notifier_block __refdata mc_cpu_notifier = {
+ .notifier_call = mc_cpu_callback,
+};
+
+#ifdef MODULE
+/* Autoload on Intel and AMD systems */
+static const struct x86_cpu_id __initconst microcode_id[] = {
+#ifdef CONFIG_MICROCODE_INTEL
+ { X86_VENDOR_INTEL, X86_FAMILY_ANY, X86_MODEL_ANY, },
+#endif
+#ifdef CONFIG_MICROCODE_AMD
+ { X86_VENDOR_AMD, X86_FAMILY_ANY, X86_MODEL_ANY, },
+#endif
+ {}
+};
+MODULE_DEVICE_TABLE(x86cpu, microcode_id);
+#endif
+
+static struct attribute *cpu_root_microcode_attrs[] = {
+ &dev_attr_reload.attr,
+ NULL
+};
+
+static struct attribute_group cpu_root_microcode_group = {
+ .name = "microcode",
+ .attrs = cpu_root_microcode_attrs,
+};
+
+static int __init microcode_init(void)
+{
+ struct cpuinfo_x86 *c = &cpu_data(0);
+ int error;
+
+ if (c->x86_vendor == X86_VENDOR_INTEL)
+ microcode_ops = init_intel_microcode();
+ else if (c->x86_vendor == X86_VENDOR_AMD)
+ microcode_ops = init_amd_microcode();
+ else
+ pr_err("no support for this CPU vendor\n");
+
+ if (!microcode_ops)
+ return -ENODEV;
+
+ microcode_pdev = platform_device_register_simple("microcode", -1,
+ NULL, 0);
+ if (IS_ERR(microcode_pdev))
+ return PTR_ERR(microcode_pdev);
+
+ get_online_cpus();
+ mutex_lock(&microcode_mutex);
+
+ error = subsys_interface_register(&mc_cpu_interface);
+ if (!error)
+ perf_check_microcode();
+ mutex_unlock(&microcode_mutex);
+ put_online_cpus();
+
+ if (error)
+ goto out_pdev;
+
+ error = sysfs_create_group(&cpu_subsys.dev_root->kobj,
+ &cpu_root_microcode_group);
+
+ if (error) {
+ pr_err("Error creating microcode group!\n");
+ goto out_driver;
+ }
+
+ error = microcode_dev_init();
+ if (error)
+ goto out_ucode_group;
+
+ register_syscore_ops(&mc_syscore_ops);
+ register_hotcpu_notifier(&mc_cpu_notifier);
+
+ pr_info("Microcode Update Driver: v" MICROCODE_VERSION
+ " <tigran@aivazian.fsnet.co.uk>, Peter Oruba\n");
+
+ return 0;
+
+ out_ucode_group:
+ sysfs_remove_group(&cpu_subsys.dev_root->kobj,
+ &cpu_root_microcode_group);
+
+ out_driver:
+ get_online_cpus();
+ mutex_lock(&microcode_mutex);
+
+ subsys_interface_unregister(&mc_cpu_interface);
+
+ mutex_unlock(&microcode_mutex);
+ put_online_cpus();
+
+ out_pdev:
+ platform_device_unregister(microcode_pdev);
+ return error;
+
+}
+module_init(microcode_init);
+
+static void __exit microcode_exit(void)
+{
+ struct cpuinfo_x86 *c = &cpu_data(0);
+
+ microcode_dev_exit();
+
+ unregister_hotcpu_notifier(&mc_cpu_notifier);
+ unregister_syscore_ops(&mc_syscore_ops);
+
+ sysfs_remove_group(&cpu_subsys.dev_root->kobj,
+ &cpu_root_microcode_group);
+
+ get_online_cpus();
+ mutex_lock(&microcode_mutex);
+
+ subsys_interface_unregister(&mc_cpu_interface);
+
+ mutex_unlock(&microcode_mutex);
+ put_online_cpus();
+
+ platform_device_unregister(microcode_pdev);
+
+ microcode_ops = NULL;
+
+ if (c->x86_vendor == X86_VENDOR_AMD)
+ exit_amd_microcode();
+
+ pr_info("Microcode Update Driver: v" MICROCODE_VERSION " removed.\n");
+}
+module_exit(microcode_exit);
diff --git a/arch/x86/kernel/cpu/microcode/core_early.c b/arch/x86/kernel/cpu/microcode/core_early.c
new file mode 100644
index 000000000000..be7f8514f577
--- /dev/null
+++ b/arch/x86/kernel/cpu/microcode/core_early.c
@@ -0,0 +1,141 @@
+/*
+ * X86 CPU microcode early update for Linux
+ *
+ * Copyright (C) 2012 Fenghua Yu <fenghua.yu@intel.com>
+ * H Peter Anvin" <hpa@zytor.com>
+ *
+ * This driver allows to early upgrade microcode on Intel processors
+ * belonging to IA-32 family - PentiumPro, Pentium II,
+ * Pentium III, Xeon, Pentium 4, etc.
+ *
+ * Reference: Section 9.11 of Volume 3, IA-32 Intel Architecture
+ * Software Developer's Manual.
+ *
+ * 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/module.h>
+#include <asm/microcode_intel.h>
+#include <asm/microcode_amd.h>
+#include <asm/processor.h>
+
+#define QCHAR(a, b, c, d) ((a) + ((b) << 8) + ((c) << 16) + ((d) << 24))
+#define CPUID_INTEL1 QCHAR('G', 'e', 'n', 'u')
+#define CPUID_INTEL2 QCHAR('i', 'n', 'e', 'I')
+#define CPUID_INTEL3 QCHAR('n', 't', 'e', 'l')
+#define CPUID_AMD1 QCHAR('A', 'u', 't', 'h')
+#define CPUID_AMD2 QCHAR('e', 'n', 't', 'i')
+#define CPUID_AMD3 QCHAR('c', 'A', 'M', 'D')
+
+#define CPUID_IS(a, b, c, ebx, ecx, edx) \
+ (!((ebx ^ (a))|(edx ^ (b))|(ecx ^ (c))))
+
+/*
+ * In early loading microcode phase on BSP, boot_cpu_data is not set up yet.
+ * x86_vendor() gets vendor id for BSP.
+ *
+ * In 32 bit AP case, accessing boot_cpu_data needs linear address. To simplify
+ * coding, we still use x86_vendor() to get vendor id for AP.
+ *
+ * x86_vendor() gets vendor information directly through cpuid.
+ */
+static int x86_vendor(void)
+{
+ u32 eax = 0x00000000;
+ u32 ebx, ecx = 0, edx;
+
+ native_cpuid(&eax, &ebx, &ecx, &edx);
+
+ if (CPUID_IS(CPUID_INTEL1, CPUID_INTEL2, CPUID_INTEL3, ebx, ecx, edx))
+ return X86_VENDOR_INTEL;
+
+ if (CPUID_IS(CPUID_AMD1, CPUID_AMD2, CPUID_AMD3, ebx, ecx, edx))
+ return X86_VENDOR_AMD;
+
+ return X86_VENDOR_UNKNOWN;
+}
+
+static int x86_family(void)
+{
+ u32 eax = 0x00000001;
+ u32 ebx, ecx = 0, edx;
+ int x86;
+
+ native_cpuid(&eax, &ebx, &ecx, &edx);
+
+ x86 = (eax >> 8) & 0xf;
+ if (x86 == 15)
+ x86 += (eax >> 20) & 0xff;
+
+ return x86;
+}
+
+void __init load_ucode_bsp(void)
+{
+ int vendor, x86;
+
+ if (!have_cpuid_p())
+ return;
+
+ vendor = x86_vendor();
+ x86 = x86_family();
+
+ switch (vendor) {
+ case X86_VENDOR_INTEL:
+ if (x86 >= 6)
+ load_ucode_intel_bsp();
+ break;
+ case X86_VENDOR_AMD:
+ if (x86 >= 0x10)
+ load_ucode_amd_bsp();
+ break;
+ default:
+ break;
+ }
+}
+
+void load_ucode_ap(void)
+{
+ int vendor, x86;
+
+ if (!have_cpuid_p())
+ return;
+
+ vendor = x86_vendor();
+ x86 = x86_family();
+
+ switch (vendor) {
+ case X86_VENDOR_INTEL:
+ if (x86 >= 6)
+ load_ucode_intel_ap();
+ break;
+ case X86_VENDOR_AMD:
+ if (x86 >= 0x10)
+ load_ucode_amd_ap();
+ break;
+ default:
+ break;
+ }
+}
+
+int __init save_microcode_in_initrd(void)
+{
+ struct cpuinfo_x86 *c = &boot_cpu_data;
+
+ switch (c->x86_vendor) {
+ case X86_VENDOR_INTEL:
+ if (c->x86 >= 6)
+ save_microcode_in_initrd_intel();
+ break;
+ case X86_VENDOR_AMD:
+ if (c->x86 >= 0x10)
+ save_microcode_in_initrd_amd();
+ break;
+ default:
+ break;
+ }
+
+ return 0;
+}
diff --git a/arch/x86/kernel/cpu/microcode/intel.c b/arch/x86/kernel/cpu/microcode/intel.c
new file mode 100644
index 000000000000..5fb2cebf556b
--- /dev/null
+++ b/arch/x86/kernel/cpu/microcode/intel.c
@@ -0,0 +1,333 @@
+/*
+ * Intel CPU Microcode Update Driver for Linux
+ *
+ * Copyright (C) 2000-2006 Tigran Aivazian <tigran@aivazian.fsnet.co.uk>
+ * 2006 Shaohua Li <shaohua.li@intel.com>
+ *
+ * This driver allows to upgrade microcode on Intel processors
+ * belonging to IA-32 family - PentiumPro, Pentium II,
+ * Pentium III, Xeon, Pentium 4, etc.
+ *
+ * Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture
+ * Software Developer's Manual
+ * Order Number 253668 or free download from:
+ *
+ * http://developer.intel.com/Assets/PDF/manual/253668.pdf
+ *
+ * For more information, go to http://www.urbanmyth.org/microcode
+ *
+ * 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.
+ *
+ * 1.0 16 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ * Initial release.
+ * 1.01 18 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ * Added read() support + cleanups.
+ * 1.02 21 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ * Added 'device trimming' support. open(O_WRONLY) zeroes
+ * and frees the saved copy of applied microcode.
+ * 1.03 29 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ * Made to use devfs (/dev/cpu/microcode) + cleanups.
+ * 1.04 06 Jun 2000, Simon Trimmer <simon@veritas.com>
+ * Added misc device support (now uses both devfs and misc).
+ * Added MICROCODE_IOCFREE ioctl to clear memory.
+ * 1.05 09 Jun 2000, Simon Trimmer <simon@veritas.com>
+ * Messages for error cases (non Intel & no suitable microcode).
+ * 1.06 03 Aug 2000, Tigran Aivazian <tigran@veritas.com>
+ * Removed ->release(). Removed exclusive open and status bitmap.
+ * Added microcode_rwsem to serialize read()/write()/ioctl().
+ * Removed global kernel lock usage.
+ * 1.07 07 Sep 2000, Tigran Aivazian <tigran@veritas.com>
+ * Write 0 to 0x8B msr and then cpuid before reading revision,
+ * so that it works even if there were no update done by the
+ * BIOS. Otherwise, reading from 0x8B gives junk (which happened
+ * to be 0 on my machine which is why it worked even when I
+ * disabled update by the BIOS)
+ * Thanks to Eric W. Biederman <ebiederman@lnxi.com> for the fix.
+ * 1.08 11 Dec 2000, Richard Schaal <richard.schaal@intel.com> and
+ * Tigran Aivazian <tigran@veritas.com>
+ * Intel Pentium 4 processor support and bugfixes.
+ * 1.09 30 Oct 2001, Tigran Aivazian <tigran@veritas.com>
+ * Bugfix for HT (Hyper-Threading) enabled processors
+ * whereby processor resources are shared by all logical processors
+ * in a single CPU package.
+ * 1.10 28 Feb 2002 Asit K Mallick <asit.k.mallick@intel.com> and
+ * Tigran Aivazian <tigran@veritas.com>,
+ * Serialize updates as required on HT processors due to
+ * speculative nature of implementation.
+ * 1.11 22 Mar 2002 Tigran Aivazian <tigran@veritas.com>
+ * Fix the panic when writing zero-length microcode chunk.
+ * 1.12 29 Sep 2003 Nitin Kamble <nitin.a.kamble@intel.com>,
+ * Jun Nakajima <jun.nakajima@intel.com>
+ * Support for the microcode updates in the new format.
+ * 1.13 10 Oct 2003 Tigran Aivazian <tigran@veritas.com>
+ * Removed ->read() method and obsoleted MICROCODE_IOCFREE ioctl
+ * because we no longer hold a copy of applied microcode
+ * in kernel memory.
+ * 1.14 25 Jun 2004 Tigran Aivazian <tigran@veritas.com>
+ * Fix sigmatch() macro to handle old CPUs with pf == 0.
+ * Thanks to Stuart Swales for pointing out this bug.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/firmware.h>
+#include <linux/uaccess.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/vmalloc.h>
+
+#include <asm/microcode_intel.h>
+#include <asm/processor.h>
+#include <asm/msr.h>
+
+MODULE_DESCRIPTION("Microcode Update Driver");
+MODULE_AUTHOR("Tigran Aivazian <tigran@aivazian.fsnet.co.uk>");
+MODULE_LICENSE("GPL");
+
+static int collect_cpu_info(int cpu_num, struct cpu_signature *csig)
+{
+ struct cpuinfo_x86 *c = &cpu_data(cpu_num);
+ unsigned int val[2];
+
+ memset(csig, 0, sizeof(*csig));
+
+ csig->sig = cpuid_eax(0x00000001);
+
+ if ((c->x86_model >= 5) || (c->x86 > 6)) {
+ /* get processor flags from MSR 0x17 */
+ rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]);
+ csig->pf = 1 << ((val[1] >> 18) & 7);
+ }
+
+ csig->rev = c->microcode;
+ pr_info("CPU%d sig=0x%x, pf=0x%x, revision=0x%x\n",
+ cpu_num, csig->sig, csig->pf, csig->rev);
+
+ return 0;
+}
+
+/*
+ * return 0 - no update found
+ * return 1 - found update
+ */
+static int get_matching_mc(struct microcode_intel *mc_intel, int cpu)
+{
+ struct cpu_signature cpu_sig;
+ unsigned int csig, cpf, crev;
+
+ collect_cpu_info(cpu, &cpu_sig);
+
+ csig = cpu_sig.sig;
+ cpf = cpu_sig.pf;
+ crev = cpu_sig.rev;
+
+ return get_matching_microcode(csig, cpf, mc_intel, crev);
+}
+
+int apply_microcode(int cpu)
+{
+ struct microcode_intel *mc_intel;
+ struct ucode_cpu_info *uci;
+ unsigned int val[2];
+ int cpu_num = raw_smp_processor_id();
+ struct cpuinfo_x86 *c = &cpu_data(cpu_num);
+
+ uci = ucode_cpu_info + cpu;
+ mc_intel = uci->mc;
+
+ /* We should bind the task to the CPU */
+ BUG_ON(cpu_num != cpu);
+
+ if (mc_intel == NULL)
+ return 0;
+
+ /*
+ * Microcode on this CPU could be updated earlier. Only apply the
+ * microcode patch in mc_intel when it is newer than the one on this
+ * CPU.
+ */
+ if (get_matching_mc(mc_intel, cpu) == 0)
+ return 0;
+
+ /* write microcode via MSR 0x79 */
+ wrmsr(MSR_IA32_UCODE_WRITE,
+ (unsigned long) mc_intel->bits,
+ (unsigned long) mc_intel->bits >> 16 >> 16);
+ wrmsr(MSR_IA32_UCODE_REV, 0, 0);
+
+ /* As documented in the SDM: Do a CPUID 1 here */
+ sync_core();
+
+ /* get the current revision from MSR 0x8B */
+ rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]);
+
+ if (val[1] != mc_intel->hdr.rev) {
+ pr_err("CPU%d update to revision 0x%x failed\n",
+ cpu_num, mc_intel->hdr.rev);
+ return -1;
+ }
+ pr_info("CPU%d updated to revision 0x%x, date = %04x-%02x-%02x\n",
+ cpu_num, val[1],
+ mc_intel->hdr.date & 0xffff,
+ mc_intel->hdr.date >> 24,
+ (mc_intel->hdr.date >> 16) & 0xff);
+
+ uci->cpu_sig.rev = val[1];
+ c->microcode = val[1];
+
+ return 0;
+}
+
+static enum ucode_state generic_load_microcode(int cpu, void *data, size_t size,
+ int (*get_ucode_data)(void *, const void *, size_t))
+{
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+ u8 *ucode_ptr = data, *new_mc = NULL, *mc = NULL;
+ int new_rev = uci->cpu_sig.rev;
+ unsigned int leftover = size;
+ enum ucode_state state = UCODE_OK;
+ unsigned int curr_mc_size = 0;
+ unsigned int csig, cpf;
+
+ while (leftover) {
+ struct microcode_header_intel mc_header;
+ unsigned int mc_size;
+
+ if (get_ucode_data(&mc_header, ucode_ptr, sizeof(mc_header)))
+ break;
+
+ mc_size = get_totalsize(&mc_header);
+ if (!mc_size || mc_size > leftover) {
+ pr_err("error! Bad data in microcode data file\n");
+ break;
+ }
+
+ /* For performance reasons, reuse mc area when possible */
+ if (!mc || mc_size > curr_mc_size) {
+ vfree(mc);
+ mc = vmalloc(mc_size);
+ if (!mc)
+ break;
+ curr_mc_size = mc_size;
+ }
+
+ if (get_ucode_data(mc, ucode_ptr, mc_size) ||
+ microcode_sanity_check(mc, 1) < 0) {
+ break;
+ }
+
+ csig = uci->cpu_sig.sig;
+ cpf = uci->cpu_sig.pf;
+ if (get_matching_microcode(csig, cpf, mc, new_rev)) {
+ vfree(new_mc);
+ new_rev = mc_header.rev;
+ new_mc = mc;
+ mc = NULL; /* trigger new vmalloc */
+ }
+
+ ucode_ptr += mc_size;
+ leftover -= mc_size;
+ }
+
+ vfree(mc);
+
+ if (leftover) {
+ vfree(new_mc);
+ state = UCODE_ERROR;
+ goto out;
+ }
+
+ if (!new_mc) {
+ state = UCODE_NFOUND;
+ goto out;
+ }
+
+ vfree(uci->mc);
+ uci->mc = (struct microcode_intel *)new_mc;
+
+ /*
+ * If early loading microcode is supported, save this mc into
+ * permanent memory. So it will be loaded early when a CPU is hot added
+ * or resumes.
+ */
+ save_mc_for_early(new_mc);
+
+ pr_debug("CPU%d found a matching microcode update with version 0x%x (current=0x%x)\n",
+ cpu, new_rev, uci->cpu_sig.rev);
+out:
+ return state;
+}
+
+static int get_ucode_fw(void *to, const void *from, size_t n)
+{
+ memcpy(to, from, n);
+ return 0;
+}
+
+static enum ucode_state request_microcode_fw(int cpu, struct device *device,
+ bool refresh_fw)
+{
+ char name[30];
+ struct cpuinfo_x86 *c = &cpu_data(cpu);
+ const struct firmware *firmware;
+ enum ucode_state ret;
+
+ sprintf(name, "intel-ucode/%02x-%02x-%02x",
+ c->x86, c->x86_model, c->x86_mask);
+
+ if (request_firmware(&firmware, name, device)) {
+ pr_debug("data file %s load failed\n", name);
+ return UCODE_NFOUND;
+ }
+
+ ret = generic_load_microcode(cpu, (void *)firmware->data,
+ firmware->size, &get_ucode_fw);
+
+ release_firmware(firmware);
+
+ return ret;
+}
+
+static int get_ucode_user(void *to, const void *from, size_t n)
+{
+ return copy_from_user(to, from, n);
+}
+
+static enum ucode_state
+request_microcode_user(int cpu, const void __user *buf, size_t size)
+{
+ return generic_load_microcode(cpu, (void *)buf, size, &get_ucode_user);
+}
+
+static void microcode_fini_cpu(int cpu)
+{
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+ vfree(uci->mc);
+ uci->mc = NULL;
+}
+
+static struct microcode_ops microcode_intel_ops = {
+ .request_microcode_user = request_microcode_user,
+ .request_microcode_fw = request_microcode_fw,
+ .collect_cpu_info = collect_cpu_info,
+ .apply_microcode = apply_microcode,
+ .microcode_fini_cpu = microcode_fini_cpu,
+};
+
+struct microcode_ops * __init init_intel_microcode(void)
+{
+ struct cpuinfo_x86 *c = &cpu_data(0);
+
+ if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 ||
+ cpu_has(c, X86_FEATURE_IA64)) {
+ pr_err("Intel CPU family 0x%x not supported\n", c->x86);
+ return NULL;
+ }
+
+ return &microcode_intel_ops;
+}
+
diff --git a/arch/x86/kernel/cpu/microcode/intel_early.c b/arch/x86/kernel/cpu/microcode/intel_early.c
new file mode 100644
index 000000000000..18f739129e72
--- /dev/null
+++ b/arch/x86/kernel/cpu/microcode/intel_early.c
@@ -0,0 +1,787 @@
+/*
+ * Intel CPU microcode early update for Linux
+ *
+ * Copyright (C) 2012 Fenghua Yu <fenghua.yu@intel.com>
+ * H Peter Anvin" <hpa@zytor.com>
+ *
+ * This allows to early upgrade microcode on Intel processors
+ * belonging to IA-32 family - PentiumPro, Pentium II,
+ * Pentium III, Xeon, Pentium 4, etc.
+ *
+ * Reference: Section 9.11 of Volume 3, IA-32 Intel Architecture
+ * Software Developer's Manual.
+ *
+ * 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/module.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/earlycpio.h>
+#include <linux/initrd.h>
+#include <linux/cpu.h>
+#include <asm/msr.h>
+#include <asm/microcode_intel.h>
+#include <asm/processor.h>
+#include <asm/tlbflush.h>
+#include <asm/setup.h>
+
+unsigned long mc_saved_in_initrd[MAX_UCODE_COUNT];
+struct mc_saved_data {
+ unsigned int mc_saved_count;
+ struct microcode_intel **mc_saved;
+} mc_saved_data;
+
+static enum ucode_state
+generic_load_microcode_early(struct microcode_intel **mc_saved_p,
+ unsigned int mc_saved_count,
+ struct ucode_cpu_info *uci)
+{
+ struct microcode_intel *ucode_ptr, *new_mc = NULL;
+ int new_rev = uci->cpu_sig.rev;
+ enum ucode_state state = UCODE_OK;
+ unsigned int mc_size;
+ struct microcode_header_intel *mc_header;
+ unsigned int csig = uci->cpu_sig.sig;
+ unsigned int cpf = uci->cpu_sig.pf;
+ int i;
+
+ for (i = 0; i < mc_saved_count; i++) {
+ ucode_ptr = mc_saved_p[i];
+
+ mc_header = (struct microcode_header_intel *)ucode_ptr;
+ mc_size = get_totalsize(mc_header);
+ if (get_matching_microcode(csig, cpf, ucode_ptr, new_rev)) {
+ new_rev = mc_header->rev;
+ new_mc = ucode_ptr;
+ }
+ }
+
+ if (!new_mc) {
+ state = UCODE_NFOUND;
+ goto out;
+ }
+
+ uci->mc = (struct microcode_intel *)new_mc;
+out:
+ return state;
+}
+
+static void
+microcode_pointer(struct microcode_intel **mc_saved,
+ unsigned long *mc_saved_in_initrd,
+ unsigned long initrd_start, int mc_saved_count)
+{
+ int i;
+
+ for (i = 0; i < mc_saved_count; i++)
+ mc_saved[i] = (struct microcode_intel *)
+ (mc_saved_in_initrd[i] + initrd_start);
+}
+
+#ifdef CONFIG_X86_32
+static void
+microcode_phys(struct microcode_intel **mc_saved_tmp,
+ struct mc_saved_data *mc_saved_data)
+{
+ int i;
+ struct microcode_intel ***mc_saved;
+
+ mc_saved = (struct microcode_intel ***)
+ __pa_nodebug(&mc_saved_data->mc_saved);
+ for (i = 0; i < mc_saved_data->mc_saved_count; i++) {
+ struct microcode_intel *p;
+
+ p = *(struct microcode_intel **)
+ __pa_nodebug(mc_saved_data->mc_saved + i);
+ mc_saved_tmp[i] = (struct microcode_intel *)__pa_nodebug(p);
+ }
+}
+#endif
+
+static enum ucode_state
+load_microcode(struct mc_saved_data *mc_saved_data,
+ unsigned long *mc_saved_in_initrd,
+ unsigned long initrd_start,
+ struct ucode_cpu_info *uci)
+{
+ struct microcode_intel *mc_saved_tmp[MAX_UCODE_COUNT];
+ unsigned int count = mc_saved_data->mc_saved_count;
+
+ if (!mc_saved_data->mc_saved) {
+ microcode_pointer(mc_saved_tmp, mc_saved_in_initrd,
+ initrd_start, count);
+
+ return generic_load_microcode_early(mc_saved_tmp, count, uci);
+ } else {
+#ifdef CONFIG_X86_32
+ microcode_phys(mc_saved_tmp, mc_saved_data);
+ return generic_load_microcode_early(mc_saved_tmp, count, uci);
+#else
+ return generic_load_microcode_early(mc_saved_data->mc_saved,
+ count, uci);
+#endif
+ }
+}
+
+static u8 get_x86_family(unsigned long sig)
+{
+ u8 x86;
+
+ x86 = (sig >> 8) & 0xf;
+
+ if (x86 == 0xf)
+ x86 += (sig >> 20) & 0xff;
+
+ return x86;
+}
+
+static u8 get_x86_model(unsigned long sig)
+{
+ u8 x86, x86_model;
+
+ x86 = get_x86_family(sig);
+ x86_model = (sig >> 4) & 0xf;
+
+ if (x86 == 0x6 || x86 == 0xf)
+ x86_model += ((sig >> 16) & 0xf) << 4;
+
+ return x86_model;
+}
+
+/*
+ * Given CPU signature and a microcode patch, this function finds if the
+ * microcode patch has matching family and model with the CPU.
+ */
+static enum ucode_state
+matching_model_microcode(struct microcode_header_intel *mc_header,
+ unsigned long sig)
+{
+ u8 x86, x86_model;
+ u8 x86_ucode, x86_model_ucode;
+ struct extended_sigtable *ext_header;
+ unsigned long total_size = get_totalsize(mc_header);
+ unsigned long data_size = get_datasize(mc_header);
+ int ext_sigcount, i;
+ struct extended_signature *ext_sig;
+
+ x86 = get_x86_family(sig);
+ x86_model = get_x86_model(sig);
+
+ x86_ucode = get_x86_family(mc_header->sig);
+ x86_model_ucode = get_x86_model(mc_header->sig);
+
+ if (x86 == x86_ucode && x86_model == x86_model_ucode)
+ return UCODE_OK;
+
+ /* Look for ext. headers: */
+ if (total_size <= data_size + MC_HEADER_SIZE)
+ return UCODE_NFOUND;
+
+ ext_header = (struct extended_sigtable *)
+ mc_header + data_size + MC_HEADER_SIZE;
+ ext_sigcount = ext_header->count;
+ ext_sig = (void *)ext_header + EXT_HEADER_SIZE;
+
+ for (i = 0; i < ext_sigcount; i++) {
+ x86_ucode = get_x86_family(ext_sig->sig);
+ x86_model_ucode = get_x86_model(ext_sig->sig);
+
+ if (x86 == x86_ucode && x86_model == x86_model_ucode)
+ return UCODE_OK;
+
+ ext_sig++;
+ }
+
+ return UCODE_NFOUND;
+}
+
+static int
+save_microcode(struct mc_saved_data *mc_saved_data,
+ struct microcode_intel **mc_saved_src,
+ unsigned int mc_saved_count)
+{
+ int i, j;
+ struct microcode_intel **mc_saved_p;
+ int ret;
+
+ if (!mc_saved_count)
+ return -EINVAL;
+
+ /*
+ * Copy new microcode data.
+ */
+ mc_saved_p = kmalloc(mc_saved_count*sizeof(struct microcode_intel *),
+ GFP_KERNEL);
+ if (!mc_saved_p)
+ return -ENOMEM;
+
+ for (i = 0; i < mc_saved_count; i++) {
+ struct microcode_intel *mc = mc_saved_src[i];
+ struct microcode_header_intel *mc_header = &mc->hdr;
+ unsigned long mc_size = get_totalsize(mc_header);
+ mc_saved_p[i] = kmalloc(mc_size, GFP_KERNEL);
+ if (!mc_saved_p[i]) {
+ ret = -ENOMEM;
+ goto err;
+ }
+ if (!mc_saved_src[i]) {
+ ret = -EINVAL;
+ goto err;
+ }
+ memcpy(mc_saved_p[i], mc, mc_size);
+ }
+
+ /*
+ * Point to newly saved microcode.
+ */
+ mc_saved_data->mc_saved = mc_saved_p;
+ mc_saved_data->mc_saved_count = mc_saved_count;
+
+ return 0;
+
+err:
+ for (j = 0; j <= i; j++)
+ kfree(mc_saved_p[j]);
+ kfree(mc_saved_p);
+
+ return ret;
+}
+
+/*
+ * A microcode patch in ucode_ptr is saved into mc_saved
+ * - if it has matching signature and newer revision compared to an existing
+ * patch mc_saved.
+ * - or if it is a newly discovered microcode patch.
+ *
+ * The microcode patch should have matching model with CPU.
+ */
+static void _save_mc(struct microcode_intel **mc_saved, u8 *ucode_ptr,
+ unsigned int *mc_saved_count_p)
+{
+ int i;
+ int found = 0;
+ unsigned int mc_saved_count = *mc_saved_count_p;
+ struct microcode_header_intel *mc_header;
+
+ mc_header = (struct microcode_header_intel *)ucode_ptr;
+ for (i = 0; i < mc_saved_count; i++) {
+ unsigned int sig, pf;
+ unsigned int new_rev;
+ struct microcode_header_intel *mc_saved_header =
+ (struct microcode_header_intel *)mc_saved[i];
+ sig = mc_saved_header->sig;
+ pf = mc_saved_header->pf;
+ new_rev = mc_header->rev;
+
+ if (get_matching_sig(sig, pf, ucode_ptr, new_rev)) {
+ found = 1;
+ if (update_match_revision(mc_header, new_rev)) {
+ /*
+ * Found an older ucode saved before.
+ * Replace the older one with this newer
+ * one.
+ */
+ mc_saved[i] =
+ (struct microcode_intel *)ucode_ptr;
+ break;
+ }
+ }
+ }
+ if (i >= mc_saved_count && !found)
+ /*
+ * This ucode is first time discovered in ucode file.
+ * Save it to memory.
+ */
+ mc_saved[mc_saved_count++] =
+ (struct microcode_intel *)ucode_ptr;
+
+ *mc_saved_count_p = mc_saved_count;
+}
+
+/*
+ * Get microcode matching with BSP's model. Only CPUs with the same model as
+ * BSP can stay in the platform.
+ */
+static enum ucode_state __init
+get_matching_model_microcode(int cpu, unsigned long start,
+ void *data, size_t size,
+ struct mc_saved_data *mc_saved_data,
+ unsigned long *mc_saved_in_initrd,
+ struct ucode_cpu_info *uci)
+{
+ u8 *ucode_ptr = data;
+ unsigned int leftover = size;
+ enum ucode_state state = UCODE_OK;
+ unsigned int mc_size;
+ struct microcode_header_intel *mc_header;
+ struct microcode_intel *mc_saved_tmp[MAX_UCODE_COUNT];
+ unsigned int mc_saved_count = mc_saved_data->mc_saved_count;
+ int i;
+
+ while (leftover) {
+ mc_header = (struct microcode_header_intel *)ucode_ptr;
+
+ mc_size = get_totalsize(mc_header);
+ if (!mc_size || mc_size > leftover ||
+ microcode_sanity_check(ucode_ptr, 0) < 0)
+ break;
+
+ leftover -= mc_size;
+
+ /*
+ * Since APs with same family and model as the BSP may boot in
+ * the platform, we need to find and save microcode patches
+ * with the same family and model as the BSP.
+ */
+ if (matching_model_microcode(mc_header, uci->cpu_sig.sig) !=
+ UCODE_OK) {
+ ucode_ptr += mc_size;
+ continue;
+ }
+
+ _save_mc(mc_saved_tmp, ucode_ptr, &mc_saved_count);
+
+ ucode_ptr += mc_size;
+ }
+
+ if (leftover) {
+ state = UCODE_ERROR;
+ goto out;
+ }
+
+ if (mc_saved_count == 0) {
+ state = UCODE_NFOUND;
+ goto out;
+ }
+
+ for (i = 0; i < mc_saved_count; i++)
+ mc_saved_in_initrd[i] = (unsigned long)mc_saved_tmp[i] - start;
+
+ mc_saved_data->mc_saved_count = mc_saved_count;
+out:
+ return state;
+}
+
+static int collect_cpu_info_early(struct ucode_cpu_info *uci)
+{
+ unsigned int val[2];
+ u8 x86, x86_model;
+ struct cpu_signature csig;
+ unsigned int eax, ebx, ecx, edx;
+
+ csig.sig = 0;
+ csig.pf = 0;
+ csig.rev = 0;
+
+ memset(uci, 0, sizeof(*uci));
+
+ eax = 0x00000001;
+ ecx = 0;
+ native_cpuid(&eax, &ebx, &ecx, &edx);
+ csig.sig = eax;
+
+ x86 = get_x86_family(csig.sig);
+ x86_model = get_x86_model(csig.sig);
+
+ if ((x86_model >= 5) || (x86 > 6)) {
+ /* get processor flags from MSR 0x17 */
+ native_rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]);
+ csig.pf = 1 << ((val[1] >> 18) & 7);
+ }
+ native_wrmsr(MSR_IA32_UCODE_REV, 0, 0);
+
+ /* As documented in the SDM: Do a CPUID 1 here */
+ sync_core();
+
+ /* get the current revision from MSR 0x8B */
+ native_rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]);
+
+ csig.rev = val[1];
+
+ uci->cpu_sig = csig;
+ uci->valid = 1;
+
+ return 0;
+}
+
+#ifdef DEBUG
+static void __ref show_saved_mc(void)
+{
+ int i, j;
+ unsigned int sig, pf, rev, total_size, data_size, date;
+ struct ucode_cpu_info uci;
+
+ if (mc_saved_data.mc_saved_count == 0) {
+ pr_debug("no micorcode data saved.\n");
+ return;
+ }
+ pr_debug("Total microcode saved: %d\n", mc_saved_data.mc_saved_count);
+
+ collect_cpu_info_early(&uci);
+
+ sig = uci.cpu_sig.sig;
+ pf = uci.cpu_sig.pf;
+ rev = uci.cpu_sig.rev;
+ pr_debug("CPU%d: sig=0x%x, pf=0x%x, rev=0x%x\n",
+ smp_processor_id(), sig, pf, rev);
+
+ for (i = 0; i < mc_saved_data.mc_saved_count; i++) {
+ struct microcode_header_intel *mc_saved_header;
+ struct extended_sigtable *ext_header;
+ int ext_sigcount;
+ struct extended_signature *ext_sig;
+
+ mc_saved_header = (struct microcode_header_intel *)
+ mc_saved_data.mc_saved[i];
+ sig = mc_saved_header->sig;
+ pf = mc_saved_header->pf;
+ rev = mc_saved_header->rev;
+ total_size = get_totalsize(mc_saved_header);
+ data_size = get_datasize(mc_saved_header);
+ date = mc_saved_header->date;
+
+ pr_debug("mc_saved[%d]: sig=0x%x, pf=0x%x, rev=0x%x, toal size=0x%x, date = %04x-%02x-%02x\n",
+ i, sig, pf, rev, total_size,
+ date & 0xffff,
+ date >> 24,
+ (date >> 16) & 0xff);
+
+ /* Look for ext. headers: */
+ if (total_size <= data_size + MC_HEADER_SIZE)
+ continue;
+
+ ext_header = (struct extended_sigtable *)
+ mc_saved_header + data_size + MC_HEADER_SIZE;
+ ext_sigcount = ext_header->count;
+ ext_sig = (void *)ext_header + EXT_HEADER_SIZE;
+
+ for (j = 0; j < ext_sigcount; j++) {
+ sig = ext_sig->sig;
+ pf = ext_sig->pf;
+
+ pr_debug("\tExtended[%d]: sig=0x%x, pf=0x%x\n",
+ j, sig, pf);
+
+ ext_sig++;
+ }
+
+ }
+}
+#else
+static inline void show_saved_mc(void)
+{
+}
+#endif
+
+#if defined(CONFIG_MICROCODE_INTEL_EARLY) && defined(CONFIG_HOTPLUG_CPU)
+static DEFINE_MUTEX(x86_cpu_microcode_mutex);
+/*
+ * Save this mc into mc_saved_data. So it will be loaded early when a CPU is
+ * hot added or resumes.
+ *
+ * Please make sure this mc should be a valid microcode patch before calling
+ * this function.
+ */
+int save_mc_for_early(u8 *mc)
+{
+ struct microcode_intel *mc_saved_tmp[MAX_UCODE_COUNT];
+ unsigned int mc_saved_count_init;
+ unsigned int mc_saved_count;
+ struct microcode_intel **mc_saved;
+ int ret = 0;
+ int i;
+
+ /*
+ * Hold hotplug lock so mc_saved_data is not accessed by a CPU in
+ * hotplug.
+ */
+ mutex_lock(&x86_cpu_microcode_mutex);
+
+ mc_saved_count_init = mc_saved_data.mc_saved_count;
+ mc_saved_count = mc_saved_data.mc_saved_count;
+ mc_saved = mc_saved_data.mc_saved;
+
+ if (mc_saved && mc_saved_count)
+ memcpy(mc_saved_tmp, mc_saved,
+ mc_saved_count * sizeof(struct mirocode_intel *));
+ /*
+ * Save the microcode patch mc in mc_save_tmp structure if it's a newer
+ * version.
+ */
+
+ _save_mc(mc_saved_tmp, mc, &mc_saved_count);
+
+ /*
+ * Save the mc_save_tmp in global mc_saved_data.
+ */
+ ret = save_microcode(&mc_saved_data, mc_saved_tmp, mc_saved_count);
+ if (ret) {
+ pr_err("Cannot save microcode patch.\n");
+ goto out;
+ }
+
+ show_saved_mc();
+
+ /*
+ * Free old saved microcod data.
+ */
+ if (mc_saved) {
+ for (i = 0; i < mc_saved_count_init; i++)
+ kfree(mc_saved[i]);
+ kfree(mc_saved);
+ }
+
+out:
+ mutex_unlock(&x86_cpu_microcode_mutex);
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(save_mc_for_early);
+#endif
+
+static __initdata char ucode_name[] = "kernel/x86/microcode/GenuineIntel.bin";
+static __init enum ucode_state
+scan_microcode(unsigned long start, unsigned long end,
+ struct mc_saved_data *mc_saved_data,
+ unsigned long *mc_saved_in_initrd,
+ struct ucode_cpu_info *uci)
+{
+ unsigned int size = end - start + 1;
+ struct cpio_data cd;
+ long offset = 0;
+#ifdef CONFIG_X86_32
+ char *p = (char *)__pa_nodebug(ucode_name);
+#else
+ char *p = ucode_name;
+#endif
+
+ cd.data = NULL;
+ cd.size = 0;
+
+ cd = find_cpio_data(p, (void *)start, size, &offset);
+ if (!cd.data)
+ return UCODE_ERROR;
+
+
+ return get_matching_model_microcode(0, start, cd.data, cd.size,
+ mc_saved_data, mc_saved_in_initrd,
+ uci);
+}
+
+/*
+ * Print ucode update info.
+ */
+static void
+print_ucode_info(struct ucode_cpu_info *uci, unsigned int date)
+{
+ int cpu = smp_processor_id();
+
+ pr_info("CPU%d microcode updated early to revision 0x%x, date = %04x-%02x-%02x\n",
+ cpu,
+ uci->cpu_sig.rev,
+ date & 0xffff,
+ date >> 24,
+ (date >> 16) & 0xff);
+}
+
+#ifdef CONFIG_X86_32
+
+static int delay_ucode_info;
+static int current_mc_date;
+
+/*
+ * Print early updated ucode info after printk works. This is delayed info dump.
+ */
+void show_ucode_info_early(void)
+{
+ struct ucode_cpu_info uci;
+
+ if (delay_ucode_info) {
+ collect_cpu_info_early(&uci);
+ print_ucode_info(&uci, current_mc_date);
+ delay_ucode_info = 0;
+ }
+}
+
+/*
+ * At this point, we can not call printk() yet. Keep microcode patch number in
+ * mc_saved_data.mc_saved and delay printing microcode info in
+ * show_ucode_info_early() until printk() works.
+ */
+static void print_ucode(struct ucode_cpu_info *uci)
+{
+ struct microcode_intel *mc_intel;
+ int *delay_ucode_info_p;
+ int *current_mc_date_p;
+
+ mc_intel = uci->mc;
+ if (mc_intel == NULL)
+ return;
+
+ delay_ucode_info_p = (int *)__pa_nodebug(&delay_ucode_info);
+ current_mc_date_p = (int *)__pa_nodebug(&current_mc_date);
+
+ *delay_ucode_info_p = 1;
+ *current_mc_date_p = mc_intel->hdr.date;
+}
+#else
+
+/*
+ * Flush global tlb. We only do this in x86_64 where paging has been enabled
+ * already and PGE should be enabled as well.
+ */
+static inline void flush_tlb_early(void)
+{
+ __native_flush_tlb_global_irq_disabled();
+}
+
+static inline void print_ucode(struct ucode_cpu_info *uci)
+{
+ struct microcode_intel *mc_intel;
+
+ mc_intel = uci->mc;
+ if (mc_intel == NULL)
+ return;
+
+ print_ucode_info(uci, mc_intel->hdr.date);
+}
+#endif
+
+static int apply_microcode_early(struct mc_saved_data *mc_saved_data,
+ struct ucode_cpu_info *uci)
+{
+ struct microcode_intel *mc_intel;
+ unsigned int val[2];
+
+ mc_intel = uci->mc;
+ if (mc_intel == NULL)
+ return 0;
+
+ /* write microcode via MSR 0x79 */
+ native_wrmsr(MSR_IA32_UCODE_WRITE,
+ (unsigned long) mc_intel->bits,
+ (unsigned long) mc_intel->bits >> 16 >> 16);
+ native_wrmsr(MSR_IA32_UCODE_REV, 0, 0);
+
+ /* As documented in the SDM: Do a CPUID 1 here */
+ sync_core();
+
+ /* get the current revision from MSR 0x8B */
+ native_rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]);
+ if (val[1] != mc_intel->hdr.rev)
+ return -1;
+
+#ifdef CONFIG_X86_64
+ /* Flush global tlb. This is precaution. */
+ flush_tlb_early();
+#endif
+ uci->cpu_sig.rev = val[1];
+
+ print_ucode(uci);
+
+ return 0;
+}
+
+/*
+ * This function converts microcode patch offsets previously stored in
+ * mc_saved_in_initrd to pointers and stores the pointers in mc_saved_data.
+ */
+int __init save_microcode_in_initrd_intel(void)
+{
+ unsigned int count = mc_saved_data.mc_saved_count;
+ struct microcode_intel *mc_saved[MAX_UCODE_COUNT];
+ int ret = 0;
+
+ if (count == 0)
+ return ret;
+
+ microcode_pointer(mc_saved, mc_saved_in_initrd, initrd_start, count);
+ ret = save_microcode(&mc_saved_data, mc_saved, count);
+ if (ret)
+ pr_err("Cannot save microcode patches from initrd.\n");
+
+ show_saved_mc();
+
+ return ret;
+}
+
+static void __init
+_load_ucode_intel_bsp(struct mc_saved_data *mc_saved_data,
+ unsigned long *mc_saved_in_initrd,
+ unsigned long initrd_start_early,
+ unsigned long initrd_end_early,
+ struct ucode_cpu_info *uci)
+{
+ collect_cpu_info_early(uci);
+ scan_microcode(initrd_start_early, initrd_end_early, mc_saved_data,
+ mc_saved_in_initrd, uci);
+ load_microcode(mc_saved_data, mc_saved_in_initrd,
+ initrd_start_early, uci);
+ apply_microcode_early(mc_saved_data, uci);
+}
+
+void __init
+load_ucode_intel_bsp(void)
+{
+ u64 ramdisk_image, ramdisk_size;
+ unsigned long initrd_start_early, initrd_end_early;
+ struct ucode_cpu_info uci;
+#ifdef CONFIG_X86_32
+ struct boot_params *boot_params_p;
+
+ boot_params_p = (struct boot_params *)__pa_nodebug(&boot_params);
+ ramdisk_image = boot_params_p->hdr.ramdisk_image;
+ ramdisk_size = boot_params_p->hdr.ramdisk_size;
+ initrd_start_early = ramdisk_image;
+ initrd_end_early = initrd_start_early + ramdisk_size;
+
+ _load_ucode_intel_bsp(
+ (struct mc_saved_data *)__pa_nodebug(&mc_saved_data),
+ (unsigned long *)__pa_nodebug(&mc_saved_in_initrd),
+ initrd_start_early, initrd_end_early, &uci);
+#else
+ ramdisk_image = boot_params.hdr.ramdisk_image;
+ ramdisk_size = boot_params.hdr.ramdisk_size;
+ initrd_start_early = ramdisk_image + PAGE_OFFSET;
+ initrd_end_early = initrd_start_early + ramdisk_size;
+
+ _load_ucode_intel_bsp(&mc_saved_data, mc_saved_in_initrd,
+ initrd_start_early, initrd_end_early, &uci);
+#endif
+}
+
+void load_ucode_intel_ap(void)
+{
+ struct mc_saved_data *mc_saved_data_p;
+ struct ucode_cpu_info uci;
+ unsigned long *mc_saved_in_initrd_p;
+ unsigned long initrd_start_addr;
+#ifdef CONFIG_X86_32
+ unsigned long *initrd_start_p;
+
+ mc_saved_in_initrd_p =
+ (unsigned long *)__pa_nodebug(mc_saved_in_initrd);
+ mc_saved_data_p = (struct mc_saved_data *)__pa_nodebug(&mc_saved_data);
+ initrd_start_p = (unsigned long *)__pa_nodebug(&initrd_start);
+ initrd_start_addr = (unsigned long)__pa_nodebug(*initrd_start_p);
+#else
+ mc_saved_data_p = &mc_saved_data;
+ mc_saved_in_initrd_p = mc_saved_in_initrd;
+ initrd_start_addr = initrd_start;
+#endif
+
+ /*
+ * If there is no valid ucode previously saved in memory, no need to
+ * update ucode on this AP.
+ */
+ if (mc_saved_data_p->mc_saved_count == 0)
+ return;
+
+ collect_cpu_info_early(&uci);
+ load_microcode(mc_saved_data_p, mc_saved_in_initrd_p,
+ initrd_start_addr, &uci);
+ apply_microcode_early(mc_saved_data_p, &uci);
+}
diff --git a/arch/x86/kernel/cpu/microcode/intel_lib.c b/arch/x86/kernel/cpu/microcode/intel_lib.c
new file mode 100644
index 000000000000..ce69320d0179
--- /dev/null
+++ b/arch/x86/kernel/cpu/microcode/intel_lib.c
@@ -0,0 +1,174 @@
+/*
+ * Intel CPU Microcode Update Driver for Linux
+ *
+ * Copyright (C) 2012 Fenghua Yu <fenghua.yu@intel.com>
+ * H Peter Anvin" <hpa@zytor.com>
+ *
+ * This driver allows to upgrade microcode on Intel processors
+ * belonging to IA-32 family - PentiumPro, Pentium II,
+ * Pentium III, Xeon, Pentium 4, etc.
+ *
+ * Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture
+ * Software Developer's Manual
+ * Order Number 253668 or free download from:
+ *
+ * http://developer.intel.com/Assets/PDF/manual/253668.pdf
+ *
+ * For more information, go to http://www.urbanmyth.org/microcode
+ *
+ * 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/firmware.h>
+#include <linux/uaccess.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+#include <asm/microcode_intel.h>
+#include <asm/processor.h>
+#include <asm/msr.h>
+
+static inline int
+update_match_cpu(unsigned int csig, unsigned int cpf,
+ unsigned int sig, unsigned int pf)
+{
+ return (!sigmatch(sig, csig, pf, cpf)) ? 0 : 1;
+}
+
+int
+update_match_revision(struct microcode_header_intel *mc_header, int rev)
+{
+ return (mc_header->rev <= rev) ? 0 : 1;
+}
+
+int microcode_sanity_check(void *mc, int print_err)
+{
+ unsigned long total_size, data_size, ext_table_size;
+ struct microcode_header_intel *mc_header = mc;
+ struct extended_sigtable *ext_header = NULL;
+ int sum, orig_sum, ext_sigcount = 0, i;
+ struct extended_signature *ext_sig;
+
+ total_size = get_totalsize(mc_header);
+ data_size = get_datasize(mc_header);
+
+ if (data_size + MC_HEADER_SIZE > total_size) {
+ if (print_err)
+ pr_err("error! Bad data size in microcode data file\n");
+ return -EINVAL;
+ }
+
+ if (mc_header->ldrver != 1 || mc_header->hdrver != 1) {
+ if (print_err)
+ pr_err("error! Unknown microcode update format\n");
+ return -EINVAL;
+ }
+ ext_table_size = total_size - (MC_HEADER_SIZE + data_size);
+ if (ext_table_size) {
+ if ((ext_table_size < EXT_HEADER_SIZE)
+ || ((ext_table_size - EXT_HEADER_SIZE) % EXT_SIGNATURE_SIZE)) {
+ if (print_err)
+ pr_err("error! Small exttable size in microcode data file\n");
+ return -EINVAL;
+ }
+ ext_header = mc + MC_HEADER_SIZE + data_size;
+ if (ext_table_size != exttable_size(ext_header)) {
+ if (print_err)
+ pr_err("error! Bad exttable size in microcode data file\n");
+ return -EFAULT;
+ }
+ ext_sigcount = ext_header->count;
+ }
+
+ /* check extended table checksum */
+ if (ext_table_size) {
+ int ext_table_sum = 0;
+ int *ext_tablep = (int *)ext_header;
+
+ i = ext_table_size / DWSIZE;
+ while (i--)
+ ext_table_sum += ext_tablep[i];
+ if (ext_table_sum) {
+ if (print_err)
+ pr_warn("aborting, bad extended signature table checksum\n");
+ return -EINVAL;
+ }
+ }
+
+ /* calculate the checksum */
+ orig_sum = 0;
+ i = (MC_HEADER_SIZE + data_size) / DWSIZE;
+ while (i--)
+ orig_sum += ((int *)mc)[i];
+ if (orig_sum) {
+ if (print_err)
+ pr_err("aborting, bad checksum\n");
+ return -EINVAL;
+ }
+ if (!ext_table_size)
+ return 0;
+ /* check extended signature checksum */
+ for (i = 0; i < ext_sigcount; i++) {
+ ext_sig = (void *)ext_header + EXT_HEADER_SIZE +
+ EXT_SIGNATURE_SIZE * i;
+ sum = orig_sum
+ - (mc_header->sig + mc_header->pf + mc_header->cksum)
+ + (ext_sig->sig + ext_sig->pf + ext_sig->cksum);
+ if (sum) {
+ if (print_err)
+ pr_err("aborting, bad checksum\n");
+ return -EINVAL;
+ }
+ }
+ return 0;
+}
+EXPORT_SYMBOL_GPL(microcode_sanity_check);
+
+/*
+ * return 0 - no update found
+ * return 1 - found update
+ */
+int get_matching_sig(unsigned int csig, int cpf, void *mc, int rev)
+{
+ struct microcode_header_intel *mc_header = mc;
+ struct extended_sigtable *ext_header;
+ unsigned long total_size = get_totalsize(mc_header);
+ int ext_sigcount, i;
+ struct extended_signature *ext_sig;
+
+ if (update_match_cpu(csig, cpf, mc_header->sig, mc_header->pf))
+ return 1;
+
+ /* Look for ext. headers: */
+ if (total_size <= get_datasize(mc_header) + MC_HEADER_SIZE)
+ return 0;
+
+ ext_header = mc + get_datasize(mc_header) + MC_HEADER_SIZE;
+ ext_sigcount = ext_header->count;
+ ext_sig = (void *)ext_header + EXT_HEADER_SIZE;
+
+ for (i = 0; i < ext_sigcount; i++) {
+ if (update_match_cpu(csig, cpf, ext_sig->sig, ext_sig->pf))
+ return 1;
+ ext_sig++;
+ }
+ return 0;
+}
+
+/*
+ * return 0 - no update found
+ * return 1 - found update
+ */
+int get_matching_microcode(unsigned int csig, int cpf, void *mc, int rev)
+{
+ struct microcode_header_intel *mc_header = mc;
+
+ if (!update_match_revision(mc_header, rev))
+ return 0;
+
+ return get_matching_sig(csig, cpf, mc, rev);
+}
+EXPORT_SYMBOL_GPL(get_matching_microcode);
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 8e132931614d..b88645191fe5 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -1883,21 +1883,27 @@ static struct pmu pmu = {
void arch_perf_update_userpage(struct perf_event_mmap_page *userpg, u64 now)
{
+ struct cyc2ns_data *data;
+
userpg->cap_user_time = 0;
userpg->cap_user_time_zero = 0;
userpg->cap_user_rdpmc = x86_pmu.attr_rdpmc;
userpg->pmc_width = x86_pmu.cntval_bits;
- if (!sched_clock_stable)
+ if (!sched_clock_stable())
return;
+ data = cyc2ns_read_begin();
+
userpg->cap_user_time = 1;
- userpg->time_mult = this_cpu_read(cyc2ns);
- userpg->time_shift = CYC2NS_SCALE_FACTOR;
- userpg->time_offset = this_cpu_read(cyc2ns_offset) - now;
+ userpg->time_mult = data->cyc2ns_mul;
+ userpg->time_shift = data->cyc2ns_shift;
+ userpg->time_offset = data->cyc2ns_offset - now;
userpg->cap_user_time_zero = 1;
- userpg->time_zero = this_cpu_read(cyc2ns_offset);
+ userpg->time_zero = data->cyc2ns_offset;
+
+ cyc2ns_read_end(data);
}
/*
diff --git a/arch/x86/kernel/cpu/perf_event_amd_ibs.c b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
index e09f0bfb7b8f..4b8e4d3cd6ea 100644
--- a/arch/x86/kernel/cpu/perf_event_amd_ibs.c
+++ b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
@@ -10,6 +10,7 @@
#include <linux/module.h>
#include <linux/pci.h>
#include <linux/ptrace.h>
+#include <linux/syscore_ops.h>
#include <asm/apic.h>
@@ -816,6 +817,18 @@ out:
return ret;
}
+static void ibs_eilvt_setup(void)
+{
+ /*
+ * Force LVT offset assignment for family 10h: The offsets are
+ * not assigned by the BIOS for this family, so the OS is
+ * responsible for doing it. If the OS assignment fails, fall
+ * back to BIOS settings and try to setup this.
+ */
+ if (boot_cpu_data.x86 == 0x10)
+ force_ibs_eilvt_setup();
+}
+
static inline int get_ibs_lvt_offset(void)
{
u64 val;
@@ -851,6 +864,36 @@ static void clear_APIC_ibs(void *dummy)
setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_FIX, 1);
}
+#ifdef CONFIG_PM
+
+static int perf_ibs_suspend(void)
+{
+ clear_APIC_ibs(NULL);
+ return 0;
+}
+
+static void perf_ibs_resume(void)
+{
+ ibs_eilvt_setup();
+ setup_APIC_ibs(NULL);
+}
+
+static struct syscore_ops perf_ibs_syscore_ops = {
+ .resume = perf_ibs_resume,
+ .suspend = perf_ibs_suspend,
+};
+
+static void perf_ibs_pm_init(void)
+{
+ register_syscore_ops(&perf_ibs_syscore_ops);
+}
+
+#else
+
+static inline void perf_ibs_pm_init(void) { }
+
+#endif
+
static int
perf_ibs_cpu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
{
@@ -877,18 +920,12 @@ static __init int amd_ibs_init(void)
if (!caps)
return -ENODEV; /* ibs not supported by the cpu */
- /*
- * Force LVT offset assignment for family 10h: The offsets are
- * not assigned by the BIOS for this family, so the OS is
- * responsible for doing it. If the OS assignment fails, fall
- * back to BIOS settings and try to setup this.
- */
- if (boot_cpu_data.x86 == 0x10)
- force_ibs_eilvt_setup();
+ ibs_eilvt_setup();
if (!ibs_eilvt_valid())
goto out;
+ perf_ibs_pm_init();
get_online_cpus();
ibs_caps = caps;
/* make ibs_caps visible to other cpus: */
diff --git a/arch/x86/kernel/cpu/perf_event_intel_rapl.c b/arch/x86/kernel/cpu/perf_event_intel_rapl.c
new file mode 100644
index 000000000000..5ad35ad94d0f
--- /dev/null
+++ b/arch/x86/kernel/cpu/perf_event_intel_rapl.c
@@ -0,0 +1,679 @@
+/*
+ * perf_event_intel_rapl.c: support Intel RAPL energy consumption counters
+ * Copyright (C) 2013 Google, Inc., Stephane Eranian
+ *
+ * Intel RAPL interface is specified in the IA-32 Manual Vol3b
+ * section 14.7.1 (September 2013)
+ *
+ * RAPL provides more controls than just reporting energy consumption
+ * however here we only expose the 3 energy consumption free running
+ * counters (pp0, pkg, dram).
+ *
+ * Each of those counters increments in a power unit defined by the
+ * RAPL_POWER_UNIT MSR. On SandyBridge, this unit is 1/(2^16) Joules
+ * but it can vary.
+ *
+ * Counter to rapl events mappings:
+ *
+ * pp0 counter: consumption of all physical cores (power plane 0)
+ * event: rapl_energy_cores
+ * perf code: 0x1
+ *
+ * pkg counter: consumption of the whole processor package
+ * event: rapl_energy_pkg
+ * perf code: 0x2
+ *
+ * dram counter: consumption of the dram domain (servers only)
+ * event: rapl_energy_dram
+ * perf code: 0x3
+ *
+ * dram counter: consumption of the builtin-gpu domain (client only)
+ * event: rapl_energy_gpu
+ * perf code: 0x4
+ *
+ * We manage those counters as free running (read-only). They may be
+ * use simultaneously by other tools, such as turbostat.
+ *
+ * The events only support system-wide mode counting. There is no
+ * sampling support because it does not make sense and is not
+ * supported by the RAPL hardware.
+ *
+ * Because we want to avoid floating-point operations in the kernel,
+ * the events are all reported in fixed point arithmetic (32.32).
+ * Tools must adjust the counts to convert them to Watts using
+ * the duration of the measurement. Tools may use a function such as
+ * ldexp(raw_count, -32);
+ */
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/perf_event.h>
+#include <asm/cpu_device_id.h>
+#include "perf_event.h"
+
+/*
+ * RAPL energy status counters
+ */
+#define RAPL_IDX_PP0_NRG_STAT 0 /* all cores */
+#define INTEL_RAPL_PP0 0x1 /* pseudo-encoding */
+#define RAPL_IDX_PKG_NRG_STAT 1 /* entire package */
+#define INTEL_RAPL_PKG 0x2 /* pseudo-encoding */
+#define RAPL_IDX_RAM_NRG_STAT 2 /* DRAM */
+#define INTEL_RAPL_RAM 0x3 /* pseudo-encoding */
+#define RAPL_IDX_PP1_NRG_STAT 3 /* DRAM */
+#define INTEL_RAPL_PP1 0x4 /* pseudo-encoding */
+
+/* Clients have PP0, PKG */
+#define RAPL_IDX_CLN (1<<RAPL_IDX_PP0_NRG_STAT|\
+ 1<<RAPL_IDX_PKG_NRG_STAT|\
+ 1<<RAPL_IDX_PP1_NRG_STAT)
+
+/* Servers have PP0, PKG, RAM */
+#define RAPL_IDX_SRV (1<<RAPL_IDX_PP0_NRG_STAT|\
+ 1<<RAPL_IDX_PKG_NRG_STAT|\
+ 1<<RAPL_IDX_RAM_NRG_STAT)
+
+/*
+ * event code: LSB 8 bits, passed in attr->config
+ * any other bit is reserved
+ */
+#define RAPL_EVENT_MASK 0xFFULL
+
+#define DEFINE_RAPL_FORMAT_ATTR(_var, _name, _format) \
+static ssize_t __rapl_##_var##_show(struct kobject *kobj, \
+ struct kobj_attribute *attr, \
+ char *page) \
+{ \
+ BUILD_BUG_ON(sizeof(_format) >= PAGE_SIZE); \
+ return sprintf(page, _format "\n"); \
+} \
+static struct kobj_attribute format_attr_##_var = \
+ __ATTR(_name, 0444, __rapl_##_var##_show, NULL)
+
+#define RAPL_EVENT_DESC(_name, _config) \
+{ \
+ .attr = __ATTR(_name, 0444, rapl_event_show, NULL), \
+ .config = _config, \
+}
+
+#define RAPL_CNTR_WIDTH 32 /* 32-bit rapl counters */
+
+struct rapl_pmu {
+ spinlock_t lock;
+ int hw_unit; /* 1/2^hw_unit Joule */
+ int n_active; /* number of active events */
+ struct list_head active_list;
+ struct pmu *pmu; /* pointer to rapl_pmu_class */
+ ktime_t timer_interval; /* in ktime_t unit */
+ struct hrtimer hrtimer;
+};
+
+static struct pmu rapl_pmu_class;
+static cpumask_t rapl_cpu_mask;
+static int rapl_cntr_mask;
+
+static DEFINE_PER_CPU(struct rapl_pmu *, rapl_pmu);
+static DEFINE_PER_CPU(struct rapl_pmu *, rapl_pmu_to_free);
+
+static inline u64 rapl_read_counter(struct perf_event *event)
+{
+ u64 raw;
+ rdmsrl(event->hw.event_base, raw);
+ return raw;
+}
+
+static inline u64 rapl_scale(u64 v)
+{
+ /*
+ * scale delta to smallest unit (1/2^32)
+ * users must then scale back: count * 1/(1e9*2^32) to get Joules
+ * or use ldexp(count, -32).
+ * Watts = Joules/Time delta
+ */
+ return v << (32 - __get_cpu_var(rapl_pmu)->hw_unit);
+}
+
+static u64 rapl_event_update(struct perf_event *event)
+{
+ struct hw_perf_event *hwc = &event->hw;
+ u64 prev_raw_count, new_raw_count;
+ s64 delta, sdelta;
+ int shift = RAPL_CNTR_WIDTH;
+
+again:
+ prev_raw_count = local64_read(&hwc->prev_count);
+ rdmsrl(event->hw.event_base, new_raw_count);
+
+ if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
+ new_raw_count) != prev_raw_count) {
+ cpu_relax();
+ goto again;
+ }
+
+ /*
+ * Now we have the new raw value and have updated the prev
+ * timestamp already. We can now calculate the elapsed delta
+ * (event-)time and add that to the generic event.
+ *
+ * Careful, not all hw sign-extends above the physical width
+ * of the count.
+ */
+ delta = (new_raw_count << shift) - (prev_raw_count << shift);
+ delta >>= shift;
+
+ sdelta = rapl_scale(delta);
+
+ local64_add(sdelta, &event->count);
+
+ return new_raw_count;
+}
+
+static void rapl_start_hrtimer(struct rapl_pmu *pmu)
+{
+ __hrtimer_start_range_ns(&pmu->hrtimer,
+ pmu->timer_interval, 0,
+ HRTIMER_MODE_REL_PINNED, 0);
+}
+
+static void rapl_stop_hrtimer(struct rapl_pmu *pmu)
+{
+ hrtimer_cancel(&pmu->hrtimer);
+}
+
+static enum hrtimer_restart rapl_hrtimer_handle(struct hrtimer *hrtimer)
+{
+ struct rapl_pmu *pmu = __get_cpu_var(rapl_pmu);
+ struct perf_event *event;
+ unsigned long flags;
+
+ if (!pmu->n_active)
+ return HRTIMER_NORESTART;
+
+ spin_lock_irqsave(&pmu->lock, flags);
+
+ list_for_each_entry(event, &pmu->active_list, active_entry) {
+ rapl_event_update(event);
+ }
+
+ spin_unlock_irqrestore(&pmu->lock, flags);
+
+ hrtimer_forward_now(hrtimer, pmu->timer_interval);
+
+ return HRTIMER_RESTART;
+}
+
+static void rapl_hrtimer_init(struct rapl_pmu *pmu)
+{
+ struct hrtimer *hr = &pmu->hrtimer;
+
+ hrtimer_init(hr, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+ hr->function = rapl_hrtimer_handle;
+}
+
+static void __rapl_pmu_event_start(struct rapl_pmu *pmu,
+ struct perf_event *event)
+{
+ if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED)))
+ return;
+
+ event->hw.state = 0;
+
+ list_add_tail(&event->active_entry, &pmu->active_list);
+
+ local64_set(&event->hw.prev_count, rapl_read_counter(event));
+
+ pmu->n_active++;
+ if (pmu->n_active == 1)
+ rapl_start_hrtimer(pmu);
+}
+
+static void rapl_pmu_event_start(struct perf_event *event, int mode)
+{
+ struct rapl_pmu *pmu = __get_cpu_var(rapl_pmu);
+ unsigned long flags;
+
+ spin_lock_irqsave(&pmu->lock, flags);
+ __rapl_pmu_event_start(pmu, event);
+ spin_unlock_irqrestore(&pmu->lock, flags);
+}
+
+static void rapl_pmu_event_stop(struct perf_event *event, int mode)
+{
+ struct rapl_pmu *pmu = __get_cpu_var(rapl_pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ unsigned long flags;
+
+ spin_lock_irqsave(&pmu->lock, flags);
+
+ /* mark event as deactivated and stopped */
+ if (!(hwc->state & PERF_HES_STOPPED)) {
+ WARN_ON_ONCE(pmu->n_active <= 0);
+ pmu->n_active--;
+ if (pmu->n_active == 0)
+ rapl_stop_hrtimer(pmu);
+
+ list_del(&event->active_entry);
+
+ WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+ hwc->state |= PERF_HES_STOPPED;
+ }
+
+ /* check if update of sw counter is necessary */
+ if ((mode & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+ /*
+ * Drain the remaining delta count out of a event
+ * that we are disabling:
+ */
+ rapl_event_update(event);
+ hwc->state |= PERF_HES_UPTODATE;
+ }
+
+ spin_unlock_irqrestore(&pmu->lock, flags);
+}
+
+static int rapl_pmu_event_add(struct perf_event *event, int mode)
+{
+ struct rapl_pmu *pmu = __get_cpu_var(rapl_pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ unsigned long flags;
+
+ spin_lock_irqsave(&pmu->lock, flags);
+
+ hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+
+ if (mode & PERF_EF_START)
+ __rapl_pmu_event_start(pmu, event);
+
+ spin_unlock_irqrestore(&pmu->lock, flags);
+
+ return 0;
+}
+
+static void rapl_pmu_event_del(struct perf_event *event, int flags)
+{
+ rapl_pmu_event_stop(event, PERF_EF_UPDATE);
+}
+
+static int rapl_pmu_event_init(struct perf_event *event)
+{
+ u64 cfg = event->attr.config & RAPL_EVENT_MASK;
+ int bit, msr, ret = 0;
+
+ /* only look at RAPL events */
+ if (event->attr.type != rapl_pmu_class.type)
+ return -ENOENT;
+
+ /* check only supported bits are set */
+ if (event->attr.config & ~RAPL_EVENT_MASK)
+ return -EINVAL;
+
+ /*
+ * check event is known (determines counter)
+ */
+ switch (cfg) {
+ case INTEL_RAPL_PP0:
+ bit = RAPL_IDX_PP0_NRG_STAT;
+ msr = MSR_PP0_ENERGY_STATUS;
+ break;
+ case INTEL_RAPL_PKG:
+ bit = RAPL_IDX_PKG_NRG_STAT;
+ msr = MSR_PKG_ENERGY_STATUS;
+ break;
+ case INTEL_RAPL_RAM:
+ bit = RAPL_IDX_RAM_NRG_STAT;
+ msr = MSR_DRAM_ENERGY_STATUS;
+ break;
+ case INTEL_RAPL_PP1:
+ bit = RAPL_IDX_PP1_NRG_STAT;
+ msr = MSR_PP1_ENERGY_STATUS;
+ break;
+ default:
+ return -EINVAL;
+ }
+ /* check event supported */
+ if (!(rapl_cntr_mask & (1 << bit)))
+ return -EINVAL;
+
+ /* unsupported modes and filters */
+ if (event->attr.exclude_user ||
+ event->attr.exclude_kernel ||
+ event->attr.exclude_hv ||
+ event->attr.exclude_idle ||
+ event->attr.exclude_host ||
+ event->attr.exclude_guest ||
+ event->attr.sample_period) /* no sampling */
+ return -EINVAL;
+
+ /* must be done before validate_group */
+ event->hw.event_base = msr;
+ event->hw.config = cfg;
+ event->hw.idx = bit;
+
+ return ret;
+}
+
+static void rapl_pmu_event_read(struct perf_event *event)
+{
+ rapl_event_update(event);
+}
+
+static ssize_t rapl_get_attr_cpumask(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ int n = cpulist_scnprintf(buf, PAGE_SIZE - 2, &rapl_cpu_mask);
+
+ buf[n++] = '\n';
+ buf[n] = '\0';
+ return n;
+}
+
+static DEVICE_ATTR(cpumask, S_IRUGO, rapl_get_attr_cpumask, NULL);
+
+static struct attribute *rapl_pmu_attrs[] = {
+ &dev_attr_cpumask.attr,
+ NULL,
+};
+
+static struct attribute_group rapl_pmu_attr_group = {
+ .attrs = rapl_pmu_attrs,
+};
+
+EVENT_ATTR_STR(energy-cores, rapl_cores, "event=0x01");
+EVENT_ATTR_STR(energy-pkg , rapl_pkg, "event=0x02");
+EVENT_ATTR_STR(energy-ram , rapl_ram, "event=0x03");
+EVENT_ATTR_STR(energy-gpu , rapl_gpu, "event=0x04");
+
+EVENT_ATTR_STR(energy-cores.unit, rapl_cores_unit, "Joules");
+EVENT_ATTR_STR(energy-pkg.unit , rapl_pkg_unit, "Joules");
+EVENT_ATTR_STR(energy-ram.unit , rapl_ram_unit, "Joules");
+EVENT_ATTR_STR(energy-gpu.unit , rapl_gpu_unit, "Joules");
+
+/*
+ * we compute in 0.23 nJ increments regardless of MSR
+ */
+EVENT_ATTR_STR(energy-cores.scale, rapl_cores_scale, "2.3283064365386962890625e-10");
+EVENT_ATTR_STR(energy-pkg.scale, rapl_pkg_scale, "2.3283064365386962890625e-10");
+EVENT_ATTR_STR(energy-ram.scale, rapl_ram_scale, "2.3283064365386962890625e-10");
+EVENT_ATTR_STR(energy-gpu.scale, rapl_gpu_scale, "2.3283064365386962890625e-10");
+
+static struct attribute *rapl_events_srv_attr[] = {
+ EVENT_PTR(rapl_cores),
+ EVENT_PTR(rapl_pkg),
+ EVENT_PTR(rapl_ram),
+
+ EVENT_PTR(rapl_cores_unit),
+ EVENT_PTR(rapl_pkg_unit),
+ EVENT_PTR(rapl_ram_unit),
+
+ EVENT_PTR(rapl_cores_scale),
+ EVENT_PTR(rapl_pkg_scale),
+ EVENT_PTR(rapl_ram_scale),
+ NULL,
+};
+
+static struct attribute *rapl_events_cln_attr[] = {
+ EVENT_PTR(rapl_cores),
+ EVENT_PTR(rapl_pkg),
+ EVENT_PTR(rapl_gpu),
+
+ EVENT_PTR(rapl_cores_unit),
+ EVENT_PTR(rapl_pkg_unit),
+ EVENT_PTR(rapl_gpu_unit),
+
+ EVENT_PTR(rapl_cores_scale),
+ EVENT_PTR(rapl_pkg_scale),
+ EVENT_PTR(rapl_gpu_scale),
+ NULL,
+};
+
+static struct attribute_group rapl_pmu_events_group = {
+ .name = "events",
+ .attrs = NULL, /* patched at runtime */
+};
+
+DEFINE_RAPL_FORMAT_ATTR(event, event, "config:0-7");
+static struct attribute *rapl_formats_attr[] = {
+ &format_attr_event.attr,
+ NULL,
+};
+
+static struct attribute_group rapl_pmu_format_group = {
+ .name = "format",
+ .attrs = rapl_formats_attr,
+};
+
+const struct attribute_group *rapl_attr_groups[] = {
+ &rapl_pmu_attr_group,
+ &rapl_pmu_format_group,
+ &rapl_pmu_events_group,
+ NULL,
+};
+
+static struct pmu rapl_pmu_class = {
+ .attr_groups = rapl_attr_groups,
+ .task_ctx_nr = perf_invalid_context, /* system-wide only */
+ .event_init = rapl_pmu_event_init,
+ .add = rapl_pmu_event_add, /* must have */
+ .del = rapl_pmu_event_del, /* must have */
+ .start = rapl_pmu_event_start,
+ .stop = rapl_pmu_event_stop,
+ .read = rapl_pmu_event_read,
+};
+
+static void rapl_cpu_exit(int cpu)
+{
+ struct rapl_pmu *pmu = per_cpu(rapl_pmu, cpu);
+ int i, phys_id = topology_physical_package_id(cpu);
+ int target = -1;
+
+ /* find a new cpu on same package */
+ for_each_online_cpu(i) {
+ if (i == cpu)
+ continue;
+ if (phys_id == topology_physical_package_id(i)) {
+ target = i;
+ break;
+ }
+ }
+ /*
+ * clear cpu from cpumask
+ * if was set in cpumask and still some cpu on package,
+ * then move to new cpu
+ */
+ if (cpumask_test_and_clear_cpu(cpu, &rapl_cpu_mask) && target >= 0)
+ cpumask_set_cpu(target, &rapl_cpu_mask);
+
+ WARN_ON(cpumask_empty(&rapl_cpu_mask));
+ /*
+ * migrate events and context to new cpu
+ */
+ if (target >= 0)
+ perf_pmu_migrate_context(pmu->pmu, cpu, target);
+
+ /* cancel overflow polling timer for CPU */
+ rapl_stop_hrtimer(pmu);
+}
+
+static void rapl_cpu_init(int cpu)
+{
+ int i, phys_id = topology_physical_package_id(cpu);
+
+ /* check if phys_is is already covered */
+ for_each_cpu(i, &rapl_cpu_mask) {
+ if (phys_id == topology_physical_package_id(i))
+ return;
+ }
+ /* was not found, so add it */
+ cpumask_set_cpu(cpu, &rapl_cpu_mask);
+}
+
+static int rapl_cpu_prepare(int cpu)
+{
+ struct rapl_pmu *pmu = per_cpu(rapl_pmu, cpu);
+ int phys_id = topology_physical_package_id(cpu);
+ u64 ms;
+
+ if (pmu)
+ return 0;
+
+ if (phys_id < 0)
+ return -1;
+
+ pmu = kzalloc_node(sizeof(*pmu), GFP_KERNEL, cpu_to_node(cpu));
+ if (!pmu)
+ return -1;
+
+ spin_lock_init(&pmu->lock);
+
+ INIT_LIST_HEAD(&pmu->active_list);
+
+ /*
+ * grab power unit as: 1/2^unit Joules
+ *
+ * we cache in local PMU instance
+ */
+ rdmsrl(MSR_RAPL_POWER_UNIT, pmu->hw_unit);
+ pmu->hw_unit = (pmu->hw_unit >> 8) & 0x1FULL;
+ pmu->pmu = &rapl_pmu_class;
+
+ /*
+ * use reference of 200W for scaling the timeout
+ * to avoid missing counter overflows.
+ * 200W = 200 Joules/sec
+ * divide interval by 2 to avoid lockstep (2 * 100)
+ * if hw unit is 32, then we use 2 ms 1/200/2
+ */
+ if (pmu->hw_unit < 32)
+ ms = (1000 / (2 * 100)) * (1ULL << (32 - pmu->hw_unit - 1));
+ else
+ ms = 2;
+
+ pmu->timer_interval = ms_to_ktime(ms);
+
+ rapl_hrtimer_init(pmu);
+
+ /* set RAPL pmu for this cpu for now */
+ per_cpu(rapl_pmu, cpu) = pmu;
+ per_cpu(rapl_pmu_to_free, cpu) = NULL;
+
+ return 0;
+}
+
+static void rapl_cpu_kfree(int cpu)
+{
+ struct rapl_pmu *pmu = per_cpu(rapl_pmu_to_free, cpu);
+
+ kfree(pmu);
+
+ per_cpu(rapl_pmu_to_free, cpu) = NULL;
+}
+
+static int rapl_cpu_dying(int cpu)
+{
+ struct rapl_pmu *pmu = per_cpu(rapl_pmu, cpu);
+
+ if (!pmu)
+ return 0;
+
+ per_cpu(rapl_pmu, cpu) = NULL;
+
+ per_cpu(rapl_pmu_to_free, cpu) = pmu;
+
+ return 0;
+}
+
+static int rapl_cpu_notifier(struct notifier_block *self,
+ unsigned long action, void *hcpu)
+{
+ unsigned int cpu = (long)hcpu;
+
+ switch (action & ~CPU_TASKS_FROZEN) {
+ case CPU_UP_PREPARE:
+ rapl_cpu_prepare(cpu);
+ break;
+ case CPU_STARTING:
+ rapl_cpu_init(cpu);
+ break;
+ case CPU_UP_CANCELED:
+ case CPU_DYING:
+ rapl_cpu_dying(cpu);
+ break;
+ case CPU_ONLINE:
+ case CPU_DEAD:
+ rapl_cpu_kfree(cpu);
+ break;
+ case CPU_DOWN_PREPARE:
+ rapl_cpu_exit(cpu);
+ break;
+ default:
+ break;
+ }
+
+ return NOTIFY_OK;
+}
+
+static const struct x86_cpu_id rapl_cpu_match[] = {
+ [0] = { .vendor = X86_VENDOR_INTEL, .family = 6 },
+ [1] = {},
+};
+
+static int __init rapl_pmu_init(void)
+{
+ struct rapl_pmu *pmu;
+ int cpu, ret;
+
+ /*
+ * check for Intel processor family 6
+ */
+ if (!x86_match_cpu(rapl_cpu_match))
+ return 0;
+
+ /* check supported CPU */
+ switch (boot_cpu_data.x86_model) {
+ case 42: /* Sandy Bridge */
+ case 58: /* Ivy Bridge */
+ case 60: /* Haswell */
+ case 69: /* Haswell-Celeron */
+ rapl_cntr_mask = RAPL_IDX_CLN;
+ rapl_pmu_events_group.attrs = rapl_events_cln_attr;
+ break;
+ case 45: /* Sandy Bridge-EP */
+ case 62: /* IvyTown */
+ rapl_cntr_mask = RAPL_IDX_SRV;
+ rapl_pmu_events_group.attrs = rapl_events_srv_attr;
+ break;
+
+ default:
+ /* unsupported */
+ return 0;
+ }
+ get_online_cpus();
+
+ for_each_online_cpu(cpu) {
+ rapl_cpu_prepare(cpu);
+ rapl_cpu_init(cpu);
+ }
+
+ perf_cpu_notifier(rapl_cpu_notifier);
+
+ ret = perf_pmu_register(&rapl_pmu_class, "power", -1);
+ if (WARN_ON(ret)) {
+ pr_info("RAPL PMU detected, registration failed (%d), RAPL PMU disabled\n", ret);
+ put_online_cpus();
+ return -1;
+ }
+
+ pmu = __get_cpu_var(rapl_pmu);
+
+ pr_info("RAPL PMU detected, hw unit 2^-%d Joules,"
+ " API unit is 2^-32 Joules,"
+ " %d fixed counters"
+ " %llu ms ovfl timer\n",
+ pmu->hw_unit,
+ hweight32(rapl_cntr_mask),
+ ktime_to_ms(pmu->timer_interval));
+
+ put_online_cpus();
+
+ return 0;
+}
+device_initcall(rapl_pmu_init);
diff --git a/arch/x86/kernel/cpu/transmeta.c b/arch/x86/kernel/cpu/transmeta.c
index aa0430d69b90..3fa0e5ad86b4 100644
--- a/arch/x86/kernel/cpu/transmeta.c
+++ b/arch/x86/kernel/cpu/transmeta.c
@@ -1,6 +1,5 @@
#include <linux/kernel.h>
#include <linux/mm.h>
-#include <linux/init.h>
#include <asm/processor.h>
#include <asm/msr.h>
#include "cpu.h"
diff --git a/arch/x86/kernel/cpu/umc.c b/arch/x86/kernel/cpu/umc.c
index 75c5ad5d35cc..ef9c2a0078bd 100644
--- a/arch/x86/kernel/cpu/umc.c
+++ b/arch/x86/kernel/cpu/umc.c
@@ -1,5 +1,4 @@
#include <linux/kernel.h>
-#include <linux/init.h>
#include <asm/processor.h>
#include "cpu.h"