Merge branch 'x86/nmi' into x86/core, to merge dependent commits

Prepare to resolve conflicts with an upstream series of fixes that conflict
with pending x86 changes:

  6f5bf947bab0 Merge tag 'its-for-linus-20250509' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Signed-off-by: Ingo Molnar <mingo@kernel.org>

7 files changed, 121 insertions, 107 deletions

diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h
index f677382093f3..79d88d12c8fb 100644
--- a/arch/x86/include/asm/nmi.h
+++ b/arch/x86/include/asm/nmi.h
@@ -14,12 +14,26 @@ extern void release_perfctr_nmi(unsigned int);
 extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
-extern int unknown_nmi_panic;
-
 #endif /* CONFIG_X86_LOCAL_APIC */
 
+extern int unknown_nmi_panic;
+extern int panic_on_unrecovered_nmi;
+extern int panic_on_io_nmi;
+
+/* NMI handler flags */
 #define NMI_FLAG_FIRST	1
 
+/**
+ * enum - NMI types.
+ * @NMI_LOCAL:    Local NMI, CPU-specific NMI generated by the Local APIC.
+ * @NMI_UNKNOWN:  Unknown NMI, the source of the NMI may not be identified.
+ * @NMI_SERR:     System Error NMI, typically triggered by PCI errors.
+ * @NMI_IO_CHECK: I/O Check NMI, related to I/O errors.
+ * @NMI_MAX:      Maximum value for NMI types.
+ *
+ * NMI types are used to categorize NMIs and to dispatch them to the
+ * appropriate handler.
+ */
 enum {
 	NMI_LOCAL=0,
 	NMI_UNKNOWN,
@@ -28,6 +42,7 @@ enum {
 	NMI_MAX
 };
 
+/* NMI handler return values */
 #define NMI_DONE	0
 #define NMI_HANDLED	1
 
@@ -41,6 +56,25 @@ struct nmiaction {
 	const char		*name;
 };
 
+/**
+ * register_nmi_handler - Register a handler for a specific NMI type
+ * @t:    NMI type (e.g. NMI_LOCAL)
+ * @fn:   The NMI handler
+ * @fg:   Flags associated with the NMI handler
+ * @n:    Name of the NMI handler
+ * @init: Optional __init* attributes for struct nmiaction
+ *
+ * Adds the provided handler to the list of handlers for the specified
+ * NMI type. Handlers flagged with NMI_FLAG_FIRST would be executed first.
+ *
+ * Sometimes the source of an NMI can't be reliably determined which
+ * results in an NMI being tagged as "unknown". Register an additional
+ * handler using the NMI type - NMI_UNKNOWN to handle such cases. The
+ * caller would get one last chance to assume responsibility for the
+ * NMI.
+ *
+ * Return: 0 on success, or an error code on failure.
+ */
 #define register_nmi_handler(t, fn, fg, n, init...)	\
 ({							\
 	static struct nmiaction init fn##_na = {	\
@@ -54,7 +88,16 @@ struct nmiaction {
 
 int __register_nmi_handler(unsigned int, struct nmiaction *);
 
-void unregister_nmi_handler(unsigned int, const char *);
+/**
+ * unregister_nmi_handler - Unregister a handler for a specific NMI type
+ * @type: NMI type (e.g. NMI_LOCAL)
+ * @name: Name of the NMI handler used during registration
+ *
+ * Removes the handler associated with the specified NMI type from the
+ * NMI handler list. The "name" is used as a lookup key to identify the
+ * handler.
+ */
+void unregister_nmi_handler(unsigned int type, const char *name);
 
 void set_emergency_nmi_handler(unsigned int type, nmi_handler_t handler);
 
diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h
index 213cf5379a5a..36698cc9fb44 100644
--- a/arch/x86/include/asm/x86_init.h
+++ b/arch/x86/include/asm/x86_init.h
@@ -292,6 +292,7 @@ struct x86_hyper_runtime {
  * @set_wallclock:		set time back to HW clock
  * @is_untracked_pat_range	exclude from PAT logic
  * @nmi_init			enable NMI on cpus
+ * @get_nmi_reason		get the reason an NMI was received
  * @save_sched_clock_state:	save state for sched_clock() on suspend
  * @restore_sched_clock_state:	restore state for sched_clock() on resume
  * @apic_post_init:		adjust apic if needed
diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
index c6fefd4585f8..71ee20102a8a 100644
--- a/arch/x86/kernel/dumpstack.c
+++ b/arch/x86/kernel/dumpstack.c
@@ -23,8 +23,6 @@
 #include <asm/stacktrace.h>
 #include <asm/unwind.h>
 
-int panic_on_unrecovered_nmi;
-int panic_on_io_nmi;
 static int die_counter;
 
 static struct pt_regs exec_summary_regs;
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index 9a95d00f1423..be93ec7255bf 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -49,27 +49,20 @@ struct nmi_desc {
 	struct list_head head;
 };
 
-static struct nmi_desc nmi_desc[NMI_MAX] = 
-{
-	{
-		.lock = __RAW_SPIN_LOCK_UNLOCKED(&nmi_desc[0].lock),
-		.head = LIST_HEAD_INIT(nmi_desc[0].head),
-	},
-	{
-		.lock = __RAW_SPIN_LOCK_UNLOCKED(&nmi_desc[1].lock),
-		.head = LIST_HEAD_INIT(nmi_desc[1].head),
-	},
-	{
-		.lock = __RAW_SPIN_LOCK_UNLOCKED(&nmi_desc[2].lock),
-		.head = LIST_HEAD_INIT(nmi_desc[2].head),
-	},
-	{
-		.lock = __RAW_SPIN_LOCK_UNLOCKED(&nmi_desc[3].lock),
-		.head = LIST_HEAD_INIT(nmi_desc[3].head),
-	},
+#define NMI_DESC_INIT(type) { \
+	.lock = __RAW_SPIN_LOCK_UNLOCKED(&nmi_desc[type].lock), \
+	.head = LIST_HEAD_INIT(nmi_desc[type].head), \
+}
 
+static struct nmi_desc nmi_desc[NMI_MAX] = {
+	NMI_DESC_INIT(NMI_LOCAL),
+	NMI_DESC_INIT(NMI_UNKNOWN),
+	NMI_DESC_INIT(NMI_SERR),
+	NMI_DESC_INIT(NMI_IO_CHECK),
 };
 
+#define nmi_to_desc(type) (&nmi_desc[type])
+
 struct nmi_stats {
 	unsigned int normal;
 	unsigned int unknown;
@@ -91,6 +84,9 @@ static DEFINE_PER_CPU(struct nmi_stats, nmi_stats);
 static int ignore_nmis __read_mostly;
 
 int unknown_nmi_panic;
+int panic_on_unrecovered_nmi;
+int panic_on_io_nmi;
+
 /*
  * Prevent NMI reason port (0x61) being accessed simultaneously, can
  * only be used in NMI handler.
@@ -104,8 +100,6 @@ static int __init setup_unknown_nmi_panic(char *str)
 }
 __setup("unknown_nmi_panic", setup_unknown_nmi_panic);
 
-#define nmi_to_desc(type) (&nmi_desc[type])
-
 static u64 nmi_longest_ns = 1 * NSEC_PER_MSEC;
 
 static int __init nmi_warning_debugfs(void)
@@ -125,12 +119,12 @@ static void nmi_check_duration(struct nmiaction *action, u64 duration)
 
 	action->max_duration = duration;
 
-	remainder_ns = do_div(duration, (1000 * 1000));
-	decimal_msecs = remainder_ns / 1000;
+	/* Convert duration from nsec to msec */
+	remainder_ns = do_div(duration, NSEC_PER_MSEC);
+	decimal_msecs = remainder_ns / NSEC_PER_USEC;
 
-	printk_ratelimited(KERN_INFO
-		"INFO: NMI handler (%ps) took too long to run: %lld.%03d msecs\n",
-		action->handler, duration, decimal_msecs);
+	pr_info_ratelimited("INFO: NMI handler (%ps) took too long to run: %lld.%03d msecs\n",
+			    action->handler, duration, decimal_msecs);
 }
 
 static int nmi_handle(unsigned int type, struct pt_regs *regs)
@@ -333,10 +327,9 @@ unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
 	int handled;
 
 	/*
-	 * Use 'false' as back-to-back NMIs are dealt with one level up.
-	 * Of course this makes having multiple 'unknown' handlers useless
-	 * as only the first one is ever run (unless it can actually determine
-	 * if it caused the NMI)
+	 * As a last resort, let the "unknown" handlers make a
+	 * best-effort attempt to figure out if they can claim
+	 * responsibility for this Unknown NMI.
 	 */
 	handled = nmi_handle(NMI_UNKNOWN, regs);
 	if (handled) {
@@ -366,17 +359,18 @@ static noinstr void default_do_nmi(struct pt_regs *regs)
 	bool b2b = false;
 
 	/*
-	 * CPU-specific NMI must be processed before non-CPU-specific
-	 * NMI, otherwise we may lose it, because the CPU-specific
-	 * NMI can not be detected/processed on other CPUs.
-	 */
-
-	/*
-	 * Back-to-back NMIs are interesting because they can either
-	 * be two NMI or more than two NMIs (any thing over two is dropped
-	 * due to NMI being edge-triggered).  If this is the second half
-	 * of the back-to-back NMI, assume we dropped things and process
-	 * more handlers.  Otherwise reset the 'swallow' NMI behaviour
+	 * Back-to-back NMIs are detected by comparing the RIP of the
+	 * current NMI with that of the previous NMI. If it is the same,
+	 * it is assumed that the CPU did not have a chance to jump back
+	 * into a non-NMI context and execute code in between the two
+	 * NMIs.
+	 *
+	 * They are interesting because even if there are more than two,
+	 * only a maximum of two can be detected (anything over two is
+	 * dropped due to NMI being edge-triggered). If this is the
+	 * second half of the back-to-back NMI, assume we dropped things
+	 * and process more handlers. Otherwise, reset the 'swallow' NMI
+	 * behavior.
 	 */
 	if (regs->ip == __this_cpu_read(last_nmi_rip))
 		b2b = true;
@@ -390,6 +384,11 @@ static noinstr void default_do_nmi(struct pt_regs *regs)
 	if (microcode_nmi_handler_enabled() && microcode_nmi_handler())
 		goto out;
 
+	/*
+	 * CPU-specific NMI must be processed before non-CPU-specific
+	 * NMI, otherwise we may lose it, because the CPU-specific
+	 * NMI can not be detected/processed on other CPUs.
+	 */
 	handled = nmi_handle(NMI_LOCAL, regs);
 	__this_cpu_add(nmi_stats.normal, handled);
 	if (handled) {
@@ -426,13 +425,14 @@ static noinstr void default_do_nmi(struct pt_regs *regs)
 			pci_serr_error(reason, regs);
 		else if (reason & NMI_REASON_IOCHK)
 			io_check_error(reason, regs);
-#ifdef CONFIG_X86_32
+
 		/*
 		 * Reassert NMI in case it became active
 		 * meanwhile as it's edge-triggered:
 		 */
-		reassert_nmi();
-#endif
+		if (IS_ENABLED(CONFIG_X86_32))
+			reassert_nmi();
+
 		__this_cpu_add(nmi_stats.external, 1);
 		raw_spin_unlock(&nmi_reason_lock);
 		goto out;
@@ -751,4 +751,3 @@ void local_touch_nmi(void)
 {
 	__this_cpu_write(last_nmi_rip, 0);
 }
-EXPORT_SYMBOL_GPL(local_touch_nmi);
diff --git a/arch/x86/kernel/nmi_selftest.c b/arch/x86/kernel/nmi_selftest.c
index e93a8545c74d..a010e9d062bf 100644
--- a/arch/x86/kernel/nmi_selftest.c
+++ b/arch/x86/kernel/nmi_selftest.c
@@ -1,7 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
 /*
- * arch/x86/kernel/nmi-selftest.c
- *
  * Testsuite for NMI: IPIs
  *
  * Started by Don Zickus:
@@ -30,7 +28,6 @@ static DECLARE_BITMAP(nmi_ipi_mask, NR_CPUS) __initdata;
 
 static int __initdata testcase_total;
 static int __initdata testcase_successes;
-static int __initdata expected_testcase_failures;
 static int __initdata unexpected_testcase_failures;
 static int __initdata unexpected_testcase_unknowns;
 
@@ -120,26 +117,22 @@ static void __init dotest(void (*testcase_fn)(void), int expected)
 		unexpected_testcase_failures++;
 
 		if (nmi_fail == FAILURE)
-			printk(KERN_CONT "FAILED |");
+			pr_cont("FAILED |");
 		else if (nmi_fail == TIMEOUT)
-			printk(KERN_CONT "TIMEOUT|");
+			pr_cont("TIMEOUT|");
 		else
-			printk(KERN_CONT "ERROR  |");
+			pr_cont("ERROR  |");
 		dump_stack();
 	} else {
 		testcase_successes++;
-		printk(KERN_CONT "  ok  |");
+		pr_cont("  ok  |");
 	}
-	testcase_total++;
+	pr_cont("\n");
 
+	testcase_total++;
 	reset_nmi();
 }
 
-static inline void __init print_testname(const char *testname)
-{
-	printk("%12s:", testname);
-}
-
 void __init nmi_selftest(void)
 {
 	init_nmi_testsuite();
@@ -147,38 +140,25 @@ void __init nmi_selftest(void)
         /*
 	 * Run the testsuite:
 	 */
-	printk("----------------\n");
-	printk("| NMI testsuite:\n");
-	printk("--------------------\n");
+	pr_info("----------------\n");
+	pr_info("| NMI testsuite:\n");
+	pr_info("--------------------\n");
 
-	print_testname("remote IPI");
+	pr_info("%12s:", "remote IPI");
 	dotest(remote_ipi, SUCCESS);
-	printk(KERN_CONT "\n");
-	print_testname("local IPI");
+
+	pr_info("%12s:", "local IPI");
 	dotest(local_ipi, SUCCESS);
-	printk(KERN_CONT "\n");
 
 	cleanup_nmi_testsuite();
 
+	pr_info("--------------------\n");
 	if (unexpected_testcase_failures) {
-		printk("--------------------\n");
-		printk("BUG: %3d unexpected failures (out of %3d) - debugging disabled! |\n",
+		pr_info("BUG: %3d unexpected failures (out of %3d) - debugging disabled! |\n",
 			unexpected_testcase_failures, testcase_total);
-		printk("-----------------------------------------------------------------\n");
-	} else if (expected_testcase_failures && testcase_successes) {
-		printk("--------------------\n");
-		printk("%3d out of %3d testcases failed, as expected. |\n",
-			expected_testcase_failures, testcase_total);
-		printk("----------------------------------------------------\n");
-	} else if (expected_testcase_failures && !testcase_successes) {
-		printk("--------------------\n");
-		printk("All %3d testcases failed, as expected. |\n",
-			expected_testcase_failures);
-		printk("----------------------------------------\n");
 	} else {
-		printk("--------------------\n");
-		printk("Good, all %3d testcases passed! |\n",
+		pr_info("Good, all %3d testcases passed! |\n",
 			testcase_successes);
-		printk("---------------------------------\n");
 	}
+	pr_info("-----------------------------------------------------------------\n");
 }
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index e0cf1595a0ab..7d9ed79a93c0 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -11,6 +11,7 @@
 #include <linux/crash_dump.h>
 #include <linux/dma-map-ops.h>
 #include <linux/efi.h>
+#include <linux/hugetlb.h>
 #include <linux/ima.h>
 #include <linux/init_ohci1394_dma.h>
 #include <linux/initrd.h>
@@ -18,21 +19,19 @@
 #include <linux/memblock.h>
 #include <linux/panic_notifier.h>
 #include <linux/pci.h>
+#include <linux/random.h>
 #include <linux/root_dev.h>
-#include <linux/hugetlb.h>
-#include <linux/tboot.h>
-#include <linux/usb/xhci-dbgp.h>
 #include <linux/static_call.h>
 #include <linux/swiotlb.h>
-#include <linux/random.h>
+#include <linux/tboot.h>
+#include <linux/usb/xhci-dbgp.h>
+#include <linux/vmalloc.h>
 
 #include <uapi/linux/mount.h>
 
 #include <xen/xen.h>
 
 #include <asm/apic.h>
-#include <asm/efi.h>
-#include <asm/numa.h>
 #include <asm/bios_ebda.h>
 #include <asm/bugs.h>
 #include <asm/cacheinfo.h>
@@ -47,18 +46,16 @@
 #include <asm/mce.h>
 #include <asm/memtype.h>
 #include <asm/mtrr.h>
-#include <asm/realmode.h>
+#include <asm/nmi.h>
+#include <asm/numa.h>
 #include <asm/olpc_ofw.h>
 #include <asm/pci-direct.h>
 #include <asm/prom.h>
 #include <asm/proto.h>
+#include <asm/realmode.h>
 #include <asm/thermal.h>
 #include <asm/unwind.h>
 #include <asm/vsyscall.h>
-#include <linux/vmalloc.h>
-#if defined(CONFIG_X86_LOCAL_APIC)
-#include <asm/nmi.h>
-#endif
 
 /*
  * max_low_pfn_mapped: highest directly mapped pfn < 4 GB
@@ -152,6 +149,13 @@ int bootloader_type, bootloader_version;
 
 static const struct ctl_table x86_sysctl_table[] = {
 	{
+		.procname       = "unknown_nmi_panic",
+		.data           = &unknown_nmi_panic,
+		.maxlen         = sizeof(int),
+		.mode           = 0644,
+		.proc_handler   = proc_dointvec,
+	},
+	{
 		.procname	= "panic_on_unrecovered_nmi",
 		.data		= &panic_on_unrecovered_nmi,
 		.maxlen		= sizeof(int),
@@ -186,15 +190,6 @@ static const struct ctl_table x86_sysctl_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec,
 	},
-#if defined(CONFIG_X86_LOCAL_APIC)
-	{
-		.procname       = "unknown_nmi_panic",
-		.data           = &unknown_nmi_panic,
-		.maxlen         = sizeof(int),
-		.mode           = 0644,
-		.proc_handler   = proc_dointvec,
-	},
-#endif
 #if defined(CONFIG_ACPI_SLEEP)
 	{
 		.procname	= "acpi_video_flags",
diff --git a/include/linux/panic.h b/include/linux/panic.h
index 2494d51707ef..4adc65766935 100644
--- a/include/linux/panic.h
+++ b/include/linux/panic.h
@@ -20,8 +20,6 @@ extern bool panic_triggering_all_cpu_backtrace;
 extern int panic_timeout;
 extern unsigned long panic_print;
 extern int panic_on_oops;
-extern int panic_on_unrecovered_nmi;
-extern int panic_on_io_nmi;
 extern int panic_on_warn;
 
 extern unsigned long panic_on_taint;