Patch "x86/microcode: Synchronize late microcode loading" has been added to the 4.14-stable tree

[<gregkh@linuxfoundation.org>]

This is a note to let you know that I've just added the patch titled

    x86/microcode: Synchronize late microcode loading

to the 4.14-stable tree which can be found at:
    http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
     x86-microcode-synchronize-late-microcode-loading.patch
and it can be found in the queue-4.14 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@vger.kernel.org> know about it.


>From a5321aec6412b20b5ad15db2d6b916c05349dbff Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Wed, 28 Feb 2018 11:28:46 +0100
Subject: x86/microcode: Synchronize late microcode loading

From: Ashok Raj <ashok.raj@intel.com>

commit a5321aec6412b20b5ad15db2d6b916c05349dbff upstream.

Original idea by Ashok, completely rewritten by Borislav.

Before you read any further: the early loading method is still the
preferred one and you should always do that. The following patch is
improving the late loading mechanism for long running jobs and cloud use
cases.

Gather all cores and serialize the microcode update on them by doing it
one-by-one to make the late update process as reliable as possible and
avoid potential issues caused by the microcode update.

[ Borislav: Rewrite completely. ]

Co-developed-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Tom Lendacky <thomas.lendacky@amd.com>
Tested-by: Ashok Raj <ashok.raj@intel.com>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Arjan Van De Ven <arjan.van.de.ven@intel.com>
Link: https://lkml.kernel.org/r/20180228102846.13447-8-bp@alien8.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 arch/x86/kernel/cpu/microcode/core.c |  118 +++++++++++++++++++++++++++--------
 1 file changed, 92 insertions(+), 26 deletions(-)

--- a/arch/x86/kernel/cpu/microcode/core.c
+++ b/arch/x86/kernel/cpu/microcode/core.c
@@ -22,13 +22,16 @@
 #define pr_fmt(fmt) "microcode: " fmt
 
 #include <linux/platform_device.h>
+#include <linux/stop_machine.h>
 #include <linux/syscore_ops.h>
 #include <linux/miscdevice.h>
 #include <linux/capability.h>
 #include <linux/firmware.h>
 #include <linux/kernel.h>
+#include <linux/delay.h>
 #include <linux/mutex.h>
 #include <linux/cpu.h>
+#include <linux/nmi.h>
 #include <linux/fs.h>
 #include <linux/mm.h>
 
@@ -64,6 +67,11 @@ LIST_HEAD(microcode_cache);
  */
 static DEFINE_MUTEX(microcode_mutex);
 
+/*
+ * Serialize late loading so that CPUs get updated one-by-one.
+ */
+static DEFINE_SPINLOCK(update_lock);
+
 struct ucode_cpu_info		ucode_cpu_info[NR_CPUS];
 
 struct cpu_info_ctx {
@@ -486,6 +494,19 @@ static void __exit microcode_dev_exit(vo
 /* fake device for request_firmware */
 static struct platform_device	*microcode_pdev;
 
+/*
+ * Late loading dance. Why the heavy-handed stomp_machine effort?
+ *
+ * - HT siblings must be idle and not execute other code while the other sibling
+ *   is loading microcode in order to avoid any negative interactions caused by
+ *   the loading.
+ *
+ * - In addition, microcode update on the cores must be serialized until this
+ *   requirement can be relaxed in the future. Right now, this is conservative
+ *   and good.
+ */
+#define SPINUNIT 100 /* 100 nsec */
+
 static int check_online_cpus(void)
 {
 	if (num_online_cpus() == num_present_cpus())
@@ -496,23 +517,85 @@ static int check_online_cpus(void)
 	return -EINVAL;
 }
 
-static enum ucode_state reload_for_cpu(int cpu)
+static atomic_t late_cpus;
+
+/*
+ * Returns:
+ * < 0 - on error
+ *   0 - no update done
+ *   1 - microcode was updated
+ */
+static int __reload_late(void *info)
 {
-	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+	unsigned int timeout = NSEC_PER_SEC;
+	int all_cpus = num_online_cpus();
+	int cpu = smp_processor_id();
+	enum ucode_state err;
+	int ret = 0;
+
+	atomic_dec(&late_cpus);
+
+	/*
+	 * Wait for all CPUs to arrive. A load will not be attempted unless all
+	 * CPUs show up.
+	 * */
+	while (atomic_read(&late_cpus)) {
+		if (timeout < SPINUNIT) {
+			pr_err("Timeout while waiting for CPUs rendezvous, remaining: %d\n",
+				atomic_read(&late_cpus));
+			return -1;
+		}
+
+		ndelay(SPINUNIT);
+		timeout -= SPINUNIT;
+
+		touch_nmi_watchdog();
+	}
+
+	spin_lock(&update_lock);
+	apply_microcode_local(&err);
+	spin_unlock(&update_lock);
+
+	if (err > UCODE_NFOUND) {
+		pr_warn("Error reloading microcode on CPU %d\n", cpu);
+		ret = -1;
+	} else if (err == UCODE_UPDATED) {
+		ret = 1;
+	}
 
-	if (!uci->valid)
-		return UCODE_OK;
+	atomic_inc(&late_cpus);
 
-	return apply_microcode_on_target(cpu);
+	while (atomic_read(&late_cpus) != all_cpus)
+		cpu_relax();
+
+	return ret;
+}
+
+/*
+ * Reload microcode late on all CPUs. Wait for a sec until they
+ * all gather together.
+ */
+static int microcode_reload_late(void)
+{
+	int ret;
+
+	atomic_set(&late_cpus, num_online_cpus());
+
+	ret = stop_machine_cpuslocked(__reload_late, NULL, cpu_online_mask);
+	if (ret < 0)
+		return ret;
+	else if (ret > 0)
+		microcode_check();
+
+	return ret;
 }
 
 static ssize_t reload_store(struct device *dev,
 			    struct device_attribute *attr,
 			    const char *buf, size_t size)
 {
-	int cpu, bsp = boot_cpu_data.cpu_index;
 	enum ucode_state tmp_ret = UCODE_OK;
-	bool do_callback = false;
+	int bsp = boot_cpu_data.cpu_index;
 	unsigned long val;
 	ssize_t ret = 0;
 
@@ -534,30 +617,13 @@ static ssize_t reload_store(struct devic
 		goto put;
 
 	mutex_lock(&microcode_mutex);
-
-	for_each_online_cpu(cpu) {
-		tmp_ret = reload_for_cpu(cpu);
-		if (tmp_ret > UCODE_NFOUND) {
-			pr_warn("Error reloading microcode on CPU %d\n", cpu);
-
-			/* set retval for the first encountered reload error */
-			if (!ret)
-				ret = -EINVAL;
-		}
-
-		if (tmp_ret == UCODE_UPDATED)
-			do_callback = true;
-	}
-
-	if (!ret && do_callback)
-		microcode_check();
-
+	ret = microcode_reload_late();
 	mutex_unlock(&microcode_mutex);
 
 put:
 	put_online_cpus();
 
-	if (!ret)
+	if (ret >= 0)
 		ret = size;
 
 	return ret;


Patches currently in stable-queue which might be from ashok.raj@intel.com are

queue-4.14/x86-microcode-intel-check-microcode-revision-before-updating-sibling-threads.patch
queue-4.14/x86-microcode-attempt-late-loading-only-when-new-microcode-is-present.patch
queue-4.14/x86-microcode-propagate-return-value-from-updating-functions.patch
queue-4.14/x86-cpu-check-cpu-feature-bits-after-microcode-upgrade.patch
queue-4.14/x86-microcode-intel-writeback-and-invalidate-caches-before-updating-microcode.patch
queue-4.14/x86-microcode-intel-look-into-the-patch-cache-first.patch
queue-4.14/x86-microcode-request-microcode-on-the-bsp.patch
queue-4.14/x86-microcode-get-rid-of-struct-apply_microcode_ctx.patch
queue-4.14/x86-microcode-fix-cpu-synchronization-routine.patch
queue-4.14/x86-microcode-synchronize-late-microcode-loading.patch
queue-4.14/x86-microcode-do-not-upload-microcode-if-cpus-are-offline.patch
queue-4.14/x86-cpu-add-a-microcode-loader-callback.patch