Skip to contentSkip to conversation

4.19.249-rc1 review

PATCH In review Greg Kroah-Hartman 8 participants 242 messages Jun 23, 2022 – Jun 25, 2022 Page 2 of 3

Subsystems (5)
lore
Share & tools
lore
author maintainer
#99 lore raw

PATCH 99/234 random: group initialization wait functions

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 5f1bb112006b104b3e2a1e1b39bbb9b2617581e6 upstream.

This pulls all of the readiness waiting-focused functions into the first labeled section.

No functional changes.

---
 drivers/char/random.c | 1015 +++++++++++++++++++++++++-------------------------
 1 file changed, 527 insertions(+), 488 deletions(-)
drivers/char/random.c +527 -488
@@ -201,126 +201,144 @@
 #include <asm/irq_regs.h>
 #include <asm/io.h>
 
-enum {
-	POOL_BITS = BLAKE2S_HASH_SIZE * 8,
-	POOL_MIN_BITS = POOL_BITS /* No point in settling for less. */
-};
-
-/*
- * Static global variables
- */
-static DECLARE_WAIT_QUEUE_HEAD(random_write_wait);
-static struct fasync_struct *fasync;
-
-static DEFINE_SPINLOCK(random_ready_list_lock);
-static LIST_HEAD(random_ready_list);
+/*********************************************************************
+ *
+ * Initialization and readiness waiting.
+ *
+ * Much of the RNG infrastructure is devoted to various dependencies
+ * being able to wait until the RNG has collected enough entropy and
+ * is ready for safe consumption.
+ *
+ *********************************************************************/
 
 /*
  * crng_init =  0 --> Uninitialized
  *		1 --> Initialized
  *		2 --> Initialized from input_pool
  *
- * crng_init is protected by primary_crng->lock, and only increases
+ * crng_init is protected by base_crng->lock, and only increases
  * its value (from 0->1->2).
  */
 static int crng_init = 0;
 #define crng_ready() (likely(crng_init > 1))
-static int crng_init_cnt = 0;
-static void process_random_ready_list(void);
-static void _get_random_bytes(void *buf, size_t nbytes);
+/* Various types of waiters for crng_init->2 transition. */
+static DECLARE_WAIT_QUEUE_HEAD(crng_init_wait);
+static struct fasync_struct *fasync;
+static DEFINE_SPINLOCK(random_ready_list_lock);
+static LIST_HEAD(random_ready_list);
 
+/* Control how we warn userspace. */
 static struct ratelimit_state unseeded_warning =
 	RATELIMIT_STATE_INIT("warn_unseeded_randomness", HZ, 3);
 static struct ratelimit_state urandom_warning =
 	RATELIMIT_STATE_INIT("warn_urandom_randomness", HZ, 3);
-
 static int ratelimit_disable __read_mostly;
-
 module_param_named(ratelimit_disable, ratelimit_disable, int, 0644);
 MODULE_PARM_DESC(ratelimit_disable, "Disable random ratelimit suppression");
 
-/**********************************************************************
- *
- * OS independent entropy store.   Here are the functions which handle
- * storing entropy in an entropy pool.
+/*
+ * Returns whether or not the input pool has been seeded and thus guaranteed
+ * to supply cryptographically secure random numbers. This applies to: the
+ * /dev/urandom device, the get_random_bytes function, and the get_random_{u32,
+ * ,u64,int,long} family of functions.
  *
- **********************************************************************/
-
-static struct {
-	struct blake2s_state hash;
-	spinlock_t lock;
-	unsigned int entropy_count;
-} input_pool = {
-	.hash.h = { BLAKE2S_IV0 ^ (0x01010000 | BLAKE2S_HASH_SIZE),
-		    BLAKE2S_IV1, BLAKE2S_IV2, BLAKE2S_IV3, BLAKE2S_IV4,
-		    BLAKE2S_IV5, BLAKE2S_IV6, BLAKE2S_IV7 },
-	.hash.outlen = BLAKE2S_HASH_SIZE,
-	.lock = __SPIN_LOCK_UNLOCKED(input_pool.lock),
-};
-
-static void extract_entropy(void *buf, size_t nbytes);
-static bool drain_entropy(void *buf, size_t nbytes);
+ * Returns: true if the input pool has been seeded.
+ *          false if the input pool has not been seeded.
+ */
+bool rng_is_initialized(void)
+{
+	return crng_ready();
+}
+EXPORT_SYMBOL(rng_is_initialized);
 
-static void crng_reseed(void);
+/* Used by wait_for_random_bytes(), and considered an entropy collector, below. */
+static void try_to_generate_entropy(void);
 
 /*
- * This function adds bytes into the entropy "pool".  It does not
- * update the entropy estimate.  The caller should call
- * credit_entropy_bits if this is appropriate.
+ * Wait for the input pool to be seeded and thus guaranteed to supply
+ * cryptographically secure random numbers. This applies to: the /dev/urandom
+ * device, the get_random_bytes function, and the get_random_{u32,u64,int,long}
+ * family of functions. Using any of these functions without first calling
+ * this function forfeits the guarantee of security.
+ *
+ * Returns: 0 if the input pool has been seeded.
+ *          -ERESTARTSYS if the function was interrupted by a signal.
  */
-static void _mix_pool_bytes(const void *in, size_t nbytes)
+int wait_for_random_bytes(void)
 {
-	blake2s_update(&input_pool.hash, in, nbytes);
-}
+	if (likely(crng_ready()))
+		return 0;
 
-static void mix_pool_bytes(const void *in, size_t nbytes)
-{
-	unsigned long flags;
+	do {
+		int ret;
+		ret = wait_event_interruptible_timeout(crng_init_wait, crng_ready(), HZ);
+		if (ret)
+			return ret > 0 ? 0 : ret;
 
-	spin_lock_irqsave(&input_pool.lock, flags);
-	_mix_pool_bytes(in, nbytes);
-	spin_unlock_irqrestore(&input_pool.lock, flags);
-}
+		try_to_generate_entropy();
+	} while (!crng_ready());
 
-struct fast_pool {
-	union {
-		u32 pool32[4];
-		u64 pool64[2];
-	};
-	unsigned long last;
-	u16 reg_idx;
-	u8 count;
-};
+	return 0;
+}
+EXPORT_SYMBOL(wait_for_random_bytes);
 
 /*
- * This is a fast mixing routine used by the interrupt randomness
- * collector.  It's hardcoded for an 128 bit pool and assumes that any
- * locks that might be needed are taken by the caller.
+ * Add a callback function that will be invoked when the input
+ * pool is initialised.
+ *
+ * returns: 0 if callback is successfully added
+ *	    -EALREADY if pool is already initialised (callback not called)
+ *	    -ENOENT if module for callback is not alive
  */
-static void fast_mix(u32 pool[4])
+int add_random_ready_callback(struct random_ready_callback *rdy)
 {
-	u32 a = pool[0],	b = pool[1];
-	u32 c = pool[2],	d = pool[3];
+	struct module *owner;
+	unsigned long flags;
+	int err = -EALREADY;
 
-	a += b;			c += d;
-	b = rol32(b, 6);	d = rol32(d, 27);
-	d ^= a;			b ^= c;
+	if (crng_ready())
+		return err;
 
-	a += b;			c += d;
-	b = rol32(b, 16);	d = rol32(d, 14);
-	d ^= a;			b ^= c;
+	owner = rdy->owner;
+	if (!try_module_get(owner))
+		return -ENOENT;
 
-	a += b;			c += d;
-	b = rol32(b, 6);	d = rol32(d, 27);
-	d ^= a;			b ^= c;
+	spin_lock_irqsave(&random_ready_list_lock, flags);
+	if (crng_ready())
+		goto out;
 
-	a += b;			c += d;
-	b = rol32(b, 16);	d = rol32(d, 14);
-	d ^= a;			b ^= c;
+	owner = NULL;
 
-	pool[0] = a;  pool[1] = b;
-	pool[2] = c;  pool[3] = d;
+	list_add(&rdy->list, &random_ready_list);
+	err = 0;
+
+out:
+	spin_unlock_irqrestore(&random_ready_list_lock, flags);
+
+	module_put(owner);
+
+	return err;
+}
+EXPORT_SYMBOL(add_random_ready_callback);
+
+/*
+ * Delete a previously registered readiness callback function.
+ */
+void del_random_ready_callback(struct random_ready_callback *rdy)
+{
+	unsigned long flags;
+	struct module *owner = NULL;
+
+	spin_lock_irqsave(&random_ready_list_lock, flags);
+	if (!list_empty(&rdy->list)) {
+		list_del_init(&rdy->list);
+		owner = rdy->owner;
+	}
+	spin_unlock_irqrestore(&random_ready_list_lock, flags);
+
+	module_put(owner);
 }
+EXPORT_SYMBOL(del_random_ready_callback);
 
 static void process_random_ready_list(void)
 {
@@ -338,27 +356,51 @@ static void process_random_ready_list(vo
 	spin_unlock_irqrestore(&random_ready_list_lock, flags);
 }
 
-static void credit_entropy_bits(size_t nbits)
+#define warn_unseeded_randomness(previous) \
+	_warn_unseeded_randomness(__func__, (void *)_RET_IP_, (previous))
+
+static void _warn_unseeded_randomness(const char *func_name, void *caller, void **previous)
 {
-	unsigned int entropy_count, orig, add;
+#ifdef CONFIG_WARN_ALL_UNSEEDED_RANDOM
+	const bool print_once = false;
+#else
+	static bool print_once __read_mostly;
+#endif
 
-	if (!nbits)
+	if (print_once || crng_ready() ||
+	    (previous && (caller == READ_ONCE(*previous))))
 		return;
-
-	add = min_t(size_t, nbits, POOL_BITS);
-
-	do {
-		orig = READ_ONCE(input_pool.entropy_count);
-		entropy_count = min_t(unsigned int, POOL_BITS, orig + add);
-	} while (cmpxchg(&input_pool.entropy_count, orig, entropy_count) != orig);
-
-	if (crng_init < 2 && entropy_count >= POOL_MIN_BITS)
-		crng_reseed();
+	WRITE_ONCE(*previous, caller);
+#ifndef CONFIG_WARN_ALL_UNSEEDED_RANDOM
+	print_once = true;
+#endif
+	if (__ratelimit(&unseeded_warning))
+		printk_deferred(KERN_NOTICE "random: %s called from %pS with crng_init=%d\n",
+				func_name, caller, crng_init);
 }
 
+
 /*********************************************************************
  *
- * CRNG using CHACHA20
+ * Fast key erasure RNG, the "crng".
+ *
+ * These functions expand entropy from the entropy extractor into
+ * long streams for external consumption using the "fast key erasure"
+ * RNG described at <https://blog.cr.yp.to/20170723-random.html>.
+ *
+ * There are a few exported interfaces for use by other drivers:
+ *
+ *	void get_random_bytes(void *buf, size_t nbytes)
+ *	u32 get_random_u32()
+ *	u64 get_random_u64()
+ *	unsigned int get_random_int()
+ *	unsigned long get_random_long()
+ *
+ * These interfaces will return the requested number of random bytes
+ * into the given buffer or as a return value. This is equivalent to
+ * a read from /dev/urandom. The integer family of functions may be
+ * higher performance for one-off random integers, because they do a
+ * bit of buffering.
  *
  *********************************************************************/
 
@@ -385,72 +427,14 @@ static DEFINE_PER_CPU(struct crng, crngs
 	.generation = ULONG_MAX
 };
 
-static DECLARE_WAIT_QUEUE_HEAD(crng_init_wait);
+/* Used by crng_reseed() to extract a new seed from the input pool. */
+static bool drain_entropy(void *buf, size_t nbytes);
 
 /*
- * crng_fast_load() can be called by code in the interrupt service
- * path.  So we can't afford to dilly-dally. Returns the number of
- * bytes processed from cp.
+ * This extracts a new crng key from the input pool, but only if there is a
+ * sufficient amount of entropy available, in order to mitigate bruteforcing
+ * of newly added bits.
  */
-static size_t crng_fast_load(const void *cp, size_t len)
-{
-	unsigned long flags;
-	const u8 *src = (const u8 *)cp;
-	size_t ret = 0;
-
-	if (!spin_trylock_irqsave(&base_crng.lock, flags))
-		return 0;
-	if (crng_init != 0) {
-		spin_unlock_irqrestore(&base_crng.lock, flags);
-		return 0;
-	}
-	while (len > 0 && crng_init_cnt < CRNG_INIT_CNT_THRESH) {
-		base_crng.key[crng_init_cnt % sizeof(base_crng.key)] ^= *src;
-		src++; crng_init_cnt++; len--; ret++;
-	}
-	if (crng_init_cnt >= CRNG_INIT_CNT_THRESH) {
-		++base_crng.generation;
-		crng_init = 1;
-	}
-	spin_unlock_irqrestore(&base_crng.lock, flags);
-	if (crng_init == 1)
-		pr_notice("fast init done\n");
-	return ret;
-}
-
-/*
- * crng_slow_load() is called by add_device_randomness, which has two
- * attributes.  (1) We can't trust the buffer passed to it is
- * guaranteed to be unpredictable (so it might not have any entropy at
- * all), and (2) it doesn't have the performance constraints of
- * crng_fast_load().
- *
- * So, we simply hash the contents in with the current key. Finally,
- * we do *not* advance crng_init_cnt since buffer we may get may be
- * something like a fixed DMI table (for example), which might very
- * well be unique to the machine, but is otherwise unvarying.
- */
-static void crng_slow_load(const void *cp, size_t len)
-{
-	unsigned long flags;
-	struct blake2s_state hash;
-
-	blake2s_init(&hash, sizeof(base_crng.key));
-
-	if (!spin_trylock_irqsave(&base_crng.lock, flags))
-		return;
-	if (crng_init != 0) {
-		spin_unlock_irqrestore(&base_crng.lock, flags);
-		return;
-	}
-
-	blake2s_update(&hash, base_crng.key, sizeof(base_crng.key));
-	blake2s_update(&hash, cp, len);
-	blake2s_final(&hash, base_crng.key);
-
-	spin_unlock_irqrestore(&base_crng.lock, flags);
-}
-
 static void crng_reseed(void)
 {
 	unsigned long flags;
@@ -500,13 +484,11 @@ static void crng_reseed(void)
 }
 
 /*
- * The general form here is based on a "fast key erasure RNG" from
- * <https://blog.cr.yp.to/20170723-random.html>. It generates a ChaCha
- * block using the provided key, and then immediately overwites that
- * key with half the block. It returns the resultant ChaCha state to the
- * user, along with the second half of the block containing 32 bytes of
- * random data that may be used; random_data_len may not be greater than
- * 32.
+ * This generates a ChaCha block using the provided key, and then
+ * immediately overwites that key with half the block. It returns
+ * the resultant ChaCha state to the user, along with the second
+ * half of the block containing 32 bytes of random data that may
+ * be used; random_data_len may not be greater than 32.
  */
 static void crng_fast_key_erasure(u8 key[CHACHA20_KEY_SIZE],
 				  u32 chacha_state[CHACHA20_BLOCK_SIZE / sizeof(u32)],
@@ -593,6 +575,126 @@ static void crng_make_state(u32 chacha_s
 	local_irq_restore(flags);
 }
 
+/*
+ * This function is for crng_init == 0 only.
+ *
+ * crng_fast_load() can be called by code in the interrupt service
+ * path.  So we can't afford to dilly-dally. Returns the number of
+ * bytes processed from cp.
+ */
+static size_t crng_fast_load(const void *cp, size_t len)
+{
+	static int crng_init_cnt = 0;
+	unsigned long flags;
+	const u8 *src = (const u8 *)cp;
+	size_t ret = 0;
+
+	if (!spin_trylock_irqsave(&base_crng.lock, flags))
+		return 0;
+	if (crng_init != 0) {
+		spin_unlock_irqrestore(&base_crng.lock, flags);
+		return 0;
+	}
+	while (len > 0 && crng_init_cnt < CRNG_INIT_CNT_THRESH) {
+		base_crng.key[crng_init_cnt % sizeof(base_crng.key)] ^= *src;
+		src++; crng_init_cnt++; len--; ret++;
+	}
+	if (crng_init_cnt >= CRNG_INIT_CNT_THRESH) {
+		++base_crng.generation;
+		crng_init = 1;
+	}
+	spin_unlock_irqrestore(&base_crng.lock, flags);
+	if (crng_init == 1)
+		pr_notice("fast init done\n");
+	return ret;
+}
+
+/*
+ * This function is for crng_init == 0 only.
+ *
+ * crng_slow_load() is called by add_device_randomness, which has two
+ * attributes.  (1) We can't trust the buffer passed to it is
+ * guaranteed to be unpredictable (so it might not have any entropy at
+ * all), and (2) it doesn't have the performance constraints of
+ * crng_fast_load().
+ *
+ * So, we simply hash the contents in with the current key. Finally,
+ * we do *not* advance crng_init_cnt since buffer we may get may be
+ * something like a fixed DMI table (for example), which might very
+ * well be unique to the machine, but is otherwise unvarying.
+ */
+static void crng_slow_load(const void *cp, size_t len)
+{
+	unsigned long flags;
+	struct blake2s_state hash;
+
+	blake2s_init(&hash, sizeof(base_crng.key));
+
+	if (!spin_trylock_irqsave(&base_crng.lock, flags))
+		return;
+	if (crng_init != 0) {
+		spin_unlock_irqrestore(&base_crng.lock, flags);
+		return;
+	}
+
+	blake2s_update(&hash, base_crng.key, sizeof(base_crng.key));
+	blake2s_update(&hash, cp, len);
+	blake2s_final(&hash, base_crng.key);
+
+	spin_unlock_irqrestore(&base_crng.lock, flags);
+}
+
+static void _get_random_bytes(void *buf, size_t nbytes)
+{
+	u32 chacha_state[CHACHA20_BLOCK_SIZE / sizeof(u32)];
+	u8 tmp[CHACHA20_BLOCK_SIZE];
+	size_t len;
+
+	if (!nbytes)
+		return;
+
+	len = min_t(size_t, 32, nbytes);
+	crng_make_state(chacha_state, buf, len);
+	nbytes -= len;
+	buf += len;
+
+	while (nbytes) {
+		if (nbytes < CHACHA20_BLOCK_SIZE) {
+			chacha20_block(chacha_state, tmp);
+			memcpy(buf, tmp, nbytes);
+			memzero_explicit(tmp, sizeof(tmp));
+			break;
+		}
+
+		chacha20_block(chacha_state, buf);
+		if (unlikely(chacha_state[12] == 0))
+			++chacha_state[13];
+		nbytes -= CHACHA20_BLOCK_SIZE;
+		buf += CHACHA20_BLOCK_SIZE;
+	}
+
+	memzero_explicit(chacha_state, sizeof(chacha_state));
+}
+
+/*
+ * This function is the exported kernel interface.  It returns some
+ * number of good random numbers, suitable for key generation, seeding
+ * TCP sequence numbers, etc.  It does not rely on the hardware random
+ * number generator.  For random bytes direct from the hardware RNG
+ * (when available), use get_random_bytes_arch(). In order to ensure
+ * that the randomness provided by this function is okay, the function
+ * wait_for_random_bytes() should be called and return 0 at least once
+ * at any point prior.
+ */
+void get_random_bytes(void *buf, size_t nbytes)
+{
+	static void *previous;
+
+	warn_unseeded_randomness(&previous);
+	_get_random_bytes(buf, nbytes);
+}
+EXPORT_SYMBOL(get_random_bytes);
+
 static ssize_t get_random_bytes_user(void __user *buf, size_t nbytes)
 {
 	bool large_request = nbytes > 256;
@@ -640,6 +742,265 @@ static ssize_t get_random_bytes_user(voi
 	return ret;
 }
 
+/*
+ * Batched entropy returns random integers. The quality of the random
+ * number is good as /dev/urandom. In order to ensure that the randomness
+ * provided by this function is okay, the function wait_for_random_bytes()
+ * should be called and return 0 at least once at any point prior.
+ */
+struct batched_entropy {
+	union {
+		/*
+		 * We make this 1.5x a ChaCha block, so that we get the
+		 * remaining 32 bytes from fast key erasure, plus one full
+		 * block from the detached ChaCha state. We can increase
+		 * the size of this later if needed so long as we keep the
+		 * formula of (integer_blocks + 0.5) * CHACHA20_BLOCK_SIZE.
+		 */
+		u64 entropy_u64[CHACHA20_BLOCK_SIZE * 3 / (2 * sizeof(u64))];
+		u32 entropy_u32[CHACHA20_BLOCK_SIZE * 3 / (2 * sizeof(u32))];
+	};
+	unsigned long generation;
+	unsigned int position;
+};
+
+
+static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u64) = {
+	.position = UINT_MAX
+};
+
+u64 get_random_u64(void)
+{
+	u64 ret;
+	unsigned long flags;
+	struct batched_entropy *batch;
+	static void *previous;
+	unsigned long next_gen;
+
+	warn_unseeded_randomness(&previous);
+
+	local_irq_save(flags);
+	batch = raw_cpu_ptr(&batched_entropy_u64);
+
+	next_gen = READ_ONCE(base_crng.generation);
+	if (batch->position >= ARRAY_SIZE(batch->entropy_u64) ||
+	    next_gen != batch->generation) {
+		_get_random_bytes(batch->entropy_u64, sizeof(batch->entropy_u64));
+		batch->position = 0;
+		batch->generation = next_gen;
+	}
+
+	ret = batch->entropy_u64[batch->position];
+	batch->entropy_u64[batch->position] = 0;
+	++batch->position;
+	local_irq_restore(flags);
+	return ret;
+}
+EXPORT_SYMBOL(get_random_u64);
+
+static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u32) = {
+	.position = UINT_MAX
+};
+
+u32 get_random_u32(void)
+{
+	u32 ret;
+	unsigned long flags;
+	struct batched_entropy *batch;
+	static void *previous;
+	unsigned long next_gen;
+
+	warn_unseeded_randomness(&previous);
+
+	local_irq_save(flags);
+	batch = raw_cpu_ptr(&batched_entropy_u32);
+
+	next_gen = READ_ONCE(base_crng.generation);
+	if (batch->position >= ARRAY_SIZE(batch->entropy_u32) ||
+	    next_gen != batch->generation) {
+		_get_random_bytes(batch->entropy_u32, sizeof(batch->entropy_u32));
+		batch->position = 0;
+		batch->generation = next_gen;
+	}
+
+	ret = batch->entropy_u32[batch->position];
+	batch->entropy_u32[batch->position] = 0;
+	++batch->position;
+	local_irq_restore(flags);
+	return ret;
+}
+EXPORT_SYMBOL(get_random_u32);
+
+/**
+ * randomize_page - Generate a random, page aligned address
+ * @start:	The smallest acceptable address the caller will take.
+ * @range:	The size of the area, starting at @start, within which the
+ *		random address must fall.
+ *
+ * If @start + @range would overflow, @range is capped.
+ *
+ * NOTE: Historical use of randomize_range, which this replaces, presumed that
+ * @start was already page aligned.  We now align it regardless.
+ *
+ * Return: A page aligned address within [start, start + range).  On error,
+ * @start is returned.
+ */
+unsigned long randomize_page(unsigned long start, unsigned long range)
+{
+	if (!PAGE_ALIGNED(start)) {
+		range -= PAGE_ALIGN(start) - start;
+		start = PAGE_ALIGN(start);
+	}
+
+	if (start > ULONG_MAX - range)
+		range = ULONG_MAX - start;
+
+	range >>= PAGE_SHIFT;
+
+	if (range == 0)
+		return start;
+
+	return start + (get_random_long() % range << PAGE_SHIFT);
+}
+
+/*
+ * This function will use the architecture-specific hardware random
+ * number generator if it is available. It is not recommended for
+ * use. Use get_random_bytes() instead. It returns the number of
+ * bytes filled in.
+ */
+size_t __must_check get_random_bytes_arch(void *buf, size_t nbytes)
+{
+	size_t left = nbytes;
+	u8 *p = buf;
+
+	while (left) {
+		unsigned long v;
+		size_t chunk = min_t(size_t, left, sizeof(unsigned long));
+
+		if (!arch_get_random_long(&v))
+			break;
+
+		memcpy(p, &v, chunk);
+		p += chunk;
+		left -= chunk;
+	}
+
+	return nbytes - left;
+}
+EXPORT_SYMBOL(get_random_bytes_arch);
+
+enum {
+	POOL_BITS = BLAKE2S_HASH_SIZE * 8,
+	POOL_MIN_BITS = POOL_BITS /* No point in settling for less. */
+};
+
+/*
+ * Static global variables
+ */
+static DECLARE_WAIT_QUEUE_HEAD(random_write_wait);
+
+/**********************************************************************
+ *
+ * OS independent entropy store.   Here are the functions which handle
+ * storing entropy in an entropy pool.
+ *
+ **********************************************************************/
+
+static struct {
+	struct blake2s_state hash;
+	spinlock_t lock;
+	unsigned int entropy_count;
+} input_pool = {
+	.hash.h = { BLAKE2S_IV0 ^ (0x01010000 | BLAKE2S_HASH_SIZE),
+		    BLAKE2S_IV1, BLAKE2S_IV2, BLAKE2S_IV3, BLAKE2S_IV4,
+		    BLAKE2S_IV5, BLAKE2S_IV6, BLAKE2S_IV7 },
+	.hash.outlen = BLAKE2S_HASH_SIZE,
+	.lock = __SPIN_LOCK_UNLOCKED(input_pool.lock),
+};
+
+static void extract_entropy(void *buf, size_t nbytes);
+static bool drain_entropy(void *buf, size_t nbytes);
+
+static void crng_reseed(void);
+
+/*
+ * This function adds bytes into the entropy "pool".  It does not
+ * update the entropy estimate.  The caller should call
+ * credit_entropy_bits if this is appropriate.
+ */
+static void _mix_pool_bytes(const void *in, size_t nbytes)
+{
+	blake2s_update(&input_pool.hash, in, nbytes);
+}
+
+static void mix_pool_bytes(const void *in, size_t nbytes)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&input_pool.lock, flags);
+	_mix_pool_bytes(in, nbytes);
+	spin_unlock_irqrestore(&input_pool.lock, flags);
+}
+
+struct fast_pool {
+	union {
+		u32 pool32[4];
+		u64 pool64[2];
+	};
+	unsigned long last;
+	u16 reg_idx;
+	u8 count;
+};
+
+/*
+ * This is a fast mixing routine used by the interrupt randomness
+ * collector.  It's hardcoded for an 128 bit pool and assumes that any
+ * locks that might be needed are taken by the caller.
+ */
+static void fast_mix(u32 pool[4])
+{
+	u32 a = pool[0],	b = pool[1];
+	u32 c = pool[2],	d = pool[3];
+
+	a += b;			c += d;
+	b = rol32(b, 6);	d = rol32(d, 27);
+	d ^= a;			b ^= c;
+
+	a += b;			c += d;
+	b = rol32(b, 16);	d = rol32(d, 14);
+	d ^= a;			b ^= c;
+
+	a += b;			c += d;
+	b = rol32(b, 6);	d = rol32(d, 27);
+	d ^= a;			b ^= c;
+
+	a += b;			c += d;
+	b = rol32(b, 16);	d = rol32(d, 14);
+	d ^= a;			b ^= c;
+
+	pool[0] = a;  pool[1] = b;
+	pool[2] = c;  pool[3] = d;
+}
+
+static void credit_entropy_bits(size_t nbits)
+{
+	unsigned int entropy_count, orig, add;
+
+	if (!nbits)
+		return;
+
+	add = min_t(size_t, nbits, POOL_BITS);
+
+	do {
+		orig = READ_ONCE(input_pool.entropy_count);
+		entropy_count = min_t(unsigned int, POOL_BITS, orig + add);
+	} while (cmpxchg(&input_pool.entropy_count, orig, entropy_count) != orig);
+
+	if (crng_init < 2 && entropy_count >= POOL_MIN_BITS)
+		crng_reseed();
+}
+
 /*********************************************************************
  *
  * Entropy input management
@@ -907,80 +1268,6 @@ static bool drain_entropy(void *buf, siz
 	return true;
 }
 
-#define warn_unseeded_randomness(previous) \
-	_warn_unseeded_randomness(__func__, (void *)_RET_IP_, (previous))
-
-static void _warn_unseeded_randomness(const char *func_name, void *caller, void **previous)
-{
-#ifdef CONFIG_WARN_ALL_UNSEEDED_RANDOM
-	const bool print_once = false;
-#else
-	static bool print_once __read_mostly;
-#endif
-
-	if (print_once || crng_ready() ||
-	    (previous && (caller == READ_ONCE(*previous))))
-		return;
-	WRITE_ONCE(*previous, caller);
-#ifndef CONFIG_WARN_ALL_UNSEEDED_RANDOM
-	print_once = true;
-#endif
-	if (__ratelimit(&unseeded_warning))
-		printk_deferred(KERN_NOTICE "random: %s called from %pS with crng_init=%d\n",
-				func_name, caller, crng_init);
-}
-
-/*
- * This function is the exported kernel interface.  It returns some
- * number of good random numbers, suitable for key generation, seeding
- * TCP sequence numbers, etc.  It does not rely on the hardware random
- * number generator.  For random bytes direct from the hardware RNG
- * (when available), use get_random_bytes_arch(). In order to ensure
- * that the randomness provided by this function is okay, the function
- * wait_for_random_bytes() should be called and return 0 at least once
- * at any point prior.
- */
-static void _get_random_bytes(void *buf, size_t nbytes)
-{
-	u32 chacha_state[CHACHA20_BLOCK_SIZE / sizeof(u32)];
-	u8 tmp[CHACHA20_BLOCK_SIZE];
-	size_t len;
-
-	if (!nbytes)
-		return;
-
-	len = min_t(size_t, 32, nbytes);
-	crng_make_state(chacha_state, buf, len);
-	nbytes -= len;
-	buf += len;
-
-	while (nbytes) {
-		if (nbytes < CHACHA20_BLOCK_SIZE) {
-			chacha20_block(chacha_state, tmp);
-			memcpy(buf, tmp, nbytes);
-			memzero_explicit(tmp, sizeof(tmp));
-			break;
-		}
-
-		chacha20_block(chacha_state, buf);
-		if (unlikely(chacha_state[12] == 0))
-			++chacha_state[13];
-		nbytes -= CHACHA20_BLOCK_SIZE;
-		buf += CHACHA20_BLOCK_SIZE;
-	}
-
-	memzero_explicit(chacha_state, sizeof(chacha_state));
-}
-
-void get_random_bytes(void *buf, size_t nbytes)
-{
-	static void *previous;
-
-	warn_unseeded_randomness(&previous);
-	_get_random_bytes(buf, nbytes);
-}
-EXPORT_SYMBOL(get_random_bytes);
-
 /*
  * Each time the timer fires, we expect that we got an unpredictable
  * jump in the cycle counter. Even if the timer is running on another
@@ -1030,134 +1317,6 @@ static void try_to_generate_entropy(void
 	mix_pool_bytes(&stack.now, sizeof(stack.now));
 }
 
-/*
- * Wait for the urandom pool to be seeded and thus guaranteed to supply
- * cryptographically secure random numbers. This applies to: the /dev/urandom
- * device, the get_random_bytes function, and the get_random_{u32,u64,int,long}
- * family of functions. Using any of these functions without first calling
- * this function forfeits the guarantee of security.
- *
- * Returns: 0 if the urandom pool has been seeded.
- *          -ERESTARTSYS if the function was interrupted by a signal.
- */
-int wait_for_random_bytes(void)
-{
-	if (likely(crng_ready()))
-		return 0;
-
-	do {
-		int ret;
-		ret = wait_event_interruptible_timeout(crng_init_wait, crng_ready(), HZ);
-		if (ret)
-			return ret > 0 ? 0 : ret;
-
-		try_to_generate_entropy();
-	} while (!crng_ready());
-
-	return 0;
-}
-EXPORT_SYMBOL(wait_for_random_bytes);
-
-/*
- * Returns whether or not the urandom pool has been seeded and thus guaranteed
- * to supply cryptographically secure random numbers. This applies to: the
- * /dev/urandom device, the get_random_bytes function, and the get_random_{u32,
- * ,u64,int,long} family of functions.
- *
- * Returns: true if the urandom pool has been seeded.
- *          false if the urandom pool has not been seeded.
- */
-bool rng_is_initialized(void)
-{
-	return crng_ready();
-}
-EXPORT_SYMBOL(rng_is_initialized);
-
-/*
- * Add a callback function that will be invoked when the nonblocking
- * pool is initialised.
- *
- * returns: 0 if callback is successfully added
- *	    -EALREADY if pool is already initialised (callback not called)
- *	    -ENOENT if module for callback is not alive
- */
-int add_random_ready_callback(struct random_ready_callback *rdy)
-{
-	struct module *owner;
-	unsigned long flags;
-	int err = -EALREADY;
-
-	if (crng_ready())
-		return err;
-
-	owner = rdy->owner;
-	if (!try_module_get(owner))
-		return -ENOENT;
-
-	spin_lock_irqsave(&random_ready_list_lock, flags);
-	if (crng_ready())
-		goto out;
-
-	owner = NULL;
-
-	list_add(&rdy->list, &random_ready_list);
-	err = 0;
-
-out:
-	spin_unlock_irqrestore(&random_ready_list_lock, flags);
-
-	module_put(owner);
-
-	return err;
-}
-EXPORT_SYMBOL(add_random_ready_callback);
-
-/*
- * Delete a previously registered readiness callback function.
- */
-void del_random_ready_callback(struct random_ready_callback *rdy)
-{
-	unsigned long flags;
-	struct module *owner = NULL;
-
-	spin_lock_irqsave(&random_ready_list_lock, flags);
-	if (!list_empty(&rdy->list)) {
-		list_del_init(&rdy->list);
-		owner = rdy->owner;
-	}
-	spin_unlock_irqrestore(&random_ready_list_lock, flags);
-
-	module_put(owner);
-}
-EXPORT_SYMBOL(del_random_ready_callback);
-
-/*
- * This function will use the architecture-specific hardware random
- * number generator if it is available. It is not recommended for
- * use. Use get_random_bytes() instead. It returns the number of
- * bytes filled in.
- */
-size_t __must_check get_random_bytes_arch(void *buf, size_t nbytes)
-{
-	size_t left = nbytes;
-	u8 *p = buf;
-
-	while (left) {
-		unsigned long v;
-		size_t chunk = min_t(size_t, left, sizeof(unsigned long));
-
-		if (!arch_get_random_long(&v))
-			break;
-
-		memcpy(p, &v, chunk);
-		p += chunk;
-		left -= chunk;
-	}
-
-	return nbytes - left;
-}
-EXPORT_SYMBOL(get_random_bytes_arch);
-
 static bool trust_cpu __ro_after_init = IS_ENABLED(CONFIG_RANDOM_TRUST_CPU);
 static int __init parse_trust_cpu(char *arg)
 {
@@ -1508,126 +1667,6 @@ struct ctl_table random_table[] = {
 };
 #endif	/* CONFIG_SYSCTL */
 
-struct batched_entropy {
-	union {
-		/*
-		 * We make this 1.5x a ChaCha block, so that we get the
-		 * remaining 32 bytes from fast key erasure, plus one full
-		 * block from the detached ChaCha state. We can increase
-		 * the size of this later if needed so long as we keep the
-		 * formula of (integer_blocks + 0.5) * CHACHA20_BLOCK_SIZE.
-		 */
-		u64 entropy_u64[CHACHA20_BLOCK_SIZE * 3 / (2 * sizeof(u64))];
-		u32 entropy_u32[CHACHA20_BLOCK_SIZE * 3 / (2 * sizeof(u32))];
-	};
-	unsigned long generation;
-	unsigned int position;
-};
-
-/*
- * Get a random word for internal kernel use only. The quality of the random
- * number is good as /dev/urandom. In order to ensure that the randomness
- * provided by this function is okay, the function wait_for_random_bytes()
- * should be called and return 0 at least once at any point prior.
- */
-static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u64) = {
-	.position = UINT_MAX
-};
-
-u64 get_random_u64(void)
-{
-	u64 ret;
-	unsigned long flags;
-	struct batched_entropy *batch;
-	static void *previous;
-	unsigned long next_gen;
-
-	warn_unseeded_randomness(&previous);
-
-	local_irq_save(flags);
-	batch = raw_cpu_ptr(&batched_entropy_u64);
-
-	next_gen = READ_ONCE(base_crng.generation);
-	if (batch->position >= ARRAY_SIZE(batch->entropy_u64) ||
-	    next_gen != batch->generation) {
-		_get_random_bytes(batch->entropy_u64, sizeof(batch->entropy_u64));
-		batch->position = 0;
-		batch->generation = next_gen;
-	}
-
-	ret = batch->entropy_u64[batch->position];
-	batch->entropy_u64[batch->position] = 0;
-	++batch->position;
-	local_irq_restore(flags);
-	return ret;
-}
-EXPORT_SYMBOL(get_random_u64);
-
-static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u32) = {
-	.position = UINT_MAX
-};
-
-u32 get_random_u32(void)
-{
-	u32 ret;
-	unsigned long flags;
-	struct batched_entropy *batch;
-	static void *previous;
-	unsigned long next_gen;
-
-	warn_unseeded_randomness(&previous);
-
-	local_irq_save(flags);
-	batch = raw_cpu_ptr(&batched_entropy_u32);
-
-	next_gen = READ_ONCE(base_crng.generation);
-	if (batch->position >= ARRAY_SIZE(batch->entropy_u32) ||
-	    next_gen != batch->generation) {
-		_get_random_bytes(batch->entropy_u32, sizeof(batch->entropy_u32));
-		batch->position = 0;
-		batch->generation = next_gen;
-	}
-
-	ret = batch->entropy_u32[batch->position];
-	batch->entropy_u32[batch->position] = 0;
-	++batch->position;
-	local_irq_restore(flags);
-	return ret;
-}
-EXPORT_SYMBOL(get_random_u32);
-
-/**
- * randomize_page - Generate a random, page aligned address
- * @start:	The smallest acceptable address the caller will take.
- * @range:	The size of the area, starting at @start, within which the
- *		random address must fall.
- *
- * If @start + @range would overflow, @range is capped.
- *
- * NOTE: Historical use of randomize_range, which this replaces, presumed that
- * @start was already page aligned.  We now align it regardless.
- *
- * Return: A page aligned address within [start, start + range).  On error,
- * @start is returned.
- */
-unsigned long randomize_page(unsigned long start, unsigned long range)
-{
-	if (!PAGE_ALIGNED(start)) {
-		range -= PAGE_ALIGN(start) - start;
-		start = PAGE_ALIGN(start);
-	}
-
-	if (start > ULONG_MAX - range)
-		range = ULONG_MAX - start;
-
-	range >>= PAGE_SHIFT;
-
-	if (range == 0)
-		return start;
-
-	return start + (get_random_long() % range << PAGE_SHIFT);
-}
-
 /* Interface for in-kernel drivers of true hardware RNGs.
  * Those devices may produce endless random bits and will be throttled
  * when our pool is full.
author maintainer
#100 lore raw

PATCH 100/234 random: group entropy extraction functions

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit a5ed7cb1a7732ef11959332d507889fbc39ebbb4 upstream.

This pulls all of the entropy extraction-focused functions into the third labeled section.

No functional changes.

---
 drivers/char/random.c |  216 +++++++++++++++++++++++++-------------------------
 1 file changed, 109 insertions(+), 107 deletions(-)
drivers/char/random.c +109 -107
@@ -890,23 +890,36 @@ size_t __must_check get_random_bytes_arc
 }
 EXPORT_SYMBOL(get_random_bytes_arch);
 
+
+/**********************************************************************
+ *
+ * Entropy accumulation and extraction routines.
+ *
+ * Callers may add entropy via:
+ *
+ *     static void mix_pool_bytes(const void *in, size_t nbytes)
+ *
+ * After which, if added entropy should be credited:
+ *
+ *     static void credit_entropy_bits(size_t nbits)
+ *
+ * Finally, extract entropy via these two, with the latter one
+ * setting the entropy count to zero and extracting only if there
+ * is POOL_MIN_BITS entropy credited prior:
+ *
+ *     static void extract_entropy(void *buf, size_t nbytes)
+ *     static bool drain_entropy(void *buf, size_t nbytes)
+ *
+ **********************************************************************/
+
 enum {
 	POOL_BITS = BLAKE2S_HASH_SIZE * 8,
 	POOL_MIN_BITS = POOL_BITS /* No point in settling for less. */
 };
 
-/*
- * Static global variables
- */
+/* For notifying userspace should write into /dev/random. */
 static DECLARE_WAIT_QUEUE_HEAD(random_write_wait);
 
-/**********************************************************************
- *
- * OS independent entropy store.   Here are the functions which handle
- * storing entropy in an entropy pool.
- *
- **********************************************************************/
-
 static struct {
 	struct blake2s_state hash;
 	spinlock_t lock;
@@ -919,28 +932,106 @@ static struct {
 	.lock = __SPIN_LOCK_UNLOCKED(input_pool.lock),
 };
 
-static void extract_entropy(void *buf, size_t nbytes);
-static bool drain_entropy(void *buf, size_t nbytes);
-
-static void crng_reseed(void);
+static void _mix_pool_bytes(const void *in, size_t nbytes)
+{
+	blake2s_update(&input_pool.hash, in, nbytes);
+}
 
 /*
  * This function adds bytes into the entropy "pool".  It does not
  * update the entropy estimate.  The caller should call
  * credit_entropy_bits if this is appropriate.
  */
-static void _mix_pool_bytes(const void *in, size_t nbytes)
+static void mix_pool_bytes(const void *in, size_t nbytes)
 {
-	blake2s_update(&input_pool.hash, in, nbytes);
+	unsigned long flags;
+
+	spin_lock_irqsave(&input_pool.lock, flags);
+	_mix_pool_bytes(in, nbytes);
+	spin_unlock_irqrestore(&input_pool.lock, flags);
 }
 
-static void mix_pool_bytes(const void *in, size_t nbytes)
+static void credit_entropy_bits(size_t nbits)
+{
+	unsigned int entropy_count, orig, add;
+
+	if (!nbits)
+		return;
+
+	add = min_t(size_t, nbits, POOL_BITS);
+
+	do {
+		orig = READ_ONCE(input_pool.entropy_count);
+		entropy_count = min_t(unsigned int, POOL_BITS, orig + add);
+	} while (cmpxchg(&input_pool.entropy_count, orig, entropy_count) != orig);
+
+	if (crng_init < 2 && entropy_count >= POOL_MIN_BITS)
+		crng_reseed();
+}
+
+/*
+ * This is an HKDF-like construction for using the hashed collected entropy
+ * as a PRF key, that's then expanded block-by-block.
+ */
+static void extract_entropy(void *buf, size_t nbytes)
 {
 	unsigned long flags;
+	u8 seed[BLAKE2S_HASH_SIZE], next_key[BLAKE2S_HASH_SIZE];
+	struct {
+		unsigned long rdseed[32 / sizeof(long)];
+		size_t counter;
+	} block;
+	size_t i;
+
+	for (i = 0; i < ARRAY_SIZE(block.rdseed); ++i) {
+		if (!arch_get_random_seed_long(&block.rdseed[i]) &&
+		    !arch_get_random_long(&block.rdseed[i]))
+			block.rdseed[i] = random_get_entropy();
+	}
 
 	spin_lock_irqsave(&input_pool.lock, flags);
-	_mix_pool_bytes(in, nbytes);
+
+	/* seed = HASHPRF(last_key, entropy_input) */
+	blake2s_final(&input_pool.hash, seed);
+
+	/* next_key = HASHPRF(seed, RDSEED || 0) */
+	block.counter = 0;
+	blake2s(next_key, (u8 *)&block, seed, sizeof(next_key), sizeof(block), sizeof(seed));
+	blake2s_init_key(&input_pool.hash, BLAKE2S_HASH_SIZE, next_key, sizeof(next_key));
+
 	spin_unlock_irqrestore(&input_pool.lock, flags);
+	memzero_explicit(next_key, sizeof(next_key));
+
+	while (nbytes) {
+		i = min_t(size_t, nbytes, BLAKE2S_HASH_SIZE);
+		/* output = HASHPRF(seed, RDSEED || ++counter) */
+		++block.counter;
+		blake2s(buf, (u8 *)&block, seed, i, sizeof(block), sizeof(seed));
+		nbytes -= i;
+		buf += i;
+	}
+
+	memzero_explicit(seed, sizeof(seed));
+	memzero_explicit(&block, sizeof(block));
+}
+
+/*
+ * First we make sure we have POOL_MIN_BITS of entropy in the pool, and then we
+ * set the entropy count to zero (but don't actually touch any data). Only then
+ * can we extract a new key with extract_entropy().
+ */
+static bool drain_entropy(void *buf, size_t nbytes)
+{
+	unsigned int entropy_count;
+	do {
+		entropy_count = READ_ONCE(input_pool.entropy_count);
+		if (entropy_count < POOL_MIN_BITS)
+			return false;
+	} while (cmpxchg(&input_pool.entropy_count, entropy_count, 0) != entropy_count);
+	extract_entropy(buf, nbytes);
+	wake_up_interruptible(&random_write_wait);
+	kill_fasync(&fasync, SIGIO, POLL_OUT);
+	return true;
 }
 
 struct fast_pool {
@@ -983,24 +1074,6 @@ static void fast_mix(u32 pool[4])
 	pool[2] = c;  pool[3] = d;
 }
 
-static void credit_entropy_bits(size_t nbits)
-{
-	unsigned int entropy_count, orig, add;
-
-	if (!nbits)
-		return;
-
-	add = min_t(size_t, nbits, POOL_BITS);
-
-	do {
-		orig = READ_ONCE(input_pool.entropy_count);
-		entropy_count = min_t(unsigned int, POOL_BITS, orig + add);
-	} while (cmpxchg(&input_pool.entropy_count, orig, entropy_count) != orig);
-
-	if (crng_init < 2 && entropy_count >= POOL_MIN_BITS)
-		crng_reseed();
-}
-
 /*********************************************************************
  *
  * Entropy input management
@@ -1197,77 +1270,6 @@ void add_disk_randomness(struct gendisk
 EXPORT_SYMBOL_GPL(add_disk_randomness);
 #endif
 
-/*********************************************************************
- *
- * Entropy extraction routines
- *
- *********************************************************************/
-
-/*
- * This is an HKDF-like construction for using the hashed collected entropy
- * as a PRF key, that's then expanded block-by-block.
- */
-static void extract_entropy(void *buf, size_t nbytes)
-{
-	unsigned long flags;
-	u8 seed[BLAKE2S_HASH_SIZE], next_key[BLAKE2S_HASH_SIZE];
-	struct {
-		unsigned long rdseed[32 / sizeof(long)];
-		size_t counter;
-	} block;
-	size_t i;
-
-	for (i = 0; i < ARRAY_SIZE(block.rdseed); ++i) {
-		if (!arch_get_random_seed_long(&block.rdseed[i]) &&
-		    !arch_get_random_long(&block.rdseed[i]))
-			block.rdseed[i] = random_get_entropy();
-	}
-
-	spin_lock_irqsave(&input_pool.lock, flags);
-
-	/* seed = HASHPRF(last_key, entropy_input) */
-	blake2s_final(&input_pool.hash, seed);
-
-	/* next_key = HASHPRF(seed, RDSEED || 0) */
-	block.counter = 0;
-	blake2s(next_key, (u8 *)&block, seed, sizeof(next_key), sizeof(block), sizeof(seed));
-	blake2s_init_key(&input_pool.hash, BLAKE2S_HASH_SIZE, next_key, sizeof(next_key));
-
-	spin_unlock_irqrestore(&input_pool.lock, flags);
-	memzero_explicit(next_key, sizeof(next_key));
-
-	while (nbytes) {
-		i = min_t(size_t, nbytes, BLAKE2S_HASH_SIZE);
-		/* output = HASHPRF(seed, RDSEED || ++counter) */
-		++block.counter;
-		blake2s(buf, (u8 *)&block, seed, i, sizeof(block), sizeof(seed));
-		nbytes -= i;
-		buf += i;
-	}
-
-	memzero_explicit(seed, sizeof(seed));
-	memzero_explicit(&block, sizeof(block));
-}
-
-/*
- * First we make sure we have POOL_MIN_BITS of entropy in the pool, and then we
- * set the entropy count to zero (but don't actually touch any data). Only then
- * can we extract a new key with extract_entropy().
- */
-static bool drain_entropy(void *buf, size_t nbytes)
-{
-	unsigned int entropy_count;
-	do {
-		entropy_count = READ_ONCE(input_pool.entropy_count);
-		if (entropy_count < POOL_MIN_BITS)
-			return false;
-	} while (cmpxchg(&input_pool.entropy_count, entropy_count, 0) != entropy_count);
-	extract_entropy(buf, nbytes);
-	wake_up_interruptible(&random_write_wait);
-	kill_fasync(&fasync, SIGIO, POLL_OUT);
-	return true;
-}
-
 /*
  * Each time the timer fires, we expect that we got an unpredictable
  * jump in the cycle counter. Even if the timer is running on another
author maintainer
#101 lore raw

PATCH 101/234 random: group entropy collection functions

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 92c653cf14400946f376a29b828d6af7e01f38dd upstream.

This pulls all of the entropy collection-focused functions into the fourth labeled section.

No functional changes.

---
 drivers/char/random.c |  370 +++++++++++++++++++++++++++-----------------------
 1 file changed, 206 insertions(+), 164 deletions(-)
drivers/char/random.c +206 -164
@@ -1034,60 +1034,112 @@ static bool drain_entropy(void *buf, siz
 	return true;
 }
 
-struct fast_pool {
-	union {
-		u32 pool32[4];
-		u64 pool64[2];
-	};
-	unsigned long last;
-	u16 reg_idx;
-	u8 count;
-};
+
+/**********************************************************************
+ *
+ * Entropy collection routines.
+ *
+ * The following exported functions are used for pushing entropy into
+ * the above entropy accumulation routines:
+ *
+ *	void add_device_randomness(const void *buf, size_t size);
+ *	void add_input_randomness(unsigned int type, unsigned int code,
+ *	                          unsigned int value);
+ *	void add_disk_randomness(struct gendisk *disk);
+ *	void add_hwgenerator_randomness(const void *buffer, size_t count,
+ *					size_t entropy);
+ *	void add_bootloader_randomness(const void *buf, size_t size);
+ *	void add_interrupt_randomness(int irq);
+ *
+ * add_device_randomness() adds data to the input pool that
+ * is likely to differ between two devices (or possibly even per boot).
+ * This would be things like MAC addresses or serial numbers, or the
+ * read-out of the RTC. This does *not* credit any actual entropy to
+ * the pool, but it initializes the pool to different values for devices
+ * that might otherwise be identical and have very little entropy
+ * available to them (particularly common in the embedded world).
+ *
+ * add_input_randomness() uses the input layer interrupt timing, as well
+ * as the event type information from the hardware.
+ *
+ * add_disk_randomness() uses what amounts to the seek time of block
+ * layer request events, on a per-disk_devt basis, as input to the
+ * entropy pool. Note that high-speed solid state drives with very low
+ * seek times do not make for good sources of entropy, as their seek
+ * times are usually fairly consistent.
+ *
+ * The above two routines try to estimate how many bits of entropy
+ * to credit. They do this by keeping track of the first and second
+ * order deltas of the event timings.
+ *
+ * add_hwgenerator_randomness() is for true hardware RNGs, and will credit
+ * entropy as specified by the caller. If the entropy pool is full it will
+ * block until more entropy is needed.
+ *
+ * add_bootloader_randomness() is the same as add_hwgenerator_randomness() or
+ * add_device_randomness(), depending on whether or not the configuration
+ * option CONFIG_RANDOM_TRUST_BOOTLOADER is set.
+ *
+ * add_interrupt_randomness() uses the interrupt timing as random
+ * inputs to the entropy pool. Using the cycle counters and the irq source
+ * as inputs, it feeds the input pool roughly once a second or after 64
+ * interrupts, crediting 1 bit of entropy for whichever comes first.
+ *
+ **********************************************************************/
+
+static bool trust_cpu __ro_after_init = IS_ENABLED(CONFIG_RANDOM_TRUST_CPU);
+static int __init parse_trust_cpu(char *arg)
+{
+	return kstrtobool(arg, &trust_cpu);
+}
+early_param("random.trust_cpu", parse_trust_cpu);
 
 /*
- * This is a fast mixing routine used by the interrupt randomness
- * collector.  It's hardcoded for an 128 bit pool and assumes that any
- * locks that might be needed are taken by the caller.
+ * The first collection of entropy occurs at system boot while interrupts
+ * are still turned off. Here we push in RDSEED, a timestamp, and utsname().
+ * Depending on the above configuration knob, RDSEED may be considered
+ * sufficient for initialization. Note that much earlier setup may already
+ * have pushed entropy into the input pool by the time we get here.
  */
-static void fast_mix(u32 pool[4])
+int __init rand_initialize(void)
 {
-	u32 a = pool[0],	b = pool[1];
-	u32 c = pool[2],	d = pool[3];
-
-	a += b;			c += d;
-	b = rol32(b, 6);	d = rol32(d, 27);
-	d ^= a;			b ^= c;
+	size_t i;
+	ktime_t now = ktime_get_real();
+	bool arch_init = true;
+	unsigned long rv;
 
-	a += b;			c += d;
-	b = rol32(b, 16);	d = rol32(d, 14);
-	d ^= a;			b ^= c;
+	for (i = 0; i < BLAKE2S_BLOCK_SIZE; i += sizeof(rv)) {
+		if (!arch_get_random_seed_long_early(&rv) &&
+		    !arch_get_random_long_early(&rv)) {
+			rv = random_get_entropy();
+			arch_init = false;
+		}
+		mix_pool_bytes(&rv, sizeof(rv));
+	}
+	mix_pool_bytes(&now, sizeof(now));
+	mix_pool_bytes(utsname(), sizeof(*(utsname())));
 
-	a += b;			c += d;
-	b = rol32(b, 6);	d = rol32(d, 27);
-	d ^= a;			b ^= c;
+	extract_entropy(base_crng.key, sizeof(base_crng.key));
+	++base_crng.generation;
 
-	a += b;			c += d;
-	b = rol32(b, 16);	d = rol32(d, 14);
-	d ^= a;			b ^= c;
+	if (arch_init && trust_cpu && crng_init < 2) {
+		crng_init = 2;
+		pr_notice("crng init done (trusting CPU's manufacturer)\n");
+	}
 
-	pool[0] = a;  pool[1] = b;
-	pool[2] = c;  pool[3] = d;
+	if (ratelimit_disable) {
+		urandom_warning.interval = 0;
+		unseeded_warning.interval = 0;
+	}
+	return 0;
 }
 
-/*********************************************************************
- *
- * Entropy input management
- *
- *********************************************************************/
-
 /* There is one of these per entropy source */
 struct timer_rand_state {
 	cycles_t last_time;
 	long last_delta, last_delta2;
 };
 
-#define INIT_TIMER_RAND_STATE { INITIAL_JIFFIES, };
-
 /*
  * Add device- or boot-specific data to the input pool to help
  * initialize it.
@@ -1111,8 +1163,6 @@ void add_device_randomness(const void *b
 }
 EXPORT_SYMBOL(add_device_randomness);
 
-static struct timer_rand_state input_timer_state = INIT_TIMER_RAND_STATE;
-
 /*
  * This function adds entropy to the entropy "pool" by using timing
  * delays.  It uses the timer_rand_state structure to make an estimate
@@ -1174,8 +1224,9 @@ void add_input_randomness(unsigned int t
 			  unsigned int value)
 {
 	static unsigned char last_value;
+	static struct timer_rand_state input_timer_state = { INITIAL_JIFFIES };
 
-	/* ignore autorepeat and the like */
+	/* Ignore autorepeat and the like. */
 	if (value == last_value)
 		return;
 
@@ -1185,6 +1236,119 @@ void add_input_randomness(unsigned int t
 }
 EXPORT_SYMBOL_GPL(add_input_randomness);
 
+#ifdef CONFIG_BLOCK
+void add_disk_randomness(struct gendisk *disk)
+{
+	if (!disk || !disk->random)
+		return;
+	/* First major is 1, so we get >= 0x200 here. */
+	add_timer_randomness(disk->random, 0x100 + disk_devt(disk));
+}
+EXPORT_SYMBOL_GPL(add_disk_randomness);
+
+void rand_initialize_disk(struct gendisk *disk)
+{
+	struct timer_rand_state *state;
+
+	/*
+	 * If kzalloc returns null, we just won't use that entropy
+	 * source.
+	 */
+	state = kzalloc(sizeof(struct timer_rand_state), GFP_KERNEL);
+	if (state) {
+		state->last_time = INITIAL_JIFFIES;
+		disk->random = state;
+	}
+}
+#endif
+
+/*
+ * Interface for in-kernel drivers of true hardware RNGs.
+ * Those devices may produce endless random bits and will be throttled
+ * when our pool is full.
+ */
+void add_hwgenerator_randomness(const void *buffer, size_t count,
+				size_t entropy)
+{
+	if (unlikely(crng_init == 0)) {
+		size_t ret = crng_fast_load(buffer, count);
+		mix_pool_bytes(buffer, ret);
+		count -= ret;
+		buffer += ret;
+		if (!count || crng_init == 0)
+			return;
+	}
+
+	/*
+	 * Throttle writing if we're above the trickle threshold.
+	 * We'll be woken up again once below POOL_MIN_BITS, when
+	 * the calling thread is about to terminate, or once
+	 * CRNG_RESEED_INTERVAL has elapsed.
+	 */
+	wait_event_interruptible_timeout(random_write_wait,
+			!system_wq || kthread_should_stop() ||
+			input_pool.entropy_count < POOL_MIN_BITS,
+			CRNG_RESEED_INTERVAL);
+	mix_pool_bytes(buffer, count);
+	credit_entropy_bits(entropy);
+}
+EXPORT_SYMBOL_GPL(add_hwgenerator_randomness);
+
+/*
+ * Handle random seed passed by bootloader.
+ * If the seed is trustworthy, it would be regarded as hardware RNGs. Otherwise
+ * it would be regarded as device data.
+ * The decision is controlled by CONFIG_RANDOM_TRUST_BOOTLOADER.
+ */
+void add_bootloader_randomness(const void *buf, size_t size)
+{
+	if (IS_ENABLED(CONFIG_RANDOM_TRUST_BOOTLOADER))
+		add_hwgenerator_randomness(buf, size, size * 8);
+	else
+		add_device_randomness(buf, size);
+}
+EXPORT_SYMBOL_GPL(add_bootloader_randomness);
+
+struct fast_pool {
+	union {
+		u32 pool32[4];
+		u64 pool64[2];
+	};
+	unsigned long last;
+	u16 reg_idx;
+	u8 count;
+};
+
+/*
+ * This is a fast mixing routine used by the interrupt randomness
+ * collector. It's hardcoded for an 128 bit pool and assumes that any
+ * locks that might be needed are taken by the caller.
+ */
+static void fast_mix(u32 pool[4])
+{
+	u32 a = pool[0],	b = pool[1];
+	u32 c = pool[2],	d = pool[3];
+
+	a += b;			c += d;
+	b = rol32(b, 6);	d = rol32(d, 27);
+	d ^= a;			b ^= c;
+
+	a += b;			c += d;
+	b = rol32(b, 16);	d = rol32(d, 14);
+	d ^= a;			b ^= c;
+
+	a += b;			c += d;
+	b = rol32(b, 6);	d = rol32(d, 27);
+	d ^= a;			b ^= c;
+
+	a += b;			c += d;
+	b = rol32(b, 16);	d = rol32(d, 14);
+	d ^= a;			b ^= c;
+
+	pool[0] = a;  pool[1] = b;
+	pool[2] = c;  pool[3] = d;
+}
+
 static DEFINE_PER_CPU(struct fast_pool, irq_randomness);
 
 static u32 get_reg(struct fast_pool *f, struct pt_regs *regs)
@@ -1254,22 +1418,11 @@ void add_interrupt_randomness(int irq)
 
 	fast_pool->count = 0;
 
-	/* award one bit for the contents of the fast pool */
+	/* Award one bit for the contents of the fast pool. */
 	credit_entropy_bits(1);
 }
 EXPORT_SYMBOL_GPL(add_interrupt_randomness);
 
-#ifdef CONFIG_BLOCK
-void add_disk_randomness(struct gendisk *disk)
-{
-	if (!disk || !disk->random)
-		return;
-	/* first major is 1, so we get >= 0x200 here */
-	add_timer_randomness(disk->random, 0x100 + disk_devt(disk));
-}
-EXPORT_SYMBOL_GPL(add_disk_randomness);
-#endif
-
 /*
  * Each time the timer fires, we expect that we got an unpredictable
  * jump in the cycle counter. Even if the timer is running on another
@@ -1319,73 +1472,6 @@ static void try_to_generate_entropy(void
 	mix_pool_bytes(&stack.now, sizeof(stack.now));
 }
 
-static bool trust_cpu __ro_after_init = IS_ENABLED(CONFIG_RANDOM_TRUST_CPU);
-static int __init parse_trust_cpu(char *arg)
-{
-	return kstrtobool(arg, &trust_cpu);
-}
-early_param("random.trust_cpu", parse_trust_cpu);
-
-/*
- * Note that setup_arch() may call add_device_randomness()
- * long before we get here. This allows seeding of the pools
- * with some platform dependent data very early in the boot
- * process. But it limits our options here. We must use
- * statically allocated structures that already have all
- * initializations complete at compile time. We should also
- * take care not to overwrite the precious per platform data
- * we were given.
- */
-int __init rand_initialize(void)
-{
-	size_t i;
-	ktime_t now = ktime_get_real();
-	bool arch_init = true;
-	unsigned long rv;
-
-	for (i = 0; i < BLAKE2S_BLOCK_SIZE; i += sizeof(rv)) {
-		if (!arch_get_random_seed_long_early(&rv) &&
-		    !arch_get_random_long_early(&rv)) {
-			rv = random_get_entropy();
-			arch_init = false;
-		}
-		mix_pool_bytes(&rv, sizeof(rv));
-	}
-	mix_pool_bytes(&now, sizeof(now));
-	mix_pool_bytes(utsname(), sizeof(*(utsname())));
-
-	extract_entropy(base_crng.key, sizeof(base_crng.key));
-	++base_crng.generation;
-
-	if (arch_init && trust_cpu && crng_init < 2) {
-		crng_init = 2;
-		pr_notice("crng init done (trusting CPU's manufacturer)\n");
-	}
-
-	if (ratelimit_disable) {
-		urandom_warning.interval = 0;
-		unseeded_warning.interval = 0;
-	}
-	return 0;
-}
-
-#ifdef CONFIG_BLOCK
-void rand_initialize_disk(struct gendisk *disk)
-{
-	struct timer_rand_state *state;
-
-	/*
-	 * If kzalloc returns null, we just won't use that entropy
-	 * source.
-	 */
-	state = kzalloc(sizeof(struct timer_rand_state), GFP_KERNEL);
-	if (state) {
-		state->last_time = INITIAL_JIFFIES;
-		disk->random = state;
-	}
-}
-#endif
-
 static ssize_t urandom_read(struct file *file, char __user *buf, size_t nbytes,
 			    loff_t *ppos)
 {
@@ -1668,47 +1754,3 @@ struct ctl_table random_table[] = {
 	{ }
 };
 #endif	/* CONFIG_SYSCTL */
-
-/* Interface for in-kernel drivers of true hardware RNGs.
- * Those devices may produce endless random bits and will be throttled
- * when our pool is full.
- */
-void add_hwgenerator_randomness(const void *buffer, size_t count,
-				size_t entropy)
-{
-	if (unlikely(crng_init == 0)) {
-		size_t ret = crng_fast_load(buffer, count);
-		mix_pool_bytes(buffer, ret);
-		count -= ret;
-		buffer += ret;
-		if (!count || crng_init == 0)
-			return;
-	}
-
-	/* Throttle writing if we're above the trickle threshold.
-	 * We'll be woken up again once below POOL_MIN_BITS, when
-	 * the calling thread is about to terminate, or once
-	 * CRNG_RESEED_INTERVAL has elapsed.
-	 */
-	wait_event_interruptible_timeout(random_write_wait,
-			!system_wq || kthread_should_stop() ||
-			input_pool.entropy_count < POOL_MIN_BITS,
-			CRNG_RESEED_INTERVAL);
-	mix_pool_bytes(buffer, count);
-	credit_entropy_bits(entropy);
-}
-EXPORT_SYMBOL_GPL(add_hwgenerator_randomness);
-
-/* Handle random seed passed by bootloader.
- * If the seed is trustworthy, it would be regarded as hardware RNGs. Otherwise
- * it would be regarded as device data.
- * The decision is controlled by CONFIG_RANDOM_TRUST_BOOTLOADER.
- */
-void add_bootloader_randomness(const void *buf, size_t size)
-{
-	if (IS_ENABLED(CONFIG_RANDOM_TRUST_BOOTLOADER))
-		add_hwgenerator_randomness(buf, size, size * 8);
-	else
-		add_device_randomness(buf, size);
-}
-EXPORT_SYMBOL_GPL(add_bootloader_randomness);
author maintainer
#102 lore raw

PATCH 102/234 random: group userspace read/write functions

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit a6adf8e7a605250b911e94793fd077933709ff9e upstream.

This pulls all of the userspace read/write-focused functions into the fifth labeled section.

No functional changes.

---
 drivers/char/random.c |  125 ++++++++++++++++++++++++++++++--------------------
 1 file changed, 77 insertions(+), 48 deletions(-)
drivers/char/random.c +77 -48
@@ -1472,30 +1472,61 @@ static void try_to_generate_entropy(void
 	mix_pool_bytes(&stack.now, sizeof(stack.now));
 }
 
-static ssize_t urandom_read(struct file *file, char __user *buf, size_t nbytes,
-			    loff_t *ppos)
+
+/**********************************************************************
+ *
+ * Userspace reader/writer interfaces.
+ *
+ * getrandom(2) is the primary modern interface into the RNG and should
+ * be used in preference to anything else.
+ *
+ * Reading from /dev/random has the same functionality as calling
+ * getrandom(2) with flags=0. In earlier versions, however, it had
+ * vastly different semantics and should therefore be avoided, to
+ * prevent backwards compatibility issues.
+ *
+ * Reading from /dev/urandom has the same functionality as calling
+ * getrandom(2) with flags=GRND_INSECURE. Because it does not block
+ * waiting for the RNG to be ready, it should not be used.
+ *
+ * Writing to either /dev/random or /dev/urandom adds entropy to
+ * the input pool but does not credit it.
+ *
+ * Polling on /dev/random indicates when the RNG is initialized, on
+ * the read side, and when it wants new entropy, on the write side.
+ *
+ * Both /dev/random and /dev/urandom have the same set of ioctls for
+ * adding entropy, getting the entropy count, zeroing the count, and
+ * reseeding the crng.
+ *
+ **********************************************************************/
+
+SYSCALL_DEFINE3(getrandom, char __user *, buf, size_t, count, unsigned int,
+		flags)
 {
-	static int maxwarn = 10;
+	if (flags & ~(GRND_NONBLOCK | GRND_RANDOM | GRND_INSECURE))
+		return -EINVAL;
 
-	if (!crng_ready() && maxwarn > 0) {
-		maxwarn--;
-		if (__ratelimit(&urandom_warning))
-			pr_notice("%s: uninitialized urandom read (%zd bytes read)\n",
-				  current->comm, nbytes);
-	}
+	/*
+	 * Requesting insecure and blocking randomness at the same time makes
+	 * no sense.
+	 */
+	if ((flags & (GRND_INSECURE | GRND_RANDOM)) == (GRND_INSECURE | GRND_RANDOM))
+		return -EINVAL;
 
-	return get_random_bytes_user(buf, nbytes);
-}
+	if (count > INT_MAX)
+		count = INT_MAX;
 
-static ssize_t random_read(struct file *file, char __user *buf, size_t nbytes,
-			   loff_t *ppos)
-{
-	int ret;
+	if (!(flags & GRND_INSECURE) && !crng_ready()) {
+		int ret;
 
-	ret = wait_for_random_bytes();
-	if (ret != 0)
-		return ret;
-	return get_random_bytes_user(buf, nbytes);
+		if (flags & GRND_NONBLOCK)
+			return -EAGAIN;
+		ret = wait_for_random_bytes();
+		if (unlikely(ret))
+			return ret;
+	}
+	return get_random_bytes_user(buf, count);
 }
 
 static __poll_t random_poll(struct file *file, poll_table *wait)
@@ -1547,6 +1578,32 @@ static ssize_t random_write(struct file
 	return (ssize_t)count;
 }
 
+static ssize_t urandom_read(struct file *file, char __user *buf, size_t nbytes,
+			    loff_t *ppos)
+{
+	static int maxwarn = 10;
+
+	if (!crng_ready() && maxwarn > 0) {
+		maxwarn--;
+		if (__ratelimit(&urandom_warning))
+			pr_notice("%s: uninitialized urandom read (%zd bytes read)\n",
+				  current->comm, nbytes);
+	}
+
+	return get_random_bytes_user(buf, nbytes);
+}
+
+static ssize_t random_read(struct file *file, char __user *buf, size_t nbytes,
+			   loff_t *ppos)
+{
+	int ret;
+
+	ret = wait_for_random_bytes();
+	if (ret != 0)
+		return ret;
+	return get_random_bytes_user(buf, nbytes);
+}
+
 static long random_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
 {
 	int size, ent_count;
@@ -1555,7 +1612,7 @@ static long random_ioctl(struct file *f,
 
 	switch (cmd) {
 	case RNDGETENTCNT:
-		/* inherently racy, no point locking */
+		/* Inherently racy, no point locking. */
 		if (put_user(input_pool.entropy_count, p))
 			return -EFAULT;
 		return 0;
@@ -1629,34 +1686,6 @@ const struct file_operations urandom_fop
 	.llseek = noop_llseek,
 };
 
-SYSCALL_DEFINE3(getrandom, char __user *, buf, size_t, count, unsigned int,
-		flags)
-{
-	if (flags & ~(GRND_NONBLOCK | GRND_RANDOM | GRND_INSECURE))
-		return -EINVAL;
-
-	/*
-	 * Requesting insecure and blocking randomness at the same time makes
-	 * no sense.
-	 */
-	if ((flags & (GRND_INSECURE | GRND_RANDOM)) == (GRND_INSECURE | GRND_RANDOM))
-		return -EINVAL;
-
-	if (count > INT_MAX)
-		count = INT_MAX;
-
-	if (!(flags & GRND_INSECURE) && !crng_ready()) {
-		int ret;
-
-		if (flags & GRND_NONBLOCK)
-			return -EAGAIN;
-		ret = wait_for_random_bytes();
-		if (unlikely(ret))
-			return ret;
-	}
-	return get_random_bytes_user(buf, count);
-}
-
 /********************************************************************
  *
  * Sysctl interface
author maintainer
#103 lore raw

PATCH 103/234 random: group sysctl functions

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 0deff3c43206c24e746b1410f11125707ad3040e upstream.

This pulls all of the sysctl-focused functions into the sixth labeled section.

No functional changes.

---
 drivers/char/random.c |   35 ++++++++++++++++++++++++++++++-----
 1 file changed, 30 insertions(+), 5 deletions(-)
drivers/char/random.c +30 -5
@@ -1686,9 +1686,34 @@ const struct file_operations urandom_fop
 	.llseek = noop_llseek,
 };
 
+
 /********************************************************************
  *
- * Sysctl interface
+ * Sysctl interface.
+ *
+ * These are partly unused legacy knobs with dummy values to not break
+ * userspace and partly still useful things. They are usually accessible
+ * in /proc/sys/kernel/random/ and are as follows:
+ *
+ * - boot_id - a UUID representing the current boot.
+ *
+ * - uuid - a random UUID, different each time the file is read.
+ *
+ * - poolsize - the number of bits of entropy that the input pool can
+ *   hold, tied to the POOL_BITS constant.
+ *
+ * - entropy_avail - the number of bits of entropy currently in the
+ *   input pool. Always <= poolsize.
+ *
+ * - write_wakeup_threshold - the amount of entropy in the input pool
+ *   below which write polls to /dev/random will unblock, requesting
+ *   more entropy, tied to the POOL_MIN_BITS constant. It is writable
+ *   to avoid breaking old userspaces, but writing to it does not
+ *   change any behavior of the RNG.
+ *
+ * - urandom_min_reseed_secs - fixed to the meaningless value "60".
+ *   It is writable to avoid breaking old userspaces, but writing
+ *   to it does not change any behavior of the RNG.
  *
  ********************************************************************/
 
@@ -1696,8 +1721,8 @@ const struct file_operations urandom_fop
 
 #include <linux/sysctl.h>
 
-static int random_min_urandom_seed = 60;
-static int random_write_wakeup_bits = POOL_MIN_BITS;
+static int sysctl_random_min_urandom_seed = 60;
+static int sysctl_random_write_wakeup_bits = POOL_MIN_BITS;
 static int sysctl_poolsize = POOL_BITS;
 static char sysctl_bootid[16];
 
@@ -1755,14 +1780,14 @@ struct ctl_table random_table[] = {
 	},
 	{
 		.procname	= "write_wakeup_threshold",
-		.data		= &random_write_wakeup_bits,
+		.data		= &sysctl_random_write_wakeup_bits,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec,
 	},
 	{
 		.procname	= "urandom_min_reseed_secs",
-		.data		= &random_min_urandom_seed,
+		.data		= &sysctl_random_min_urandom_seed,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec,
author maintainer
#104 lore raw

PATCH 104/234 random: rewrite header introductory comment

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 5f75d9f3babea8ae0a2d06724656874f41d317f5 upstream.

Now that we've re-documented the various sections, we can remove the outdated text here and replace it with a high-level overview.

---
 drivers/char/random.c |  179 +++++---------------------------------------------
 1 file changed, 19 insertions(+), 160 deletions(-)
drivers/char/random.c +19 -160
@@ -2,168 +2,27 @@
 /*
  * Copyright (C) 2017-2022 Jason A. Donenfeld <[email protected]>. All Rights Reserved.
  * Copyright Matt Mackall <[email protected]>, 2003, 2004, 2005
- * Copyright Theodore Ts'o, 1994, 1995, 1996, 1997, 1998, 1999.  All
- * rights reserved.
- */
-
-/*
- * Exported interfaces ---- output
- * ===============================
- *
- * There are four exported interfaces; two for use within the kernel,
- * and two for use from userspace.
- *
- * Exported interfaces ---- userspace output
- * -----------------------------------------
- *
- * The userspace interfaces are two character devices /dev/random and
- * /dev/urandom.  /dev/random is suitable for use when very high
- * quality randomness is desired (for example, for key generation or
- * one-time pads), as it will only return a maximum of the number of
- * bits of randomness (as estimated by the random number generator)
- * contained in the entropy pool.
- *
- * The /dev/urandom device does not have this limit, and will return
- * as many bytes as are requested.  As more and more random bytes are
- * requested without giving time for the entropy pool to recharge,
- * this will result in random numbers that are merely cryptographically
- * strong.  For many applications, however, this is acceptable.
- *
- * Exported interfaces ---- kernel output
- * --------------------------------------
- *
- * The primary kernel interfaces are:
- *
- *	void get_random_bytes(void *buf, size_t nbytes);
- *	u32 get_random_u32()
- *	u64 get_random_u64()
- *	unsigned int get_random_int()
- *	unsigned long get_random_long()
- *
- * These interfaces will return the requested number of random bytes
- * into the given buffer or as a return value. This is equivalent to a
- * read from /dev/urandom. The get_random_{u32,u64,int,long}() family
- * of functions may be higher performance for one-off random integers,
- * because they do a bit of buffering.
- *
- * prandom_u32()
- * -------------
- *
- * For even weaker applications, see the pseudorandom generator
- * prandom_u32(), prandom_max(), and prandom_bytes().  If the random
- * numbers aren't security-critical at all, these are *far* cheaper.
- * Useful for self-tests, random error simulation, randomized backoffs,
- * and any other application where you trust that nobody is trying to
- * maliciously mess with you by guessing the "random" numbers.
- *
- * Exported interfaces ---- input
- * ==============================
- *
- * The current exported interfaces for gathering environmental noise
- * from the devices are:
- *
- *	void add_device_randomness(const void *buf, size_t size);
- *	void add_input_randomness(unsigned int type, unsigned int code,
- *                                unsigned int value);
- *	void add_interrupt_randomness(int irq);
- *	void add_disk_randomness(struct gendisk *disk);
- *	void add_hwgenerator_randomness(const void *buffer, size_t count,
- *					size_t entropy);
- *	void add_bootloader_randomness(const void *buf, size_t size);
- *
- * add_device_randomness() is for adding data to the random pool that
- * is likely to differ between two devices (or possibly even per boot).
- * This would be things like MAC addresses or serial numbers, or the
- * read-out of the RTC. This does *not* add any actual entropy to the
- * pool, but it initializes the pool to different values for devices
- * that might otherwise be identical and have very little entropy
- * available to them (particularly common in the embedded world).
- *
- * add_input_randomness() uses the input layer interrupt timing, as well as
- * the event type information from the hardware.
- *
- * add_interrupt_randomness() uses the interrupt timing as random
- * inputs to the entropy pool. Using the cycle counters and the irq source
- * as inputs, it feeds the randomness roughly once a second.
- *
- * add_disk_randomness() uses what amounts to the seek time of block
- * layer request events, on a per-disk_devt basis, as input to the
- * entropy pool. Note that high-speed solid state drives with very low
- * seek times do not make for good sources of entropy, as their seek
- * times are usually fairly consistent.
- *
- * All of these routines try to estimate how many bits of randomness a
- * particular randomness source.  They do this by keeping track of the
- * first and second order deltas of the event timings.
- *
- * add_hwgenerator_randomness() is for true hardware RNGs, and will credit
- * entropy as specified by the caller. If the entropy pool is full it will
- * block until more entropy is needed.
- *
- * add_bootloader_randomness() is the same as add_hwgenerator_randomness() or
- * add_device_randomness(), depending on whether or not the configuration
- * option CONFIG_RANDOM_TRUST_BOOTLOADER is set.
- *
- * Ensuring unpredictability at system startup
- * ============================================
- *
- * When any operating system starts up, it will go through a sequence
- * of actions that are fairly predictable by an adversary, especially
- * if the start-up does not involve interaction with a human operator.
- * This reduces the actual number of bits of unpredictability in the
- * entropy pool below the value in entropy_count.  In order to
- * counteract this effect, it helps to carry information in the
- * entropy pool across shut-downs and start-ups.  To do this, put the
- * following lines an appropriate script which is run during the boot
- * sequence:
- *
- *	echo "Initializing random number generator..."
- *	random_seed=/var/run/random-seed
- *	# Carry a random seed from start-up to start-up
- *	# Load and then save the whole entropy pool
- *	if [ -f $random_seed ]; then
- *		cat $random_seed >/dev/urandom
- *	else
- *		touch $random_seed
- *	fi
- *	chmod 600 $random_seed
- *	dd if=/dev/urandom of=$random_seed count=1 bs=512
- *
- * and the following lines in an appropriate script which is run as
- * the system is shutdown:
- *
- *	# Carry a random seed from shut-down to start-up
- *	# Save the whole entropy pool
- *	echo "Saving random seed..."
- *	random_seed=/var/run/random-seed
- *	touch $random_seed
- *	chmod 600 $random_seed
- *	dd if=/dev/urandom of=$random_seed count=1 bs=512
- *
- * For example, on most modern systems using the System V init
- * scripts, such code fragments would be found in
- * /etc/rc.d/init.d/random.  On older Linux systems, the correct script
- * location might be in /etc/rcb.d/rc.local or /etc/rc.d/rc.0.
- *
- * Effectively, these commands cause the contents of the entropy pool
- * to be saved at shut-down time and reloaded into the entropy pool at
- * start-up.  (The 'dd' in the addition to the bootup script is to
- * make sure that /etc/random-seed is different for every start-up,
- * even if the system crashes without executing rc.0.)  Even with
- * complete knowledge of the start-up activities, predicting the state
- * of the entropy pool requires knowledge of the previous history of
- * the system.
- *
- * Configuring the /dev/random driver under Linux
- * ==============================================
+ * Copyright Theodore Ts'o, 1994, 1995, 1996, 1997, 1998, 1999. All rights reserved.
  *
- * The /dev/random driver under Linux uses minor numbers 8 and 9 of
- * the /dev/mem major number (#1).  So if your system does not have
- * /dev/random and /dev/urandom created already, they can be created
- * by using the commands:
+ * This driver produces cryptographically secure pseudorandom data. It is divided
+ * into roughly six sections, each with a section header:
  *
- *	mknod /dev/random c 1 8
- *	mknod /dev/urandom c 1 9
+ *   - Initialization and readiness waiting.
+ *   - Fast key erasure RNG, the "crng".
+ *   - Entropy accumulation and extraction routines.
+ *   - Entropy collection routines.
+ *   - Userspace reader/writer interfaces.
+ *   - Sysctl interface.
+ *
+ * The high level overview is that there is one input pool, into which
+ * various pieces of data are hashed. Some of that data is then "credited" as
+ * having a certain number of bits of entropy. When enough bits of entropy are
+ * available, the hash is finalized and handed as a key to a stream cipher that
+ * expands it indefinitely for various consumers. This key is periodically
+ * refreshed as the various entropy collectors, described below, add data to the
+ * input pool and credit it. There is currently no Fortuna-like scheduler
+ * involved, which can lead to malicious entropy sources causing a premature
+ * reseed, and the entropy estimates are, at best, conservative guesses.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
author maintainer
#105 lore raw

PATCH 105/234 random: defer fast pool mixing to worker

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 58340f8e952b613e0ead0bed58b97b05bf4743c5 upstream.

On PREEMPT_RT, it's problematic to take spinlocks from hard irq handlers. We can fix this by deferring to a workqueue the dumping of the fast pool into the input pool.

We accomplish this with some careful rules on fast_pool->count:

- When it's incremented to >= 64, we schedule the work.
- If the top bit is set, we never schedule the work, even if >= 64.
- The worker is responsible for setting it back to 0 when it's done.

There are two small issues around using workqueues for this purpose that we work around.

The first issue is that mix_interrupt_randomness() might be migrated to another CPU during CPU hotplug. This issue is rectified by checking that it hasn't been migrated (after disabling irqs). If it has been migrated, then we set the count to zero, so that when the CPU comes online again, it can requeue the work. As part of this, we switch to using an atomic_t, so that the increment in the irq handler doesn't wipe out the zeroing if the CPU comes back online while this worker is running.

The second issue is that, though relatively minor in effect, we probably want to make sure we get a consistent view of the pool onto the stack, in case it's interrupted by an irq while reading. To do this, we don't reenable irqs until after the copy. There are only 18 instructions between the cli and sti, so this is a pretty tiny window.

---
 drivers/char/random.c |   63 ++++++++++++++++++++++++++++++++++++++------------
 1 file changed, 49 insertions(+), 14 deletions(-)
drivers/char/random.c +49 -14
@@ -1173,9 +1173,10 @@ struct fast_pool {
 		u32 pool32[4];
 		u64 pool64[2];
 	};
+	struct work_struct mix;
 	unsigned long last;
+	atomic_t count;
 	u16 reg_idx;
-	u8 count;
 };
 
 /*
@@ -1225,12 +1226,49 @@ static u32 get_reg(struct fast_pool *f,
 	return *ptr;
 }
 
+static void mix_interrupt_randomness(struct work_struct *work)
+{
+	struct fast_pool *fast_pool = container_of(work, struct fast_pool, mix);
+	u32 pool[4];
+
+	/* Check to see if we're running on the wrong CPU due to hotplug. */
+	local_irq_disable();
+	if (fast_pool != this_cpu_ptr(&irq_randomness)) {
+		local_irq_enable();
+		/*
+		 * If we are unlucky enough to have been moved to another CPU,
+		 * during CPU hotplug while the CPU was shutdown then we set
+		 * our count to zero atomically so that when the CPU comes
+		 * back online, it can enqueue work again. The _release here
+		 * pairs with the atomic_inc_return_acquire in
+		 * add_interrupt_randomness().
+		 */
+		atomic_set_release(&fast_pool->count, 0);
+		return;
+	}
+
+	/*
+	 * Copy the pool to the stack so that the mixer always has a
+	 * consistent view, before we reenable irqs again.
+	 */
+	memcpy(pool, fast_pool->pool32, sizeof(pool));
+	atomic_set(&fast_pool->count, 0);
+	fast_pool->last = jiffies;
+	local_irq_enable();
+
+	mix_pool_bytes(pool, sizeof(pool));
+	credit_entropy_bits(1);
+	memzero_explicit(pool, sizeof(pool));
+}
+
 void add_interrupt_randomness(int irq)
 {
+	enum { MIX_INFLIGHT = 1U << 31 };
 	struct fast_pool *fast_pool = this_cpu_ptr(&irq_randomness);
 	struct pt_regs *regs = get_irq_regs();
 	unsigned long now = jiffies;
 	cycles_t cycles = random_get_entropy();
+	unsigned int new_count;
 
 	if (cycles == 0)
 		cycles = get_reg(fast_pool, regs);
@@ -1250,12 +1288,13 @@ void add_interrupt_randomness(int irq)
 	}
 
 	fast_mix(fast_pool->pool32);
-	++fast_pool->count;
+	/* The _acquire here pairs with the atomic_set_release in mix_interrupt_randomness(). */
+	new_count = (unsigned int)atomic_inc_return_acquire(&fast_pool->count);
 
 	if (unlikely(crng_init == 0)) {
-		if (fast_pool->count >= 64 &&
+		if (new_count >= 64 &&
 		    crng_fast_load(fast_pool->pool32, sizeof(fast_pool->pool32)) > 0) {
-			fast_pool->count = 0;
+			atomic_set(&fast_pool->count, 0);
 			fast_pool->last = now;
 			if (spin_trylock(&input_pool.lock)) {
 				_mix_pool_bytes(&fast_pool->pool32, sizeof(fast_pool->pool32));
@@ -1265,20 +1304,16 @@ void add_interrupt_randomness(int irq)
 		return;
 	}
 
-	if ((fast_pool->count < 64) && !time_after(now, fast_pool->last + HZ))
+	if (new_count & MIX_INFLIGHT)
 		return;
 
-	if (!spin_trylock(&input_pool.lock))
+	if (new_count < 64 && !time_after(now, fast_pool->last + HZ))
 		return;
 
-	fast_pool->last = now;
-	_mix_pool_bytes(&fast_pool->pool32, sizeof(fast_pool->pool32));
-	spin_unlock(&input_pool.lock);
-
-	fast_pool->count = 0;
-
-	/* Award one bit for the contents of the fast pool. */
-	credit_entropy_bits(1);
+	if (unlikely(!fast_pool->mix.func))
+		INIT_WORK(&fast_pool->mix, mix_interrupt_randomness);
+	atomic_or(MIX_INFLIGHT, &fast_pool->count);
+	queue_work_on(raw_smp_processor_id(), system_highpri_wq, &fast_pool->mix);
 }
 EXPORT_SYMBOL_GPL(add_interrupt_randomness);
 
author maintainer
#106 lore raw

PATCH 106/234 random: do not take pool spinlock at boot

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit afba0b80b977b2a8f16234f2acd982f82710ba33 upstream.

Since rand_initialize() is run while interrupts are still off and nothing else is running, we don't need to repeatedly take and release the pool spinlock, especially in the RDSEED loop.

---
 drivers/char/random.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/char/random.c +3 -3
@@ -973,10 +973,10 @@ int __init rand_initialize(void)
 			rv = random_get_entropy();
 			arch_init = false;
 		}
-		mix_pool_bytes(&rv, sizeof(rv));
+		_mix_pool_bytes(&rv, sizeof(rv));
 	}
-	mix_pool_bytes(&now, sizeof(now));
-	mix_pool_bytes(utsname(), sizeof(*(utsname())));
+	_mix_pool_bytes(&now, sizeof(now));
+	_mix_pool_bytes(utsname(), sizeof(*(utsname())));
 
 	extract_entropy(base_crng.key, sizeof(base_crng.key));
 	++base_crng.generation;
author maintainer
#107 lore raw

PATCH 107/234 random: unify early init crng load accounting

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit da792c6d5f59a76c10a310c5d4c93428fd18f996 upstream.

crng_fast_load() and crng_slow_load() have different semantics:

- crng_fast_load() xors and accounts with crng_init_cnt.
- crng_slow_load() hashes and doesn't account.

However add_hwgenerator_randomness() can afford to hash (it's called from a kthread), and it should account. Additionally, ones that can afford to hash don't need to take a trylock but can take a normal lock. So, we combine these into one function, crng_pre_init_inject(), which allows us to control these in a uniform way. This will make it simpler later to simplify this all down when the time comes for that.

---
 drivers/char/random.c |  114 +++++++++++++++++++++++++-------------------------
 1 file changed, 59 insertions(+), 55 deletions(-)
drivers/char/random.c +59 -55
@@ -384,7 +384,7 @@ static void crng_make_state(u32 chacha_s
 	 * For the fast path, we check whether we're ready, unlocked first, and
 	 * then re-check once locked later. In the case where we're really not
 	 * ready, we do fast key erasure with the base_crng directly, because
-	 * this is what crng_{fast,slow}_load mutate during early init.
+	 * this is what crng_pre_init_inject() mutates during early init.
 	 */
 	if (unlikely(!crng_ready())) {
 		bool ready;
@@ -435,72 +435,75 @@ static void crng_make_state(u32 chacha_s
 }
 
 /*
- * This function is for crng_init == 0 only.
+ * This function is for crng_init == 0 only. It loads entropy directly
+ * into the crng's key, without going through the input pool. It is,
+ * generally speaking, not very safe, but we use this only at early
+ * boot time when it's better to have something there rather than
+ * nothing.
+ *
+ * There are two paths, a slow one and a fast one. The slow one
+ * hashes the input along with the current key. The fast one simply
+ * xors it in, and should only be used from interrupt context.
+ *
+ * If account is set, then the crng_init_cnt counter is incremented.
+ * This shouldn't be set by functions like add_device_randomness(),
+ * where we can't trust the buffer passed to it is guaranteed to be
+ * unpredictable (so it might not have any entropy at all).
  *
- * crng_fast_load() can be called by code in the interrupt service
- * path.  So we can't afford to dilly-dally. Returns the number of
- * bytes processed from cp.
+ * Returns the number of bytes processed from input, which is bounded
+ * by CRNG_INIT_CNT_THRESH if account is true.
  */
-static size_t crng_fast_load(const void *cp, size_t len)
+static size_t crng_pre_init_inject(const void *input, size_t len,
+				   bool fast, bool account)
 {
 	static int crng_init_cnt = 0;
 	unsigned long flags;
-	const u8 *src = (const u8 *)cp;
-	size_t ret = 0;
 
-	if (!spin_trylock_irqsave(&base_crng.lock, flags))
-		return 0;
+	if (fast) {
+		if (!spin_trylock_irqsave(&base_crng.lock, flags))
+			return 0;
+	} else {
+		spin_lock_irqsave(&base_crng.lock, flags);
+	}
+
 	if (crng_init != 0) {
 		spin_unlock_irqrestore(&base_crng.lock, flags);
 		return 0;
 	}
-	while (len > 0 && crng_init_cnt < CRNG_INIT_CNT_THRESH) {
-		base_crng.key[crng_init_cnt % sizeof(base_crng.key)] ^= *src;
-		src++; crng_init_cnt++; len--; ret++;
-	}
-	if (crng_init_cnt >= CRNG_INIT_CNT_THRESH) {
-		++base_crng.generation;
-		crng_init = 1;
-	}
-	spin_unlock_irqrestore(&base_crng.lock, flags);
-	if (crng_init == 1)
-		pr_notice("fast init done\n");
-	return ret;
-}
 
-/*
- * This function is for crng_init == 0 only.
- *
- * crng_slow_load() is called by add_device_randomness, which has two
- * attributes.  (1) We can't trust the buffer passed to it is
- * guaranteed to be unpredictable (so it might not have any entropy at
- * all), and (2) it doesn't have the performance constraints of
- * crng_fast_load().
- *
- * So, we simply hash the contents in with the current key. Finally,
- * we do *not* advance crng_init_cnt since buffer we may get may be
- * something like a fixed DMI table (for example), which might very
- * well be unique to the machine, but is otherwise unvarying.
- */
-static void crng_slow_load(const void *cp, size_t len)
-{
-	unsigned long flags;
-	struct blake2s_state hash;
+	if (account)
+		len = min_t(size_t, len, CRNG_INIT_CNT_THRESH - crng_init_cnt);
 
-	blake2s_init(&hash, sizeof(base_crng.key));
-
-	if (!spin_trylock_irqsave(&base_crng.lock, flags))
-		return;
-	if (crng_init != 0) {
-		spin_unlock_irqrestore(&base_crng.lock, flags);
-		return;
+	if (fast) {
+		const u8 *src = input;
+		size_t i;
+
+		for (i = 0; i < len; ++i)
+			base_crng.key[(crng_init_cnt + i) %
+				      sizeof(base_crng.key)] ^= src[i];
+	} else {
+		struct blake2s_state hash;
+
+		blake2s_init(&hash, sizeof(base_crng.key));
+		blake2s_update(&hash, base_crng.key, sizeof(base_crng.key));
+		blake2s_update(&hash, input, len);
+		blake2s_final(&hash, base_crng.key);
+	}
+
+	if (account) {
+		crng_init_cnt += len;
+		if (crng_init_cnt >= CRNG_INIT_CNT_THRESH) {
+			++base_crng.generation;
+			crng_init = 1;
+		}
 	}
 
-	blake2s_update(&hash, base_crng.key, sizeof(base_crng.key));
-	blake2s_update(&hash, cp, len);
-	blake2s_final(&hash, base_crng.key);
-
 	spin_unlock_irqrestore(&base_crng.lock, flags);
+
+	if (crng_init == 1)
+		pr_notice("fast init done\n");
+
+	return len;
 }
 
 static void _get_random_bytes(void *buf, size_t nbytes)
@@ -1013,7 +1016,7 @@ void add_device_randomness(const void *b
 	unsigned long flags;
 
 	if (!crng_ready() && size)
-		crng_slow_load(buf, size);
+		crng_pre_init_inject(buf, size, false, false);
 
 	spin_lock_irqsave(&input_pool.lock, flags);
 	_mix_pool_bytes(buf, size);
@@ -1130,7 +1133,7 @@ void add_hwgenerator_randomness(const vo
 				size_t entropy)
 {
 	if (unlikely(crng_init == 0)) {
-		size_t ret = crng_fast_load(buffer, count);
+		size_t ret = crng_pre_init_inject(buffer, count, false, true);
 		mix_pool_bytes(buffer, ret);
 		count -= ret;
 		buffer += ret;
@@ -1293,7 +1296,8 @@ void add_interrupt_randomness(int irq)
 
 	if (unlikely(crng_init == 0)) {
 		if (new_count >= 64 &&
-		    crng_fast_load(fast_pool->pool32, sizeof(fast_pool->pool32)) > 0) {
+		    crng_pre_init_inject(fast_pool->pool32, sizeof(fast_pool->pool32),
+					 true, true) > 0) {
 			atomic_set(&fast_pool->count, 0);
 			fast_pool->last = now;
 			if (spin_trylock(&input_pool.lock)) {
author maintainer
#108 lore raw

PATCH 108/234 random: check for crng_init == 0 in add_device_randomness()

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 1daf2f387652bf3a7044aea042f5023b3f6b189b upstream.

This has no real functional change, as crng_pre_init_inject() (and before that, crng_slow_init()) always checks for == 0, not >= 2. So correct the outer unlocked change to reflect that. Before this used crng_ready(), which was not correct.

---
 drivers/char/random.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/char/random.c +1 -1
@@ -1015,7 +1015,7 @@ void add_device_randomness(const void *b
 	unsigned long time = random_get_entropy() ^ jiffies;
 	unsigned long flags;
 
-	if (!crng_ready() && size)
+	if (crng_init == 0 && size)
 		crng_pre_init_inject(buf, size, false, false);
 
 	spin_lock_irqsave(&input_pool.lock, flags);
author maintainer
#109 lore raw

PATCH 109/234 random: pull add_hwgenerator_randomness() declaration into random.h

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit b777c38239fec5a528e59f55b379e31b1a187524 upstream.

add_hwgenerator_randomness() is a function implemented and documented inside of random.c. It is the way that hardware RNGs push data into it. Therefore, it should be declared in random.h. Otherwise sparse complains with:

random.c:1137:6: warning: symbol 'add_hwgenerator_randomness' was not declared. Should it be static?

The alternative would be to include hw_random.h into random.c, but that wouldn't really be good for anything except slowing down compile time.

---
 drivers/char/hw_random/core.c |    1 +
 include/linux/hw_random.h     |    2 --
 include/linux/random.h        |    2 ++
 3 files changed, 3 insertions(+), 2 deletions(-)
drivers/char/hw_random/core.c +1 -0
@@ -15,6 +15,7 @@
 #include <linux/err.h>
 #include <linux/fs.h>
 #include <linux/hw_random.h>
+#include <linux/random.h>
 #include <linux/kernel.h>
 #include <linux/kthread.h>
 #include <linux/sched/signal.h>
include/linux/hw_random.h +0 -2
@@ -59,7 +59,5 @@ extern int devm_hwrng_register(struct de
 /** Unregister a Hardware Random Number Generator driver. */
 extern void hwrng_unregister(struct hwrng *rng);
 extern void devm_hwrng_unregister(struct device *dve, struct hwrng *rng);
-/** Feed random bits into the pool. */
-extern void add_hwgenerator_randomness(const void *buffer, size_t count, size_t entropy);
 
 #endif /* LINUX_HWRANDOM_H_ */
include/linux/random.h +2 -0
@@ -32,6 +32,8 @@ static inline void add_latent_entropy(vo
 extern void add_input_randomness(unsigned int type, unsigned int code,
 				 unsigned int value) __latent_entropy;
 extern void add_interrupt_randomness(int irq) __latent_entropy;
+extern void add_hwgenerator_randomness(const void *buffer, size_t count,
+				       size_t entropy);
 
 extern void get_random_bytes(void *buf, size_t nbytes);
 extern int wait_for_random_bytes(void);
author maintainer
#110 lore raw

PATCH 110/234 random: clear fast pool, crng, and batches in cpuhp bring up

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 3191dd5a1179ef0fad5a050a1702ae98b6251e8f upstream.

For the irq randomness fast pool, rather than having to use expensive atomics, which were visibly the most expensive thing in the entire irq handler, simply take care of the extreme edge case of resetting count to zero in the cpuhp online handler, just after workqueues have been reenabled. This simplifies the code a bit and lets us use vanilla variables rather than atomics, and performance should be improved.

As well, very early on when the CPU comes up, while interrupts are still disabled, we clear out the per-cpu crng and its batches, so that it always starts with fresh randomness.

---
 drivers/char/random.c      |   62 ++++++++++++++++++++++++++++++++++-----------
 include/linux/cpuhotplug.h |    2 +
 include/linux/random.h     |    5 +++
 kernel/cpu.c               |   11 +++++++
 4 files changed, 65 insertions(+), 15 deletions(-)
drivers/char/random.c +47 -15
@@ -693,6 +693,25 @@ u32 get_random_u32(void)
 }
 EXPORT_SYMBOL(get_random_u32);
 
+#ifdef CONFIG_SMP
+/*
+ * This function is called when the CPU is coming up, with entry
+ * CPUHP_RANDOM_PREPARE, which comes before CPUHP_WORKQUEUE_PREP.
+ */
+int random_prepare_cpu(unsigned int cpu)
+{
+	/*
+	 * When the cpu comes back online, immediately invalidate both
+	 * the per-cpu crng and all batches, so that we serve fresh
+	 * randomness.
+	 */
+	per_cpu_ptr(&crngs, cpu)->generation = ULONG_MAX;
+	per_cpu_ptr(&batched_entropy_u32, cpu)->position = UINT_MAX;
+	per_cpu_ptr(&batched_entropy_u64, cpu)->position = UINT_MAX;
+	return 0;
+}
+#endif
+
 /**
  * randomize_page - Generate a random, page aligned address
  * @start:	The smallest acceptable address the caller will take.
@@ -1178,7 +1197,7 @@ struct fast_pool {
 	};
 	struct work_struct mix;
 	unsigned long last;
-	atomic_t count;
+	unsigned int count;
 	u16 reg_idx;
 };
 
@@ -1214,6 +1233,29 @@ static void fast_mix(u32 pool[4])
 
 static DEFINE_PER_CPU(struct fast_pool, irq_randomness);
 
+#ifdef CONFIG_SMP
+/*
+ * This function is called when the CPU has just come online, with
+ * entry CPUHP_AP_RANDOM_ONLINE, just after CPUHP_AP_WORKQUEUE_ONLINE.
+ */
+int random_online_cpu(unsigned int cpu)
+{
+	/*
+	 * During CPU shutdown and before CPU onlining, add_interrupt_
+	 * randomness() may schedule mix_interrupt_randomness(), and
+	 * set the MIX_INFLIGHT flag. However, because the worker can
+	 * be scheduled on a different CPU during this period, that
+	 * flag will never be cleared. For that reason, we zero out
+	 * the flag here, which runs just after workqueues are onlined
+	 * for the CPU again. This also has the effect of setting the
+	 * irq randomness count to zero so that new accumulated irqs
+	 * are fresh.
+	 */
+	per_cpu_ptr(&irq_randomness, cpu)->count = 0;
+	return 0;
+}
+#endif
+
 static u32 get_reg(struct fast_pool *f, struct pt_regs *regs)
 {
 	u32 *ptr = (u32 *)regs;
@@ -1238,15 +1280,6 @@ static void mix_interrupt_randomness(str
 	local_irq_disable();
 	if (fast_pool != this_cpu_ptr(&irq_randomness)) {
 		local_irq_enable();
-		/*
-		 * If we are unlucky enough to have been moved to another CPU,
-		 * during CPU hotplug while the CPU was shutdown then we set
-		 * our count to zero atomically so that when the CPU comes
-		 * back online, it can enqueue work again. The _release here
-		 * pairs with the atomic_inc_return_acquire in
-		 * add_interrupt_randomness().
-		 */
-		atomic_set_release(&fast_pool->count, 0);
 		return;
 	}
 
@@ -1255,7 +1288,7 @@ static void mix_interrupt_randomness(str
 	 * consistent view, before we reenable irqs again.
 	 */
 	memcpy(pool, fast_pool->pool32, sizeof(pool));
-	atomic_set(&fast_pool->count, 0);
+	fast_pool->count = 0;
 	fast_pool->last = jiffies;
 	local_irq_enable();
 
@@ -1291,14 +1324,13 @@ void add_interrupt_randomness(int irq)
 	}
 
 	fast_mix(fast_pool->pool32);
-	/* The _acquire here pairs with the atomic_set_release in mix_interrupt_randomness(). */
-	new_count = (unsigned int)atomic_inc_return_acquire(&fast_pool->count);
+	new_count = ++fast_pool->count;
 
 	if (unlikely(crng_init == 0)) {
 		if (new_count >= 64 &&
 		    crng_pre_init_inject(fast_pool->pool32, sizeof(fast_pool->pool32),
 					 true, true) > 0) {
-			atomic_set(&fast_pool->count, 0);
+			fast_pool->count = 0;
 			fast_pool->last = now;
 			if (spin_trylock(&input_pool.lock)) {
 				_mix_pool_bytes(&fast_pool->pool32, sizeof(fast_pool->pool32));
@@ -1316,7 +1348,7 @@ void add_interrupt_randomness(int irq)
 
 	if (unlikely(!fast_pool->mix.func))
 		INIT_WORK(&fast_pool->mix, mix_interrupt_randomness);
-	atomic_or(MIX_INFLIGHT, &fast_pool->count);
+	fast_pool->count |= MIX_INFLIGHT;
 	queue_work_on(raw_smp_processor_id(), system_highpri_wq, &fast_pool->mix);
 }
 EXPORT_SYMBOL_GPL(add_interrupt_randomness);
include/linux/cpuhotplug.h +2 -0
@@ -60,6 +60,7 @@ enum cpuhp_state {
 	CPUHP_LUSTRE_CFS_DEAD,
 	CPUHP_AP_ARM_CACHE_B15_RAC_DEAD,
 	CPUHP_PADATA_DEAD,
+	CPUHP_RANDOM_PREPARE,
 	CPUHP_WORKQUEUE_PREP,
 	CPUHP_POWER_NUMA_PREPARE,
 	CPUHP_HRTIMERS_PREPARE,
@@ -171,6 +172,7 @@ enum cpuhp_state {
 	CPUHP_AP_PERF_POWERPC_THREAD_IMC_ONLINE,
 	CPUHP_AP_WATCHDOG_ONLINE,
 	CPUHP_AP_WORKQUEUE_ONLINE,
+	CPUHP_AP_RANDOM_ONLINE,
 	CPUHP_AP_RCUTREE_ONLINE,
 	CPUHP_AP_BASE_CACHEINFO_ONLINE,
 	CPUHP_AP_ONLINE_DYN,
include/linux/random.h +5 -0
@@ -156,4 +156,9 @@ static inline bool __init arch_get_rando
 }
 #endif
 
+#ifdef CONFIG_SMP
+extern int random_prepare_cpu(unsigned int cpu);
+extern int random_online_cpu(unsigned int cpu);
+#endif
+
 #endif /* _LINUX_RANDOM_H */
kernel/cpu.c +11 -0
@@ -32,6 +32,7 @@
 #include <linux/slab.h>
 #include <linux/percpu-rwsem.h>
 #include <linux/cpuset.h>
+#include <linux/random.h>
 
 #include <trace/events/power.h>
 #define CREATE_TRACE_POINTS
@@ -1404,6 +1405,11 @@ static struct cpuhp_step cpuhp_hp_states
 		.startup.single		= perf_event_init_cpu,
 		.teardown.single	= perf_event_exit_cpu,
 	},
+	[CPUHP_RANDOM_PREPARE] = {
+		.name			= "random:prepare",
+		.startup.single		= random_prepare_cpu,
+		.teardown.single	= NULL,
+	},
 	[CPUHP_WORKQUEUE_PREP] = {
 		.name			= "workqueue:prepare",
 		.startup.single		= workqueue_prepare_cpu,
@@ -1520,6 +1526,11 @@ static struct cpuhp_step cpuhp_hp_states
 		.startup.single		= workqueue_online_cpu,
 		.teardown.single	= workqueue_offline_cpu,
 	},
+	[CPUHP_AP_RANDOM_ONLINE] = {
+		.name			= "random:online",
+		.startup.single		= random_online_cpu,
+		.teardown.single	= NULL,
+	},
 	[CPUHP_AP_RCUTREE_ONLINE] = {
 		.name			= "RCU/tree:online",
 		.startup.single		= rcutree_online_cpu,
author maintainer
#111 lore raw

PATCH 111/234 random: round-robin registers as ulong, not u32

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit da3951ebdcd1cb1d5c750e08cd05aee7b0c04d9a upstream.

When the interrupt handler does not have a valid cycle counter, it calls get_reg() to read a register from the irq stack, in round-robin. Currently it does this assuming that registers are 32-bit. This is _probably_ the case, and probably all platforms without cycle counters are in fact 32-bit platforms. But maybe not, and either way, it's not quite correct. This commit fixes that to deal with unsigned long rather than u32.

---
 drivers/char/random.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/char/random.c +3 -3
@@ -1256,15 +1256,15 @@ int random_online_cpu(unsigned int cpu)
 }
 #endif
 
-static u32 get_reg(struct fast_pool *f, struct pt_regs *regs)
+static unsigned long get_reg(struct fast_pool *f, struct pt_regs *regs)
 {
-	u32 *ptr = (u32 *)regs;
+	unsigned long *ptr = (unsigned long *)regs;
 	unsigned int idx;
 
 	if (regs == NULL)
 		return 0;
 	idx = READ_ONCE(f->reg_idx);
-	if (idx >= sizeof(struct pt_regs) / sizeof(u32))
+	if (idx >= sizeof(struct pt_regs) / sizeof(unsigned long))
 		idx = 0;
 	ptr += idx++;
 	WRITE_ONCE(f->reg_idx, idx);
author maintainer
#112 lore raw

PATCH 112/234 random: only wake up writers after zap if threshold was passed

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit a3f9e8910e1584d7725ef7d5ac870920d42d0bb4 upstream.

The only time that we need to wake up /dev/random writers on RNDCLEARPOOL/RNDZAPPOOL is when we're changing from a value that is greater than or equal to POOL_MIN_BITS to zero, because if we're changing from below POOL_MIN_BITS to zero, the writers are already unblocked.

---
 drivers/char/random.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/char/random.c +1 -1
@@ -1577,7 +1577,7 @@ static long random_ioctl(struct file *f,
 		 */
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
-		if (xchg(&input_pool.entropy_count, 0)) {
+		if (xchg(&input_pool.entropy_count, 0) >= POOL_MIN_BITS) {
 			wake_up_interruptible(&random_write_wait);
 			kill_fasync(&fasync, SIGIO, POLL_OUT);
 		}
author maintainer
#113 lore raw

PATCH 113/234 random: cleanup UUID handling

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 64276a9939ff414f2f0db38036cf4e1a0a703394 upstream.

Rather than hard coding various lengths, we can use the right constants. Strings should be char * while buffers should be u8 *. Rather than have a nonsensical and unused maxlength, just remove it. Finally, use snprintf instead of sprintf, just out of good hygiene.

As well, remove the old comment about returning a binary UUID via the binary sysctl syscall. That syscall was removed from the kernel in 5.5, and actually, the "uuid_strategy" function and related infrastructure for even serving it via the binary sysctl syscall was removed with 894d2491153a ("sysctl drivers: Remove dead binary sysctl support") back in 2.6.33.

---
 drivers/char/random.c |   29 +++++++++++++----------------
 1 file changed, 13 insertions(+), 16 deletions(-)
drivers/char/random.c +13 -16
@@ -1654,22 +1654,25 @@ const struct file_operations urandom_fop
 static int sysctl_random_min_urandom_seed = 60;
 static int sysctl_random_write_wakeup_bits = POOL_MIN_BITS;
 static int sysctl_poolsize = POOL_BITS;
-static char sysctl_bootid[16];
+static u8 sysctl_bootid[UUID_SIZE];
 
 /*
  * This function is used to return both the bootid UUID, and random
- * UUID.  The difference is in whether table->data is NULL; if it is,
+ * UUID. The difference is in whether table->data is NULL; if it is,
  * then a new UUID is generated and returned to the user.
- *
- * If the user accesses this via the proc interface, the UUID will be
- * returned as an ASCII string in the standard UUID format; if via the
- * sysctl system call, as 16 bytes of binary data.
  */
 static int proc_do_uuid(struct ctl_table *table, int write,
 			void __user *buffer, size_t *lenp, loff_t *ppos)
 {
-	struct ctl_table fake_table;
-	unsigned char buf[64], tmp_uuid[16], *uuid;
+	u8 tmp_uuid[UUID_SIZE], *uuid;
+	char uuid_string[UUID_STRING_LEN + 1];
+	struct ctl_table fake_table = {
+		.data = uuid_string,
+		.maxlen = UUID_STRING_LEN
+	};
+
+	if (write)
+		return -EPERM;
 
 	uuid = table->data;
 	if (!uuid) {
@@ -1684,12 +1687,8 @@ static int proc_do_uuid(struct ctl_table
 		spin_unlock(&bootid_spinlock);
 	}
 
-	sprintf(buf, "%pU", uuid);
-
-	fake_table.data = buf;
-	fake_table.maxlen = sizeof(buf);
-
-	return proc_dostring(&fake_table, write, buffer, lenp, ppos);
+	snprintf(uuid_string, sizeof(uuid_string), "%pU", uuid);
+	return proc_dostring(&fake_table, 0, buffer, lenp, ppos);
 }
 
 extern struct ctl_table random_table[];
@@ -1725,13 +1724,11 @@ struct ctl_table random_table[] = {
 	{
 		.procname	= "boot_id",
 		.data		= &sysctl_bootid,
-		.maxlen		= 16,
 		.mode		= 0444,
 		.proc_handler	= proc_do_uuid,
 	},
 	{
 		.procname	= "uuid",
-		.maxlen		= 16,
 		.mode		= 0444,
 		.proc_handler	= proc_do_uuid,
 	},
author maintainer
#114 lore raw

PATCH 114/234 random: unify cycles_t and jiffies usage and types

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit abded93ec1e9692920fe309f07f40bd1035f2940 upstream.

random_get_entropy() returns a cycles_t, not an unsigned long, which is sometimes 64 bits on various 32-bit platforms, including x86. Conversely, jiffies is always unsigned long. This commit fixes things to use cycles_t for fields that use random_get_entropy(), named "cycles", and unsigned long for fields that use jiffies, named "now". It's also good to mix in a cycles_t and a jiffies in the same way for both add_device_randomness and add_timer_randomness, rather than using xor in one case. Finally, we unify the order of these volatile reads, always reading the more precise cycles counter, and then jiffies, so that the cycle counter is as close to the event as possible.

---
 drivers/char/random.c |   56 ++++++++++++++++++++++++--------------------------
 1 file changed, 27 insertions(+), 29 deletions(-)
drivers/char/random.c +27 -29
@@ -1015,12 +1015,6 @@ int __init rand_initialize(void)
 	return 0;
 }
 
-/* There is one of these per entropy source */
-struct timer_rand_state {
-	cycles_t last_time;
-	long last_delta, last_delta2;
-};
-
 /*
  * Add device- or boot-specific data to the input pool to help
  * initialize it.
@@ -1031,19 +1025,26 @@ struct timer_rand_state {
  */
 void add_device_randomness(const void *buf, size_t size)
 {
-	unsigned long time = random_get_entropy() ^ jiffies;
-	unsigned long flags;
+	cycles_t cycles = random_get_entropy();
+	unsigned long flags, now = jiffies;
 
 	if (crng_init == 0 && size)
 		crng_pre_init_inject(buf, size, false, false);
 
 	spin_lock_irqsave(&input_pool.lock, flags);
+	_mix_pool_bytes(&cycles, sizeof(cycles));
+	_mix_pool_bytes(&now, sizeof(now));
 	_mix_pool_bytes(buf, size);
-	_mix_pool_bytes(&time, sizeof(time));
 	spin_unlock_irqrestore(&input_pool.lock, flags);
 }
 EXPORT_SYMBOL(add_device_randomness);
 
+/* There is one of these per entropy source */
+struct timer_rand_state {
+	unsigned long last_time;
+	long last_delta, last_delta2;
+};
+
 /*
  * This function adds entropy to the entropy "pool" by using timing
  * delays.  It uses the timer_rand_state structure to make an estimate
@@ -1052,29 +1053,26 @@ EXPORT_SYMBOL(add_device_randomness);
  * The number "num" is also added to the pool - it should somehow describe
  * the type of event which just happened.  This is currently 0-255 for
  * keyboard scan codes, and 256 upwards for interrupts.
- *
  */
 static void add_timer_randomness(struct timer_rand_state *state, unsigned int num)
 {
-	struct {
-		long jiffies;
-		unsigned int cycles;
-		unsigned int num;
-	} sample;
+	cycles_t cycles = random_get_entropy();
+	unsigned long flags, now = jiffies;
 	long delta, delta2, delta3;
 
-	sample.jiffies = jiffies;
-	sample.cycles = random_get_entropy();
-	sample.num = num;
-	mix_pool_bytes(&sample, sizeof(sample));
+	spin_lock_irqsave(&input_pool.lock, flags);
+	_mix_pool_bytes(&cycles, sizeof(cycles));
+	_mix_pool_bytes(&now, sizeof(now));
+	_mix_pool_bytes(&num, sizeof(num));
+	spin_unlock_irqrestore(&input_pool.lock, flags);
 
 	/*
 	 * Calculate number of bits of randomness we probably added.
 	 * We take into account the first, second and third-order deltas
 	 * in order to make our estimate.
 	 */
-	delta = sample.jiffies - READ_ONCE(state->last_time);
-	WRITE_ONCE(state->last_time, sample.jiffies);
+	delta = now - READ_ONCE(state->last_time);
+	WRITE_ONCE(state->last_time, now);
 
 	delta2 = delta - READ_ONCE(state->last_delta);
 	WRITE_ONCE(state->last_delta, delta);
@@ -1300,10 +1298,10 @@ static void mix_interrupt_randomness(str
 void add_interrupt_randomness(int irq)
 {
 	enum { MIX_INFLIGHT = 1U << 31 };
+	cycles_t cycles = random_get_entropy();
+	unsigned long now = jiffies;
 	struct fast_pool *fast_pool = this_cpu_ptr(&irq_randomness);
 	struct pt_regs *regs = get_irq_regs();
-	unsigned long now = jiffies;
-	cycles_t cycles = random_get_entropy();
 	unsigned int new_count;
 
 	if (cycles == 0)
@@ -1378,28 +1376,28 @@ static void entropy_timer(struct timer_l
 static void try_to_generate_entropy(void)
 {
 	struct {
-		unsigned long now;
+		cycles_t cycles;
 		struct timer_list timer;
 	} stack;
 
-	stack.now = random_get_entropy();
+	stack.cycles = random_get_entropy();
 
 	/* Slow counter - or none. Don't even bother */
-	if (stack.now == random_get_entropy())
+	if (stack.cycles == random_get_entropy())
 		return;
 
 	timer_setup_on_stack(&stack.timer, entropy_timer, 0);
 	while (!crng_ready()) {
 		if (!timer_pending(&stack.timer))
 			mod_timer(&stack.timer, jiffies + 1);
-		mix_pool_bytes(&stack.now, sizeof(stack.now));
+		mix_pool_bytes(&stack.cycles, sizeof(stack.cycles));
 		schedule();
-		stack.now = random_get_entropy();
+		stack.cycles = random_get_entropy();
 	}
 
 	del_timer_sync(&stack.timer);
 	destroy_timer_on_stack(&stack.timer);
-	mix_pool_bytes(&stack.now, sizeof(stack.now));
+	mix_pool_bytes(&stack.cycles, sizeof(stack.cycles));
 }
 
 
author maintainer
#115 lore raw

PATCH 115/234 random: do crng pre-init loading in worker rather than irq

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit c2a7de4feb6e09f23af7accc0f882a8fa92e7ae5 upstream.

Taking spinlocks from IRQ context is generally problematic for PREEMPT_RT. That is, in part, why we take trylocks instead. However, a spin_try_lock() is also problematic since another spin_lock() invocation can potentially PI-boost the wrong task, as the spin_try_lock() is invoked from an IRQ-context, so the task on CPU (random task or idle) is not the actual owner.

Additionally, by deferring the crng pre-init loading to the worker, we can use the cryptographic hash function rather than xor, which is perhaps a meaningful difference when considering this data has only been through the relatively weak fast_mix() function.

The biggest downside of this approach is that the pre-init loading is now deferred until later, which means things that need random numbers after interrupts are enabled, but before workqueues are running -- or before this particular worker manages to run -- are going to get into trouble. Hopefully in the real world, this window is rather small, especially since this code won't run until 64 interrupts had occurred.

---
 drivers/char/random.c |   65 ++++++++++++++------------------------------------
 1 file changed, 19 insertions(+), 46 deletions(-)
drivers/char/random.c +19 -46
@@ -441,10 +441,6 @@ static void crng_make_state(u32 chacha_s
  * boot time when it's better to have something there rather than
  * nothing.
  *
- * There are two paths, a slow one and a fast one. The slow one
- * hashes the input along with the current key. The fast one simply
- * xors it in, and should only be used from interrupt context.
- *
  * If account is set, then the crng_init_cnt counter is incremented.
  * This shouldn't be set by functions like add_device_randomness(),
  * where we can't trust the buffer passed to it is guaranteed to be
@@ -453,19 +449,15 @@ static void crng_make_state(u32 chacha_s
  * Returns the number of bytes processed from input, which is bounded
  * by CRNG_INIT_CNT_THRESH if account is true.
  */
-static size_t crng_pre_init_inject(const void *input, size_t len,
-				   bool fast, bool account)
+static size_t crng_pre_init_inject(const void *input, size_t len, bool account)
 {
 	static int crng_init_cnt = 0;
+	struct blake2s_state hash;
 	unsigned long flags;
 
-	if (fast) {
-		if (!spin_trylock_irqsave(&base_crng.lock, flags))
-			return 0;
-	} else {
-		spin_lock_irqsave(&base_crng.lock, flags);
-	}
+	blake2s_init(&hash, sizeof(base_crng.key));
 
+	spin_lock_irqsave(&base_crng.lock, flags);
 	if (crng_init != 0) {
 		spin_unlock_irqrestore(&base_crng.lock, flags);
 		return 0;
@@ -474,21 +466,9 @@ static size_t crng_pre_init_inject(const
 	if (account)
 		len = min_t(size_t, len, CRNG_INIT_CNT_THRESH - crng_init_cnt);
 
-	if (fast) {
-		const u8 *src = input;
-		size_t i;
-
-		for (i = 0; i < len; ++i)
-			base_crng.key[(crng_init_cnt + i) %
-				      sizeof(base_crng.key)] ^= src[i];
-	} else {
-		struct blake2s_state hash;
-
-		blake2s_init(&hash, sizeof(base_crng.key));
-		blake2s_update(&hash, base_crng.key, sizeof(base_crng.key));
-		blake2s_update(&hash, input, len);
-		blake2s_final(&hash, base_crng.key);
-	}
+	blake2s_update(&hash, base_crng.key, sizeof(base_crng.key));
+	blake2s_update(&hash, input, len);
+	blake2s_final(&hash, base_crng.key);
 
 	if (account) {
 		crng_init_cnt += len;
@@ -1029,7 +1009,7 @@ void add_device_randomness(const void *b
 	unsigned long flags, now = jiffies;
 
 	if (crng_init == 0 && size)
-		crng_pre_init_inject(buf, size, false, false);
+		crng_pre_init_inject(buf, size, false);
 
 	spin_lock_irqsave(&input_pool.lock, flags);
 	_mix_pool_bytes(&cycles, sizeof(cycles));
@@ -1150,7 +1130,7 @@ void add_hwgenerator_randomness(const vo
 				size_t entropy)
 {
 	if (unlikely(crng_init == 0)) {
-		size_t ret = crng_pre_init_inject(buffer, count, false, true);
+		size_t ret = crng_pre_init_inject(buffer, count, true);
 		mix_pool_bytes(buffer, ret);
 		count -= ret;
 		buffer += ret;
@@ -1290,8 +1270,14 @@ static void mix_interrupt_randomness(str
 	fast_pool->last = jiffies;
 	local_irq_enable();
 
-	mix_pool_bytes(pool, sizeof(pool));
-	credit_entropy_bits(1);
+	if (unlikely(crng_init == 0)) {
+		crng_pre_init_inject(pool, sizeof(pool), true);
+		mix_pool_bytes(pool, sizeof(pool));
+	} else {
+		mix_pool_bytes(pool, sizeof(pool));
+		credit_entropy_bits(1);
+	}
+
 	memzero_explicit(pool, sizeof(pool));
 }
 
@@ -1324,24 +1310,11 @@ void add_interrupt_randomness(int irq)
 	fast_mix(fast_pool->pool32);
 	new_count = ++fast_pool->count;
 
-	if (unlikely(crng_init == 0)) {
-		if (new_count >= 64 &&
-		    crng_pre_init_inject(fast_pool->pool32, sizeof(fast_pool->pool32),
-					 true, true) > 0) {
-			fast_pool->count = 0;
-			fast_pool->last = now;
-			if (spin_trylock(&input_pool.lock)) {
-				_mix_pool_bytes(&fast_pool->pool32, sizeof(fast_pool->pool32));
-				spin_unlock(&input_pool.lock);
-			}
-		}
-		return;
-	}
-
 	if (new_count & MIX_INFLIGHT)
 		return;
 
-	if (new_count < 64 && !time_after(now, fast_pool->last + HZ))
+	if (new_count < 64 && (!time_after(now, fast_pool->last + HZ) ||
+			       unlikely(crng_init == 0)))
 		return;
 
 	if (unlikely(!fast_pool->mix.func))
author maintainer
#116 lore raw

PATCH 116/234 random: give sysctl_random_min_urandom_seed a more sensible value

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit d0efdf35a6a71d307a250199af6fce122a7c7e11 upstream.

This isn't used by anything or anywhere, but we can't delete it due to compatibility. So at least give it the correct value of what it's supposed to be instead of a garbage one.

---
 drivers/char/random.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/char/random.c +2 -2
@@ -1612,7 +1612,7 @@ const struct file_operations urandom_fop
  *   to avoid breaking old userspaces, but writing to it does not
  *   change any behavior of the RNG.
  *
- * - urandom_min_reseed_secs - fixed to the meaningless value "60".
+ * - urandom_min_reseed_secs - fixed to the value CRNG_RESEED_INTERVAL.
  *   It is writable to avoid breaking old userspaces, but writing
  *   to it does not change any behavior of the RNG.
  *
@@ -1622,7 +1622,7 @@ const struct file_operations urandom_fop
 
 #include <linux/sysctl.h>
 
-static int sysctl_random_min_urandom_seed = 60;
+static int sysctl_random_min_urandom_seed = CRNG_RESEED_INTERVAL / HZ;
 static int sysctl_random_write_wakeup_bits = POOL_MIN_BITS;
 static int sysctl_poolsize = POOL_BITS;
 static u8 sysctl_bootid[UUID_SIZE];
author maintainer
#117 lore raw

PATCH 117/234 random: dont let 644 read-only sysctls be written to

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 77553cf8f44863b31da242cf24671d76ddb61597 upstream.

We leave around these old sysctls for compatibility, and we keep them "writable" for compatibility, but even after writing, we should keep reporting the same value. This is consistent with how userspaces tend to use sysctl_random_write_wakeup_bits, writing to it, and then later reading from it and using the value.

---
 drivers/char/random.c |   11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)
drivers/char/random.c +9 -2
@@ -1662,6 +1662,13 @@ static int proc_do_uuid(struct ctl_table
 	return proc_dostring(&fake_table, 0, buffer, lenp, ppos);
 }
 
+/* The same as proc_dointvec, but writes don't change anything. */
+static int proc_do_rointvec(struct ctl_table *table, int write, void __user *buffer,
+			    size_t *lenp, loff_t *ppos)
+{
+	return write ? 0 : proc_dointvec(table, 0, buffer, lenp, ppos);
+}
+
 extern struct ctl_table random_table[];
 struct ctl_table random_table[] = {
 	{
@@ -1683,14 +1690,14 @@ struct ctl_table random_table[] = {
 		.data		= &sysctl_random_write_wakeup_bits,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= proc_do_rointvec,
 	},
 	{
 		.procname	= "urandom_min_reseed_secs",
 		.data		= &sysctl_random_min_urandom_seed,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= proc_do_rointvec,
 	},
 	{
 		.procname	= "boot_id",
author maintainer
#118 lore raw

PATCH 118/234 random: replace custom notifier chain with standard one

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 5acd35487dc911541672b3ffc322851769c32a56 upstream.

We previously rolled our own randomness readiness notifier, which only has two users in the whole kernel. Replace this with a more standard atomic notifier block that serves the same purpose with less code. Also unexport the symbols, because no modules use it, only unconditional builtins. The only drawback is that it's possible for a notification handler returning the "stop" code to prevent further processing, but given that there are only two users, and that we're unexporting this anyway, that doesn't seem like a significant drawback for the simplification we receive here.

Cc Greg Kroah-Hartman <[email protected]>
Cc Theodore Ts'o <[email protected]>
[Jason: for stable, also backported to crypto/drbg.c, not unexporting.]
---
 crypto/drbg.c          |   17 +++++-------
 drivers/char/random.c  |   69 ++++++++++++++-----------------------------------
 include/crypto/drbg.h  |    2 -
 include/linux/random.h |   10 ++-----
 lib/random32.c         |   13 +++++----
 lib/vsprintf.c         |   10 ++++---
 6 files changed, 47 insertions(+), 74 deletions(-)
crypto/drbg.c +8 -9
@@ -1388,12 +1388,13 @@ static int drbg_generate_long(struct drb
 	return 0;
 }
 
-static void drbg_schedule_async_seed(struct random_ready_callback *rdy)
+static int drbg_schedule_async_seed(struct notifier_block *nb, unsigned long action, void *data)
 {
-	struct drbg_state *drbg = container_of(rdy, struct drbg_state,
+	struct drbg_state *drbg = container_of(nb, struct drbg_state,
 					       random_ready);
 
 	schedule_work(&drbg->seed_work);
+	return 0;
 }
 
 static int drbg_prepare_hrng(struct drbg_state *drbg)
@@ -1406,10 +1407,8 @@ static int drbg_prepare_hrng(struct drbg
 
 	INIT_WORK(&drbg->seed_work, drbg_async_seed);
 
-	drbg->random_ready.owner = THIS_MODULE;
-	drbg->random_ready.func = drbg_schedule_async_seed;
-
-	err = add_random_ready_callback(&drbg->random_ready);
+	drbg->random_ready.notifier_call = drbg_schedule_async_seed;
+	err = register_random_ready_notifier(&drbg->random_ready);
 
 	switch (err) {
 	case 0:
@@ -1420,7 +1419,7 @@ static int drbg_prepare_hrng(struct drbg
 		/* fall through */
 
 	default:
-		drbg->random_ready.func = NULL;
+		drbg->random_ready.notifier_call = NULL;
 		return err;
 	}
 
@@ -1526,8 +1525,8 @@ free_everything:
  */
 static int drbg_uninstantiate(struct drbg_state *drbg)
 {
-	if (drbg->random_ready.func) {
-		del_random_ready_callback(&drbg->random_ready);
+	if (drbg->random_ready.notifier_call) {
+		unregister_random_ready_notifier(&drbg->random_ready);
 		cancel_work_sync(&drbg->seed_work);
 		crypto_free_rng(drbg->jent);
 		drbg->jent = NULL;
drivers/char/random.c +21 -48

101 lines not on this page · raw message

include/crypto/drbg.h +1 -1
@@ -134,7 +134,7 @@ struct drbg_state {
 	const struct drbg_state_ops *d_ops;
 	const struct drbg_core *core;
 	struct drbg_string test_data;
-	struct random_ready_callback random_ready;
+	struct notifier_block random_ready;
 };
 
 static inline __u8 drbg_statelen(struct drbg_state *drbg)
include/linux/random.h +3 -7
@@ -10,11 +10,7 @@
 
 #include <uapi/linux/random.h>
 
-struct random_ready_callback {
-	struct list_head list;
-	void (*func)(struct random_ready_callback *rdy);
-	struct module *owner;
-};
+struct notifier_block;
 
 extern void add_device_randomness(const void *, size_t);
 extern void add_bootloader_randomness(const void *, size_t);
@@ -39,8 +35,8 @@ extern void get_random_bytes(void *buf,
 extern int wait_for_random_bytes(void);
 extern int __init rand_initialize(void);
 extern bool rng_is_initialized(void);
-extern int add_random_ready_callback(struct random_ready_callback *rdy);
-extern void del_random_ready_callback(struct random_ready_callback *rdy);
+extern int register_random_ready_notifier(struct notifier_block *nb);
+extern int unregister_random_ready_notifier(struct notifier_block *nb);
 extern size_t __must_check get_random_bytes_arch(void *buf, size_t nbytes);
 
 #ifndef MODULE
lib/random32.c +8 -5
@@ -40,6 +40,7 @@
 #include <linux/sched.h>
 #include <linux/bitops.h>
 #include <linux/slab.h>
+#include <linux/notifier.h>
 #include <asm/unaligned.h>
 
 /**
@@ -546,9 +547,11 @@ static void prandom_reseed(struct timer_
  * To avoid worrying about whether it's safe to delay that interrupt
  * long enough to seed all CPUs, just schedule an immediate timer event.
  */
-static void prandom_timer_start(struct random_ready_callback *unused)
+static int prandom_timer_start(struct notifier_block *nb,
+			       unsigned long action, void *data)
 {
 	mod_timer(&seed_timer, jiffies);
+	return 0;
 }
 
 /*
@@ -557,13 +560,13 @@ static void prandom_timer_start(struct r
  */
 static int __init prandom_init_late(void)
 {
-	static struct random_ready_callback random_ready = {
-		.func = prandom_timer_start
+	static struct notifier_block random_ready = {
+		.notifier_call = prandom_timer_start
 	};
-	int ret = add_random_ready_callback(&random_ready);
+	int ret = register_random_ready_notifier(&random_ready);
 
 	if (ret == -EALREADY) {
-		prandom_timer_start(&random_ready);
+		prandom_timer_start(&random_ready, 0, NULL);
 		ret = 0;
 	}
 	return ret;
lib/vsprintf.c +6 -4
@@ -1700,14 +1700,16 @@ static void enable_ptr_key_workfn(struct
 
 static DECLARE_WORK(enable_ptr_key_work, enable_ptr_key_workfn);
 
-static void fill_random_ptr_key(struct random_ready_callback *unused)
+static int fill_random_ptr_key(struct notifier_block *nb,
+			       unsigned long action, void *data)
 {
 	/* This may be in an interrupt handler. */
 	queue_work(system_unbound_wq, &enable_ptr_key_work);
+	return 0;
 }
 
-static struct random_ready_callback random_ready = {
-	.func = fill_random_ptr_key
+static struct notifier_block random_ready = {
+	.notifier_call = fill_random_ptr_key
 };
 
 static int __init initialize_ptr_random(void)
@@ -1721,7 +1723,7 @@ static int __init initialize_ptr_random(
 		return 0;
 	}
 
-	ret = add_random_ready_callback(&random_ready);
+	ret = register_random_ready_notifier(&random_ready);
 	if (!ret) {
 		return 0;
 	} else if (ret == -EALREADY) {
author maintainer
#119 lore raw

PATCH 119/234 random: use SipHash as interrupt entropy accumulator

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit f5eab0e2db4f881fb2b62b3fdad5b9be673dd7ae upstream.

The current fast_mix() function is a piece of classic mailing list crypto, where it just sort of sprung up by an anonymous author without a lot of real analysis of what precisely it was accomplishing. As an ARX permutation alone, there are some easily searchable differential trails in it, and as a means of preventing malicious interrupts, it completely fails, since it xors new data into the entire state every time. It can't really be analyzed as a random permutation, because it clearly isn't, and it can't be analyzed as an interesting linear algebraic structure either, because it's also not that. There really is very little one can say about it in terms of entropy accumulation. It might diffuse bits, some of the time, maybe, we hope, I guess. But for the most part, it fails to accomplish anything concrete.

As a reminder, the simple goal of add_interrupt_randomness() is to simply accumulate entropy until ~64 interrupts have elapsed, and then dump it into the main input pool, which uses a cryptographic hash.

It would be nice to have something cryptographically strong in the interrupt handler itself, in case a malicious interrupt compromises a per-cpu fast pool within the 64 interrupts / 1 second window, and then inside of that same window somehow can control its return address and cycle counter, even if that's a bit far fetched. However, with a very CPU-limited budget, actually doing that remains an active research project (and perhaps there'll be something useful for Linux to come out of it). And while the abundance of caution would be nice, this isn't *currently* the security model, and we don't yet have a fast enough solution to make it our security model. Plus there's not exactly a pressing need to do that. (And for the avoidance of doubt, the actual cluster of 64 accumulated interrupts still gets dumped into our cryptographically secure input pool.)

So, for now we are going to stick with the existing interrupt security model, which assumes that each cluster of 64 interrupt data samples is mostly non-malicious and not colluding with an infoleaker. With this as our goal, we have a few more choices, simply aiming to accumulate entropy, while discarding the least amount of it.

We know from <https://eprint.iacr.org/2019/198> that random oracles, instantiated as computational hash functions, make good entropy accumulators and extractors, which is the justification for using BLAKE2s in the main input pool. As mentioned, we don't have that luxury here, but we also don't have the same security model requirements, because we're assuming that there aren't malicious inputs. A pseudorandom function instance can approximately behave like a random oracle, provided that the key is uniformly random. But since we're not concerned with malicious inputs, we can pick a fixed key, which is not secret, knowing that "nature" won't interact with a sufficiently chosen fixed key by accident. So we pick a PRF with a fixed initial key, and accumulate into it continuously, dumping the result every 64 interrupts into our cryptographically secure input pool.

For this, we make use of SipHash-1-x on 64-bit and HalfSipHash-1-x on 32-bit, which are already in use in the kernel's hsiphash family of functions and achieve the same performance as the function they replace. It would be nice to do two rounds, but we don't exactly have the CPU budget handy for that, and one round alone is already sufficient.

As mentioned, we start with a fixed initial key (zeros is fine), and allow SipHash's symmetry breaking constants to turn that into a useful starting point. Also, since we're dumping the result (or half of it on 64-bit so as to tax our hash function the same amount on all platforms) into the cryptographically secure input pool, there's no point in finalizing SipHash's output, since it'll wind up being finalized by something much stronger. This means that all we need to do is use the ordinary round function word-by-word, as normal SipHash does. Simplified, the flow is as follows:

Initialize:

    siphash_state_t state;
    siphash_init(&state, key={0, 0, 0, 0});

Update (accumulate) on interrupt:

    siphash_update(&state, interrupt_data_and_timing);

Dump into input pool after 64 interrupts:

    blake2s_update(&input_pool, &state, sizeof(state) / 2);

The result of all of this is that the security model is unchanged from before -- we assume non-malicious inputs -- yet we now implement that model with a stronger argument. I would like to emphasize, again, that the purpose of this commit is to improve the existing design, by making it analyzable, without changing any fundamental assumptions. There may well be value down the road in changing up the existing design, using something cryptographically strong, or simply using a ring buffer of samples rather than having a fast_mix() at all, or changing which and how much data we collect each interrupt so that we can use something linear, or a variety of other ideas. This commit does not invalidate the potential for those in the future.

For example, in the future, if we're able to characterize the data we're collecting on each interrupt, we may be able to inch toward information theoretic accumulators. <https://eprint.iacr.org/2021/523> shows that s = ror32(s, 7) ^ x and s = ror64(s, 19) ^ x make very good accumulators for 2-monotone distributions, which would apply to timestamp counters, like random_get_entropy() or jiffies, but would not apply to our current combination of the two values, or to the various function addresses and register values we mix in. Alternatively,
<https://eprint.iacr.org/2021/1002> shows that max-period linear functions with no non-trivial invariant subspace make good extractors, used in the form s = f(s) ^ x. However, this only works if the input data is both identical and independent, and obviously a collection of address values and counters fails; so it goes with theoretical papers. Future directions here may involve trying to characterize more precisely what we actually need to collect in the interrupt handler, and building something specific around that.

However, as mentioned, the morass of data we're gathering at the interrupt handler presently defies characterization, and so we use SipHash for now, which works well and performs well.

---
 drivers/char/random.c |   94 +++++++++++++++++++++++++++++---------------------
 1 file changed, 55 insertions(+), 39 deletions(-)
drivers/char/random.c +55 -39
@@ -1142,48 +1142,51 @@ void add_bootloader_randomness(const voi
 EXPORT_SYMBOL_GPL(add_bootloader_randomness);
 
 struct fast_pool {
-	union {
-		u32 pool32[4];
-		u64 pool64[2];
-	};
 	struct work_struct mix;
+	unsigned long pool[4];
 	unsigned long last;
 	unsigned int count;
 	u16 reg_idx;
 };
 
+static DEFINE_PER_CPU(struct fast_pool, irq_randomness) = {
+#ifdef CONFIG_64BIT
+	/* SipHash constants */
+	.pool = { 0x736f6d6570736575UL, 0x646f72616e646f6dUL,
+		  0x6c7967656e657261UL, 0x7465646279746573UL }
+#else
+	/* HalfSipHash constants */
+	.pool = { 0, 0, 0x6c796765U, 0x74656462U }
+#endif
+};
+
 /*
- * This is a fast mixing routine used by the interrupt randomness
- * collector. It's hardcoded for an 128 bit pool and assumes that any
- * locks that might be needed are taken by the caller.
+ * This is [Half]SipHash-1-x, starting from an empty key. Because
+ * the key is fixed, it assumes that its inputs are non-malicious,
+ * and therefore this has no security on its own. s represents the
+ * 128 or 256-bit SipHash state, while v represents a 128-bit input.
  */
-static void fast_mix(u32 pool[4])
+static void fast_mix(unsigned long s[4], const unsigned long *v)
 {
-	u32 a = pool[0],	b = pool[1];
-	u32 c = pool[2],	d = pool[3];
-
-	a += b;			c += d;
-	b = rol32(b, 6);	d = rol32(d, 27);
-	d ^= a;			b ^= c;
-
-	a += b;			c += d;
-	b = rol32(b, 16);	d = rol32(d, 14);
-	d ^= a;			b ^= c;
-
-	a += b;			c += d;
-	b = rol32(b, 6);	d = rol32(d, 27);
-	d ^= a;			b ^= c;
-
-	a += b;			c += d;
-	b = rol32(b, 16);	d = rol32(d, 14);
-	d ^= a;			b ^= c;
+	size_t i;
 
-	pool[0] = a;  pool[1] = b;
-	pool[2] = c;  pool[3] = d;
+	for (i = 0; i < 16 / sizeof(long); ++i) {
+		s[3] ^= v[i];
+#ifdef CONFIG_64BIT
+		s[0] += s[1]; s[1] = rol64(s[1], 13); s[1] ^= s[0]; s[0] = rol64(s[0], 32);
+		s[2] += s[3]; s[3] = rol64(s[3], 16); s[3] ^= s[2];
+		s[0] += s[3]; s[3] = rol64(s[3], 21); s[3] ^= s[0];
+		s[2] += s[1]; s[1] = rol64(s[1], 17); s[1] ^= s[2]; s[2] = rol64(s[2], 32);
+#else
+		s[0] += s[1]; s[1] = rol32(s[1],  5); s[1] ^= s[0]; s[0] = rol32(s[0], 16);
+		s[2] += s[3]; s[3] = rol32(s[3],  8); s[3] ^= s[2];
+		s[0] += s[3]; s[3] = rol32(s[3],  7); s[3] ^= s[0];
+		s[2] += s[1]; s[1] = rol32(s[1], 13); s[1] ^= s[2]; s[2] = rol32(s[2], 16);
+#endif
+		s[0] ^= v[i];
+	}
 }
 
-static DEFINE_PER_CPU(struct fast_pool, irq_randomness);
-
 #ifdef CONFIG_SMP
 /*
  * This function is called when the CPU has just come online, with
@@ -1225,7 +1228,15 @@ static unsigned long get_reg(struct fast
 static void mix_interrupt_randomness(struct work_struct *work)
 {
 	struct fast_pool *fast_pool = container_of(work, struct fast_pool, mix);
-	u32 pool[4];
+	/*
+	 * The size of the copied stack pool is explicitly 16 bytes so that we
+	 * tax mix_pool_byte()'s compression function the same amount on all
+	 * platforms. This means on 64-bit we copy half the pool into this,
+	 * while on 32-bit we copy all of it. The entropy is supposed to be
+	 * sufficiently dispersed between bits that in the sponge-like
+	 * half case, on average we don't wind up "losing" some.
+	 */
+	u8 pool[16];
 
 	/* Check to see if we're running on the wrong CPU due to hotplug. */
 	local_irq_disable();
@@ -1238,7 +1249,7 @@ static void mix_interrupt_randomness(str
 	 * Copy the pool to the stack so that the mixer always has a
 	 * consistent view, before we reenable irqs again.
 	 */
-	memcpy(pool, fast_pool->pool32, sizeof(pool));
+	memcpy(pool, fast_pool->pool, sizeof(pool));
 	fast_pool->count = 0;
 	fast_pool->last = jiffies;
 	local_irq_enable();
@@ -1262,25 +1273,30 @@ void add_interrupt_randomness(int irq)
 	struct fast_pool *fast_pool = this_cpu_ptr(&irq_randomness);
 	struct pt_regs *regs = get_irq_regs();
 	unsigned int new_count;
+	union {
+		u32 u32[4];
+		u64 u64[2];
+		unsigned long longs[16 / sizeof(long)];
+	} irq_data;
 
 	if (cycles == 0)
 		cycles = get_reg(fast_pool, regs);
 
 	if (sizeof(cycles) == 8)
-		fast_pool->pool64[0] ^= cycles ^ rol64(now, 32) ^ irq;
+		irq_data.u64[0] = cycles ^ rol64(now, 32) ^ irq;
 	else {
-		fast_pool->pool32[0] ^= cycles ^ irq;
-		fast_pool->pool32[1] ^= now;
+		irq_data.u32[0] = cycles ^ irq;
+		irq_data.u32[1] = now;
 	}
 
 	if (sizeof(unsigned long) == 8)
-		fast_pool->pool64[1] ^= regs ? instruction_pointer(regs) : _RET_IP_;
+		irq_data.u64[1] = regs ? instruction_pointer(regs) : _RET_IP_;
 	else {
-		fast_pool->pool32[2] ^= regs ? instruction_pointer(regs) : _RET_IP_;
-		fast_pool->pool32[3] ^= get_reg(fast_pool, regs);
+		irq_data.u32[2] = regs ? instruction_pointer(regs) : _RET_IP_;
+		irq_data.u32[3] = get_reg(fast_pool, regs);
 	}
 
-	fast_mix(fast_pool->pool32);
+	fast_mix(fast_pool->pool, irq_data.longs);
 	new_count = ++fast_pool->count;
 
 	if (new_count & MIX_INFLIGHT)
author maintainer
#120 lore raw

PATCH 120/234 random: make consistent usage of crng_ready()

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit a96cfe2d427064325ecbf56df8816c6b871ec285 upstream.

Rather than sometimes checking crng_init < 2, we should always use the crng_ready() macro, so that should we change anything later, it's consistent. Additionally, that macro already has a likely() around it, which means we don't need to open code our own likely() and unlikely() annotations.

---
 drivers/char/random.c |   19 +++++++------------
 1 file changed, 7 insertions(+), 12 deletions(-)
drivers/char/random.c +7 -12
@@ -125,18 +125,13 @@ static void try_to_generate_entropy(void
  */
 int wait_for_random_bytes(void)
 {
-	if (likely(crng_ready()))
-		return 0;
-
-	do {
+	while (!crng_ready()) {
 		int ret;
 		ret = wait_event_interruptible_timeout(crng_init_wait, crng_ready(), HZ);
 		if (ret)
 			return ret > 0 ? 0 : ret;
-
 		try_to_generate_entropy();
-	} while (!crng_ready());
-
+	}
 	return 0;
 }
 EXPORT_SYMBOL(wait_for_random_bytes);
@@ -291,7 +286,7 @@ static void crng_reseed(void)
 		++next_gen;
 	WRITE_ONCE(base_crng.generation, next_gen);
 	WRITE_ONCE(base_crng.birth, jiffies);
-	if (crng_init < 2) {
+	if (!crng_ready()) {
 		crng_init = 2;
 		finalize_init = true;
 	}
@@ -359,7 +354,7 @@ static void crng_make_state(u32 chacha_s
 	 * ready, we do fast key erasure with the base_crng directly, because
 	 * this is what crng_pre_init_inject() mutates during early init.
 	 */
-	if (unlikely(!crng_ready())) {
+	if (!crng_ready()) {
 		bool ready;
 
 		spin_lock_irqsave(&base_crng.lock, flags);
@@ -799,7 +794,7 @@ static void credit_entropy_bits(size_t n
 		entropy_count = min_t(unsigned int, POOL_BITS, orig + add);
 	} while (cmpxchg(&input_pool.entropy_count, orig, entropy_count) != orig);
 
-	if (crng_init < 2 && entropy_count >= POOL_MIN_BITS)
+	if (!crng_ready() && entropy_count >= POOL_MIN_BITS)
 		crng_reseed();
 }
 
@@ -956,7 +951,7 @@ int __init rand_initialize(void)
 	extract_entropy(base_crng.key, sizeof(base_crng.key));
 	++base_crng.generation;
 
-	if (arch_init && trust_cpu && crng_init < 2) {
+	if (arch_init && trust_cpu && !crng_ready()) {
 		crng_init = 2;
 		pr_notice("crng init done (trusting CPU's manufacturer)\n");
 	}
@@ -1545,7 +1540,7 @@ static long random_ioctl(struct file *f,
 	case RNDRESEEDCRNG:
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
-		if (crng_init < 2)
+		if (!crng_ready())
 			return -ENODATA;
 		crng_reseed();
 		return 0;
author maintainer
#121 lore raw

PATCH 121/234 random: reseed more often immediately after booting

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 7a7ff644aeaf071d433caffb3b8ea57354b55bd3 upstream.

In order to chip away at the "premature first" problem, we augment our existing entropy accounting with more frequent reseedings at boot.

The idea is that at boot, we're getting entropy from various places, and we're not very sure which of early boot entropy is good and which isn't. Even when we're crediting the entropy, we're still not totally certain that it's any good. Since boot is the one time (aside from a compromise) that we have zero entropy, it's important that we shepherd entropy into the crng fairly often.

At the same time, we don't want a "premature next" problem, whereby an attacker can brute force individual bits of added entropy. In lieu of going full-on Fortuna (for now), we can pick a simpler strategy of just reseeding more often during the first 5 minutes after boot. This is still bounded by the 256-bit entropy credit requirement, so we'll skip a reseeding if we haven't reached that, but in case entropy /is/ coming in, this ensures that it makes its way into the crng rather rapidly during these early stages.

Ordinarily we reseed if the previous reseeding is 300 seconds old. This commit changes things so that for the first 600 seconds of boot time, we reseed if the previous reseeding is uptime / 2 seconds old. That means that we'll reseed at the very least double the uptime of the previous reseeding.

---
 drivers/char/random.c |   28 +++++++++++++++++++++++++---
 1 file changed, 25 insertions(+), 3 deletions(-)
drivers/char/random.c +25 -3
@@ -336,6 +336,28 @@ static void crng_fast_key_erasure(u8 key
 }
 
 /*
+ * Return whether the crng seed is considered to be sufficiently
+ * old that a reseeding might be attempted. This happens if the last
+ * reseeding was CRNG_RESEED_INTERVAL ago, or during early boot, at
+ * an interval proportional to the uptime.
+ */
+static bool crng_has_old_seed(void)
+{
+	static bool early_boot = true;
+	unsigned long interval = CRNG_RESEED_INTERVAL;
+
+	if (unlikely(READ_ONCE(early_boot))) {
+		time64_t uptime = ktime_get_seconds();
+		if (uptime >= CRNG_RESEED_INTERVAL / HZ * 2)
+			WRITE_ONCE(early_boot, false);
+		else
+			interval = max_t(unsigned int, 5 * HZ,
+					 (unsigned int)uptime / 2 * HZ);
+	}
+	return time_after(jiffies, READ_ONCE(base_crng.birth) + interval);
+}
+
+/*
  * This function returns a ChaCha state that you may use for generating
  * random data. It also returns up to 32 bytes on its own of random data
  * that may be used; random_data_len may not be greater than 32.
@@ -368,10 +390,10 @@ static void crng_make_state(u32 chacha_s
 	}
 
 	/*
-	 * If the base_crng is more than 5 minutes old, we reseed, which
-	 * in turn bumps the generation counter that we check below.
+	 * If the base_crng is old enough, we try to reseed, which in turn
+	 * bumps the generation counter that we check below.
 	 */
-	if (unlikely(time_after(jiffies, READ_ONCE(base_crng.birth) + CRNG_RESEED_INTERVAL)))
+	if (unlikely(crng_has_old_seed()))
 		crng_reseed();
 
 	local_irq_save(flags);
author maintainer
#122 lore raw

PATCH 122/234 random: check for signal and try earlier when generating entropy

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 3e504d2026eb6c8762cd6040ae57db166516824a upstream.

Rather than waiting a full second in an interruptable waiter before trying to generate entropy, try to generate entropy first and wait second. While waiting one second might give an extra second for getting entropy from elsewhere, we're already pretty late in the init process here, and whatever else is generating entropy will still continue to contribute. This has implications on signal handling: we call try_to_generate_entropy() from wait_for_random_bytes(), and wait_for_random_bytes() always uses wait_event_interruptible_timeout() when waiting, since it's called by userspace code in restartable contexts, where signals can pend. Since try_to_generate_entropy() now runs first, if a signal is pending, it's necessary for try_to_generate_entropy() to check for signals, since it won't hit the wait until after try_to_generate_entropy() has returned. And even before this change, when entering a busy loop in try_to_generate_entropy(), we should have been checking to see if any signals are pending, so that a process doesn't get stuck in that loop longer than expected.

---
 drivers/char/random.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/char/random.c +3 -2
@@ -127,10 +127,11 @@ int wait_for_random_bytes(void)
 {
 	while (!crng_ready()) {
 		int ret;
+
+		try_to_generate_entropy();
 		ret = wait_event_interruptible_timeout(crng_init_wait, crng_ready(), HZ);
 		if (ret)
 			return ret > 0 ? 0 : ret;
-		try_to_generate_entropy();
 	}
 	return 0;
 }
@@ -1366,7 +1367,7 @@ static void try_to_generate_entropy(void
 		return;
 
 	timer_setup_on_stack(&stack.timer, entropy_timer, 0);
-	while (!crng_ready()) {
+	while (!crng_ready() && !signal_pending(current)) {
 		if (!timer_pending(&stack.timer))
 			mod_timer(&stack.timer, jiffies + 1);
 		mix_pool_bytes(&stack.cycles, sizeof(stack.cycles));
author maintainer
#123 lore raw

PATCH 123/234 random: skip fast_init if hwrng provides large chunk of entropy

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit af704c856e888fb044b058d731d61b46eeec499d upstream.

At boot time, EFI calls add_bootloader_randomness(), which in turn calls add_hwgenerator_randomness(). Currently add_hwgenerator_randomness() feeds the first 64 bytes of randomness to the "fast init" non-crypto-grade phase. But if add_hwgenerator_randomness() gets called with more than POOL_MIN_BITS of entropy, there's no point in passing it off to the "fast init" stage, since that's enough entropy to bootstrap the real RNG. The "fast init" stage is just there to provide _something_ in the case where we don't have enough entropy to properly bootstrap the RNG. But if we do have enough entropy to bootstrap the RNG, the current logic doesn't serve a purpose. So, in the case where we're passed greater than or equal to POOL_MIN_BITS of entropy, this commit makes us skip the "fast init" phase.

---
 drivers/char/random.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/char/random.c +1 -1
@@ -1120,7 +1120,7 @@ void rand_initialize_disk(struct gendisk
 void add_hwgenerator_randomness(const void *buffer, size_t count,
 				size_t entropy)
 {
-	if (unlikely(crng_init == 0)) {
+	if (unlikely(crng_init == 0 && entropy < POOL_MIN_BITS)) {
 		size_t ret = crng_pre_init_inject(buffer, count, true);
 		mix_pool_bytes(buffer, ret);
 		count -= ret;
author maintainer
#124 lore raw

PATCH 124/234 random: treat bootloader trust toggle the same way as cpu trust toggle

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit d97c68d178fbf8aaaf21b69b446f2dfb13909316 upstream.

If CONFIG_RANDOM_TRUST_CPU is set, the RNG initializes using RDRAND. But, the user can disable (or enable) this behavior by setting random.trust_cpu=0/1 on the kernel command line. This allows system builders to do reasonable things while avoiding howls from tinfoil hatters. (Or vice versa.)

CONFIG_RANDOM_TRUST_BOOTLOADER is basically the same thing, but regards the seed passed via EFI or device tree, which might come from RDRAND or a TPM or somewhere else. In order to allow distros to more easily enable this while avoiding those same howls (or vice versa), this commit adds the corresponding random.trust_bootloader=0/1 toggle.

---
 Documentation/admin-guide/kernel-parameters.txt |    6 ++++++
 drivers/char/Kconfig                            |    3 ++-
 drivers/char/random.c                           |    8 +++++++-
 3 files changed, 15 insertions(+), 2 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +6 -0
@@ -3697,6 +3697,12 @@
 			fully seed the kernel's CRNG. Default is controlled
 			by CONFIG_RANDOM_TRUST_CPU.
 
+	random.trust_bootloader={on,off}
+			[KNL] Enable or disable trusting the use of a
+			seed passed by the bootloader (if available) to
+			fully seed the kernel's CRNG. Default is controlled
+			by CONFIG_RANDOM_TRUST_BOOTLOADER.
+
 	ras=option[,option,...]	[KNL] RAS-specific options
 
 		cec_disable	[X86]
drivers/char/Kconfig +2 -1
@@ -576,4 +576,5 @@ config RANDOM_TRUST_BOOTLOADER
 	device randomness. Say Y here to assume the entropy provided by the
 	booloader is trustworthy so it will be added to the kernel's entropy
 	pool. Otherwise, say N here so it will be regarded as device input that
-	only mixes the entropy pool.
\ No newline at end of file
+	only mixes the entropy pool. This can also be configured at boot with
+	"random.trust_bootloader=on/off".
drivers/char/random.c +7 -1
@@ -940,11 +940,17 @@ static bool drain_entropy(void *buf, siz
  **********************************************************************/
 
 static bool trust_cpu __ro_after_init = IS_ENABLED(CONFIG_RANDOM_TRUST_CPU);
+static bool trust_bootloader __ro_after_init = IS_ENABLED(CONFIG_RANDOM_TRUST_BOOTLOADER);
 static int __init parse_trust_cpu(char *arg)
 {
 	return kstrtobool(arg, &trust_cpu);
 }
+static int __init parse_trust_bootloader(char *arg)
+{
+	return kstrtobool(arg, &trust_bootloader);
+}
 early_param("random.trust_cpu", parse_trust_cpu);
+early_param("random.trust_bootloader", parse_trust_bootloader);
 
 /*
  * The first collection of entropy occurs at system boot while interrupts
@@ -1152,7 +1158,7 @@ EXPORT_SYMBOL_GPL(add_hwgenerator_random
  */
 void add_bootloader_randomness(const void *buf, size_t size)
 {
-	if (IS_ENABLED(CONFIG_RANDOM_TRUST_BOOTLOADER))
+	if (trust_bootloader)
 		add_hwgenerator_randomness(buf, size, size * 8);
 	else
 		add_device_randomness(buf, size);
author maintainer
#125 lore raw

PATCH 125/234 random: re-add removed comment about get_random_{u32,u64} reseeding

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit dd7aa36e535797926d8eb311da7151919130139d upstream.

The comment about get_random_{u32,u64}() not invoking reseeding got added in an unrelated commit, that then was recently reverted by 0313bc278dac ("Revert "random: block in /dev/urandom""). So this adds that little comment snippet back, and improves the wording a bit too.

---
 drivers/char/random.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/char/random.c +4 -3
@@ -226,9 +226,10 @@ static void _warn_unseeded_randomness(co
  *
  * These interfaces will return the requested number of random bytes
  * into the given buffer or as a return value. This is equivalent to
- * a read from /dev/urandom. The integer family of functions may be
- * higher performance for one-off random integers, because they do a
- * bit of buffering.
+ * a read from /dev/urandom. The u32, u64, int, and long family of
+ * functions may be higher performance for one-off random integers,
+ * because they do a bit of buffering and do not invoke reseeding
+ * until the buffer is emptied.
  *
  *********************************************************************/
 
author maintainer
#126 lore raw

PATCH 126/234 random: mix build-time latent entropy into pool at init

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 1754abb3e7583c570666fa1e1ee5b317e88c89a0 upstream.

Prior, the "input_pool_data" array needed no real initialization, and so it was easy to mark it with __latent_entropy to populate it during compile-time. In switching to using a hash function, this required us to specifically initialize it to some specific state, which means we dropped the __latent_entropy attribute. An unfortunate side effect was this meant the pool was no longer seeded using compile-time random data. In order to bring this back, we declare an array in rand_initialize() with __latent_entropy and call mix_pool_bytes() on that at init, which accomplishes the same thing as before. We make this __initconst, so that it doesn't take up space at runtime after init.

Fixes 6e8ec2552c7d ("random: use computational hash for entropy extraction")
---
 drivers/char/random.c |    5 +++++
 1 file changed, 5 insertions(+)
drivers/char/random.c +5 -0
@@ -967,6 +967,11 @@ int __init rand_initialize(void)
 	bool arch_init = true;
 	unsigned long rv;
 
+#if defined(LATENT_ENTROPY_PLUGIN)
+	static const u8 compiletime_seed[BLAKE2S_BLOCK_SIZE] __initconst __latent_entropy;
+	_mix_pool_bytes(compiletime_seed, sizeof(compiletime_seed));
+#endif
+
 	for (i = 0; i < BLAKE2S_BLOCK_SIZE; i += sizeof(rv)) {
 		if (!arch_get_random_seed_long_early(&rv) &&
 		    !arch_get_random_long_early(&rv)) {
author maintainer
#127 lore raw

PATCH 127/234 random: do not split fast init input in add_hwgenerator_randomness()

In reply to Greg Kroah-Hartman

From: Jan Varho <[email protected]>

commit 527a9867af29ff89f278d037db704e0ed50fb666 upstream.

add_hwgenerator_randomness() tries to only use the required amount of input for fast init, but credits all the entropy, rather than a fraction of it. Since it's hard to determine how much entropy is left over out of a non-unformly random sample, either give it all to fast init or credit it, but don't attempt to do both. In the process, we can clean up the injection code to no longer need to return a value.

[Jason: expanded commit message]
Fixes 73c7733f122e ("random: do not throw away excess input to crng_fast_load")
Cc [email protected] # 5.17+, requires af704c856e88
---
 drivers/char/random.c |   23 ++++++-----------------
 1 file changed, 6 insertions(+), 17 deletions(-)
drivers/char/random.c +6 -17
@@ -437,11 +437,8 @@ static void crng_make_state(u32 chacha_s
  * This shouldn't be set by functions like add_device_randomness(),
  * where we can't trust the buffer passed to it is guaranteed to be
  * unpredictable (so it might not have any entropy at all).
- *
- * Returns the number of bytes processed from input, which is bounded
- * by CRNG_INIT_CNT_THRESH if account is true.
  */
-static size_t crng_pre_init_inject(const void *input, size_t len, bool account)
+static void crng_pre_init_inject(const void *input, size_t len, bool account)
 {
 	static int crng_init_cnt = 0;
 	struct blake2s_state hash;
@@ -452,18 +449,15 @@ static size_t crng_pre_init_inject(const
 	spin_lock_irqsave(&base_crng.lock, flags);
 	if (crng_init != 0) {
 		spin_unlock_irqrestore(&base_crng.lock, flags);
-		return 0;
+		return;
 	}
 
-	if (account)
-		len = min_t(size_t, len, CRNG_INIT_CNT_THRESH - crng_init_cnt);
-
 	blake2s_update(&hash, base_crng.key, sizeof(base_crng.key));
 	blake2s_update(&hash, input, len);
 	blake2s_final(&hash, base_crng.key);
 
 	if (account) {
-		crng_init_cnt += len;
+		crng_init_cnt += min_t(size_t, len, CRNG_INIT_CNT_THRESH - crng_init_cnt);
 		if (crng_init_cnt >= CRNG_INIT_CNT_THRESH) {
 			++base_crng.generation;
 			crng_init = 1;
@@ -474,8 +468,6 @@ static size_t crng_pre_init_inject(const
 
 	if (crng_init == 1)
 		pr_notice("fast init done\n");
-
-	return len;
 }
 
 static void _get_random_bytes(void *buf, size_t nbytes)
@@ -1133,12 +1125,9 @@ void add_hwgenerator_randomness(const vo
 				size_t entropy)
 {
 	if (unlikely(crng_init == 0 && entropy < POOL_MIN_BITS)) {
-		size_t ret = crng_pre_init_inject(buffer, count, true);
-		mix_pool_bytes(buffer, ret);
-		count -= ret;
-		buffer += ret;
-		if (!count || crng_init == 0)
-			return;
+		crng_pre_init_inject(buffer, count, true);
+		mix_pool_bytes(buffer, count);
+		return;
 	}
 
 	/*
author maintainer
#128 lore raw

PATCH 128/234 random: do not allow user to keep crng key around on stack

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit aba120cc101788544aa3e2c30c8da88513892350 upstream.

The fast key erasure RNG design relies on the key that's used to be used and then discarded. We do this, making judicious use of memzero_explicit(). However, reads to /dev/urandom and calls to getrandom() involve a copy_to_user(), and userspace can use FUSE or userfaultfd, or make a massive call, dynamically remap memory addresses as it goes, and set the process priority to idle, in order to keep a kernel stack alive indefinitely. By probing /proc/sys/kernel/random/entropy_avail to learn when the crng key is refreshed, a malicious userspace could mount this attack every 5 minutes thereafter, breaking the crng's forward secrecy.

In order to fix this, we just overwrite the stack's key with the first 32 bytes of the "free" fast key erasure output. If we're returning <= 32 bytes to the user, then we can still return those bytes directly, so that short reads don't become slower. And for long reads, the difference is hopefully lost in the amortization, so it doesn't change much, with that amortization helping variously for medium reads.

We don't need to do this for get_random_bytes() and the various kernel-space callers, and later, if we ever switch to always batching, this won't be necessary either, so there's no need to change the API of these functions.

Cc Theodore Ts'o <[email protected]>
Fixes c92e040d575a ("random: add backtracking protection to the CRNG")
Fixes 186873c549df ("random: use simpler fast key erasure flow on per-cpu keys")
---
 drivers/char/random.c |   35 +++++++++++++++++++++++------------
 1 file changed, 23 insertions(+), 12 deletions(-)
drivers/char/random.c +23 -12
@@ -532,19 +532,29 @@ static ssize_t get_random_bytes_user(voi
 	if (!nbytes)
 		return 0;
 
-	len = min_t(size_t, 32, nbytes);
-	crng_make_state(chacha_state, output, len);
-
-	if (copy_to_user(buf, output, len))
-		return -EFAULT;
-	nbytes -= len;
-	buf += len;
-	ret += len;
+	/*
+	 * Immediately overwrite the ChaCha key at index 4 with random
+	 * bytes, in case userspace causes copy_to_user() below to sleep
+	 * forever, so that we still retain forward secrecy in that case.
+	 */
+	crng_make_state(chacha_state, (u8 *)&chacha_state[4], CHACHA_KEY_SIZE);
+	/*
+	 * However, if we're doing a read of len <= 32, we don't need to
+	 * use chacha_state after, so we can simply return those bytes to
+	 * the user directly.
+	 */
+	if (nbytes <= CHACHA_KEY_SIZE) {
+		ret = copy_to_user(buf, &chacha_state[4], nbytes) ? -EFAULT : nbytes;
+		goto out_zero_chacha;
+	}
 
-	while (nbytes) {
+	do {
 		if (large_request && need_resched()) {
-			if (signal_pending(current))
+			if (signal_pending(current)) {
+				if (!ret)
+					ret = -ERESTARTSYS;
 				break;
+			}
 			schedule();
 		}
 
@@ -561,10 +571,11 @@ static ssize_t get_random_bytes_user(voi
 		nbytes -= len;
 		buf += len;
 		ret += len;
-	}
+	} while (nbytes);
 
-	memzero_explicit(chacha_state, sizeof(chacha_state));
 	memzero_explicit(output, sizeof(output));
+out_zero_chacha:
+	memzero_explicit(chacha_state, sizeof(chacha_state));
 	return ret;
 }
 
author maintainer
#129 lore raw

PATCH 129/234 random: check for signal_pending() outside of need_resched() check

In reply to Greg Kroah-Hartman

From: Jann Horn <[email protected]>

commit 1448769c9cdb69ad65287f4f7ab58bc5f2f5d7ba upstream.

signal_pending() checks TIF_NOTIFY_SIGNAL and TIF_SIGPENDING, which signal that the task should bail out of the syscall when possible. This is a separate concept from need_resched(), which checks TIF_NEED_RESCHED, signaling that the task should preempt.

In particular, with the current code, the signal_pending() bailout probably won't work reliably.

Change this to look like other functions that read lots of data, such as read_zero().

Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 drivers/char/random.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/char/random.c +2 -2
@@ -549,13 +549,13 @@ static ssize_t get_random_bytes_user(voi
 	}
 
 	do {
-		if (large_request && need_resched()) {
+		if (large_request) {
 			if (signal_pending(current)) {
 				if (!ret)
 					ret = -ERESTARTSYS;
 				break;
 			}
-			schedule();
+			cond_resched();
 		}
 
 		chacha20_block(chacha_state, output);
author maintainer
#130 lore raw

PATCH 130/234 random: check for signals every PAGE_SIZE chunk of /dev/[u]random

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit e3c1c4fd9e6d14059ed93ebfe15e1c57793b1a05 upstream.

In 1448769c9cdb ("random: check for signal_pending() outside of need_resched() check"), Jann pointed out that we previously were only checking the TIF_NOTIFY_SIGNAL and TIF_SIGPENDING flags if the process had TIF_NEED_RESCHED set, which meant in practice, super long reads to /dev/[u]random would delay signal handling by a long time. I tried this using the below program, and indeed I wasn't able to interrupt a /dev/urandom read until after several megabytes had been read. The bug he fixed has always been there, and so code that reads from /dev/urandom without checking the return value of read() has mostly worked for a long time, for most sizes, not just for <= 256.

Maybe it makes sense to keep that code working. The reason it was so small prior, ignoring the fact that it didn't work anyway, was likely because /dev/random used to block, and that could happen for pretty large lengths of time while entropy was gathered. But now, it's just a chacha20 call, which is extremely fast and is just operating on pure data, without having to wait for some external event. In that sense, /dev/[u]random is a lot more like /dev/zero.

Taking a page out of /dev/zero's read_zero() function, it always returns at least one chunk, and then checks for signals after each chunk. Chunk sizes there are of length PAGE_SIZE. Let's just copy the same thing for /dev/[u]random, and check for signals and cond_resched() for every PAGE_SIZE amount of data. This makes the behavior more consistent with expectations, and should mitigate the impact of Jann's fix for the age-old signal check bug.

---- test program ----

  #include <unistd.h>
  #include <signal.h>
  #include <stdio.h>
  #include <sys/random.h>

  static unsigned char x[~0U];

static void handle(int) { }

  int main(int argc, char *argv[])
  {
    pid_t pid = getpid(), child;
    signal(SIGUSR1, handle);
    if (!(child = fork())) {
      for (;;)
        kill(pid, SIGUSR1);
    }
    pause();
    printf("interrupted after reading %zd bytes\n", getrandom(x, sizeof(x), 0));
    kill(child, SIGTERM);
    return 0;
  }
---
 drivers/char/random.c |   41 ++++++++++++++++++++---------------------
 1 file changed, 20 insertions(+), 21 deletions(-)
drivers/char/random.c +20 -21
@@ -523,9 +523,7 @@ EXPORT_SYMBOL(get_random_bytes);
 
 static ssize_t get_random_bytes_user(void __user *buf, size_t nbytes)
 {
-	bool large_request = nbytes > 256;
-	ssize_t ret = 0;
-	size_t len;
+	size_t len, left, ret = 0;
 	u32 chacha_state[CHACHA20_BLOCK_SIZE / sizeof(u32)];
 	u8 output[CHACHA20_BLOCK_SIZE];
 
@@ -537,46 +535,47 @@ static ssize_t get_random_bytes_user(voi
 	 * bytes, in case userspace causes copy_to_user() below to sleep
 	 * forever, so that we still retain forward secrecy in that case.
 	 */
-	crng_make_state(chacha_state, (u8 *)&chacha_state[4], CHACHA_KEY_SIZE);
+	crng_make_state(chacha_state, (u8 *)&chacha_state[4], CHACHA20_KEY_SIZE);
 	/*
 	 * However, if we're doing a read of len <= 32, we don't need to
 	 * use chacha_state after, so we can simply return those bytes to
 	 * the user directly.
 	 */
-	if (nbytes <= CHACHA_KEY_SIZE) {
-		ret = copy_to_user(buf, &chacha_state[4], nbytes) ? -EFAULT : nbytes;
+	if (nbytes <= CHACHA20_KEY_SIZE) {
+		ret = nbytes - copy_to_user(buf, &chacha_state[4], nbytes);
 		goto out_zero_chacha;
 	}
 
-	do {
-		if (large_request) {
-			if (signal_pending(current)) {
-				if (!ret)
-					ret = -ERESTARTSYS;
-				break;
-			}
-			cond_resched();
-		}
-
+	for (;;) {
 		chacha20_block(chacha_state, output);
 		if (unlikely(chacha_state[12] == 0))
 			++chacha_state[13];
 
 		len = min_t(size_t, nbytes, CHACHA20_BLOCK_SIZE);
-		if (copy_to_user(buf, output, len)) {
-			ret = -EFAULT;
+		left = copy_to_user(buf, output, len);
+		if (left) {
+			ret += len - left;
 			break;
 		}
 
-		nbytes -= len;
 		buf += len;
 		ret += len;
-	} while (nbytes);
+		nbytes -= len;
+		if (!nbytes)
+			break;
+
+		BUILD_BUG_ON(PAGE_SIZE % CHACHA20_BLOCK_SIZE != 0);
+		if (ret % PAGE_SIZE == 0) {
+			if (signal_pending(current))
+				break;
+			cond_resched();
+		}
+	}
 
 	memzero_explicit(output, sizeof(output));
 out_zero_chacha:
 	memzero_explicit(chacha_state, sizeof(chacha_state));
-	return ret;
+	return ret ? ret : -EFAULT;
 }
 
 /*
author maintainer
#131 lore raw

PATCH 131/234 random: make random_get_entropy() return an unsigned long

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit b0c3e796f24b588b862b61ce235d3c9417dc8983 upstream.

Some implementations were returning type unsigned long, while others that fell back to get_cycles() were implicitly returning a cycles_t or an untyped constant int literal. That makes for weird and confusing code, and basically all code in the kernel already handled it like it was an unsigned long. I recently tried to handle it as the largest type it could be, a cycles_t, but doing so doesn't really help with much.

Instead let's just make random_get_entropy() return an unsigned long all the time. This also matches the commonly used arch_get_random_long() function, so now RDRAND and RDTSC return the same sized integer, which means one can fallback to the other more gracefully.

---
 drivers/char/random.c |   20 +++++++-------------
 include/linux/timex.h |    2 +-
 2 files changed, 8 insertions(+), 14 deletions(-)
drivers/char/random.c +7 -13
@@ -1010,7 +1010,7 @@ int __init rand_initialize(void)
  */
 void add_device_randomness(const void *buf, size_t size)
 {
-	cycles_t cycles = random_get_entropy();
+	unsigned long cycles = random_get_entropy();
 	unsigned long flags, now = jiffies;
 
 	if (crng_init == 0 && size)
@@ -1041,8 +1041,7 @@ struct timer_rand_state {
  */
 static void add_timer_randomness(struct timer_rand_state *state, unsigned int num)
 {
-	cycles_t cycles = random_get_entropy();
-	unsigned long flags, now = jiffies;
+	unsigned long cycles = random_get_entropy(), now = jiffies, flags;
 	long delta, delta2, delta3;
 
 	spin_lock_irqsave(&input_pool.lock, flags);
@@ -1297,8 +1296,7 @@ static void mix_interrupt_randomness(str
 void add_interrupt_randomness(int irq)
 {
 	enum { MIX_INFLIGHT = 1U << 31 };
-	cycles_t cycles = random_get_entropy();
-	unsigned long now = jiffies;
+	unsigned long cycles = random_get_entropy(), now = jiffies;
 	struct fast_pool *fast_pool = this_cpu_ptr(&irq_randomness);
 	struct pt_regs *regs = get_irq_regs();
 	unsigned int new_count;
@@ -1311,16 +1309,12 @@ void add_interrupt_randomness(int irq)
 	if (cycles == 0)
 		cycles = get_reg(fast_pool, regs);
 
-	if (sizeof(cycles) == 8)
+	if (sizeof(unsigned long) == 8) {
 		irq_data.u64[0] = cycles ^ rol64(now, 32) ^ irq;
-	else {
+		irq_data.u64[1] = regs ? instruction_pointer(regs) : _RET_IP_;
+	} else {
 		irq_data.u32[0] = cycles ^ irq;
 		irq_data.u32[1] = now;
-	}
-
-	if (sizeof(unsigned long) == 8)
-		irq_data.u64[1] = regs ? instruction_pointer(regs) : _RET_IP_;
-	else {
 		irq_data.u32[2] = regs ? instruction_pointer(regs) : _RET_IP_;
 		irq_data.u32[3] = get_reg(fast_pool, regs);
 	}
@@ -1367,7 +1361,7 @@ static void entropy_timer(struct timer_l
 static void try_to_generate_entropy(void)
 {
 	struct {
-		cycles_t cycles;
+		unsigned long cycles;
 		struct timer_list timer;
 	} stack;
 
include/linux/timex.h +1 -1
@@ -75,7 +75,7 @@
  * By default we use get_cycles() for this purpose, but individual
  * architectures may override this in their asm/timex.h header file.
  */
-#define random_get_entropy()	get_cycles()
+#define random_get_entropy()	((unsigned long)get_cycles())
 #endif
 
 /*
author maintainer
#132 lore raw

PATCH 132/234 random: document crng_fast_key_erasure() destination possibility

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 8717627d6ac53251ee012c3c7aca392f29f38a42 upstream.

This reverts 35a33ff3807d ("random: use memmove instead of memcpy for remaining 32 bytes"), which was made on a totally bogus basis. The thing it was worried about overlapping came from the stack, not from one of its arguments, as Eric pointed out.

But the fact that this confusion even happened draws attention to the fact that it's a bit non-obvious that the random_data parameter can alias chacha_state, and in fact should do so when the caller can't rely on the stack being cleared in a timely manner. So this commit documents that.

---
 drivers/char/random.c |    7 +++++++
 1 file changed, 7 insertions(+)
drivers/char/random.c +7 -0
@@ -318,6 +318,13 @@ static void crng_reseed(void)
  * the resultant ChaCha state to the user, along with the second
  * half of the block containing 32 bytes of random data that may
  * be used; random_data_len may not be greater than 32.
+ *
+ * The returned ChaCha state contains within it a copy of the old
+ * key value, at index 4, so the state should always be zeroed out
+ * immediately after using in order to maintain forward secrecy.
+ * If the state cannot be erased in a timely manner, then it is
+ * safer to set the random_data parameter to &chacha_state[4] so
+ * that this function overwrites it before returning.
  */
 static void crng_fast_key_erasure(u8 key[CHACHA20_KEY_SIZE],
 				  u32 chacha_state[CHACHA20_BLOCK_SIZE / sizeof(u32)],
author maintainer
#133 lore raw

PATCH 133/234 random: fix sysctl documentation nits

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 069c4ea6871c18bd368f27756e0f91ffb524a788 upstream.

A semicolon was missing, and the almost-alphabetical-but-not ordering was confusing, so regroup these by category instead.

---
 Documentation/sysctl/kernel.txt |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
Documentation/sysctl/kernel.txt +4 -4
@@ -835,6 +835,9 @@ This is a directory, with the following
 * ``boot_id``: a UUID generated the first time this is retrieved, and
   unvarying after that;
 
+* ``uuid``: a UUID generated every time this is retrieved (this can
+  thus be used to generate UUIDs at will);
+
 * ``entropy_avail``: the pool's entropy count, in bits;
 
 * ``poolsize``: the entropy pool size, in bits;
@@ -842,10 +845,7 @@ This is a directory, with the following
 * ``urandom_min_reseed_secs``: obsolete (used to determine the minimum
   number of seconds between urandom pool reseeding). This file is
   writable for compatibility purposes, but writing to it has no effect
-  on any RNG behavior.
-
-* ``uuid``: a UUID generated every time this is retrieved (this can
-  thus be used to generate UUIDs at will);
+  on any RNG behavior;
 
 * ``write_wakeup_threshold``: when the entropy count drops below this
   (as a number of bits), processes waiting to write to ``/dev/random``
author maintainer
#134 lore raw

PATCH 134/234 init: call time_init() before rand_initialize()

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit fe222a6ca2d53c38433cba5d3be62a39099e708e upstream.

Currently time_init() is called after rand_initialize(), but rand_initialize() makes use of the timer on various platforms, and sometimes this timer needs to be initialized by time_init() first. In order for random_get_entropy() to not return zero during early boot when it's potentially used as an entropy source, reverse the order of these two calls. The block doing random initialization was right before time_init() before, so changing the order shouldn't have any complicated effects.

---
 init/main.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
init/main.c +2 -1
@@ -633,11 +633,13 @@ asmlinkage __visible void __init start_k
 	hrtimers_init();
 	softirq_init();
 	timekeeping_init();
+	time_init();
 
 	/*
 	 * For best initial stack canary entropy, prepare it after:
 	 * - setup_arch() for any UEFI RNG entropy and boot cmdline access
 	 * - timekeeping_init() for ktime entropy used in rand_initialize()
+	 * - time_init() for making random_get_entropy() work on some platforms
 	 * - rand_initialize() to get any arch-specific entropy like RDRAND
 	 * - add_latent_entropy() to get any latent entropy
 	 * - adding command line entropy
@@ -647,7 +649,6 @@ asmlinkage __visible void __init start_k
 	add_device_randomness(command_line, strlen(command_line));
 	boot_init_stack_canary();
 
-	time_init();
 	perf_event_init();
 	profile_init();
 	call_function_init();
author maintainer
#135 lore raw

PATCH 135/234 ia64: define get_cycles macro for arch-override

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 57c0900b91d8891ab43f0e6b464d059fda51d102 upstream.

Itanium defines a get_cycles() function, but it does not do the usual #define get_cycles get_cycles dance, making it impossible for generic code to see if an arch-specific function was defined. While the get_cycles() ifdef is not currently used, the following timekeeping patch in this series will depend on the macro existing (or not existing) when defining random_get_entropy().

---
 arch/ia64/include/asm/timex.h |    1 +
 1 file changed, 1 insertion(+)
arch/ia64/include/asm/timex.h +1 -0
@@ -39,6 +39,7 @@ get_cycles (void)
 	ret = ia64_getreg(_IA64_REG_AR_ITC);
 	return ret;
 }
+#define get_cycles get_cycles
 
 extern void ia64_cpu_local_tick (void);
 extern unsigned long long ia64_native_sched_clock (void);
author maintainer
#136 lore raw

PATCH 136/234 s390: define get_cycles macro for arch-override

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 2e3df523256cb9836de8441e9c791a796759bb3c upstream.

S390x defines a get_cycles() function, but it does not do the usual #define get_cycles get_cycles dance, making it impossible for generic code to see if an arch-specific function was defined. While the get_cycles() ifdef is not currently used, the following timekeeping patch in this series will depend on the macro existing (or not existing) when defining random_get_entropy().

Cc Thomas Gleixner <[email protected]>
Cc Arnd Bergmann <[email protected]>
Cc Vasily Gorbik <[email protected]>
Cc Alexander Gordeev <[email protected]>
Cc Christian Borntraeger <[email protected]>
Cc Sven Schnelle <[email protected]>
---
 arch/s390/include/asm/timex.h |    1 +
 1 file changed, 1 insertion(+)
arch/s390/include/asm/timex.h +1 -0
@@ -177,6 +177,7 @@ static inline cycles_t get_cycles(void)
 {
 	return (cycles_t) get_tod_clock() >> 2;
 }
+#define get_cycles get_cycles
 
 int get_phys_clock(unsigned long *clock);
 void init_cpu_timer(void);
author maintainer
#137 lore raw

PATCH 137/234 parisc: define get_cycles macro for arch-override

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 8865bbe6ba1120e67f72201b7003a16202cd42be upstream.

PA-RISC defines a get_cycles() function, but it does not do the usual #define get_cycles get_cycles dance, making it impossible for generic code to see if an arch-specific function was defined. While the get_cycles() ifdef is not currently used, the following timekeeping patch in this series will depend on the macro existing (or not existing) when defining random_get_entropy().

---
 arch/parisc/include/asm/timex.h |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
arch/parisc/include/asm/timex.h +2 -1
@@ -12,9 +12,10 @@
 
 typedef unsigned long cycles_t;
 
-static inline cycles_t get_cycles (void)
+static inline cycles_t get_cycles(void)
 {
 	return mfctl(16);
 }
+#define get_cycles get_cycles
 
 #endif
author maintainer
#138 lore raw

PATCH 138/234 alpha: define get_cycles macro for arch-override

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 1097710bc9660e1e588cf2186a35db3d95c4d258 upstream.

Alpha defines a get_cycles() function, but it does not do the usual #define get_cycles get_cycles dance, making it impossible for generic code to see if an arch-specific function was defined. While the get_cycles() ifdef is not currently used, the following timekeeping patch in this series will depend on the macro existing (or not existing) when defining random_get_entropy().

---
 arch/alpha/include/asm/timex.h |    1 +
 1 file changed, 1 insertion(+)
arch/alpha/include/asm/timex.h +1 -0
@@ -28,5 +28,6 @@ static inline cycles_t get_cycles (void)
 	__asm__ __volatile__ ("rpcc %0" : "=r"(ret));
 	return ret;
 }
+#define get_cycles get_cycles
 
 #endif
author maintainer
#139 lore raw

PATCH 139/234 powerpc: define get_cycles macro for arch-override

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 408835832158df0357e18e96da7f2d1ed6b80e7f upstream.

PowerPC defines a get_cycles() function, but it does not do the usual #define get_cycles get_cycles dance, making it impossible for generic code to see if an arch-specific function was defined. While the get_cycles() ifdef is not currently used, the following timekeeping patch in this series will depend on the macro existing (or not existing) when defining random_get_entropy().

Cc Thomas Gleixner <[email protected]>
Cc Arnd Bergmann <[email protected]>
Cc Benjamin Herrenschmidt <[email protected]>
Cc Paul Mackerras <[email protected]>
---
 arch/powerpc/include/asm/timex.h |    1 +
 1 file changed, 1 insertion(+)
arch/powerpc/include/asm/timex.h +1 -0
@@ -50,6 +50,7 @@ static inline cycles_t get_cycles(void)
 	return ret;
 #endif
 }
+#define get_cycles get_cycles
 
 #endif	/* __KERNEL__ */
 #endif	/* _ASM_POWERPC_TIMEX_H */
author maintainer
#140 lore raw

PATCH 140/234 timekeeping: Add raw clock fallback for random_get_entropy()

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 1366992e16bddd5e2d9a561687f367f9f802e2e4 upstream.

The addition of random_get_entropy_fallback() provides access to whichever time source has the highest frequency, which is useful for gathering entropy on platforms without available cycle counters. It's not necessarily as good as being able to quickly access a cycle counter that the CPU has, but it's still something, even when it falls back to being jiffies-based.

In the event that a given arch does not define get_cycles(), falling back to the get_cycles() default implementation that returns 0 is really not the best we can do. Instead, at least calling random_get_entropy_fallback() would be preferable, because that always needs to return _something_, even falling back to jiffies eventually. It's not as though random_get_entropy_fallback() is super high precision or guaranteed to be entropic, but basically anything that's not zero all the time is better than returning zero all the time.

Finally, since random_get_entropy_fallback() is used during extremely early boot when randomizing freelists in mm_init(), it can be called before timekeeping has been initialized. In that case there really is nothing we can do; jiffies hasn't even started ticking yet. So just give up and return 0.

---
 include/linux/timex.h     |    8 ++++++++
 kernel/time/timekeeping.c |   15 +++++++++++++++
 2 files changed, 23 insertions(+)
include/linux/timex.h +8 -0
@@ -62,6 +62,8 @@
 #include <linux/types.h>
 #include <linux/param.h>
 
+unsigned long random_get_entropy_fallback(void);
+
 #include <asm/timex.h>
 
 #ifndef random_get_entropy
@@ -74,8 +76,14 @@
  *
  * By default we use get_cycles() for this purpose, but individual
  * architectures may override this in their asm/timex.h header file.
+ * If a given arch does not have get_cycles(), then we fallback to
+ * using random_get_entropy_fallback().
  */
+#ifdef get_cycles
 #define random_get_entropy()	((unsigned long)get_cycles())
+#else
+#define random_get_entropy()	random_get_entropy_fallback()
+#endif
 #endif
 
 /*
kernel/time/timekeeping.c +15 -0
@@ -22,6 +22,7 @@
 #include <linux/clocksource.h>
 #include <linux/jiffies.h>
 #include <linux/time.h>
+#include <linux/timex.h>
 #include <linux/tick.h>
 #include <linux/stop_machine.h>
 #include <linux/pvclock_gtod.h>
@@ -2308,6 +2309,20 @@ static int timekeeping_validate_timex(co
 	return 0;
 }
 
+/**
+ * random_get_entropy_fallback - Returns the raw clock source value,
+ * used by random.c for platforms with no valid random_get_entropy().
+ */
+unsigned long random_get_entropy_fallback(void)
+{
+	struct tk_read_base *tkr = &tk_core.timekeeper.tkr_mono;
+	struct clocksource *clock = READ_ONCE(tkr->clock);
+
+	if (unlikely(timekeeping_suspended || !clock))
+		return 0;
+	return clock->read(clock);
+}
+EXPORT_SYMBOL_GPL(random_get_entropy_fallback);
 
 /**
  * do_adjtimex() - Accessor function to NTP __do_adjtimex function
author maintainer
#141 lore raw

PATCH 141/234 m68k: use fallback for random_get_entropy() instead of zero

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 0f392c95391f2d708b12971a07edaa7973f9eece upstream.

In the event that random_get_entropy() can't access a cycle counter or similar, falling back to returning 0 is really not the best we can do. Instead, at least calling random_get_entropy_fallback() would be preferable, because that always needs to return _something_, even falling back to jiffies eventually. It's not as though random_get_entropy_fallback() is super high precision or guaranteed to be entropic, but basically anything that's not zero all the time is better than returning zero all the time.

---
 arch/m68k/include/asm/timex.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/m68k/include/asm/timex.h +1 -1
@@ -35,7 +35,7 @@ static inline unsigned long random_get_e
 {
 	if (mach_random_get_entropy)
 		return mach_random_get_entropy();
-	return 0;
+	return random_get_entropy_fallback();
 }
 #define random_get_entropy	random_get_entropy
 
author maintainer
#142 lore raw

PATCH 142/234 mips: use fallback for random_get_entropy() instead of just c0 random

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 1c99c6a7c3c599a68321b01b9ec243215ede5a68 upstream.

For situations in which we don't have a c0 counter register available, we've been falling back to reading the c0 "random" register, which is usually bounded by the amount of TLB entries and changes every other cycle or so. This means it wraps extremely often. We can do better by combining this fast-changing counter with a potentially slower-changing counter from random_get_entropy_fallback() in the more significant bits. This commit combines the two, taking into account that the changing bits are in a different bit position depending on the CPU model. In addition, we previously were falling back to 0 for ancient CPUs that Linux does not support anyway; remove that dead path entirely.

---
 arch/mips/include/asm/timex.h |   17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)
arch/mips/include/asm/timex.h +8 -9
@@ -76,25 +76,24 @@ static inline cycles_t get_cycles(void)
 	else
 		return 0;	/* no usable counter */
 }
+#define get_cycles get_cycles
 
 /*
  * Like get_cycles - but where c0_count is not available we desperately
  * use c0_random in an attempt to get at least a little bit of entropy.
- *
- * R6000 and R6000A neither have a count register nor a random register.
- * That leaves no entropy source in the CPU itself.
  */
 static inline unsigned long random_get_entropy(void)
 {
-	unsigned int prid = read_c0_prid();
-	unsigned int imp = prid & PRID_IMP_MASK;
+	unsigned int c0_random;
 
-	if (can_use_mips_counter(prid))
+	if (can_use_mips_counter(read_c0_prid()))
 		return read_c0_count();
-	else if (likely(imp != PRID_IMP_R6000 && imp != PRID_IMP_R6000A))
-		return read_c0_random();
+
+	if (cpu_has_3kex)
+		c0_random = (read_c0_random() >> 8) & 0x3f;
 	else
-		return 0;	/* no usable register */
+		c0_random = read_c0_random() & 0x3f;
+	return (random_get_entropy_fallback() << 6) | (0x3f - c0_random);
 }
 #define random_get_entropy random_get_entropy
 
author maintainer
#143 lore raw

PATCH 143/234 arm: use fallback for random_get_entropy() instead of zero

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit ff8a8f59c99f6a7c656387addc4d9f2247d75077 upstream.

In the event that random_get_entropy() can't access a cycle counter or similar, falling back to returning 0 is really not the best we can do. Instead, at least calling random_get_entropy_fallback() would be preferable, because that always needs to return _something_, even falling back to jiffies eventually. It's not as though random_get_entropy_fallback() is super high precision or guaranteed to be entropic, but basically anything that's not zero all the time is better than returning zero all the time.

---
 arch/arm/include/asm/timex.h |    1 +
 1 file changed, 1 insertion(+)
arch/arm/include/asm/timex.h +1 -0
@@ -14,5 +14,6 @@
 
 typedef unsigned long cycles_t;
 #define get_cycles()	({ cycles_t c; read_current_timer(&c) ? 0 : c; })
+#define random_get_entropy() (((unsigned long)get_cycles()) ?: random_get_entropy_fallback())
 
 #endif
author maintainer
#144 lore raw

PATCH 144/234 nios2: use fallback for random_get_entropy() instead of zero

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit c04e72700f2293013dab40208e809369378f224c upstream.

In the event that random_get_entropy() can't access a cycle counter or similar, falling back to returning 0 is really not the best we can do. Instead, at least calling random_get_entropy_fallback() would be preferable, because that always needs to return _something_, even falling back to jiffies eventually. It's not as though random_get_entropy_fallback() is super high precision or guaranteed to be entropic, but basically anything that's not zero all the time is better than returning zero all the time.

---
 arch/nios2/include/asm/timex.h |    3 +++
 1 file changed, 3 insertions(+)
arch/nios2/include/asm/timex.h +3 -0
@@ -20,5 +20,8 @@
 typedef unsigned long cycles_t;
 
 extern cycles_t get_cycles(void);
+#define get_cycles get_cycles
+
+#define random_get_entropy() (((unsigned long)get_cycles()) ?: random_get_entropy_fallback())
 
 #endif
author maintainer
#145 lore raw

PATCH 145/234 x86/tsc: Use fallback for random_get_entropy() instead of zero

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 3bd4abc07a267e6a8b33d7f8717136e18f921c53 upstream.

In the event that random_get_entropy() can't access a cycle counter or similar, falling back to returning 0 is suboptimal. Instead, fallback to calling random_get_entropy_fallback(), which isn't extremely high precision or guaranteed to be entropic, but is certainly better than returning zero all the time.

If CONFIG_X86_TSC=n, then it's possible for the kernel to run on systems without RDTSC, such as 486 and certain 586, so the fallback code is only required for that case.

As well, fix up both the new function and the get_cycles() function from which it was derived to use cpu_feature_enabled() rather than boot_cpu_has(), and use !IS_ENABLED() instead of #ifndef.

---
 arch/x86/include/asm/timex.h |    9 +++++++++
 arch/x86/include/asm/tsc.h   |    7 +++----
 2 files changed, 12 insertions(+), 4 deletions(-)
arch/x86/include/asm/timex.h +9 -0
@@ -5,6 +5,15 @@
 #include <asm/processor.h>
 #include <asm/tsc.h>
 
+static inline unsigned long random_get_entropy(void)
+{
+	if (!IS_ENABLED(CONFIG_X86_TSC) &&
+	    !cpu_feature_enabled(X86_FEATURE_TSC))
+		return random_get_entropy_fallback();
+	return rdtsc();
+}
+#define random_get_entropy random_get_entropy
+
 /* Assume we use the PIT time source for the clock tick */
 #define CLOCK_TICK_RATE		PIT_TICK_RATE
 
arch/x86/include/asm/tsc.h +3 -4
@@ -22,13 +22,12 @@ extern void disable_TSC(void);
 
 static inline cycles_t get_cycles(void)
 {
-#ifndef CONFIG_X86_TSC
-	if (!boot_cpu_has(X86_FEATURE_TSC))
+	if (!IS_ENABLED(CONFIG_X86_TSC) &&
+	    !cpu_feature_enabled(X86_FEATURE_TSC))
 		return 0;
-#endif
-
 	return rdtsc();
 }
+#define get_cycles get_cycles
 
 extern struct system_counterval_t convert_art_to_tsc(u64 art);
 extern struct system_counterval_t convert_art_ns_to_tsc(u64 art_ns);
author maintainer
#146 lore raw

PATCH 146/234 um: use fallback for random_get_entropy() instead of zero

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 9f13fb0cd11ed2327abff69f6501a2c124c88b5a upstream.

In the event that random_get_entropy() can't access a cycle counter or similar, falling back to returning 0 is really not the best we can do. Instead, at least calling random_get_entropy_fallback() would be preferable, because that always needs to return _something_, even falling back to jiffies eventually. It's not as though random_get_entropy_fallback() is super high precision or guaranteed to be entropic, but basically anything that's not zero all the time is better than returning zero all the time.

This is accomplished by just including the asm-generic code like on other architectures, which means we can get rid of the empty stub function here.

---
 arch/um/include/asm/timex.h |    9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)
arch/um/include/asm/timex.h +2 -7
@@ -2,13 +2,8 @@
 #ifndef __UM_TIMEX_H
 #define __UM_TIMEX_H
 
-typedef unsigned long cycles_t;
-
-static inline cycles_t get_cycles (void)
-{
-	return 0;
-}
-
 #define CLOCK_TICK_RATE (HZ)
 
+#include <asm-generic/timex.h>
+
 #endif
author maintainer
#147 lore raw

PATCH 147/234 sparc: use fallback for random_get_entropy() instead of zero

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit ac9756c79797bb98972736b13cfb239fd2cffb79 upstream.

In the event that random_get_entropy() can't access a cycle counter or similar, falling back to returning 0 is really not the best we can do. Instead, at least calling random_get_entropy_fallback() would be preferable, because that always needs to return _something_, even falling back to jiffies eventually. It's not as though random_get_entropy_fallback() is super high precision or guaranteed to be entropic, but basically anything that's not zero all the time is better than returning zero all the time.

This is accomplished by just including the asm-generic code like on other architectures, which means we can get rid of the empty stub function here.

Cc Thomas Gleixner <[email protected]>
Cc Arnd Bergmann <[email protected]>
Cc David S. Miller <[email protected]>
---
 arch/sparc/include/asm/timex_32.h |    4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
arch/sparc/include/asm/timex_32.h +1 -3
@@ -9,8 +9,6 @@
 
 #define CLOCK_TICK_RATE	1193180 /* Underlying HZ */
 
-/* XXX Maybe do something better at some point... -DaveM */
-typedef unsigned long cycles_t;
-#define get_cycles()	(0)
+#include <asm-generic/timex.h>
 
 #endif
author maintainer
#148 lore raw

PATCH 148/234 xtensa: use fallback for random_get_entropy() instead of zero

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit e10e2f58030c5c211d49042a8c2a1b93d40b2ffb upstream.

In the event that random_get_entropy() can't access a cycle counter or similar, falling back to returning 0 is really not the best we can do. Instead, at least calling random_get_entropy_fallback() would be preferable, because that always needs to return _something_, even falling back to jiffies eventually. It's not as though random_get_entropy_fallback() is super high precision or guaranteed to be entropic, but basically anything that's not zero all the time is better than returning zero all the time.

This is accomplished by just including the asm-generic code like on other architectures, which means we can get rid of the empty stub function here.

---
 arch/xtensa/include/asm/timex.h |    6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
arch/xtensa/include/asm/timex.h +2 -4
@@ -30,10 +30,6 @@
 
 extern unsigned long ccount_freq;
 
-typedef unsigned long long cycles_t;
-
-#define get_cycles()	(0)
-
 void local_timer_setup(unsigned cpu);
 
 /*
@@ -69,4 +65,6 @@ static inline void set_linux_timer (unsi
 	WSR_CCOMPARE(LINUX_TIMER, ccompare);
 }
 
+#include <asm-generic/timex.h>
+
 #endif	/* _XTENSA_TIMEX_H */
author maintainer
#149 lore raw

PATCH 149/234 random: insist on random_get_entropy() existing in order to simplify

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 4b758eda851eb9336ca86a0041a4d3da55f66511 upstream.

All platforms are now guaranteed to provide some value for random_get_entropy(). In case some bug leads to this not being so, we print a warning, because that indicates that something is really very wrong (and likely other things are impacted too). This should never be hit, but it's a good and cheap way of finding out if something ever is problematic.

Since we now have viable fallback code for random_get_entropy() on all platforms, which is, in the worst case, not worse than jiffies, we can count on getting the best possible value out of it. That means there's no longer a use for using jiffies as entropy input. It also means we no longer have a reason for doing the round-robin register flow in the IRQ handler, which was always of fairly dubious value.

Instead we can greatly simplify the IRQ handler inputs and also unify the construction between 64-bits and 32-bits. We now collect the cycle counter and the return address, since those are the two things that matter. Because the return address and the irq number are likely related, to the extent we mix in the irq number, we can just xor it into the top unchanging bytes of the return address, rather than the bottom changing bytes of the cycle counter as before. Then, we can do a fixed 2 rounds of SipHash/HSipHash. Finally, we use the same construction of hashing only half of the [H]SipHash state on 32-bit and 64-bit. We're not actually discarding any entropy, since that entropy is carried through until the next time. And more importantly, it lets us do the same sponge-like construction everywhere.

---
 drivers/char/random.c |   86 +++++++++++++++-----------------------------------
 1 file changed, 26 insertions(+), 60 deletions(-)
drivers/char/random.c +26 -60
@@ -1017,15 +1017,14 @@ int __init rand_initialize(void)
  */
 void add_device_randomness(const void *buf, size_t size)
 {
-	unsigned long cycles = random_get_entropy();
-	unsigned long flags, now = jiffies;
+	unsigned long entropy = random_get_entropy();
+	unsigned long flags;
 
 	if (crng_init == 0 && size)
 		crng_pre_init_inject(buf, size, false);
 
 	spin_lock_irqsave(&input_pool.lock, flags);
-	_mix_pool_bytes(&cycles, sizeof(cycles));
-	_mix_pool_bytes(&now, sizeof(now));
+	_mix_pool_bytes(&entropy, sizeof(entropy));
 	_mix_pool_bytes(buf, size);
 	spin_unlock_irqrestore(&input_pool.lock, flags);
 }
@@ -1048,12 +1047,11 @@ struct timer_rand_state {
  */
 static void add_timer_randomness(struct timer_rand_state *state, unsigned int num)
 {
-	unsigned long cycles = random_get_entropy(), now = jiffies, flags;
+	unsigned long entropy = random_get_entropy(), now = jiffies, flags;
 	long delta, delta2, delta3;
 
 	spin_lock_irqsave(&input_pool.lock, flags);
-	_mix_pool_bytes(&cycles, sizeof(cycles));
-	_mix_pool_bytes(&now, sizeof(now));
+	_mix_pool_bytes(&entropy, sizeof(entropy));
 	_mix_pool_bytes(&num, sizeof(num));
 	spin_unlock_irqrestore(&input_pool.lock, flags);
 
@@ -1181,7 +1179,6 @@ struct fast_pool {
 	unsigned long pool[4];
 	unsigned long last;
 	unsigned int count;
-	u16 reg_idx;
 };
 
 static DEFINE_PER_CPU(struct fast_pool, irq_randomness) = {
@@ -1199,13 +1196,13 @@ static DEFINE_PER_CPU(struct fast_pool,
  * This is [Half]SipHash-1-x, starting from an empty key. Because
  * the key is fixed, it assumes that its inputs are non-malicious,
  * and therefore this has no security on its own. s represents the
- * 128 or 256-bit SipHash state, while v represents a 128-bit input.
+ * four-word SipHash state, while v represents a two-word input.
  */
-static void fast_mix(unsigned long s[4], const unsigned long *v)
+static void fast_mix(unsigned long s[4], const unsigned long v[2])
 {
 	size_t i;
 
-	for (i = 0; i < 16 / sizeof(long); ++i) {
+	for (i = 0; i < 2; ++i) {
 		s[3] ^= v[i];
 #ifdef CONFIG_64BIT
 		s[0] += s[1]; s[1] = rol64(s[1], 13); s[1] ^= s[0]; s[0] = rol64(s[0], 32);
@@ -1245,33 +1242,17 @@ int random_online_cpu(unsigned int cpu)
 }
 #endif
 
-static unsigned long get_reg(struct fast_pool *f, struct pt_regs *regs)
-{
-	unsigned long *ptr = (unsigned long *)regs;
-	unsigned int idx;
-
-	if (regs == NULL)
-		return 0;
-	idx = READ_ONCE(f->reg_idx);
-	if (idx >= sizeof(struct pt_regs) / sizeof(unsigned long))
-		idx = 0;
-	ptr += idx++;
-	WRITE_ONCE(f->reg_idx, idx);
-	return *ptr;
-}
-
 static void mix_interrupt_randomness(struct work_struct *work)
 {
 	struct fast_pool *fast_pool = container_of(work, struct fast_pool, mix);
 	/*
-	 * The size of the copied stack pool is explicitly 16 bytes so that we
-	 * tax mix_pool_byte()'s compression function the same amount on all
-	 * platforms. This means on 64-bit we copy half the pool into this,
-	 * while on 32-bit we copy all of it. The entropy is supposed to be
-	 * sufficiently dispersed between bits that in the sponge-like
-	 * half case, on average we don't wind up "losing" some.
+	 * The size of the copied stack pool is explicitly 2 longs so that we
+	 * only ever ingest half of the siphash output each time, retaining
+	 * the other half as the next "key" that carries over. The entropy is
+	 * supposed to be sufficiently dispersed between bits so on average
+	 * we don't wind up "losing" some.
 	 */
-	u8 pool[16];
+	unsigned long pool[2];
 
 	/* Check to see if we're running on the wrong CPU due to hotplug. */
 	local_irq_disable();
@@ -1303,36 +1284,21 @@ static void mix_interrupt_randomness(str
 void add_interrupt_randomness(int irq)
 {
 	enum { MIX_INFLIGHT = 1U << 31 };
-	unsigned long cycles = random_get_entropy(), now = jiffies;
+	unsigned long entropy = random_get_entropy();
 	struct fast_pool *fast_pool = this_cpu_ptr(&irq_randomness);
 	struct pt_regs *regs = get_irq_regs();
 	unsigned int new_count;
-	union {
-		u32 u32[4];
-		u64 u64[2];
-		unsigned long longs[16 / sizeof(long)];
-	} irq_data;
-
-	if (cycles == 0)
-		cycles = get_reg(fast_pool, regs);
-
-	if (sizeof(unsigned long) == 8) {
-		irq_data.u64[0] = cycles ^ rol64(now, 32) ^ irq;
-		irq_data.u64[1] = regs ? instruction_pointer(regs) : _RET_IP_;
-	} else {
-		irq_data.u32[0] = cycles ^ irq;
-		irq_data.u32[1] = now;
-		irq_data.u32[2] = regs ? instruction_pointer(regs) : _RET_IP_;
-		irq_data.u32[3] = get_reg(fast_pool, regs);
-	}
 
-	fast_mix(fast_pool->pool, irq_data.longs);
+	fast_mix(fast_pool->pool, (unsigned long[2]){
+		entropy,
+		(regs ? instruction_pointer(regs) : _RET_IP_) ^ swab(irq)
+	});
 	new_count = ++fast_pool->count;
 
 	if (new_count & MIX_INFLIGHT)
 		return;
 
-	if (new_count < 64 && (!time_after(now, fast_pool->last + HZ) ||
+	if (new_count < 64 && (!time_is_before_jiffies(fast_pool->last + HZ) ||
 			       unlikely(crng_init == 0)))
 		return;
 
@@ -1368,28 +1334,28 @@ static void entropy_timer(struct timer_l
 static void try_to_generate_entropy(void)
 {
 	struct {
-		unsigned long cycles;
+		unsigned long entropy;
 		struct timer_list timer;
 	} stack;
 
-	stack.cycles = random_get_entropy();
+	stack.entropy = random_get_entropy();
 
 	/* Slow counter - or none. Don't even bother */
-	if (stack.cycles == random_get_entropy())
+	if (stack.entropy == random_get_entropy())
 		return;
 
 	timer_setup_on_stack(&stack.timer, entropy_timer, 0);
 	while (!crng_ready() && !signal_pending(current)) {
 		if (!timer_pending(&stack.timer))
 			mod_timer(&stack.timer, jiffies + 1);
-		mix_pool_bytes(&stack.cycles, sizeof(stack.cycles));
+		mix_pool_bytes(&stack.entropy, sizeof(stack.entropy));
 		schedule();
-		stack.cycles = random_get_entropy();
+		stack.entropy = random_get_entropy();
 	}
 
 	del_timer_sync(&stack.timer);
 	destroy_timer_on_stack(&stack.timer);
-	mix_pool_bytes(&stack.cycles, sizeof(stack.cycles));
+	mix_pool_bytes(&stack.entropy, sizeof(stack.entropy));
 }
 
 
author maintainer
#150 lore raw

PATCH 150/234 random: do not use batches when !crng_ready()

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit cbe89e5a375a51bbb952929b93fa973416fea74e upstream.

It's too hard to keep the batches synchronized, and pointless anyway, since in !crng_ready(), we're updating the base_crng key really often, where batching only hurts. So instead, if the crng isn't ready, just call into get_random_bytes(). At this stage nothing is performance critical anyhow.

---
 drivers/char/random.c |  158 ++++++++++++++++++--------------------------------
 1 file changed, 59 insertions(+), 99 deletions(-)
drivers/char/random.c +59 -99
@@ -233,10 +233,7 @@ static void _warn_unseeded_randomness(co
  *
  *********************************************************************/
 
-enum {
-	CRNG_RESEED_INTERVAL = 300 * HZ,
-	CRNG_INIT_CNT_THRESH = 2 * CHACHA20_KEY_SIZE
-};
+enum { CRNG_RESEED_INTERVAL = 300 * HZ };
 
 static struct {
 	u8 key[CHACHA20_KEY_SIZE] __aligned(__alignof__(long));
@@ -258,6 +255,8 @@ static DEFINE_PER_CPU(struct crng, crngs
 
 /* Used by crng_reseed() to extract a new seed from the input pool. */
 static bool drain_entropy(void *buf, size_t nbytes);
+/* Used by crng_make_state() to extract a new seed when crng_init==0. */
+static void extract_entropy(void *buf, size_t nbytes);
 
 /*
  * This extracts a new crng key from the input pool, but only if there is a
@@ -382,17 +381,20 @@ static void crng_make_state(u32 chacha_s
 	/*
 	 * For the fast path, we check whether we're ready, unlocked first, and
 	 * then re-check once locked later. In the case where we're really not
-	 * ready, we do fast key erasure with the base_crng directly, because
-	 * this is what crng_pre_init_inject() mutates during early init.
+	 * ready, we do fast key erasure with the base_crng directly, extracting
+	 * when crng_init==0.
 	 */
 	if (!crng_ready()) {
 		bool ready;
 
 		spin_lock_irqsave(&base_crng.lock, flags);
 		ready = crng_ready();
-		if (!ready)
+		if (!ready) {
+			if (crng_init == 0)
+				extract_entropy(base_crng.key, sizeof(base_crng.key));
 			crng_fast_key_erasure(base_crng.key, chacha_state,
 					      random_data, random_data_len);
+		}
 		spin_unlock_irqrestore(&base_crng.lock, flags);
 		if (!ready)
 			return;
@@ -433,50 +435,6 @@ static void crng_make_state(u32 chacha_s
 	local_irq_restore(flags);
 }
 
-/*
- * This function is for crng_init == 0 only. It loads entropy directly
- * into the crng's key, without going through the input pool. It is,
- * generally speaking, not very safe, but we use this only at early
- * boot time when it's better to have something there rather than
- * nothing.
- *
- * If account is set, then the crng_init_cnt counter is incremented.
- * This shouldn't be set by functions like add_device_randomness(),
- * where we can't trust the buffer passed to it is guaranteed to be
- * unpredictable (so it might not have any entropy at all).
- */
-static void crng_pre_init_inject(const void *input, size_t len, bool account)
-{
-	static int crng_init_cnt = 0;
-	struct blake2s_state hash;
-	unsigned long flags;
-
-	blake2s_init(&hash, sizeof(base_crng.key));
-
-	spin_lock_irqsave(&base_crng.lock, flags);
-	if (crng_init != 0) {
-		spin_unlock_irqrestore(&base_crng.lock, flags);
-		return;
-	}
-
-	blake2s_update(&hash, base_crng.key, sizeof(base_crng.key));
-	blake2s_update(&hash, input, len);
-	blake2s_final(&hash, base_crng.key);
-
-	if (account) {
-		crng_init_cnt += min_t(size_t, len, CRNG_INIT_CNT_THRESH - crng_init_cnt);
-		if (crng_init_cnt >= CRNG_INIT_CNT_THRESH) {
-			++base_crng.generation;
-			crng_init = 1;
-		}
-	}
-
-	spin_unlock_irqrestore(&base_crng.lock, flags);
-
-	if (crng_init == 1)
-		pr_notice("fast init done\n");
-}
-
 static void _get_random_bytes(void *buf, size_t nbytes)
 {
 	u32 chacha_state[CHACHA20_BLOCK_SIZE / sizeof(u32)];
@@ -622,6 +580,11 @@ u64 get_random_u64(void)
 
 	warn_unseeded_randomness(&previous);
 
+	if  (!crng_ready()) {
+		_get_random_bytes(&ret, sizeof(ret));
+		return ret;
+	}
+
 	local_irq_save(flags);
 	batch = raw_cpu_ptr(&batched_entropy_u64);
 
@@ -655,6 +618,11 @@ u32 get_random_u32(void)
 
 	warn_unseeded_randomness(&previous);
 
+	if  (!crng_ready()) {
+		_get_random_bytes(&ret, sizeof(ret));
+		return ret;
+	}
+
 	local_irq_save(flags);
 	batch = raw_cpu_ptr(&batched_entropy_u32);
 
@@ -776,7 +744,8 @@ EXPORT_SYMBOL(get_random_bytes_arch);
 
 enum {
 	POOL_BITS = BLAKE2S_HASH_SIZE * 8,
-	POOL_MIN_BITS = POOL_BITS /* No point in settling for less. */
+	POOL_MIN_BITS = POOL_BITS, /* No point in settling for less. */
+	POOL_FAST_INIT_BITS = POOL_MIN_BITS / 2
 };
 
 /* For notifying userspace should write into /dev/random. */
@@ -813,24 +782,6 @@ static void mix_pool_bytes(const void *i
 	spin_unlock_irqrestore(&input_pool.lock, flags);
 }
 
-static void credit_entropy_bits(size_t nbits)
-{
-	unsigned int entropy_count, orig, add;
-
-	if (!nbits)
-		return;
-
-	add = min_t(size_t, nbits, POOL_BITS);
-
-	do {
-		orig = READ_ONCE(input_pool.entropy_count);
-		entropy_count = min_t(unsigned int, POOL_BITS, orig + add);
-	} while (cmpxchg(&input_pool.entropy_count, orig, entropy_count) != orig);
-
-	if (!crng_ready() && entropy_count >= POOL_MIN_BITS)
-		crng_reseed();
-}
-
 /*
  * This is an HKDF-like construction for using the hashed collected entropy
  * as a PRF key, that's then expanded block-by-block.
@@ -896,6 +847,33 @@ static bool drain_entropy(void *buf, siz
 	return true;
 }
 
+static void credit_entropy_bits(size_t nbits)
+{
+	unsigned int entropy_count, orig, add;
+	unsigned long flags;
+
+	if (!nbits)
+		return;
+
+	add = min_t(size_t, nbits, POOL_BITS);
+
+	do {
+		orig = READ_ONCE(input_pool.entropy_count);
+		entropy_count = min_t(unsigned int, POOL_BITS, orig + add);
+	} while (cmpxchg(&input_pool.entropy_count, orig, entropy_count) != orig);
+
+	if (!crng_ready() && entropy_count >= POOL_MIN_BITS)
+		crng_reseed();
+	else if (unlikely(crng_init == 0 && entropy_count >= POOL_FAST_INIT_BITS)) {
+		spin_lock_irqsave(&base_crng.lock, flags);
+		if (crng_init == 0) {
+			extract_entropy(base_crng.key, sizeof(base_crng.key));
+			crng_init = 1;
+		}
+		spin_unlock_irqrestore(&base_crng.lock, flags);
+	}
+}
+
 
 /**********************************************************************
  *
@@ -938,9 +916,9 @@ static bool drain_entropy(void *buf, siz
  * entropy as specified by the caller. If the entropy pool is full it will
  * block until more entropy is needed.
  *
- * add_bootloader_randomness() is the same as add_hwgenerator_randomness() or
- * add_device_randomness(), depending on whether or not the configuration
- * option CONFIG_RANDOM_TRUST_BOOTLOADER is set.
+ * add_bootloader_randomness() is called by bootloader drivers, such as EFI
+ * and device tree, and credits its input depending on whether or not the
+ * configuration option CONFIG_RANDOM_TRUST_BOOTLOADER is set.
  *
  * add_interrupt_randomness() uses the interrupt timing as random
  * inputs to the entropy pool. Using the cycle counters and the irq source
@@ -1020,9 +998,6 @@ void add_device_randomness(const void *b
 	unsigned long entropy = random_get_entropy();
 	unsigned long flags;
 
-	if (crng_init == 0 && size)
-		crng_pre_init_inject(buf, size, false);
-
 	spin_lock_irqsave(&input_pool.lock, flags);
 	_mix_pool_bytes(&entropy, sizeof(entropy));
 	_mix_pool_bytes(buf, size);
@@ -1138,12 +1113,6 @@ void rand_initialize_disk(struct gendisk
 void add_hwgenerator_randomness(const void *buffer, size_t count,
 				size_t entropy)
 {
-	if (unlikely(crng_init == 0 && entropy < POOL_MIN_BITS)) {
-		crng_pre_init_inject(buffer, count, true);
-		mix_pool_bytes(buffer, count);
-		return;
-	}
-
 	/*
 	 * Throttle writing if we're above the trickle threshold.
 	 * We'll be woken up again once below POOL_MIN_BITS, when
@@ -1151,7 +1120,7 @@ void add_hwgenerator_randomness(const vo
 	 * CRNG_RESEED_INTERVAL has elapsed.
 	 */
 	wait_event_interruptible_timeout(random_write_wait,
-			!system_wq || kthread_should_stop() ||
+			kthread_should_stop() ||
 			input_pool.entropy_count < POOL_MIN_BITS,
 			CRNG_RESEED_INTERVAL);
 	mix_pool_bytes(buffer, count);
@@ -1160,17 +1129,14 @@ void add_hwgenerator_randomness(const vo
 EXPORT_SYMBOL_GPL(add_hwgenerator_randomness);
 
 /*
- * Handle random seed passed by bootloader.
- * If the seed is trustworthy, it would be regarded as hardware RNGs. Otherwise
- * it would be regarded as device data.
- * The decision is controlled by CONFIG_RANDOM_TRUST_BOOTLOADER.
+ * Handle random seed passed by bootloader, and credit it if
+ * CONFIG_RANDOM_TRUST_BOOTLOADER is set.
  */
 void add_bootloader_randomness(const void *buf, size_t size)
 {
+	mix_pool_bytes(buf, size);
 	if (trust_bootloader)
-		add_hwgenerator_randomness(buf, size, size * 8);
-	else
-		add_device_randomness(buf, size);
+		credit_entropy_bits(size * 8);
 }
 EXPORT_SYMBOL_GPL(add_bootloader_randomness);
 
@@ -1270,13 +1236,8 @@ static void mix_interrupt_randomness(str
 	fast_pool->last = jiffies;
 	local_irq_enable();
 
-	if (unlikely(crng_init == 0)) {
-		crng_pre_init_inject(pool, sizeof(pool), true);
-		mix_pool_bytes(pool, sizeof(pool));
-	} else {
-		mix_pool_bytes(pool, sizeof(pool));
-		credit_entropy_bits(1);
-	}
+	mix_pool_bytes(pool, sizeof(pool));
+	credit_entropy_bits(1);
 
 	memzero_explicit(pool, sizeof(pool));
 }
@@ -1298,8 +1259,7 @@ void add_interrupt_randomness(int irq)
 	if (new_count & MIX_INFLIGHT)
 		return;
 
-	if (new_count < 64 && (!time_is_before_jiffies(fast_pool->last + HZ) ||
-			       unlikely(crng_init == 0)))
+	if (new_count < 64 && !time_is_before_jiffies(fast_pool->last + HZ))
 		return;
 
 	if (unlikely(!fast_pool->mix.func))
author maintainer
#151 lore raw

PATCH 151/234 random: do not pretend to handle premature next security model

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit e85c0fc1d94c52483a603651748d4c76d6aa1c6b upstream.

Per the thread linked below, "premature next" is not considered to be a realistic threat model, and leads to more serious security problems.

"Premature next" is the scenario in which:

- Attacker compromises the current state of a fully initialized RNG via some kind of infoleak.
- New bits of entropy are added directly to the key used to generate the /dev/urandom stream, without any buffering or pooling.
- Attacker then, somehow having read access to /dev/urandom, samples RNG output and brute forces the individual new bits that were added.
- Result: the RNG never "recovers" from the initial compromise, a so-called violation of what academics term "post-compromise security".

The usual solutions to this involve some form of delaying when entropy gets mixed into the crng. With Fortuna, this involves multiple input buckets. With what the Linux RNG was trying to do prior, this involves entropy estimation.

However, by delaying when entropy gets mixed in, it also means that RNG compromises are extremely dangerous during the window of time before the RNG has gathered enough entropy, during which time nonces may become predictable (or repeated), ephemeral keys may not be secret, and so forth. Moreover, it's unclear how realistic "premature next" is from an attack perspective, if these attacks even make sense in practice.

Put together -- and discussed in more detail in the thread below -- these constitute grounds for just doing away with the current code that pretends to handle premature next. I say "pretends" because it wasn't doing an especially great job at it either; should we change our mind about this direction, we would probably implement Fortuna to "fix" the "problem", in which case, removing the pretend solution still makes sense.

This also reduces the crng reseed period from 5 minutes down to 1 minute. The rationale from the thread might lead us toward reducing that even further in the future (or even eliminating it), but that remains a topic of a future commit.

At a high level, this patch changes semantics from:

Before: Seed for the first time after 256 "bits" of estimated entropy have been accumulated since the system booted. Thereafter, reseed once every five minutes, but only if 256 new "bits" have been accumulated since the last reseeding.

After: Seed for the first time after 256 "bits" of estimated entropy have been accumulated since the system booted. Thereafter, reseed once every minute.

Most of this patch is renaming and removing: POOL_MIN_BITS becomes POOL_INIT_BITS, credit_entropy_bits() becomes credit_init_bits(), crng_reseed() loses its "force" parameter since it's now always true, the drain_entropy() function no longer has any use so it's removed, entropy estimation is skipped if we've already init'd, the various notifiers for "low on entropy" are now only active prior to init, and finally, some documentation comments are cleaned up here and there.

---
 drivers/char/random.c |  174 +++++++++++++++++---------------------------------
 1 file changed, 62 insertions(+), 112 deletions(-)
drivers/char/random.c +62 -112
@@ -15,14 +15,12 @@
  *   - Sysctl interface.
  *
  * The high level overview is that there is one input pool, into which
- * various pieces of data are hashed. Some of that data is then "credited" as
- * having a certain number of bits of entropy. When enough bits of entropy are
- * available, the hash is finalized and handed as a key to a stream cipher that
- * expands it indefinitely for various consumers. This key is periodically
- * refreshed as the various entropy collectors, described below, add data to the
- * input pool and credit it. There is currently no Fortuna-like scheduler
- * involved, which can lead to malicious entropy sources causing a premature
- * reseed, and the entropy estimates are, at best, conservative guesses.
+ * various pieces of data are hashed. Prior to initialization, some of that
+ * data is then "credited" as having a certain number of bits of entropy.
+ * When enough bits of entropy are available, the hash is finalized and
+ * handed as a key to a stream cipher that expands it indefinitely for
+ * various consumers. This key is periodically refreshed as the various
+ * entropy collectors, described below, add data to the input pool.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
@@ -233,7 +231,10 @@ static void _warn_unseeded_randomness(co
  *
  *********************************************************************/
 
-enum { CRNG_RESEED_INTERVAL = 300 * HZ };
+enum {
+	CRNG_RESEED_START_INTERVAL = HZ,
+	CRNG_RESEED_INTERVAL = 60 * HZ
+};
 
 static struct {
 	u8 key[CHACHA20_KEY_SIZE] __aligned(__alignof__(long));
@@ -253,16 +254,10 @@ static DEFINE_PER_CPU(struct crng, crngs
 	.generation = ULONG_MAX
 };
 
-/* Used by crng_reseed() to extract a new seed from the input pool. */
-static bool drain_entropy(void *buf, size_t nbytes);
-/* Used by crng_make_state() to extract a new seed when crng_init==0. */
+/* Used by crng_reseed() and crng_make_state() to extract a new seed from the input pool. */
 static void extract_entropy(void *buf, size_t nbytes);
 
-/*
- * This extracts a new crng key from the input pool, but only if there is a
- * sufficient amount of entropy available, in order to mitigate bruteforcing
- * of newly added bits.
- */
+/* This extracts a new crng key from the input pool. */
 static void crng_reseed(void)
 {
 	unsigned long flags;
@@ -270,9 +265,7 @@ static void crng_reseed(void)
 	u8 key[CHACHA20_KEY_SIZE];
 	bool finalize_init = false;
 
-	/* Only reseed if we can, to prevent brute forcing a small amount of new bits. */
-	if (!drain_entropy(key, sizeof(key)))
-		return;
+	extract_entropy(key, sizeof(key));
 
 	/*
 	 * We copy the new key into the base_crng, overwriting the old one,
@@ -344,10 +337,10 @@ static void crng_fast_key_erasure(u8 key
 }
 
 /*
- * Return whether the crng seed is considered to be sufficiently
- * old that a reseeding might be attempted. This happens if the last
- * reseeding was CRNG_RESEED_INTERVAL ago, or during early boot, at
- * an interval proportional to the uptime.
+ * Return whether the crng seed is considered to be sufficiently old
+ * that a reseeding is needed. This happens if the last reseeding
+ * was CRNG_RESEED_INTERVAL ago, or during early boot, at an interval
+ * proportional to the uptime.
  */
 static bool crng_has_old_seed(void)
 {
@@ -359,7 +352,7 @@ static bool crng_has_old_seed(void)
 		if (uptime >= CRNG_RESEED_INTERVAL / HZ * 2)
 			WRITE_ONCE(early_boot, false);
 		else
-			interval = max_t(unsigned int, 5 * HZ,
+			interval = max_t(unsigned int, CRNG_RESEED_START_INTERVAL,
 					 (unsigned int)uptime / 2 * HZ);
 	}
 	return time_after(jiffies, READ_ONCE(base_crng.birth) + interval);
@@ -401,8 +394,8 @@ static void crng_make_state(u32 chacha_s
 	}
 
 	/*
-	 * If the base_crng is old enough, we try to reseed, which in turn
-	 * bumps the generation counter that we check below.
+	 * If the base_crng is old enough, we reseed, which in turn bumps the
+	 * generation counter that we check below.
 	 */
 	if (unlikely(crng_has_old_seed()))
 		crng_reseed();
@@ -731,30 +724,24 @@ EXPORT_SYMBOL(get_random_bytes_arch);
  *
  * After which, if added entropy should be credited:
  *
- *     static void credit_entropy_bits(size_t nbits)
+ *     static void credit_init_bits(size_t nbits)
  *
- * Finally, extract entropy via these two, with the latter one
- * setting the entropy count to zero and extracting only if there
- * is POOL_MIN_BITS entropy credited prior:
+ * Finally, extract entropy via:
  *
  *     static void extract_entropy(void *buf, size_t nbytes)
- *     static bool drain_entropy(void *buf, size_t nbytes)
  *
  **********************************************************************/
 
 enum {
 	POOL_BITS = BLAKE2S_HASH_SIZE * 8,
-	POOL_MIN_BITS = POOL_BITS, /* No point in settling for less. */
-	POOL_FAST_INIT_BITS = POOL_MIN_BITS / 2
+	POOL_INIT_BITS = POOL_BITS, /* No point in settling for less. */
+	POOL_FAST_INIT_BITS = POOL_INIT_BITS / 2
 };
 
-/* For notifying userspace should write into /dev/random. */
-static DECLARE_WAIT_QUEUE_HEAD(random_write_wait);
-
 static struct {
 	struct blake2s_state hash;
 	spinlock_t lock;
-	unsigned int entropy_count;
+	unsigned int init_bits;
 } input_pool = {
 	.hash.h = { BLAKE2S_IV0 ^ (0x01010000 | BLAKE2S_HASH_SIZE),
 		    BLAKE2S_IV1, BLAKE2S_IV2, BLAKE2S_IV3, BLAKE2S_IV4,
@@ -769,9 +756,9 @@ static void _mix_pool_bytes(const void *
 }
 
 /*
- * This function adds bytes into the entropy "pool".  It does not
- * update the entropy estimate.  The caller should call
- * credit_entropy_bits if this is appropriate.
+ * This function adds bytes into the input pool. It does not
+ * update the initialization bit counter; the caller should call
+ * credit_init_bits if this is appropriate.
  */
 static void mix_pool_bytes(const void *in, size_t nbytes)
 {
@@ -828,43 +815,24 @@ static void extract_entropy(void *buf, s
 	memzero_explicit(&block, sizeof(block));
 }
 
-/*
- * First we make sure we have POOL_MIN_BITS of entropy in the pool, and then we
- * set the entropy count to zero (but don't actually touch any data). Only then
- * can we extract a new key with extract_entropy().
- */
-static bool drain_entropy(void *buf, size_t nbytes)
-{
-	unsigned int entropy_count;
-	do {
-		entropy_count = READ_ONCE(input_pool.entropy_count);
-		if (entropy_count < POOL_MIN_BITS)
-			return false;
-	} while (cmpxchg(&input_pool.entropy_count, entropy_count, 0) != entropy_count);
-	extract_entropy(buf, nbytes);
-	wake_up_interruptible(&random_write_wait);
-	kill_fasync(&fasync, SIGIO, POLL_OUT);
-	return true;
-}
-
-static void credit_entropy_bits(size_t nbits)
+static void credit_init_bits(size_t nbits)
 {
-	unsigned int entropy_count, orig, add;
+	unsigned int init_bits, orig, add;
 	unsigned long flags;
 
-	if (!nbits)
+	if (crng_ready() || !nbits)
 		return;
 
 	add = min_t(size_t, nbits, POOL_BITS);
 
 	do {
-		orig = READ_ONCE(input_pool.entropy_count);
-		entropy_count = min_t(unsigned int, POOL_BITS, orig + add);
-	} while (cmpxchg(&input_pool.entropy_count, orig, entropy_count) != orig);
+		orig = READ_ONCE(input_pool.init_bits);
+		init_bits = min_t(unsigned int, POOL_BITS, orig + add);
+	} while (cmpxchg(&input_pool.init_bits, orig, init_bits) != orig);
 
-	if (!crng_ready() && entropy_count >= POOL_MIN_BITS)
+	if (!crng_ready() && init_bits >= POOL_INIT_BITS)
 		crng_reseed();
-	else if (unlikely(crng_init == 0 && entropy_count >= POOL_FAST_INIT_BITS)) {
+	else if (unlikely(crng_init == 0 && init_bits >= POOL_FAST_INIT_BITS)) {
 		spin_lock_irqsave(&base_crng.lock, flags);
 		if (crng_init == 0) {
 			extract_entropy(base_crng.key, sizeof(base_crng.key));
@@ -970,13 +938,10 @@ int __init rand_initialize(void)
 	_mix_pool_bytes(&now, sizeof(now));
 	_mix_pool_bytes(utsname(), sizeof(*(utsname())));
 
-	extract_entropy(base_crng.key, sizeof(base_crng.key));
-	++base_crng.generation;
-
-	if (arch_init && trust_cpu && !crng_ready()) {
-		crng_init = 2;
-		pr_notice("crng init done (trusting CPU's manufacturer)\n");
-	}
+	if (crng_ready())
+		crng_reseed();
+	else if (arch_init && trust_cpu)
+		credit_init_bits(BLAKE2S_BLOCK_SIZE * 8);
 
 	if (ratelimit_disable) {
 		urandom_warning.interval = 0;
@@ -1030,6 +995,9 @@ static void add_timer_randomness(struct
 	_mix_pool_bytes(&num, sizeof(num));
 	spin_unlock_irqrestore(&input_pool.lock, flags);
 
+	if (crng_ready())
+		return;
+
 	/*
 	 * Calculate number of bits of randomness we probably added.
 	 * We take into account the first, second and third-order deltas
@@ -1060,7 +1028,7 @@ static void add_timer_randomness(struct
 	 * Round down by 1 bit on general principles,
 	 * and limit entropy estimate to 12 bits.
 	 */
-	credit_entropy_bits(min_t(unsigned int, fls(delta >> 1), 11));
+	credit_init_bits(min_t(unsigned int, fls(delta >> 1), 11));
 }
 
 void add_input_randomness(unsigned int type, unsigned int code,
@@ -1113,18 +1081,15 @@ void rand_initialize_disk(struct gendisk
 void add_hwgenerator_randomness(const void *buffer, size_t count,
 				size_t entropy)
 {
+	mix_pool_bytes(buffer, count);
+	credit_init_bits(entropy);
+
 	/*
-	 * Throttle writing if we're above the trickle threshold.
-	 * We'll be woken up again once below POOL_MIN_BITS, when
-	 * the calling thread is about to terminate, or once
-	 * CRNG_RESEED_INTERVAL has elapsed.
+	 * Throttle writing to once every CRNG_RESEED_INTERVAL, unless
+	 * we're not yet initialized.
 	 */
-	wait_event_interruptible_timeout(random_write_wait,
-			kthread_should_stop() ||
-			input_pool.entropy_count < POOL_MIN_BITS,
-			CRNG_RESEED_INTERVAL);
-	mix_pool_bytes(buffer, count);
-	credit_entropy_bits(entropy);
+	if (!kthread_should_stop() && crng_ready())
+		schedule_timeout_interruptible(CRNG_RESEED_INTERVAL);
 }
 EXPORT_SYMBOL_GPL(add_hwgenerator_randomness);
 
@@ -1136,7 +1101,7 @@ void add_bootloader_randomness(const voi
 {
 	mix_pool_bytes(buf, size);
 	if (trust_bootloader)
-		credit_entropy_bits(size * 8);
+		credit_init_bits(size * 8);
 }
 EXPORT_SYMBOL_GPL(add_bootloader_randomness);
 
@@ -1237,7 +1202,7 @@ static void mix_interrupt_randomness(str
 	local_irq_enable();
 
 	mix_pool_bytes(pool, sizeof(pool));
-	credit_entropy_bits(1);
+	credit_init_bits(1);
 
 	memzero_explicit(pool, sizeof(pool));
 }
@@ -1284,7 +1249,7 @@ EXPORT_SYMBOL_GPL(add_interrupt_randomne
  */
 static void entropy_timer(struct timer_list *t)
 {
-	credit_entropy_bits(1);
+	credit_init_bits(1);
 }
 
 /*
@@ -1377,16 +1342,8 @@ SYSCALL_DEFINE3(getrandom, char __user *
 
 static __poll_t random_poll(struct file *file, poll_table *wait)
 {
-	__poll_t mask;
-
 	poll_wait(file, &crng_init_wait, wait);
-	poll_wait(file, &random_write_wait, wait);
-	mask = 0;
-	if (crng_ready())
-		mask |= EPOLLIN | EPOLLRDNORM;
-	if (input_pool.entropy_count < POOL_MIN_BITS)
-		mask |= EPOLLOUT | EPOLLWRNORM;
-	return mask;
+	return crng_ready() ? EPOLLIN | EPOLLRDNORM : EPOLLOUT | EPOLLWRNORM;
 }
 
 static int write_pool(const char __user *ubuf, size_t count)
@@ -1459,7 +1416,7 @@ static long random_ioctl(struct file *f,
 	switch (cmd) {
 	case RNDGETENTCNT:
 		/* Inherently racy, no point locking. */
-		if (put_user(input_pool.entropy_count, p))
+		if (put_user(input_pool.init_bits, p))
 			return -EFAULT;
 		return 0;
 	case RNDADDTOENTCNT:
@@ -1469,7 +1426,7 @@ static long random_ioctl(struct file *f,
 			return -EFAULT;
 		if (ent_count < 0)
 			return -EINVAL;
-		credit_entropy_bits(ent_count);
+		credit_init_bits(ent_count);
 		return 0;
 	case RNDADDENTROPY:
 		if (!capable(CAP_SYS_ADMIN))
@@ -1483,20 +1440,13 @@ static long random_ioctl(struct file *f,
 		retval = write_pool((const char __user *)p, size);
 		if (retval < 0)
 			return retval;
-		credit_entropy_bits(ent_count);
+		credit_init_bits(ent_count);
 		return 0;
 	case RNDZAPENTCNT:
 	case RNDCLEARPOOL:
-		/*
-		 * Clear the entropy pool counters. We no longer clear
-		 * the entropy pool, as that's silly.
-		 */
+		/* No longer has any effect. */
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
-		if (xchg(&input_pool.entropy_count, 0) >= POOL_MIN_BITS) {
-			wake_up_interruptible(&random_write_wait);
-			kill_fasync(&fasync, SIGIO, POLL_OUT);
-		}
 		return 0;
 	case RNDRESEEDCRNG:
 		if (!capable(CAP_SYS_ADMIN))
@@ -1553,7 +1503,7 @@ const struct file_operations urandom_fop
  *
  * - write_wakeup_threshold - the amount of entropy in the input pool
  *   below which write polls to /dev/random will unblock, requesting
- *   more entropy, tied to the POOL_MIN_BITS constant. It is writable
+ *   more entropy, tied to the POOL_INIT_BITS constant. It is writable
  *   to avoid breaking old userspaces, but writing to it does not
  *   change any behavior of the RNG.
  *
@@ -1568,7 +1518,7 @@ const struct file_operations urandom_fop
 #include <linux/sysctl.h>
 
 static int sysctl_random_min_urandom_seed = CRNG_RESEED_INTERVAL / HZ;
-static int sysctl_random_write_wakeup_bits = POOL_MIN_BITS;
+static int sysctl_random_write_wakeup_bits = POOL_INIT_BITS;
 static int sysctl_poolsize = POOL_BITS;
 static u8 sysctl_bootid[UUID_SIZE];
 
@@ -1625,7 +1575,7 @@ struct ctl_table random_table[] = {
 	},
 	{
 		.procname	= "entropy_avail",
-		.data		= &input_pool.entropy_count,
+		.data		= &input_pool.init_bits,
 		.maxlen		= sizeof(int),
 		.mode		= 0444,
 		.proc_handler	= proc_dointvec,
author maintainer
#152 lore raw

PATCH 152/234 random: order timer entropy functions below interrupt functions

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit a4b5c26b79ffdfcfb816c198f2fc2b1e7b5b580f upstream.

There are no code changes here; this is just a reordering of functions, so that in subsequent commits, the timer entropy functions can call into the interrupt ones.

---
 drivers/char/random.c |  238 +++++++++++++++++++++++++-------------------------
 1 file changed, 119 insertions(+), 119 deletions(-)
drivers/char/random.c +119 -119
@@ -851,13 +851,13 @@ static void credit_init_bits(size_t nbit
  * the above entropy accumulation routines:
  *
  *	void add_device_randomness(const void *buf, size_t size);
- *	void add_input_randomness(unsigned int type, unsigned int code,
- *	                          unsigned int value);
- *	void add_disk_randomness(struct gendisk *disk);
  *	void add_hwgenerator_randomness(const void *buffer, size_t count,
  *					size_t entropy);
  *	void add_bootloader_randomness(const void *buf, size_t size);
  *	void add_interrupt_randomness(int irq);
+ *	void add_input_randomness(unsigned int type, unsigned int code,
+ *	                          unsigned int value);
+ *	void add_disk_randomness(struct gendisk *disk);
  *
  * add_device_randomness() adds data to the input pool that
  * is likely to differ between two devices (or possibly even per boot).
@@ -867,19 +867,6 @@ static void credit_init_bits(size_t nbit
  * that might otherwise be identical and have very little entropy
  * available to them (particularly common in the embedded world).
  *
- * add_input_randomness() uses the input layer interrupt timing, as well
- * as the event type information from the hardware.
- *
- * add_disk_randomness() uses what amounts to the seek time of block
- * layer request events, on a per-disk_devt basis, as input to the
- * entropy pool. Note that high-speed solid state drives with very low
- * seek times do not make for good sources of entropy, as their seek
- * times are usually fairly consistent.
- *
- * The above two routines try to estimate how many bits of entropy
- * to credit. They do this by keeping track of the first and second
- * order deltas of the event timings.
- *
  * add_hwgenerator_randomness() is for true hardware RNGs, and will credit
  * entropy as specified by the caller. If the entropy pool is full it will
  * block until more entropy is needed.
@@ -893,6 +880,19 @@ static void credit_init_bits(size_t nbit
  * as inputs, it feeds the input pool roughly once a second or after 64
  * interrupts, crediting 1 bit of entropy for whichever comes first.
  *
+ * add_input_randomness() uses the input layer interrupt timing, as well
+ * as the event type information from the hardware.
+ *
+ * add_disk_randomness() uses what amounts to the seek time of block
+ * layer request events, on a per-disk_devt basis, as input to the
+ * entropy pool. Note that high-speed solid state drives with very low
+ * seek times do not make for good sources of entropy, as their seek
+ * times are usually fairly consistent.
+ *
+ * The last two routines try to estimate how many bits of entropy
+ * to credit. They do this by keeping track of the first and second
+ * order deltas of the event timings.
+ *
  **********************************************************************/
 
 static bool trust_cpu __ro_after_init = IS_ENABLED(CONFIG_RANDOM_TRUST_CPU);
@@ -970,109 +970,6 @@ void add_device_randomness(const void *b
 }
 EXPORT_SYMBOL(add_device_randomness);
 
-/* There is one of these per entropy source */
-struct timer_rand_state {
-	unsigned long last_time;
-	long last_delta, last_delta2;
-};
-
-/*
- * This function adds entropy to the entropy "pool" by using timing
- * delays.  It uses the timer_rand_state structure to make an estimate
- * of how many bits of entropy this call has added to the pool.
- *
- * The number "num" is also added to the pool - it should somehow describe
- * the type of event which just happened.  This is currently 0-255 for
- * keyboard scan codes, and 256 upwards for interrupts.
- */
-static void add_timer_randomness(struct timer_rand_state *state, unsigned int num)
-{
-	unsigned long entropy = random_get_entropy(), now = jiffies, flags;
-	long delta, delta2, delta3;
-
-	spin_lock_irqsave(&input_pool.lock, flags);
-	_mix_pool_bytes(&entropy, sizeof(entropy));
-	_mix_pool_bytes(&num, sizeof(num));
-	spin_unlock_irqrestore(&input_pool.lock, flags);
-
-	if (crng_ready())
-		return;
-
-	/*
-	 * Calculate number of bits of randomness we probably added.
-	 * We take into account the first, second and third-order deltas
-	 * in order to make our estimate.
-	 */
-	delta = now - READ_ONCE(state->last_time);
-	WRITE_ONCE(state->last_time, now);
-
-	delta2 = delta - READ_ONCE(state->last_delta);
-	WRITE_ONCE(state->last_delta, delta);
-
-	delta3 = delta2 - READ_ONCE(state->last_delta2);
-	WRITE_ONCE(state->last_delta2, delta2);
-
-	if (delta < 0)
-		delta = -delta;
-	if (delta2 < 0)
-		delta2 = -delta2;
-	if (delta3 < 0)
-		delta3 = -delta3;
-	if (delta > delta2)
-		delta = delta2;
-	if (delta > delta3)
-		delta = delta3;
-
-	/*
-	 * delta is now minimum absolute delta.
-	 * Round down by 1 bit on general principles,
-	 * and limit entropy estimate to 12 bits.
-	 */
-	credit_init_bits(min_t(unsigned int, fls(delta >> 1), 11));
-}
-
-void add_input_randomness(unsigned int type, unsigned int code,
-			  unsigned int value)
-{
-	static unsigned char last_value;
-	static struct timer_rand_state input_timer_state = { INITIAL_JIFFIES };
-
-	/* Ignore autorepeat and the like. */
-	if (value == last_value)
-		return;
-
-	last_value = value;
-	add_timer_randomness(&input_timer_state,
-			     (type << 4) ^ code ^ (code >> 4) ^ value);
-}
-EXPORT_SYMBOL_GPL(add_input_randomness);
-
-#ifdef CONFIG_BLOCK
-void add_disk_randomness(struct gendisk *disk)
-{
-	if (!disk || !disk->random)
-		return;
-	/* First major is 1, so we get >= 0x200 here. */
-	add_timer_randomness(disk->random, 0x100 + disk_devt(disk));
-}
-EXPORT_SYMBOL_GPL(add_disk_randomness);
-
-void rand_initialize_disk(struct gendisk *disk)
-{
-	struct timer_rand_state *state;
-
-	/*
-	 * If kzalloc returns null, we just won't use that entropy
-	 * source.
-	 */
-	state = kzalloc(sizeof(struct timer_rand_state), GFP_KERNEL);
-	if (state) {
-		state->last_time = INITIAL_JIFFIES;
-		disk->random = state;
-	}
-}
-#endif
-
 /*
  * Interface for in-kernel drivers of true hardware RNGs.
  * Those devices may produce endless random bits and will be throttled
@@ -1234,6 +1131,109 @@ void add_interrupt_randomness(int irq)
 }
 EXPORT_SYMBOL_GPL(add_interrupt_randomness);
 
+/* There is one of these per entropy source */
+struct timer_rand_state {
+	unsigned long last_time;
+	long last_delta, last_delta2;
+};
+
+/*
+ * This function adds entropy to the entropy "pool" by using timing
+ * delays.  It uses the timer_rand_state structure to make an estimate
+ * of how many bits of entropy this call has added to the pool.
+ *
+ * The number "num" is also added to the pool - it should somehow describe
+ * the type of event which just happened.  This is currently 0-255 for
+ * keyboard scan codes, and 256 upwards for interrupts.
+ */
+static void add_timer_randomness(struct timer_rand_state *state, unsigned int num)
+{
+	unsigned long entropy = random_get_entropy(), now = jiffies, flags;
+	long delta, delta2, delta3;
+
+	spin_lock_irqsave(&input_pool.lock, flags);
+	_mix_pool_bytes(&entropy, sizeof(entropy));
+	_mix_pool_bytes(&num, sizeof(num));
+	spin_unlock_irqrestore(&input_pool.lock, flags);
+
+	if (crng_ready())
+		return;
+
+	/*
+	 * Calculate number of bits of randomness we probably added.
+	 * We take into account the first, second and third-order deltas
+	 * in order to make our estimate.
+	 */
+	delta = now - READ_ONCE(state->last_time);
+	WRITE_ONCE(state->last_time, now);
+
+	delta2 = delta - READ_ONCE(state->last_delta);
+	WRITE_ONCE(state->last_delta, delta);
+
+	delta3 = delta2 - READ_ONCE(state->last_delta2);
+	WRITE_ONCE(state->last_delta2, delta2);
+
+	if (delta < 0)
+		delta = -delta;
+	if (delta2 < 0)
+		delta2 = -delta2;
+	if (delta3 < 0)
+		delta3 = -delta3;
+	if (delta > delta2)
+		delta = delta2;
+	if (delta > delta3)
+		delta = delta3;
+
+	/*
+	 * delta is now minimum absolute delta.
+	 * Round down by 1 bit on general principles,
+	 * and limit entropy estimate to 12 bits.
+	 */
+	credit_init_bits(min_t(unsigned int, fls(delta >> 1), 11));
+}
+
+void add_input_randomness(unsigned int type, unsigned int code,
+			  unsigned int value)
+{
+	static unsigned char last_value;
+	static struct timer_rand_state input_timer_state = { INITIAL_JIFFIES };
+
+	/* Ignore autorepeat and the like. */
+	if (value == last_value)
+		return;
+
+	last_value = value;
+	add_timer_randomness(&input_timer_state,
+			     (type << 4) ^ code ^ (code >> 4) ^ value);
+}
+EXPORT_SYMBOL_GPL(add_input_randomness);
+
+#ifdef CONFIG_BLOCK
+void add_disk_randomness(struct gendisk *disk)
+{
+	if (!disk || !disk->random)
+		return;
+	/* First major is 1, so we get >= 0x200 here. */
+	add_timer_randomness(disk->random, 0x100 + disk_devt(disk));
+}
+EXPORT_SYMBOL_GPL(add_disk_randomness);
+
+void rand_initialize_disk(struct gendisk *disk)
+{
+	struct timer_rand_state *state;
+
+	/*
+	 * If kzalloc returns null, we just won't use that entropy
+	 * source.
+	 */
+	state = kzalloc(sizeof(struct timer_rand_state), GFP_KERNEL);
+	if (state) {
+		state->last_time = INITIAL_JIFFIES;
+		disk->random = state;
+	}
+}
+#endif
+
 /*
  * Each time the timer fires, we expect that we got an unpredictable
  * jump in the cycle counter. Even if the timer is running on another
author maintainer
#153 lore raw

PATCH 153/234 random: do not use input pool from hard IRQs

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit e3e33fc2ea7fcefd0d761db9d6219f83b4248f5c upstream.

Years ago, a separate fast pool was added for interrupts, so that the cost associated with taking the input pool spinlocks and mixing into it would be avoided in places where latency is critical. However, one oversight was that add_input_randomness() and add_disk_randomness() still sometimes are called directly from the interrupt handler, rather than being deferred to a thread. This means that some unlucky interrupts will be caught doing a blake2s_compress() call and potentially spinning on input_pool.lock, which can also be taken by unprivileged users by writing into /dev/urandom.

In order to fix this, add_timer_randomness() now checks whether it is being called from a hard IRQ and if so, just mixes into the per-cpu IRQ fast pool using fast_mix(), which is much faster and can be done lock-free. A nice consequence of this, as well, is that it means hard IRQ context FPU support is likely no longer useful.

The entropy estimation algorithm used by add_timer_randomness() is also somewhat different than the one used for add_interrupt_randomness(). The former looks at deltas of deltas of deltas, while the latter just waits for 64 interrupts for one bit or for one second since the last bit. In order to bridge these, and since add_interrupt_randomness() runs after an add_timer_randomness() that's called from hard IRQ, we add to the fast pool credit the related amount, and then subtract one to account for add_interrupt_randomness()'s contribution.

A downside of this, however, is that the num argument is potentially attacker controlled, which puts a bit more pressure on the fast_mix() sponge to do more than it's really intended to do. As a mitigating factor, the first 96 bits of input aren't attacker controlled (a cycle counter followed by zeros), which means it's essentially two rounds of siphash rather than one, which is somewhat better. It's also not that much different from add_interrupt_randomness()'s use of the irq stack instruction pointer register.

Cc Thomas Gleixner <[email protected]>
Cc Filipe Manana <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Borislav Petkov <[email protected]>
Cc Theodore Ts'o <[email protected]>
---
 drivers/char/random.c |   51 +++++++++++++++++++++++++++++++++++---------------
 1 file changed, 36 insertions(+), 15 deletions(-)
drivers/char/random.c +36 -15
@@ -1081,6 +1081,7 @@ static void mix_interrupt_randomness(str
 	 * we don't wind up "losing" some.
 	 */
 	unsigned long pool[2];
+	unsigned int count;
 
 	/* Check to see if we're running on the wrong CPU due to hotplug. */
 	local_irq_disable();
@@ -1094,12 +1095,13 @@ static void mix_interrupt_randomness(str
 	 * consistent view, before we reenable irqs again.
 	 */
 	memcpy(pool, fast_pool->pool, sizeof(pool));
+	count = fast_pool->count;
 	fast_pool->count = 0;
 	fast_pool->last = jiffies;
 	local_irq_enable();
 
 	mix_pool_bytes(pool, sizeof(pool));
-	credit_init_bits(1);
+	credit_init_bits(max(1u, (count & U16_MAX) / 64));
 
 	memzero_explicit(pool, sizeof(pool));
 }
@@ -1139,22 +1141,30 @@ struct timer_rand_state {
 
 /*
  * This function adds entropy to the entropy "pool" by using timing
- * delays.  It uses the timer_rand_state structure to make an estimate
- * of how many bits of entropy this call has added to the pool.
- *
- * The number "num" is also added to the pool - it should somehow describe
- * the type of event which just happened.  This is currently 0-255 for
- * keyboard scan codes, and 256 upwards for interrupts.
+ * delays. It uses the timer_rand_state structure to make an estimate
+ * of how many bits of entropy this call has added to the pool. The
+ * value "num" is also added to the pool; it should somehow describe
+ * the type of event that just happened.
  */
 static void add_timer_randomness(struct timer_rand_state *state, unsigned int num)
 {
 	unsigned long entropy = random_get_entropy(), now = jiffies, flags;
 	long delta, delta2, delta3;
+	unsigned int bits;
 
-	spin_lock_irqsave(&input_pool.lock, flags);
-	_mix_pool_bytes(&entropy, sizeof(entropy));
-	_mix_pool_bytes(&num, sizeof(num));
-	spin_unlock_irqrestore(&input_pool.lock, flags);
+	/*
+	 * If we're in a hard IRQ, add_interrupt_randomness() will be called
+	 * sometime after, so mix into the fast pool.
+	 */
+	if (in_irq()) {
+		fast_mix(this_cpu_ptr(&irq_randomness)->pool,
+			 (unsigned long[2]){ entropy, num });
+	} else {
+		spin_lock_irqsave(&input_pool.lock, flags);
+		_mix_pool_bytes(&entropy, sizeof(entropy));
+		_mix_pool_bytes(&num, sizeof(num));
+		spin_unlock_irqrestore(&input_pool.lock, flags);
+	}
 
 	if (crng_ready())
 		return;
@@ -1185,11 +1195,22 @@ static void add_timer_randomness(struct
 		delta = delta3;
 
 	/*
-	 * delta is now minimum absolute delta.
-	 * Round down by 1 bit on general principles,
-	 * and limit entropy estimate to 12 bits.
+	 * delta is now minimum absolute delta. Round down by 1 bit
+	 * on general principles, and limit entropy estimate to 11 bits.
+	 */
+	bits = min(fls(delta >> 1), 11);
+
+	/*
+	 * As mentioned above, if we're in a hard IRQ, add_interrupt_randomness()
+	 * will run after this, which uses a different crediting scheme of 1 bit
+	 * per every 64 interrupts. In order to let that function do accounting
+	 * close to the one in this function, we credit a full 64/64 bit per bit,
+	 * and then subtract one to account for the extra one added.
 	 */
-	credit_init_bits(min_t(unsigned int, fls(delta >> 1), 11));
+	if (in_irq())
+		this_cpu_ptr(&irq_randomness)->count += max(1u, bits * 64) - 1;
+	else
+		credit_init_bits(bits);
 }
 
 void add_input_randomness(unsigned int type, unsigned int code,
author maintainer
#154 lore raw

PATCH 154/234 random: help compiler out with fast_mix() by using simpler arguments

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 791332b3cbb080510954a4c152ce02af8832eac9 upstream.

Now that fast_mix() has more than one caller, gcc no longer inlines it. That's fine. But it also doesn't handle the compound literal argument we pass it very efficiently, nor does it handle the loop as well as it could. So just expand the code to spell out this function so that it generates the same code as it did before. Performance-wise, this now behaves as it did before the last commit. The difference in actual code size on x86 is 45 bytes, which is less than a cache line.

---
 drivers/char/random.c |   44 +++++++++++++++++++++++---------------------
 1 file changed, 23 insertions(+), 21 deletions(-)
drivers/char/random.c +23 -21
@@ -1026,25 +1026,30 @@ static DEFINE_PER_CPU(struct fast_pool,
  * and therefore this has no security on its own. s represents the
  * four-word SipHash state, while v represents a two-word input.
  */
-static void fast_mix(unsigned long s[4], const unsigned long v[2])
+static void fast_mix(unsigned long s[4], unsigned long v1, unsigned long v2)
 {
-	size_t i;
-
-	for (i = 0; i < 2; ++i) {
-		s[3] ^= v[i];
 #ifdef CONFIG_64BIT
-		s[0] += s[1]; s[1] = rol64(s[1], 13); s[1] ^= s[0]; s[0] = rol64(s[0], 32);
-		s[2] += s[3]; s[3] = rol64(s[3], 16); s[3] ^= s[2];
-		s[0] += s[3]; s[3] = rol64(s[3], 21); s[3] ^= s[0];
-		s[2] += s[1]; s[1] = rol64(s[1], 17); s[1] ^= s[2]; s[2] = rol64(s[2], 32);
+#define PERM() do { \
+	s[0] += s[1]; s[1] = rol64(s[1], 13); s[1] ^= s[0]; s[0] = rol64(s[0], 32); \
+	s[2] += s[3]; s[3] = rol64(s[3], 16); s[3] ^= s[2]; \
+	s[0] += s[3]; s[3] = rol64(s[3], 21); s[3] ^= s[0]; \
+	s[2] += s[1]; s[1] = rol64(s[1], 17); s[1] ^= s[2]; s[2] = rol64(s[2], 32); \
+} while (0)
 #else
-		s[0] += s[1]; s[1] = rol32(s[1],  5); s[1] ^= s[0]; s[0] = rol32(s[0], 16);
-		s[2] += s[3]; s[3] = rol32(s[3],  8); s[3] ^= s[2];
-		s[0] += s[3]; s[3] = rol32(s[3],  7); s[3] ^= s[0];
-		s[2] += s[1]; s[1] = rol32(s[1], 13); s[1] ^= s[2]; s[2] = rol32(s[2], 16);
+#define PERM() do { \
+	s[0] += s[1]; s[1] = rol32(s[1],  5); s[1] ^= s[0]; s[0] = rol32(s[0], 16); \
+	s[2] += s[3]; s[3] = rol32(s[3],  8); s[3] ^= s[2]; \
+	s[0] += s[3]; s[3] = rol32(s[3],  7); s[3] ^= s[0]; \
+	s[2] += s[1]; s[1] = rol32(s[1], 13); s[1] ^= s[2]; s[2] = rol32(s[2], 16); \
+} while (0)
 #endif
-		s[0] ^= v[i];
-	}
+
+	s[3] ^= v1;
+	PERM();
+	s[0] ^= v1;
+	s[3] ^= v2;
+	PERM();
+	s[0] ^= v2;
 }
 
 #ifdef CONFIG_SMP
@@ -1114,10 +1119,8 @@ void add_interrupt_randomness(int irq)
 	struct pt_regs *regs = get_irq_regs();
 	unsigned int new_count;
 
-	fast_mix(fast_pool->pool, (unsigned long[2]){
-		entropy,
-		(regs ? instruction_pointer(regs) : _RET_IP_) ^ swab(irq)
-	});
+	fast_mix(fast_pool->pool, entropy,
+		 (regs ? instruction_pointer(regs) : _RET_IP_) ^ swab(irq));
 	new_count = ++fast_pool->count;
 
 	if (new_count & MIX_INFLIGHT)
@@ -1157,8 +1160,7 @@ static void add_timer_randomness(struct
 	 * sometime after, so mix into the fast pool.
 	 */
 	if (in_irq()) {
-		fast_mix(this_cpu_ptr(&irq_randomness)->pool,
-			 (unsigned long[2]){ entropy, num });
+		fast_mix(this_cpu_ptr(&irq_randomness)->pool, entropy, num);
 	} else {
 		spin_lock_irqsave(&input_pool.lock, flags);
 		_mix_pool_bytes(&entropy, sizeof(entropy));
author maintainer
#155 lore raw

PATCH 155/234 siphash: use one source of truth for siphash permutations

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit e73aaae2fa9024832e1f42e30c787c7baf61d014 upstream.

The SipHash family of permutations is currently used in three places:

- siphash.c itself, used in the ordinary way it was intended.
- random32.c, in a construction from an anonymous contributor.
- random.c, as part of its fast_mix function.

Each one of these places reinvents the wheel with the same C code, same rotation constants, and same symmetry-breaking constants.

This commit tidies things up a bit by placing macros for the permutations and constants into siphash.h, where each of the three .c users can access them. It also leaves a note dissuading more users of them from emerging.

---
 drivers/char/random.c   |   30 +++++++-----------------------
 include/linux/prandom.h |   23 +++++++----------------
 include/linux/siphash.h |   28 ++++++++++++++++++++++++++++
 lib/siphash.c           |   32 ++++++++++----------------------
 4 files changed, 52 insertions(+), 61 deletions(-)
drivers/char/random.c +7 -23
@@ -51,6 +51,7 @@
 #include <linux/completion.h>
 #include <linux/uuid.h>
 #include <linux/uaccess.h>
+#include <linux/siphash.h>
 #include <crypto/chacha20.h>
 #include <crypto/blake2s.h>
 #include <asm/processor.h>
@@ -1011,12 +1012,11 @@ struct fast_pool {
 
 static DEFINE_PER_CPU(struct fast_pool, irq_randomness) = {
 #ifdef CONFIG_64BIT
-	/* SipHash constants */
-	.pool = { 0x736f6d6570736575UL, 0x646f72616e646f6dUL,
-		  0x6c7967656e657261UL, 0x7465646279746573UL }
+#define FASTMIX_PERM SIPHASH_PERMUTATION
+	.pool = { SIPHASH_CONST_0, SIPHASH_CONST_1, SIPHASH_CONST_2, SIPHASH_CONST_3 }
 #else
-	/* HalfSipHash constants */
-	.pool = { 0, 0, 0x6c796765U, 0x74656462U }
+#define FASTMIX_PERM HSIPHASH_PERMUTATION
+	.pool = { HSIPHASH_CONST_0, HSIPHASH_CONST_1, HSIPHASH_CONST_2, HSIPHASH_CONST_3 }
 #endif
 };
 
@@ -1028,27 +1028,11 @@ static DEFINE_PER_CPU(struct fast_pool,
  */
 static void fast_mix(unsigned long s[4], unsigned long v1, unsigned long v2)
 {
-#ifdef CONFIG_64BIT
-#define PERM() do { \
-	s[0] += s[1]; s[1] = rol64(s[1], 13); s[1] ^= s[0]; s[0] = rol64(s[0], 32); \
-	s[2] += s[3]; s[3] = rol64(s[3], 16); s[3] ^= s[2]; \
-	s[0] += s[3]; s[3] = rol64(s[3], 21); s[3] ^= s[0]; \
-	s[2] += s[1]; s[1] = rol64(s[1], 17); s[1] ^= s[2]; s[2] = rol64(s[2], 32); \
-} while (0)
-#else
-#define PERM() do { \
-	s[0] += s[1]; s[1] = rol32(s[1],  5); s[1] ^= s[0]; s[0] = rol32(s[0], 16); \
-	s[2] += s[3]; s[3] = rol32(s[3],  8); s[3] ^= s[2]; \
-	s[0] += s[3]; s[3] = rol32(s[3],  7); s[3] ^= s[0]; \
-	s[2] += s[1]; s[1] = rol32(s[1], 13); s[1] ^= s[2]; s[2] = rol32(s[2], 16); \
-} while (0)
-#endif
-
 	s[3] ^= v1;
-	PERM();
+	FASTMIX_PERM(s[0], s[1], s[2], s[3]);
 	s[0] ^= v1;
 	s[3] ^= v2;
-	PERM();
+	FASTMIX_PERM(s[0], s[1], s[2], s[3]);
 	s[0] ^= v2;
 }
 
include/linux/prandom.h +7 -16
@@ -10,6 +10,7 @@
 
 #include <linux/types.h>
 #include <linux/percpu.h>
+#include <linux/siphash.h>
 
 u32 prandom_u32(void);
 void prandom_bytes(void *buf, size_t nbytes);
@@ -21,15 +22,10 @@ void prandom_reseed_late(void);
  * The core SipHash round function.  Each line can be executed in
  * parallel given enough CPU resources.
  */
-#define PRND_SIPROUND(v0, v1, v2, v3) ( \
-	v0 += v1, v1 = rol64(v1, 13),  v2 += v3, v3 = rol64(v3, 16), \
-	v1 ^= v0, v0 = rol64(v0, 32),  v3 ^= v2,                     \
-	v0 += v3, v3 = rol64(v3, 21),  v2 += v1, v1 = rol64(v1, 17), \
-	v3 ^= v0,                      v1 ^= v2, v2 = rol64(v2, 32)  \
-)
+#define PRND_SIPROUND(v0, v1, v2, v3) SIPHASH_PERMUTATION(v0, v1, v2, v3)
 
-#define PRND_K0 (0x736f6d6570736575 ^ 0x6c7967656e657261)
-#define PRND_K1 (0x646f72616e646f6d ^ 0x7465646279746573)
+#define PRND_K0 (SIPHASH_CONST_0 ^ SIPHASH_CONST_2)
+#define PRND_K1 (SIPHASH_CONST_1 ^ SIPHASH_CONST_3)
 
 #elif BITS_PER_LONG == 32
 /*
@@ -37,14 +33,9 @@ void prandom_reseed_late(void);
  * This is weaker, but 32-bit machines are not used for high-traffic
  * applications, so there is less output for an attacker to analyze.
  */
-#define PRND_SIPROUND(v0, v1, v2, v3) ( \
-	v0 += v1, v1 = rol32(v1,  5),  v2 += v3, v3 = rol32(v3,  8), \
-	v1 ^= v0, v0 = rol32(v0, 16),  v3 ^= v2,                     \
-	v0 += v3, v3 = rol32(v3,  7),  v2 += v1, v1 = rol32(v1, 13), \
-	v3 ^= v0,                      v1 ^= v2, v2 = rol32(v2, 16)  \
-)
-#define PRND_K0 0x6c796765
-#define PRND_K1 0x74656462
+#define PRND_SIPROUND(v0, v1, v2, v3) HSIPHASH_PERMUTATION(v0, v1, v2, v3)
+#define PRND_K0 (HSIPHASH_CONST_0 ^ HSIPHASH_CONST_2)
+#define PRND_K1 (HSIPHASH_CONST_1 ^ HSIPHASH_CONST_3)
 
 #else
 #error Unsupported BITS_PER_LONG
include/linux/siphash.h +28 -0
@@ -136,4 +136,32 @@ static inline u32 hsiphash(const void *d
 	return ___hsiphash_aligned(data, len, key);
 }
 
+/*
+ * These macros expose the raw SipHash and HalfSipHash permutations.
+ * Do not use them directly! If you think you have a use for them,
+ * be sure to CC the maintainer of this file explaining why.
+ */
+
+#define SIPHASH_PERMUTATION(a, b, c, d) ( \
+	(a) += (b), (b) = rol64((b), 13), (b) ^= (a), (a) = rol64((a), 32), \
+	(c) += (d), (d) = rol64((d), 16), (d) ^= (c), \
+	(a) += (d), (d) = rol64((d), 21), (d) ^= (a), \
+	(c) += (b), (b) = rol64((b), 17), (b) ^= (c), (c) = rol64((c), 32))
+
+#define SIPHASH_CONST_0 0x736f6d6570736575ULL
+#define SIPHASH_CONST_1 0x646f72616e646f6dULL
+#define SIPHASH_CONST_2 0x6c7967656e657261ULL
+#define SIPHASH_CONST_3 0x7465646279746573ULL
+
+#define HSIPHASH_PERMUTATION(a, b, c, d) ( \
+	(a) += (b), (b) = rol32((b), 5), (b) ^= (a), (a) = rol32((a), 16), \
+	(c) += (d), (d) = rol32((d), 8), (d) ^= (c), \
+	(a) += (d), (d) = rol32((d), 7), (d) ^= (a), \
+	(c) += (b), (b) = rol32((b), 13), (b) ^= (c), (c) = rol32((c), 16))
+
+#define HSIPHASH_CONST_0 0U
+#define HSIPHASH_CONST_1 0U
+#define HSIPHASH_CONST_2 0x6c796765U
+#define HSIPHASH_CONST_3 0x74656462U
+
 #endif /* _LINUX_SIPHASH_H */
lib/siphash.c +10 -22
@@ -18,19 +18,13 @@
 #include <asm/word-at-a-time.h>
 #endif
 
-#define SIPROUND \
-	do { \
-	v0 += v1; v1 = rol64(v1, 13); v1 ^= v0; v0 = rol64(v0, 32); \
-	v2 += v3; v3 = rol64(v3, 16); v3 ^= v2; \
-	v0 += v3; v3 = rol64(v3, 21); v3 ^= v0; \
-	v2 += v1; v1 = rol64(v1, 17); v1 ^= v2; v2 = rol64(v2, 32); \
-	} while (0)
+#define SIPROUND SIPHASH_PERMUTATION(v0, v1, v2, v3)
 
 #define PREAMBLE(len) \
-	u64 v0 = 0x736f6d6570736575ULL; \
-	u64 v1 = 0x646f72616e646f6dULL; \
-	u64 v2 = 0x6c7967656e657261ULL; \
-	u64 v3 = 0x7465646279746573ULL; \
+	u64 v0 = SIPHASH_CONST_0; \
+	u64 v1 = SIPHASH_CONST_1; \
+	u64 v2 = SIPHASH_CONST_2; \
+	u64 v3 = SIPHASH_CONST_3; \
 	u64 b = ((u64)(len)) << 56; \
 	v3 ^= key->key[1]; \
 	v2 ^= key->key[0]; \
@@ -389,19 +383,13 @@ u32 hsiphash_4u32(const u32 first, const
 }
 EXPORT_SYMBOL(hsiphash_4u32);
 #else
-#define HSIPROUND \
-	do { \
-	v0 += v1; v1 = rol32(v1, 5); v1 ^= v0; v0 = rol32(v0, 16); \
-	v2 += v3; v3 = rol32(v3, 8); v3 ^= v2; \
-	v0 += v3; v3 = rol32(v3, 7); v3 ^= v0; \
-	v2 += v1; v1 = rol32(v1, 13); v1 ^= v2; v2 = rol32(v2, 16); \
-	} while (0)
+#define HSIPROUND HSIPHASH_PERMUTATION(v0, v1, v2, v3)
 
 #define HPREAMBLE(len) \
-	u32 v0 = 0; \
-	u32 v1 = 0; \
-	u32 v2 = 0x6c796765U; \
-	u32 v3 = 0x74656462U; \
+	u32 v0 = HSIPHASH_CONST_0; \
+	u32 v1 = HSIPHASH_CONST_1; \
+	u32 v2 = HSIPHASH_CONST_2; \
+	u32 v3 = HSIPHASH_CONST_3; \
 	u32 b = ((u32)(len)) << 24; \
 	v3 ^= key->key[1]; \
 	v2 ^= key->key[0]; \
author maintainer
#156 lore raw

PATCH 156/234 random: use symbolic constants for crng_init states

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit e3d2c5e79a999aa4e7d6f0127e16d3da5a4ff70d upstream.

crng_init represents a state machine, with three states, and various rules for transitions. For the longest time, we've been managing these with "0", "1", and "2", and expecting people to figure it out. To make the code more obvious, replace these with proper enum values representing the transition, and then redocument what each of these states mean.

---
 drivers/char/random.c |   38 +++++++++++++++++++-------------------
 1 file changed, 19 insertions(+), 19 deletions(-)
drivers/char/random.c +19 -19
@@ -70,16 +70,16 @@
  *********************************************************************/
 
 /*
- * crng_init =  0 --> Uninitialized
- *		1 --> Initialized
- *		2 --> Initialized from input_pool
- *
  * crng_init is protected by base_crng->lock, and only increases
- * its value (from 0->1->2).
+ * its value (from empty->early->ready).
  */
-static int crng_init = 0;
-#define crng_ready() (likely(crng_init > 1))
-/* Various types of waiters for crng_init->2 transition. */
+static enum {
+	CRNG_EMPTY = 0, /* Little to no entropy collected */
+	CRNG_EARLY = 1, /* At least POOL_EARLY_BITS collected */
+	CRNG_READY = 2  /* Fully initialized with POOL_READY_BITS collected */
+} crng_init = CRNG_EMPTY;
+#define crng_ready() (likely(crng_init >= CRNG_READY))
+/* Various types of waiters for crng_init->CRNG_READY transition. */
 static DECLARE_WAIT_QUEUE_HEAD(crng_init_wait);
 static struct fasync_struct *fasync;
 static DEFINE_SPINLOCK(random_ready_chain_lock);
@@ -282,7 +282,7 @@ static void crng_reseed(void)
 	WRITE_ONCE(base_crng.generation, next_gen);
 	WRITE_ONCE(base_crng.birth, jiffies);
 	if (!crng_ready()) {
-		crng_init = 2;
+		crng_init = CRNG_READY;
 		finalize_init = true;
 	}
 	spin_unlock_irqrestore(&base_crng.lock, flags);
@@ -376,7 +376,7 @@ static void crng_make_state(u32 chacha_s
 	 * For the fast path, we check whether we're ready, unlocked first, and
 	 * then re-check once locked later. In the case where we're really not
 	 * ready, we do fast key erasure with the base_crng directly, extracting
-	 * when crng_init==0.
+	 * when crng_init is CRNG_EMPTY.
 	 */
 	if (!crng_ready()) {
 		bool ready;
@@ -384,7 +384,7 @@ static void crng_make_state(u32 chacha_s
 		spin_lock_irqsave(&base_crng.lock, flags);
 		ready = crng_ready();
 		if (!ready) {
-			if (crng_init == 0)
+			if (crng_init == CRNG_EMPTY)
 				extract_entropy(base_crng.key, sizeof(base_crng.key));
 			crng_fast_key_erasure(base_crng.key, chacha_state,
 					      random_data, random_data_len);
@@ -735,8 +735,8 @@ EXPORT_SYMBOL(get_random_bytes_arch);
 
 enum {
 	POOL_BITS = BLAKE2S_HASH_SIZE * 8,
-	POOL_INIT_BITS = POOL_BITS, /* No point in settling for less. */
-	POOL_FAST_INIT_BITS = POOL_INIT_BITS / 2
+	POOL_READY_BITS = POOL_BITS, /* When crng_init->CRNG_READY */
+	POOL_EARLY_BITS = POOL_READY_BITS / 2 /* When crng_init->CRNG_EARLY */
 };
 
 static struct {
@@ -831,13 +831,13 @@ static void credit_init_bits(size_t nbit
 		init_bits = min_t(unsigned int, POOL_BITS, orig + add);
 	} while (cmpxchg(&input_pool.init_bits, orig, init_bits) != orig);
 
-	if (!crng_ready() && init_bits >= POOL_INIT_BITS)
+	if (!crng_ready() && init_bits >= POOL_READY_BITS)
 		crng_reseed();
-	else if (unlikely(crng_init == 0 && init_bits >= POOL_FAST_INIT_BITS)) {
+	else if (unlikely(crng_init == CRNG_EMPTY && init_bits >= POOL_EARLY_BITS)) {
 		spin_lock_irqsave(&base_crng.lock, flags);
-		if (crng_init == 0) {
+		if (crng_init == CRNG_EMPTY) {
 			extract_entropy(base_crng.key, sizeof(base_crng.key));
-			crng_init = 1;
+			crng_init = CRNG_EARLY;
 		}
 		spin_unlock_irqrestore(&base_crng.lock, flags);
 	}
@@ -1510,7 +1510,7 @@ const struct file_operations urandom_fop
  *
  * - write_wakeup_threshold - the amount of entropy in the input pool
  *   below which write polls to /dev/random will unblock, requesting
- *   more entropy, tied to the POOL_INIT_BITS constant. It is writable
+ *   more entropy, tied to the POOL_READY_BITS constant. It is writable
  *   to avoid breaking old userspaces, but writing to it does not
  *   change any behavior of the RNG.
  *
@@ -1525,7 +1525,7 @@ const struct file_operations urandom_fop
 #include <linux/sysctl.h>
 
 static int sysctl_random_min_urandom_seed = CRNG_RESEED_INTERVAL / HZ;
-static int sysctl_random_write_wakeup_bits = POOL_INIT_BITS;
+static int sysctl_random_write_wakeup_bits = POOL_READY_BITS;
 static int sysctl_poolsize = POOL_BITS;
 static u8 sysctl_bootid[UUID_SIZE];
 
author maintainer
#157 lore raw

PATCH 157/234 random: avoid initializing twice in credit race

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit fed7ef061686cc813b1f3d8d0edc6c35b4d3537b upstream.

Since all changes of crng_init now go through credit_init_bits(), we can fix a long standing race in which two concurrent callers of credit_init_bits() have the new bit count >= some threshold, but are doing so with crng_init as a lower threshold, checked outside of a lock, resulting in crng_reseed() or similar being called twice.

In order to fix this, we can use the original cmpxchg value of the bit count, and only change crng_init when the bit count transitions from below a threshold to meeting the threshold.

---
 drivers/char/random.c |   48 ++++++++++++++++++++++--------------------------
 1 file changed, 22 insertions(+), 26 deletions(-)
drivers/char/random.c +22 -26
@@ -264,7 +264,6 @@ static void crng_reseed(void)
 	unsigned long flags;
 	unsigned long next_gen;
 	u8 key[CHACHA20_KEY_SIZE];
-	bool finalize_init = false;
 
 	extract_entropy(key, sizeof(key));
 
@@ -281,28 +280,10 @@ static void crng_reseed(void)
 		++next_gen;
 	WRITE_ONCE(base_crng.generation, next_gen);
 	WRITE_ONCE(base_crng.birth, jiffies);
-	if (!crng_ready()) {
+	if (!crng_ready())
 		crng_init = CRNG_READY;
-		finalize_init = true;
-	}
 	spin_unlock_irqrestore(&base_crng.lock, flags);
 	memzero_explicit(key, sizeof(key));
-	if (finalize_init) {
-		process_random_ready_list();
-		wake_up_interruptible(&crng_init_wait);
-		kill_fasync(&fasync, SIGIO, POLL_IN);
-		pr_notice("crng init done\n");
-		if (unseeded_warning.missed) {
-			pr_notice("%d get_random_xx warning(s) missed due to ratelimiting\n",
-				  unseeded_warning.missed);
-			unseeded_warning.missed = 0;
-		}
-		if (urandom_warning.missed) {
-			pr_notice("%d urandom warning(s) missed due to ratelimiting\n",
-				  urandom_warning.missed);
-			urandom_warning.missed = 0;
-		}
-	}
 }
 
 /*
@@ -818,7 +799,7 @@ static void extract_entropy(void *buf, s
 
 static void credit_init_bits(size_t nbits)
 {
-	unsigned int init_bits, orig, add;
+	unsigned int new, orig, add;
 	unsigned long flags;
 
 	if (crng_ready() || !nbits)
@@ -828,13 +809,28 @@ static void credit_init_bits(size_t nbit
 
 	do {
 		orig = READ_ONCE(input_pool.init_bits);
-		init_bits = min_t(unsigned int, POOL_BITS, orig + add);
-	} while (cmpxchg(&input_pool.init_bits, orig, init_bits) != orig);
+		new = min_t(unsigned int, POOL_BITS, orig + add);
+	} while (cmpxchg(&input_pool.init_bits, orig, new) != orig);
 
-	if (!crng_ready() && init_bits >= POOL_READY_BITS)
-		crng_reseed();
-	else if (unlikely(crng_init == CRNG_EMPTY && init_bits >= POOL_EARLY_BITS)) {
+	if (orig < POOL_READY_BITS && new >= POOL_READY_BITS) {
+		crng_reseed(); /* Sets crng_init to CRNG_READY under base_crng.lock. */
+		process_random_ready_list();
+		wake_up_interruptible(&crng_init_wait);
+		kill_fasync(&fasync, SIGIO, POLL_IN);
+		pr_notice("crng init done\n");
+		if (unseeded_warning.missed) {
+			pr_notice("%d get_random_xx warning(s) missed due to ratelimiting\n",
+				  unseeded_warning.missed);
+			unseeded_warning.missed = 0;
+		}
+		if (urandom_warning.missed) {
+			pr_notice("%d urandom warning(s) missed due to ratelimiting\n",
+				  urandom_warning.missed);
+			urandom_warning.missed = 0;
+		}
+	} else if (orig < POOL_EARLY_BITS && new >= POOL_EARLY_BITS) {
 		spin_lock_irqsave(&base_crng.lock, flags);
+		/* Check if crng_init is CRNG_EMPTY, to avoid race with crng_reseed(). */
 		if (crng_init == CRNG_EMPTY) {
 			extract_entropy(base_crng.key, sizeof(base_crng.key));
 			crng_init = CRNG_EARLY;
author maintainer
#158 lore raw

PATCH 158/234 random: remove ratelimiting for in-kernel unseeded randomness

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit cc1e127bfa95b5fb2f9307e7168bf8b2b45b4c5e upstream.

The CONFIG_WARN_ALL_UNSEEDED_RANDOM debug option controls whether the kernel warns about all unseeded randomness or just the first instance. There's some complicated rate limiting and comparison to the previous caller, such that even with CONFIG_WARN_ALL_UNSEEDED_RANDOM enabled, developers still don't see all the messages or even an accurate count of how many were missed. This is the result of basically parallel mechanisms aimed at accomplishing more or less the same thing, added at different points in random.c history, which sort of compete with the first-instance-only limiting we have now.

It turns out, however, that nobody cares about the first unseeded randomness instance of in-kernel users. The same first user has been there for ages now, and nobody is doing anything about it. It isn't even clear that anybody _can_ do anything about it. Most places that can do something about it have switched over to using get_random_bytes_wait() or wait_for_random_bytes(), which is the right thing to do, but there is still much code that needs randomness sometimes during init, and as a geeneral rule, if you're not using one of the _wait functions or the readiness notifier callback, you're bound to be doing it wrong just based on that fact alone.

So warning about this same first user that can't easily change is simply not an effective mechanism for anything at all. Users can't do anything about it, as the Kconfig text points out -- the problem isn't in userspace code -- and kernel developers don't or more often can't react to it.

Instead, show the warning for all instances when CONFIG_WARN_ALL_UNSEEDED_RANDOM is set, so that developers can debug things need be, or if it isn't set, don't show a warning at all.

At the same time, CONFIG_WARN_ALL_UNSEEDED_RANDOM now implies setting random.ratelimit_disable=1 on by default, since if you care about one you probably care about the other too. And we can clean up usage around the related urandom_warning ratelimiter as well (whose behavior isn't changing), so that it properly counts missed messages after the 10 message threshold is reached.

---
 drivers/char/random.c |   61 ++++++++++++++------------------------------------
 lib/Kconfig.debug     |    3 --
 2 files changed, 19 insertions(+), 45 deletions(-)
drivers/char/random.c +18 -43
@@ -86,11 +86,10 @@ static DEFINE_SPINLOCK(random_ready_chai
 static RAW_NOTIFIER_HEAD(random_ready_chain);
 
 /* Control how we warn userspace. */
-static struct ratelimit_state unseeded_warning =
-	RATELIMIT_STATE_INIT("warn_unseeded_randomness", HZ, 3);
 static struct ratelimit_state urandom_warning =
 	RATELIMIT_STATE_INIT("warn_urandom_randomness", HZ, 3);
-static int ratelimit_disable __read_mostly;
+static int ratelimit_disable __read_mostly =
+	IS_ENABLED(CONFIG_WARN_ALL_UNSEEDED_RANDOM);
 module_param_named(ratelimit_disable, ratelimit_disable, int, 0644);
 MODULE_PARM_DESC(ratelimit_disable, "Disable random ratelimit suppression");
 
@@ -183,27 +182,15 @@ static void process_random_ready_list(vo
 	spin_unlock_irqrestore(&random_ready_chain_lock, flags);
 }
 
-#define warn_unseeded_randomness(previous) \
-	_warn_unseeded_randomness(__func__, (void *)_RET_IP_, (previous))
+#define warn_unseeded_randomness() \
+	_warn_unseeded_randomness(__func__, (void *)_RET_IP_)
 
-static void _warn_unseeded_randomness(const char *func_name, void *caller, void **previous)
+static void _warn_unseeded_randomness(const char *func_name, void *caller)
 {
-#ifdef CONFIG_WARN_ALL_UNSEEDED_RANDOM
-	const bool print_once = false;
-#else
-	static bool print_once __read_mostly;
-#endif
-
-	if (print_once || crng_ready() ||
-	    (previous && (caller == READ_ONCE(*previous))))
+	if (!IS_ENABLED(CONFIG_WARN_ALL_UNSEEDED_RANDOM) || crng_ready())
 		return;
-	WRITE_ONCE(*previous, caller);
-#ifndef CONFIG_WARN_ALL_UNSEEDED_RANDOM
-	print_once = true;
-#endif
-	if (__ratelimit(&unseeded_warning))
-		printk_deferred(KERN_NOTICE "random: %s called from %pS with crng_init=%d\n",
-				func_name, caller, crng_init);
+	printk_deferred(KERN_NOTICE "random: %s called from %pS with crng_init=%d\n",
+			func_name, caller, crng_init);
 }
 
 
@@ -454,9 +441,7 @@ static void _get_random_bytes(void *buf,
  */
 void get_random_bytes(void *buf, size_t nbytes)
 {
-	static void *previous;
-
-	warn_unseeded_randomness(&previous);
+	warn_unseeded_randomness();
 	_get_random_bytes(buf, nbytes);
 }
 EXPORT_SYMBOL(get_random_bytes);
@@ -550,10 +535,9 @@ u64 get_random_u64(void)
 	u64 ret;
 	unsigned long flags;
 	struct batched_entropy *batch;
-	static void *previous;
 	unsigned long next_gen;
 
-	warn_unseeded_randomness(&previous);
+	warn_unseeded_randomness();
 
 	if  (!crng_ready()) {
 		_get_random_bytes(&ret, sizeof(ret));
@@ -588,10 +572,9 @@ u32 get_random_u32(void)
 	u32 ret;
 	unsigned long flags;
 	struct batched_entropy *batch;
-	static void *previous;
 	unsigned long next_gen;
 
-	warn_unseeded_randomness(&previous);
+	warn_unseeded_randomness();
 
 	if  (!crng_ready()) {
 		_get_random_bytes(&ret, sizeof(ret));
@@ -818,16 +801,9 @@ static void credit_init_bits(size_t nbit
 		wake_up_interruptible(&crng_init_wait);
 		kill_fasync(&fasync, SIGIO, POLL_IN);
 		pr_notice("crng init done\n");
-		if (unseeded_warning.missed) {
-			pr_notice("%d get_random_xx warning(s) missed due to ratelimiting\n",
-				  unseeded_warning.missed);
-			unseeded_warning.missed = 0;
-		}
-		if (urandom_warning.missed) {
+		if (urandom_warning.missed)
 			pr_notice("%d urandom warning(s) missed due to ratelimiting\n",
 				  urandom_warning.missed);
-			urandom_warning.missed = 0;
-		}
 	} else if (orig < POOL_EARLY_BITS && new >= POOL_EARLY_BITS) {
 		spin_lock_irqsave(&base_crng.lock, flags);
 		/* Check if crng_init is CRNG_EMPTY, to avoid race with crng_reseed(). */
@@ -940,10 +916,6 @@ int __init rand_initialize(void)
 	else if (arch_init && trust_cpu)
 		credit_init_bits(BLAKE2S_BLOCK_SIZE * 8);
 
-	if (ratelimit_disable) {
-		urandom_warning.interval = 0;
-		unseeded_warning.interval = 0;
-	}
 	return 0;
 }
 
@@ -1389,11 +1361,14 @@ static ssize_t urandom_read(struct file
 {
 	static int maxwarn = 10;
 
-	if (!crng_ready() && maxwarn > 0) {
-		maxwarn--;
-		if (__ratelimit(&urandom_warning))
+	if (!crng_ready()) {
+		if (!ratelimit_disable && maxwarn <= 0)
+			++urandom_warning.missed;
+		else if (ratelimit_disable || __ratelimit(&urandom_warning)) {
+			--maxwarn;
 			pr_notice("%s: uninitialized urandom read (%zd bytes read)\n",
 				  current->comm, nbytes);
+		}
 	}
 
 	return get_random_bytes_user(buf, nbytes);
lib/Kconfig.debug +1 -2
@@ -1277,8 +1277,7 @@ config WARN_ALL_UNSEEDED_RANDOM
 	  so architecture maintainers really need to do what they can
 	  to get the CRNG seeded sooner after the system is booted.
 	  However, since users cannot do anything actionable to
-	  address this, by default the kernel will issue only a single
-	  warning for the first use of unseeded randomness.
+	  address this, by default this option is disabled.
 
 	  Say Y here if you want to receive warnings for all uses of
 	  unseeded randomness.  This will be of use primarily for
author maintainer
#159 lore raw

PATCH 159/234 random: use proper jiffies comparison macro

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 8a5b8a4a4ceb353b4dd5bafd09e2b15751bcdb51 upstream.

This expands to exactly the same code that it replaces, but makes things consistent by using the same macro for jiffy comparisons throughout.

---
 drivers/char/random.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/char/random.c +1 -1
@@ -324,7 +324,7 @@ static bool crng_has_old_seed(void)
 			interval = max_t(unsigned int, CRNG_RESEED_START_INTERVAL,
 					 (unsigned int)uptime / 2 * HZ);
 	}
-	return time_after(jiffies, READ_ONCE(base_crng.birth) + interval);
+	return time_is_before_jiffies(READ_ONCE(base_crng.birth) + interval);
 }
 
 /*
author maintainer
#160 lore raw

PATCH 160/234 random: handle latent entropy and command line from random_init()

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 2f14062bb14b0fcfcc21e6dc7d5b5c0d25966164 upstream.

Currently, start_kernel() adds latent entropy and the command line to the entropy bool *after* the RNG has been initialized, deferring when it's actually used by things like stack canaries until the next time the pool is seeded. This surely is not intended.

Rather than splitting up which entropy gets added where and when between start_kernel() and random_init(), just do everything in random_init(), which should eliminate these kinds of bugs in the future.

While we're at it, rename the awkwardly titled "rand_initialize()" to the more standard "random_init()" nomenclature.

---
 drivers/char/random.c  |   13 ++++++++-----
 include/linux/random.h |   16 +++++++---------
 init/main.c            |   10 +++-------
 3 files changed, 18 insertions(+), 21 deletions(-)
drivers/char/random.c +8 -5
@@ -883,12 +883,13 @@ early_param("random.trust_bootloader", p
 
 /*
  * The first collection of entropy occurs at system boot while interrupts
- * are still turned off. Here we push in RDSEED, a timestamp, and utsname().
- * Depending on the above configuration knob, RDSEED may be considered
- * sufficient for initialization. Note that much earlier setup may already
- * have pushed entropy into the input pool by the time we get here.
+ * are still turned off. Here we push in latent entropy, RDSEED, a timestamp,
+ * utsname(), and the command line. Depending on the above configuration knob,
+ * RDSEED may be considered sufficient for initialization. Note that much
+ * earlier setup may already have pushed entropy into the input pool by the
+ * time we get here.
  */
-int __init rand_initialize(void)
+int __init random_init(const char *command_line)
 {
 	size_t i;
 	ktime_t now = ktime_get_real();
@@ -910,6 +911,8 @@ int __init rand_initialize(void)
 	}
 	_mix_pool_bytes(&now, sizeof(now));
 	_mix_pool_bytes(utsname(), sizeof(*(utsname())));
+	_mix_pool_bytes(command_line, strlen(command_line));
+	add_latent_entropy();
 
 	if (crng_ready())
 		crng_reseed();
include/linux/random.h +7 -9
@@ -14,26 +14,24 @@ struct notifier_block;
 
 extern void add_device_randomness(const void *, size_t);
 extern void add_bootloader_randomness(const void *, size_t);
+extern void add_input_randomness(unsigned int type, unsigned int code,
+				 unsigned int value) __latent_entropy;
+extern void add_interrupt_randomness(int irq) __latent_entropy;
+extern void add_hwgenerator_randomness(const void *buffer, size_t count,
+				       size_t entropy);
 
 #if defined(LATENT_ENTROPY_PLUGIN) && !defined(__CHECKER__)
 static inline void add_latent_entropy(void)
 {
-	add_device_randomness((const void *)&latent_entropy,
-			      sizeof(latent_entropy));
+	add_device_randomness((const void *)&latent_entropy, sizeof(latent_entropy));
 }
 #else
 static inline void add_latent_entropy(void) {}
 #endif
 
-extern void add_input_randomness(unsigned int type, unsigned int code,
-				 unsigned int value) __latent_entropy;
-extern void add_interrupt_randomness(int irq) __latent_entropy;
-extern void add_hwgenerator_randomness(const void *buffer, size_t count,
-				       size_t entropy);
-
 extern void get_random_bytes(void *buf, size_t nbytes);
 extern int wait_for_random_bytes(void);
-extern int __init rand_initialize(void);
+extern int __init random_init(const char *command_line);
 extern bool rng_is_initialized(void);
 extern int register_random_ready_notifier(struct notifier_block *nb);
 extern int unregister_random_ready_notifier(struct notifier_block *nb);
init/main.c +3 -7
@@ -638,15 +638,11 @@ asmlinkage __visible void __init start_k
 	/*
 	 * For best initial stack canary entropy, prepare it after:
 	 * - setup_arch() for any UEFI RNG entropy and boot cmdline access
-	 * - timekeeping_init() for ktime entropy used in rand_initialize()
+	 * - timekeeping_init() for ktime entropy used in random_init()
 	 * - time_init() for making random_get_entropy() work on some platforms
-	 * - rand_initialize() to get any arch-specific entropy like RDRAND
-	 * - add_latent_entropy() to get any latent entropy
-	 * - adding command line entropy
+	 * - random_init() to initialize the RNG from from early entropy sources
 	 */
-	rand_initialize();
-	add_latent_entropy();
-	add_device_randomness(command_line, strlen(command_line));
+	random_init(command_line);
 	boot_init_stack_canary();
 
 	perf_event_init();
author maintainer
#161 lore raw

PATCH 161/234 random: credit architectural init the exact amount

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 12e45a2a6308105469968951e6d563e8f4fea187 upstream.

RDRAND and RDSEED can fail sometimes, which is fine. We currently initialize the RNG with 512 bits of RDRAND/RDSEED. We only need 256 bits of those to succeed in order to initialize the RNG. Instead of the current "all or nothing" approach, actually credit these contributions the amount that is actually contributed.

---
 drivers/char/random.c |   12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/char/random.c +6 -6
@@ -891,9 +891,8 @@ early_param("random.trust_bootloader", p
  */
 int __init random_init(const char *command_line)
 {
-	size_t i;
 	ktime_t now = ktime_get_real();
-	bool arch_init = true;
+	unsigned int i, arch_bytes;
 	unsigned long rv;
 
 #if defined(LATENT_ENTROPY_PLUGIN)
@@ -901,11 +900,12 @@ int __init random_init(const char *comma
 	_mix_pool_bytes(compiletime_seed, sizeof(compiletime_seed));
 #endif
 
-	for (i = 0; i < BLAKE2S_BLOCK_SIZE; i += sizeof(rv)) {
+	for (i = 0, arch_bytes = BLAKE2S_BLOCK_SIZE;
+	     i < BLAKE2S_BLOCK_SIZE; i += sizeof(rv)) {
 		if (!arch_get_random_seed_long_early(&rv) &&
 		    !arch_get_random_long_early(&rv)) {
 			rv = random_get_entropy();
-			arch_init = false;
+			arch_bytes -= sizeof(rv);
 		}
 		_mix_pool_bytes(&rv, sizeof(rv));
 	}
@@ -916,8 +916,8 @@ int __init random_init(const char *comma
 
 	if (crng_ready())
 		crng_reseed();
-	else if (arch_init && trust_cpu)
-		credit_init_bits(BLAKE2S_BLOCK_SIZE * 8);
+	else if (trust_cpu)
+		credit_init_bits(arch_bytes * 8);
 
 	return 0;
 }
author maintainer
#162 lore raw

PATCH 162/234 random: use static branch for crng_ready()

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit f5bda35fba615ace70a656d4700423fa6c9bebee upstream.

Since crng_ready() is only false briefly during initialization and then forever after becomes true, we don't need to evaluate it after, making it a prime candidate for a static branch.

One complication, however, is that it changes state in a particular call to credit_init_bits(), which might be made from atomic context, which means we must kick off a workqueue to change the static key. Further complicating things, credit_init_bits() may be called sufficiently early on in system initialization such that system_wq is NULL.

Fortunately, there exists the nice function execute_in_process_context(), which will immediately execute the function if !in_interrupt(), and otherwise defer it to a workqueue. During early init, before workqueues are available, in_interrupt() is always false, because interrupts haven't even been enabled yet, which means the function in that case executes immediately. Later on, after workqueues are available, in_interrupt() might be true, but in that case, the work is queued in system_wq and all goes well.

---
 drivers/char/random.c |   16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)
drivers/char/random.c +12 -4
@@ -77,8 +77,9 @@ static enum {
 	CRNG_EMPTY = 0, /* Little to no entropy collected */
 	CRNG_EARLY = 1, /* At least POOL_EARLY_BITS collected */
 	CRNG_READY = 2  /* Fully initialized with POOL_READY_BITS collected */
-} crng_init = CRNG_EMPTY;
-#define crng_ready() (likely(crng_init >= CRNG_READY))
+} crng_init __read_mostly = CRNG_EMPTY;
+static DEFINE_STATIC_KEY_FALSE(crng_is_ready);
+#define crng_ready() (static_branch_likely(&crng_is_ready) || crng_init >= CRNG_READY)
 /* Various types of waiters for crng_init->CRNG_READY transition. */
 static DECLARE_WAIT_QUEUE_HEAD(crng_init_wait);
 static struct fasync_struct *fasync;
@@ -108,6 +109,11 @@ bool rng_is_initialized(void)
 }
 EXPORT_SYMBOL(rng_is_initialized);
 
+static void crng_set_ready(struct work_struct *work)
+{
+	static_branch_enable(&crng_is_ready);
+}
+
 /* Used by wait_for_random_bytes(), and considered an entropy collector, below. */
 static void try_to_generate_entropy(void);
 
@@ -267,7 +273,7 @@ static void crng_reseed(void)
 		++next_gen;
 	WRITE_ONCE(base_crng.generation, next_gen);
 	WRITE_ONCE(base_crng.birth, jiffies);
-	if (!crng_ready())
+	if (!static_branch_likely(&crng_is_ready))
 		crng_init = CRNG_READY;
 	spin_unlock_irqrestore(&base_crng.lock, flags);
 	memzero_explicit(key, sizeof(key));
@@ -782,6 +788,7 @@ static void extract_entropy(void *buf, s
 
 static void credit_init_bits(size_t nbits)
 {
+	static struct execute_work set_ready;
 	unsigned int new, orig, add;
 	unsigned long flags;
 
@@ -797,6 +804,7 @@ static void credit_init_bits(size_t nbit
 
 	if (orig < POOL_READY_BITS && new >= POOL_READY_BITS) {
 		crng_reseed(); /* Sets crng_init to CRNG_READY under base_crng.lock. */
+		execute_in_process_context(crng_set_ready, &set_ready);
 		process_random_ready_list();
 		wake_up_interruptible(&crng_init_wait);
 		kill_fasync(&fasync, SIGIO, POLL_IN);
@@ -1306,7 +1314,7 @@ SYSCALL_DEFINE3(getrandom, char __user *
 	if (count > INT_MAX)
 		count = INT_MAX;
 
-	if (!(flags & GRND_INSECURE) && !crng_ready()) {
+	if (!crng_ready() && !(flags & GRND_INSECURE)) {
 		int ret;
 
 		if (flags & GRND_NONBLOCK)
author maintainer
#163 lore raw

PATCH 163/234 random: remove extern from functions in header

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 7782cfeca7d420e8bb707613d4cfb0f7ff29bb3a upstream.

Accoriding to the kernel style guide, having extern on functions in headers is old school and deprecated, and doesn't add anything. So remove them from random.h, and tidy up the file a little bit too.

---
 include/linux/random.h |   71 +++++++++++++++++++------------------------------
 1 file changed, 28 insertions(+), 43 deletions(-)
include/linux/random.h +28 -43
@@ -12,13 +12,12 @@
 
 struct notifier_block;
 
-extern void add_device_randomness(const void *, size_t);
-extern void add_bootloader_randomness(const void *, size_t);
-extern void add_input_randomness(unsigned int type, unsigned int code,
-				 unsigned int value) __latent_entropy;
-extern void add_interrupt_randomness(int irq) __latent_entropy;
-extern void add_hwgenerator_randomness(const void *buffer, size_t count,
-				       size_t entropy);
+void add_device_randomness(const void *, size_t);
+void add_bootloader_randomness(const void *, size_t);
+void add_input_randomness(unsigned int type, unsigned int code,
+			  unsigned int value) __latent_entropy;
+void add_interrupt_randomness(int irq) __latent_entropy;
+void add_hwgenerator_randomness(const void *buffer, size_t count, size_t entropy);
 
 #if defined(LATENT_ENTROPY_PLUGIN) && !defined(__CHECKER__)
 static inline void add_latent_entropy(void)
@@ -26,21 +25,11 @@ static inline void add_latent_entropy(vo
 	add_device_randomness((const void *)&latent_entropy, sizeof(latent_entropy));
 }
 #else
-static inline void add_latent_entropy(void) {}
-#endif
-
-extern void get_random_bytes(void *buf, size_t nbytes);
-extern int wait_for_random_bytes(void);
-extern int __init random_init(const char *command_line);
-extern bool rng_is_initialized(void);
-extern int register_random_ready_notifier(struct notifier_block *nb);
-extern int unregister_random_ready_notifier(struct notifier_block *nb);
-extern size_t __must_check get_random_bytes_arch(void *buf, size_t nbytes);
-
-#ifndef MODULE
-extern const struct file_operations random_fops, urandom_fops;
+static inline void add_latent_entropy(void) { }
 #endif
 
+void get_random_bytes(void *buf, size_t nbytes);
+size_t __must_check get_random_bytes_arch(void *buf, size_t nbytes);
 u32 get_random_u32(void);
 u64 get_random_u64(void);
 static inline unsigned int get_random_int(void)
@@ -72,11 +61,17 @@ static inline unsigned long get_random_l
 
 static inline unsigned long get_random_canary(void)
 {
-	unsigned long val = get_random_long();
-
-	return val & CANARY_MASK;
+	return get_random_long() & CANARY_MASK;
 }
 
+unsigned long randomize_page(unsigned long start, unsigned long range);
+
+int __init random_init(const char *command_line);
+bool rng_is_initialized(void);
+int wait_for_random_bytes(void);
+int register_random_ready_notifier(struct notifier_block *nb);
+int unregister_random_ready_notifier(struct notifier_block *nb);
+
 /* Calls wait_for_random_bytes() and then calls get_random_bytes(buf, nbytes).
  * Returns the result of the call to wait_for_random_bytes. */
 static inline int get_random_bytes_wait(void *buf, size_t nbytes)
@@ -100,8 +95,6 @@ declare_get_random_var_wait(int)
 declare_get_random_var_wait(long)
 #undef declare_get_random_var
 
-unsigned long randomize_page(unsigned long start, unsigned long range);
-
 /*
  * This is designed to be standalone for just prandom
  * users, but for now we include it from <linux/random.h>
@@ -112,22 +105,10 @@ unsigned long randomize_page(unsigned lo
 #ifdef CONFIG_ARCH_RANDOM
 # include <asm/archrandom.h>
 #else
-static inline bool __must_check arch_get_random_long(unsigned long *v)
-{
-	return false;
-}
-static inline bool __must_check arch_get_random_int(unsigned int *v)
-{
-	return false;
-}
-static inline bool __must_check arch_get_random_seed_long(unsigned long *v)
-{
-	return false;
-}
-static inline bool __must_check arch_get_random_seed_int(unsigned int *v)
-{
-	return false;
-}
+static inline bool __must_check arch_get_random_long(unsigned long *v) { return false; }
+static inline bool __must_check arch_get_random_int(unsigned int *v) { return false; }
+static inline bool __must_check arch_get_random_seed_long(unsigned long *v) { return false; }
+static inline bool __must_check arch_get_random_seed_int(unsigned int *v) { return false; }
 #endif
 
 /*
@@ -151,8 +132,12 @@ static inline bool __init arch_get_rando
 #endif
 
 #ifdef CONFIG_SMP
-extern int random_prepare_cpu(unsigned int cpu);
-extern int random_online_cpu(unsigned int cpu);
+int random_prepare_cpu(unsigned int cpu);
+int random_online_cpu(unsigned int cpu);
+#endif
+
+#ifndef MODULE
+extern const struct file_operations random_fops, urandom_fops;
 #endif
 
 #endif /* _LINUX_RANDOM_H */
author maintainer
#164 lore raw

PATCH 164/234 random: use proper return types on get_random_{int,long}_wait()

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 7c3a8a1db5e03d02cc0abb3357a84b8b326dfac3 upstream.

Before these were returning signed values, but the API is intended to be used with unsigned values.

---
 drivers/char/random.c  |  195 +++++++++++++++++++++++--------------------------
 include/linux/random.h |   24 +++---
 2 files changed, 107 insertions(+), 112 deletions(-)
drivers/char/random.c +95 -100
@@ -210,7 +210,7 @@ static void _warn_unseeded_randomness(co
  *
  * There are a few exported interfaces for use by other drivers:
  *
- *	void get_random_bytes(void *buf, size_t nbytes)
+ *	void get_random_bytes(void *buf, size_t len)
  *	u32 get_random_u32()
  *	u64 get_random_u64()
  *	unsigned int get_random_int()
@@ -249,7 +249,7 @@ static DEFINE_PER_CPU(struct crng, crngs
 };
 
 /* Used by crng_reseed() and crng_make_state() to extract a new seed from the input pool. */
-static void extract_entropy(void *buf, size_t nbytes);
+static void extract_entropy(void *buf, size_t len);
 
 /* This extracts a new crng key from the input pool. */
 static void crng_reseed(void)
@@ -403,24 +403,24 @@ static void crng_make_state(u32 chacha_s
 	local_irq_restore(flags);
 }
 
-static void _get_random_bytes(void *buf, size_t nbytes)
+static void _get_random_bytes(void *buf, size_t len)
 {
 	u32 chacha_state[CHACHA20_BLOCK_SIZE / sizeof(u32)];
 	u8 tmp[CHACHA20_BLOCK_SIZE];
-	size_t len;
+	size_t first_block_len;
 
-	if (!nbytes)
+	if (!len)
 		return;
 
-	len = min_t(size_t, 32, nbytes);
-	crng_make_state(chacha_state, buf, len);
-	nbytes -= len;
-	buf += len;
+	first_block_len = min_t(size_t, 32, len);
+	crng_make_state(chacha_state, buf, first_block_len);
+	len -= first_block_len;
+	buf += first_block_len;
 
-	while (nbytes) {
-		if (nbytes < CHACHA20_BLOCK_SIZE) {
+	while (len) {
+		if (len < CHACHA20_BLOCK_SIZE) {
 			chacha20_block(chacha_state, tmp);
-			memcpy(buf, tmp, nbytes);
+			memcpy(buf, tmp, len);
 			memzero_explicit(tmp, sizeof(tmp));
 			break;
 		}
@@ -428,7 +428,7 @@ static void _get_random_bytes(void *buf,
 		chacha20_block(chacha_state, buf);
 		if (unlikely(chacha_state[12] == 0))
 			++chacha_state[13];
-		nbytes -= CHACHA20_BLOCK_SIZE;
+		len -= CHACHA20_BLOCK_SIZE;
 		buf += CHACHA20_BLOCK_SIZE;
 	}
 
@@ -445,20 +445,20 @@ static void _get_random_bytes(void *buf,
  * wait_for_random_bytes() should be called and return 0 at least once
  * at any point prior.
  */
-void get_random_bytes(void *buf, size_t nbytes)
+void get_random_bytes(void *buf, size_t len)
 {
 	warn_unseeded_randomness();
-	_get_random_bytes(buf, nbytes);
+	_get_random_bytes(buf, len);
 }
 EXPORT_SYMBOL(get_random_bytes);
 
-static ssize_t get_random_bytes_user(void __user *buf, size_t nbytes)
+static ssize_t get_random_bytes_user(void __user *ubuf, size_t len)
 {
-	size_t len, left, ret = 0;
+	size_t block_len, left, ret = 0;
 	u32 chacha_state[CHACHA20_BLOCK_SIZE / sizeof(u32)];
 	u8 output[CHACHA20_BLOCK_SIZE];
 
-	if (!nbytes)
+	if (!len)
 		return 0;
 
 	/*
@@ -472,8 +472,8 @@ static ssize_t get_random_bytes_user(voi
 	 * use chacha_state after, so we can simply return those bytes to
 	 * the user directly.
 	 */
-	if (nbytes <= CHACHA20_KEY_SIZE) {
-		ret = nbytes - copy_to_user(buf, &chacha_state[4], nbytes);
+	if (len <= CHACHA20_KEY_SIZE) {
+		ret = len - copy_to_user(ubuf, &chacha_state[4], len);
 		goto out_zero_chacha;
 	}
 
@@ -482,17 +482,17 @@ static ssize_t get_random_bytes_user(voi
 		if (unlikely(chacha_state[12] == 0))
 			++chacha_state[13];
 
-		len = min_t(size_t, nbytes, CHACHA20_BLOCK_SIZE);
-		left = copy_to_user(buf, output, len);
+		block_len = min_t(size_t, len, CHACHA20_BLOCK_SIZE);
+		left = copy_to_user(ubuf, output, block_len);
 		if (left) {
-			ret += len - left;
+			ret += block_len - left;
 			break;
 		}
 
-		buf += len;
-		ret += len;
-		nbytes -= len;
-		if (!nbytes)
+		ubuf += block_len;
+		ret += block_len;
+		len -= block_len;
+		if (!len)
 			break;
 
 		BUILD_BUG_ON(PAGE_SIZE % CHACHA20_BLOCK_SIZE != 0);
@@ -663,24 +663,24 @@ unsigned long randomize_page(unsigned lo
  * use. Use get_random_bytes() instead. It returns the number of
  * bytes filled in.
  */
-size_t __must_check get_random_bytes_arch(void *buf, size_t nbytes)
+size_t __must_check get_random_bytes_arch(void *buf, size_t len)
 {
-	size_t left = nbytes;
+	size_t left = len;
 	u8 *p = buf;
 
 	while (left) {
 		unsigned long v;
-		size_t chunk = min_t(size_t, left, sizeof(unsigned long));
+		size_t block_len = min_t(size_t, left, sizeof(unsigned long));
 
 		if (!arch_get_random_long(&v))
 			break;
 
-		memcpy(p, &v, chunk);
-		p += chunk;
-		left -= chunk;
+		memcpy(p, &v, block_len);
+		p += block_len;
+		left -= block_len;
 	}
 
-	return nbytes - left;
+	return len - left;
 }
 EXPORT_SYMBOL(get_random_bytes_arch);
 
@@ -691,15 +691,15 @@ EXPORT_SYMBOL(get_random_bytes_arch);
  *
  * Callers may add entropy via:
  *
- *     static void mix_pool_bytes(const void *in, size_t nbytes)
+ *     static void mix_pool_bytes(const void *buf, size_t len)
  *
  * After which, if added entropy should be credited:
  *
- *     static void credit_init_bits(size_t nbits)
+ *     static void credit_init_bits(size_t bits)
  *
  * Finally, extract entropy via:
  *
- *     static void extract_entropy(void *buf, size_t nbytes)
+ *     static void extract_entropy(void *buf, size_t len)
  *
  **********************************************************************/
 
@@ -721,9 +721,9 @@ static struct {
 	.lock = __SPIN_LOCK_UNLOCKED(input_pool.lock),
 };
 
-static void _mix_pool_bytes(const void *in, size_t nbytes)
+static void _mix_pool_bytes(const void *buf, size_t len)
 {
-	blake2s_update(&input_pool.hash, in, nbytes);
+	blake2s_update(&input_pool.hash, buf, len);
 }
 
 /*
@@ -731,12 +731,12 @@ static void _mix_pool_bytes(const void *
  * update the initialization bit counter; the caller should call
  * credit_init_bits if this is appropriate.
  */
-static void mix_pool_bytes(const void *in, size_t nbytes)
+static void mix_pool_bytes(const void *buf, size_t len)
 {
 	unsigned long flags;
 
 	spin_lock_irqsave(&input_pool.lock, flags);
-	_mix_pool_bytes(in, nbytes);
+	_mix_pool_bytes(buf, len);
 	spin_unlock_irqrestore(&input_pool.lock, flags);
 }
 
@@ -744,7 +744,7 @@ static void mix_pool_bytes(const void *i
  * This is an HKDF-like construction for using the hashed collected entropy
  * as a PRF key, that's then expanded block-by-block.
  */
-static void extract_entropy(void *buf, size_t nbytes)
+static void extract_entropy(void *buf, size_t len)
 {
 	unsigned long flags;
 	u8 seed[BLAKE2S_HASH_SIZE], next_key[BLAKE2S_HASH_SIZE];
@@ -773,12 +773,12 @@ static void extract_entropy(void *buf, s
 	spin_unlock_irqrestore(&input_pool.lock, flags);
 	memzero_explicit(next_key, sizeof(next_key));
 
-	while (nbytes) {
-		i = min_t(size_t, nbytes, BLAKE2S_HASH_SIZE);
+	while (len) {
+		i = min_t(size_t, len, BLAKE2S_HASH_SIZE);
 		/* output = HASHPRF(seed, RDSEED || ++counter) */
 		++block.counter;
 		blake2s(buf, (u8 *)&block, seed, i, sizeof(block), sizeof(seed));
-		nbytes -= i;
+		len -= i;
 		buf += i;
 	}
 
@@ -786,16 +786,16 @@ static void extract_entropy(void *buf, s
 	memzero_explicit(&block, sizeof(block));
 }
 
-static void credit_init_bits(size_t nbits)
+static void credit_init_bits(size_t bits)
 {
 	static struct execute_work set_ready;
 	unsigned int new, orig, add;
 	unsigned long flags;
 
-	if (crng_ready() || !nbits)
+	if (crng_ready() || !bits)
 		return;
 
-	add = min_t(size_t, nbits, POOL_BITS);
+	add = min_t(size_t, bits, POOL_BITS);
 
 	do {
 		orig = READ_ONCE(input_pool.init_bits);
@@ -831,13 +831,11 @@ static void credit_init_bits(size_t nbit
  * The following exported functions are used for pushing entropy into
  * the above entropy accumulation routines:
  *
- *	void add_device_randomness(const void *buf, size_t size);
- *	void add_hwgenerator_randomness(const void *buffer, size_t count,
- *					size_t entropy);
- *	void add_bootloader_randomness(const void *buf, size_t size);
+ *	void add_device_randomness(const void *buf, size_t len);
+ *	void add_hwgenerator_randomness(const void *buf, size_t len, size_t entropy);
+ *	void add_bootloader_randomness(const void *buf, size_t len);
  *	void add_interrupt_randomness(int irq);
- *	void add_input_randomness(unsigned int type, unsigned int code,
- *	                          unsigned int value);
+ *	void add_input_randomness(unsigned int type, unsigned int code, unsigned int value);
  *	void add_disk_randomness(struct gendisk *disk);
  *
  * add_device_randomness() adds data to the input pool that
@@ -901,7 +899,7 @@ int __init random_init(const char *comma
 {
 	ktime_t now = ktime_get_real();
 	unsigned int i, arch_bytes;
-	unsigned long rv;
+	unsigned long entropy;
 
 #if defined(LATENT_ENTROPY_PLUGIN)
 	static const u8 compiletime_seed[BLAKE2S_BLOCK_SIZE] __initconst __latent_entropy;
@@ -909,13 +907,13 @@ int __init random_init(const char *comma
 #endif
 
 	for (i = 0, arch_bytes = BLAKE2S_BLOCK_SIZE;
-	     i < BLAKE2S_BLOCK_SIZE; i += sizeof(rv)) {
-		if (!arch_get_random_seed_long_early(&rv) &&
-		    !arch_get_random_long_early(&rv)) {
-			rv = random_get_entropy();
-			arch_bytes -= sizeof(rv);
+	     i < BLAKE2S_BLOCK_SIZE; i += sizeof(entropy)) {
+		if (!arch_get_random_seed_long_early(&entropy) &&
+		    !arch_get_random_long_early(&entropy)) {
+			entropy = random_get_entropy();
+			arch_bytes -= sizeof(entropy);
 		}
-		_mix_pool_bytes(&rv, sizeof(rv));
+		_mix_pool_bytes(&entropy, sizeof(entropy));
 	}
 	_mix_pool_bytes(&now, sizeof(now));
 	_mix_pool_bytes(utsname(), sizeof(*(utsname())));
@@ -938,14 +936,14 @@ int __init random_init(const char *comma
  * the entropy pool having similar initial state across largely
  * identical devices.
  */
-void add_device_randomness(const void *buf, size_t size)
+void add_device_randomness(const void *buf, size_t len)
 {
 	unsigned long entropy = random_get_entropy();
 	unsigned long flags;
 
 	spin_lock_irqsave(&input_pool.lock, flags);
 	_mix_pool_bytes(&entropy, sizeof(entropy));
-	_mix_pool_bytes(buf, size);
+	_mix_pool_bytes(buf, len);
 	spin_unlock_irqrestore(&input_pool.lock, flags);
 }
 EXPORT_SYMBOL(add_device_randomness);
@@ -955,10 +953,9 @@ EXPORT_SYMBOL(add_device_randomness);
  * Those devices may produce endless random bits and will be throttled
  * when our pool is full.
  */
-void add_hwgenerator_randomness(const void *buffer, size_t count,
-				size_t entropy)
+void add_hwgenerator_randomness(const void *buf, size_t len, size_t entropy)
 {
-	mix_pool_bytes(buffer, count);
+	mix_pool_bytes(buf, len);
 	credit_init_bits(entropy);
 
 	/*
@@ -974,11 +971,11 @@ EXPORT_SYMBOL_GPL(add_hwgenerator_random
  * Handle random seed passed by bootloader, and credit it if
  * CONFIG_RANDOM_TRUST_BOOTLOADER is set.
  */
-void add_bootloader_randomness(const void *buf, size_t size)
+void add_bootloader_randomness(const void *buf, size_t len)
 {
-	mix_pool_bytes(buf, size);
+	mix_pool_bytes(buf, len);
 	if (trust_bootloader)
-		credit_init_bits(size * 8);
+		credit_init_bits(len * 8);
 }
 EXPORT_SYMBOL_GPL(add_bootloader_randomness);
 
@@ -1178,8 +1175,7 @@ static void add_timer_randomness(struct
 		credit_init_bits(bits);
 }
 
-void add_input_randomness(unsigned int type, unsigned int code,
-			  unsigned int value)
+void add_input_randomness(unsigned int type, unsigned int code, unsigned int value)
 {
 	static unsigned char last_value;
 	static struct timer_rand_state input_timer_state = { INITIAL_JIFFIES };
@@ -1298,8 +1294,7 @@ static void try_to_generate_entropy(void
  *
  **********************************************************************/
 
-SYSCALL_DEFINE3(getrandom, char __user *, buf, size_t, count, unsigned int,
-		flags)
+SYSCALL_DEFINE3(getrandom, char __user *, ubuf, size_t, len, unsigned int, flags)
 {
 	if (flags & ~(GRND_NONBLOCK | GRND_RANDOM | GRND_INSECURE))
 		return -EINVAL;
@@ -1311,8 +1306,8 @@ SYSCALL_DEFINE3(getrandom, char __user *
 	if ((flags & (GRND_INSECURE | GRND_RANDOM)) == (GRND_INSECURE | GRND_RANDOM))
 		return -EINVAL;
 
-	if (count > INT_MAX)
-		count = INT_MAX;
+	if (len > INT_MAX)
+		len = INT_MAX;
 
 	if (!crng_ready() && !(flags & GRND_INSECURE)) {
 		int ret;
@@ -1323,7 +1318,7 @@ SYSCALL_DEFINE3(getrandom, char __user *
 		if (unlikely(ret))
 			return ret;
 	}
-	return get_random_bytes_user(buf, count);
+	return get_random_bytes_user(ubuf, len);
 }
 
 static __poll_t random_poll(struct file *file, poll_table *wait)
@@ -1332,21 +1327,21 @@ static __poll_t random_poll(struct file
 	return crng_ready() ? EPOLLIN | EPOLLRDNORM : EPOLLOUT | EPOLLWRNORM;
 }
 
-static int write_pool(const char __user *ubuf, size_t count)
+static int write_pool(const char __user *ubuf, size_t len)
 {
-	size_t len;
+	size_t block_len;
 	int ret = 0;
 	u8 block[BLAKE2S_BLOCK_SIZE];
 
-	while (count) {
-		len = min(count, sizeof(block));
-		if (copy_from_user(block, ubuf, len)) {
+	while (len) {
+		block_len = min(len, sizeof(block));
+		if (copy_from_user(block, ubuf, block_len)) {
 			ret = -EFAULT;
 			goto out;
 		}
-		count -= len;
-		ubuf += len;
-		mix_pool_bytes(block, len);
+		len -= block_len;
+		ubuf += block_len;
+		mix_pool_bytes(block, block_len);
 		cond_resched();
 	}
 
@@ -1355,20 +1350,20 @@ out:
 	return ret;
 }
 
-static ssize_t random_write(struct file *file, const char __user *buffer,
-			    size_t count, loff_t *ppos)
+static ssize_t random_write(struct file *file, const char __user *ubuf,
+			    size_t len, loff_t *ppos)
 {
 	int ret;
 
-	ret = write_pool(buffer, count);
+	ret = write_pool(ubuf, len);
 	if (ret)
 		return ret;
 
-	return (ssize_t)count;
+	return (ssize_t)len;
 }
 
-static ssize_t urandom_read(struct file *file, char __user *buf, size_t nbytes,
-			    loff_t *ppos)
+static ssize_t urandom_read(struct file *file, char __user *ubuf,
+			    size_t len, loff_t *ppos)
 {
 	static int maxwarn = 10;
 
@@ -1378,22 +1373,22 @@ static ssize_t urandom_read(struct file
 		else if (ratelimit_disable || __ratelimit(&urandom_warning)) {
 			--maxwarn;
 			pr_notice("%s: uninitialized urandom read (%zd bytes read)\n",
-				  current->comm, nbytes);
+				  current->comm, len);
 		}
 	}
 
-	return get_random_bytes_user(buf, nbytes);
+	return get_random_bytes_user(ubuf, len);
 }
 
-static ssize_t random_read(struct file *file, char __user *buf, size_t nbytes,
-			   loff_t *ppos)
+static ssize_t random_read(struct file *file, char __user *ubuf,
+			   size_t len, loff_t *ppos)
 {
 	int ret;
 
 	ret = wait_for_random_bytes();
 	if (ret != 0)
 		return ret;
-	return get_random_bytes_user(buf, nbytes);
+	return get_random_bytes_user(ubuf, len);
 }
 
 static long random_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
@@ -1516,8 +1511,8 @@ static u8 sysctl_bootid[UUID_SIZE];
  * UUID. The difference is in whether table->data is NULL; if it is,
  * then a new UUID is generated and returned to the user.
  */
-static int proc_do_uuid(struct ctl_table *table, int write,
-			void __user *buffer, size_t *lenp, loff_t *ppos)
+static int proc_do_uuid(struct ctl_table *table, int write, void __user *buf,
+			size_t *lenp, loff_t *ppos)
 {
 	u8 tmp_uuid[UUID_SIZE], *uuid;
 	char uuid_string[UUID_STRING_LEN + 1];
@@ -1543,14 +1538,14 @@ static int proc_do_uuid(struct ctl_table
 	}
 
 	snprintf(uuid_string, sizeof(uuid_string), "%pU", uuid);
-	return proc_dostring(&fake_table, 0, buffer, lenp, ppos);
+	return proc_dostring(&fake_table, 0, buf, lenp, ppos);
 }
 
 /* The same as proc_dointvec, but writes don't change anything. */
-static int proc_do_rointvec(struct ctl_table *table, int write, void __user *buffer,
+static int proc_do_rointvec(struct ctl_table *table, int write, void __user *buf,
 			    size_t *lenp, loff_t *ppos)
 {
-	return write ? 0 : proc_dointvec(table, 0, buffer, lenp, ppos);
+	return write ? 0 : proc_dointvec(table, 0, buf, lenp, ppos);
 }
 
 extern struct ctl_table random_table[];
include/linux/random.h +12 -12
@@ -12,12 +12,12 @@
 
 struct notifier_block;
 
-void add_device_randomness(const void *, size_t);
-void add_bootloader_randomness(const void *, size_t);
+void add_device_randomness(const void *buf, size_t len);
+void add_bootloader_randomness(const void *buf, size_t len);
 void add_input_randomness(unsigned int type, unsigned int code,
 			  unsigned int value) __latent_entropy;
 void add_interrupt_randomness(int irq) __latent_entropy;
-void add_hwgenerator_randomness(const void *buffer, size_t count, size_t entropy);
+void add_hwgenerator_randomness(const void *buf, size_t len, size_t entropy);
 
 #if defined(LATENT_ENTROPY_PLUGIN) && !defined(__CHECKER__)
 static inline void add_latent_entropy(void)
@@ -28,8 +28,8 @@ static inline void add_latent_entropy(vo
 static inline void add_latent_entropy(void) { }
 #endif
 
-void get_random_bytes(void *buf, size_t nbytes);
-size_t __must_check get_random_bytes_arch(void *buf, size_t nbytes);
+void get_random_bytes(void *buf, size_t len);
+size_t __must_check get_random_bytes_arch(void *buf, size_t len);
 u32 get_random_u32(void);
 u64 get_random_u64(void);
 static inline unsigned int get_random_int(void)
@@ -81,18 +81,18 @@ static inline int get_random_bytes_wait(
 	return ret;
 }
 
-#define declare_get_random_var_wait(var) \
-	static inline int get_random_ ## var ## _wait(var *out) { \
+#define declare_get_random_var_wait(name, ret_type) \
+	static inline int get_random_ ## name ## _wait(ret_type *out) { \
 		int ret = wait_for_random_bytes(); \
 		if (unlikely(ret)) \
 			return ret; \
-		*out = get_random_ ## var(); \
+		*out = get_random_ ## name(); \
 		return 0; \
 	}
-declare_get_random_var_wait(u32)
-declare_get_random_var_wait(u64)
-declare_get_random_var_wait(int)
-declare_get_random_var_wait(long)
+declare_get_random_var_wait(u32, u32)
+declare_get_random_var_wait(u64, u32)
+declare_get_random_var_wait(int, unsigned int)
+declare_get_random_var_wait(long, unsigned long)
 #undef declare_get_random_var
 
 /*
author maintainer
#165 lore raw

PATCH 165/234 random: move initialization functions out of hot pages

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 560181c27b582557d633ecb608110075433383af upstream.

Much of random.c is devoted to initializing the rng and accounting for when a sufficient amount of entropy has been added. In a perfect world, this would all happen during init, and so we could mark these functions as __init. But in reality, this isn't the case: sometimes the rng only finishes initializing some seconds after system init is finished.

For this reason, at the moment, a whole host of functions that are only used relatively close to system init and then never again are intermixed with functions that are used in hot code all the time. This creates more cache misses than necessary.

In order to pack the hot code closer together, this commit moves the initialization functions that can't be marked as __init into .text.unlikely by way of the __cold attribute.

Of particular note is moving credit_init_bits() into a macro wrapper that inlines the crng_ready() static branch check. This avoids a function call to a nop+ret, and most notably prevents extra entropy arithmetic from being computed in mix_interrupt_randomness().

[ Jason: for stable, made sure the printk_deferred was a pr_notice,
because those caused problems on ≤ 4.19 according to commit logs. ]

---
 drivers/char/random.c |   40 ++++++++++++++++++----------------------
 1 file changed, 18 insertions(+), 22 deletions(-)
drivers/char/random.c +18 -22
@@ -109,7 +109,7 @@ bool rng_is_initialized(void)
 }
 EXPORT_SYMBOL(rng_is_initialized);
 
-static void crng_set_ready(struct work_struct *work)
+static void __cold crng_set_ready(struct work_struct *work)
 {
 	static_branch_enable(&crng_is_ready);
 }
@@ -148,7 +148,7 @@ EXPORT_SYMBOL(wait_for_random_bytes);
  * returns: 0 if callback is successfully added
  *	    -EALREADY if pool is already initialised (callback not called)
  */
-int register_random_ready_notifier(struct notifier_block *nb)
+int __cold register_random_ready_notifier(struct notifier_block *nb)
 {
 	unsigned long flags;
 	int ret = -EALREADY;
@@ -167,7 +167,7 @@ EXPORT_SYMBOL(register_random_ready_noti
 /*
  * Delete a previously registered readiness callback function.
  */
-int unregister_random_ready_notifier(struct notifier_block *nb)
+int __cold unregister_random_ready_notifier(struct notifier_block *nb)
 {
 	unsigned long flags;
 	int ret;
@@ -179,7 +179,7 @@ int unregister_random_ready_notifier(str
 }
 EXPORT_SYMBOL(unregister_random_ready_notifier);
 
-static void process_random_ready_list(void)
+static void __cold process_random_ready_list(void)
 {
 	unsigned long flags;
 
@@ -189,15 +189,9 @@ static void process_random_ready_list(vo
 }
 
 #define warn_unseeded_randomness() \
-	_warn_unseeded_randomness(__func__, (void *)_RET_IP_)
-
-static void _warn_unseeded_randomness(const char *func_name, void *caller)
-{
-	if (!IS_ENABLED(CONFIG_WARN_ALL_UNSEEDED_RANDOM) || crng_ready())
-		return;
-	printk_deferred(KERN_NOTICE "random: %s called from %pS with crng_init=%d\n",
-			func_name, caller, crng_init);
-}
+	if (IS_ENABLED(CONFIG_WARN_ALL_UNSEEDED_RANDOM) && !crng_ready()) \
+		pr_notice("%s called from %pS with crng_init=%d\n", \
+			  __func__, (void *)_RET_IP_, crng_init)
 
 
 /*********************************************************************
@@ -611,7 +605,7 @@ EXPORT_SYMBOL(get_random_u32);
  * This function is called when the CPU is coming up, with entry
  * CPUHP_RANDOM_PREPARE, which comes before CPUHP_WORKQUEUE_PREP.
  */
-int random_prepare_cpu(unsigned int cpu)
+int __cold random_prepare_cpu(unsigned int cpu)
 {
 	/*
 	 * When the cpu comes back online, immediately invalidate both
@@ -786,13 +780,15 @@ static void extract_entropy(void *buf, s
 	memzero_explicit(&block, sizeof(block));
 }
 
-static void credit_init_bits(size_t bits)
+#define credit_init_bits(bits) if (!crng_ready()) _credit_init_bits(bits)
+
+static void __cold _credit_init_bits(size_t bits)
 {
 	static struct execute_work set_ready;
 	unsigned int new, orig, add;
 	unsigned long flags;
 
-	if (crng_ready() || !bits)
+	if (!bits)
 		return;
 
 	add = min_t(size_t, bits, POOL_BITS);
@@ -971,7 +967,7 @@ EXPORT_SYMBOL_GPL(add_hwgenerator_random
  * Handle random seed passed by bootloader, and credit it if
  * CONFIG_RANDOM_TRUST_BOOTLOADER is set.
  */
-void add_bootloader_randomness(const void *buf, size_t len)
+void __cold add_bootloader_randomness(const void *buf, size_t len)
 {
 	mix_pool_bytes(buf, len);
 	if (trust_bootloader)
@@ -1017,7 +1013,7 @@ static void fast_mix(unsigned long s[4],
  * This function is called when the CPU has just come online, with
  * entry CPUHP_AP_RANDOM_ONLINE, just after CPUHP_AP_WORKQUEUE_ONLINE.
  */
-int random_online_cpu(unsigned int cpu)
+int __cold random_online_cpu(unsigned int cpu)
 {
 	/*
 	 * During CPU shutdown and before CPU onlining, add_interrupt_
@@ -1172,7 +1168,7 @@ static void add_timer_randomness(struct
 	if (in_irq())
 		this_cpu_ptr(&irq_randomness)->count += max(1u, bits * 64) - 1;
 	else
-		credit_init_bits(bits);
+		_credit_init_bits(bits);
 }
 
 void add_input_randomness(unsigned int type, unsigned int code, unsigned int value)
@@ -1200,7 +1196,7 @@ void add_disk_randomness(struct gendisk
 }
 EXPORT_SYMBOL_GPL(add_disk_randomness);
 
-void rand_initialize_disk(struct gendisk *disk)
+void __cold rand_initialize_disk(struct gendisk *disk)
 {
 	struct timer_rand_state *state;
 
@@ -1229,7 +1225,7 @@ void rand_initialize_disk(struct gendisk
  *
  * So the re-arming always happens in the entropy loop itself.
  */
-static void entropy_timer(struct timer_list *t)
+static void __cold entropy_timer(struct timer_list *t)
 {
 	credit_init_bits(1);
 }
@@ -1238,7 +1234,7 @@ static void entropy_timer(struct timer_l
  * If we have an actual cycle counter, see if we can
  * generate enough entropy with timing noise
  */
-static void try_to_generate_entropy(void)
+static void __cold try_to_generate_entropy(void)
 {
 	struct {
 		unsigned long entropy;
author maintainer
#166 lore raw

PATCH 166/234 random: move randomize_page() into mm where it belongs

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 5ad7dd882e45d7fe432c32e896e2aaa0b21746ea upstream.

randomize_page is an mm function. It is documented like one. It contains the history of one. It has the naming convention of one. It looks just like another very similar function in mm, randomize_stack_top(). And it has always been maintained and updated by mm people. There is no need for it to be in random.c. In the "which shape does not look like the other ones" test, pointing to randomize_page() is correct.

So move randomize_page() into mm/util.c, right next to the similar randomize_stack_top() function.

This commit contains no actual code changes.

---
 drivers/char/random.c  |  238 ++++++++++++++++---------------------------------
 include/linux/mm.h     |    2 
 include/linux/random.h |    2 
 mm/util.c              |   33 ++++++
 4 files changed, 117 insertions(+), 158 deletions(-)
drivers/char/random.c +82 -156
@@ -52,6 +52,7 @@
 #include <linux/uuid.h>
 #include <linux/uaccess.h>
 #include <linux/siphash.h>
+#include <linux/uio.h>
 #include <crypto/chacha20.h>
 #include <crypto/blake2s.h>
 #include <asm/processor.h>
@@ -446,13 +447,13 @@ void get_random_bytes(void *buf, size_t
 }
 EXPORT_SYMBOL(get_random_bytes);
 
-static ssize_t get_random_bytes_user(void __user *ubuf, size_t len)
+static ssize_t get_random_bytes_user(struct iov_iter *iter)
 {
-	size_t block_len, left, ret = 0;
 	u32 chacha_state[CHACHA20_BLOCK_SIZE / sizeof(u32)];
-	u8 output[CHACHA20_BLOCK_SIZE];
+	u8 block[CHACHA20_BLOCK_SIZE];
+	size_t ret = 0, copied;
 
-	if (!len)
+	if (unlikely(!iov_iter_count(iter)))
 		return 0;
 
 	/*
@@ -466,30 +467,22 @@ static ssize_t get_random_bytes_user(voi
 	 * use chacha_state after, so we can simply return those bytes to
 	 * the user directly.
 	 */
-	if (len <= CHACHA20_KEY_SIZE) {
-		ret = len - copy_to_user(ubuf, &chacha_state[4], len);
+	if (iov_iter_count(iter) <= CHACHA20_KEY_SIZE) {
+		ret = copy_to_iter(&chacha_state[4], CHACHA20_KEY_SIZE, iter);
 		goto out_zero_chacha;
 	}
 
 	for (;;) {
-		chacha20_block(chacha_state, output);
+		chacha20_block(chacha_state, block);
 		if (unlikely(chacha_state[12] == 0))
 			++chacha_state[13];
 
-		block_len = min_t(size_t, len, CHACHA20_BLOCK_SIZE);
-		left = copy_to_user(ubuf, output, block_len);
-		if (left) {
-			ret += block_len - left;
+		copied = copy_to_iter(block, sizeof(block), iter);
+		ret += copied;
+		if (!iov_iter_count(iter) || copied != sizeof(block))
 			break;
-		}
 
-		ubuf += block_len;
-		ret += block_len;
-		len -= block_len;
-		if (!len)
-			break;
-
-		BUILD_BUG_ON(PAGE_SIZE % CHACHA20_BLOCK_SIZE != 0);
+		BUILD_BUG_ON(PAGE_SIZE % sizeof(block) != 0);
 		if (ret % PAGE_SIZE == 0) {
 			if (signal_pending(current))
 				break;
@@ -497,7 +490,7 @@ static ssize_t get_random_bytes_user(voi
 		}
 	}
 
-	memzero_explicit(output, sizeof(output));
+	memzero_explicit(block, sizeof(block));
 out_zero_chacha:
 	memzero_explicit(chacha_state, sizeof(chacha_state));
 	return ret ? ret : -EFAULT;
@@ -509,96 +502,60 @@ out_zero_chacha:
  * provided by this function is okay, the function wait_for_random_bytes()
  * should be called and return 0 at least once at any point prior.
  */
-struct batched_entropy {
-	union {
-		/*
-		 * We make this 1.5x a ChaCha block, so that we get the
-		 * remaining 32 bytes from fast key erasure, plus one full
-		 * block from the detached ChaCha state. We can increase
-		 * the size of this later if needed so long as we keep the
-		 * formula of (integer_blocks + 0.5) * CHACHA20_BLOCK_SIZE.
-		 */
-		u64 entropy_u64[CHACHA20_BLOCK_SIZE * 3 / (2 * sizeof(u64))];
-		u32 entropy_u32[CHACHA20_BLOCK_SIZE * 3 / (2 * sizeof(u32))];
-	};
-	unsigned long generation;
-	unsigned int position;
-};
-
-
-static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u64) = {
-	.position = UINT_MAX
-};
-
-u64 get_random_u64(void)
-{
-	u64 ret;
-	unsigned long flags;
-	struct batched_entropy *batch;
-	unsigned long next_gen;
-
-	warn_unseeded_randomness();
-
-	if  (!crng_ready()) {
-		_get_random_bytes(&ret, sizeof(ret));
-		return ret;
-	}
-
-	local_irq_save(flags);
-	batch = raw_cpu_ptr(&batched_entropy_u64);
-
-	next_gen = READ_ONCE(base_crng.generation);
-	if (batch->position >= ARRAY_SIZE(batch->entropy_u64) ||
-	    next_gen != batch->generation) {
-		_get_random_bytes(batch->entropy_u64, sizeof(batch->entropy_u64));
-		batch->position = 0;
-		batch->generation = next_gen;
-	}
-
-	ret = batch->entropy_u64[batch->position];
-	batch->entropy_u64[batch->position] = 0;
-	++batch->position;
-	local_irq_restore(flags);
-	return ret;
-}
-EXPORT_SYMBOL(get_random_u64);
-
-static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u32) = {
-	.position = UINT_MAX
-};
-
-u32 get_random_u32(void)
-{
-	u32 ret;
-	unsigned long flags;
-	struct batched_entropy *batch;
-	unsigned long next_gen;
 
-	warn_unseeded_randomness();
+#define DEFINE_BATCHED_ENTROPY(type)						\
+struct batch_ ##type {								\
+	/*									\
+	 * We make this 1.5x a ChaCha block, so that we get the			\
+	 * remaining 32 bytes from fast key erasure, plus one full		\
+	 * block from the detached ChaCha state. We can increase		\
+	 * the size of this later if needed so long as we keep the		\
+	 * formula of (integer_blocks + 0.5) * CHACHA20_BLOCK_SIZE.		\
+	 */									\
+	type entropy[CHACHA20_BLOCK_SIZE * 3 / (2 * sizeof(type))];		\
+	unsigned long generation;						\
+	unsigned int position;							\
+};										\
+										\
+static DEFINE_PER_CPU(struct batch_ ##type, batched_entropy_ ##type) = {	\
+	.position = UINT_MAX							\
+};										\
+										\
+type get_random_ ##type(void)							\
+{										\
+	type ret;								\
+	unsigned long flags;							\
+	struct batch_ ##type *batch;						\
+	unsigned long next_gen;							\
+										\
+	warn_unseeded_randomness();						\
+										\
+	if  (!crng_ready()) {							\
+		_get_random_bytes(&ret, sizeof(ret));				\
+		return ret;							\
+	}									\
+										\
+	local_irq_save(flags);		\
+	batch = raw_cpu_ptr(&batched_entropy_##type);				\
+										\
+	next_gen = READ_ONCE(base_crng.generation);				\
+	if (batch->position >= ARRAY_SIZE(batch->entropy) ||			\
+	    next_gen != batch->generation) {					\
+		_get_random_bytes(batch->entropy, sizeof(batch->entropy));	\
+		batch->position = 0;						\
+		batch->generation = next_gen;					\
+	}									\
+										\
+	ret = batch->entropy[batch->position];					\
+	batch->entropy[batch->position] = 0;					\
+	++batch->position;							\
+	local_irq_restore(flags);		\
+	return ret;								\
+}										\
+EXPORT_SYMBOL(get_random_ ##type);
 
-	if  (!crng_ready()) {
-		_get_random_bytes(&ret, sizeof(ret));
-		return ret;
-	}
-
-	local_irq_save(flags);
-	batch = raw_cpu_ptr(&batched_entropy_u32);
-
-	next_gen = READ_ONCE(base_crng.generation);
-	if (batch->position >= ARRAY_SIZE(batch->entropy_u32) ||
-	    next_gen != batch->generation) {
-		_get_random_bytes(batch->entropy_u32, sizeof(batch->entropy_u32));
-		batch->position = 0;
-		batch->generation = next_gen;
-	}
-
-	ret = batch->entropy_u32[batch->position];
-	batch->entropy_u32[batch->position] = 0;
-	++batch->position;
-	local_irq_restore(flags);
-	return ret;
-}
-EXPORT_SYMBOL(get_random_u32);
+DEFINE_BATCHED_ENTROPY(u64)
+DEFINE_BATCHED_ENTROPY(u32)
 
 #ifdef CONFIG_SMP
 /*
@@ -619,38 +576,6 @@ int __cold random_prepare_cpu(unsigned i
 }
 #endif
 
-/**
- * randomize_page - Generate a random, page aligned address
- * @start:	The smallest acceptable address the caller will take.
- * @range:	The size of the area, starting at @start, within which the
- *		random address must fall.
- *
- * If @start + @range would overflow, @range is capped.
- *
- * NOTE: Historical use of randomize_range, which this replaces, presumed that
- * @start was already page aligned.  We now align it regardless.
- *
- * Return: A page aligned address within [start, start + range).  On error,
- * @start is returned.
- */
-unsigned long randomize_page(unsigned long start, unsigned long range)
-{
-	if (!PAGE_ALIGNED(start)) {
-		range -= PAGE_ALIGN(start) - start;
-		start = PAGE_ALIGN(start);
-	}
-
-	if (start > ULONG_MAX - range)
-		range = ULONG_MAX - start;
-
-	range >>= PAGE_SHIFT;
-
-	if (range == 0)
-		return start;
-
-	return start + (get_random_long() % range << PAGE_SHIFT);
-}
-
 /*
  * This function will use the architecture-specific hardware random
  * number generator if it is available. It is not recommended for
@@ -1292,6 +1217,10 @@ static void __cold try_to_generate_entro
 
 SYSCALL_DEFINE3(getrandom, char __user *, ubuf, size_t, len, unsigned int, flags)
 {
+	struct iov_iter iter;
+	struct iovec iov;
+	int ret;
+
 	if (flags & ~(GRND_NONBLOCK | GRND_RANDOM | GRND_INSECURE))
 		return -EINVAL;
 
@@ -1302,19 +1231,18 @@ SYSCALL_DEFINE3(getrandom, char __user *
 	if ((flags & (GRND_INSECURE | GRND_RANDOM)) == (GRND_INSECURE | GRND_RANDOM))
 		return -EINVAL;
 
-	if (len > INT_MAX)
-		len = INT_MAX;
-
 	if (!crng_ready() && !(flags & GRND_INSECURE)) {
-		int ret;
-
 		if (flags & GRND_NONBLOCK)
 			return -EAGAIN;
 		ret = wait_for_random_bytes();
 		if (unlikely(ret))
 			return ret;
 	}
-	return get_random_bytes_user(ubuf, len);
+
+	ret = import_single_range(READ, ubuf, len, &iov, &iter);
+	if (unlikely(ret))
+		return ret;
+	return get_random_bytes_user(&iter);
 }
 
 static __poll_t random_poll(struct file *file, poll_table *wait)
@@ -1358,8 +1286,7 @@ static ssize_t random_write(struct file
 	return (ssize_t)len;
 }
 
-static ssize_t urandom_read(struct file *file, char __user *ubuf,
-			    size_t len, loff_t *ppos)
+static ssize_t urandom_read_iter(struct kiocb *kiocb, struct iov_iter *iter)
 {
 	static int maxwarn = 10;
 
@@ -1368,23 +1295,22 @@ static ssize_t urandom_read(struct file
 			++urandom_warning.missed;
 		else if (ratelimit_disable || __ratelimit(&urandom_warning)) {
 			--maxwarn;
-			pr_notice("%s: uninitialized urandom read (%zd bytes read)\n",
-				  current->comm, len);
+			pr_notice("%s: uninitialized urandom read (%zu bytes read)\n",
+				  current->comm, iov_iter_count(iter));
 		}
 	}
 
-	return get_random_bytes_user(ubuf, len);
+	return get_random_bytes_user(iter);
 }
 
-static ssize_t random_read(struct file *file, char __user *ubuf,
-			   size_t len, loff_t *ppos)
+static ssize_t random_read_iter(struct kiocb *kiocb, struct iov_iter *iter)
 {
 	int ret;
 
 	ret = wait_for_random_bytes();
 	if (ret != 0)
 		return ret;
-	return get_random_bytes_user(ubuf, len);
+	return get_random_bytes_user(iter);
 }
 
 static long random_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
@@ -1446,7 +1372,7 @@ static int random_fasync(int fd, struct
 }
 
 const struct file_operations random_fops = {
-	.read = random_read,
+	.read_iter = random_read_iter,
 	.write = random_write,
 	.poll = random_poll,
 	.unlocked_ioctl = random_ioctl,
@@ -1455,7 +1381,7 @@ const struct file_operations random_fops
 };
 
 const struct file_operations urandom_fops = {
-	.read = urandom_read,
+	.read_iter = urandom_read_iter,
 	.write = random_write,
 	.unlocked_ioctl = random_ioctl,
 	.fasync = random_fasync,
include/linux/mm.h +2 -0
@@ -2308,6 +2308,8 @@ extern int install_special_mapping(struc
 				   unsigned long addr, unsigned long len,
 				   unsigned long flags, struct page **pages);
 
+unsigned long randomize_page(unsigned long start, unsigned long range);
+
 extern unsigned long get_unmapped_area(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
 
 extern unsigned long mmap_region(struct file *file, unsigned long addr,
include/linux/random.h +0 -2
@@ -64,8 +64,6 @@ static inline unsigned long get_random_c
 	return get_random_long() & CANARY_MASK;
 }
 
-unsigned long randomize_page(unsigned long start, unsigned long range);
-
 int __init random_init(const char *command_line);
 bool rng_is_initialized(void);
 int wait_for_random_bytes(void);
mm/util.c +33 -0
@@ -14,6 +14,7 @@
 #include <linux/hugetlb.h>
 #include <linux/vmalloc.h>
 #include <linux/userfaultfd_k.h>
+#include <linux/random.h>
 
 #include <asm/sections.h>
 #include <linux/uaccess.h>
@@ -286,6 +287,38 @@ int vma_is_stack_for_current(struct vm_a
 	return (vma->vm_start <= KSTK_ESP(t) && vma->vm_end >= KSTK_ESP(t));
 }
 
+/**
+ * randomize_page - Generate a random, page aligned address
+ * @start:	The smallest acceptable address the caller will take.
+ * @range:	The size of the area, starting at @start, within which the
+ *		random address must fall.
+ *
+ * If @start + @range would overflow, @range is capped.
+ *
+ * NOTE: Historical use of randomize_range, which this replaces, presumed that
+ * @start was already page aligned.  We now align it regardless.
+ *
+ * Return: A page aligned address within [start, start + range).  On error,
+ * @start is returned.
+ */
+unsigned long randomize_page(unsigned long start, unsigned long range)
+{
+	if (!PAGE_ALIGNED(start)) {
+		range -= PAGE_ALIGN(start) - start;
+		start = PAGE_ALIGN(start);
+	}
+
+	if (start > ULONG_MAX - range)
+		range = ULONG_MAX - start;
+
+	range >>= PAGE_SHIFT;
+
+	if (range == 0)
+		return start;
+
+	return start + (get_random_long() % range << PAGE_SHIFT);
+}
+
 #if defined(CONFIG_MMU) && !defined(HAVE_ARCH_PICK_MMAP_LAYOUT)
 void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack)
 {
author maintainer
#167 lore raw

PATCH 167/234 random: convert to using fops->write_iter()

In reply to Greg Kroah-Hartman

From: Jens Axboe <[email protected]>

commit 22b0a222af4df8ee9bb8e07013ab44da9511b047 upstream.

Now that the read side has been converted to fix a regression with splice, convert the write side as well to have some symmetry in the interface used (and help deprecate ->write()).

[Jason: cleaned up random_ioctl a bit, require full writes in
RNDADDENTROPY since it's crediting entropy, simplify control flow of write_pool(), and incorporate suggestions from Al.]

---
 drivers/char/random.c |   67 ++++++++++++++++++++++++++------------------------
 1 file changed, 35 insertions(+), 32 deletions(-)
drivers/char/random.c +35 -32
@@ -1251,39 +1251,31 @@ static __poll_t random_poll(struct file
 	return crng_ready() ? EPOLLIN | EPOLLRDNORM : EPOLLOUT | EPOLLWRNORM;
 }
 
-static int write_pool(const char __user *ubuf, size_t len)
+static ssize_t write_pool(struct iov_iter *iter)
 {
-	size_t block_len;
-	int ret = 0;
 	u8 block[BLAKE2S_BLOCK_SIZE];
+	ssize_t ret = 0;
+	size_t copied;
 
-	while (len) {
-		block_len = min(len, sizeof(block));
-		if (copy_from_user(block, ubuf, block_len)) {
-			ret = -EFAULT;
-			goto out;
-		}
-		len -= block_len;
-		ubuf += block_len;
-		mix_pool_bytes(block, block_len);
+	if (unlikely(!iov_iter_count(iter)))
+		return 0;
+
+	for (;;) {
+		copied = copy_from_iter(block, sizeof(block), iter);
+		ret += copied;
+		mix_pool_bytes(block, copied);
+		if (!iov_iter_count(iter) || copied != sizeof(block))
+			break;
 		cond_resched();
 	}
 
-out:
 	memzero_explicit(block, sizeof(block));
-	return ret;
+	return ret ? ret : -EFAULT;
 }
 
-static ssize_t random_write(struct file *file, const char __user *ubuf,
-			    size_t len, loff_t *ppos)
+static ssize_t random_write_iter(struct kiocb *kiocb, struct iov_iter *iter)
 {
-	int ret;
-
-	ret = write_pool(ubuf, len);
-	if (ret)
-		return ret;
-
-	return (ssize_t)len;
+	return write_pool(iter);
 }
 
 static ssize_t urandom_read_iter(struct kiocb *kiocb, struct iov_iter *iter)
@@ -1315,9 +1307,8 @@ static ssize_t random_read_iter(struct k
 
 static long random_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
 {
-	int size, ent_count;
 	int __user *p = (int __user *)arg;
-	int retval;
+	int ent_count;
 
 	switch (cmd) {
 	case RNDGETENTCNT:
@@ -1334,20 +1325,32 @@ static long random_ioctl(struct file *f,
 			return -EINVAL;
 		credit_init_bits(ent_count);
 		return 0;
-	case RNDADDENTROPY:
+	case RNDADDENTROPY: {
+		struct iov_iter iter;
+		struct iovec iov;
+		ssize_t ret;
+		int len;
+
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
 		if (get_user(ent_count, p++))
 			return -EFAULT;
 		if (ent_count < 0)
 			return -EINVAL;
-		if (get_user(size, p++))
+		if (get_user(len, p++))
+			return -EFAULT;
+		ret = import_single_range(WRITE, p, len, &iov, &iter);
+		if (unlikely(ret))
+			return ret;
+		ret = write_pool(&iter);
+		if (unlikely(ret < 0))
+			return ret;
+		/* Since we're crediting, enforce that it was all written into the pool. */
+		if (unlikely(ret != len))
 			return -EFAULT;
-		retval = write_pool((const char __user *)p, size);
-		if (retval < 0)
-			return retval;
 		credit_init_bits(ent_count);
 		return 0;
+	}
 	case RNDZAPENTCNT:
 	case RNDCLEARPOOL:
 		/* No longer has any effect. */
@@ -1373,7 +1376,7 @@ static int random_fasync(int fd, struct
 
 const struct file_operations random_fops = {
 	.read_iter = random_read_iter,
-	.write = random_write,
+	.write_iter = random_write_iter,
 	.poll = random_poll,
 	.unlocked_ioctl = random_ioctl,
 	.fasync = random_fasync,
@@ -1382,7 +1385,7 @@ const struct file_operations random_fops
 
 const struct file_operations urandom_fops = {
 	.read_iter = urandom_read_iter,
-	.write = random_write,
+	.write_iter = random_write_iter,
 	.unlocked_ioctl = random_ioctl,
 	.fasync = random_fasync,
 	.llseek = noop_llseek,
author maintainer
#168 lore raw

PATCH 168/234 random: wire up fops->splice_{read,write}_iter()

In reply to Greg Kroah-Hartman

From: Jens Axboe <[email protected]>

commit 79025e727a846be6fd215ae9cdb654368ac3f9a6 upstream.

Now that random/urandom is using {read,write}_iter, we can wire it up to using the generic splice handlers.

Fixes 36e2c7421f02 ("fs: don't allow splice read/write without explicit ops")

[Jason: added the splice_write path. Note that sendfile() and such still
does not work for read, though it does for write, because of a file type restriction in splice_direct_to_actor(), which I'll address separately.]

---
 drivers/char/random.c |    4 ++++
 1 file changed, 4 insertions(+)
drivers/char/random.c +4 -0
@@ -1381,6 +1381,8 @@ const struct file_operations random_fops
 	.unlocked_ioctl = random_ioctl,
 	.fasync = random_fasync,
 	.llseek = noop_llseek,
+	.splice_read = generic_file_splice_read,
+	.splice_write = iter_file_splice_write,
 };
 
 const struct file_operations urandom_fops = {
@@ -1389,6 +1391,8 @@ const struct file_operations urandom_fop
 	.unlocked_ioctl = random_ioctl,
 	.fasync = random_fasync,
 	.llseek = noop_llseek,
+	.splice_read = generic_file_splice_read,
+	.splice_write = iter_file_splice_write,
 };
 
 
author maintainer
#169 lore raw

PATCH 169/234 random: check for signals after page of pool writes

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 1ce6c8d68f8ac587f54d0a271ac594d3d51f3efb upstream.

get_random_bytes_user() checks for signals after producing a PAGE_SIZE worth of output, just like /dev/zero does. write_pool() is doing basically the same work (actually, slightly more expensive), and so should stop to check for signals in the same way. Let's also name it write_pool_user() to match get_random_bytes_user(), so this won't be misused in the future.

Before this patch, massive writes to /dev/urandom would tie up the process for an extremely long time and make it unterminatable. After, it can be successfully interrupted. The following test program can be used to see this works as intended:

  #include <unistd.h>
  #include <fcntl.h>
  #include <signal.h>
  #include <stdio.h>

  static unsigned char x[~0U];

static void handle(int) { }

  int main(int argc, char *argv[])
  {
    pid_t pid = getpid(), child;
    int fd;
    signal(SIGUSR1, handle);
    if (!(child = fork())) {
      for (;;)
        kill(pid, SIGUSR1);
    }
    fd = open("/dev/urandom", O_WRONLY);
    pause();
    printf("interrupted after writing %zd bytes\n", write(fd, x, sizeof(x)));
    close(fd);
    kill(child, SIGTERM);
    return 0;
  }

Result before: "interrupted after writing 2147479552 bytes" Result after: "interrupted after writing 4096 bytes"

---
 drivers/char/random.c |   14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)
drivers/char/random.c +10 -4
@@ -1251,7 +1251,7 @@ static __poll_t random_poll(struct file
 	return crng_ready() ? EPOLLIN | EPOLLRDNORM : EPOLLOUT | EPOLLWRNORM;
 }
 
-static ssize_t write_pool(struct iov_iter *iter)
+static ssize_t write_pool_user(struct iov_iter *iter)
 {
 	u8 block[BLAKE2S_BLOCK_SIZE];
 	ssize_t ret = 0;
@@ -1266,7 +1266,13 @@ static ssize_t write_pool(struct iov_ite
 		mix_pool_bytes(block, copied);
 		if (!iov_iter_count(iter) || copied != sizeof(block))
 			break;
-		cond_resched();
+
+		BUILD_BUG_ON(PAGE_SIZE % sizeof(block) != 0);
+		if (ret % PAGE_SIZE == 0) {
+			if (signal_pending(current))
+				break;
+			cond_resched();
+		}
 	}
 
 	memzero_explicit(block, sizeof(block));
@@ -1275,7 +1281,7 @@ static ssize_t write_pool(struct iov_ite
 
 static ssize_t random_write_iter(struct kiocb *kiocb, struct iov_iter *iter)
 {
-	return write_pool(iter);
+	return write_pool_user(iter);
 }
 
 static ssize_t urandom_read_iter(struct kiocb *kiocb, struct iov_iter *iter)
@@ -1342,7 +1348,7 @@ static long random_ioctl(struct file *f,
 		ret = import_single_range(WRITE, p, len, &iov, &iter);
 		if (unlikely(ret))
 			return ret;
-		ret = write_pool(&iter);
+		ret = write_pool_user(&iter);
 		if (unlikely(ret < 0))
 			return ret;
 		/* Since we're crediting, enforce that it was all written into the pool. */
author maintainer
#170 lore raw

PATCH 170/234 Revert "random: use static branch for crng_ready()"

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

This reverts upstream commit f5bda35fba615ace70a656d4700423fa6c9bebee from stable. It's not essential and will take some time during 5.19 to work out properly.

---
 drivers/char/random.c |   12 ++----------
 1 file changed, 2 insertions(+), 10 deletions(-)
drivers/char/random.c +2 -10
@@ -79,8 +79,7 @@ static enum {
 	CRNG_EARLY = 1, /* At least POOL_EARLY_BITS collected */
 	CRNG_READY = 2  /* Fully initialized with POOL_READY_BITS collected */
 } crng_init __read_mostly = CRNG_EMPTY;
-static DEFINE_STATIC_KEY_FALSE(crng_is_ready);
-#define crng_ready() (static_branch_likely(&crng_is_ready) || crng_init >= CRNG_READY)
+#define crng_ready() (likely(crng_init >= CRNG_READY))
 /* Various types of waiters for crng_init->CRNG_READY transition. */
 static DECLARE_WAIT_QUEUE_HEAD(crng_init_wait);
 static struct fasync_struct *fasync;
@@ -110,11 +109,6 @@ bool rng_is_initialized(void)
 }
 EXPORT_SYMBOL(rng_is_initialized);
 
-static void __cold crng_set_ready(struct work_struct *work)
-{
-	static_branch_enable(&crng_is_ready);
-}
-
 /* Used by wait_for_random_bytes(), and considered an entropy collector, below. */
 static void try_to_generate_entropy(void);
 
@@ -268,7 +262,7 @@ static void crng_reseed(void)
 		++next_gen;
 	WRITE_ONCE(base_crng.generation, next_gen);
 	WRITE_ONCE(base_crng.birth, jiffies);
-	if (!static_branch_likely(&crng_is_ready))
+	if (!crng_ready())
 		crng_init = CRNG_READY;
 	spin_unlock_irqrestore(&base_crng.lock, flags);
 	memzero_explicit(key, sizeof(key));
@@ -709,7 +703,6 @@ static void extract_entropy(void *buf, s
 
 static void __cold _credit_init_bits(size_t bits)
 {
-	static struct execute_work set_ready;
 	unsigned int new, orig, add;
 	unsigned long flags;
 
@@ -725,7 +718,6 @@ static void __cold _credit_init_bits(siz
 
 	if (orig < POOL_READY_BITS && new >= POOL_READY_BITS) {
 		crng_reseed(); /* Sets crng_init to CRNG_READY under base_crng.lock. */
-		execute_in_process_context(crng_set_ready, &set_ready);
 		process_random_ready_list();
 		wake_up_interruptible(&crng_init_wait);
 		kill_fasync(&fasync, SIGIO, POLL_IN);
author maintainer
#171 lore raw

PATCH 171/234 crypto: drbg - add FIPS 140-2 CTRNG for noise source

In reply to Greg Kroah-Hartman

From: Stephan Mueller <[email protected]>

commit db07cd26ac6a418dc2823187958edcfdb415fa83 upstream.

FIPS 140-2 section 4.9.2 requires a continuous self test of the noise source. Up to kernel 4.8 drivers/char/random.c provided this continuous self test. Afterwards it was moved to a location that is inconsistent with the FIPS 140-2 requirements. The relevant patch was e192be9d9a30555aae2ca1dc3aad37cba484cd4a .

Thus, the FIPS 140-2 CTRNG is added to the DRBG when it obtains the seed. This patch resurrects the function drbg_fips_continous_test that existed some time ago and applies it to the noise sources. The patch that removed the drbg_fips_continous_test was b3614763059b82c26bdd02ffcb1c016c1132aad0 .

The Jitter RNG implements its own FIPS 140-2 self test and thus does not need to be subjected to the test in the DRBG.

The patch contains a tiny fix to ensure proper zeroization in case of an error during the Jitter RNG data gathering.

---
 crypto/drbg.c         |   94 ++++++++++++++++++++++++++++++++++++++++++++++++--
 include/crypto/drbg.h |    2 +
 2 files changed, 93 insertions(+), 3 deletions(-)
crypto/drbg.c +91 -3
@@ -220,6 +220,57 @@ static inline unsigned short drbg_sec_st
 }
 
 /*
+ * FIPS 140-2 continuous self test for the noise source
+ * The test is performed on the noise source input data. Thus, the function
+ * implicitly knows the size of the buffer to be equal to the security
+ * strength.
+ *
+ * Note, this function disregards the nonce trailing the entropy data during
+ * initial seeding.
+ *
+ * drbg->drbg_mutex must have been taken.
+ *
+ * @drbg DRBG handle
+ * @entropy buffer of seed data to be checked
+ *
+ * return:
+ *	0 on success
+ *	-EAGAIN on when the CTRNG is not yet primed
+ *	< 0 on error
+ */
+static int drbg_fips_continuous_test(struct drbg_state *drbg,
+				     const unsigned char *entropy)
+{
+	unsigned short entropylen = drbg_sec_strength(drbg->core->flags);
+	int ret = 0;
+
+	if (!IS_ENABLED(CONFIG_CRYPTO_FIPS))
+		return 0;
+
+	/* skip test if we test the overall system */
+	if (list_empty(&drbg->test_data.list))
+		return 0;
+	/* only perform test in FIPS mode */
+	if (!fips_enabled)
+		return 0;
+
+	if (!drbg->fips_primed) {
+		/* Priming of FIPS test */
+		memcpy(drbg->prev, entropy, entropylen);
+		drbg->fips_primed = true;
+		/* priming: another round is needed */
+		return -EAGAIN;
+	}
+	ret = memcmp(drbg->prev, entropy, entropylen);
+	if (!ret)
+		panic("DRBG continuous self test failed\n");
+	memcpy(drbg->prev, entropy, entropylen);
+
+	/* the test shall pass when the two values are not equal */
+	return 0;
+}
+
+/*
  * Convert an integer into a byte representation of this integer.
  * The byte representation is big-endian
  *
@@ -998,6 +1049,22 @@ static inline int __drbg_seed(struct drb
 	return ret;
 }
 
+static inline int drbg_get_random_bytes(struct drbg_state *drbg,
+					unsigned char *entropy,
+					unsigned int entropylen)
+{
+	int ret;
+
+	do {
+		get_random_bytes(entropy, entropylen);
+		ret = drbg_fips_continuous_test(drbg, entropy);
+		if (ret && ret != -EAGAIN)
+			return ret;
+	} while (ret);
+
+	return 0;
+}
+
 static void drbg_async_seed(struct work_struct *work)
 {
 	struct drbg_string data;
@@ -1006,16 +1073,20 @@ static void drbg_async_seed(struct work_
 					       seed_work);
 	unsigned int entropylen = drbg_sec_strength(drbg->core->flags);
 	unsigned char entropy[32];
+	int ret;
 
 	BUG_ON(!entropylen);
 	BUG_ON(entropylen > sizeof(entropy));
-	get_random_bytes(entropy, entropylen);
 
 	drbg_string_fill(&data, entropy, entropylen);
 	list_add_tail(&data.list, &seedlist);
 
 	mutex_lock(&drbg->drbg_mutex);
 
+	ret = drbg_get_random_bytes(drbg, entropy, entropylen);
+	if (ret)
+		goto unlock;
+
 	/* If nonblocking pool is initialized, deactivate Jitter RNG */
 	crypto_free_rng(drbg->jent);
 	drbg->jent = NULL;
@@ -1030,6 +1101,7 @@ static void drbg_async_seed(struct work_
 	if (drbg->seeded)
 		drbg->reseed_threshold = drbg_max_requests(drbg);
 
+unlock:
 	mutex_unlock(&drbg->drbg_mutex);
 
 	memzero_explicit(entropy, entropylen);
@@ -1081,7 +1153,9 @@ static int drbg_seed(struct drbg_state *
 		BUG_ON((entropylen * 2) > sizeof(entropy));
 
 		/* Get seed from in-kernel /dev/urandom */
-		get_random_bytes(entropy, entropylen);
+		ret = drbg_get_random_bytes(drbg, entropy, entropylen);
+		if (ret)
+			goto out;
 
 		if (!drbg->jent) {
 			drbg_string_fill(&data1, entropy, entropylen);
@@ -1094,7 +1168,7 @@ static int drbg_seed(struct drbg_state *
 						   entropylen);
 			if (ret) {
 				pr_devel("DRBG: jent failed with %d\n", ret);
-				return ret;
+				goto out;
 			}
 
 			drbg_string_fill(&data1, entropy, entropylen * 2);
@@ -1121,6 +1195,7 @@ static int drbg_seed(struct drbg_state *
 
 	ret = __drbg_seed(drbg, &seedlist, reseed);
 
+out:
 	memzero_explicit(entropy, entropylen * 2);
 
 	return ret;
@@ -1142,6 +1217,11 @@ static inline void drbg_dealloc_state(st
 	drbg->reseed_ctr = 0;
 	drbg->d_ops = NULL;
 	drbg->core = NULL;
+	if (IS_ENABLED(CONFIG_CRYPTO_FIPS)) {
+		kzfree(drbg->prev);
+		drbg->prev = NULL;
+		drbg->fips_primed = false;
+	}
 }
 
 /*
@@ -1211,6 +1291,14 @@ static inline int drbg_alloc_state(struc
 		drbg->scratchpad = PTR_ALIGN(drbg->scratchpadbuf, ret + 1);
 	}
 
+	if (IS_ENABLED(CONFIG_CRYPTO_FIPS)) {
+		drbg->prev = kzalloc(drbg_sec_strength(drbg->core->flags),
+				     GFP_KERNEL);
+		if (!drbg->prev)
+			goto fini;
+		drbg->fips_primed = false;
+	}
+
 	return 0;
 
 fini:
include/crypto/drbg.h +2 -0
@@ -129,6 +129,8 @@ struct drbg_state {
 
 	bool seeded;		/* DRBG fully seeded? */
 	bool pr;		/* Prediction resistance enabled? */
+	bool fips_primed;	/* Continuous test primed? */
+	unsigned char *prev;	/* FIPS 140-2 continuous test value */
 	struct work_struct seed_work;	/* asynchronous seeding support */
 	struct crypto_rng *jent;
 	const struct drbg_state_ops *d_ops;
author maintainer
#172 lore raw

PATCH 172/234 crypto: drbg - always seeded with SP800-90B compliant noise source

In reply to Greg Kroah-Hartman

From: "Stephan M�ller" <[email protected]>

commit 97f2650e504033376e8813691cb6eccf73151676 upstream.

As the Jitter RNG provides an SP800-90B compliant noise source, use this noise source always for the (re)seeding of the DRBG.

To make sure the DRBG is always properly seeded, the reseed threshold is reduced to 1<<20 generate operations.

The Jitter RNG may report health test failures. Such health test failures are treated as transient as follows. The DRBG will not reseed from the Jitter RNG (but from get_random_bytes) in case of a health test failure. Though, it produces the requested random number.

The Jitter RNG has a failure counter where at most 1024 consecutive resets due to a health test failure are considered as a transient error. If more consecutive resets are required, the Jitter RNG will return a permanent error which is returned to the caller by the DRBG. With this approach, the worst case reseed threshold is significantly lower than mandated by SP800-90A in order to seed with an SP800-90B noise source:
the DRBG has a reseed threshold of 2^20 * 1024 = 2^30 generate requests.

Yet, in case of a transient Jitter RNG health test failure, the DRBG is seeded with the data obtained from get_random_bytes.

However, if the Jitter RNG fails during the initial seeding operation even due to a health test error, the DRBG will send an error to the caller because at that time, the DRBG has received no seed that is SP800-90B compliant.

---
 crypto/drbg.c         |   26 +++++++++++++++++++-------
 include/crypto/drbg.h |    6 +-----
 2 files changed, 20 insertions(+), 12 deletions(-)
crypto/drbg.c +19 -7
@@ -1087,10 +1087,6 @@ static void drbg_async_seed(struct work_
 	if (ret)
 		goto unlock;
 
-	/* If nonblocking pool is initialized, deactivate Jitter RNG */
-	crypto_free_rng(drbg->jent);
-	drbg->jent = NULL;
-
 	/* Set seeded to false so that if __drbg_seed fails the
 	 * next generate call will trigger a reseed.
 	 */
@@ -1168,7 +1164,23 @@ static int drbg_seed(struct drbg_state *
 						   entropylen);
 			if (ret) {
 				pr_devel("DRBG: jent failed with %d\n", ret);
-				goto out;
+
+				/*
+				 * Do not treat the transient failure of the
+				 * Jitter RNG as an error that needs to be
+				 * reported. The combined number of the
+				 * maximum reseed threshold times the maximum
+				 * number of Jitter RNG transient errors is
+				 * less than the reseed threshold required by
+				 * SP800-90A allowing us to treat the
+				 * transient errors as such.
+				 *
+				 * However, we mandate that at least the first
+				 * seeding operation must succeed with the
+				 * Jitter RNG.
+				 */
+				if (!reseed || ret != -EAGAIN)
+					goto out;
 			}
 
 			drbg_string_fill(&data1, entropy, entropylen * 2);
@@ -1493,6 +1505,8 @@ static int drbg_prepare_hrng(struct drbg
 	if (list_empty(&drbg->test_data.list))
 		return 0;
 
+	drbg->jent = crypto_alloc_rng("jitterentropy_rng", 0, 0);
+
 	INIT_WORK(&drbg->seed_work, drbg_async_seed);
 
 	drbg->random_ready.notifier_call = drbg_schedule_async_seed;
@@ -1511,8 +1525,6 @@ static int drbg_prepare_hrng(struct drbg
 		return err;
 	}
 
-	drbg->jent = crypto_alloc_rng("jitterentropy_rng", 0, 0);
-
 	/*
 	 * Require frequent reseeds until the seed source is fully
 	 * initialized.
include/crypto/drbg.h +1 -5
@@ -184,11 +184,7 @@ static inline size_t drbg_max_addtl(stru
 static inline size_t drbg_max_requests(struct drbg_state *drbg)
 {
 	/* SP800-90A requires 2**48 maximum requests before reseeding */
-#if (__BITS_PER_LONG == 32)
-	return SIZE_MAX;
-#else
-	return (1UL<<48);
-#endif
+	return (1<<20);
 }
 
 /*
author maintainer
#173 lore raw

PATCH 173/234 crypto: drbg - prepare for more fine-grained tracking of seeding state

In reply to Greg Kroah-Hartman

From: Nicolai Stange <[email protected]>

commit ce8ce31b2c5c8b18667784b8c515650c65d57b4e upstream.

There are two different randomness sources the DRBGs are getting seeded from, namely the jitterentropy source (if enabled) and get_random_bytes(). At initial DRBG seeding time during boot, the latter might not have collected sufficient entropy for seeding itself yet and thus, the DRBG implementation schedules a reseed work from a random_ready_callback once that has happened. This is particularly important for the !->pr DRBG instances, for which (almost) no further reseeds are getting triggered during their lifetime.

Because collecting data from the jitterentropy source is a rather expensive operation, the aforementioned asynchronously scheduled reseed work restricts itself to get_random_bytes() only. That is, it in some sense amends the initial DRBG seed derived from jitterentropy output at full (estimated) entropy with fresh randomness obtained from get_random_bytes() once that has been seeded with sufficient entropy itself.

With the advent of rng_is_initialized(), there is no real need for doing the reseed operation from an asynchronously scheduled work anymore and a subsequent patch will make it synchronous by moving it next to related logic already present in drbg_generate().

However, for tracking whether a full reseed including the jitterentropy source is required or a "partial" reseed involving only get_random_bytes() would be sufficient already, the boolean struct drbg_state's ->seeded member must become a tristate value.

Prepare for this by introducing the new enum drbg_seed_state and change struct drbg_state's ->seeded member's type from bool to that type.

For facilitating review, enum drbg_seed_state is made to only contain two members corresponding to the former ->seeded values of false and true resp. at this point: DRBG_SEED_STATE_UNSEEDED and DRBG_SEED_STATE_FULL. A third one for tracking the intermediate state of "seeded from jitterentropy only" will be introduced with a subsequent patch.

There is no change in behaviour at this point.

---
 crypto/drbg.c         |   19 ++++++++++---------
 include/crypto/drbg.h |    7 ++++++-
 2 files changed, 16 insertions(+), 10 deletions(-)
crypto/drbg.c +10 -9
@@ -1042,7 +1042,7 @@ static inline int __drbg_seed(struct drb
 	if (ret)
 		return ret;
 
-	drbg->seeded = true;
+	drbg->seeded = DRBG_SEED_STATE_FULL;
 	/* 10.1.1.2 / 10.1.1.3 step 5 */
 	drbg->reseed_ctr = 1;
 
@@ -1087,14 +1087,14 @@ static void drbg_async_seed(struct work_
 	if (ret)
 		goto unlock;
 
-	/* Set seeded to false so that if __drbg_seed fails the
-	 * next generate call will trigger a reseed.
+	/* Reset ->seeded so that if __drbg_seed fails the next
+	 * generate call will trigger a reseed.
 	 */
-	drbg->seeded = false;
+	drbg->seeded = DRBG_SEED_STATE_UNSEEDED;
 
 	__drbg_seed(drbg, &seedlist, true);
 
-	if (drbg->seeded)
+	if (drbg->seeded == DRBG_SEED_STATE_FULL)
 		drbg->reseed_threshold = drbg_max_requests(drbg);
 
 unlock:
@@ -1383,13 +1383,14 @@ static int drbg_generate(struct drbg_sta
 	 * here. The spec is a bit convoluted here, we make it simpler.
 	 */
 	if (drbg->reseed_threshold < drbg->reseed_ctr)
-		drbg->seeded = false;
+		drbg->seeded = DRBG_SEED_STATE_UNSEEDED;
 
-	if (drbg->pr || !drbg->seeded) {
+	if (drbg->pr || drbg->seeded == DRBG_SEED_STATE_UNSEEDED) {
 		pr_devel("DRBG: reseeding before generation (prediction "
 			 "resistance: %s, state %s)\n",
 			 drbg->pr ? "true" : "false",
-			 drbg->seeded ? "seeded" : "unseeded");
+			 (drbg->seeded ==  DRBG_SEED_STATE_FULL ?
+			  "seeded" : "unseeded"));
 		/* 9.3.1 steps 7.1 through 7.3 */
 		len = drbg_seed(drbg, addtl, true);
 		if (len)
@@ -1574,7 +1575,7 @@ static int drbg_instantiate(struct drbg_
 	if (!drbg->core) {
 		drbg->core = &drbg_cores[coreref];
 		drbg->pr = pr;
-		drbg->seeded = false;
+		drbg->seeded = DRBG_SEED_STATE_UNSEEDED;
 		drbg->reseed_threshold = drbg_max_requests(drbg);
 
 		ret = drbg_alloc_state(drbg);
include/crypto/drbg.h +6 -1
@@ -105,6 +105,11 @@ struct drbg_test_data {
 	struct drbg_string *testentropy; /* TEST PARAMETER: test entropy */
 };
 
+enum drbg_seed_state {
+	DRBG_SEED_STATE_UNSEEDED,
+	DRBG_SEED_STATE_FULL,
+};
+
 struct drbg_state {
 	struct mutex drbg_mutex;	/* lock around DRBG */
 	unsigned char *V;	/* internal state 10.1.1.1 1a) */
@@ -127,7 +132,7 @@ struct drbg_state {
 	struct crypto_wait ctr_wait;		/* CTR mode async wait obj */
 	struct scatterlist sg_in, sg_out;	/* CTR mode SGLs */
 
-	bool seeded;		/* DRBG fully seeded? */
+	enum drbg_seed_state seeded;		/* DRBG fully seeded? */
 	bool pr;		/* Prediction resistance enabled? */
 	bool fips_primed;	/* Continuous test primed? */
 	unsigned char *prev;	/* FIPS 140-2 continuous test value */
author maintainer
#174 lore raw

PATCH 174/234 crypto: drbg - track whether DRBG was seeded with !rng_is_initialized()

In reply to Greg Kroah-Hartman

From: Nicolai Stange <[email protected]>

commit 2bcd25443868aa8863779a6ebc6c9319633025d2 upstream.

Currently, the DRBG implementation schedules asynchronous works from random_ready_callbacks for reseeding the DRBG instances with output from get_random_bytes() once the latter has sufficient entropy available.

However, as the get_random_bytes() initialization state can get queried by means of rng_is_initialized() now, there is no real need for this asynchronous reseeding logic anymore and it's better to keep things simple by doing it synchronously when needed instead, i.e. from drbg_generate() once rng_is_initialized() has flipped to true.

Of course, for this to work, drbg_generate() would need some means by which it can tell whether or not rng_is_initialized() has flipped to true since the last seeding from get_random_bytes(). Or equivalently, whether or not the last seed from get_random_bytes() has happened when rng_is_initialized() was still evaluating to false.

As it currently stands, enum drbg_seed_state allows for the representation of two different DRBG seeding states: DRBG_SEED_STATE_UNSEEDED and DRBG_SEED_STATE_FULL. The former makes drbg_generate() to invoke a full reseeding operation involving both, the rather expensive jitterentropy as well as the get_random_bytes() randomness sources. The DRBG_SEED_STATE_FULL state on the other hand implies that no reseeding at all is required for a !->pr DRBG variant.

Introduce the new DRBG_SEED_STATE_PARTIAL state to enum drbg_seed_state for representing the condition that a DRBG was being seeded when rng_is_initialized() had still been false. In particular, this new state implies that
- the given DRBG instance has been fully seeded from the jitterentropy source (if enabled)
- and drbg_generate() is supposed to reseed from get_random_bytes() *only* once rng_is_initialized() turns to true.

Up to now, the __drbg_seed() helper used to set the given DRBG instance's ->seeded state to constant DRBG_SEED_STATE_FULL. Introduce a new argument allowing for the specification of the to be written ->seeded value instead. Make the first of its two callers, drbg_seed(), determine the appropriate value based on rng_is_initialized(). The remaining caller, drbg_async_seed(), is known to get invoked only once rng_is_initialized() is true, hence let it pass constant DRBG_SEED_STATE_FULL for the new argument to __drbg_seed().

There is no change in behaviour, except for that the pr_devel() in drbg_generate() would now report "unseeded" for ->pr DRBG instances which had last been seeded when rng_is_initialized() was still evaluating to false.

---
 crypto/drbg.c         |   12 ++++++++----
 include/crypto/drbg.h |    1 +
 2 files changed, 9 insertions(+), 4 deletions(-)
crypto/drbg.c +8 -4
@@ -1035,14 +1035,14 @@ static const struct drbg_state_ops drbg_
  ******************************************************************/
 
 static inline int __drbg_seed(struct drbg_state *drbg, struct list_head *seed,
-			      int reseed)
+			      int reseed, enum drbg_seed_state new_seed_state)
 {
 	int ret = drbg->d_ops->update(drbg, seed, reseed);
 
 	if (ret)
 		return ret;
 
-	drbg->seeded = DRBG_SEED_STATE_FULL;
+	drbg->seeded = new_seed_state;
 	/* 10.1.1.2 / 10.1.1.3 step 5 */
 	drbg->reseed_ctr = 1;
 
@@ -1092,7 +1092,7 @@ static void drbg_async_seed(struct work_
 	 */
 	drbg->seeded = DRBG_SEED_STATE_UNSEEDED;
 
-	__drbg_seed(drbg, &seedlist, true);
+	__drbg_seed(drbg, &seedlist, true, DRBG_SEED_STATE_FULL);
 
 	if (drbg->seeded == DRBG_SEED_STATE_FULL)
 		drbg->reseed_threshold = drbg_max_requests(drbg);
@@ -1122,6 +1122,7 @@ static int drbg_seed(struct drbg_state *
 	unsigned int entropylen = drbg_sec_strength(drbg->core->flags);
 	struct drbg_string data1;
 	LIST_HEAD(seedlist);
+	enum drbg_seed_state new_seed_state = DRBG_SEED_STATE_FULL;
 
 	/* 9.1 / 9.2 / 9.3.1 step 3 */
 	if (pers && pers->len > (drbg_max_addtl(drbg))) {
@@ -1149,6 +1150,9 @@ static int drbg_seed(struct drbg_state *
 		BUG_ON((entropylen * 2) > sizeof(entropy));
 
 		/* Get seed from in-kernel /dev/urandom */
+		if (!rng_is_initialized())
+			new_seed_state = DRBG_SEED_STATE_PARTIAL;
+
 		ret = drbg_get_random_bytes(drbg, entropy, entropylen);
 		if (ret)
 			goto out;
@@ -1205,7 +1209,7 @@ static int drbg_seed(struct drbg_state *
 		memset(drbg->C, 0, drbg_statelen(drbg));
 	}
 
-	ret = __drbg_seed(drbg, &seedlist, reseed);
+	ret = __drbg_seed(drbg, &seedlist, reseed, new_seed_state);
 
 out:
 	memzero_explicit(entropy, entropylen * 2);
include/crypto/drbg.h +1 -0
@@ -107,6 +107,7 @@ struct drbg_test_data {
 
 enum drbg_seed_state {
 	DRBG_SEED_STATE_UNSEEDED,
+	DRBG_SEED_STATE_PARTIAL, /* Seeded with !rng_is_initialized() */
 	DRBG_SEED_STATE_FULL,
 };
 
author maintainer
#175 lore raw

PATCH 175/234 crypto: drbg - move dynamic ->reseed_threshold adjustments to __drbg_seed()

In reply to Greg Kroah-Hartman

From: Nicolai Stange <[email protected]>

commit 262d83a4290c331cd4f617a457408bdb82fbb738 upstream.

Since commit 42ea507fae1a ("crypto: drbg - reseed often if seedsource is degraded"), the maximum seed lifetime represented by ->reseed_threshold gets temporarily lowered if the get_random_bytes() source cannot provide sufficient entropy yet, as is common during boot, and restored back to the original value again once that has changed.

More specifically, if the add_random_ready_callback() invoked from drbg_prepare_hrng() in the course of DRBG instantiation does not return -EALREADY, that is, if get_random_bytes() has not been fully initialized at this point yet, drbg_prepare_hrng() will lower ->reseed_threshold to a value of 50. The drbg_async_seed() scheduled from said random_ready_callback will eventually restore the original value.

A future patch will replace the random_ready_callback based notification mechanism and thus, there will be no add_random_ready_callback() return value anymore which could get compared to -EALREADY.

However, there's __drbg_seed() which gets invoked in the course of both, the DRBG instantiation as well as the eventual reseeding from get_random_bytes() in aforementioned drbg_async_seed(), if any. Moreover, it knows about the get_random_bytes() initialization state by the time the seed data had been obtained from it: the new_seed_state argument introduced with the previous patch would get set to DRBG_SEED_STATE_PARTIAL in case get_random_bytes() had not been fully initialized yet and to DRBG_SEED_STATE_FULL otherwise. Thus, __drbg_seed() provides a convenient alternative for managing that ->reseed_threshold lowering and restoring at a central place.

Move all ->reseed_threshold adjustment code from drbg_prepare_hrng() and drbg_async_seed() respectively to __drbg_seed(). Make __drbg_seed() lower the ->reseed_threshold to 50 in case its new_seed_state argument equals DRBG_SEED_STATE_PARTIAL and let it restore the original value otherwise.

There is no change in behaviour.

---
 crypto/drbg.c |   29 ++++++++++++++++++++---------
 1 file changed, 20 insertions(+), 9 deletions(-)
crypto/drbg.c +20 -9
@@ -1046,6 +1046,26 @@ static inline int __drbg_seed(struct drb
 	/* 10.1.1.2 / 10.1.1.3 step 5 */
 	drbg->reseed_ctr = 1;
 
+	switch (drbg->seeded) {
+	case DRBG_SEED_STATE_UNSEEDED:
+		/* Impossible, but handle it to silence compiler warnings. */
+	case DRBG_SEED_STATE_PARTIAL:
+		/*
+		 * Require frequent reseeds until the seed source is
+		 * fully initialized.
+		 */
+		drbg->reseed_threshold = 50;
+		break;
+
+	case DRBG_SEED_STATE_FULL:
+		/*
+		 * Seed source has become fully initialized, frequent
+		 * reseeds no longer required.
+		 */
+		drbg->reseed_threshold = drbg_max_requests(drbg);
+		break;
+	}
+
 	return ret;
 }
 
@@ -1094,9 +1114,6 @@ static void drbg_async_seed(struct work_
 
 	__drbg_seed(drbg, &seedlist, true, DRBG_SEED_STATE_FULL);
 
-	if (drbg->seeded == DRBG_SEED_STATE_FULL)
-		drbg->reseed_threshold = drbg_max_requests(drbg);
-
 unlock:
 	mutex_unlock(&drbg->drbg_mutex);
 
@@ -1530,12 +1547,6 @@ static int drbg_prepare_hrng(struct drbg
 		return err;
 	}
 
-	/*
-	 * Require frequent reseeds until the seed source is fully
-	 * initialized.
-	 */
-	drbg->reseed_threshold = 50;
-
 	return err;
 }
 
author maintainer
#176 lore raw

PATCH 176/234 crypto: drbg - always try to free Jitter RNG instance

In reply to Greg Kroah-Hartman

From: "Stephan M�ller" <[email protected]>

commit 819966c06b759022e9932f328284314d9272b9f3 upstream.

The Jitter RNG is unconditionally allocated as a seed source follwoing the patch 97f2650e5040. Thus, the instance must always be deallocated.

Reported-by [email protected]
Fixes 97f2650e5040 ("crypto: drbg - always seeded with SP800-90B ...")
---
 crypto/drbg.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
crypto/drbg.c +4 -2
@@ -1644,10 +1644,12 @@ static int drbg_uninstantiate(struct drb
 	if (drbg->random_ready.notifier_call) {
 		unregister_random_ready_notifier(&drbg->random_ready);
 		cancel_work_sync(&drbg->seed_work);
-		crypto_free_rng(drbg->jent);
-		drbg->jent = NULL;
 	}
 
+	if (!IS_ERR_OR_NULL(drbg->jent))
+		crypto_free_rng(drbg->jent);
+	drbg->jent = NULL;
+
 	if (drbg->d_ops)
 		drbg->d_ops->crypto_fini(drbg);
 	drbg_dealloc_state(drbg);
author maintainer
#177 lore raw

PATCH 177/234 crypto: drbg - make reseeding from get_random_bytes() synchronous

In reply to Greg Kroah-Hartman

From: Nicolai Stange <[email protected]>

commit 074bcd4000e0d812bc253f86fedc40f81ed59ccc upstream.

get_random_bytes() usually hasn't full entropy available by the time DRBG instances are first getting seeded from it during boot. Thus, the DRBG implementation registers random_ready_callbacks which would in turn schedule some work for reseeding the DRBGs once get_random_bytes() has sufficient entropy available.

For reference, the relevant history around handling DRBG (re)seeding in the context of a not yet fully seeded get_random_bytes() is:

commit 16b369a91d0d ("random: Blocking API for accessing
nonblocking_pool")
commit 4c7879907edd ("crypto: drbg - add async seeding operation")

commit 205a525c3342 ("random: Add callback API for random pool
readiness")
commit 57225e679788 ("crypto: drbg - Use callback API for random
readiness")
commit c2719503f5e1 ("random: Remove kernel blocking API")

However, some time later, the initialization state of get_random_bytes() has been made queryable via rng_is_initialized() introduced with commit 9a47249d444d ("random: Make crng state queryable"). This primitive now allows for streamlining the DRBG reseeding from get_random_bytes() by replacing that aforementioned asynchronous work scheduling from random_ready_callbacks with some simpler, synchronous code in drbg_generate() next to the related logic already present therein. Apart from improving overall code readability, this change will also enable DRBG users to rely on wait_for_random_bytes() for ensuring that the initial seeding has completed, if desired.

The previous patches already laid the grounds by making drbg_seed() to record at each DRBG instance whether it was being seeded at a time when rng_is_initialized() still had been false as indicated by ->seeded == DRBG_SEED_STATE_PARTIAL.

All that remains to be done now is to make drbg_generate() check for this condition, determine whether rng_is_initialized() has flipped to true in the meanwhile and invoke a reseed from get_random_bytes() if so.

Make this move:
- rename the former drbg_async_seed() work handler, i.e. the one in charge of reseeding a DRBG instance from get_random_bytes(), to "drbg_seed_from_random()",
- change its signature as appropriate, i.e. make it take a struct drbg_state rather than a work_struct and change its return type from "void" to "int" in order to allow for passing error information from e.g. its __drbg_seed() invocation onwards to callers,
- make drbg_generate() invoke this drbg_seed_from_random() once it encounters a DRBG instance with ->seeded == DRBG_SEED_STATE_PARTIAL by the time rng_is_initialized() has flipped to true and
- prune everything related to the former, random_ready_callback based mechanism.

As drbg_seed_from_random() is now getting invoked from drbg_generate() with the ->drbg_mutex being held, it must not attempt to recursively grab it once again. Remove the corresponding mutex operations from what is now drbg_seed_from_random(). Furthermore, as drbg_seed_from_random() can now report errors directly to its caller, there's no need for it to temporarily switch the DRBG's ->seeded state to DRBG_SEED_STATE_UNSEEDED so that a failure of the subsequently invoked __drbg_seed() will get signaled to drbg_generate(). Don't do it then.

[Jason: for stable, undid the modifications for the backport of 5acd3548.]
---
 crypto/drbg.c         |   61 +++++++++-----------------------------------------
 drivers/char/random.c |    2 -
 include/crypto/drbg.h |    2 -
 3 files changed, 11 insertions(+), 54 deletions(-)
crypto/drbg.c +11 -50
@@ -1085,12 +1085,10 @@ static inline int drbg_get_random_bytes(
 	return 0;
 }
 
-static void drbg_async_seed(struct work_struct *work)
+static int drbg_seed_from_random(struct drbg_state *drbg)
 {
 	struct drbg_string data;
 	LIST_HEAD(seedlist);
-	struct drbg_state *drbg = container_of(work, struct drbg_state,
-					       seed_work);
 	unsigned int entropylen = drbg_sec_strength(drbg->core->flags);
 	unsigned char entropy[32];
 	int ret;
@@ -1101,23 +1099,15 @@ static void drbg_async_seed(struct work_
 	drbg_string_fill(&data, entropy, entropylen);
 	list_add_tail(&data.list, &seedlist);
 
-	mutex_lock(&drbg->drbg_mutex);
-
 	ret = drbg_get_random_bytes(drbg, entropy, entropylen);
 	if (ret)
-		goto unlock;
-
-	/* Reset ->seeded so that if __drbg_seed fails the next
-	 * generate call will trigger a reseed.
-	 */
-	drbg->seeded = DRBG_SEED_STATE_UNSEEDED;
+		goto out;
 
-	__drbg_seed(drbg, &seedlist, true, DRBG_SEED_STATE_FULL);
-
-unlock:
-	mutex_unlock(&drbg->drbg_mutex);
+	ret = __drbg_seed(drbg, &seedlist, true, DRBG_SEED_STATE_FULL);
 
+out:
 	memzero_explicit(entropy, entropylen);
+	return ret;
 }
 
 /*
@@ -1418,6 +1408,11 @@ static int drbg_generate(struct drbg_sta
 			goto err;
 		/* 9.3.1 step 7.4 */
 		addtl = NULL;
+	} else if (rng_is_initialized() &&
+		   drbg->seeded == DRBG_SEED_STATE_PARTIAL) {
+		len = drbg_seed_from_random(drbg);
+		if (len)
+			goto err;
 	}
 
 	if (addtl && 0 < addtl->len)
@@ -1510,44 +1505,15 @@ static int drbg_generate_long(struct drb
 	return 0;
 }
 
-static int drbg_schedule_async_seed(struct notifier_block *nb, unsigned long action, void *data)
-{
-	struct drbg_state *drbg = container_of(nb, struct drbg_state,
-					       random_ready);
-
-	schedule_work(&drbg->seed_work);
-	return 0;
-}
-
 static int drbg_prepare_hrng(struct drbg_state *drbg)
 {
-	int err;
-
 	/* We do not need an HRNG in test mode. */
 	if (list_empty(&drbg->test_data.list))
 		return 0;
 
 	drbg->jent = crypto_alloc_rng("jitterentropy_rng", 0, 0);
 
-	INIT_WORK(&drbg->seed_work, drbg_async_seed);
-
-	drbg->random_ready.notifier_call = drbg_schedule_async_seed;
-	err = register_random_ready_notifier(&drbg->random_ready);
-
-	switch (err) {
-	case 0:
-		break;
-
-	case -EALREADY:
-		err = 0;
-		/* fall through */
-
-	default:
-		drbg->random_ready.notifier_call = NULL;
-		return err;
-	}
-
-	return err;
+	return 0;
 }
 
 /*
@@ -1641,11 +1607,6 @@ free_everything:
  */
 static int drbg_uninstantiate(struct drbg_state *drbg)
 {
-	if (drbg->random_ready.notifier_call) {
-		unregister_random_ready_notifier(&drbg->random_ready);
-		cancel_work_sync(&drbg->seed_work);
-	}
-
 	if (!IS_ERR_OR_NULL(drbg->jent))
 		crypto_free_rng(drbg->jent);
 	drbg->jent = NULL;
drivers/char/random.c +0 -2
@@ -157,7 +157,6 @@ int __cold register_random_ready_notifie
 	spin_unlock_irqrestore(&random_ready_chain_lock, flags);
 	return ret;
 }
-EXPORT_SYMBOL(register_random_ready_notifier);
 
 /*
  * Delete a previously registered readiness callback function.
@@ -172,7 +171,6 @@ int __cold unregister_random_ready_notif
 	spin_unlock_irqrestore(&random_ready_chain_lock, flags);
 	return ret;
 }
-EXPORT_SYMBOL(unregister_random_ready_notifier);
 
 static void __cold process_random_ready_list(void)
 {
include/crypto/drbg.h +0 -2
@@ -137,12 +137,10 @@ struct drbg_state {
 	bool pr;		/* Prediction resistance enabled? */
 	bool fips_primed;	/* Continuous test primed? */
 	unsigned char *prev;	/* FIPS 140-2 continuous test value */
-	struct work_struct seed_work;	/* asynchronous seeding support */
 	struct crypto_rng *jent;
 	const struct drbg_state_ops *d_ops;
 	const struct drbg_core *core;
 	struct drbg_string test_data;
-	struct notifier_block random_ready;
 };
 
 static inline __u8 drbg_statelen(struct drbg_state *drbg)
author maintainer
#178 lore raw

PATCH 178/234 random: avoid checking crng_ready() twice in random_init()

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 9b29b6b20376ab64e1b043df6301d8a92378e631 upstream.

The current flow expands to:

    if (crng_ready())
       ...
    else if (...)
        if (!crng_ready())
            ...

The second crng_ready() call is redundant, but can't so easily be optimized out by the compiler.

This commit simplifies that to:

    if (crng_ready()
        ...
    else if (...)
        ...
Fixes 560181c27b58 ("random: move initialization functions out of hot pages")
Cc Dominik Brodowski <[email protected]>
---
 drivers/char/random.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/char/random.c +1 -1
@@ -834,7 +834,7 @@ int __init random_init(const char *comma
 	if (crng_ready())
 		crng_reseed();
 	else if (trust_cpu)
-		credit_init_bits(arch_bytes * 8);
+		_credit_init_bits(arch_bytes * 8);
 
 	return 0;
 }
author maintainer
#179 lore raw

PATCH 179/234 random: mark bootloader randomness code as __init

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 39e0f991a62ed5efabd20711a7b6e7da92603170 upstream.

add_bootloader_randomness() and the variables it touches are only used during __init and not after, so mark these as __init. At the same time, unexport this, since it's only called by other __init code that's built-in.

Fixes 428826f5358c ("fdt: add support for rng-seed")
---
 drivers/char/random.c  |    7 +++----
 include/linux/random.h |    2 +-
 2 files changed, 4 insertions(+), 5 deletions(-)
drivers/char/random.c +3 -4
@@ -785,8 +785,8 @@ static void __cold _credit_init_bits(siz
  *
  **********************************************************************/
 
-static bool trust_cpu __ro_after_init = IS_ENABLED(CONFIG_RANDOM_TRUST_CPU);
-static bool trust_bootloader __ro_after_init = IS_ENABLED(CONFIG_RANDOM_TRUST_BOOTLOADER);
+static bool trust_cpu __initdata = IS_ENABLED(CONFIG_RANDOM_TRUST_CPU);
+static bool trust_bootloader __initdata = IS_ENABLED(CONFIG_RANDOM_TRUST_BOOTLOADER);
 static int __init parse_trust_cpu(char *arg)
 {
 	return kstrtobool(arg, &trust_cpu);
@@ -882,13 +882,12 @@ EXPORT_SYMBOL_GPL(add_hwgenerator_random
  * Handle random seed passed by bootloader, and credit it if
  * CONFIG_RANDOM_TRUST_BOOTLOADER is set.
  */
-void __cold add_bootloader_randomness(const void *buf, size_t len)
+void __init add_bootloader_randomness(const void *buf, size_t len)
 {
 	mix_pool_bytes(buf, len);
 	if (trust_bootloader)
 		credit_init_bits(len * 8);
 }
-EXPORT_SYMBOL_GPL(add_bootloader_randomness);
 
 struct fast_pool {
 	struct work_struct mix;
include/linux/random.h +1 -1
@@ -13,7 +13,7 @@
 struct notifier_block;
 
 void add_device_randomness(const void *buf, size_t len);
-void add_bootloader_randomness(const void *buf, size_t len);
+void __init add_bootloader_randomness(const void *buf, size_t len);
 void add_input_randomness(unsigned int type, unsigned int code,
 			  unsigned int value) __latent_entropy;
 void add_interrupt_randomness(int irq) __latent_entropy;
author maintainer
#180 lore raw

PATCH 180/234 random: account for arch randomness in bits

In reply to Greg Kroah-Hartman

From: "Jason A. Donenfeld" <[email protected]>

commit 77fc95f8c0dc9e1f8e620ec14d2fb65028fb7adc upstream.

Rather than accounting in bytes and multiplying (shifting), we can just account in bits and avoid the shift. The main motivation for this is there are other patches in flux that expand this code a bit, and avoiding the duplication of "* 8" everywhere makes things a bit clearer.

Fixes 12e45a2a6308 ("random: credit architectural init the exact amount")
---
 drivers/char/random.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/char/random.c +4 -4
@@ -809,7 +809,7 @@ early_param("random.trust_bootloader", p
 int __init random_init(const char *command_line)
 {
 	ktime_t now = ktime_get_real();
-	unsigned int i, arch_bytes;
+	unsigned int i, arch_bits;
 	unsigned long entropy;
 
 #if defined(LATENT_ENTROPY_PLUGIN)
@@ -817,12 +817,12 @@ int __init random_init(const char *comma
 	_mix_pool_bytes(compiletime_seed, sizeof(compiletime_seed));
 #endif
 
-	for (i = 0, arch_bytes = BLAKE2S_BLOCK_SIZE;
+	for (i = 0, arch_bits = BLAKE2S_BLOCK_SIZE * 8;
 	     i < BLAKE2S_BLOCK_SIZE; i += sizeof(entropy)) {
 		if (!arch_get_random_seed_long_early(&entropy) &&
 		    !arch_get_random_long_early(&entropy)) {
 			entropy = random_get_entropy();
-			arch_bytes -= sizeof(entropy);
+			arch_bits -= sizeof(entropy) * 8;
 		}
 		_mix_pool_bytes(&entropy, sizeof(entropy));
 	}
@@ -834,7 +834,7 @@ int __init random_init(const char *comma
 	if (crng_ready())
 		crng_reseed();
 	else if (trust_cpu)
-		_credit_init_bits(arch_bytes * 8);
+		_credit_init_bits(arch_bits);
 
 	return 0;
 }
author maintainer
#181 lore raw

PATCH 181/234 powerpc/kasan: Silence KASAN warnings in __get_wchan()

In reply to Greg Kroah-Hartman

From: He Ying <[email protected]>

[ Upstream commit a1b29ba2f2c171b9bea73be993bfdf0a62d37d15 ]

The following KASAN warning was reported in our kernel.

  BUG: KASAN: stack-out-of-bounds in get_wchan+0x188/0x250
  Read of size 4 at addr d216f958 by task ps/14437

  CPU: 3 PID: 14437 Comm: ps Tainted: G           O      5.10.0 #1
  Call Trace:
  [daa63858] [c0654348] dump_stack+0x9c/0xe4 (unreliable)
  [daa63888] [c035cf0c] print_address_description.constprop.3+0x8c/0x570
  [daa63908] [c035d6bc] kasan_report+0x1ac/0x218
  [daa63948] [c00496e8] get_wchan+0x188/0x250
  [daa63978] [c0461ec8] do_task_stat+0xce8/0xe60
  [daa63b98] [c0455ac8] proc_single_show+0x98/0x170
  [daa63bc8] [c03cab8c] seq_read_iter+0x1ec/0x900
  [daa63c38] [c03cb47c] seq_read+0x1dc/0x290
  [daa63d68] [c037fc94] vfs_read+0x164/0x510
  [daa63ea8] [c03808e4] ksys_read+0x144/0x1d0
  [daa63f38] [c005b1dc] ret_from_syscall+0x0/0x38
  --- interrupt: c00 at 0x8fa8f4
      LR = 0x8fa8cc

  The buggy address belongs to the page:
  page:98ebcdd2 refcount:0 mapcount:0 mapping:00000000 index:0x2 pfn:0x1216f
  flags: 0x0()
  raw: 00000000 00000000 01010122 00000000 00000002 00000000 ffffffff 00000000
  raw: 00000000
  page dumped because: kasan: bad access detected

  Memory state around the buggy address:
   d216f800: 00 00 00 00 00 f1 f1 f1 f1 00 00 00 00 00 00 00
   d216f880: f2 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
  >d216f900: 00 00 00 00 00 00 00 00 00 00 00 f1 f1 f1 f1 00
                                            ^
   d216f980: f2 f2 f2 f2 f2 f2 f2 00 00 00 00 00 00 00 00 00
   d216fa00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

After looking into this issue, I find the buggy address belongs to the task stack region. It seems KASAN has something wrong. I look into the code of __get_wchan in x86 architecture and find the same issue has been resolved by the commit f7d27c35ddff ("x86/mm, kasan: Silence KASAN warnings in get_wchan()"). The solution could be applied to powerpc architecture too.

As Andrey Ryabinin said, get_wchan() is racy by design, it may access volatile stack of running task, thus it may access redzone in a stack frame and cause KASAN to warn about this.

Use READ_ONCE_NOCHECK() to silence these warnings.

---
 arch/powerpc/kernel/process.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/kernel/process.c +2 -2
index 02b69a68139c..56c33285b1df 100644
@@ -2017,12 +2017,12 @@ unsigned long get_wchan(struct task_struct *p)
 		return 0;
 
 	do {
-		sp = *(unsigned long *)sp;
+		sp = READ_ONCE_NOCHECK(*(unsigned long *)sp);
 		if (!validate_sp(sp, p, STACK_FRAME_OVERHEAD) ||
 		    p->state == TASK_RUNNING)
 			return 0;
 		if (count > 0) {
-			ip = ((unsigned long *)sp)[STACK_FRAME_LR_SAVE];
+			ip = READ_ONCE_NOCHECK(((unsigned long *)sp)[STACK_FRAME_LR_SAVE]);
 			if (!in_sched_functions(ip))
 				return ip;
 		}
Signature · 1 line
2.35.1
author maintainer
#182 lore raw

PATCH 182/234 ASoC: cs42l52: Fix TLV scales for mixer controls

In reply to Greg Kroah-Hartman

From: Charles Keepax <[email protected]>

[ Upstream commit 8bf5aabf524eec61013e506f764a0b2652dc5665 ]

The datasheet specifies the range of the mixer volumes as between -51.5dB and 12dB with a 0.5dB step. Update the TLVs for this.

---
 sound/soc/codecs/cs42l52.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
sound/soc/codecs/cs42l52.c +2 -2
index 3d83c1be1292..9a3180e71bd8 100644
@@ -141,7 +141,7 @@ static DECLARE_TLV_DB_SCALE(mic_tlv, 1600, 100, 0);
 
 static DECLARE_TLV_DB_SCALE(pga_tlv, -600, 50, 0);
 
-static DECLARE_TLV_DB_SCALE(mix_tlv, -50, 50, 0);
+static DECLARE_TLV_DB_SCALE(mix_tlv, -5150, 50, 0);
 
 static DECLARE_TLV_DB_SCALE(beep_tlv, -56, 200, 0);
 
@@ -368,7 +368,7 @@ static const struct snd_kcontrol_new cs42l52_snd_controls[] = {
 			      CS42L52_ADCB_VOL, 0, 0xA0, 0x78, ipd_tlv),
 	SOC_DOUBLE_R_SX_TLV("ADC Mixer Volume",
 			     CS42L52_ADCA_MIXER_VOL, CS42L52_ADCB_MIXER_VOL,
-				0, 0x19, 0x7F, ipd_tlv),
+				0, 0x19, 0x7F, mix_tlv),
 
 	SOC_DOUBLE("ADC Switch", CS42L52_ADC_MISC_CTL, 0, 1, 1, 0),
 
Signature · 1 line
2.35.1
author maintainer
#183 lore raw

PATCH 183/234 ASoC: cs53l30: Correct number of volume levels on SX controls

In reply to Greg Kroah-Hartman

From: Charles Keepax <[email protected]>

[ Upstream commit 7fbd6dd68127927e844912a16741016d432a0737 ]

This driver specified the maximum value rather than the number of volume levels on the SX controls, this is incorrect, so correct them.

---
 sound/soc/codecs/cs53l30.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
sound/soc/codecs/cs53l30.c +8 -8
index 8995ea45b4ca..86e93904b001 100644
@@ -351,22 +351,22 @@ static const struct snd_kcontrol_new cs53l30_snd_controls[] = {
 	SOC_ENUM("ADC2 NG Delay", adc2_ng_delay_enum),
 
 	SOC_SINGLE_SX_TLV("ADC1A PGA Volume",
-		    CS53L30_ADC1A_AFE_CTL, 0, 0x34, 0x18, pga_tlv),
+		    CS53L30_ADC1A_AFE_CTL, 0, 0x34, 0x24, pga_tlv),
 	SOC_SINGLE_SX_TLV("ADC1B PGA Volume",
-		    CS53L30_ADC1B_AFE_CTL, 0, 0x34, 0x18, pga_tlv),
+		    CS53L30_ADC1B_AFE_CTL, 0, 0x34, 0x24, pga_tlv),
 	SOC_SINGLE_SX_TLV("ADC2A PGA Volume",
-		    CS53L30_ADC2A_AFE_CTL, 0, 0x34, 0x18, pga_tlv),
+		    CS53L30_ADC2A_AFE_CTL, 0, 0x34, 0x24, pga_tlv),
 	SOC_SINGLE_SX_TLV("ADC2B PGA Volume",
-		    CS53L30_ADC2B_AFE_CTL, 0, 0x34, 0x18, pga_tlv),
+		    CS53L30_ADC2B_AFE_CTL, 0, 0x34, 0x24, pga_tlv),
 
 	SOC_SINGLE_SX_TLV("ADC1A Digital Volume",
-		    CS53L30_ADC1A_DIG_VOL, 0, 0xA0, 0x0C, dig_tlv),
+		    CS53L30_ADC1A_DIG_VOL, 0, 0xA0, 0x6C, dig_tlv),
 	SOC_SINGLE_SX_TLV("ADC1B Digital Volume",
-		    CS53L30_ADC1B_DIG_VOL, 0, 0xA0, 0x0C, dig_tlv),
+		    CS53L30_ADC1B_DIG_VOL, 0, 0xA0, 0x6C, dig_tlv),
 	SOC_SINGLE_SX_TLV("ADC2A Digital Volume",
-		    CS53L30_ADC2A_DIG_VOL, 0, 0xA0, 0x0C, dig_tlv),
+		    CS53L30_ADC2A_DIG_VOL, 0, 0xA0, 0x6C, dig_tlv),
 	SOC_SINGLE_SX_TLV("ADC2B Digital Volume",
-		    CS53L30_ADC2B_DIG_VOL, 0, 0xA0, 0x0C, dig_tlv),
+		    CS53L30_ADC2B_DIG_VOL, 0, 0xA0, 0x6C, dig_tlv),
 };
 
 static const struct snd_soc_dapm_widget cs53l30_dapm_widgets[] = {
Signature · 1 line
2.35.1
author maintainer
#184 lore raw

PATCH 184/234 ASoC: cs42l52: Correct TLV for Bypass Volume

In reply to Greg Kroah-Hartman

From: Charles Keepax <[email protected]>

[ Upstream commit 91e90c712fade0b69cdff7cc6512f6099bd18ae5 ]

The Bypass Volume is accidentally using a -6dB minimum TLV rather than the correct -60dB minimum. Add a new TLV to correct this.

---
 sound/soc/codecs/cs42l52.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
sound/soc/codecs/cs42l52.c +3 -1
index 9a3180e71bd8..de311299432b 100644
@@ -141,6 +141,8 @@ static DECLARE_TLV_DB_SCALE(mic_tlv, 1600, 100, 0);
 
 static DECLARE_TLV_DB_SCALE(pga_tlv, -600, 50, 0);
 
+static DECLARE_TLV_DB_SCALE(pass_tlv, -6000, 50, 0);
+
 static DECLARE_TLV_DB_SCALE(mix_tlv, -5150, 50, 0);
 
 static DECLARE_TLV_DB_SCALE(beep_tlv, -56, 200, 0);
@@ -355,7 +357,7 @@ static const struct snd_kcontrol_new cs42l52_snd_controls[] = {
 			      CS42L52_SPKB_VOL, 0, 0x40, 0xC0, hl_tlv),
 
 	SOC_DOUBLE_R_SX_TLV("Bypass Volume", CS42L52_PASSTHRUA_VOL,
-			      CS42L52_PASSTHRUB_VOL, 0, 0x88, 0x90, pga_tlv),
+			      CS42L52_PASSTHRUB_VOL, 0, 0x88, 0x90, pass_tlv),
 
 	SOC_DOUBLE("Bypass Mute", CS42L52_MISC_CTL, 4, 5, 1, 0),
 
Signature · 1 line
2.35.1
author maintainer
#185 lore raw

PATCH 185/234 ASoC: cs42l56: Correct typo in minimum level for SX volume controls

In reply to Greg Kroah-Hartman

From: Charles Keepax <[email protected]>

[ Upstream commit a8928ada9b96944cadd8b65d191e33199fd38782 ]

A couple of the SX volume controls specify 0x84 as the lowest volume value, however the correct value from the datasheet is 0x44. The datasheet don't include spaces in the value it displays as binary so this was almost certainly just a typo reading 1000100.

---
 sound/soc/codecs/cs42l56.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
sound/soc/codecs/cs42l56.c +2 -2
index 04f89b751304..deaad703a7db 100644
@@ -403,9 +403,9 @@ static const struct snd_kcontrol_new cs42l56_snd_controls[] = {
 	SOC_DOUBLE("ADC Boost Switch", CS42L56_GAIN_BIAS_CTL, 3, 2, 1, 1),
 
 	SOC_DOUBLE_R_SX_TLV("Headphone Volume", CS42L56_HPA_VOLUME,
-			      CS42L56_HPB_VOLUME, 0, 0x84, 0x48, hl_tlv),
+			      CS42L56_HPB_VOLUME, 0, 0x44, 0x48, hl_tlv),
 	SOC_DOUBLE_R_SX_TLV("LineOut Volume", CS42L56_LOA_VOLUME,
-			      CS42L56_LOB_VOLUME, 0, 0x84, 0x48, hl_tlv),
+			      CS42L56_LOB_VOLUME, 0, 0x44, 0x48, hl_tlv),
 
 	SOC_SINGLE_TLV("Bass Shelving Volume", CS42L56_TONE_CTL,
 			0, 0x00, 1, tone_tlv),
Signature · 1 line
2.35.1
author maintainer
#186 lore raw

PATCH 186/234 ata: libata-core: fix NULL pointer deref in ata_host_alloc_pinfo()

In reply to Greg Kroah-Hartman

From: Sergey Shtylyov <[email protected]>

[ Upstream commit bf476fe22aa1851bab4728e0c49025a6a0bea307 ]

In an unlikely (and probably wrong?) case that the 'ppi' parameter of ata_host_alloc_pinfo() points to an array starting with a NULL pointer, there's going to be a kernel oops as the 'pi' local variable won't get reassigned from the initial value of NULL. Initialize 'pi' instead to '&ata_dummy_port_info' to fix the possible kernel oops for good...

Found by Linux Verification Center (linuxtesting.org) with the SVACE static analysis tool.

---
 drivers/ata/libata-core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/ata/libata-core.c +2 -2
index 0c10d9557754..b0dea0702c74 100644
@@ -6253,7 +6253,7 @@ struct ata_host *ata_host_alloc_pinfo(struct device *dev,
 				      const struct ata_port_info * const * ppi,
 				      int n_ports)
 {
-	const struct ata_port_info *pi;
+	const struct ata_port_info *pi = &ata_dummy_port_info;
 	struct ata_host *host;
 	int i, j;
 
@@ -6261,7 +6261,7 @@ struct ata_host *ata_host_alloc_pinfo(struct device *dev,
 	if (!host)
 		return NULL;
 
-	for (i = 0, j = 0, pi = NULL; i < host->n_ports; i++) {
+	for (i = 0, j = 0; i < host->n_ports; i++) {
 		struct ata_port *ap = host->ports[i];
 
 		if (ppi[j])
Signature · 1 line
2.35.1
author maintainer
#187 lore raw

PATCH 187/234 ASoC: wm8962: Fix suspend while playing music

In reply to Greg Kroah-Hartman

From: Adam Ford <[email protected]>

[ Upstream commit d1f5272c0f7d2e53c6f2480f46725442776f5f78 ]

If the audio CODEC is playing sound when the system is suspended, it can be left in a state which throws the following error:

wm8962 3-001a: ASoC: error at soc_component_read_no_lock on wm8962.3-001a: -16

Once this error has occurred, the audio will not work again until rebooted.

Fix this by configuring SET_SYSTEM_SLEEP_PM_OPS.

---
 sound/soc/codecs/wm8962.c | 1 +
 1 file changed, 1 insertion(+)
sound/soc/codecs/wm8962.c +1 -0
index dde015fd70a4..3f75cb3209ff 100644
@@ -3861,6 +3861,7 @@ static int wm8962_runtime_suspend(struct device *dev)
 #endif
 
 static const struct dev_pm_ops wm8962_pm = {
+	SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, pm_runtime_force_resume)
 	SET_RUNTIME_PM_OPS(wm8962_runtime_suspend, wm8962_runtime_resume, NULL)
 };
 
Signature · 1 line
2.35.1
author maintainer
#188 lore raw

PATCH 188/234 ASoC: es8328: Fix event generation for deemphasis control

In reply to Greg Kroah-Hartman

From: Mark Brown <[email protected]>

[ Upstream commit 8259610c2ec01c5cbfb61882ae176aabacac9c19 ]

Currently the put() method for the deemphasis control returns 0 when a new value is written to the control even if the value changed, meaning events are not generated. Fix this, skip the work of updating the value when it is unchanged and then return 1 after having done so.

---
 sound/soc/codecs/es8328.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
sound/soc/codecs/es8328.c +4 -1
index 3afa163f7652..dcb01889e177 100644
@@ -165,13 +165,16 @@ static int es8328_put_deemph(struct snd_kcontrol *kcontrol,
 	if (deemph > 1)
 		return -EINVAL;
 
+	if (es8328->deemph == deemph)
+		return 0;
+
 	ret = es8328_set_deemph(component);
 	if (ret < 0)
 		return ret;
 
 	es8328->deemph = deemph;
 
-	return 0;
+	return 1;
 }
 
 
Signature · 1 line
2.35.1
author maintainer
#189 lore raw

PATCH 189/234 ASoC: wm_adsp: Fix event generation for wm_adsp_fw_put()

In reply to Greg Kroah-Hartman

From: Mark Brown <[email protected]>

[ Upstream commit 2abdf9f80019e8244d3806ed0e1c9f725e50b452 ]

Currently wm_adsp_fw_put() returns 0 rather than 1 when updating the value of the control, meaning that no event is generated to userspace. Fix this by setting the default return value to 1, the code already exits early with a return value of 0 if the value is unchanged.

---
 sound/soc/codecs/wm_adsp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/soc/codecs/wm_adsp.c +1 -1
index 02c557e1f779..c5b0b56d9c94 100644
@@ -697,7 +697,7 @@ int wm_adsp_fw_put(struct snd_kcontrol *kcontrol,
 	struct snd_soc_component *component = snd_soc_kcontrol_component(kcontrol);
 	struct soc_enum *e = (struct soc_enum *)kcontrol->private_value;
 	struct wm_adsp *dsp = snd_soc_component_get_drvdata(component);
-	int ret = 0;
+	int ret = 1;
 
 	if (ucontrol->value.enumerated.item[0] == dsp[e->shift_l].fw)
 		return 0;
Signature · 1 line
2.35.1
author maintainer
#190 lore raw

PATCH 190/234 scsi: vmw_pvscsi: Expand vcpuHint to 16 bits

In reply to Greg Kroah-Hartman

From: Wentao Wang <[email protected]>

[ Upstream commit cf71d59c2eceadfcde0fb52e237990a0909880d7 ]

vcpuHint has been expanded to 16 bit on host to enable routing to more CPUs. Guest side should align with the change. This change has been tested with hosts with 8-bit and 16-bit vcpuHint, on both platforms host side can get correct value.

---
 drivers/scsi/vmw_pvscsi.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/scsi/vmw_pvscsi.h +2 -2
index 75966d3f326e..d87c12324c03 100644
@@ -333,8 +333,8 @@ struct PVSCSIRingReqDesc {
 	u8	tag;
 	u8	bus;
 	u8	target;
-	u8	vcpuHint;
-	u8	unused[59];
+	u16	vcpuHint;
+	u8	unused[58];
 } __packed;
 
 /*
Signature · 1 line
2.35.1
author maintainer
#191 lore raw

PATCH 191/234 scsi: lpfc: Fix port stuck in bypassed state after LIP in PT2PT topology

In reply to Greg Kroah-Hartman

From: James Smart <[email protected]>

[ Upstream commit 336d63615466b4c06b9401c987813fd19bdde39b ]

After issuing a LIP, a specific target vendor does not ACC the FLOGI that lpfc sends. However, it does send its own FLOGI that lpfc ACCs. The target then establishes the port IDs by sending a PLOGI. lpfc PLOGI_ACCs and starts the RPI registration for DID 0x000001. The target then sends a LOGO to the fabric DID. lpfc is currently treating the LOGO from the fabric DID as a link down and cleans up all the ndlps. The ndlp for DID 0x000001 is put back into NPR and discovery stops, leaving the port in stuck in bypassed mode.

Change lpfc behavior such that if a LOGO is received for the fabric DID in PT2PT topology skip the lpfc_linkdown_port() routine and just move the fabric DID back to NPR.

---
 drivers/scsi/lpfc/lpfc_nportdisc.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/lpfc/lpfc_nportdisc.c +2 -1
index 9442fb30e7cd..f666518d84b0 100644
@@ -662,7 +662,8 @@ lpfc_rcv_logo(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp,
 	else
 		lpfc_els_rsp_acc(vport, ELS_CMD_ACC, cmdiocb, ndlp, NULL);
 	if (ndlp->nlp_DID == Fabric_DID) {
-		if (vport->port_state <= LPFC_FDISC)
+		if (vport->port_state <= LPFC_FDISC ||
+		    vport->fc_flag & FC_PT2PT)
 			goto out;
 		lpfc_linkdown_port(vport);
 		spin_lock_irq(shost->host_lock);
Signature · 1 line
2.35.1
author maintainer
#192 lore raw

PATCH 192/234 scsi: ipr: Fix missing/incorrect resource cleanup in error case

In reply to Greg Kroah-Hartman

From: Chengguang Xu <[email protected]>

[ Upstream commit d64c491911322af1dcada98e5b9ee0d87e8c8fee ]

Fix missing resource cleanup (when '(--i) == 0') for error case in ipr_alloc_mem() and skip incorrect resource cleanup (when '(--i) == 0') for error case in ipr_request_other_msi_irqs() because variable i started from 1.

---
 drivers/scsi/ipr.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/scsi/ipr.c +2 -2
index 1b04a8223eb0..c6cde552b995 100644
@@ -9783,7 +9783,7 @@ static int ipr_alloc_mem(struct ipr_ioa_cfg *ioa_cfg)
 					GFP_KERNEL);
 
 		if (!ioa_cfg->hrrq[i].host_rrq)  {
-			while (--i > 0)
+			while (--i >= 0)
 				dma_free_coherent(&pdev->dev,
 					sizeof(u32) * ioa_cfg->hrrq[i].size,
 					ioa_cfg->hrrq[i].host_rrq,
@@ -10056,7 +10056,7 @@ static int ipr_request_other_msi_irqs(struct ipr_ioa_cfg *ioa_cfg,
 			ioa_cfg->vectors_info[i].desc,
 			&ioa_cfg->hrrq[i]);
 		if (rc) {
-			while (--i >= 0)
+			while (--i > 0)
 				free_irq(pci_irq_vector(pdev, i),
 					&ioa_cfg->hrrq[i]);
 			return rc;
Signature · 1 line
2.35.1
author maintainer
#193 lore raw

PATCH 193/234 scsi: pmcraid: Fix missing resource cleanup in error case

In reply to Greg Kroah-Hartman

From: Chengguang Xu <[email protected]>

[ Upstream commit ec1e8adcbdf661c57c395bca342945f4f815add7 ]

Fix missing resource cleanup (when '(--i) == 0') for error case in pmcraid_register_interrupt_handler().

---
 drivers/scsi/pmcraid.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/pmcraid.c +1 -1
index 4e86994e10e8..6e96229c58e0 100644
@@ -4559,7 +4559,7 @@ pmcraid_register_interrupt_handler(struct pmcraid_instance *pinstance)
 	return 0;
 
 out_unwind:
-	while (--i > 0)
+	while (--i >= 0)
 		free_irq(pci_irq_vector(pdev, i), &pinstance->hrrq_vector[i]);
 	pci_free_irq_vectors(pdev);
 	return rc;
Signature · 1 line
2.35.1
author maintainer
#194 lore raw

PATCH 194/234 virtio-mmio: fix missing put_device() when vm_cmdline_parent registration failed

In reply to Greg Kroah-Hartman

From: chengkaitao <[email protected]>

[ Upstream commit a58a7f97ba11391d2d0d408e0b24f38d86ae748e ]

The reference must be released when device_register(&vm_cmdline_parent) failed. Add the corresponding 'put_device()' in the error handling path.

Message-Id: <[email protected]>

---
 drivers/virtio/virtio_mmio.c | 1 +
 1 file changed, 1 insertion(+)
drivers/virtio/virtio_mmio.c +1 -0
index 4cd9ea5c75be..c69c755bf553 100644
@@ -663,6 +663,7 @@ static int vm_cmdline_set(const char *device,
 	if (!vm_cmdline_parent_registered) {
 		err = device_register(&vm_cmdline_parent);
 		if (err) {
+			put_device(&vm_cmdline_parent);
 			pr_err("Failed to register parent device!\n");
 			return err;
 		}
Signature · 1 line
2.35.1
author maintainer
#195 lore raw

PATCH 195/234 nfc: nfcmrvl: Fix memory leak in nfcmrvl_play_deferred

In reply to Greg Kroah-Hartman

From: Xiaohui Zhang <[email protected]>

[ Upstream commit 8a4d480702b71184fabcf379b80bf7539716752e ]

Similar to the handling of play_deferred in commit 19cfe912c37b ("Bluetooth: btusb: Fix memory leak in play_deferred"), we thought a patch might be needed here as well.

Currently usb_submit_urb is called directly to submit deferred tx urbs after unanchor them.

So the usb_giveback_urb_bh would failed to unref it in usb_unanchor_urb and cause memory leak.

Put those urbs in tx_anchor to avoid the leak, and also fix the error handling.

---
 drivers/nfc/nfcmrvl/usb.c | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)
drivers/nfc/nfcmrvl/usb.c +14 -2
index 888e298f610b..f26986eb53f1 100644
@@ -401,13 +401,25 @@ static void nfcmrvl_play_deferred(struct nfcmrvl_usb_drv_data *drv_data)
 	int err;
 
 	while ((urb = usb_get_from_anchor(&drv_data->deferred))) {
+		usb_anchor_urb(urb, &drv_data->tx_anchor);
+
 		err = usb_submit_urb(urb, GFP_ATOMIC);
-		if (err)
+		if (err) {
+			kfree(urb->setup_packet);
+			usb_unanchor_urb(urb);
+			usb_free_urb(urb);
 			break;
+		}
 
 		drv_data->tx_in_flight++;
+		usb_free_urb(urb);
+	}
+
+	/* Cleanup the rest deferred urbs. */
+	while ((urb = usb_get_from_anchor(&drv_data->deferred))) {
+		kfree(urb->setup_packet);
+		usb_free_urb(urb);
 	}
-	usb_scuttle_anchored_urbs(&drv_data->deferred);
 }
 
 static int nfcmrvl_resume(struct usb_interface *intf)
Signature · 1 line
2.35.1
author maintainer
#196 lore raw

PATCH 196/234 ipv6: Fix signed integer overflow in l2tp_ip6_sendmsg

In reply to Greg Kroah-Hartman

From: Wang Yufen <[email protected]>

[ Upstream commit f638a84afef3dfe10554c51820c16e39a278c915 ]

When len >= INT_MAX - transhdrlen, ulen = len + transhdrlen will be overflow. To fix, we can follow what udpv6 does and subtract the transhdrlen from the max.

---
 net/l2tp/l2tp_ip6.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
net/l2tp/l2tp_ip6.c +3 -2
index 2ff25c445b82..9dae10d8880c 100644
@@ -519,14 +519,15 @@ static int l2tp_ip6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	struct ipcm6_cookie ipc6;
 	int addr_len = msg->msg_namelen;
 	int transhdrlen = 4; /* zero session-id */
-	int ulen = len + transhdrlen;
+	int ulen;
 	int err;
 
 	/* Rough check on arithmetic overflow,
 	   better check is made in ip6_append_data().
 	 */
-	if (len > INT_MAX)
+	if (len > INT_MAX - transhdrlen)
 		return -EMSGSIZE;
+	ulen = len + transhdrlen;
 
 	/* Mirror BSD error message compatibility */
 	if (msg->msg_flags & MSG_OOB)
Signature · 1 line
2.35.1
author maintainer
#197 lore raw

PATCH 197/234 net: ethernet: mtk_eth_soc: fix misuse of mem alloc interface netdev[napi]_alloc_frag

In reply to Greg Kroah-Hartman

From: Chen Lin <[email protected]>

[ Upstream commit 2f2c0d2919a14002760f89f4e02960c735a316d2 ]

When rx_flag == MTK_RX_FLAGS_HWLRO,
rx_data_len = MTK_MAX_LRO_RX_LENGTH(4096 * 3) > PAGE_SIZE. netdev_alloc_frag is for alloction of page fragment only. Reference to other drivers and Documentation/vm/page_frags.rst

Branch to use __get_free_pages when ring->frag_size > PAGE_SIZE.

---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)
drivers/net/ethernet/mediatek/mtk_eth_soc.c +19 -2
index f2eaf8c13cc2..53cff913abf0 100644
@@ -597,6 +597,17 @@ static inline void mtk_rx_get_desc(struct mtk_rx_dma *rxd,
 	rxd->rxd4 = READ_ONCE(dma_rxd->rxd4);
 }
 
+static void *mtk_max_lro_buf_alloc(gfp_t gfp_mask)
+{
+	unsigned int size = mtk_max_frag_size(MTK_MAX_LRO_RX_LENGTH);
+	unsigned long data;
+
+	data = __get_free_pages(gfp_mask | __GFP_COMP | __GFP_NOWARN,
+				get_order(size));
+
+	return (void *)data;
+}
+
 /* the qdma core needs scratch memory to be setup */
 static int mtk_init_fq_dma(struct mtk_eth *eth)
 {
@@ -1005,7 +1016,10 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget,
 			goto release_desc;
 
 		/* alloc new buffer */
-		new_data = napi_alloc_frag(ring->frag_size);
+		if (ring->frag_size <= PAGE_SIZE)
+			new_data = napi_alloc_frag(ring->frag_size);
+		else
+			new_data = mtk_max_lro_buf_alloc(GFP_ATOMIC);
 		if (unlikely(!new_data)) {
 			netdev->stats.rx_dropped++;
 			goto release_desc;
@@ -1312,7 +1326,10 @@ static int mtk_rx_alloc(struct mtk_eth *eth, int ring_no, int rx_flag)
 		return -ENOMEM;
 
 	for (i = 0; i < rx_dma_size; i++) {
-		ring->data[i] = netdev_alloc_frag(ring->frag_size);
+		if (ring->frag_size <= PAGE_SIZE)
+			ring->data[i] = netdev_alloc_frag(ring->frag_size);
+		else
+			ring->data[i] = mtk_max_lro_buf_alloc(GFP_KERNEL);
 		if (!ring->data[i])
 			return -ENOMEM;
 	}
Signature · 1 line
2.35.1
author maintainer
#198 lore raw

PATCH 198/234 random: credit cpu and bootloader seeds by default

In reply to Greg Kroah-Hartman

From: Jason A. Donenfeld <[email protected]>

[ Upstream commit 846bb97e131d7938847963cca00657c995b1fce1 ]

This commit changes the default Kconfig values of RANDOM_TRUST_CPU and RANDOM_TRUST_BOOTLOADER to be Y by default. It does not change any existing configs or change any kernel behavior. The reason for this is several fold.

As background, I recently had an email thread with the kernel maintainers of Fedora/RHEL, Debian, Ubuntu, Gentoo, Arch, NixOS, Alpine, SUSE, and Void as recipients. I noted that some distros trust RDRAND, some trust EFI, and some trust both, and I asked why or why not. There wasn't really much of a "debate" but rather an interesting discussion of what the historical reasons have been for this, and it came up that some distros just missed the introduction of the bootloader Kconfig knob, while another didn't want to enable it until there was a boot time switch to turn it off for more concerned users (which has since been added). The result of the rather uneventful discussion is that every major Linux distro enables these two options by default.

While I didn't have really too strong of an opinion going into this thread -- and I mostly wanted to learn what the distros' thinking was one way or another -- ultimately I think their choice was a decent enough one for a default option (which can be disabled at boot time). I'll try to summarize the pros and cons:

Pros:

- The RNG machinery gets initialized super quickly, and there's no messing around with subsequent blocking behavior.

- The bootloader mechanism is used by kexec in order for the prior kernel to initialize the RNG of the next kernel, which increases the entropy available to early boot daemons of the next kernel.

- Previous objections related to backdoors centered around Dual_EC_DRBG-like kleptographic systems, in which observing some amount of the output stream enables an adversary holding the right key to determine the entire output stream.

This used to be a partially justified concern, because RDRAND output was mixed into the output stream in varying ways, some of which may have lacked pre-image resistance (e.g. XOR or an LFSR).

But this is no longer the case. Now, all usage of RDRAND and bootloader seeds go through a cryptographic hash function. This means that the CPU would have to compute a hash pre-image, which is not considered to be feasible (otherwise the hash function would be terribly broken).

- More generally, if the CPU is backdoored, the RNG is probably not the realistic vector of choice for an attacker.

- These CPU or bootloader seeds are far from being the only source of entropy. Rather, there is generally a pretty huge amount of entropy, not all of which is credited, especially on CPUs that support instructions like RDRAND. In other words, assuming RDRAND outputs all zeros, an attacker would *still* have to accurately model every single other entropy source also in use.

- The RNG now reseeds itself quite rapidly during boot, starting at 2 seconds, then 4, then 8, then 16, and so forth, so that other sources of entropy get used without much delay.

- Paranoid users can set random.trust_{cpu,bootloader}=no in the kernel command line, and paranoid system builders can set the Kconfig options to N, so there's no reduction or restriction of optionality.

- It's a practical default.

- All the distros have it set this way. Microsoft and Apple trust it too. Bandwagon.

Cons:

- RDRAND *could* still be backdoored with something like a fixed key or limited space serial number seed or another indexable scheme like that. (However, it's hard to imagine threat models where the CPU is backdoored like this, yet people are still okay making *any* computations with it or connecting it to networks, etc.)

- RDRAND *could* be defective, rather than backdoored, and produce garbage that is in one way or another insufficient for crypto.

- Suggesting a *reduction* in paranoia, as this commit effectively does, may cause some to question my personal integrity as a "security person".

- Bootloader seeds and RDRAND are generally very difficult if not all together impossible to audit.

Keep in mind that this doesn't actually change any behavior. This is just a change in the default Kconfig value. The distros already are shipping kernels that set things this way.

Ard made an additional argument in [1]:

We're at the mercy of firmware and micro-architecture anyway, given that we are also relying on it to ensure that every instruction in the kernel's executable image has been faithfully copied to memory, and that the CPU implements those instructions as documented. So I don't think firmware or ISA bugs related to RNGs deserve special treatment - if they are broken, we should quirk around them like we usually do. So enabling these by default is a step in the right direction IMHO.

In [2], Phil pointed out that having this disabled masked a bug that CI otherwise would have caught:

A clean 5.15.45 boots cleanly, whereas a downstream kernel shows the static key warning (but it does go on to boot). The significant difference is that our defconfigs set CONFIG_RANDOM_TRUST_BOOTLOADER=y defining that on top of multi_v7_defconfig demonstrates the issue on a clean 5.15.45. Conversely, not setting that option in a downstream kernel build avoids the warning

[1] https://lore.kernel.org/lkml/CAMj1kXGi+ieviFjXv9zQBSaGyyzeGW_VpMpTLJK8PJb2QHEQ-w@mail.gmail.com/
[2] https://lore.kernel.org/lkml/[email protected]/

---
 drivers/char/Kconfig | 54 +++++++++++++++++++++++++++-----------------
 1 file changed, 33 insertions(+), 21 deletions(-)
drivers/char/Kconfig +33 -21
index 094548183f8b..f55c9bbd58fb 100644
@@ -552,29 +552,41 @@ config ADI
 	  and SSM (Silicon Secured Memory).  Intended consumers of this
 	  driver include crash and makedumpfile.
 
-endmenu
-
 config RANDOM_TRUST_CPU
-	bool "Trust the CPU manufacturer to initialize Linux's CRNG"
-	depends on X86 || S390 || PPC
-	default n
+	bool "Initialize RNG using CPU RNG instructions"
+	default y
+	depends on ARCH_RANDOM
 	help
-	Assume that CPU manufacturer (e.g., Intel or AMD for RDSEED or
-	RDRAND, IBM for the S390 and Power PC architectures) is trustworthy
-	for the purposes of initializing Linux's CRNG.  Since this is not
-	something that can be independently audited, this amounts to trusting
-	that CPU manufacturer (perhaps with the insistence or mandate
-	of a Nation State's intelligence or law enforcement agencies)
-	has not installed a hidden back door to compromise the CPU's
-	random number generation facilities. This can also be configured
-	at boot with "random.trust_cpu=on/off".
+	  Initialize the RNG using random numbers supplied by the CPU's
+	  RNG instructions (e.g. RDRAND), if supported and available. These
+	  random numbers are never used directly, but are rather hashed into
+	  the main input pool, and this happens regardless of whether or not
+	  this option is enabled. Instead, this option controls whether the
+	  they are credited and hence can initialize the RNG. Additionally,
+	  other sources of randomness are always used, regardless of this
+	  setting.  Enabling this implies trusting that the CPU can supply high
+	  quality and non-backdoored random numbers.
+
+	  Say Y here unless you have reason to mistrust your CPU or believe
+	  its RNG facilities may be faulty. This may also be configured at
+	  boot time with "random.trust_cpu=on/off".
 
 config RANDOM_TRUST_BOOTLOADER
-	bool "Trust the bootloader to initialize Linux's CRNG"
+	bool "Initialize RNG using bootloader-supplied seed"
+	default y
 	help
-	Some bootloaders can provide entropy to increase the kernel's initial
-	device randomness. Say Y here to assume the entropy provided by the
-	booloader is trustworthy so it will be added to the kernel's entropy
-	pool. Otherwise, say N here so it will be regarded as device input that
-	only mixes the entropy pool. This can also be configured at boot with
-	"random.trust_bootloader=on/off".
+	  Initialize the RNG using a seed supplied by the bootloader or boot
+	  environment (e.g. EFI or a bootloader-generated device tree). This
+	  seed is not used directly, but is rather hashed into the main input
+	  pool, and this happens regardless of whether or not this option is
+	  enabled. Instead, this option controls whether the seed is credited
+	  and hence can initialize the RNG. Additionally, other sources of
+	  randomness are always used, regardless of this setting. Enabling
+	  this implies trusting that the bootloader can supply high quality and
+	  non-backdoored seeds.
+
+	  Say Y here unless you have reason to mistrust your bootloader or
+	  believe its RNG facilities may be faulty. This may also be configured
+	  at boot time with "random.trust_bootloader=on/off".
+
+endmenu
Signature · 1 line
2.35.1
author maintainer
#199 lore raw

PATCH 199/234 pNFS: Dont keep retrying if the server replied NFS4ERR_LAYOUTUNAVAILABLE

In reply to Greg Kroah-Hartman

From: Trond Myklebust <[email protected]>

[ Upstream commit fe44fb23d6ccde4c914c44ef74ab8d9d9ba02bea ]

If the server tells us that a pNFS layout is not available for a specific file, then we should not keep pounding it with further layoutget requests.

Fixes 183d9e7b112a ("pnfs: rework LAYOUTGET retry handling")
---
 fs/nfs/pnfs.c | 6 ++++++
 1 file changed, 6 insertions(+)
fs/nfs/pnfs.c +6 -0
index a7d638bfb46b..cfb1fe5dfb1e 100644
@@ -2045,6 +2045,12 @@ pnfs_update_layout(struct inode *ino,
 		case -ERECALLCONFLICT:
 		case -EAGAIN:
 			break;
+		case -ENODATA:
+			/* The server returned NFS4ERR_LAYOUTUNAVAILABLE */
+			pnfs_layout_set_fail_bit(
+				lo, pnfs_iomode_to_fail_bit(iomode));
+			lseg = NULL;
+			goto out_put_layout_hdr;
 		default:
 			if (!nfs_error_is_fatal(PTR_ERR(lseg))) {
 				pnfs_layout_clear_fail_bit(lo, pnfs_iomode_to_fail_bit(iomode));
Signature · 1 line
2.35.1
author maintainer
#200 lore raw

PATCH 200/234 i40e: Fix adding ADQ filter to TC0

In reply to Greg Kroah-Hartman

From: Grzegorz Szczurek <[email protected]>

[ Upstream commit c3238d36c3a2be0a29a9d848d6c51e1b14be6692 ]

Procedure of configure tc flower filters erroneously allows to create filters on TC0 where unfiltered packets are also directed by default. Issue was caused by insufficient checks of hw_tc parameter specifying the hardware traffic class to pass matching packets to.

Fix checking hw_tc parameter which blocks creation of filters on TC0.

Fixes 2f4b411a3d67 ("i40e: Enable cloud filters via tc-flower")
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/net/ethernet/intel/i40e/i40e_main.c +5 -0
index 21ea0cdea666..3615c6533cf4 100644
@@ -7508,6 +7508,11 @@ static int i40e_configure_clsflower(struct i40e_vsi *vsi,
 		return -EOPNOTSUPP;
 	}
 
+	if (!tc) {
+		dev_err(&pf->pdev->dev, "Unable to add filter because of invalid destination");
+		return -EINVAL;
+	}
+
 	if (test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state) ||
 	    test_bit(__I40E_RESET_INTR_RECEIVED, pf->state))
 		return -EBUSY;
Signature · 1 line
2.35.1
author maintainer
#201 lore raw

PATCH 201/234 i40e: Fix call trace in setup_tx_descriptors

In reply to Greg Kroah-Hartman

From: Aleksandr Loktionov <[email protected]>

[ Upstream commit fd5855e6b1358e816710afee68a1d2bc685176ca ]

After PF reset and ethtool -t there was call trace in dmesg sometimes leading to panic. When there was some time, around 5 seconds, between reset and test there were no errors.

Problem was that pf reset calls i40e_vsi_close in prep_for_reset and ethtool -t calls i40e_vsi_close in diag_test. If there was not enough time between those commands the second i40e_vsi_close starts before previous i40e_vsi_close was done which leads to crash.

Add check to diag_test if pf is in reset and don't start offline tests if it is true.
Add netif_info("testing failed") into unhappy path of i40e_diag_test()

Fixes e17bc411aea8 ("i40e: Disable offline diagnostics if VFs are enabled")
Fixes 510efb2682b3 ("i40e: Fix ethtool offline diagnostic with netqueues")
---
 .../net/ethernet/intel/i40e/i40e_ethtool.c    | 25 +++++++++++++------
 1 file changed, 17 insertions(+), 8 deletions(-)
drivers/net/ethernet/intel/i40e/i40e_ethtool.c +17 -8
index 5242d3dfeb22..6a70e62836f8 100644
@@ -2195,15 +2195,16 @@ static void i40e_diag_test(struct net_device *netdev,
 
 		set_bit(__I40E_TESTING, pf->state);
 
+		if (test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state) ||
+		    test_bit(__I40E_RESET_INTR_RECEIVED, pf->state)) {
+			dev_warn(&pf->pdev->dev,
+				 "Cannot start offline testing when PF is in reset state.\n");
+			goto skip_ol_tests;
+		}
+
 		if (i40e_active_vfs(pf) || i40e_active_vmdqs(pf)) {
 			dev_warn(&pf->pdev->dev,
 				 "Please take active VFs and Netqueues offline and restart the adapter before running NIC diagnostics\n");
-			data[I40E_ETH_TEST_REG]		= 1;
-			data[I40E_ETH_TEST_EEPROM]	= 1;
-			data[I40E_ETH_TEST_INTR]	= 1;
-			data[I40E_ETH_TEST_LINK]	= 1;
-			eth_test->flags |= ETH_TEST_FL_FAILED;
-			clear_bit(__I40E_TESTING, pf->state);
 			goto skip_ol_tests;
 		}
 
@@ -2250,9 +2251,17 @@ static void i40e_diag_test(struct net_device *netdev,
 		data[I40E_ETH_TEST_INTR] = 0;
 	}
 
-skip_ol_tests:
-
 	netif_info(pf, drv, netdev, "testing finished\n");
+	return;
+
+skip_ol_tests:
+	data[I40E_ETH_TEST_REG]		= 1;
+	data[I40E_ETH_TEST_EEPROM]	= 1;
+	data[I40E_ETH_TEST_INTR]	= 1;
+	data[I40E_ETH_TEST_LINK]	= 1;
+	eth_test->flags |= ETH_TEST_FL_FAILED;
+	clear_bit(__I40E_TESTING, pf->state);
+	netif_info(pf, drv, netdev, "testing failed\n");
 }
 
 static void i40e_get_wol(struct net_device *netdev,
Signature · 1 line
2.35.1
author maintainer
#202 lore raw

PATCH 202/234 tty: goldfish: Fix free_irq() on remove

In reply to Greg Kroah-Hartman

From: Vincent Whitchurch <[email protected]>

[ Upstream commit 499e13aac6c762e1e828172b0f0f5275651d6512 ]

Pass the correct dev_id to free_irq() to fix this splat when the driver is unbound:

 WARNING: CPU: 0 PID: 30 at kernel/irq/manage.c:1895 free_irq
 Trying to free already-free IRQ 65
 Call Trace:
  warn_slowpath_fmt
  free_irq
  goldfish_tty_remove
  platform_remove
  device_remove
  device_release_driver_internal
  device_driver_detach
  unbind_store
  drv_attr_store
  ...
Fixes 465893e18878e119 ("tty: goldfish: support platform_device with id -1")
---
 drivers/tty/goldfish.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/tty/goldfish.c +1 -1
index abc84d84f638..9180ca5e4dcd 100644
@@ -428,7 +428,7 @@ static int goldfish_tty_remove(struct platform_device *pdev)
 	tty_unregister_device(goldfish_tty_driver, qtty->console.index);
 	iounmap(qtty->base);
 	qtty->base = NULL;
-	free_irq(qtty->irq, pdev);
+	free_irq(qtty->irq, qtty);
 	tty_port_destroy(&qtty->port);
 	goldfish_tty_current_line_count--;
 	if (goldfish_tty_current_line_count == 0)
Signature · 1 line
2.35.1
author maintainer
#203 lore raw

PATCH 203/234 misc: atmel-ssc: Fix IRQ check in ssc_probe

In reply to Greg Kroah-Hartman

From: Miaoqian Lin <[email protected]>

[ Upstream commit 1c245358ce0b13669f6d1625f7a4e05c41f28980 ]

platform_get_irq() returns negative error number instead 0 on failure. And the doc of platform_get_irq() provides a usage example:

    int irq = platform_get_irq(pdev, 0);
    if (irq < 0)
        return irq;

Fix the check of return value to catch errors correctly.

Fixes eb1f2930609b ("Driver for the Atmel on-chip SSC on AT32AP and AT91")
---
 drivers/misc/atmel-ssc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/misc/atmel-ssc.c +2 -2
index f9caf233e2cc..48521861beb5 100644
@@ -235,9 +235,9 @@ static int ssc_probe(struct platform_device *pdev)
 	clk_disable_unprepare(ssc->clk);
 
 	ssc->irq = platform_get_irq(pdev, 0);
-	if (!ssc->irq) {
+	if (ssc->irq < 0) {
 		dev_dbg(&pdev->dev, "could not get irq\n");
-		return -ENXIO;
+		return ssc->irq;
 	}
 
 	mutex_lock(&user_lock);
Signature · 1 line
2.35.1
author maintainer
#204 lore raw

PATCH 204/234 mlxsw: spectrum_cnt: Reorder counter pools

In reply to Greg Kroah-Hartman

From: Petr Machata <[email protected]>

[ Upstream commit 4b7a632ac4e7101ceefee8484d5c2ca505d347b3 ]

Both RIF and ACL flow counters use a 24-bit SW-managed counter address to communicate which counter they want to bind.

In a number of Spectrum FW releases, binding a RIF counter is broken and slices the counter index to 16 bits. As a result, on Spectrum-2 and above, no more than about 410 RIF counters can be effectively used. This translates to 205 netdevices for which L3 HW stats can be enabled. (This does not happen on Spectrum-1, because there are fewer counters available overall and the counter index never exceeds 16 bits.)

Binding counters to ACLs does not have this issue. Therefore reorder the counter allocation scheme so that RIF counters come first and therefore get lower indices that are below the 16-bit barrier.

Fixes 98e60dce4da1 ("Merge branch 'mlxsw-Introduce-initial-Spectrum-2-support'")
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.h +1 -1
index 81465e267b10..b7eb3674e285 100644
@@ -7,8 +7,8 @@
 #include "spectrum.h"
 
 enum mlxsw_sp_counter_sub_pool_id {
-	MLXSW_SP_COUNTER_SUB_POOL_FLOW,
 	MLXSW_SP_COUNTER_SUB_POOL_RIF,
+	MLXSW_SP_COUNTER_SUB_POOL_FLOW,
 };
 
 int mlxsw_sp_counter_alloc(struct mlxsw_sp *mlxsw_sp,
Signature · 1 line
2.35.1
author maintainer
#205 lore raw

PATCH 205/234 net: bgmac: Fix an erroneous kfree() in bgmac_remove()

In reply to Greg Kroah-Hartman

From: Christophe JAILLET <[email protected]>

[ Upstream commit d7dd6eccfbc95ac47a12396f84e7e1b361db654b ]

'bgmac' is part of a managed resource allocated with bgmac_alloc(). It should not be freed explicitly.

Remove the erroneous kfree() from the .remove() function.

---
 drivers/net/ethernet/broadcom/bgmac-bcma.c | 1 -
 1 file changed, 1 deletion(-)
drivers/net/ethernet/broadcom/bgmac-bcma.c +0 -1
index 6fe074c1588b..77de92eb08b2 100644
@@ -323,7 +323,6 @@ static void bgmac_remove(struct bcma_device *core)
 	bcma_mdio_mii_unregister(bgmac->mii_bus);
 	bgmac_enet_remove(bgmac);
 	bcma_set_drvdata(core, NULL);
-	kfree(bgmac);
 }
 
 static struct bcma_driver bgmac_bcma_driver = {
Signature · 1 line
2.35.1
author maintainer
#206 lore raw

PATCH 206/234 arm64: ftrace: fix branch range checks

In reply to Greg Kroah-Hartman

From: Mark Rutland <[email protected]>

[ Upstream commit 3eefdf9d1e406f3da47470b2854347009ffcb6fa ]

The branch range checks in ftrace_make_call() and ftrace_make_nop() are incorrect, erroneously permitting a forwards branch of 128M and erroneously rejecting a backwards branch of 128M.

This is because both functions calculate the offset backwards, calculating the offset *from* the target *to* the branch, rather than the other way around as the later comparisons expect.

If an out-of-range branch were erroeously permitted, this would later be rejected by aarch64_insn_gen_branch_imm() as branch_imm_common() checks the bounds correctly, resulting in warnings and the placement of a BRK instruction. Note that this can only happen for a forwards branch of exactly 128M, and so the caller would need to be exactly 128M bytes below the relevant ftrace trampoline.

If an in-range branch were erroeously rejected, then:

* For modules when CONFIG_ARM64_MODULE_PLTS=y, this would result in the use of a PLT entry, which is benign.

Note that this is the common case, as this is selected by CONFIG_RANDOMIZE_BASE (and therefore RANDOMIZE_MODULE_REGION_FULL), which distributions typically seelct. This is also selected by CONFIG_ARM64_ERRATUM_843419.

* For modules when CONFIG_ARM64_MODULE_PLTS=n, this would result in internal ftrace failures.

* For core kernel text, this would result in internal ftrace failues.

Note that for this to happen, the kernel text would need to be at least 128M bytes in size, and typical configurations are smaller tha this.

Fix this by calculating the offset *from* the branch *to* the target in both functions.

Fixes f8af0b364e24 ("arm64: ftrace: don't validate branch via PLT in ftrace_make_nop()")
Fixes e71a4e1bebaf ("arm64: ftrace: add support for far branches to dynamic ftrace")
Cc Ard Biesheuvel <[email protected]>
Cc Will Deacon <[email protected]>
---
 arch/arm64/kernel/ftrace.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/kernel/ftrace.c +2 -2
index b6618391be8c..4254d7808def 100644
@@ -72,7 +72,7 @@ int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr)
 {
 	unsigned long pc = rec->ip;
 	u32 old, new;
-	long offset = (long)pc - (long)addr;
+	long offset = (long)addr - (long)pc;
 
 	if (offset < -SZ_128M || offset >= SZ_128M) {
 #ifdef CONFIG_ARM64_MODULE_PLTS
@@ -151,7 +151,7 @@ int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec,
 	unsigned long pc = rec->ip;
 	bool validate = true;
 	u32 old = 0, new;
-	long offset = (long)pc - (long)addr;
+	long offset = (long)addr - (long)pc;
 
 	if (offset < -SZ_128M || offset >= SZ_128M) {
 #ifdef CONFIG_ARM64_MODULE_PLTS
Signature · 1 line
2.35.1
author maintainer
#207 lore raw

PATCH 207/234 certs/blacklist_hashes.c: fix const confusion in certs blacklist

In reply to Greg Kroah-Hartman

From: Masahiro Yamada <[email protected]>

[ Upstream commit 6a1c3767d82ed8233de1263aa7da81595e176087 ]

This file fails to compile as follows:

CC certs/blacklist_hashes.o
certs/blacklist_hashes.c:4:1: error: ignoring attribute ‘section (".init.data")’ because it conflicts with previous ‘section (".init.rodata")’ [-Werror=attributes]

    4 | const char __initdata *const blacklist_hashes[] = {
      | ^~~~~
In file included from certs/blacklist_hashes.c:2:
certs/blacklist.h:5:38: note: previous declaration here
    5 | extern const char __initconst *const blacklist_hashes[];
      |                                      ^~~~~~~~~~~~~~~~

Apply the same fix as commit 2be04df5668d ("certs/blacklist_nohashes.c:
fix const confusion in certs blacklist").

Fixes 734114f8782f ("KEYS: Add a system blacklist keyring")
---
 certs/blacklist_hashes.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
certs/blacklist_hashes.c +1 -1
index 344892337be0..d5961aa3d338 100644
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include "blacklist.h"
 
-const char __initdata *const blacklist_hashes[] = {
+const char __initconst *const blacklist_hashes[] = {
 #include CONFIG_SYSTEM_BLACKLIST_HASH_LIST
 	, NULL
 };
Signature · 1 line
2.35.1
author maintainer
#208 lore raw

PATCH 208/234 faddr2line: Fix overlapping text section failures, the sequel

In reply to Greg Kroah-Hartman

From: Josh Poimboeuf <[email protected]>

[ Upstream commit dcea997beed694cbd8705100ca1a6eb0d886de69 ]

If a function lives in a section other than .text, but .text also exists in the object, faddr2line may wrongly assume .text. This can result in comically wrong output. For example:

  $ scripts/faddr2line vmlinux.o enter_from_user_mode+0x1c
  enter_from_user_mode+0x1c/0x30:
  find_next_bit at /home/jpoimboe/git/linux/./include/linux/find.h:40
  (inlined by) perf_clear_dirty_counters at /home/jpoimboe/git/linux/arch/x86/events/core.c:2504

Fix it by passing the section name to addr2line, unless the object file is vmlinux, in which case the symbol table uses absolute addresses.

---
 scripts/faddr2line | 45 ++++++++++++++++++++++++++++++++++-----------
 1 file changed, 34 insertions(+), 11 deletions(-)
scripts/faddr2line +34 -11
index 226c3f559dc5..2571caac3156 100755
@@ -95,17 +95,25 @@ __faddr2line() {
 	local print_warnings=$4
 
 	local sym_name=${func_addr%+*}
-	local offset=${func_addr#*+}
-	offset=${offset%/*}
+	local func_offset=${func_addr#*+}
+	func_offset=${func_offset%/*}
 	local user_size=
+	local file_type
+	local is_vmlinux=0
 	[[ $func_addr =~ "/" ]] && user_size=${func_addr#*/}
 
-	if [[ -z $sym_name ]] || [[ -z $offset ]] || [[ $sym_name = $func_addr ]]; then
+	if [[ -z $sym_name ]] || [[ -z $func_offset ]] || [[ $sym_name = $func_addr ]]; then
 		warn "bad func+offset $func_addr"
 		DONE=1
 		return
 	fi
 
+	# vmlinux uses absolute addresses in the section table rather than
+	# section offsets.
+	local file_type=$(${READELF} --file-header $objfile |
+		${AWK} '$1 == "Type:" { print $2; exit }')
+	[[ $file_type = "EXEC" ]] && is_vmlinux=1
+
 	# Go through each of the object's symbols which match the func name.
 	# In rare cases there might be duplicates, in which case we print all
 	# matches.
@@ -114,9 +122,11 @@ __faddr2line() {
 		local sym_addr=0x${fields[1]}
 		local sym_elf_size=${fields[2]}
 		local sym_sec=${fields[6]}
+		local sec_size
+		local sec_name
 
 		# Get the section size:
-		local sec_size=$(${READELF} --section-headers --wide $objfile |
+		sec_size=$(${READELF} --section-headers --wide $objfile |
 			sed 's/\[ /\[/' |
 			${AWK} -v sec=$sym_sec '$1 == "[" sec "]" { print "0x" $6; exit }')
 
@@ -126,6 +136,17 @@ __faddr2line() {
 			return
 		fi
 
+		# Get the section name:
+		sec_name=$(${READELF} --section-headers --wide $objfile |
+			sed 's/\[ /\[/' |
+			${AWK} -v sec=$sym_sec '$1 == "[" sec "]" { print $2; exit }')
+
+		if [[ -z $sec_name ]]; then
+			warn "bad section name: section: $sym_sec"
+			DONE=1
+			return
+		fi
+
 		# Calculate the symbol size.
 		#
 		# Unfortunately we can't use the ELF size, because kallsyms
@@ -174,10 +195,10 @@ __faddr2line() {
 
 		sym_size=0x$(printf %x $sym_size)
 
-		# Calculate the section address from user-supplied offset:
-		local addr=$(($sym_addr + $offset))
+		# Calculate the address from user-supplied offset:
+		local addr=$(($sym_addr + $func_offset))
 		if [[ -z $addr ]] || [[ $addr = 0 ]]; then
-			warn "bad address: $sym_addr + $offset"
+			warn "bad address: $sym_addr + $func_offset"
 			DONE=1
 			return
 		fi
@@ -191,9 +212,9 @@ __faddr2line() {
 		fi
 
 		# Make sure the provided offset is within the symbol's range:
-		if [[ $offset -gt $sym_size ]]; then
+		if [[ $func_offset -gt $sym_size ]]; then
 			[[ $print_warnings = 1 ]] &&
-				echo "skipping $sym_name address at $addr due to size mismatch ($offset > $sym_size)"
+				echo "skipping $sym_name address at $addr due to size mismatch ($func_offset > $sym_size)"
 			continue
 		fi
 
@@ -202,11 +223,13 @@ __faddr2line() {
 		[[ $FIRST = 0 ]] && echo
 		FIRST=0
 
-		echo "$sym_name+$offset/$sym_size:"
+		echo "$sym_name+$func_offset/$sym_size:"
 
 		# Pass section address to addr2line and strip absolute paths
 		# from the output:
-		local output=$(${ADDR2LINE} -fpie $objfile $addr | sed "s; $dir_prefix\(\./\)*; ;")
+		local args="--functions --pretty-print --inlines --exe=$objfile"
+		[[ $is_vmlinux = 0 ]] && args="$args --section=$sec_name"
+		local output=$(${ADDR2LINE} $args $addr | sed "s; $dir_prefix\(\./\)*; ;")
 		[[ -z $output ]] && continue
 
 		# Default output (non --list):
Signature · 1 line
2.35.1
author maintainer
#209 lore raw

PATCH 209/234 irqchip/gic/realview: Fix refcount leak in realview_gic_of_init

In reply to Greg Kroah-Hartman

From: Miaoqian Lin <[email protected]>

[ Upstream commit f4b98e314888cc51486421bcf6d52852452ea48b ]

of_find_matching_node_and_match() returns a node pointer with refcount incremented, we should use of_node_put() on it when not need anymore. Add missing of_node_put() to avoid refcount leak.

Fixes 82b0a434b436 ("irqchip/gic/realview: Support more RealView DCC variants")
---
 drivers/irqchip/irq-gic-realview.c | 1 +
 1 file changed, 1 insertion(+)
drivers/irqchip/irq-gic-realview.c +1 -0
index b4c1924f0255..38fab02ffe9d 100644
@@ -57,6 +57,7 @@ realview_gic_of_init(struct device_node *node, struct device_node *parent)
 
 	/* The PB11MPCore GIC needs to be configured in the syscon */
 	map = syscon_node_to_regmap(np);
+	of_node_put(np);
 	if (!IS_ERR(map)) {
 		/* new irq mode with no DCC */
 		regmap_write(map, REALVIEW_SYS_LOCK_OFFSET,
Signature · 1 line
2.35.1
author maintainer
#210 lore raw

PATCH 210/234 irqchip/gic-v3: Fix refcount leak in gic_populate_ppi_partitions

In reply to Greg Kroah-Hartman

From: Miaoqian Lin <[email protected]>

[ Upstream commit fa1ad9d4cc47ca2470cd904ad4519f05d7e43a2b ]

of_find_node_by_phandle() returns a node pointer with refcount incremented, we should use of_node_put() on it when not need anymore. Add missing of_node_put() to avoid refcount leak.

Fixes e3825ba1af3a ("irqchip/gic-v3: Add support for partitioned PPIs")
---
 drivers/irqchip/irq-gic-v3.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/irqchip/irq-gic-v3.c +4 -1
index 05b9a4cdc8fd..8d8b8d192e2e 100644
@@ -1205,12 +1205,15 @@ static void __init gic_populate_ppi_partitions(struct device_node *gic_node)
 				continue;
 
 			cpu = of_cpu_node_to_id(cpu_node);
-			if (WARN_ON(cpu < 0))
+			if (WARN_ON(cpu < 0)) {
+				of_node_put(cpu_node);
 				continue;
+			}
 
 			pr_cont("%pOF[%d] ", cpu_node, cpu);
 
 			cpumask_set_cpu(cpu, &part->mask);
+			of_node_put(cpu_node);
 		}
 
 		pr_cont("}\n");
Signature · 1 line
2.35.1