[PATCH 4.14 073/237] crypto: chacha20 - Fix chacha20_block() keystream alignment (again)

From: Greg Kroah-Hartman
Date: Thu Jun 23 2022 - 13:25:50 EST


From: Eric Biggers <ebiggers@xxxxxxxxxx>

[ Upstream commit a5e9f557098e54af44ade5d501379be18435bfbf ]

In commit 9f480faec58c ("crypto: chacha20 - Fix keystream alignment for
chacha20_block()"), I had missed that chacha20_block() can be called
directly on the buffer passed to get_random_bytes(), which can have any
alignment. So, while my commit didn't break anything, it didn't fully
solve the alignment problems.

Revert my solution and just update chacha20_block() to use
put_unaligned_le32(), so the output buffer need not be aligned.
This is simpler, and on many CPUs it's the same speed.

But, I kept the 'tmp' buffers in extract_crng_user() and
_get_random_bytes() 4-byte aligned, since that alignment is actually
needed for _crng_backtrack_protect() too.

Reported-by: Stephan Müller <smueller@xxxxxxxxxx>
Cc: Theodore Ts'o <tytso@xxxxxxx>
Signed-off-by: Eric Biggers <ebiggers@xxxxxxxxxx>
Signed-off-by: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
crypto/chacha20_generic.c | 7 ++++---
drivers/char/random.c | 24 ++++++++++++------------
include/crypto/chacha20.h | 3 +--
lib/chacha20.c | 6 +++---
4 files changed, 20 insertions(+), 20 deletions(-)

--- a/crypto/chacha20_generic.c
+++ b/crypto/chacha20_generic.c
@@ -22,20 +22,21 @@ static inline u32 le32_to_cpuvp(const vo
static void chacha20_docrypt(u32 *state, u8 *dst, const u8 *src,
unsigned int bytes)
{
- u32 stream[CHACHA20_BLOCK_WORDS];
+ /* aligned to potentially speed up crypto_xor() */
+ u8 stream[CHACHA20_BLOCK_SIZE] __aligned(sizeof(long));

if (dst != src)
memcpy(dst, src, bytes);

while (bytes >= CHACHA20_BLOCK_SIZE) {
chacha20_block(state, stream);
- crypto_xor(dst, (const u8 *)stream, CHACHA20_BLOCK_SIZE);
+ crypto_xor(dst, stream, CHACHA20_BLOCK_SIZE);
bytes -= CHACHA20_BLOCK_SIZE;
dst += CHACHA20_BLOCK_SIZE;
}
if (bytes) {
chacha20_block(state, stream);
- crypto_xor(dst, (const u8 *)stream, bytes);
+ crypto_xor(dst, stream, bytes);
}
}

--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -485,9 +485,9 @@ static int crng_init_cnt = 0;
static unsigned long crng_global_init_time = 0;
#define CRNG_INIT_CNT_THRESH (2*CHACHA20_KEY_SIZE)
static void _extract_crng(struct crng_state *crng,
- __u32 out[CHACHA20_BLOCK_WORDS]);
+ __u8 out[CHACHA20_BLOCK_SIZE]);
static void _crng_backtrack_protect(struct crng_state *crng,
- __u32 tmp[CHACHA20_BLOCK_WORDS], int used);
+ __u8 tmp[CHACHA20_BLOCK_SIZE], int used);
static void process_random_ready_list(void);
static void _get_random_bytes(void *buf, int nbytes);

@@ -986,7 +986,7 @@ static void crng_reseed(struct crng_stat
unsigned long flags;
int i, num;
union {
- __u32 block[CHACHA20_BLOCK_WORDS];
+ __u8 block[CHACHA20_BLOCK_SIZE];
__u32 key[8];
} buf;

@@ -1014,7 +1014,7 @@ static void crng_reseed(struct crng_stat
}

static void _extract_crng(struct crng_state *crng,
- __u32 out[CHACHA20_BLOCK_WORDS])
+ __u8 out[CHACHA20_BLOCK_SIZE])
{
unsigned long flags, init_time;

@@ -1032,7 +1032,7 @@ static void _extract_crng(struct crng_st
spin_unlock_irqrestore(&crng->lock, flags);
}

-static void extract_crng(__u32 out[CHACHA20_BLOCK_WORDS])
+static void extract_crng(__u8 out[CHACHA20_BLOCK_SIZE])
{
_extract_crng(select_crng(), out);
}
@@ -1042,7 +1042,7 @@ static void extract_crng(__u32 out[CHACH
* enough) to mutate the CRNG key to provide backtracking protection.
*/
static void _crng_backtrack_protect(struct crng_state *crng,
- __u32 tmp[CHACHA20_BLOCK_WORDS], int used)
+ __u8 tmp[CHACHA20_BLOCK_SIZE], int used)
{
unsigned long flags;
__u32 *s, *d;
@@ -1054,14 +1054,14 @@ static void _crng_backtrack_protect(stru
used = 0;
}
spin_lock_irqsave(&crng->lock, flags);
- s = &tmp[used / sizeof(__u32)];
+ s = (__u32 *) &tmp[used];
d = &crng->state[4];
for (i=0; i < 8; i++)
*d++ ^= *s++;
spin_unlock_irqrestore(&crng->lock, flags);
}

-static void crng_backtrack_protect(__u32 tmp[CHACHA20_BLOCK_WORDS], int used)
+static void crng_backtrack_protect(__u8 tmp[CHACHA20_BLOCK_SIZE], int used)
{
_crng_backtrack_protect(select_crng(), tmp, used);
}
@@ -1069,7 +1069,7 @@ static void crng_backtrack_protect(__u32
static ssize_t extract_crng_user(void __user *buf, size_t nbytes)
{
ssize_t ret = 0, i = CHACHA20_BLOCK_SIZE;
- __u32 tmp[CHACHA20_BLOCK_WORDS];
+ __u8 tmp[CHACHA20_BLOCK_SIZE] __aligned(4);
int large_request = (nbytes > 256);

while (nbytes) {
@@ -1529,7 +1529,7 @@ static void _warn_unseeded_randomness(co
*/
static void _get_random_bytes(void *buf, int nbytes)
{
- __u32 tmp[CHACHA20_BLOCK_WORDS];
+ __u8 tmp[CHACHA20_BLOCK_SIZE] __aligned(4);

trace_get_random_bytes(nbytes, _RET_IP_);

@@ -2116,7 +2116,7 @@ u64 get_random_u64(void)
batch = raw_cpu_ptr(&batched_entropy_u64);
spin_lock_irqsave(&batch->batch_lock, flags);
if (batch->position % ARRAY_SIZE(batch->entropy_u64) == 0) {
- extract_crng((__u32 *)batch->entropy_u64);
+ extract_crng((u8 *)batch->entropy_u64);
batch->position = 0;
}
ret = batch->entropy_u64[batch->position++];
@@ -2140,7 +2140,7 @@ u32 get_random_u32(void)
batch = raw_cpu_ptr(&batched_entropy_u32);
spin_lock_irqsave(&batch->batch_lock, flags);
if (batch->position % ARRAY_SIZE(batch->entropy_u32) == 0) {
- extract_crng(batch->entropy_u32);
+ extract_crng((u8 *)batch->entropy_u32);
batch->position = 0;
}
ret = batch->entropy_u32[batch->position++];
--- a/include/crypto/chacha20.h
+++ b/include/crypto/chacha20.h
@@ -13,13 +13,12 @@
#define CHACHA20_IV_SIZE 16
#define CHACHA20_KEY_SIZE 32
#define CHACHA20_BLOCK_SIZE 64
-#define CHACHA20_BLOCK_WORDS (CHACHA20_BLOCK_SIZE / sizeof(u32))

struct chacha20_ctx {
u32 key[8];
};

-void chacha20_block(u32 *state, u32 *stream);
+void chacha20_block(u32 *state, u8 *stream);
void crypto_chacha20_init(u32 *state, struct chacha20_ctx *ctx, u8 *iv);
int crypto_chacha20_setkey(struct crypto_skcipher *tfm, const u8 *key,
unsigned int keysize);
--- a/lib/chacha20.c
+++ b/lib/chacha20.c
@@ -21,9 +21,9 @@ static inline u32 rotl32(u32 v, u8 n)
return (v << n) | (v >> (sizeof(v) * 8 - n));
}

-void chacha20_block(u32 *state, u32 *stream)
+void chacha20_block(u32 *state, u8 *stream)
{
- u32 x[16], *out = stream;
+ u32 x[16];
int i;

for (i = 0; i < ARRAY_SIZE(x); i++)
@@ -72,7 +72,7 @@ void chacha20_block(u32 *state, u32 *str
}

for (i = 0; i < ARRAY_SIZE(x); i++)
- out[i] = cpu_to_le32(x[i] + state[i]);
+ put_unaligned_le32(x[i] + state[i], &stream[i * sizeof(u32)]);

state[12]++;
}