BACKPORT: crypto: chacha20 - Fix chacha20_block() keystream alignment (again)
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@chronox.de>
Cc: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
(cherry picked from commit a5e9f557098e54af44ade5d501379be18435bfbf)
Conflicts:
drivers/char/random.c
Bug: 112008522
Test: As series, see Ic61c13b53facfd2173065be715a7ee5f3af8760b
Change-Id: Ic355d2416330ae2f4a50cb7064633810e35a93bf
Signed-off-by: Eric Biggers <ebiggers@google.com>
diff --git a/crypto/chacha20_generic.c b/crypto/chacha20_generic.c
index 1a48d3a..94a2048 100644
--- a/crypto/chacha20_generic.c
+++ b/crypto/chacha20_generic.c
@@ -19,20 +19,21 @@
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);
}
}
diff --git a/drivers/char/random.c b/drivers/char/random.c
index 4fc600d..d2ff213 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -440,9 +440,9 @@
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 struct ratelimit_state unseeded_warning =
@@ -868,7 +868,7 @@
unsigned long flags;
int i, num;
union {
- __u32 block[CHACHA20_BLOCK_WORDS];
+ __u8 block[CHACHA20_BLOCK_SIZE];
__u32 key[8];
} buf;
@@ -926,7 +926,7 @@
}
static void _extract_crng(struct crng_state *crng,
- __u32 out[CHACHA20_BLOCK_WORDS])
+ __u8 out[CHACHA20_BLOCK_SIZE])
{
unsigned long v, flags;
@@ -943,7 +943,7 @@
spin_unlock_irqrestore(&crng->lock, flags);
}
-static void extract_crng(__u32 out[CHACHA20_BLOCK_WORDS])
+static void extract_crng(__u8 out[CHACHA20_BLOCK_SIZE])
{
struct crng_state *crng = NULL;
@@ -961,7 +961,7 @@
* 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;
@@ -973,14 +973,14 @@
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)
{
struct crng_state *crng = NULL;
@@ -996,7 +996,7 @@
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) {
@@ -1564,7 +1564,7 @@
*/
void get_random_bytes(void *buf, int nbytes)
{
- __u32 tmp[CHACHA20_BLOCK_WORDS];
+ __u8 tmp[CHACHA20_BLOCK_SIZE] __aligned(4);
#if DEBUG_RANDOM_BOOT > 0
if (!crng_ready())
@@ -2132,7 +2132,7 @@
batch = &get_cpu_var(batched_entropy_long);
if (batch->position % ARRAY_SIZE(batch->entropy_long) == 0) {
- extract_crng((__u32 *)batch->entropy_long);
+ extract_crng((u8 *)batch->entropy_long);
batch->position = 0;
}
ret = batch->entropy_long[batch->position++];
@@ -2158,7 +2158,7 @@
batch = &get_cpu_var(batched_entropy_int);
if (batch->position % ARRAY_SIZE(batch->entropy_int) == 0) {
- extract_crng((__u32 *)batch->entropy_int);
+ extract_crng((u8 *)batch->entropy_int);
batch->position = 0;
}
ret = batch->entropy_int[batch->position++];
diff --git a/include/crypto/chacha20.h b/include/crypto/chacha20.h
index 254298b..01543a9 100644
--- a/include/crypto/chacha20.h
+++ b/include/crypto/chacha20.h
@@ -11,13 +11,12 @@
#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_tfm *tfm, const u8 *key,
unsigned int keysize);
diff --git a/lib/chacha20.c b/lib/chacha20.c
index c1cc50f..d907fec 100644
--- a/lib/chacha20.c
+++ b/lib/chacha20.c
@@ -16,9 +16,9 @@
#include <asm/unaligned.h>
#include <crypto/chacha20.h>
-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++)
@@ -67,7 +67,7 @@
}
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]++;
}