crypto: crypto_xor - use helpers for unaligned accesses

Dereferencing a misaligned pointer is undefined behavior in C, and may
result in codegen on architectures such as ARM that trigger alignments
traps and expensive fixups in software.

Instead, use the get_aligned()/put_aligned() accessors, which are cheap
or even completely free when CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS=y.

In the converse case, the prior alignment checks ensure that the casts
are safe, and so no unaligned accessors are necessary.

Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This commit is contained in:
Ard Biesheuvel 2022-02-23 08:07:01 +01:00 committed by Herbert Xu
parent 4920a4a726
commit 7976c14925
2 changed files with 29 additions and 5 deletions

View File

@ -1022,7 +1022,13 @@ void __crypto_xor(u8 *dst, const u8 *src1, const u8 *src2, unsigned int len)
} }
while (IS_ENABLED(CONFIG_64BIT) && len >= 8 && !(relalign & 7)) { while (IS_ENABLED(CONFIG_64BIT) && len >= 8 && !(relalign & 7)) {
*(u64 *)dst = *(u64 *)src1 ^ *(u64 *)src2; if (IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)) {
u64 l = get_unaligned((u64 *)src1) ^
get_unaligned((u64 *)src2);
put_unaligned(l, (u64 *)dst);
} else {
*(u64 *)dst = *(u64 *)src1 ^ *(u64 *)src2;
}
dst += 8; dst += 8;
src1 += 8; src1 += 8;
src2 += 8; src2 += 8;
@ -1030,7 +1036,13 @@ void __crypto_xor(u8 *dst, const u8 *src1, const u8 *src2, unsigned int len)
} }
while (len >= 4 && !(relalign & 3)) { while (len >= 4 && !(relalign & 3)) {
*(u32 *)dst = *(u32 *)src1 ^ *(u32 *)src2; if (IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)) {
u32 l = get_unaligned((u32 *)src1) ^
get_unaligned((u32 *)src2);
put_unaligned(l, (u32 *)dst);
} else {
*(u32 *)dst = *(u32 *)src1 ^ *(u32 *)src2;
}
dst += 4; dst += 4;
src1 += 4; src1 += 4;
src2 += 4; src2 += 4;
@ -1038,7 +1050,13 @@ void __crypto_xor(u8 *dst, const u8 *src1, const u8 *src2, unsigned int len)
} }
while (len >= 2 && !(relalign & 1)) { while (len >= 2 && !(relalign & 1)) {
*(u16 *)dst = *(u16 *)src1 ^ *(u16 *)src2; if (IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)) {
u16 l = get_unaligned((u16 *)src1) ^
get_unaligned((u16 *)src2);
put_unaligned(l, (u16 *)dst);
} else {
*(u16 *)dst = *(u16 *)src1 ^ *(u16 *)src2;
}
dst += 2; dst += 2;
src1 += 2; src1 += 2;
src2 += 2; src2 += 2;

View File

@ -13,6 +13,8 @@
#include <linux/list.h> #include <linux/list.h>
#include <linux/types.h> #include <linux/types.h>
#include <asm/unaligned.h>
/* /*
* Maximum values for blocksize and alignmask, used to allocate * Maximum values for blocksize and alignmask, used to allocate
* static buffers that are big enough for any combination of * static buffers that are big enough for any combination of
@ -154,9 +156,11 @@ static inline void crypto_xor(u8 *dst, const u8 *src, unsigned int size)
(size % sizeof(unsigned long)) == 0) { (size % sizeof(unsigned long)) == 0) {
unsigned long *d = (unsigned long *)dst; unsigned long *d = (unsigned long *)dst;
unsigned long *s = (unsigned long *)src; unsigned long *s = (unsigned long *)src;
unsigned long l;
while (size > 0) { while (size > 0) {
*d++ ^= *s++; l = get_unaligned(d) ^ get_unaligned(s++);
put_unaligned(l, d++);
size -= sizeof(unsigned long); size -= sizeof(unsigned long);
} }
} else { } else {
@ -173,9 +177,11 @@ static inline void crypto_xor_cpy(u8 *dst, const u8 *src1, const u8 *src2,
unsigned long *d = (unsigned long *)dst; unsigned long *d = (unsigned long *)dst;
unsigned long *s1 = (unsigned long *)src1; unsigned long *s1 = (unsigned long *)src1;
unsigned long *s2 = (unsigned long *)src2; unsigned long *s2 = (unsigned long *)src2;
unsigned long l;
while (size > 0) { while (size > 0) {
*d++ = *s1++ ^ *s2++; l = get_unaligned(s1++) ^ get_unaligned(s2++);
put_unaligned(l, d++);
size -= sizeof(unsigned long); size -= sizeof(unsigned long);
} }
} else { } else {