mirror/mbedtls
1
0
Fork 0
mirror of https://github.com/Mbed-TLS/mbedtls.git synced 2025-04-24 06:02:44 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Improve mbedtls_xor docs

Browse Source 
Signed-off-by: Dave Rodgman <dave.rodgman@arm.com>
This commit is contained in:
Dave Rodgman 2024-01-19 14:08:17 +00:00
parent 18d90d7519
commit 2143a4ad1f
1 changed files with 16 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
library/common.h
View File

@ -174,6 +174,14 @@ __attribute__((always_inline))
* \param a Pointer to input (buffer of at least \p n bytes) * \param a Pointer to input (buffer of at least \p n bytes)
* \param b Pointer to input (buffer of at least \p n bytes) * \param b Pointer to input (buffer of at least \p n bytes)
* \param n Number of bytes to process. * \param n Number of bytes to process.
*
* \note Depending on the situation, it may be faster to use either mbedtls_xor or
* mbedtls_xor_no_simd (these are functionally equivalent).
* If the result is used immediately after the xor operation in non-SIMD code (e.g, in
* AES-CBC), there may be additional latency to transfer the data from SIMD to scalar
* registers, and in this case, mbedtls_xor_no_simd may be faster. In other cases where
* the result is not used immediately (e.g., in AES-CTR), mbedtls_xor may be faster.
* For targets without SIMD support, they will behave the same.
*/ */
static inline void mbedtls_xor(unsigned char *r, static inline void mbedtls_xor(unsigned char *r,
const unsigned char *a, const unsigned char *a,
@ -238,6 +246,14 @@ __attribute__((always_inline))
* \param a Pointer to input (buffer of at least \p n bytes) * \param a Pointer to input (buffer of at least \p n bytes)
* \param b Pointer to input (buffer of at least \p n bytes) * \param b Pointer to input (buffer of at least \p n bytes)
* \param n Number of bytes to process. * \param n Number of bytes to process.
*
* \note Depending on the situation, it may be faster to use either mbedtls_xor or
* mbedtls_xor_no_simd (these are functionally equivalent).
* If the result is used immediately after the xor operation in non-SIMD code (e.g, in
* AES-CBC), there may be additional latency to transfer the data from SIMD to scalar
* registers, and in this case, mbedtls_xor_no_simd may be faster. In other cases where
* the result is not used immediately (e.g., in AES-CTR), mbedtls_xor may be faster.
* For targets without SIMD support, they will behave the same.
*/ */
static inline void mbedtls_xor_no_simd(unsigned char *r, static inline void mbedtls_xor_no_simd(unsigned char *r,
const unsigned char *a, const unsigned char *a,