Administrator/opencv-MIRROR
1
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2026-09-12 13:23:03 -05:00
Code Issues Packages Projects Releases Wiki Activity

dnn: fix the parameter names documented for memset

Browse Source 
The four memset overloads in cuda4dnn document their first parameter as
src while it is called dest, and the directions are the wrong way round:
the output buffer is marked [in] and the 8-bit fill value is marked [out].

The prose above two of them already says dest, only the tags disagree.
This commit is contained in:
Anton Karpov
2026-08-05 15:48:52 +03:00
parent 4bee6895eb
commit f625c00f5a
2 changed files with 8 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
modules/dnn/src/cuda4dnn/csl/memory.hpp
View File

@@ -178,8 +178,8 @@ namespace cv { namespace dnn { namespace cuda4dnn { namespace csl {
/** sets device memory block to a specific 8-bit value
*
* \param[in] src device pointer
* \param[out] ch 8-bit value to fill the device memory with
* \param[out] dest device pointer
* \param[in] ch 8-bit value to fill the device memory with
*
* Exception Guarantee: Basic
*/
@@ -243,8 +243,8 @@ namespace cv { namespace dnn { namespace cuda4dnn { namespace csl {
/** sets device memory block to a specific 8-bit value asynchronously
*
* \param[in] src device pointer
* \param[out] ch 8-bit value to fill the device memory with
* \param[out] dest device pointer
* \param[in] ch 8-bit value to fill the device memory with
* \param stream CUDA stream that has to be used for the memory operation
*
* Exception Guarantee: Basic

8
modules/dnn/src/cuda4dnn/csl/pointer.hpp
View File

@@ -307,8 +307,8 @@ namespace cv { namespace dnn { namespace cuda4dnn { namespace csl {
/** sets \p n elements to \p ch in \p dest
*
* \param[in] src device pointer
* \param[out] ch 8-bit value to fill the device memory with
* \param[out] dest device pointer
* \param[in] ch 8-bit value to fill the device memory with
*
* Pre-conditions:
* - memory pointed by \p dest must be large enough to hold \p n elements
@@ -388,8 +388,8 @@ namespace cv { namespace dnn { namespace cuda4dnn { namespace csl {
/** sets \p n elements to \p ch in \p dest asynchronously
*
* \param[in] src device pointer
* \param[out] ch 8-bit value to fill the device memory with
* \param[out] dest device pointer
* \param[in] ch 8-bit value to fill the device memory with
* \param stream CUDA stream that has to be used for the memory operation
*
* Pre-conditions: