blender
/
blender
136
407
Fork 589
Code Issues 6.3k Pull Requests 771 Projects 19 Wiki Activity

Alternative Upload geometry data in parallel to multiple GPUs using the "Multi-Device" #107552

Open
William Leeson wants to merge 137 commits from leesonw/blender-cluster:upload_changed into main
pull from: leesonw/blender-cluster:upload_changed
merge into: blender:main

When changing the target branch, be careful to rebase the branch in your fork to match. See documentation.
Conversation 2 Commits 137 Files Changed 43 +4 -14
2 changed files with 4 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit a875bc9e45 - Show all commits

6
extern/cuew/include/cuew.h vendored
View File

@ -886,11 +886,11 @@ typedef enum {
} nvrtcResult;
typedef struct _nvrtcProgram* nvrtcProgram;
// FRL_CGR
/* Function types. */
typedef void CUDAAPI tnvtxRangePushA(const char *msg);
typedef void CUDAAPI tnvtxRangePop();
// FRL_CGR
typedef CUresult CUDAAPI tcuGetErrorString(CUresult error, const char** pStr);
typedef CUresult CUDAAPI tcuGetErrorName(CUresult error, const char** pStr);
typedef CUresult CUDAAPI tcuInit(unsigned int Flags);
@ -1382,10 +1382,8 @@ enum {
enum {
CUEW_INIT_CUDA = 1,
// FRL_CGR
CUEW_INIT_NVRTC = 2,
CUEW_INIT_NVTX = 4,
// FRL_CGR
};
int cuewInit(cuuint32_t flags);

12
extern/cuew/src/cuew.c vendored
View File

@ -66,21 +66,18 @@ typedef void* DynamicLibrary;
_LIBRARY_FIND_CHECKED(nvrtc_lib, name)
#define NVRTC_LIBRARY_FIND(name) _LIBRARY_FIND(nvrtc_lib, name)
// FRL_CGR
#define NVTX_LIBRARY_FIND_CHECKED(name) \
_LIBRARY_FIND_CHECKED(nvrtc_lib, name)
#define NVTX_LIBRARY_FIND(name) _LIBRARY_FIND(nvtx_lib, name)
// FRL_CGR
static DynamicLibrary cuda_lib;
static DynamicLibrary nvrtc_lib;
static DynamicLibrary nvtx_lib;
/* Function definitions. */
// FRL_CGR
tnvtxRangePushA *nvtxRangePushA;
tnvtxRangePop *nvtxRangePop;
// FRL_CGR
tcuGetErrorString *cuGetErrorString;
tcuGetErrorName *cuGetErrorName;
@ -623,7 +620,6 @@ static int cuewCudaInit(void) {
return result;
}
// FRL_CGR
static void cuewExitNvtx(void) {
if (nvrtc_lib != NULL) {
/* Ignore errors. */
@ -631,7 +627,6 @@ static void cuewExitNvtx(void) {
nvtx_lib = NULL;
}
}
// FRL_CGR
static void cuewExitNvrtc(void) {
if (nvrtc_lib != NULL) {
@ -703,7 +698,6 @@ static int cuewNvrtcInit(void) {
return result;
}
// FRL_CGR
static int cuewNvtxInit(void) {
/* Library paths. */
#ifdef _WIN32
@ -752,7 +746,6 @@ static int cuewNvtxInit(void) {
result = CUEW_SUCCESS;
return result;
}
// FRL_CGR
int cuewInit(cuuint32_t flags) {
int result = CUEW_SUCCESS;
@ -770,14 +763,13 @@ int cuewInit(cuuint32_t flags) {
return result;
}
}
// FRL_CGR
if(flags & CUEW_INIT_NVTX) {
result = cuewNvtxInit();
if(result != CUEW_SUCCESS) {
return result;
}
}
// FRL_CGR
return result;
}