Fixed bug 4557 - SDL_SIMDAlloc and *Free should be in the public inte…

…rface Martin Gerhardy These functions are really useful and should get exposed imo.
libsdl-org · Jun 8, 2019 · 8728ce4 · 8728ce4
1 parent 56b7f4c
commit 8728ce4
Show file tree

Hide file tree

Showing 7 changed files with 69 additions and 91 deletions.
diff --git a/include/SDL_cpuinfo.h b/include/SDL_cpuinfo.h
@@ -196,6 +196,69 @@ extern DECLSPEC SDL_bool SDLCALL SDL_HasNEON(void);
  */
 extern DECLSPEC int SDLCALL SDL_GetSystemRAM(void);
 
+/**
+ * \brief Report the alignment this system needs for SIMD allocations.
+ *
+ * This will return the minimum number of bytes to which a pointer must be
+ *  aligned to be compatible with SIMD instructions on the current machine.
+ *  For example, if the machine supports SSE only, it will return 16, but if
+ *  it supports AVX-512F, it'll return 64 (etc). This only reports values for
+ *  instruction sets SDL knows about, so if your SDL build doesn't have
+ *  SDL_HasAVX512F(), then it might return 16 for the SSE support it sees and
+ *  not 64 for the AVX-512 instructions that exist but SDL doesn't know about.
+ *  Plan accordingly.
+ */
+extern DECLSPEC size_t SDLCALL SDL_SIMDGetAlignment(void);
+
+/**
+ * \brief Allocate memory in a SIMD-friendly way.
+ *
+ * This will allocate a block of memory that is suitable for use with SIMD
+ *  instructions. Specifically, it will be properly aligned and padded for
+ *  the system's supported vector instructions.
+ *
+ * The memory returned will be padded such that it is safe to read or write
+ *  an incomplete vector at the end of the memory block. This can be useful
+ *  so you don't have to drop back to a scalar fallback at the end of your
+ *  SIMD processing loop to deal with the final elements without overflowing
+ *  the allocated buffer.
+ *
+ * You must free this memory with SDL_FreeSIMD(), not free() or SDL_free()
+ *  or delete[], etc.
+ *
+ * Note that SDL will only deal with SIMD instruction sets it is aware of;
+ *  for example, SDL 2.0.8 knows that SSE wants 16-byte vectors
+ *  (SDL_HasSSE()), and AVX2 wants 32 bytes (SDL_HasAVX2()), but doesn't
+ *  know that AVX-512 wants 64. To be clear: if you can't decide to use an
+ *  instruction set with an SDL_Has*() function, don't use that instruction
+ *  set with memory allocated through here.
+ *
+ * SDL_AllocSIMD(0) will return a non-NULL pointer, assuming the system isn't
+ *  out of memory.
+ *
+ *  \param len The length, in bytes, of the block to allocated. The actual
+ *             allocated block might be larger due to padding, etc.
+ * \return Pointer to newly-allocated block, NULL if out of memory.
+ *
+ * \sa SDL_SIMDAlignment
+ * \sa SDL_SIMDFree
+ */
+extern DECLSPEC void * SDLCALL SDL_SIMDAlloc(const size_t len);
+
+/**
+ * \brief Deallocate memory obtained from SDL_SIMDAlloc
+ *
+ * It is not valid to use this function on a pointer from anything but
+ *  SDL_SIMDAlloc(). It can't be used on pointers from malloc, realloc,
+ *  SDL_malloc, memalign, new[], etc.
+ *
+ * However, SDL_SIMDFree(NULL) is a legal no-op.
+ *
+ * \sa SDL_SIMDAlloc
+ */
+extern DECLSPEC void SDLCALL SDL_SIMDFree(void *ptr);
+
+/* vi: set ts=4 sw=4 expandtab: */
 /* Ends C function definitions when using C++ */
 #ifdef __cplusplus
 }

diff --git a/src/cpuinfo/SDL_cpuinfo.c b/src/cpuinfo/SDL_cpuinfo.c
@@ -22,7 +22,6 @@
 #include "SDL_config.h"
 #else
 #include "../SDL_internal.h"
-#include "SDL_simd.h"
 #endif
 
 #if defined(__WIN32__) || defined(__WINRT__)

diff --git a/src/cpuinfo/SDL_simd.h b/src/cpuinfo/SDL_simd.h
diff --git a/src/dynapi/SDL_dynapi_overrides.h b/src/dynapi/SDL_dynapi_overrides.h
@@ -714,3 +714,6 @@
 #define SDL_RenderCopyExF SDL_RenderCopyExF_REAL
 #define SDL_GetTouchDeviceType SDL_GetTouchDeviceType_REAL
 #define SDL_UIKitRunApp SDL_UIKitRunApp_REAL
+#define SDL_SIMDGetAlignment SDL_SIMDGetAlignment_REAL
+#define SDL_SIMDAlloc SDL_SIMDAlloc_REAL
+#define SDL_SIMDFree SDL_SIMDFree_REAL
diff --git a/src/dynapi/SDL_dynapi_procs.h b/src/dynapi/SDL_dynapi_procs.h
@@ -770,3 +770,6 @@ SDL_DYNAPI_PROC(SDL_TouchDeviceType,SDL_GetTouchDeviceType,(SDL_TouchID a),(a),r
 #ifdef __IPHONEOS__
 SDL_DYNAPI_PROC(int,SDL_UIKitRunApp,(int a, char *b, SDL_main_func c),(a,b,c),return)
 #endif
+SDL_DYNAPI_PROC(size_t,SDL_SIMDGetAlignment,(void),(),return)
+SDL_DYNAPI_PROC(void*,SDL_SIMDAlloc,(const size_t a),(a),return)
+SDL_DYNAPI_PROC(void,SDL_SIMDFree,(void *a),(a),)
diff --git a/src/video/SDL_RLEaccel.c b/src/video/SDL_RLEaccel.c
@@ -89,7 +89,6 @@
 #include "SDL_sysvideo.h"
 #include "SDL_blit.h"
 #include "SDL_RLEaccel_c.h"
-#include "../cpuinfo/SDL_simd.h"
 
 #ifndef MIN
 #define MIN(a, b) ((a) < (b) ? (a) : (b))

diff --git a/src/video/SDL_surface.c b/src/video/SDL_surface.c
@@ -26,7 +26,6 @@
 #include "SDL_RLEaccel_c.h"
 #include "SDL_pixels_c.h"
 #include "SDL_yuv_c.h"
-#include "../cpuinfo/SDL_simd.h"
 
 
 /* Check to make sure we can safely check multiplication of surface w and pitch and it won't overflow size_t */