libavutil: Make LOCAL_ALIGNED(xx be equal to LOCAL_ALIGNED_xx(

Previously, the former form always produced a manually aligned,
padded buffer, while the latter can use DECLARE_ALIGNED, if that
amount of stack alignment is supported.

libavutil/internal.h needs to include mem.h, since it uses
the DECLARE_ALIGNED macro.
Signed-off-by: Martin Storsjö <martin@martin.st>

libavutil/internal.h

@@ -38,6 +38,7 @@
 #include "attributes.h"
 #include "dict.h"
 #include "macros.h"
+#include "mem.h"
 #include "pixfmt.h"
 
 #if ARCH_X86
@@ -97,24 +98,24 @@
     DECLARE_ALIGNED(a, t, la_##v) s o;                  \
     t (*v) o = la_##v
 
-#define LOCAL_ALIGNED(a, t, v, ...) E1(LOCAL_ALIGNED_A(a, t, v, __VA_ARGS__,,))
+#define LOCAL_ALIGNED(a, t, v, ...) LOCAL_ALIGNED_##a(t, v, __VA_ARGS__)
 
 #if HAVE_LOCAL_ALIGNED_8
 #   define LOCAL_ALIGNED_8(t, v, ...) E1(LOCAL_ALIGNED_D(8, t, v, __VA_ARGS__,,))
 #else
-#   define LOCAL_ALIGNED_8(t, v, ...) LOCAL_ALIGNED(8, t, v, __VA_ARGS__)
+#   define LOCAL_ALIGNED_8(t, v, ...) E1(LOCAL_ALIGNED_A(8, t, v, __VA_ARGS__,,))
 #endif
 
 #if HAVE_LOCAL_ALIGNED_16
 #   define LOCAL_ALIGNED_16(t, v, ...) E1(LOCAL_ALIGNED_D(16, t, v, __VA_ARGS__,,))
 #else
-#   define LOCAL_ALIGNED_16(t, v, ...) LOCAL_ALIGNED(16, t, v, __VA_ARGS__)
+#   define LOCAL_ALIGNED_16(t, v, ...) E1(LOCAL_ALIGNED_A(16, t, v, __VA_ARGS__,,))
 #endif
 
 #if HAVE_LOCAL_ALIGNED_32
 #   define LOCAL_ALIGNED_32(t, v, ...) E1(LOCAL_ALIGNED_D(32, t, v, __VA_ARGS__,,))
 #else
-#   define LOCAL_ALIGNED_32(t, v, ...) LOCAL_ALIGNED(32, t, v, __VA_ARGS__)
+#   define LOCAL_ALIGNED_32(t, v, ...) E1(LOCAL_ALIGNED_A(32, t, v, __VA_ARGS__,,))
 #endif
 
 #define FF_ALLOC_OR_GOTO(ctx, p, size, label)\