evas: enable NEON-optimized code for aarch64.

Summary:
Add new define, BUILD_NEON_INTRINSICS to control whether NEON inline code or
NEON intrinsics should be built.

GCC NEON intrinsics can be built both for armv7 and armv8. However NEON inline
code can be built only for armv7.

@feature

Reviewers: raster, stefan_schmidt, cedric

Subscribers: cedric, stefan_schmidt

Projects: #efl

Differential Revision: https://phab.enlightenment.org/D2309

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>

1 files changed, 9 insertions, 0 deletions

diff --git a/src/lib/evas/common/evas_op_copy/op_copy_color_neon.c b/src/lib/evas/common/evas_op_copy/op_copy_color_neon.c
index 96310cdf3a..009bd750ea 100644
--- a/src/lib/evas/common/evas_op_copy/op_copy_color_neon.c
+++ b/src/lib/evas/common/evas_op_copy/op_copy_color_neon.c
@@ -3,6 +3,14 @@
 #ifdef BUILD_NEON
 static void
 _op_copy_c_dp_neon(DATA32 *s, DATA8 *m, DATA32 c, DATA32 *d, int l) {
+#ifdef BUILD_NEON_INTRINSICS
+   DATA32 *e;
+   UNROLL8_PLD_WHILE(d, l, e,
+                     {
+                        *d = c;
+                        d++;
+                     });
+#else
 #define AP "COPY_C_DP_"
    uint32_t *e = d + l,*tmp;
    asm volatile (
@@ -85,6 +93,7 @@ _op_copy_c_dp_neon(DATA32 *s, DATA8 *m, DATA32 c, DATA32 *d, int l) {
 
 
    );
+#endif
 }
 
 #define _op_copy_cn_dp_neon _op_copy_c_dp_neon