summaryrefslogtreecommitdiff
path: root/libavresample/arm/audio_convert_neon.S
diff options
context:
space:
mode:
authorMans Rullgard <mans@mansr.com>2012-09-09 19:25:43 +0100
committerMans Rullgard <mans@mansr.com>2012-09-13 19:15:42 +0100
commit643b5b794f095a1d679568a3a6b3106a1dfd6cdb (patch)
tree23e3e93ab7fcac1ffbd61373cca38195ab5db381 /libavresample/arm/audio_convert_neon.S
parentca6b544ac9d07dcd4d151d5729f0bf52f4f8653d (diff)
ARM: libavresample: NEON optimised flat float to s16 conversion
Signed-off-by: Mans Rullgard <mans@mansr.com>
Diffstat (limited to 'libavresample/arm/audio_convert_neon.S')
-rw-r--r--libavresample/arm/audio_convert_neon.S66
1 files changed, 66 insertions, 0 deletions
diff --git a/libavresample/arm/audio_convert_neon.S b/libavresample/arm/audio_convert_neon.S
new file mode 100644
index 0000000000..e82f8c633f
--- /dev/null
+++ b/libavresample/arm/audio_convert_neon.S
@@ -0,0 +1,66 @@
+/*
+ * Copyright (c) 2008 Mans Rullgard <mans@mansr.com>
+ *
+ * This file is part of Libav.
+ *
+ * Libav is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * Libav is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with Libav; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include "config.h"
+#include "libavutil/arm/asm.S"
+
+function ff_conv_flt_to_s16_neon, export=1
+ subs r2, r2, #8
+ vld1.32 {q0}, [r1,:128]!
+ vcvt.s32.f32 q8, q0, #31
+ vld1.32 {q1}, [r1,:128]!
+ vcvt.s32.f32 q9, q1, #31
+ beq 3f
+ bics r12, r2, #15
+ beq 2f
+1: subs r12, r12, #16
+ vqrshrn.s32 d4, q8, #16
+ vld1.32 {q0}, [r1,:128]!
+ vcvt.s32.f32 q0, q0, #31
+ vqrshrn.s32 d5, q9, #16
+ vld1.32 {q1}, [r1,:128]!
+ vcvt.s32.f32 q1, q1, #31
+ vqrshrn.s32 d6, q0, #16
+ vst1.16 {q2}, [r0,:128]!
+ vqrshrn.s32 d7, q1, #16
+ vld1.32 {q8}, [r1,:128]!
+ vcvt.s32.f32 q8, q8, #31
+ vld1.32 {q9}, [r1,:128]!
+ vcvt.s32.f32 q9, q9, #31
+ vst1.16 {q3}, [r0,:128]!
+ bne 1b
+ ands r2, r2, #15
+ beq 3f
+2: vld1.32 {q0}, [r1,:128]!
+ vqrshrn.s32 d4, q8, #16
+ vcvt.s32.f32 q0, q0, #31
+ vld1.32 {q1}, [r1,:128]!
+ vqrshrn.s32 d5, q9, #16
+ vcvt.s32.f32 q1, q1, #31
+ vqrshrn.s32 d6, q0, #16
+ vst1.16 {q2}, [r0,:128]!
+ vqrshrn.s32 d7, q1, #16
+ vst1.16 {q3}, [r0,:128]!
+ bx lr
+3: vqrshrn.s32 d4, q8, #16
+ vqrshrn.s32 d5, q9, #16
+ vst1.16 {q2}, [r0,:128]!
+ bx lr
+endfunc