From 8fbd59308b54729a191dcf3aee3388abfa7dd6e3 Mon Sep 17 00:00:00 2001
From: Francis Couture-Harpin <git@compilade.net>
Date: Fri, 28 Jun 2024 22:52:57 -0400
Subject: [PATCH] ggml-quants : attempt to fix Arm 32-bit support

---
 ggml/src/ggml-impl.h   | 11 ++++-------
 ggml/src/ggml-quants.c |  8 ++++----
 2 files changed, 8 insertions(+), 11 deletions(-)

diff --git a/ggml/src/ggml-impl.h b/ggml/src/ggml-impl.h
index 1d2336190..374b5ae04 100644
--- a/ggml/src/ggml-impl.h
+++ b/ggml/src/ggml-impl.h
@@ -177,7 +177,7 @@ typedef __fp16 ggml_fp16_internal_t;
 
 // 32-bit ARM compatibility
 
-// vaddvq_s16
+// vaddlvq_s16
 // vpaddq_s16
 // vpaddq_s32
 // vaddvq_s32
@@ -187,12 +187,9 @@ typedef __fp16 ggml_fp16_internal_t;
 // vzip1_u8
 // vzip2_u8
 
-inline static int32_t vaddvq_s16(int16x8_t v) {
-    return
-        (int32_t)vgetq_lane_s16(v, 0) + (int32_t)vgetq_lane_s16(v, 1) +
-        (int32_t)vgetq_lane_s16(v, 2) + (int32_t)vgetq_lane_s16(v, 3) +
-        (int32_t)vgetq_lane_s16(v, 4) + (int32_t)vgetq_lane_s16(v, 5) +
-        (int32_t)vgetq_lane_s16(v, 6) + (int32_t)vgetq_lane_s16(v, 7);
+inline static int32_t vaddlvq_s16(int16x8_t v) {
+    int32x4_t v0 = vreinterpretq_s32_s64(vpaddlq_s32(vpaddlq_s16(v)));
+    return vgetq_lane_s32(v0, 0) + vgetq_lane_s32(v0, 2);
 }
 
 inline static int16x8_t vpaddq_s16(int16x8_t a, int16x8_t b) {
diff --git a/ggml/src/ggml-quants.c b/ggml/src/ggml-quants.c
index 1f7460ac4..df4320f5b 100644
--- a/ggml/src/ggml-quants.c
+++ b/ggml/src/ggml-quants.c
@@ -11483,10 +11483,10 @@ void ggml_vec_dot_q1_3_q8_0(int n, float * restrict s, size_t bs, const void * r
         // WARNING: reading 3 bytes further than necessary
         const uint8x16_t x13b = vld1q_u8((const uint8_t *) x);
 
-        uint8x16_t x0 = vqtbl1q_u8(x13b, mask0);
-        uint8x16_t x1 = vqtbl1q_u8(x13b, mask1);
-        uint8x16_t x2 = vqtbl1q_u8(x13b, mask2);
-        uint8x16_t x3 = vqtbl1q_u8(x13b, mask3);
+        uint8x16_t x0 = ggml_vqtbl1q_u8(x13b, mask0);
+        uint8x16_t x1 = ggml_vqtbl1q_u8(x13b, mask1);
+        uint8x16_t x2 = ggml_vqtbl1q_u8(x13b, mask2);
+        uint8x16_t x3 = ggml_vqtbl1q_u8(x13b, mask3);
 
         x0 = vmulq_u8(x0, shift0);
         x1 = vmulq_u8(x1, shift0);