From a65c4104634911b78c99ef81c768530f792cbda9 Mon Sep 17 00:00:00 2001
From: leejet <leejet714@gmail.com>
Date: Sat, 2 Mar 2024 19:49:41 +0800
Subject: [PATCH] sync: update ggml

---
 ggml               |  2 +-
 ggml_extend.hpp    |  2 +-
 stable-diffusion.h | 15 +++++++++------
 3 files changed, 11 insertions(+), 8 deletions(-)

diff --git a/ggml b/ggml
index b94c066..9cc5cb2 160000
--- a/ggml
+++ b/ggml
@@ -1 +1 @@
-Subproject commit b94c066ad4de62bda2a10a80131d4cd8112f7c65
+Subproject commit 9cc5cb27f97821482fbb99aea1e590e55bf89151
diff --git a/ggml_extend.hpp b/ggml_extend.hpp
index ddcdd29..c2d6552 100644
--- a/ggml_extend.hpp
+++ b/ggml_extend.hpp
@@ -846,7 +846,7 @@ public:
             return NULL;
         }
         // it's performing a compute, check if backend isn't cpu
-        if (!ggml_backend_is_cpu(backend) && tensor->backend == GGML_BACKEND_CPU) {
+        if (!ggml_backend_is_cpu(backend) && tensor->backend == GGML_BACKEND_TYPE_CPU) {
             // pass input tensors to gpu memory
             auto backend_tensor = ggml_dup_tensor(compute_ctx, tensor);
 
diff --git a/stable-diffusion.h b/stable-diffusion.h
index 5d1476f..1b32111 100644
--- a/stable-diffusion.h
+++ b/stable-diffusion.h
@@ -65,17 +65,20 @@ enum sd_type_t {
     SD_TYPE_Q8_0 = 8,
     SD_TYPE_Q8_1 = 9,
     // k-quantizations
-    SD_TYPE_Q2_K    = 10,
-    SD_TYPE_Q3_K    = 11,
-    SD_TYPE_Q4_K    = 12,
-    SD_TYPE_Q5_K    = 13,
-    SD_TYPE_Q6_K    = 14,
-    SD_TYPE_Q8_K    = 15,
+    SD_TYPE_Q2_K = 10,
+    SD_TYPE_Q3_K = 11,
+    SD_TYPE_Q4_K = 12,
+    SD_TYPE_Q5_K = 13,
+    SD_TYPE_Q6_K = 14,
+    SD_TYPE_Q8_K = 15,
     SD_TYPE_IQ2_XXS = 16,
     SD_TYPE_IQ2_XS  = 17,
     SD_TYPE_IQ3_XXS = 18,
     SD_TYPE_IQ1_S   = 19,
     SD_TYPE_IQ4_NL  = 20,
+    SD_TYPE_IQ3_S   = 21,
+    SD_TYPE_IQ2_S   = 22,
+    SD_TYPE_IQ4_XS  = 23,
     SD_TYPE_I8,
     SD_TYPE_I16,
     SD_TYPE_I32,