Merge mainline - Aug 12 2024 (#17)

* Merge mainline

* Fix after merge

* Remove CI check

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>

1 files changed, 5 insertions, 1 deletions

diff --git a/ggml/src/vulkan-shaders/generic_binary_head.comp b/ggml/src/vulkan-shaders/generic_binary_head.comp
index ab45d256..b6beaff1 100644
--- a/ggml/src/vulkan-shaders/generic_binary_head.comp
+++ b/ggml/src/vulkan-shaders/generic_binary_head.comp
@@ -7,7 +7,7 @@ layout (push_constant) uniform parameter
     uint ne10; uint ne11; uint ne12; uint ne13; uint nb10; uint nb11; uint nb12; uint nb13;
     uint ne20; uint ne21; uint ne22; uint ne23; uint nb20; uint nb21; uint nb22; uint nb23;
     uint d_offset;
-    float param1; float param2;
+    float param1; float param2; int param3;
 } p;
 
 layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
@@ -16,6 +16,10 @@ layout (binding = 0) readonly buffer A {A_TYPE data_a[];};
 layout (binding = 1) readonly buffer B {B_TYPE data_b[];};
 layout (binding = 2) writeonly buffer D {D_TYPE data_d[];};
 
+uint get_idx() {
+    return gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
+}
+
 uint src0_idx(uint idx) {
     const uint i03 = idx / (p.ne02*p.ne01*p.ne00);
     const uint i03_offset = i03 * p.ne02*p.ne01*p.ne00;