cuda, metal : fix nans in soft_max (#5574)

* cuda : fix nans in soft_max

* metal : fix nans in soft_max

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

2 files changed, 8 insertions, 8 deletions

diff --git a/ggml-cuda.cu b/ggml-cuda.cu
index 933ebbc4..eef21350 100644
--- a/ggml-cuda.cu
+++ b/ggml-cuda.cu
@@ -6205,7 +6205,7 @@ static __global__ void soft_max_f32(const float * x, const float * mask, const f
         const int ix = rowx*ncols + col;
         const int iy = rowy*ncols + col;
 
-        const float val = x[ix]*scale + (mask ? mask[iy] : 0.0f) + slope*pos[col];
+        const float val = x[ix]*scale + (mask ? mask[iy] : 0.0f) + (pos ? slope*pos[col] : 0.0f);
 
         vals[col] = val;
         max_val = max(max_val, val);
@@ -9170,17 +9170,17 @@ static void ggml_cuda_op_soft_max(
     memcpy(&max_bias, (float *) dst->op_params + 1, sizeof(float));
 
     // positions tensor
-    float * src2_dd = dst_dd; // default to avoid null checks in the kernel
+    float * src2_dd = nullptr;
     cuda_pool_alloc<float> src2_f;
 
     ggml_tensor * src2 = dst->src[2];
     const bool use_src2 = src2 != nullptr;
 
     if (use_src2) {
-        const bool src2_on_device = use_src2 && src2->backend == GGML_BACKEND_GPU;
-        ggml_tensor_extra_gpu * src2_extra = use_src2 ? (ggml_tensor_extra_gpu *) src2->extra : nullptr;
+        const bool src2_on_device = src2->backend == GGML_BACKEND_GPU;
 
         if (src2_on_device) {
+            ggml_tensor_extra_gpu * src2_extra = (ggml_tensor_extra_gpu *) src2->extra;
             src2_dd = (float *) src2_extra->data_device[g_main_device];
         } else {
             src2_dd = src2_f.alloc(ggml_nelements(src2));
diff --git a/ggml-metal.metal b/ggml-metal.metal
index d0a85a19..f0d77d44 100644
--- a/ggml-metal.metal
+++ b/ggml-metal.metal
@@ -392,7 +392,7 @@ kernel void kernel_soft_max(
     float lmax = -INFINITY;
 
     for (int i00 = tpitg; i00 < ne00; i00 += ntg) {
-        lmax = MAX(lmax, psrc0[i00]*scale + (pmask ? pmask[i00] : 0.0f) + slope*ppos[i00]);
+        lmax = MAX(lmax, psrc0[i00]*scale + (pmask ? pmask[i00] : 0.0f) + (ppos ? slope*ppos[i00] : 0.0f));
     }
 
     // find the max value in the block
@@ -417,7 +417,7 @@ kernel void kernel_soft_max(
     // parallel sum
     float lsum = 0.0f;
     for (int i00 = tpitg; i00 < ne00; i00 += ntg) {
-        const float exp_psrc0 = exp((psrc0[i00]*scale + (pmask ? pmask[i00] : 0.0f) + slope*ppos[i00]) - max_val);
+        const float exp_psrc0 = exp((psrc0[i00]*scale + (pmask ? pmask[i00] : 0.0f) + (ppos ? slope*ppos[i00] : 0.0f)) - max_val);
         lsum += exp_psrc0;
         pdst[i00] = exp_psrc0;
     }
@@ -495,7 +495,7 @@ kernel void kernel_soft_max_4(
     float4 lmax4 = -INFINITY;
 
     for (int i00 = tpitg; i00 < ne00/4; i00 += ntg) {
-        lmax4 = fmax(lmax4, psrc4[i00]*scale + (pmask ? pmask[i00] : 0.0f) + slope*ppos[i00]);
+        lmax4 = fmax(lmax4, psrc4[i00]*scale + (pmask ? pmask[i00] : 0.0f) + (ppos ? slope*ppos[i00] : 0.0f));
     }
 
     const float lmax = MAX(MAX(lmax4[0], lmax4[1]), MAX(lmax4[2], lmax4[3]));
@@ -521,7 +521,7 @@ kernel void kernel_soft_max_4(
     // parallel sum
     float4 lsum4 = 0.0f;
     for (int i00 = tpitg; i00 < ne00/4; i00 += ntg) {
-        const float4 exp_psrc4 = exp((psrc4[i00]*scale + (pmask ? pmask[i00] : 0.0f) + slope*ppos[i00]) - max_val);
+        const float4 exp_psrc4 = exp((psrc4[i00]*scale + (pmask ? pmask[i00] : 0.0f) + (ppos ? slope*ppos[i00] : 0.0f)) - max_val);
         lsum4 += exp_psrc4;
         pdst4[i00] = exp_psrc4;
     }