From: Jeff Bolz <redacted>
Date: Sun, 21 Dec 2025 09:27:34 +0000 (-0600)
Subject: vulkan/cuda: fix topk_moe with exp_probs_b (llama/18071)
X-Git-Tag: upstream/1.8.3~106
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=f407c5e562b5b54799293225531e0196ed395d21;p=pkg%2Fggml%2Fsources%2Fwhisper.cpp

vulkan/cuda: fix topk_moe with exp_probs_b (llama/18071)

I updated test_topk_moe to more closely match llm_graph_context::build_moe_ffn
and added coverage for exp_probs_b and some other missing combinations. This
exposed a bug in both CUDA and Vulkan backends where they were assuming the
input to argsort and the input to get_rows are the same. I'd like to optimize
this graph in another change, but for now just get it functional.

CUDA also had a bug where it got n_experts from the wrong place, leading to
GGML_ASSERT failures in some of the new tests.
---

diff --git a/ggml/src/ggml-cuda/ggml-cuda.cu b/ggml/src/ggml-cuda/ggml-cuda.cu
index ab0f6fe9..55fa2e6a 100644
--- a/ggml/src/ggml-cuda/ggml-cuda.cu
+++ b/ggml/src/ggml-cuda/ggml-cuda.cu
@@ -3076,8 +3076,11 @@ static bool ggml_cuda_can_fuse(const struct ggml_cgraph * cgraph, int node_idx,
         ggml_can_fuse_subgraph(cgraph, node_idx, ops, { node_idx + 3, node_idx + 9 })) {
         ggml_tensor * softmax = cgraph->nodes[node_idx];
         ggml_tensor * weights = cgraph->nodes[node_idx + 9];
+        ggml_tensor * get_rows = cgraph->nodes[node_idx + 4];
+        ggml_tensor * argsort = cgraph->nodes[node_idx + 2];
+        int n_expert = cgraph->nodes[node_idx]->src[0]->ne[0];
 
-        if (ggml_cuda_should_use_topk_moe(softmax, weights)) {
+        if (ggml_cuda_should_use_topk_moe(softmax, weights, get_rows, argsort, nullptr, n_expert)) {
             return true;
         }
     }
@@ -3085,7 +3088,11 @@ static bool ggml_cuda_can_fuse(const struct ggml_cgraph * cgraph, int node_idx,
     if (is_equal(topk_moe_ops, ops) && ggml_can_fuse_subgraph(cgraph, node_idx, ops, { node_idx + 3, node_idx + 4 })) {
         ggml_tensor * softmax = cgraph->nodes[node_idx];
         ggml_tensor * weights = cgraph->nodes[node_idx + 4];
-        if (ggml_cuda_should_use_topk_moe(softmax, weights)) {
+        ggml_tensor * get_rows = cgraph->nodes[node_idx + 4];
+        ggml_tensor * argsort = cgraph->nodes[node_idx + 2];
+        int n_expert = cgraph->nodes[node_idx]->src[0]->ne[0];
+
+        if (ggml_cuda_should_use_topk_moe(softmax, weights, get_rows, argsort, nullptr, n_expert)) {
             return true;
         }
     }
@@ -3094,8 +3101,11 @@ static bool ggml_cuda_can_fuse(const struct ggml_cgraph * cgraph, int node_idx,
         ggml_can_fuse_subgraph(cgraph, node_idx, ops, { node_idx + 1, node_idx + 5 })) {
         ggml_tensor * softmax = cgraph->nodes[node_idx + 4];
         ggml_tensor * weights = cgraph->nodes[node_idx + 5];
+        ggml_tensor * get_rows = cgraph->nodes[node_idx + 2];
+        ggml_tensor * argsort = cgraph->nodes[node_idx + 0];
+        int n_expert = cgraph->nodes[node_idx]->src[0]->ne[0];
 
-        if (ggml_cuda_should_use_topk_moe(softmax, weights)) {
+        if (ggml_cuda_should_use_topk_moe(softmax, weights, get_rows, argsort, nullptr, n_expert)) {
             return true;
         }
     }
diff --git a/ggml/src/ggml-cuda/topk-moe.cu b/ggml/src/ggml-cuda/topk-moe.cu
index 572379fc..48e569ef 100644
--- a/ggml/src/ggml-cuda/topk-moe.cu
+++ b/ggml/src/ggml-cuda/topk-moe.cu
@@ -268,7 +268,23 @@ void ggml_cuda_op_topk_moe(ggml_backend_cuda_context & ctx,
     }
 }
 
-bool ggml_cuda_should_use_topk_moe(const ggml_tensor * softmax, const ggml_tensor * weights, const ggml_tensor * clamp) {
+bool ggml_cuda_should_use_topk_moe(const ggml_tensor * softmax,
+                                   const ggml_tensor * weights,
+                                   const ggml_tensor * get_rows,
+                                   const ggml_tensor * argsort,
+                                   const ggml_tensor * clamp,
+                                   int n_expert) {
+    ggml_tensor * probs = get_rows->src[0];
+    if (probs->op != GGML_OP_RESHAPE) {
+        return false;
+    }
+    probs = probs->src[0];
+    ggml_tensor * selection_probs = argsort->src[0];
+
+    if (probs != selection_probs) {
+        return false;
+    }
+
     float scale    = 1.0f;
     float max_bias = 0.0f;
 
@@ -288,7 +304,6 @@ bool ggml_cuda_should_use_topk_moe(const ggml_tensor * softmax, const ggml_tenso
         return false;
     }
 
-    const int n_expert = softmax->ne[0];
     // n_expert must be a power of 2
     if ((n_expert & (n_expert - 1)) != 0 || n_expert > 512) {
         return false;
diff --git a/ggml/src/ggml-cuda/topk-moe.cuh b/ggml/src/ggml-cuda/topk-moe.cuh
index 2eff408b..6b6c13c5 100644
--- a/ggml/src/ggml-cuda/topk-moe.cuh
+++ b/ggml/src/ggml-cuda/topk-moe.cuh
@@ -11,6 +11,11 @@ void ggml_cuda_op_topk_moe(ggml_backend_cuda_context & ctx,
                            const bool                  delayed_softmax = false,
                            ggml_tensor *               weight_clamp    = nullptr);
 
-bool ggml_cuda_should_use_topk_moe(const ggml_tensor * softmax, const ggml_tensor * weights, const ggml_tensor * clamp = nullptr);
+bool ggml_cuda_should_use_topk_moe(const ggml_tensor * softmax,
+                                   const ggml_tensor * weights,
+                                   const ggml_tensor * get_rows,
+                                   const ggml_tensor * argsort,
+                                   const ggml_tensor * clamp,
+                                   int n_expert);
 
 std::initializer_list<enum ggml_op> ggml_cuda_topk_moe_ops(bool with_norm, bool delayed_softmax = false);
diff --git a/ggml/src/ggml-vulkan/ggml-vulkan.cpp b/ggml/src/ggml-vulkan/ggml-vulkan.cpp
index 1c614536..a5308fa5 100644
--- a/ggml/src/ggml-vulkan/ggml-vulkan.cpp
+++ b/ggml/src/ggml-vulkan/ggml-vulkan.cpp
@@ -12941,24 +12941,43 @@ static bool ggml_vk_can_fuse_topk_moe(ggml_backend_vk_context * ctx, const struc
 
     const ggml_tensor * softmax;
     const ggml_tensor * weights;
+    const ggml_tensor * get_rows;
+    const ggml_tensor * argsort;
 
     switch (mode) {
     case TOPK_MOE_EARLY_SOFTMAX_NORM:
         softmax = cgraph->nodes[node_idx + 0];
         weights = cgraph->nodes[node_idx + 9];
+        get_rows = cgraph->nodes[node_idx + 4];
+        argsort = cgraph->nodes[node_idx + 2];
         break;
     case TOPK_MOE_EARLY_SOFTMAX:
         softmax = cgraph->nodes[node_idx + 0];
         weights = cgraph->nodes[node_idx + 4];
+        get_rows = cgraph->nodes[node_idx + 4];
+        argsort = cgraph->nodes[node_idx + 2];
         break;
     case TOPK_MOE_LATE_SOFTMAX:
         softmax = cgraph->nodes[node_idx + 4];
         weights = cgraph->nodes[node_idx + 5];
+        get_rows = cgraph->nodes[node_idx + 2];
+        argsort = cgraph->nodes[node_idx + 0];
         break;
     default:
         return false;
     }
 
+    ggml_tensor * probs = get_rows->src[0];
+    if (probs->op != GGML_OP_RESHAPE) {
+        return false;
+    }
+    probs = probs->src[0];
+    ggml_tensor * selection_probs = argsort->src[0];
+
+    if (probs != selection_probs) {
+        return false;
+    }
+
     const float * op_params = (const float *)softmax->op_params;
 
     float scale = op_params[0];