refactor: unify ggml operation helpers around runner context (#2314)

This commit is contained in:
leejet authored and GitHub committed 2026-11-10 01:58:04 +08:00
1 parent 73fef08224
commit a3c49b13be
76 files changed
+1983 -1831

No files matched your search

+7 -7
View File
@@ -216,7 +216,7 @@ protected:
}
ggml_tensor* get_alpha(GGMLRunnerContext* ctx) {
auto mix_factor = ggml_ext_cast_f32(ctx->ggml_ctx, ctx->backend, params["mix_factor"]);
auto mix_factor = ggml_ext_cast_f32(ctx, params["mix_factor"]);
return ggml_sigmoid(ctx->ggml_ctx, mix_factor);
}
@@ -601,11 +601,11 @@ public:
int64_t H = h * p;
int64_t W = w * p;
z = ggml_reshape_4d(ctx->ggml_ctx, z, w * h, p * p, C, N); // [N, C, p*p, h*w]
z = ggml_cont(ctx->ggml_ctx, ggml_ext_torch_permute(ctx->ggml_ctx, z, 1, 0, 2, 3)); // [N, C, h*w, p*p]
z = ggml_reshape_4d(ctx->ggml_ctx, z, p, p, w, h * C * N); // [N*C*h, w, p, p]
z = ggml_cont(ctx->ggml_ctx, ggml_ext_torch_permute(ctx->ggml_ctx, z, 0, 2, 1, 3)); // [N*C*h, p, w, p]
z = ggml_reshape_4d(ctx->ggml_ctx, z, W, H, C, N); // [N, C, h*p, w*p]
z = ggml_reshape_4d(ctx->ggml_ctx, z, w * h, p * p, C, N); // [N, C, p*p, h*w]
z = ggml_cont(ctx->ggml_ctx, ggml_ext_torch_permute(ctx, z, 1, 0, 2, 3)); // [N, C, h*w, p*p]
z = ggml_reshape_4d(ctx->ggml_ctx, z, p, p, w, h * C * N); // [N*C*h, w, p, p]
z = ggml_cont(ctx->ggml_ctx, ggml_ext_torch_permute(ctx, z, 0, 2, 1, 3)); // [N*C*h, p, w, p]
z = ggml_reshape_4d(ctx->ggml_ctx, z, W, H, C, N); // [N, C, h*p, w*p]
}
if (use_quant) {
@@ -632,7 +632,7 @@ public:
// sd::ggml_graph_cut::mark_graph_cut(z, "vae.encode.final", "z");
}
if (sd_version_uses_flux2_vae(version)) {
z = ggml_ext_chunk(ctx->ggml_ctx, z, 2, 2)[0];
z = ggml_ext_chunk(ctx, z, 2, 2)[0];
// [N, C, H, W] -> [N, C*p*p, H/p, W/p]
int64_t p = 2;