From eb61e1d77012a087f3afc76ba1f95c275c5e7ab7 Mon Sep 17 00:00:00 2001
From: Ettore Di Giacinto <mudler@localai.io>
Date: Thu, 11 Jun 2026 17:17:54 +0000
Subject: [PATCH] chore(dllm): review fixes - file modes and build-matrix doc
 accuracy

Drop the stray executable bit from the Go sources and Makefile (the
sibling Go backends commit them 644; only run.sh/package.sh are
executable), and correct two documentation claims found in the final
branch review: cuda13-dllm is built for amd64 only (arm64 CUDA ships as
the l4t flavor), and package.sh is the parakeet-cpp-style stub layout
with no ldd walk.

Assisted-by: Claude Code (Fable 5)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
---
 .agents/dllm-backend.md                  | 9 +++++----
 backend/go/dllm/Makefile                 | 0
 backend/go/dllm/capi.go                  | 0
 backend/go/dllm/dllm.go                  | 0
 backend/go/dllm/dllm_test.go             | 0
 backend/go/dllm/gemma4_parser.go         | 0
 backend/go/dllm/gemma4_parser_test.go    | 0
 backend/go/dllm/gemma4_renderer.go       | 0
 backend/go/dllm/gemma4_renderer_test.go  | 0
 backend/go/dllm/main.go                  | 0
 docs/content/features/text-generation.md | 4 ++--
 11 files changed, 7 insertions(+), 6 deletions(-)
 mode change 100755 => 100644 backend/go/dllm/Makefile
 mode change 100755 => 100644 backend/go/dllm/capi.go
 mode change 100755 => 100644 backend/go/dllm/dllm.go
 mode change 100755 => 100644 backend/go/dllm/dllm_test.go
 mode change 100755 => 100644 backend/go/dllm/gemma4_parser.go
 mode change 100755 => 100644 backend/go/dllm/gemma4_parser_test.go
 mode change 100755 => 100644 backend/go/dllm/gemma4_renderer.go
 mode change 100755 => 100644 backend/go/dllm/gemma4_renderer_test.go
 mode change 100755 => 100644 backend/go/dllm/main.go

diff --git a/.agents/dllm-backend.md b/.agents/dllm-backend.md
index 6cd8c10dd..9f8586235 100644
--- a/.agents/dllm-backend.md
+++ b/.agents/dllm-backend.md
@@ -112,13 +112,14 @@ carry that coverage.
 
 ## Build matrix
 
-`cpu-dllm` (amd64 + arm64), `cuda13-dllm` (amd64 + arm64), and
-`cuda13-nvidia-l4t-arm64-dllm` (Jetson / DGX Spark GB10), via
+`cpu-dllm` (amd64 + arm64), `cuda13-dllm` (amd64), and
+`cuda13-nvidia-l4t-arm64-dllm` (arm64 CUDA: Jetson / DGX Spark GB10), via
 `.github/backend-matrix.yml`. No darwin/Metal. CUDA builds forward
 `-DDLLM_CUDA=ON` (dllm.cpp gates ggml's CUDA behind its own flag - a bare
 `-DGGML_CUDA=ON` is overridden by the cache FORCE). `libdllm.so` is
-self-contained (ggml statically absorbed, PIC), so packaging only ships the
-one .so plus the usual ldd walk.
+self-contained (ggml statically absorbed, PIC), so `package.sh` only ships
+the binary, `run.sh` and that one .so (the parakeet-cpp-style stub layout;
+no ldd walk yet).
 
 ## Known limitations
 
diff --git a/backend/go/dllm/Makefile b/backend/go/dllm/Makefile
old mode 100755
new mode 100644
diff --git a/backend/go/dllm/capi.go b/backend/go/dllm/capi.go
old mode 100755
new mode 100644
diff --git a/backend/go/dllm/dllm.go b/backend/go/dllm/dllm.go
old mode 100755
new mode 100644
diff --git a/backend/go/dllm/dllm_test.go b/backend/go/dllm/dllm_test.go
old mode 100755
new mode 100644
diff --git a/backend/go/dllm/gemma4_parser.go b/backend/go/dllm/gemma4_parser.go
old mode 100755
new mode 100644
diff --git a/backend/go/dllm/gemma4_parser_test.go b/backend/go/dllm/gemma4_parser_test.go
old mode 100755
new mode 100644
diff --git a/backend/go/dllm/gemma4_renderer.go b/backend/go/dllm/gemma4_renderer.go
old mode 100755
new mode 100644
diff --git a/backend/go/dllm/gemma4_renderer_test.go b/backend/go/dllm/gemma4_renderer_test.go
old mode 100755
new mode 100644
diff --git a/backend/go/dllm/main.go b/backend/go/dllm/main.go
old mode 100755
new mode 100644
diff --git a/docs/content/features/text-generation.md b/docs/content/features/text-generation.md
index 04b481312..b6232cbb5 100644
--- a/docs/content/features/text-generation.md
+++ b/docs/content/features/text-generation.md
@@ -676,8 +676,8 @@ This backend is **experimental**, and the engine does not yet have a prompt-KV p
 | Flavor | Hardware |
 |---|---|
 | `cpu-dllm` | CPU (amd64 + arm64) - functional but very slow on the 26B model; mainly useful for wiring tests |
-| `cuda13-dllm` | NVIDIA CUDA 13 (amd64 + arm64) |
-| `cuda13-nvidia-l4t-arm64-dllm` | NVIDIA L4T (Jetson / DGX Spark GB10) |
+| `cuda13-dllm` | NVIDIA CUDA 13 (amd64) |
+| `cuda13-nvidia-l4t-arm64-dllm` | NVIDIA L4T arm64 (Jetson / DGX Spark GB10) |
 
 macOS/Metal is not available yet.