fix(patch): make Windows arg-parse patch deterministic (drop override)

The count-guard variant of 0001-win32-arg-parse-embed-guard.patch fixed 21/25
Windows Java tests but collided on the 4 server-integration setups (Rerank,
ToolCalling, Multimodal, OpenAiCompatServer) whose argv length happened to equal
java.exe's process arg count — the guard then wrongly adopted java.exe's command
line and they kept failing with "Failed to parse model parameters". Those tests
pass on Linux/macOS, so their args are valid; the collision was the only cause.

Switch to the deterministic fix: keep the make_utf8_argv() call referenced (no
-Wunused-function) but never adopt its result, so the caller's already-UTF-8 JNI
argv is always used. A JNI library is never the process, so the GetCommandLineW
override is pure liability for us. Verified the patch applies cleanly to b9739
and the applier stays idempotent. CLAUDE.md + TODO.md updated to record the
count-guard -> removal change; upstream PR can still expose an opt-out that
preserves the standalone tools' UTF-8 fix.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SfvSZ76NW4e1qX1PjL4RKq

Author: claude

CLAUDE.md

@@ -355,7 +355,7 @@ Current patches:
 
 | Patch | Fixes |
 |-------|-------|
-| `0001-win32-arg-parse-embed-guard.patch` | Windows JNI regression from llama.cpp **#24779** (b9739): `common_params_parse` unconditionally replaced the caller's argv with the process command line (`GetCommandLineW`), so an embedded/JNI caller (`java.exe`) lost its `--model …` args → "Failed to parse model parameters". The patch guards the override to fire **only when the re-derived arg count equals `argc`** — true for the standalone `llama-*` tools (their UTF-8 CLI fix is preserved), false for a JVM host (our already-UTF-8 argv is kept). This is also the shape to PR upstream. |
+| `0001-win32-arg-parse-embed-guard.patch` | Windows JNI regression from llama.cpp **#24779** (b9739): `common_params_parse` unconditionally replaced the caller's argv with the process command line (`GetCommandLineW`), so an embedded/JNI caller (`java.exe`) lost its `--model …` args → "Failed to parse model parameters". The patch **drops the override for our build** (keeps the `make_utf8_argv()` call referenced so there's no `-Wunused-function`, but never adopts its result), so the caller's already-UTF-8 argv is always used. This is **deterministic** — an earlier count-guard variant (only override when the re-derived arg count equals `argc`) collided on the server-integration tests whose argv length happened to equal `java.exe`'s and kept them failing. The upstream PR can instead expose an opt-out / `common_params_parse_argv` that preserves the standalone tools' UTF-8 fix. |
 
 ## Upgrading/Downgrading llama.cpp Version
 

TODO.md

@@ -139,11 +139,19 @@ proving Ninja Multi-Config + MSVC works on the same tree). The two builds produc
 
 **Status: FIXED via local source patch (`patches/0001-win32-arg-parse-embed-guard.patch`).** Surfaced
 while bringing PR #248 green (the b9739 build fixes let the Windows Java jobs run to completion and
-exposed this). Resolved by **fix option 1 below** — the count-guard — applied through the generic
-`patches/` mechanism (see CLAUDE.md "Local llama.cpp source patches"), so it covers every C++ build
-and re-applies on each clean build. Still worth upstreaming (the guard, or a `common_params_parse_argv`
-companion) so the patch can eventually be dropped; until then it must be re-verified on each llama.cpp
-bump (the applier fails loud if it no longer applies).
+exposed this). Applied through the generic `patches/` mechanism (see CLAUDE.md "Local llama.cpp source
+patches"), so it covers every C++ build and re-applies on each clean build.
+
+**Note on the fix shape (count-guard → deterministic removal).** The first patch used fix option 1
+below — the count-guard (override only when the re-derived arg count equals `argc`). It fixed 21/25
+Windows Java tests, but **collided** on the 4 server-integration setups (`OpenAiServerRerank*`,
+`OpenAiServerToolCalling*`, `MultimodalIntegrationTest`, `OpenAiCompatServerIntegrationTest`) whose
+argv length happened to equal `java.exe`'s, so they kept failing with the same parse error. The patch
+was changed to **fix option 2** (drop the override entirely for our build — a JNI library is never the
+process, so the override is pure liability), which is deterministic. Still worth upstreaming as an
+opt-out / `common_params_parse_argv` that preserves the standalone tools' UTF-8 fix, so the patch can
+eventually be dropped; until then it must be re-verified on each llama.cpp bump (the applier fails loud
+if it no longer applies).
 
 **Symptom.** On **Windows x86_64 only**, every Java test that loads a real model fails in
 `LlamaModel.loadModel` (native) with `LlamaException: "Failed to parse model parameters"`

patches/0001-win32-arg-parse-embed-guard.patch

@@ -1,19 +1,25 @@
 diff --git a/common/arg.cpp b/common/arg.cpp
 --- a/common/arg.cpp
 +++ b/common/arg.cpp
-@@ -924,7 +924,14 @@ bool common_params_parse(int argc, char ** argv, common_params & params, llama_e
+@@ -924,10 +924,17 @@ bool common_params_parse(int argc, char ** argv, common_params & params, llama_e
  bool common_params_parse(int argc, char ** argv, common_params & params, llama_example ex, void(*print_usage)(int, char **)) {
  #ifdef _WIN32
      auto utf8 = make_utf8_argv();
 -    if (!utf8.ptrs.empty()) {
-+    // java-llama.cpp patch (PR #248): only adopt the process command line (GetCommandLineW) when
-+    // the caller actually passed THIS process's own argv -- i.e. the re-derived argument count
-+    // matches argc. For the standalone llama-* tools that is always true, so their UTF-8 CLI fix
-+    // (upstream llama.cpp #24779) is preserved. For an embedded JNI caller the process is java.exe
-+    // (many more args), so the counts differ and our already-UTF-8 argv (from GetStringUTFChars)
-+    // is kept instead of being silently discarded -- which otherwise makes common_params_parse_ex
-+    // parse java.exe's command line and fail with "Failed to parse model parameters".
-+    if (!utf8.ptrs.empty() && static_cast<int>(utf8.buf.size()) == argc) {
-         argc = static_cast<int>(utf8.buf.size());
-         argv = utf8.ptrs.data();
-     }
+-        argc = static_cast<int>(utf8.buf.size());
+-        argv = utf8.ptrs.data();
+-    }
++    // java-llama.cpp patch (PR #248): upstream (llama.cpp #24779) replaced the caller's argv with
++    // the process command line (GetCommandLineW) here to recover UTF-8 args for the standalone
++    // llama-* tools. libjllama is a JNI library, never the process: it already passes correct
++    // UTF-8 argv (GetStringUTFChars), and adopting GetCommandLineW discarded it -> common_params_parse
++    // parsed java.exe's command line and failed with "Failed to parse model parameters". We keep the
++    // make_utf8_argv() call (so it stays referenced -> -Wunused-function-clean) but do NOT adopt its
++    // result, so the caller's already-UTF-8 argv is always used. This is deterministic: a count-guard
++    // (only override when the re-derived arg count equals argc) collided on the server-integration
++    // tests whose argv length happened to equal java.exe's, so they kept failing. The upstream PR
++    // can instead expose an opt-out / a common_params_parse_argv that preserves the standalone fix.
++    (void) utf8;
+ #endif
+
+     auto ctx_arg = common_params_parser_init(params, ex, print_usage);