git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH v2 6/6] max_tree_depth: lower it for clangarm64 on Windows

From
Johannes Schindelin via GitGitGadget <gitgitgadget@gmail.com>
Date
Apr 23, 2025, 08:01 UTC
Message-ID
<e0e78bd5131a9efa64697cd7c0bd30965d13b41c.1745395308.git.gitgitgadget@gmail.com>
In-Reply-To
<pull.1904.v2.git.1745395308.gitgitgadget@gmail.com>
From: Johannes Schindelin <johannes.schindelin@gmx.de>

Just as in b64d78ad02ca (max_tree_depth: lower it for MSVC to avoid stack overflows, 2023-11-01), I encountered the same problem with the clang builds on Windows/ARM64.

The symptom is an exit code 127 when t6700 tries to verify that `git archive big` fails.

This exit code is reserved on Unix/Linux to mean "command not found". Unfortunately in this case, it is the fall-back chosen by Cygwin's `pinfo::status_exit()` method when encountering the NSTATUS `STATUS_STACK_OVERFLOW`, see https://github.com/cygwin/cygwin/blob/cygwin-3.6.1/winsup/cygwin/pinfo.cc#L171

I verified manually that the stack overflow always happens somewhere around tree depth 1403, therefore 1280 should be a safe bound in these instances.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
---
 environment.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
diff --git a/environment.c b/environment.c
index 9e4c7781be0..d948bb3c705 100644
--- a/environment.c
+++ b/environment.c
@@ -82,6 +82,16 @@ int max_allowed_tree_depth =
 	 * the stack overflow can occur.
 	 */
 	512;
+#elif defined(GIT_WINDOWS_NATIVE) && defined(__clang__) && defined(__aarch64__)
+	/*
+	 * Similar to Visual C, it seems that on Windows/ARM64 the clang-based
+	 * builds have a smaller stack space available. When running out of
+	 * that stack space, a `STATUS_STACK_OVERFLOW` is produced. When the
+	 * Git command was run from an MSYS2 Bash, this unfortunately results
+	 * in an exit code 127. Let's prevent that by lowering the maximal
+	 * tree depth; This value seems to be low enough.
+	 */
+	1280;
 #else
 	2048;
 #endif
-- 
gitgitgadget
Previous: Johannes Schindelin via GitGitGadget
Message 20 of 20 in “Support Windows/ARM64”
  1. 0/6 Support Windows/ARM64Johannes Schindelin via GitGitGadget, Apr 21, 2025
  2. 1/6 bswap.h: add support for built-in bswap functionsDennis Ameling via GitGitGadget, Apr 21, 2025
  3. 2/6 config.mak.uname: add support for clangarm64Dennis Ameling via GitGitGadget, Apr 21, 2025
  4. 3/6 mingw: do not use nedmalloc on Windows/ARM64Johannes Schindelin via GitGitGadget, Apr 21, 2025
  5. Patrick SteinhardtApr 22, 2025
  6. Dropping nedmalloc support? was Re: [PATCH 3/6] mingw: do not use nedmalloc on Windows/ARM64Johannes Schindelin, Apr 22, 2025
  7. 4/6 msvc: do handle builds on Windows/ARM64Johannes Schindelin via GitGitGadget, Apr 21, 2025
  8. 6/6 max_tree_depth: lower it for clangarm64 on WindowsJohannes Schindelin via GitGitGadget, Apr 21, 2025
  9. Patrick SteinhardtApr 22, 2025
  10. Johannes SchindelinApr 22, 2025
  11. 5/6 mingw(arm64): do move the `/etc/git*` locationJohannes Schindelin via GitGitGadget, Apr 21, 2025
  12. Junio C HamanoApr 21, 2025
  13. 0/6 Support Windows/ARM64Johannes Schindelin via GitGitGadget, Apr 23, 2025
  14. 1/6 bswap.h: add support for built-in bswap functionsDennis Ameling via GitGitGadget, Apr 23, 2025
  15. 2/6 config.mak.uname: add support for clangarm64Dennis Ameling via GitGitGadget, Apr 23, 2025
  16. 3/6 mingw: do not use nedmalloc on Windows/ARM64Johannes Schindelin via GitGitGadget, Apr 23, 2025
  17. Junio C HamanoApr 23, 2025
  18. 4/6 msvc: do handle builds on Windows/ARM64Johannes Schindelin via GitGitGadget, Apr 23, 2025
  19. 5/6 mingw(arm64): do move the `/etc/git*` locationJohannes Schindelin via GitGitGadget, Apr 23, 2025
  20. 6/6 max_tree_depth: lower it for clangarm64 on WindowsJohannes Schindelin via GitGitGadget, Apr 23, 2025

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.