selftests/bpf: Add a test for objects stuck in free_by_rcu_ttracegrafted

Delete all elements of BPF_F_NO_PREALLOC hash map in one batch. The first
free_bulk() starts RCU tasks trace GP and the rest of the elements are
freed while it's in flight. Wait for call_rcu_ttrace_in_progress to clear
in bpf_mem_cache of every cpu and check that free_by_rcu_ttrace and
waiting_for_gp_ttrace lists are empty.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20260930095920.601738-4-alexei.starovoitov@gmail.com
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>

1 files changed, 28 insertions, 0 deletions

diff --git a/scripts/remove-stale-files b/scripts/remove-stale-files
new file mode 100755
index 000000000..6e39fa854
--- /dev/null
+++ b/scripts/remove-stale-files
@@ -0,0 +1,28 @@
+#!/bin/sh
+
+set -e
+
+# When you move, remove or rename generated files, you probably also update
+# .gitignore and cleaning rules in the Makefile. This is the right thing
+# to do. However, people usually do 'git pull', 'git bisect', etc. without
+# running 'make clean'. Then, the stale generated files are left over, often
+# causing build issues.
+#
+# Also, 'git status' shows such stale build artifacts as untracked files.
+# What is worse, some people send a wrong patch to get them back to .gitignore
+# without checking the commit history.
+#
+# So, when you (re)move generated files, please move the cleaning rules from
+# the Makefile to this script. This is run before Kbuild starts building
+# anything, so people will not be annoyed by such garbage files.
+#
+# This script is not intended to grow endlessly. Rather, it is a temporary scrap
+# yard. Stale files stay in this file for a while (for some release cycles?),
+# then will be really dead and removed from the code base entirely.
+
+# moved to security/selinux/genheaders
+rm -f scripts/selinux/genheaders/genheaders
+
+rm -f *.spec
+
+rm -f lib/test_fortify.log