From: Nathan Chancellor <nathan@kernel.org>
To: Kees Cook <kees@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
linux-kernel@vger.kernel.org, llvm@lists.linux.dev,
stable@vger.kernel.org, Nathan Chancellor <nathan@kernel.org>
Subject: [PATCH] lkdtm/usercopy: Avoid unsequenced accesses to cache_size in lkdtm_usercopy_init()
Date: Sat, 14 Nov 2026 21:37:31 -0700 [thread overview]
Message-ID: <20261114-lkdtm-usercopy-wunsequenced-v1-1-d9f288701932@kernel.org> (raw)
After a recent improvement in clang to warn on unsequenced accesses to
volatile variables [1], an instance appears in lkdtm_usercopy_init(),
which can break the build with CONFIG_WERROR=y / W=e:
drivers/misc/lkdtm/usercopy.c:431:9: error: unsequenced accesses to volatile qualified 'cache_size' [-Werror,-Wunsequenced]
429 | kmem_cache_create_usercopy("lkdtm-usercopy", cache_size,
| ~~~~~~~~~~
430 | 0, 0,
431 | cache_size / 4,
| ^
Per C11 5.1.2.3.2, accessing a volatile object is a side effect. Per C11
6.5.2, unsequenced side effects on the same scalar object are undefined
behavior. Per C11 6.5.2.2.10/12, the evaluation of arguments in a
function call is indeterminately sequenced, meaning any side effects in
the arguments are unsequenced.
cache_size is marked as volatile to hide its value from the compile time
constant checks in the hardened usercopy routines. Refactor the
calculations based on cache_size into their own volatile variables to
keep the values hidden while avoiding unsequenced accesses on the same
object.
Cc: stable@vger.kernel.org
Fixes: e47e311843de ("lkdtm: Update usercopy tests for whitelisting")
Closes: https://github.com/ClangBuiltLinux/linux/issues/2195
Link: https://github.com/llvm/llvm-project/commit/955715460eaf9188c241c74750637fc82dfcc423 [1]
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
---
drivers/misc/lkdtm/usercopy.c | 28 ++++++++++++++++------------
1 file changed, 16 insertions(+), 12 deletions(-)
diff --git a/drivers/misc/lkdtm/usercopy.c b/drivers/misc/lkdtm/usercopy.c
index 67db57249a34..6c561dc1bcaa 100644
--- a/drivers/misc/lkdtm/usercopy.c
+++ b/drivers/misc/lkdtm/usercopy.c
@@ -19,10 +19,15 @@
* hardened usercopy checks by added "unconst" to all the const copies,
* and making sure "cache_size" isn't optimized into a const.
*/
+#define CACHE_SIZE 1024
static volatile size_t unconst;
-static volatile size_t cache_size = 1024;
+static volatile size_t cache_size = CACHE_SIZE;
static struct kmem_cache *whitelist_cache;
+/* Whitelisted window in buffer, from kmem_cache_create_usercopy. */
+static volatile size_t cache_whitelist_offset = CACHE_SIZE / 4;
+static volatile size_t cache_whitelist_size = CACHE_SIZE / 16;
+
static const unsigned char test_text[] = "This is a test.\n";
/*
@@ -204,7 +209,6 @@ static void do_usercopy_slab_whitelist(bool to_user)
unsigned long user_alloc;
unsigned char *buf = NULL;
unsigned char __user *user_addr;
- size_t offset, size;
/* Make sure cache was prepared. */
if (!whitelist_cache) {
@@ -233,31 +237,31 @@ static void do_usercopy_slab_whitelist(bool to_user)
memset(buf, 'B', cache_size);
- /* Whitelisted window in buffer, from kmem_cache_create_usercopy. */
- offset = (cache_size / 4) + unconst;
- size = (cache_size / 16) + unconst;
-
if (to_user) {
pr_info("attempting good copy_to_user inside whitelist\n");
- if (copy_to_user(user_addr, buf + offset, size)) {
+ if (copy_to_user(user_addr, buf + cache_whitelist_offset,
+ cache_whitelist_size)) {
pr_warn("copy_to_user failed unexpectedly?!\n");
goto free_user;
}
pr_info("attempting bad copy_to_user outside whitelist\n");
- if (copy_to_user(user_addr, buf + offset - 1, size)) {
+ if (copy_to_user(user_addr, buf + cache_whitelist_offset - 1,
+ cache_whitelist_size)) {
pr_warn("copy_to_user failed, but lacked Oops\n");
goto free_user;
}
} else {
pr_info("attempting good copy_from_user inside whitelist\n");
- if (copy_from_user(buf + offset, user_addr, size)) {
+ if (copy_from_user(buf + cache_whitelist_offset, user_addr,
+ cache_whitelist_size)) {
pr_warn("copy_from_user failed unexpectedly?!\n");
goto free_user;
}
pr_info("attempting bad copy_from_user outside whitelist\n");
- if (copy_from_user(buf + offset - 1, user_addr, size)) {
+ if (copy_from_user(buf + cache_whitelist_offset - 1, user_addr,
+ cache_whitelist_size)) {
pr_warn("copy_from_user failed, but lacked Oops\n");
goto free_user;
}
@@ -428,8 +432,8 @@ void __init lkdtm_usercopy_init(void)
whitelist_cache =
kmem_cache_create_usercopy("lkdtm-usercopy", cache_size,
0, 0,
- cache_size / 4,
- cache_size / 16,
+ cache_whitelist_offset,
+ cache_whitelist_size,
NULL);
}
---
base-commit: 7475ea8b58616b33c04afb976cf0fbaad9dd63fb
base-tree: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
change-id: 20261114-lkdtm-usercopy-wunsequenced-eeec86ef6500
Best regards,
--
Cheers,
Nathan
reply other threads:[~2026-11-15 4:37 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20261114-lkdtm-usercopy-wunsequenced-v1-1-d9f288701932@kernel.org \
--to=nathan@kernel.org \
--cc=arnd@arndb.de \
--cc=gregkh@linuxfoundation.org \
--cc=kees@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=llvm@lists.linux.dev \
--cc=stable@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®