The patch titled
Subject: lib/test_meminit.c: fix -Wmaybe-uninitialized false positive
has been added to the -mm tree. Its filename is
lib-test_meminit-fix-wmaybe-uninitialized-false-positive.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/lib-test_meminit-fix-wmaybe-uninitialized-false-positive.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/lib-test_meminit-fix-wmaybe-uninitialized-false-positive.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Arnd Bergmann <arnd@xxxxxxxx>
Subject: lib/test_meminit.c: fix -Wmaybe-uninitialized false positive
The conditional logic is too complicated for the compiler to fully
comprehend:
lib/test_meminit.c: In function 'test_meminit_init':
lib/test_meminit.c:236:5: error: 'buf_copy' may be used uninitialized in this function [-Werror=maybe-uninitialized]
kfree(buf_copy);
^~~~~~~~~~~~~~~
lib/test_meminit.c:201:14: note: 'buf_copy' was declared here
Simplify it by splitting out the non-rcu section.
Link: http://lkml.kernel.org/r/20190617131210.2190280-1-arnd@xxxxxxxx
Fixes: af734ee6ec85 ("lib: introduce test_meminit module")
Signed-off-by: Arnd Bergmann <arnd@xxxxxxxx>
Acked-by: Alexander Potapenko <glider@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
lib/test_meminit.c | 50 ++++++++++++++++++++++---------------------
1 file changed, 26 insertions(+), 24 deletions(-)
--- a/lib/test_meminit.c~lib-test_meminit-fix-wmaybe-uninitialized-false-positive
+++ a/lib/test_meminit.c
@@ -208,35 +208,37 @@ static int __init do_kmem_cache_size(siz
/* Check that buf is zeroed, if it must be. */
fail = check_buf(buf, size, want_ctor, want_rcu, want_zero);
fill_with_garbage_skip(buf, size, want_ctor ? CTOR_BYTES : 0);
+
+ if (!want_rcu) {
+ kmem_cache_free(c, buf);
+ continue;
+ }
+
/*
* If this is an RCU cache, use a critical section to ensure we
* can touch objects after they're freed.
*/
- if (want_rcu) {
- rcu_read_lock();
- /*
- * Copy the buffer to check that it's not wiped on
- * free().
- */
- buf_copy = kmalloc(size, GFP_KERNEL);
- if (buf_copy)
- memcpy(buf_copy, buf, size);
- }
- kmem_cache_free(c, buf);
- if (want_rcu) {
- /*
- * Check that |buf| is intact after kmem_cache_free().
- * |want_zero| is false, because we wrote garbage to
- * the buffer already.
- */
- fail |= check_buf(buf, size, want_ctor, want_rcu,
- false);
- if (buf_copy) {
- fail |= (bool)memcmp(buf, buf_copy, size);
- kfree(buf_copy);
- }
- rcu_read_unlock();
+ rcu_read_lock();
+ /*
+ * Copy the buffer to check that it's not wiped on
+ * free().
+ */
+ buf_copy = kmalloc(size, GFP_KERNEL);
+ if (buf_copy)
+ memcpy(buf_copy, buf, size);
+
+ /*
+ * Check that |buf| is intact after kmem_cache_free().
+ * |want_zero| is false, because we wrote garbage to
+ * the buffer already.
+ */
+ fail |= check_buf(buf, size, want_ctor, want_rcu,
+ false);
+ if (buf_copy) {
+ fail |= (bool)memcmp(buf, buf_copy, size);
+ kfree(buf_copy);
}
+ rcu_read_unlock();
}
kmem_cache_destroy(c);
_
Patches currently in -mm which might be from arnd@xxxxxxxx are
lib-test_meminit-fix-wmaybe-uninitialized-false-positive.patch
coda-stop-using-struct-timespec-in-user-api.patch
