Re: [PATCH v14 6/8] lib/test: memcpy_kunit: add copy_page() and copy_mc_page() tests
From: Shuai Xue <xueshuai@linux.alibaba.com>
Date: 2026-05-27 13:44:18
Also in:
linux-arm-kernel, linux-mm, lkml
On 5/18/26 4:49 PM, Ruidong Tian wrote:
quoted hunk ↗ jump to hunk
Add KUnit tests for copy_page() and copy_mc_page(), modeled after the existing memcpy_test() style: a static page-aligned src and a two-page dst, filled with random bytes plus non-zero edges, then verify byte-for-byte equality and that the adjacent page is untouched. The copy_mc_page() case additionally checks the return value is 0 on clean memory and is gated on CONFIG_ARCH_HAS_COPY_MC. Signed-off-by: Ruidong Tian <redacted> --- lib/tests/memcpy_kunit.c | 67 +++++++++++++++++++++++++++++++++++++++- 1 file changed, 66 insertions(+), 1 deletion(-)diff --git a/lib/tests/memcpy_kunit.c b/lib/tests/memcpy_kunit.c index d36933554e46..85df53ccfb0c 100644 --- a/lib/tests/memcpy_kunit.c +++ b/lib/tests/memcpy_kunit.c@@ -493,6 +493,67 @@ static void memmove_overlap_test(struct kunit *test) } } +/* --- Page-sized copy tests --- */ + +static u8 page_src[PAGE_SIZE] __aligned(PAGE_SIZE); +static u8 page_dst[PAGE_SIZE * 2] __aligned(PAGE_SIZE); +static const u8 page_zero[PAGE_SIZE] __aligned(PAGE_SIZE); + +static void init_page(struct kunit *test) +{ + /* Get many bit patterns. */ + get_random_bytes(page_src, PAGE_SIZE); + + /* Make sure we have non-zero edges. */ + set_random_nonzero(test, &page_src[0]); + set_random_nonzero(test, &page_src[PAGE_SIZE - 1]); + + /* Explicitly zero the entire destination. */ + memset(page_dst, 0, ARRAY_SIZE(page_dst));
init_page() memsets the entire two-page page_dst to zero, and the
test then asserts the second page is still zero. A buggy copy_page()
that overwrites the second page with bytes that happen to include a
zero byte would still pass.
Please fill page_dst with a non-zero sentinel (e.g. 0xA5) instead
of zero, and check the second half against that sentinel using
memchr_inv():
memset(page_dst, 0xA5, ARRAY_SIZE(page_dst));
...
KUNIT_ASSERT_TRUE_MSG(test,
!memchr_inv(page_dst + PAGE_SIZE, 0xA5, PAGE_SIZE),
"copy_page overflow into adjacent page");
This also lets you delete page_zero[] entirely, which on
PAGE_SIZE=64K configs (arm64) saves a non-trivial amount of BSS.
+}
+
+static void copy_page_test(struct kunit *test)
+{
+ init_page(test);
+
+ /* Copy. */
+ copy_page(page_dst, page_src);
+
+ /* Verify byte-for-byte exact. */
+ KUNIT_ASSERT_EQ_MSG(test,
+ memcmp(page_dst, page_src, PAGE_SIZE), 0,
+ "copy_page content mismatch with random data");
+
+ /* Verify no overflow into second page. */
+ KUNIT_ASSERT_EQ_MSG(test,
+ memcmp(page_dst + PAGE_SIZE, page_zero, PAGE_SIZE), 0,
+ "copy_page overflow into adjacent page");
+}
+
+#ifdef CONFIG_ARCH_HAS_COPY_MC
+static void copy_mc_page_test(struct kunit *test)
+{
+ int ret;
+
+ init_page(test);
+
+ /* Copy and check return value. */
+ ret = copy_mc_page(page_dst, page_src);From sashiko: Does this introduce a build regression on architectures like x86_64 or powerpc? The CONFIG_ARCH_HAS_COPY_MC flag is selected by multiple architectures, but copy_mc_page() is an arm64-specific API. Other architectures use the generic copy_mc_to_kernel() API instead. If the kernel is built with KUnit enabled on non-arm64 architectures, it appears copy_mc_page() would be an undeclared identifier. Thanks. Shuai