Merge tag 'x86_asm_for_v6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 asm updates from Borislav Petkov:

 - Move the 32-bit memmove() asm implementation out-of-line in order to
   fix a 32-bit full LTO build failure with clang where it would fail at
   register allocation.

   Move it to an asm file and clean it up while at it, similar to what
   has been already done on 64-bit

* tag 'x86_asm_for_v6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/mem: Move memmove to out of line assembler

1 files changed, 22 insertions, 0 deletions

diff --git a/lib/memcpy_kunit.c b/lib/memcpy_kunit.c
index 2b5cc70ac53f..7513e6d5dc90 100644
--- a/lib/memcpy_kunit.c
+++ b/lib/memcpy_kunit.c
@@ -105,6 +105,8 @@ static void memcpy_test(struct kunit *test)
 #undef TEST_OP
 }
 
+static unsigned char larger_array [2048];
+
 static void memmove_test(struct kunit *test)
 {
 #define TEST_OP "memmove"
@@ -179,6 +181,26 @@ static void memmove_test(struct kunit *test)
 	ptr = &overlap.data[2];
 	memmove(ptr, overlap.data, 5);
 	compare("overlapping write", overlap, overlap_expected);
+
+	/* Verify larger overlapping moves. */
+	larger_array[256] = 0xAAu;
+	/*
+	 * Test a backwards overlapping memmove first. 256 and 1024 are
+	 * important for i386 to use rep movsl.
+	 */
+	memmove(larger_array, larger_array + 256, 1024);
+	KUNIT_ASSERT_EQ(test, larger_array[0], 0xAAu);
+	KUNIT_ASSERT_EQ(test, larger_array[256], 0x00);
+	KUNIT_ASSERT_NULL(test,
+		memchr(larger_array + 1, 0xaa, ARRAY_SIZE(larger_array) - 1));
+	/* Test a forwards overlapping memmove. */
+	larger_array[0] = 0xBBu;
+	memmove(larger_array + 256, larger_array, 1024);
+	KUNIT_ASSERT_EQ(test, larger_array[0], 0xBBu);
+	KUNIT_ASSERT_EQ(test, larger_array[256], 0xBBu);
+	KUNIT_ASSERT_NULL(test, memchr(larger_array + 1, 0xBBu, 256 - 1));
+	KUNIT_ASSERT_NULL(test,
+		memchr(larger_array + 257, 0xBBu, ARRAY_SIZE(larger_array) - 257));
 #undef TEST_OP
 }