s390/test_unwind: use raw opcode instead of invalid instruction

Building with clang & LLVM_IAS=1 leads to an error:
    arch/s390/lib/test_unwind.c:179:4: error: invalid register pair
                        "       mvcl    %%r1,%%r1\n"
                        ^

The test creates an invalid instruction that would trap at runtime, but the
LLVM inline assembler tries to validate it at compile time too.

Use the raw instruction opcode instead.

Reported-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Ilie Halip <ilie.halip@gmail.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Suggested-by: Ulrich Weigand <Ulrich.Weigand@de.ibm.com>
Link: https://github.com/ClangBuiltLinux/linux/issues/1421
Link: https://lore.kernel.org/r/20211117174822.3632412-1-ilie.halip@gmail.com
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
[hca@linux.ibm.com: use illegal opcode, and update comment]
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>

diff --git a/arch/s390/lib/test_unwind.c b/arch/s390/lib/test_unwind.c
index cfc5f55..bc79733 100644 (file)
--- a/arch/s390/lib/test_unwind.c
+++ b/arch/s390/lib/test_unwind.c
@@ -173,10 +173,11 @@ static noinline int unwindme_func4(struct unwindme *u)
                }
 
                /*
-                * trigger specification exception
+                * Trigger operation exception; use insn notation to bypass
+                * llvm's integrated assembler sanity checks.
                 */
                asm volatile(
-                       "       mvcl    %%r1,%%r1\n"
+                       "       .insn   e,0x0000\n"     /* illegal opcode */
                        "0:     nopr    %%r7\n"
                        EX_TABLE(0b, 0b)
                        :);