fault-injection: make stacktrace filter works as expected

stacktrace filter is checked after others, such as fail-nth, interval and
probability.  This make it doesn't work well as expected.

Fix to running stacktrace filter before other filters.  It will speed up
fault inject testing for driver modules.

Link: https://lkml.kernel.org/r/20220817080332.1052710-5-weiyongjun1@huawei.comSigned-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Cc: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Isabella Basso <isabbasso@riseup.net>
Cc: Josh Poimboeuf <jpoimboe@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Miguel Ojeda <ojeda@kernel.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

lib/fault-inject.c

@@ -102,10 +102,16 @@ static inline bool fail_stacktrace(struct fault_attr *attr)
 
 bool should_fail_ex(struct fault_attr *attr, ssize_t size, int flags)
 {
+	bool stack_checked = false;
+
 	if (in_task()) {
 		unsigned int fail_nth = READ_ONCE(current->fail_nth);
 
 		if (fail_nth) {
+			if (!fail_stacktrace(attr))
+				return false;
+
+			stack_checked = true;
 			fail_nth--;
 			WRITE_ONCE(current->fail_nth, fail_nth);
 			if (!fail_nth)
@@ -125,6 +131,9 @@ bool should_fail_ex(struct fault_attr *attr, ssize_t size, int flags)
 	if (atomic_read(&attr->times) == 0)
 		return false;
 
+	if (!stack_checked && !fail_stacktrace(attr))
+		return false;
+
 	if (atomic_read(&attr->space) > size) {
 		atomic_sub(size, &attr->space);
 		return false;
@@ -139,9 +148,6 @@ bool should_fail_ex(struct fault_attr *attr, ssize_t size, int flags)
 	if (attr->probability <= get_random_u32_below(100))
 		return false;
 
-	if (!fail_stacktrace(attr))
-		return false;
-
 fail:
 	if (!(flags & FAULT_NOWARN))
 		fail_dump(attr);