[PATCH v2 0/1] MR3058: kerberos: Avoid potentially uninitialized variable (clang-scan)

[Fabian Maurer (＠DarkShadow44)]

When there are no caches, this variable would be uninitialized.
Reported by clang-scan.
Not sure if that can ever happen, but IMHO initializing doesn't do harm.

--
  v2: kerberos: Avoid potentially uninitialized variable (clang-scan)

https://gitlab.winehq.org/wine/wine/-/merge_requests/3058

[PATCH v2 1/1] kerberos: Avoid potentially uninitialized variable (clang-scan)

[Fabian Maurer]

From: Fabian Maurer

Signed-off-by: Fabian Maurer 
---
 dlls/kerberos/unixlib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/dlls/kerberos/unixlib.c b/dlls/kerberos/unixlib.c
index 45cf494aa70..8d9b59b58d4 100644
--- a/dlls/kerberos/unixlib.c
+++ b/dlls/kerberos/unixlib.c
@@ -305,7 +305,7 @@ static NTSTATUS copy_tickets_to_client( struct ticket_list *list, KERB_QUERY_TKT

static NTSTATUS kerberos_fill_ticket_list( struct ticket_list *list )
 {
-    NTSTATUS status;
+    NTSTATUS status = STATUS_SUCCESS;
     krb5_error_code err;
     krb5_context ctx;
     krb5_cccol_cursor cursor = NULL;

[Fabian Maurer (＠DarkShadow44)]

On Thu Jul  6 18:32:44 2023 +0000, Hans Leidekker wrote:
> The patch introduces a test failure.

Thanks, I just assumed this was unrelated... Turns out, the tests rely on the fact that the uninitialized variable is 0. I thought I was just fixing an unlikely potential issue, didn't expect this to be actually affecting tests.
Pushed an update.