[linux-nfc] [neard][PATCH v2 18/73] se: silence clang -Wcast-align warning

Fix clang warnings:

    se/ace.c:147:15: error: cast from 'uint8_t *' (aka 'unsigned char *') to 'struct seel_ace_apdu_rule *' increases required alignment from 1 to 4 [-Werror,-Wcast-align]
                    apdu_rule = (struct seel_ace_apdu_rule *)rule->apdu_rules;
                                ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    se/ace.c:790:18: error: cast from 'uint8_t *' (aka 'unsigned char *') to 'uint32_t *' (aka 'unsigned int *') increases required alignment from 1 to 4 [-Werror,-Wcast-align]
            apdu_header = *((uint32_t *) apdu);
                            ^~~~~~~~~~~~~~~~~
    se/ace.c:791:14: error: cast from 'uint8_t *' (aka 'unsigned char *') to 'struct seel_ace_apdu_rule *' increases required alignment from 1 to 4 [-Werror,-Wcast-align]
            apdu_rule = (struct seel_ace_apdu_rule *)rule->apdu_rules;
                        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Increasing alignment should be fine for most of the platforms (maybe
except some performance penalty), although there are such which might
not handle it and raise CPU exception.  I am not sure whether the code
is actually safe, but proper fixing would require bigger changes.

Signed-off-by: Krzysztof Kozlowski <[email protected]>
---
 se/ace.c | 30 +++++++++++++++++++++++++++---
 1 file changed, 27 insertions(+), 3 deletions(-)

diff --git a/se/ace.c b/se/ace.c
index c1561abae8c1..ef0e7e15fbf7 100644
--- a/se/ace.c
+++ b/se/ace.c
@@ -144,7 +144,18 @@ static void dump_rule(gpointer data, gpointer user_data)
 		uint8_t *header, *mask;
 		size_t n_rules;
 
-		apdu_rule = (struct seel_ace_apdu_rule *)rule->apdu_rules;
+		/*
+		 * (void *) to silence -Wcast-align. Code should be safe
+		 * (assuming platform handles unaliagned access) as iterations
+		 * go up to n_rules.
+		 *
+		 * TODO: Fix the problem instead of silencing with cast, so
+		 * the code would be porable.
+		 */
+		if (rule->apdu_rules_len % sizeof(struct seel_ace_apdu_rule *)) {
+			DBG("  APDU: wrong alignment (Bug, code needs fixing)");
+		}
+		apdu_rule = (struct seel_ace_apdu_rule *)(void *)rule->apdu_rules;
 		n_rules = rule->apdu_rules_len /
 				sizeof(struct seel_ace_apdu_rule);
 
@@ -787,8 +798,21 @@ static bool apdu_allowed(struct seel_ace_rule *rule,
 
 	n_rules = rule->apdu_rules_len /
 				sizeof(struct seel_ace_apdu_rule);
-	apdu_header = *((uint32_t *) apdu);
-	apdu_rule = (struct seel_ace_apdu_rule *)rule->apdu_rules;
+	/*
+	 * FIXME: apdu comes from message and where is checking for apdu_len?
+	 * The (void *) is to fix -Wcast-align but the actual problem is
+	 * whether the apdu contains enough of data.
+	 */
+	apdu_header = *((uint32_t *)(void *) apdu);
+	/*
+	 * (void *) to silence -Wcast-align. Code should be safe
+	 * (assuming platform handles unaliagned access) as iterations
+	 * go up to n_rules.
+	 *
+	 * TODO: Fix the problem instead of silencing with cast, so
+	 * the code would be porable.
+	 */
+	apdu_rule = (struct seel_ace_apdu_rule *)(void *)rule->apdu_rules;
 
 	for (i = 0; i < n_rules; i++) {
 		if ((apdu_header & apdu_rule->mask) == apdu_rule->header)
-- 
2.27.0
_______________________________________________
Linux-nfc mailing list -- [email protected]
To unsubscribe send an email to [email protected]
%(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s