adiv5.c: Fix another leak.
This commit is contained in:
parent
783ec377d9
commit
b0cf7d24bd
|
@ -317,9 +317,11 @@ static bool adiv5_component_probe(ADIv5_AP_t *ap, uint32_t addr, int recursion,
|
|||
#endif
|
||||
|
||||
for (int i = 0; i < 960; i++) {
|
||||
adiv5_dp_error(ap->dp);
|
||||
uint32_t entry = adiv5_mem_read32(ap, addr + i*4);
|
||||
if (adiv5_dp_error(ap->dp)) {
|
||||
DEBUG("%sFault reading ROM table entry\n", indent);
|
||||
DEBUG("%sFault reading ROM table entry %d\n", indent, i);
|
||||
break;
|
||||
}
|
||||
|
||||
if (entry == 0)
|
||||
|
@ -518,6 +520,7 @@ void adiv5_dp_init(ADIv5_DP_t *dp)
|
|||
if (ap->base == last_base) {
|
||||
DEBUG("AP %d: Duplicate base\n", i);
|
||||
adiv5_ap_cleanup(i);
|
||||
free(ap);
|
||||
/* FIXME: Should we expect valid APs behind duplicate ones? */
|
||||
return;
|
||||
}
|
||||
|
|
Loading…
Reference in New Issue