Hello iallouch,
The patch 706f1b5d83ca: "wifi: iwlwifi: mvm: add start mac ctdp sum
calculation debugfs handler" from Oct 4, 2023 (linux-next), leads to
the following Smatch static checker warning:
drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c:64 iwl_dbgfs_stop_ctdp_write()
error: uninitialized symbol 'force'.
drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
49 static ssize_t iwl_dbgfs_stop_ctdp_write(struct iwl_mvm *mvm, char *buf,
50 size_t count, loff_t *ppos)
51 {
52 int ret;
53 bool force;
54
55 if (!kstrtobool(buf, &force))
56 IWL_DEBUG_INFO(mvm,
57 "force start is %d [0=disabled, 1=enabled]\n",
58 force);
If kstrtobool() fails then force is uninitialized. This should be:
ret = kstrtobool(buf, &force);
if (ret)
return ret;
I feel like the IWL_DEBUG_INFO() does not add anything and should be
deleted. Right now it's printing on the success path...
59
60 /* we allow skipping cap support check and force stop ctdp
61 * statistics collection and with guerantee that it is
62 * safe to use.
63 */
--> 64 if (!force && !iwl_mvm_is_ctdp_supported(mvm))
65 return -EOPNOTSUPP;
66
67 if (!iwl_mvm_firmware_running(mvm) ||
68 mvm->fwrt.cur_fw_img != IWL_UCODE_REGULAR)
69 return -EIO;
70
71 mutex_lock(&mvm->mutex);
72 ret = iwl_mvm_ctdp_command(mvm, CTDP_CMD_OPERATION_STOP, 0);
73 mutex_unlock(&mvm->mutex);
74
75 return ret ?: count;
76 }
regards,
dan carpenter
