On 2024/4/19 23:09, Jeff Johnson wrote:
> On 4/17/2024 1:17 AM, Su Hui wrote:
>> u8 peer_debug_trigger;
>> - int ret;
>> + int ret = 0;
> this is unnecessary since this will be written in all paths that lead to the
> return that reads it
Yes, this is my fault. I will remove this in v2 patch.
>>
>> if (kstrtou8_from_user(user_buf, count, 0, &peer_debug_trigger))
>> return -EINVAL;
>> @@ -432,14 +432,12 @@ ath10k_dbg_sta_write_peer_debug_trigger(struct file *file,
>>
>> ret = ath10k_wmi_peer_set_param(ar, arsta->arvif->vdev_id, sta->addr,
>> ar->wmi.peer_param->debug, peer_debug_trigger);
>> - if (ret) {
>> + if (ret)
>> ath10k_warn(ar, "failed to set param to trigger peer tid logs for station ret: %d\n",
>> ret);
>> - goto out;
>> - }
>> out:
>> mutex_unlock(&ar->conf_mutex);
>> - return count;
>> + return ret ?: count;
>> }
>>
>> static const struct file_operations fops_peer_debug_trigger = {
> I'd suggest as an alternate solution that this function is a good candidate
> for the the cleanup.h functionality. By scoping the mutex_lock() you can
> simply return at each error location, and remove the explicit mutex_unlock().
like guard(mutex)(&ar->conf_mutex)?
Maybe I can send a separate patch for this, because there are more than
one candidates
for the cleanup.h functionality.
> But I'd accept this with the initializer change removed as well since I don't
> think ath10k has any cleanup.h usages yet.
I will send v2 patch soon , thanks for your reply!
Su Hui