After add sanity check to pass klockwork check, The spdif_mutex should be unlock before return true in check_non_pcm_per_cvt().
How do you think about a wording variant like the following?
Subject: [PATCH v2] ALSA: hda/hdmi: Unlock a mutex always before returning from check_non_pcm_per_cvt()
Change description (according to a solution alternative): The exception handling was incomplete in an if branch for a null pointer check of the variable “spdif”. Thus assign an appropriate value to the local variable “non_pcm” by using a conditional operator instead. A mutex is also appropriately unlocked then.
…
+++ b/sound/pci/hda/patch_hdmi.c @@ -1848,8 +1848,10 @@ static bool check_non_pcm_per_cvt(struct hda_codec *codec, hda_nid_t cvt_nid)
…
- if (WARN_ON(spdif == NULL))
- if (WARN_ON(spdif == NULL)) {
return true;mutex_unlock(&codec->spdif_mutex);
- } non_pcm = !!(spdif->status & IEC958_AES0_NONAUDIO);
- if (WARN_ON(spdif == NULL)) - return true; - non_pcm = !!(spdif->status & IEC958_AES0_NONAUDIO); + non_pcm = WARN_ON(!spdif) + ? true + : !!(spdif->status & IEC958_AES0_NONAUDIO); mutex_unlock(&codec->spdif_mutex); return non_pcm; }
Would you like to add the tag “Fixes” to the change description?
Regards, Markus