topology: merge eq_fir and eq_iir IPC constructors into generic "effect" constructor.  #618

Activity

1. lgirdwood

   added

   enhancementNew feature or requestNew feature or request

   on Feb 1, 2019
2. plbossart commented on Feb 1, 2019

   

   plbossart

   on Feb 1, 2019 via email

   Member

   More actions

   On 2/1/19 4:02 PM, Liam Girdwood wrote: These can be merged and reused for voice detection. All the differentiation is done in the private data. @plbossart <https://github.com/plbossart> fyi, this is what I'm thinking about changing this weekend. Seems we have unnecessary duplication for no benefit (which will lead to more duplication for other new "effects"). |/* FIR equalizer component */ struct sof_ipc_comp_eq_fir { struct sof_ipc_comp comp; struct sof_ipc_comp_config config; uint32_t size; /* reserved for future use */ uint32_t reserved[8]; unsigned char data[0]; } __packed; /* IIR equalizer component */ struct sof_ipc_comp_eq_iir { struct sof_ipc_comp comp; struct sof_ipc_comp_config config; uint32_t size; /* reserved for future use */ uint32_t reserved[8]; unsigned char data[0]; } __packed; |

   "effect" usually means PCM in and out. If you deal with non-audio/compressed data you are really talking about a processing element using non-descript binary data - or a payload. I would remove any type here, just pass the data and let the component parse it.
3. wenqingfu

   assigned

   wenqingfu

   and

   singalsu

   and unassigned

   wenqingfu

   on Feb 13, 2019
4. singalsu commented on Apr 3, 2019

   

   singalsu

   on Apr 3, 2019

   Last edited by singalsu

   Collaborator

   More actions

   The sof_ipc_comp_process has replaced in firmware the processing algorithm specific IPC structs.

   /* generic "effect", "codec" or proprietary processing component */
   struct sof_ipc_comp_process {
   	struct sof_ipc_comp comp;
   	struct sof_ipc_comp_config config;
   	uint32_t size;	/**< size of bespoke data section in bytes */
   	uint32_t type;	/**< sof_ipc_effect_type */
   
   	/* reserved for future use */
   	uint32_t reserved[7];
   
   	unsigned char data[0];
   } __attribute__((packed));
   
   

   The same change still need to be done on Linux driver side as well. It is in Linux PR #771 by @keyonjie .
5. singalsu commented on Apr 3, 2019

   

   singalsu

   on Apr 3, 2019

   Collaborator

   More actions

   This can be closed since #771 is merged now.
6. singalsu

   closed this as completedon Apr 3, 2019
7. oder-chiou

   added a commit that references this issue on May 5, 2026

   selftests/bpf: Fix selftest verif_scale_strobemeta failure with llvm22

   ...

   4f8543b
8. naveen-manohar

   added a commit that references this issue on Aug 13, 2026

   selftests/bpf: Fix selftest verif_scale_strobemeta failure with llvm22

   ...

   3667af0