> kmem_cache_destroy(hpb->map_req_cache); > @@ -1670,7 +2109,7 @@ void ufshpb_init_hpb_lu(struct ufs_hba *hba, struct > scsi_device *sdev) > if (ret) > goto out; > > - hpb = ufshpb_alloc_hpb_lu(hba, lun, &hba->ufshpb_dev, > + hpb = ufshpb_alloc_hpb_lu(hba, sdev, &hba->ufshpb_dev, > &hpb_lu_info); > if (!hpb) > goto out; In HPB2.0 device control mode, the host is expected to send HPB-WRITE-BUFFER 0x3 To informs that all HPB Regions are inactive (expect for pinned regions). Maybe a good place to do so is here, or in ufshpb_hpb_lu_prepared after you kicked the map work for pinned regions.
Either way, If you decide to do so, I would appreciate if you could align to the framework I proposed in (scsi: ufshpb: Region inactivation in host mode). This way you would have a wrapper unmap_all that would call ufshpb_issue_umap_req with buffer id 0x3, And I would have a wrapper unmap_single that would call it with buffer id 0x1. Thanks, Avri
