Re: [PATCH 3/5] nvme: refactor namespace probing

[Kanchan Joshi <joshi.k@samsung.com>]

[-- Attachment #1: Type: text/plain, Size: 1346 bytes --]

On Wed, Jul 13, 2022 at 07:49:12AM +0200, Christoph Hellwig wrote:

> static void nvme_scan_ns(struct nvme_ctrl *ctrl, unsigned nsid)
> {
>-	struct nvme_ns_ids ids = { };
>-	struct nvme_id_ns_cs_indep *id;
>+	struct nvme_ns_info info = { .nsid = nsid };
> 	struct nvme_ns *ns;
>-	bool ready = true;
>
>-	if (nvme_identify_ns_descs(ctrl, nsid, &ids))
>+	if (nvme_identify_ns_descs(ctrl, &info))
> 		return;
>
>-	if (ids.csi != NVME_CSI_NVM && !nvme_multi_css(ctrl)) {
>+	if (info.ids.csi != NVME_CSI_NVM && !nvme_multi_css(ctrl)) {
> 		dev_warn(ctrl->device,
> 			"command set not reported for nsid: %d\n", nsid);
> 		return;
> 	}
>
> 	/*
>-	 * Check if the namespace is ready.  If not ignore it, we will get an
>-	 * AEN once it becomes ready and restart the scan.
>+	 * If available try to use the Command Set Idependent Identify Namespace
>+	 * data structure to find all the generic information that is needed to
>+	 * set up a namespace.  If not fall back to the legacy version.
> 	 */
>-	if ((ctrl->cap & NVME_CAP_CRMS_CRIMS) &&
>-	    !nvme_identify_ns_cs_indep(ctrl, nsid, &id)) {
>-		ready = id->nstat & NVME_NSTAT_NRDY;
>-		kfree(id);
>+	if (ctrl->cap & NVME_CAP_CRMS_CRIMS) {
>+		if (nvme_ns_info_from_id_cs_indep(ctrl, &info))
>+			return;

Is this return fine if independent id-ns fails? Earlier behavior was to
ingore such failure.


[-- Attachment #2: Type: text/plain, Size: 0 bytes --]