Skip to contentSkip to conversation

NVMe: Fix removal in case of active namespace list scanning method

PATCH Superseded Sunad Bhandary S 2 participants 5 messages May 27, 2016 – Jun 8, 2016 NVMe

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 NVMe: Fix removal in case of active namespace list scanning method

From: Sunad Bhandary <[email protected]>

In case of the active namespace list scanning method, a namespace that is detached is not removed from the host if it was the last entry in the list. Fix this by adding a scan to validate namespaces greater than the value of prev.

This also handles the case of removing namespaces whose value exceed the device's reported number of namespaces.

Signed-off-by Sunad Bhandary S <sunad.s at samsung.com>
Reviewed-by Christoph Hellwig <hch at lst.de>

---
v1 -> v2:
Removed unnecessary braces around if condition

 drivers/nvme/host/core.c | 21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)
drivers/nvme/host/core.c +15 -6
index 2de248b..3da998f 100644
@@ -1483,6 +1483,17 @@ static void nvme_validate_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 		nvme_alloc_ns(ctrl, nsid);
 }
 
+static void nvme_remove_invalid_namespaces(struct nvme_ctrl *ctrl,
+					unsigned nsid)
+{
+	struct nvme_ns *ns, *next;
+
+	list_for_each_entry_safe(ns, next, &ctrl->namespaces, list) {
+		if (ns->ns_id > nsid)
+			nvme_ns_remove(ns);
+	}
+}
+
 static int nvme_scan_ns_list(struct nvme_ctrl *ctrl, unsigned nn)
 {
 	struct nvme_ns *ns;
@@ -1497,7 +1508,7 @@ static int nvme_scan_ns_list(struct nvme_ctrl *ctrl, unsigned nn)
 	for (i = 0; i < num_lists; i++) {
 		ret = nvme_identify_ns_list(ctrl, prev, ns_list);
 		if (ret)
-			goto out;
+			goto free;
 
 		for (j = 0; j < min(nn, 1024U); j++) {
 			nsid = le32_to_cpu(ns_list[j]);
@@ -1515,13 +1526,14 @@ static int nvme_scan_ns_list(struct nvme_ctrl *ctrl, unsigned nn)
 		nn -= j;
 	}
  out:
+	nvme_remove_invalid_namespaces(ctrl, prev);
+ free:
 	kfree(ns_list);
 	return ret;
 }
 
 static void nvme_scan_ns_sequential(struct nvme_ctrl *ctrl, unsigned nn)
 {
-	struct nvme_ns *ns, *next;
 	unsigned i;
 
 	lockdep_assert_held(&ctrl->namespaces_mutex);
@@ -1529,10 +1541,7 @@ static void nvme_scan_ns_sequential(struct nvme_ctrl *ctrl, unsigned nn)
 	for (i = 1; i <= nn; i++)
 		nvme_validate_ns(ctrl, i);
 
-	list_for_each_entry_safe(ns, next, &ctrl->namespaces, list) {
-		if (ns->ns_id > nn)
-			nvme_ns_remove(ns);
-	}
+	nvme_remove_invalid_namespaces(ctrl, nn);
 }
 
 static void nvme_scan_work(struct work_struct *work)
Signature · 1 line
1.9.3
#1 lore raw

PATCH v2 1/1 NVMe: Fix removal in case of active namespace list scanning method

Quoted 13 lines · Sunad Bhandary S

static int nvme_scan_ns_list(struct nvme_ctrl *ctrl, unsigned nn) {

drivers/nvme/host/core.c +1 -1
 	struct nvme_ns *ns;
@@ -1497,7 +1508,7 @@ static int nvme_scan_ns_list(struct nvme_ctrl *ctrl, unsigned nn)
 	for (i = 0; i < num_lists; i++) {
 		ret = nvme_identify_ns_list(ctrl, prev, ns_list);
 		if (ret)
-			goto out;
+			goto free;
 
 		for (j = 0; j < min(nn, 1024U); j++) {
 			nsid = le32_to_cpu(ns_list[j]);

drivers/nvme/host/core.c static int nvme_scan_ns_list(struct nvme_ctrl *ctrl, unsigned nn) · line 1530

Quoted 5 lines
drivers/nvme/host/core.c +2 -0
@@ -1515,13 +1526,14 @@ static int nvme_scan_ns_list(struct nvme_ctrl *ctrl, unsigned nn)
 		nn -= j;
 	}
  out:
+	nvme_remove_invalid_namespaces(ctrl, prev);
+ free:

We don't want to remove the namespaces on error here. That could remove valid, in use namespaces: if nvme_identify_ns_list fails (maybe controller just doesn't support that identify mode), we do the sequential scanning.

Just leave the 'goto free' alone, and call nvme_remove_invalid_namespaces only on success.

#2 lore raw

PATCH v2 1/1 NVMe: Fix removal in case of active namespace list scanning method

Quoted 7 lines

We don't want to remove the namespaces on error here. That could remove valid, in use namespaces: if nvme_identify_ns_list fails (maybe controller just doesn't support that identify mode), we do the sequential scanning.

Just leave the 'goto free' alone, and call
nvme_remove_invalid_namespaces only on success.

In the patch, the namespaces are not removed on failure of identify list. If nvme_identify_ns_list fails, the value is returned and sequential scanning is done. The call to nvme_remove_invalid_namespaces is done when the value of nsid in the list is 0.

Snapshot of patch:
....

	ret = nvme_identify_ns_list(ctrl, prev, ns_list);
                 if (ret)
                         goto free;
  ....
  	nsid = le32_to_cpu(ns_list[j]);
                         if (!nsid)
                                 goto out;

....

  out:
         nvme_remove_invalid_namespaces(ctrl, prev);
  free:
         kfree(ns_list);
         return ret;

This should be able to handle error scenarios or am I missing something?

Thanks and Regards,
Sunad

#3 lore raw

PATCH v2 1/1 NVMe: Fix removal in case of active namespace list scanning method

Quoted 4 lines · Sunad Bhandary

In the patch, the namespaces are not removed on failure of identify list. If nvme_identify_ns_list fails, the value is returned and sequential scanning is done. The call to nvme_remove_invalid_namespaces is done when the value of nsid in the list is 0.

Sorry, my mistake. You're right, it is good as-is.

#4 lore raw

PATCH v2 1/1 NVMe: Fix removal in case of active namespace list scanning method

Hi Jens,

Can this be picked up?

Thanks and Regards,
Sunad

Quoted 14 lines

-----Original Message-----
From: Linux-nvme [mailto:linux-nvme-bounces at lists.infradead.org] On Behalf Of Keith Busch
Sent: Tuesday, May 31, 2016 8:05 PM
To: Sunad Bhandary
Cc: hch at infradead.org; linux-nvme at lists.infradead.org
Subject: Re: Re: [PATCHv2] NVMe: Fix removal in case of active namespace list scanning method

Quoted 4 lines · Sunad Bhandary

In the patch, the namespaces are not removed on failure of identify list. If nvme_identify_ns_list fails, the value is returned and sequential scanning is done. The call to nvme_remove_invalid_namespaces is done when the value of nsid in the list is 0.

Sorry, my mistake. You're right, it is good as-is.