Skip to contentSkip to conversation

libnvdimm/bus: return the outvar 'cmd_rc' error code in __nd_ioctl()

PATCH In review Vaibhav Jain 4 participants 5 messages Jan 22, 2020 – Mar 2, 2020 libnvdimm: non-volatile memory device

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 libnvdimm/bus: return the outvar 'cmd_rc' error code in __nd_ioctl()

Presently the error code returned via out variable 'cmd_rc' from the nvdimm-bus controller function is ignored when called from __nd_ioctl() and never communicated back to user-space code that called an ioctl on dimm/bus.

This minor patch updates __nd_ioctl() to propagate the value of out variable 'cmd_rc' back to user-space in case it reports an error.

---
 drivers/nvdimm/bus.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/nvdimm/bus.c +5 -0
index a8b515968569..5b687a27fdf2 100644
@@ -1153,6 +1153,11 @@ static int __nd_ioctl(struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm,
 	if (rc < 0)
 		goto out_unlock;
 
+	if (cmd_rc < 0) {
+		rc = cmd_rc;
+		goto out_unlock;
+	}
+
 	if (!nvdimm && cmd == ND_CMD_CLEAR_ERROR && cmd_rc >= 0) {
 		struct nd_cmd_clear_error *clear_err = buf;
 
Signature · 4 lines
2.24.1
_______________________________________________
Linux-nvdimm mailing list -- [email protected]
To unsubscribe send an email to [email protected]
Reviewed-by on v1 1/1
#1 lore raw
Quoted 18 lines · Vaibhav Jain

Presently the error code returned via out variable 'cmd_rc' from the nvdimm-bus controller function is ignored when called from __nd_ioctl() and never communicated back to user-space code that called an ioctl on dimm/bus.

This minor patch updates __nd_ioctl() to propagate the value of out variable 'cmd_rc' back to user-space in case it reports an error.

---
 drivers/nvdimm/bus.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/nvdimm/bus.c +0 -0
index a8b515968569..5b687a27fdf2 100644

drivers/nvdimm/bus.c static int __nd_ioctl(struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm, · line 1162

Quoted 10 lines
drivers/nvdimm/bus.c +5 -0
@@ -1153,6 +1153,11 @@ static int __nd_ioctl(struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm,
 	if (rc < 0)
 		goto out_unlock;
 
+	if (cmd_rc < 0) {
+		rc = cmd_rc;
+		goto out_unlock;
+	}
+
 	if (!nvdimm && cmd == ND_CMD_CLEAR_ERROR && cmd_rc >= 0) {
 		struct nd_cmd_clear_error *clear_err = buf;

Looks good to me.

_______________________________________________
Linux-nvdimm mailing list -- [email protected] To unsubscribe send an email to [email protected]

#2 lore raw
Quoted 34 lines · Jeff Moyer
Quoted 18 lines · Vaibhav Jain

Presently the error code returned via out variable 'cmd_rc' from the nvdimm-bus controller function is ignored when called from __nd_ioctl() and never communicated back to user-space code that called an ioctl on dimm/bus.

This minor patch updates __nd_ioctl() to propagate the value of out variable 'cmd_rc' back to user-space in case it reports an error.

---
 drivers/nvdimm/bus.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/nvdimm/bus.c +0 -0
index a8b515968569..5b687a27fdf2 100644

drivers/nvdimm/bus.c static int __nd_ioctl(struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm, · line 1162

Quoted 10 lines
drivers/nvdimm/bus.c +5 -0
@@ -1153,6 +1153,11 @@ static int __nd_ioctl(struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm,
      if (rc < 0)
              goto out_unlock;
 
+     if (cmd_rc < 0) {
+             rc = cmd_rc;
+             goto out_unlock;
+     }
+
      if (!nvdimm && cmd == ND_CMD_CLEAR_ERROR && cmd_rc >= 0) {
              struct nd_cmd_clear_error *clear_err = buf;

Looks good to me.

Quoted 1 line

Applied.
_______________________________________________
Linux-nvdimm mailing list -- [email protected] To unsubscribe send an email to [email protected]

#3 lore raw
Quoted 36 lines · Dan Williams
Quoted 34 lines · Jeff Moyer
Quoted 18 lines · Vaibhav Jain

Presently the error code returned via out variable 'cmd_rc' from the nvdimm-bus controller function is ignored when called from __nd_ioctl() and never communicated back to user-space code that called an ioctl on dimm/bus.

This minor patch updates __nd_ioctl() to propagate the value of out variable 'cmd_rc' back to user-space in case it reports an error.

---
 drivers/nvdimm/bus.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/nvdimm/bus.c +0 -0
index a8b515968569..5b687a27fdf2 100644

drivers/nvdimm/bus.c static int __nd_ioctl(struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm, · line 1162

Quoted 10 lines
drivers/nvdimm/bus.c +5 -0
@@ -1153,6 +1153,11 @@ static int __nd_ioctl(struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm,
      if (rc < 0)
              goto out_unlock;
 
+     if (cmd_rc < 0) {
+             rc = cmd_rc;
+             goto out_unlock;
+     }
+
      if (!nvdimm && cmd == ND_CMD_CLEAR_ERROR && cmd_rc >= 0) {
              struct nd_cmd_clear_error *clear_err = buf;

Looks good to me.

Quoted 3 lines
Quoted 1 line

Applied.

Unapplied. This breaks the NVDIMM unit test, and now that I look closer you are likely overlooking the fact that cmd_rc is a translation of the firmware status, while the ioctl rc is whether the command was successfully submitted. If you want the equivalent of cmd_rc in userspace you need to translate the firmware status. See ndctl_cmd_submit_xlat() in libndctl as an example of how the equivalent of cmd_rc is generated from the firmware status.
_______________________________________________
Linux-nvdimm mailing list -- [email protected]
To unsubscribe send an email to [email protected]

#4 lore raw

Hi Dan,

Apologies for responding late to this. My response below

Quoted 41 lines · Dan Williams
Quoted 36 lines · Dan Williams
Quoted 34 lines · Jeff Moyer
Quoted 18 lines · Vaibhav Jain

Presently the error code returned via out variable 'cmd_rc' from the nvdimm-bus controller function is ignored when called from __nd_ioctl() and never communicated back to user-space code that called an ioctl on dimm/bus.

This minor patch updates __nd_ioctl() to propagate the value of out variable 'cmd_rc' back to user-space in case it reports an error.

---
 drivers/nvdimm/bus.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/nvdimm/bus.c +0 -0
index a8b515968569..5b687a27fdf2 100644

drivers/nvdimm/bus.c static int __nd_ioctl(struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm, · line 1162

Quoted 10 lines
drivers/nvdimm/bus.c +5 -0
@@ -1153,6 +1153,11 @@ static int __nd_ioctl(struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm,
      if (rc < 0)
              goto out_unlock;
 
+     if (cmd_rc < 0) {
+             rc = cmd_rc;
+             goto out_unlock;
+     }
+
      if (!nvdimm && cmd == ND_CMD_CLEAR_ERROR && cmd_rc >= 0) {
              struct nd_cmd_clear_error *clear_err = buf;

Looks good to me.

Quoted 3 lines
Quoted 1 line

Applied.

Quoted 7 lines

Unapplied. This breaks the NVDIMM unit test, and now that I look closer you are likely overlooking the fact that cmd_rc is a translation of the firmware status, while the ioctl rc is whether the command was successfully submitted. If you want the equivalent of cmd_rc in userspace you need to translate the firmware status. See ndctl_cmd_submit_xlat() in libndctl as an example of how the equivalent of cmd_rc is generated from the firmware status.

This seems to be departure from rest of the libndvdimm where a non zero value of out-var 'cmd_rc' is treated as an error and communicated back to the caller.

However I agree to the points you made that semantics for __nd_ioctl() are different hence 'cmd_rc' need not be treated the same way as others. I think it will be better if these points are documented as code-comments in this function to make it more clearer as to why negative value 'cmd_rd' will be ignored and how userspace can get hold of it if needed.

Signature · 5 lines
Vaibhav Jain <[email protected]>
Linux Technology Center, IBM India Pvt. Ltd.
_______________________________________________
Linux-nvdimm mailing list -- [email protected]
To unsubscribe send an email to [email protected]