Re: [PATCH 6/7] scsi: pm8001: use dev_and_phy_addr_same() instead of open coded

From: Jason Yan
Date: Fri Sep 23 2022 - 05:46:51 EST



On 2022/9/22 22:24, John Garry wrote:
On 17/09/2022 11:43, Jason Yan wrote:
The sas address comparation of domain device and expander phy is open
coded. Now we can replace it with dev_and_phy_addr_same().

Signed-off-by: Jason Yan <yanaijie@xxxxxxxxxx>
---
  drivers/scsi/pm8001/pm8001_sas.c | 3 +--
  1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/scsi/pm8001/pm8001_sas.c b/drivers/scsi/pm8001/pm8001_sas.c
index 8e3f2f9ddaac..bb1b1722f3ee 100644
--- a/drivers/scsi/pm8001/pm8001_sas.c
+++ b/drivers/scsi/pm8001/pm8001_sas.c
@@ -649,8 +649,7 @@ static int pm8001_dev_found_notify(struct domain_device *dev)
          for (phy_id = 0; phy_id < parent_dev->ex_dev.num_phys;

This code seems the same between many libsas LLDDs - could we factor it out into libsas? If so, then maybe those new helpers could be put in sas_internal.h

For the part of putting helpers in sas_internal.h, this needs to make the helpers exported. I think it's not worth to do this because they are very small. I'd still like to make them inline functions in libsas.h such as:


diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
index 2dbead74a2af..e9e76c898287 100644
--- a/include/scsi/libsas.h
+++ b/include/scsi/libsas.h
@@ -648,6 +648,22 @@ static inline bool sas_is_internal_abort(struct sas_task *task)
return task->task_proto == SAS_PROTOCOL_INTERNAL_ABORT;
}

+static inline int sas_find_attathed_phy(struct expander_device *ex_dev,
+ struct domain_device *dev)
+{
+ struct ex_phy *phy;
+ int phy_id;
+
+ for (phy_id = 0; phy_id < ex_dev->num_phys; phy_id++) {
+ phy = &ex_dev->ex_phy[phy_id];
+ if (SAS_ADDR(phy->attached_sas_addr)
+ == SAS_ADDR(dev->sas_addr))
+ return phy_id;
+ }
+
+ return ex_dev->num_phys;
+}
+
struct sas_domain_function_template {
/* The class calls these to notify the LLDD of an event. */
void (*lldd_port_formed)(struct asd_sas_phy *);



And the LLDDs change like:


diff --git a/drivers/scsi/pm8001/pm8001_sas.c b/drivers/scsi/pm8001/pm8001_sas.c
index 8e3f2f9ddaac..4e7350609b3d 100644
--- a/drivers/scsi/pm8001/pm8001_sas.c
+++ b/drivers/scsi/pm8001/pm8001_sas.c
@@ -645,16 +645,8 @@ static int pm8001_dev_found_notify(struct domain_device *dev)
pm8001_device->dcompletion = &completion;
if (parent_dev && dev_is_expander(parent_dev->dev_type)) {
int phy_id;
- struct ex_phy *phy;
- for (phy_id = 0; phy_id < parent_dev->ex_dev.num_phys;
- phy_id++) {
- phy = &parent_dev->ex_dev.ex_phy[phy_id];
- if (SAS_ADDR(phy->attached_sas_addr)
- == SAS_ADDR(dev->sas_addr)) {
- pm8001_device->attached_phy = phy_id;
- break;
- }
- }
+
+ phy_id = sas_find_attathed_phy(&parent_dev->ex_dev, dev);
if (phy_id == parent_dev->ex_dev.num_phys) {
pm8001_dbg(pm8001_ha, FAIL,
"Error: no attached dev:%016llx at ex:%016llx.\n",
@@ -662,6 +654,7 @@ static int pm8001_dev_found_notify(struct domain_device *dev)
SAS_ADDR(parent_dev->sas_addr));
res = -1;
}
+ pm8001_device->attached_phy = phy_id;
} else {
if (dev->dev_type == SAS_SATA_DEV) {
pm8001_device->attached_phy =


So I wonder if you have any reasons to insist exporting the helper?


Thanks,
John

          phy_id++) {
              phy = &parent_dev->ex_dev.ex_phy[phy_id];
-            if (SAS_ADDR(phy->attached_sas_addr)
-                == SAS_ADDR(dev->sas_addr)) {
+            if (dev_and_phy_addr_same(dev, phy)) {
                  pm8001_device->attached_phy = phy_id;
                  break;
              }

.