Skip to contentSkip to conversation

libsas cleanups

PATCH Applied Damien Le Moal 3 participants 14 messages Jul 25, 2025 – Aug 6, 2025

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

v4 cover libsas cleanups

Martin, John,

While debugging an issue with the pm8001 driver, I generated these cleanup patches. No functional changes overall.

These patches are against the 6.17/scsi-staging branch of the scsi tree.

Changes from v3:
- Corrected commit message of patch 2
- Added review tags

Changes from v2:
- Modified patch 2 to remove the unnecessary check using dev_is_ata()
- Added review tags

Changes from v1:
- Added patch 1
- Modified patch 2 to not inline sas_ata_wait_eh()
- Added review tags

Damien Le Moal (5):
scsi: libsas: Refactor dev_is_sata()
scsi: libsas: Simplify sas_ata_wait_eh()
scsi: libsas: Make sas_get_ata_info() static
scsi: libsas: Move declarations of internal functions to sas_internal.h
scsi: libsas: Use a bool for sas_deform_port() second argument

 drivers/scsi/libsas/sas_ata.c      | 10 +---
 drivers/scsi/libsas/sas_discover.c |  2 +-
 drivers/scsi/libsas/sas_internal.h | 78 ++++++++++++++++++++++++-
 drivers/scsi/libsas/sas_phy.c      |  6 +-
 drivers/scsi/libsas/sas_port.c     | 13 ++---
 include/scsi/sas_ata.h             | 91 +++++-------------------------
 6 files changed, 101 insertions(+), 99 deletions(-)

base-commit: 3ea3a256ed81f95ab0f3281a0e234b01a9cae605

Signature · 1 line
2.50.1
#1 lore raw

PATCH v4 1/5 scsi: libsas: Refactor dev_is_sata()

Use a switch statement in dev_is_sata() to make the code more readable (and probably slightly better than a series of or conditions). Also have this inline function return a boolean instead of an integer.

No functional changes.

---
 include/scsi/sas_ata.h | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)
include/scsi/sas_ata.h +12 -5
index 92e27e7bf088..8dddd0036f99 100644
@@ -15,10 +15,17 @@
 
 #ifdef CONFIG_SCSI_SAS_ATA
 
-static inline int dev_is_sata(struct domain_device *dev)
+static inline bool dev_is_sata(struct domain_device *dev)
 {
-	return dev->dev_type == SAS_SATA_DEV || dev->dev_type == SAS_SATA_PM ||
-	       dev->dev_type == SAS_SATA_PM_PORT || dev->dev_type == SAS_SATA_PENDING;
+	switch (dev->dev_type) {
+	case SAS_SATA_DEV:
+	case SAS_SATA_PENDING:
+	case SAS_SATA_PM:
+	case SAS_SATA_PM_PORT:
+		return true;
+	default:
+		return false;
+	}
 }
 
 int sas_get_ata_info(struct domain_device *dev, struct ex_phy *phy);
@@ -49,9 +56,9 @@ static inline void sas_ata_disabled_notice(void)
 	pr_notice_once("ATA device seen but CONFIG_SCSI_SAS_ATA=N\n");
 }
 
-static inline int dev_is_sata(struct domain_device *dev)
+static inline bool dev_is_sata(struct domain_device *dev)
 {
-	return 0;
+	return false;
 }
 static inline int sas_ata_init(struct domain_device *dev)
 {
Signature · 1 line
2.50.1
reviewer Reviewed-by on v4 1/5
#2 lore raw

Re: [PATCH v4 1/5] scsi: libsas: Refactor dev_is_sata()

Quoted 11 lines · Damien Le Moal

Use a switch statement in dev_is_sata() to make the code more readable (and probably slightly better than a series of or conditions). Also have this inline function return a boolean instead of an integer.

No functional changes.

---
  include/scsi/sas_ata.h | 17 ++++++++++++-----
  1 file changed, 12 insertions(+), 5 deletions(-)
#3 lore raw

PATCH v4 2/5 scsi: libsas: Simplify sas_ata_wait_eh()

In reply to Damien Le Moal

Simplify the code of sas_ata_wait_eh(), removing the local variable ap for the pointer to the device ata_port structure. The test using dev_is_sata() is also removed as all call sites of this function check if the device is a SATA one before calling this function.

---
 drivers/scsi/libsas/sas_ata.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)
drivers/scsi/libsas/sas_ata.c +1 -7
index 7b4e7a61965a..2cbf38b18c5c 100644
@@ -927,13 +927,7 @@ EXPORT_SYMBOL_GPL(sas_ata_schedule_reset);
 
 void sas_ata_wait_eh(struct domain_device *dev)
 {
-	struct ata_port *ap;
-
-	if (!dev_is_sata(dev))
-		return;
-
-	ap = dev->sata_dev.ap;
-	ata_port_wait_eh(ap);
+	ata_port_wait_eh(dev->sata_dev.ap);
 }
 
 void sas_ata_device_link_abort(struct domain_device *device, bool force_reset)
Signature · 1 line
2.50.1
reviewer Reviewed-by on v4 2/5
#4 lore raw

Re: [PATCH v4 2/5] scsi: libsas: Simplify sas_ata_wait_eh()

Quoted 10 lines · Damien Le Moal

Simplify the code of sas_ata_wait_eh(), removing the local variable ap for the pointer to the device ata_port structure. The test using dev_is_sata() is also removed as all call sites of this function check if the device is a SATA one before calling this function.

---
  drivers/scsi/libsas/sas_ata.c | 8 +-------
  1 file changed, 1 insertion(+), 7 deletions(-)
#5 lore raw

PATCH v4 3/5 scsi: libsas: Make sas_get_ata_info() static

In reply to Damien Le Moal

The function sas_get_ata_info() is used only in drivers/scsi/libsas/sas_ata.c. Remove its definition from include/scsi/sas_ata.h and make this function static.

No functional changes.

---
 drivers/scsi/libsas/sas_ata.c | 2 +-
 include/scsi/sas_ata.h        | 6 ------
 2 files changed, 1 insertion(+), 7 deletions(-)
drivers/scsi/libsas/sas_ata.c +1 -1
index 2cbf38b18c5c..cc093cdc9c69 100644
@@ -252,7 +252,7 @@ static int sas_get_ata_command_set(struct domain_device *dev)
 	return ata_dev_classify(&tf);
 }
 
-int sas_get_ata_info(struct domain_device *dev, struct ex_phy *phy)
+static int sas_get_ata_info(struct domain_device *dev, struct ex_phy *phy)
 {
 	if (phy->attached_tproto & SAS_PROTOCOL_STP)
 		dev->tproto = phy->attached_tproto;
include/scsi/sas_ata.h +0 -6
index 8dddd0036f99..5e3475975aee 100644
@@ -28,7 +28,6 @@ static inline bool dev_is_sata(struct domain_device *dev)
 	}
 }
 
-int sas_get_ata_info(struct domain_device *dev, struct ex_phy *phy);
 int sas_ata_init(struct domain_device *dev);
 void sas_ata_task_abort(struct sas_task *task);
 void sas_ata_strategy_handler(struct Scsi_Host *shost);
@@ -96,11 +95,6 @@ static inline void sas_resume_sata(struct asd_sas_port *port)
 {
 }
 
-static inline int sas_get_ata_info(struct domain_device *dev, struct ex_phy *phy)
-{
-	return 0;
-}
-
 static inline void sas_ata_end_eh(struct ata_port *ap)
 {
 }
Signature · 1 line
2.50.1
reviewer Reviewed-by on v4 3/5
#6 lore raw

Re: [PATCH v4 3/5] scsi: libsas: Make sas_get_ata_info() static

Quoted 6 lines · Damien Le Moal

The function sas_get_ata_info() is used only in drivers/scsi/libsas/sas_ata.c. Remove its definition from include/scsi/sas_ata.h and make this function static.

No functional changes.

Quoted 7 lines
---
  drivers/scsi/libsas/sas_ata.c | 2 +-
  include/scsi/sas_ata.h        | 6 ------
  2 files changed, 1 insertion(+), 7 deletions(-)
#7 lore raw

PATCH v4 4/5 scsi: libsas: Move declarations of internal functions to sas_internal.h

In reply to Damien Le Moal

Move the declaration of all functions used only within libsas from include/scsi/sas_ata.h to drivers/scsi/libsas/sas_internal.h.

No functional changes.

---
 drivers/scsi/libsas/sas_internal.h | 74 ++++++++++++++++++++++++++++++
 include/scsi/sas_ata.h             | 68 +--------------------------
 2 files changed, 75 insertions(+), 67 deletions(-)
drivers/scsi/libsas/sas_internal.h +74 -0
index 03d6ec1eb970..16f8d81d7531 100644
@@ -222,4 +222,78 @@ static inline void sas_put_device(struct domain_device *dev)
 	kref_put(&dev->kref, sas_free_device);
 }
 
+#ifdef CONFIG_SCSI_SAS_ATA
+
+int sas_ata_init(struct domain_device *dev);
+void sas_ata_task_abort(struct sas_task *task);
+int sas_discover_sata(struct domain_device *dev);
+int sas_ata_add_dev(struct domain_device *parent, struct ex_phy *phy,
+		    struct domain_device *child, int phy_id);
+void sas_ata_strategy_handler(struct Scsi_Host *shost);
+void sas_ata_eh(struct Scsi_Host *shost, struct list_head *work_q);
+void sas_ata_end_eh(struct ata_port *ap);
+void sas_ata_wait_eh(struct domain_device *dev);
+void sas_probe_sata(struct asd_sas_port *port);
+void sas_suspend_sata(struct asd_sas_port *port);
+void sas_resume_sata(struct asd_sas_port *port);
+
+#else
+
+static inline int sas_ata_init(struct domain_device *dev)
+{
+	return 0;
+}
+
+static inline void sas_ata_task_abort(struct sas_task *task)
+{
+}
+
+static inline void sas_ata_strategy_handler(struct Scsi_Host *shost)
+{
+}
+
+static inline void sas_ata_eh(struct Scsi_Host *shost, struct list_head *work_q)
+{
+}
+
+static inline void sas_ata_end_eh(struct ata_port *ap)
+{
+}
+
+static inline void sas_ata_wait_eh(struct domain_device *dev)
+{
+}
+
+static inline void sas_probe_sata(struct asd_sas_port *port)
+{
+}
+
+static inline void sas_suspend_sata(struct asd_sas_port *port)
+{
+}
+
+static inline void sas_resume_sata(struct asd_sas_port *port)
+{
+}
+
+static inline void sas_ata_disabled_notice(void)
+{
+	pr_notice_once("ATA device seen but CONFIG_SCSI_SAS_ATA=N\n");
+}
+
+static inline int sas_discover_sata(struct domain_device *dev)
+{
+	sas_ata_disabled_notice();
+	return -ENXIO;
+}
+
+static inline int sas_ata_add_dev(struct domain_device *parent, struct ex_phy *phy,
+				  struct domain_device *child, int phy_id)
+{
+	sas_ata_disabled_notice();
+	return -ENODEV;
+}
+
+#endif
+
 #endif /* _SAS_INTERNAL_H_ */
include/scsi/sas_ata.h +1 -67

97 lines not on this page · raw message

Signature · 1 line
2.50.1
reviewer Reviewed-by on v4 4/5
#8 lore raw

Re: [PATCH v4 4/5] scsi: libsas: Move declarations of internal functions to sas_internal.h

Quoted 5 lines · Damien Le Moal

Move the declaration of all functions used only within libsas from include/scsi/sas_ata.h to drivers/scsi/libsas/sas_internal.h.

No functional changes.

Quoted 7 lines
---
  drivers/scsi/libsas/sas_internal.h | 74 ++++++++++++++++++++++++++++++
  include/scsi/sas_ata.h             | 68 +--------------------------
  2 files changed, 75 insertions(+), 67 deletions(-)
#9 lore raw

PATCH v4 5/5 scsi: libsas: Use a bool for sas_deform_port() second argument

In reply to Damien Le Moal

Change the type of the "gone" argument of sas_deform_port() from int to bool. Simliarly, to be consistent, do the same change to the function sas_unregister_domain_devices().

No functional changes.

---
 drivers/scsi/libsas/sas_discover.c |  2 +-
 drivers/scsi/libsas/sas_internal.h |  4 ++--
 drivers/scsi/libsas/sas_phy.c      |  6 +++---
 drivers/scsi/libsas/sas_port.c     | 13 ++++++-------
 4 files changed, 12 insertions(+), 13 deletions(-)
drivers/scsi/libsas/sas_discover.c +1 -1
index 951bdc554a10..b07062db50b2 100644
@@ -406,7 +406,7 @@ void sas_unregister_dev(struct asd_sas_port *port, struct domain_device *dev)
 	}
 }
 
-void sas_unregister_domain_devices(struct asd_sas_port *port, int gone)
+void sas_unregister_domain_devices(struct asd_sas_port *port, bool gone)
 {
 	struct domain_device *dev, *n;
 
drivers/scsi/libsas/sas_internal.h +2 -2
index 16f8d81d7531..6706f2be8d27 100644
@@ -44,7 +44,7 @@ void sas_hash_addr(u8 *hashed, const u8 *sas_addr);
 int sas_discover_root_expander(struct domain_device *dev);
 
 int sas_ex_revalidate_domain(struct domain_device *dev);
-void sas_unregister_domain_devices(struct asd_sas_port *port, int gone);
+void sas_unregister_domain_devices(struct asd_sas_port *port, bool gone);
 void sas_init_disc(struct sas_discovery *disc, struct asd_sas_port *port);
 void sas_discover_event(struct asd_sas_port *port, enum discover_event ev);
 
@@ -70,7 +70,7 @@ void sas_enable_revalidation(struct sas_ha_struct *ha);
 void sas_queue_deferred_work(struct sas_ha_struct *ha);
 void __sas_drain_work(struct sas_ha_struct *ha);
 
-void sas_deform_port(struct asd_sas_phy *phy, int gone);
+void sas_deform_port(struct asd_sas_phy *phy, bool gone);
 
 void sas_porte_bytes_dmaed(struct work_struct *work);
 void sas_porte_broadcast_rcvd(struct work_struct *work);
drivers/scsi/libsas/sas_phy.c +3 -3
index 57494ac97076..635835c28ecd 100644
@@ -20,7 +20,7 @@ static void sas_phye_loss_of_signal(struct work_struct *work)
 	struct asd_sas_phy *phy = ev->phy;
 
 	phy->error = 0;
-	sas_deform_port(phy, 1);
+	sas_deform_port(phy, true);
 }
 
 static void sas_phye_oob_done(struct work_struct *work)
@@ -40,7 +40,7 @@ static void sas_phye_oob_error(struct work_struct *work)
 	struct sas_internal *i =
 		to_sas_internal(sas_ha->shost->transportt);
 
-	sas_deform_port(phy, 1);
+	sas_deform_port(phy, true);
 
 	if (!port && phy->enabled && i->dft->lldd_control_phy) {
 		phy->error++;
@@ -85,7 +85,7 @@ static void sas_phye_resume_timeout(struct work_struct *work)
 
 	phy->error = 0;
 	phy->suspended = 0;
-	sas_deform_port(phy, 1);
+	sas_deform_port(phy, true);
 }
 
 
drivers/scsi/libsas/sas_port.c +6 -7
index e3f2ed913419..de7556070048 100644
@@ -113,7 +113,7 @@ static void sas_form_port(struct asd_sas_phy *phy)
 
 	if (port) {
 		if (!phy_is_wideport_member(port, phy))
-			sas_deform_port(phy, 0);
+			sas_deform_port(phy, false);
 		else if (phy->suspended) {
 			phy->suspended = 0;
 			sas_resume_port(phy);
@@ -206,7 +206,7 @@ static void sas_form_port(struct asd_sas_phy *phy)
  * This is called when the physical link to the other phy has been
  * lost (on this phy), in Event thread context. We cannot delay here.
  */
-void sas_deform_port(struct asd_sas_phy *phy, int gone)
+void sas_deform_port(struct asd_sas_phy *phy, bool gone)
 {
 	struct sas_ha_struct *sas_ha = phy->ha;
 	struct asd_sas_port *port = phy->port;
@@ -301,7 +301,7 @@ void sas_porte_link_reset_err(struct work_struct *work)
 	struct asd_sas_event *ev = to_asd_sas_event(work);
 	struct asd_sas_phy *phy = ev->phy;
 
-	sas_deform_port(phy, 1);
+	sas_deform_port(phy, true);
 }
 
 void sas_porte_timer_event(struct work_struct *work)
@@ -309,7 +309,7 @@ void sas_porte_timer_event(struct work_struct *work)
 	struct asd_sas_event *ev = to_asd_sas_event(work);
 	struct asd_sas_phy *phy = ev->phy;
 
-	sas_deform_port(phy, 1);
+	sas_deform_port(phy, true);
 }
 
 void sas_porte_hard_reset(struct work_struct *work)
@@ -317,7 +317,7 @@ void sas_porte_hard_reset(struct work_struct *work)
 	struct asd_sas_event *ev = to_asd_sas_event(work);
 	struct asd_sas_phy *phy = ev->phy;
 
-	sas_deform_port(phy, 1);
+	sas_deform_port(phy, true);
 }
 
 /* ---------- SAS port registration ---------- */
@@ -358,8 +358,7 @@ void sas_unregister_ports(struct sas_ha_struct *sas_ha)
 
 	for (i = 0; i < sas_ha->num_phys; i++)
 		if (sas_ha->sas_phy[i]->port)
-			sas_deform_port(sas_ha->sas_phy[i], 0);
-
+			sas_deform_port(sas_ha->sas_phy[i], false);
 }
 
 const work_func_t sas_port_event_fns[PORT_NUM_EVENTS] = {
Signature · 1 line
2.50.1
reviewer Reviewed-by on v4 5/5
#10 lore raw

Re: [PATCH v4 5/5] scsi: libsas: Use a bool for sas_deform_port() second argument

Quoted 9 lines · Damien Le Moal

Change the type of the "gone" argument of sas_deform_port() from int to bool. Simliarly, to be consistent, do the same change to the function sas_unregister_domain_devices().

No functional changes.

Quoted 6 lines
---
  drivers/scsi/libsas/sas_discover.c |  2 +-
  drivers/scsi/libsas/sas_internal.h |  4 ++--
  drivers/scsi/libsas/sas_phy.c      |  6 +++---
  drivers/scsi/libsas/sas_port.c     | 13 ++++++-------
  4 files changed, 12 insertions(+), 13 deletions(-)
#11 lore raw
In reply to Damien Le Moal

Damien,

Quoted 2 lines

While debugging an issue with the pm8001 driver, I generated these cleanup patches. No functional changes overall.

Applied to 6.17/scsi-staging, thanks!

Signature · 1 line
Martin K. Petersen
#12 lore raw
In reply to Damien Le Moal
Quoted 8 lines · Damien Le Moal

Martin, John,

While debugging an issue with the pm8001 driver, I generated these cleanup patches. No functional changes overall.

These patches are against the 6.17/scsi-staging branch of the scsi tree.

[...]

Applied to 6.17/scsi-queue, thanks!

[1/5] scsi: libsas: Refactor dev_is_sata()

      https://git.kernel.org/mkp/scsi/c/54091eee08ac
[2/5] scsi: libsas: Simplify sas_ata_wait_eh()
      https://git.kernel.org/mkp/scsi/c/0dd03570512a
[3/5] scsi: libsas: Make sas_get_ata_info() static
      https://git.kernel.org/mkp/scsi/c/bd31394aabf3
[4/5] scsi: libsas: Move declarations of internal functions to sas_internal.h
      https://git.kernel.org/mkp/scsi/c/704ed03abf6b
[5/5] scsi: libsas: Use a bool for sas_deform_port() second argument
      https://git.kernel.org/mkp/scsi/c/75fe230b9bed
Signature · 1 line
Martin K. Petersen	Oracle Linux Engineering
#13 lore raw
In reply to Damien Le Moal
Quoted 8 lines · Damien Le Moal

Martin, John,

While debugging an issue with the pm8001 driver, I generated these cleanup patches. No functional changes overall.

These patches are against the 6.17/scsi-staging branch of the scsi tree.

[...]

Applied to 6.17/scsi-queue, thanks!

[1/5] scsi: libsas: Refactor dev_is_sata()

      https://git.kernel.org/mkp/scsi/c/54091eee08ac
[2/5] scsi: libsas: Simplify sas_ata_wait_eh()
      https://git.kernel.org/mkp/scsi/c/0dd03570512a
[3/5] scsi: libsas: Make sas_get_ata_info() static
      https://git.kernel.org/mkp/scsi/c/bd31394aabf3
[4/5] scsi: libsas: Move declarations of internal functions to sas_internal.h
      https://git.kernel.org/mkp/scsi/c/704ed03abf6b
[5/5] scsi: libsas: Use a bool for sas_deform_port() second argument
      https://git.kernel.org/mkp/scsi/c/75fe230b9bed
Signature · 1 line
Martin K. Petersen	Oracle Linux Engineering