Skip to contentSkip to conversation

Fix libata suspend/resume handling and code cleanup

PATCH Superseded Damien Le Moal 7 participants 80 messages Sep 11, 2023 – Sep 14, 2023

Subsystems (3)

v1 of 8 Superseded Latest is v8 →

lore
Share & tools
lore

Damien Le Moal

author maintainer
#0 lore raw

cover Fix libata suspend/resume handling and code cleanup

The first 7 patch of this series fix several issues with suspend/resume power management operations in libata. The most important change introduced is in patch 4, where the manage_start_stop of scsi devices associated with ata devices is disabled and this functionality replaced by a direct handling within libata EH of device suspend/resume (i.e. spin down for suspend and spin up for resume).

The remaining patches are code cleanup that do not introduce any significant functional change.

This series was tested on qemu and on various PCs and servers. I am CC-ing people who recently reported issues with suspend/resume. Additional testing would be much appreciated.

Of note is that there is no change to the overall suspend/resume model of libata where suspend and resume operations are tied to the ata port. This is somewhat broken as a port may have multiple devices (e.g. pata and port multiplier cases). Fixing this model (to be more similar to what libsas does) will be the next step.

Damien Le Moal (19):
ata: libata-core: Fix ata_port_request_pm() locking
ata: libata-core: Fix port and device removal
ata: libata-scsi: link ata port and scsi device
ata: libata-scsi: Disable scsi device manage_start_stop
ata: libata-scsi: Fix delayed scsi_rescan_device() execution
ata: libata-core: Do not register PM operations for SAS ports
scsi: sd: Do not issue commands to suspended disks on remove
scsi: Remove scsi device no_start_on_resume flag
ata: libata-scsi: Cleanup ata_scsi_start_stop_xlat()
ata: libata-core: Synchronize ata_port_detach() with hotplug
ata: libata-core: Detach a port devices on shutdown
ata: libata-core: Remove ata_port_suspend_async()
ata: libata-core: Remove ata_port_resume_async()
ata: libata-core: skip poweroff for devices that are runtime suspended
ata: libata-core: Do not resume ports that have been runtime suspended
ata: libata-sata: Improve ata_sas_slave_configure()
ata: libata-eh: Improve reset error messages
ata: libata-eh: Reduce "disable device" message verbosity
ata: libata: Cleanup inline DMA helper functions

 drivers/ata/libata-core.c      | 239 +++++++++++++++++++++++++--------
 drivers/ata/libata-eh.c        |  74 ++++++++--
 drivers/ata/libata-sata.c      |   5 +-
 drivers/ata/libata-scsi.c      | 143 ++++++++++----------
 drivers/ata/libata-transport.c |   9 +-
 drivers/ata/libata.h           |   7 +
 drivers/ata/pata_macio.c       |   1 +
 drivers/ata/sata_mv.c          |   1 +
 drivers/ata/sata_nv.c          |   2 +
 drivers/ata/sata_sil24.c       |   1 +
 drivers/scsi/scsi_scan.c       |  12 +-
 drivers/scsi/sd.c              |  10 +-
 include/linux/libata.h         |  27 ++--
 include/scsi/scsi_device.h     |   1 -
 include/scsi/scsi_host.h       |   2 +-
 15 files changed, 370 insertions(+), 164 deletions(-)
Signature · 1 line
2.41.0
author maintainer
#1 lore raw

PATCH 1/19 ata: libata-core: Fix ata_port_request_pm() locking

The function ata_port_request_pm() checks the port flag ATA_PFLAG_PM_PENDING and calls ata_port_wait_eh() if this flag is set to ensure that power management operations for a port are not secheduled simultaneously. However, this flag check is done without holding the port lock.

Fix this by taking the port lock on entry to the function and checking the flag under this lock. The lock is released and re-taken if ata_port_wait_eh() needs to be called.

Fixes 5ef41082912b ("ata: add ata port system PM callbacks")
---
 drivers/ata/libata-core.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)
drivers/ata/libata-core.c +9 -8
index 74314311295f..c4898483d716 100644
@@ -5040,17 +5040,20 @@ static void ata_port_request_pm(struct ata_port *ap, pm_message_t mesg,
 	struct ata_link *link;
 	unsigned long flags;
 
-	/* Previous resume operation might still be in
-	 * progress.  Wait for PM_PENDING to clear.
+	spin_lock_irqsave(ap->lock, flags);
+
+	/*
+	 * A previous PM operation might still be in progress. Wait for
+	 * ATA_PFLAG_PM_PENDING to clear.
 	 */
 	if (ap->pflags & ATA_PFLAG_PM_PENDING) {
+		spin_unlock_irqrestore(ap->lock, flags);
 		ata_port_wait_eh(ap);
+		spin_lock_irqsave(ap->lock, flags);
 		WARN_ON(ap->pflags & ATA_PFLAG_PM_PENDING);
 	}
 
-	/* request PM ops to EH */
-	spin_lock_irqsave(ap->lock, flags);
-
+	/* Request PM operation to EH */
 	ap->pm_mesg = mesg;
 	ap->pflags |= ATA_PFLAG_PM_PENDING;
 	ata_for_each_link(link, ap, HOST_FIRST) {
@@ -5062,10 +5065,8 @@ static void ata_port_request_pm(struct ata_port *ap, pm_message_t mesg,
 
 	spin_unlock_irqrestore(ap->lock, flags);
 
-	if (!async) {
+	if (!async)
 		ata_port_wait_eh(ap);
-		WARN_ON(ap->pflags & ATA_PFLAG_PM_PENDING);
-	}
 }
 
 /*
Signature · 1 line
2.41.0
Reviewed-by on v1 1/19
#2 lore raw

Re: [PATCH 01/19] ata: libata-core: Fix ata_port_request_pm() locking

Quoted 10 lines · Damien Le Moal

The function ata_port_request_pm() checks the port flag ATA_PFLAG_PM_PENDING and calls ata_port_wait_eh() if this flag is set to ensure that power management operations for a port are not secheduled simultaneously. However, this flag check is done without holding the port lock.

Fix this by taking the port lock on entry to the function and checking the flag under this lock. The lock is released and re-taken if ata_port_wait_eh() needs to be called.

Quoted 7 lines
Fixes 5ef41082912b ("ata: add ata port system PM callbacks")
---
  drivers/ata/libata-core.c | 17 +++++++++--------
  1 file changed, 9 insertions(+), 8 deletions(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
Tested-by on v1 1/19
#3 lore raw

Re: [PATCH 01/19] ata: libata-core: Fix ata_port_request_pm() locking

In reply to Damien Le Moal
Quoted 6 lines · Damien Le Moal

The function ata_port_request_pm() checks the port flag ATA_PFLAG_PM_PENDING and calls ata_port_wait_eh() if this flag is set to ensure that power management operations for a port are not secheduled simultaneously. However, this flag check is done without holding the port lock.

Quoted 7 lines

Fix this by taking the port lock on entry to the function and checking the flag under this lock. The lock is released and re-taken if ata_port_wait_eh() needs to be called.

Fixes 5ef41082912b ("ata: add ata port system PM callbacks")
author maintainer
#4 lore raw

PATCH 2/19 ata: libata-core: Fix port and device removal

In reply to Damien Le Moal

Whenever an ATA adapter driver is removed (e.g. rmmod), ata_port_detach() is called repeatedly for all the adapter ports to remove (unload) the devices attached to the port and delete the port device itself. Removing of devices is done using libata EH with the ATA_PFLAG_UNLOADING port flag set. This causes libata EH to execute ata_eh_unload() which disables all devices attached to the port.

ata_port_detach() finishes by calling scsi_remove_host() to remove the scsi host associated with the port. This function will trigger the removal of all scsi devices attached to the host and in the case of disks, calls to sd_shutdown() which will flush the device write cache and stop the device. However, given that the devices were already disabled by ata_eh_unload(), the synchronize write cache command and start stop unit commands fail. E.g. running "rmmod ahci" with first removing sd_mod results in error messages like:

ata13.00: disable device
sd 0:0:0:0: [sda] Synchronizing SCSI cache
sd 0:0:0:0: [sda] Synchronize Cache(10) failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK sd 0:0:0:0: [sda] Stopping disk
sd 0:0:0:0: [sda] Start/Stop Unit failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK

Fix this by removing all scsi devices of the ata devices connected to the port before scheduling libata EH to disable the ATA devices.

Fixes 720ba12620ee ("[PATCH] libata-hp: update unload-unplug")
---
 drivers/ata/libata-core.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)
drivers/ata/libata-core.c +20 -1
index c4898483d716..693cb3cd70cd 100644
@@ -5952,11 +5952,30 @@ static void ata_port_detach(struct ata_port *ap)
 	struct ata_link *link;
 	struct ata_device *dev;
 
-	/* tell EH we're leaving & flush EH */
+	/* Wait for any ongoing EH */
+	ata_port_wait_eh(ap);
+
+	mutex_lock(&ap->scsi_scan_mutex);
 	spin_lock_irqsave(ap->lock, flags);
+
+	/* Remove scsi devices */
+	ata_for_each_link(link, ap, HOST_FIRST) {
+		ata_for_each_dev(dev, link, ALL) {
+			if (dev->sdev) {
+				spin_unlock_irqrestore(ap->lock, flags);
+				scsi_remove_device(dev->sdev);
+				spin_lock_irqsave(ap->lock, flags);
+				dev->sdev = NULL;
+			}
+		}
+	}
+
+	/* Tell EH to disable all devices */
 	ap->pflags |= ATA_PFLAG_UNLOADING;
 	ata_port_schedule_eh(ap);
+
 	spin_unlock_irqrestore(ap->lock, flags);
+	mutex_unlock(&ap->scsi_scan_mutex);
 
 	/* wait till EH commits suicide */
 	ata_port_wait_eh(ap);
Signature · 1 line
2.41.0
#5 lore raw

Re: [PATCH 02/19] ata: libata-core: Fix port and device removal

Quoted 56 lines · Damien Le Moal

Whenever an ATA adapter driver is removed (e.g. rmmod), ata_port_detach() is called repeatedly for all the adapter ports to remove (unload) the devices attached to the port and delete the port device itself. Removing of devices is done using libata EH with the ATA_PFLAG_UNLOADING port flag set. This causes libata EH to execute ata_eh_unload() which disables all devices attached to the port.

ata_port_detach() finishes by calling scsi_remove_host() to remove the scsi host associated with the port. This function will trigger the removal of all scsi devices attached to the host and in the case of disks, calls to sd_shutdown() which will flush the device write cache and stop the device. However, given that the devices were already disabled by ata_eh_unload(), the synchronize write cache command and start stop unit commands fail. E.g. running "rmmod ahci" with first removing sd_mod results in error messages like:

ata13.00: disable device
sd 0:0:0:0: [sda] Synchronizing SCSI cache
sd 0:0:0:0: [sda] Synchronize Cache(10) failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK sd 0:0:0:0: [sda] Stopping disk
sd 0:0:0:0: [sda] Start/Stop Unit failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK

Fix this by removing all scsi devices of the ata devices connected to the port before scheduling libata EH to disable the ATA devices.

Fixes 720ba12620ee ("[PATCH] libata-hp: update unload-unplug")
---
  drivers/ata/libata-core.c | 21 ++++++++++++++++++++-
  1 file changed, 20 insertions(+), 1 deletion(-)
drivers/ata/libata-core.c +14 -1
index c4898483d716..693cb3cd70cd 100644
@@ -5952,11 +5952,30 @@ static void ata_port_detach(struct ata_port *ap)
  	struct ata_link *link;
  	struct ata_device *dev;
 
-	/* tell EH we're leaving & flush EH */
+	/* Wait for any ongoing EH */
+	ata_port_wait_eh(ap);
+
+	mutex_lock(&ap->scsi_scan_mutex);
  	spin_lock_irqsave(ap->lock, flags);
+
+	/* Remove scsi devices */
+	ata_for_each_link(link, ap, HOST_FIRST) {
+		ata_for_each_dev(dev, link, ALL) {
+			if (dev->sdev) {
+				spin_unlock_irqrestore(ap->lock, flags);
+				scsi_remove_device(dev->sdev);
+				spin_lock_irqsave(ap->lock, flags);
+				dev->sdev = NULL;
+			}

drivers/ata/libata-core.c static void ata_port_detach(struct ata_port *ap) · line 5979

Quoted 10 lines
drivers/ata/libata-core.c +6 -0
+		}
+	}
+
+	/* Tell EH to disable all devices */
  	ap->pflags |= ATA_PFLAG_UNLOADING;
  	ata_port_schedule_eh(ap);
+
  	spin_unlock_irqrestore(ap->lock, flags);
+	mutex_unlock(&ap->scsi_scan_mutex);
 

Are you sure about releasing scan_mutex after ata_port_schedule_eh()?
Technically ata_port_schedule_eh() will be calling into SCSI rescan, which would take scan_mutex ... Not that it matter much, seeing that we've disconnected all devices, but still ...

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
author maintainer
#6 lore raw

Re: [PATCH 02/19] ata: libata-core: Fix port and device removal

Quoted 64 lines · Hannes Reinecke
Quoted 63 lines · Damien Le Moal

Whenever an ATA adapter driver is removed (e.g. rmmod), ata_port_detach() is called repeatedly for all the adapter ports to remove (unload) the devices attached to the port and delete the port device itself. Removing of devices is done using libata EH with the ATA_PFLAG_UNLOADING port flag set. This causes libata EH to execute ata_eh_unload() which disables all devices attached to the port.

ata_port_detach() finishes by calling scsi_remove_host() to remove the scsi host associated with the port. This function will trigger the removal of all scsi devices attached to the host and in the case of disks, calls to sd_shutdown() which will flush the device write cache and stop the device. However, given that the devices were already disabled by ata_eh_unload(), the synchronize write cache command and start stop unit commands fail. E.g. running "rmmod ahci" with first removing sd_mod results in error messages like:

ata13.00: disable device
sd 0:0:0:0: [sda] Synchronizing SCSI cache
sd 0:0:0:0: [sda] Synchronize Cache(10) failed: Result:
hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK
sd 0:0:0:0: [sda] Stopping disk
sd 0:0:0:0: [sda] Start/Stop Unit failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK

Fix this by removing all scsi devices of the ata devices connected to the port before scheduling libata EH to disable the ATA devices.

Fixes 720ba12620ee ("[PATCH] libata-hp: update unload-unplug")

---
  drivers/ata/libata-core.c | 21 ++++++++++++++++++++-

  1 file changed, 20 insertions(+), 1 deletion(-)
drivers/ata/libata-core.c +0 -0
index c4898483d716..693cb3cd70cd 100644
@@ -5952,11 +5952,30 @@ static void ata_port_detach(struct ata_port *ap)
       struct ata_link *link;
       struct ata_device *dev;

  -    /* tell EH we're leaving & flush EH */
+    /* Wait for any ongoing EH */
+    ata_port_wait_eh(ap);
+
+    mutex_lock(&ap->scsi_scan_mutex);
      spin_lock_irqsave(ap->lock, flags);

drivers/ata/libata-core.c +14 -0
+
+    /* Remove scsi devices */
+    ata_for_each_link(link, ap, HOST_FIRST) {
+        ata_for_each_dev(dev, link, ALL) {
+            if (dev->sdev) {
+                spin_unlock_irqrestore(ap->lock, flags);
+                scsi_remove_device(dev->sdev);
+                spin_lock_irqsave(ap->lock, flags);
+                dev->sdev = NULL;
+            }
+        }
+    }
+
+    /* Tell EH to disable all devices */

      ap->pflags |= ATA_PFLAG_UNLOADING;
      ata_port_schedule_eh(ap);

Quoted 8 lines
Quoted 4 lines

+
      spin_unlock_irqrestore(ap->lock, flags);
+    mutex_unlock(&ap->scsi_scan_mutex);

Are you sure about releasing scan_mutex after ata_port_schedule_eh()? Technically ata_port_schedule_eh() will be calling into SCSI rescan, which would take scan_mutex ...
Not that it matter much, seeing that we've disconnected all devices, but still ...

UNLOADING is set and in that case, EH does nothing else than removing the devices. So I think this is OK.

Quoted 4 lines

Cheers,

Hannes

Signature · 2 lines
Damien Le Moal
Western Digital Research
Reviewed-by
#7 lore raw

Re: [PATCH 02/19] ata: libata-core: Fix port and device removal

Quoted 74 lines · Damien Le Moal
Quoted 64 lines · Hannes Reinecke
Quoted 63 lines · Damien Le Moal

Whenever an ATA adapter driver is removed (e.g. rmmod), ata_port_detach() is called repeatedly for all the adapter ports to remove (unload) the devices attached to the port and delete the port device itself. Removing of devices is done using libata EH with the ATA_PFLAG_UNLOADING port flag set. This causes libata EH to execute ata_eh_unload() which disables all devices attached to the port.

ata_port_detach() finishes by calling scsi_remove_host() to remove the scsi host associated with the port. This function will trigger the removal of all scsi devices attached to the host and in the case of disks, calls to sd_shutdown() which will flush the device write cache and stop the device. However, given that the devices were already disabled by ata_eh_unload(), the synchronize write cache command and start stop unit commands fail. E.g. running "rmmod ahci" with first removing sd_mod results in error messages like:

ata13.00: disable device
sd 0:0:0:0: [sda] Synchronizing SCSI cache
sd 0:0:0:0: [sda] Synchronize Cache(10) failed: Result:
hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK
sd 0:0:0:0: [sda] Stopping disk
sd 0:0:0:0: [sda] Start/Stop Unit failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK

Fix this by removing all scsi devices of the ata devices connected to the port before scheduling libata EH to disable the ATA devices.

Fixes 720ba12620ee ("[PATCH] libata-hp: update unload-unplug")
---
   drivers/ata/libata-core.c | 21 ++++++++++++++++++++-
   1 file changed, 20 insertions(+), 1 deletion(-)
drivers/ata/libata-core.c +18 -0
index c4898483d716..693cb3cd70cd 100644
@@ -5952,11 +5952,30 @@ static void ata_port_detach(struct ata_port *ap)
       struct ata_link *link;
       struct ata_device *dev;
   -    /* tell EH we're leaving & flush EH */
+    /* Wait for any ongoing EH */
+    ata_port_wait_eh(ap);
+
+    mutex_lock(&ap->scsi_scan_mutex);
       spin_lock_irqsave(ap->lock, flags);
+
+    /* Remove scsi devices */
+    ata_for_each_link(link, ap, HOST_FIRST) {
+        ata_for_each_dev(dev, link, ALL) {
+            if (dev->sdev) {
+                spin_unlock_irqrestore(ap->lock, flags);
+                scsi_remove_device(dev->sdev);
+                spin_lock_irqsave(ap->lock, flags);
+                dev->sdev = NULL;
+            }
+        }
+    }
+
+    /* Tell EH to disable all devices */
       ap->pflags |= ATA_PFLAG_UNLOADING;
       ata_port_schedule_eh(ap);
Quoted 8 lines

drivers/ata/libata-core.c static void ata_port_detach(struct ata_port *ap) · line 5979

Quoted 4 lines
drivers/ata/libata-core.c +2 -0
+
       spin_unlock_irqrestore(ap->lock, flags);
+    mutex_unlock(&ap->scsi_scan_mutex);
 

Are you sure about releasing scan_mutex after ata_port_schedule_eh()? Technically ata_port_schedule_eh() will be calling into SCSI rescan, which would take scan_mutex ...
Not that it matter much, seeing that we've disconnected all devices, but still ...

Quoted 3 lines

UNLOADING is set and in that case, EH does nothing else than removing the devices. So I think this is OK.

Yeah, thought so. Just wanted to mention it.

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
Tested-by on v1 2/19
#8 lore raw

Re: [PATCH 02/19] ata: libata-core: Fix port and device removal

In reply to Damien Le Moal
Quoted 22 lines · Damien Le Moal

Whenever an ATA adapter driver is removed (e.g. rmmod), ata_port_detach() is called repeatedly for all the adapter ports to remove (unload) the devices attached to the port and delete the port device itself. Removing of devices is done using libata EH with the ATA_PFLAG_UNLOADING port flag set. This causes libata EH to execute ata_eh_unload() which disables all devices attached to the port.

ata_port_detach() finishes by calling scsi_remove_host() to remove the scsi host associated with the port. This function will trigger the removal of all scsi devices attached to the host and in the case of disks, calls to sd_shutdown() which will flush the device write cache and stop the device. However, given that the devices were already disabled by ata_eh_unload(), the synchronize write cache command and start stop unit commands fail. E.g. running "rmmod ahci" with first removing sd_mod results in error messages like:

ata13.00: disable device
sd 0:0:0:0: [sda] Synchronizing SCSI cache
sd 0:0:0:0: [sda] Synchronize Cache(10) failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK sd 0:0:0:0: [sda] Stopping disk
sd 0:0:0:0: [sda] Start/Stop Unit failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK

Quoted 6 lines

Fix this by removing all scsi devices of the ata devices connected to the port before scheduling libata EH to disable the ATA devices.

Fixes 720ba12620ee ("[PATCH] libata-hp: update unload-unplug")
author maintainer
#9 lore raw

PATCH 3/19 ata: libata-scsi: link ata port and scsi device

In reply to Damien Le Moal

There is no direct device ancestry defined between an ata_device and its scsi device which prevents the power management code from correctly ordering suspend and resume operations. Create such ancestry with the ata device as the parent to ensure that the scsi device (child) is suspended before the ata device and that resume handles the ata device before the scsi device.

The parent-child (supplier-consumer) relationship is established between the ata_port (parent) and the scsi device (child) with the function device_add_link(). The parent used is not the ata_device as the PM operations are defined per port and the status of all devices connected through that port is controlled from the port operations.

The device link is established with the new function ata_scsi_dev_alloc(). This function is used to define the ->slave_alloc callback of the scsi host template of most drivers.

Fixes a19a93e4c6a9 ("scsi: core: pm: Rely on the device driver core for async power management")
---
 drivers/ata/libata-scsi.c | 46 ++++++++++++++++++++++++++++++++++-----
 drivers/ata/libata.h      |  1 +
 drivers/ata/pata_macio.c  |  1 +
 drivers/ata/sata_mv.c     |  1 +
 drivers/ata/sata_nv.c     |  2 ++
 drivers/ata/sata_sil24.c  |  1 +
 include/linux/libata.h    |  3 +++
 7 files changed, 50 insertions(+), 5 deletions(-)
drivers/ata/libata-scsi.c +41 -5
index d3f28b82c97b..f63cf6e7332e 100644
@@ -1089,6 +1089,45 @@ int ata_scsi_dev_config(struct scsi_device *sdev, struct ata_device *dev)
 	return 0;
 }
 
+int ata_scsi_dev_alloc(struct scsi_device *sdev, struct ata_port *ap)
+{
+	struct device_link *link;
+
+	ata_scsi_sdev_config(sdev);
+
+	/*
+	 * Create a link from the ata_port device to the scsi device to ensure
+	 * that PM does suspend/resume in the correct order: the scsi device is
+	 * consumer (child) and the ata port the supplier (parent).
+	 */
+	link = device_link_add(&sdev->sdev_gendev, &ap->tdev,
+			       DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE);
+	if (!link) {
+		ata_port_err(ap, "Failed to create link to scsi device %s\n",
+			     dev_name(&sdev->sdev_gendev));
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+/**
+ *	ata_scsi_slave_alloc - Early setup of SCSI device
+ *	@sdev: SCSI device to examine
+ *
+ *	This is called from scsi_alloc_sdev() when the scsi device
+ *	associated with an ATA device is scanned on a port.
+ *
+ *	LOCKING:
+ *	Defined by SCSI layer.  We don't really care.
+ */
+
+int ata_scsi_slave_alloc(struct scsi_device *sdev)
+{
+	return ata_scsi_dev_alloc(sdev, ata_shost_to_port(sdev->host));
+}
+EXPORT_SYMBOL_GPL(ata_scsi_slave_alloc);
+
 /**
  *	ata_scsi_slave_config - Set SCSI device attributes
  *	@sdev: SCSI device to examine
@@ -1105,14 +1144,11 @@ int ata_scsi_slave_config(struct scsi_device *sdev)
 {
 	struct ata_port *ap = ata_shost_to_port(sdev->host);
 	struct ata_device *dev = __ata_scsi_find_dev(ap, sdev);
-	int rc = 0;
-
-	ata_scsi_sdev_config(sdev);
 
 	if (dev)
-		rc = ata_scsi_dev_config(sdev, dev);
+		return ata_scsi_dev_config(sdev, dev);
 
-	return rc;
+	return 0;
 }
 EXPORT_SYMBOL_GPL(ata_scsi_slave_config);
 
drivers/ata/libata.h +1 -0
index 6e7d352803bd..079981e7156a 100644
@@ -111,6 +111,7 @@ extern struct ata_device *ata_scsi_find_dev(struct ata_port *ap,
 extern int ata_scsi_add_hosts(struct ata_host *host,
 			      const struct scsi_host_template *sht);
 extern void ata_scsi_scan_host(struct ata_port *ap, int sync);
+extern int ata_scsi_dev_alloc(struct scsi_device *sdev, struct ata_port *ap);
 extern int ata_scsi_offline_dev(struct ata_device *dev);
 extern bool ata_scsi_sense_is_valid(u8 sk, u8 asc, u8 ascq);
 extern void ata_scsi_set_sense(struct ata_device *dev,
drivers/ata/pata_macio.c +1 -0
index 17f6ccee53c7..32968b4cf8e4 100644
@@ -918,6 +918,7 @@ static const struct scsi_host_template pata_macio_sht = {
 	 * use 64K minus 256
 	 */
 	.max_segment_size	= MAX_DBDMA_SEG,
+	.slave_alloc		= ata_scsi_slave_alloc,
 	.slave_configure	= pata_macio_slave_config,
 	.sdev_groups		= ata_common_sdev_groups,
 	.can_queue		= ATA_DEF_QUEUE,
drivers/ata/sata_mv.c +1 -0
index d105db5c7d81..353ac7b2f14a 100644
@@ -673,6 +673,7 @@ static const struct scsi_host_template mv6_sht = {
 	.sdev_groups		= ata_ncq_sdev_groups,
 	.change_queue_depth	= ata_scsi_change_queue_depth,
 	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,
+	.slave_alloc		= ata_scsi_slave_alloc,
 	.slave_configure	= ata_scsi_slave_config
 };
 
drivers/ata/sata_nv.c +2 -0
index 0a0cee755bde..5428dc2ec5e3 100644
@@ -380,6 +380,7 @@ static const struct scsi_host_template nv_adma_sht = {
 	.can_queue		= NV_ADMA_MAX_CPBS,
 	.sg_tablesize		= NV_ADMA_SGTBL_TOTAL_LEN,
 	.dma_boundary		= NV_ADMA_DMA_BOUNDARY,
+	.slave_alloc		= ata_scsi_slave_alloc,
 	.slave_configure	= nv_adma_slave_config,
 	.sdev_groups		= ata_ncq_sdev_groups,
 	.change_queue_depth     = ata_scsi_change_queue_depth,
@@ -391,6 +392,7 @@ static const struct scsi_host_template nv_swncq_sht = {
 	.can_queue		= ATA_MAX_QUEUE - 1,
 	.sg_tablesize		= LIBATA_MAX_PRD,
 	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_alloc		= ata_scsi_slave_alloc,
 	.slave_configure	= nv_swncq_slave_config,
 	.sdev_groups		= ata_ncq_sdev_groups,
 	.change_queue_depth     = ata_scsi_change_queue_depth,
drivers/ata/sata_sil24.c +1 -0
index 142e70bfc498..e0b1b3625031 100644
@@ -381,6 +381,7 @@ static const struct scsi_host_template sil24_sht = {
 	.tag_alloc_policy	= BLK_TAG_ALLOC_FIFO,
 	.sdev_groups		= ata_ncq_sdev_groups,
 	.change_queue_depth	= ata_scsi_change_queue_depth,
+	.slave_alloc		= ata_scsi_slave_alloc,
 	.slave_configure	= ata_scsi_slave_config
 };
 
include/linux/libata.h +3 -0
index 52d58b13e5ee..c8cfea386c16 100644
@@ -1144,6 +1144,7 @@ extern int ata_std_bios_param(struct scsi_device *sdev,
 			      struct block_device *bdev,
 			      sector_t capacity, int geom[]);
 extern void ata_scsi_unlock_native_capacity(struct scsi_device *sdev);
+extern int ata_scsi_slave_alloc(struct scsi_device *sdev);
 extern int ata_scsi_slave_config(struct scsi_device *sdev);
 extern void ata_scsi_slave_destroy(struct scsi_device *sdev);
 extern int ata_scsi_change_queue_depth(struct scsi_device *sdev,
@@ -1401,12 +1402,14 @@ extern const struct attribute_group *ata_common_sdev_groups[];
 	__ATA_BASE_SHT(drv_name),				\
 	.can_queue		= ATA_DEF_QUEUE,		\
 	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,		\
+	.slave_alloc		= ata_scsi_slave_alloc,		\
 	.slave_configure	= ata_scsi_slave_config
 
 #define ATA_SUBBASE_SHT_QD(drv_name, drv_qd)			\
 	__ATA_BASE_SHT(drv_name),				\
 	.can_queue		= drv_qd,			\
 	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,		\
+	.slave_alloc		= ata_scsi_slave_alloc,		\
 	.slave_configure	= ata_scsi_slave_config
 
 #define ATA_BASE_SHT(drv_name)					\
Signature · 1 line
2.41.0
#10 lore raw

Re: [PATCH 03/19] ata: libata-scsi: link ata port and scsi device

Quoted 182 lines · Damien Le Moal

There is no direct device ancestry defined between an ata_device and its scsi device which prevents the power management code from correctly ordering suspend and resume operations. Create such ancestry with the ata device as the parent to ensure that the scsi device (child) is suspended before the ata device and that resume handles the ata device before the scsi device.

The parent-child (supplier-consumer) relationship is established between the ata_port (parent) and the scsi device (child) with the function device_add_link(). The parent used is not the ata_device as the PM operations are defined per port and the status of all devices connected through that port is controlled from the port operations.

The device link is established with the new function ata_scsi_dev_alloc(). This function is used to define the ->slave_alloc callback of the scsi host template of most drivers.

Fixes a19a93e4c6a9 ("scsi: core: pm: Rely on the device driver core for async power management")
---
  drivers/ata/libata-scsi.c | 46 ++++++++++++++++++++++++++++++++++-----
  drivers/ata/libata.h      |  1 +
  drivers/ata/pata_macio.c  |  1 +
  drivers/ata/sata_mv.c     |  1 +
  drivers/ata/sata_nv.c     |  2 ++
  drivers/ata/sata_sil24.c  |  1 +
  include/linux/libata.h    |  3 +++
  7 files changed, 50 insertions(+), 5 deletions(-)
drivers/ata/libata-scsi.c +41 -5
index d3f28b82c97b..f63cf6e7332e 100644
@@ -1089,6 +1089,45 @@ int ata_scsi_dev_config(struct scsi_device *sdev, struct ata_device *dev)
  	return 0;
  }
 
+int ata_scsi_dev_alloc(struct scsi_device *sdev, struct ata_port *ap)
+{
+	struct device_link *link;
+
+	ata_scsi_sdev_config(sdev);
+
+	/*
+	 * Create a link from the ata_port device to the scsi device to ensure
+	 * that PM does suspend/resume in the correct order: the scsi device is
+	 * consumer (child) and the ata port the supplier (parent).
+	 */
+	link = device_link_add(&sdev->sdev_gendev, &ap->tdev,
+			       DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE);
+	if (!link) {
+		ata_port_err(ap, "Failed to create link to scsi device %s\n",
+			     dev_name(&sdev->sdev_gendev));
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+/**
+ *	ata_scsi_slave_alloc - Early setup of SCSI device
+ *	@sdev: SCSI device to examine
+ *
+ *	This is called from scsi_alloc_sdev() when the scsi device
+ *	associated with an ATA device is scanned on a port.
+ *
+ *	LOCKING:
+ *	Defined by SCSI layer.  We don't really care.
+ */
+
+int ata_scsi_slave_alloc(struct scsi_device *sdev)
+{
+	return ata_scsi_dev_alloc(sdev, ata_shost_to_port(sdev->host));
+}
+EXPORT_SYMBOL_GPL(ata_scsi_slave_alloc);
+
  /**
   *	ata_scsi_slave_config - Set SCSI device attributes
   *	@sdev: SCSI device to examine
@@ -1105,14 +1144,11 @@ int ata_scsi_slave_config(struct scsi_device *sdev)
  {
  	struct ata_port *ap = ata_shost_to_port(sdev->host);
  	struct ata_device *dev = __ata_scsi_find_dev(ap, sdev);
-	int rc = 0;
-
-	ata_scsi_sdev_config(sdev);
 
  	if (dev)
-		rc = ata_scsi_dev_config(sdev, dev);
+		return ata_scsi_dev_config(sdev, dev);
 
-	return rc;
+	return 0;
  }
  EXPORT_SYMBOL_GPL(ata_scsi_slave_config);
 
drivers/ata/libata.h +1 -0
index 6e7d352803bd..079981e7156a 100644
@@ -111,6 +111,7 @@ extern struct ata_device *ata_scsi_find_dev(struct ata_port *ap,
  extern int ata_scsi_add_hosts(struct ata_host *host,
  			      const struct scsi_host_template *sht);
  extern void ata_scsi_scan_host(struct ata_port *ap, int sync);
+extern int ata_scsi_dev_alloc(struct scsi_device *sdev, struct ata_port *ap);
  extern int ata_scsi_offline_dev(struct ata_device *dev);
  extern bool ata_scsi_sense_is_valid(u8 sk, u8 asc, u8 ascq);
  extern void ata_scsi_set_sense(struct ata_device *dev,
drivers/ata/pata_macio.c +1 -0
index 17f6ccee53c7..32968b4cf8e4 100644
@@ -918,6 +918,7 @@ static const struct scsi_host_template pata_macio_sht = {
  	 * use 64K minus 256
  	 */
  	.max_segment_size	= MAX_DBDMA_SEG,
+	.slave_alloc		= ata_scsi_slave_alloc,
  	.slave_configure	= pata_macio_slave_config,
  	.sdev_groups		= ata_common_sdev_groups,
  	.can_queue		= ATA_DEF_QUEUE,
drivers/ata/sata_mv.c +1 -0
index d105db5c7d81..353ac7b2f14a 100644
@@ -673,6 +673,7 @@ static const struct scsi_host_template mv6_sht = {
  	.sdev_groups		= ata_ncq_sdev_groups,
  	.change_queue_depth	= ata_scsi_change_queue_depth,
  	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,
+	.slave_alloc		= ata_scsi_slave_alloc,
  	.slave_configure	= ata_scsi_slave_config
  };
 
drivers/ata/sata_nv.c +2 -0
index 0a0cee755bde..5428dc2ec5e3 100644
@@ -380,6 +380,7 @@ static const struct scsi_host_template nv_adma_sht = {
  	.can_queue		= NV_ADMA_MAX_CPBS,
  	.sg_tablesize		= NV_ADMA_SGTBL_TOTAL_LEN,
  	.dma_boundary		= NV_ADMA_DMA_BOUNDARY,
+	.slave_alloc		= ata_scsi_slave_alloc,
  	.slave_configure	= nv_adma_slave_config,
  	.sdev_groups		= ata_ncq_sdev_groups,
  	.change_queue_depth     = ata_scsi_change_queue_depth,
@@ -391,6 +392,7 @@ static const struct scsi_host_template nv_swncq_sht = {
  	.can_queue		= ATA_MAX_QUEUE - 1,
  	.sg_tablesize		= LIBATA_MAX_PRD,
  	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_alloc		= ata_scsi_slave_alloc,
  	.slave_configure	= nv_swncq_slave_config,
  	.sdev_groups		= ata_ncq_sdev_groups,
  	.change_queue_depth     = ata_scsi_change_queue_depth,
drivers/ata/sata_sil24.c +1 -0
index 142e70bfc498..e0b1b3625031 100644
@@ -381,6 +381,7 @@ static const struct scsi_host_template sil24_sht = {
  	.tag_alloc_policy	= BLK_TAG_ALLOC_FIFO,
  	.sdev_groups		= ata_ncq_sdev_groups,
  	.change_queue_depth	= ata_scsi_change_queue_depth,
+	.slave_alloc		= ata_scsi_slave_alloc,
  	.slave_configure	= ata_scsi_slave_config
  };
 
include/linux/libata.h +2 -0
index 52d58b13e5ee..c8cfea386c16 100644
@@ -1144,6 +1144,7 @@ extern int ata_std_bios_param(struct scsi_device *sdev,
  			      struct block_device *bdev,
  			      sector_t capacity, int geom[]);
  extern void ata_scsi_unlock_native_capacity(struct scsi_device *sdev);
+extern int ata_scsi_slave_alloc(struct scsi_device *sdev);
  extern int ata_scsi_slave_config(struct scsi_device *sdev);
  extern void ata_scsi_slave_destroy(struct scsi_device *sdev);
  extern int ata_scsi_change_queue_depth(struct scsi_device *sdev,
@@ -1401,12 +1402,14 @@ extern const struct attribute_group *ata_common_sdev_groups[];
  	__ATA_BASE_SHT(drv_name),				\
  	.can_queue		= ATA_DEF_QUEUE,		\
  	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,		\
+	.slave_alloc		= ata_scsi_slave_alloc,		\

include/linux/libata.h extern const struct attribute_group *ata_common_sdev_groups[]; · line 1415

Quoted 10 lines
include/linux/libata.h +1 -0
  	.slave_configure	= ata_scsi_slave_config
 
  #define ATA_SUBBASE_SHT_QD(drv_name, drv_qd)			\
  	__ATA_BASE_SHT(drv_name),				\
  	.can_queue		= drv_qd,			\
  	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,		\
+	.slave_alloc		= ata_scsi_slave_alloc,		\
  	.slave_configure	= ata_scsi_slave_config
 
  #define ATA_BASE_SHT(drv_name)					\

I do understand the rationale here, as the relationship between ata port and scsi devices are blurred. Question is: blurred by what? Is it the libata/SAS duality where SCSI devices will have a different
ancestry for libata and libsas?
If so, why don't we need the 'link' mechanism for SAS?
Or is it something else?

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
author maintainer
#11 lore raw

Re: [PATCH 03/19] ata: libata-scsi: link ata port and scsi device

Quoted 187 lines · Hannes Reinecke
Quoted 186 lines · Damien Le Moal

There is no direct device ancestry defined between an ata_device and its scsi device which prevents the power management code from correctly ordering suspend and resume operations. Create such ancestry with the ata device as the parent to ensure that the scsi device (child) is suspended before the ata device and that resume handles the ata device before the scsi device.

The parent-child (supplier-consumer) relationship is established between the ata_port (parent) and the scsi device (child) with the function device_add_link(). The parent used is not the ata_device as the PM operations are defined per port and the status of all devices connected through that port is controlled from the port operations.

The device link is established with the new function ata_scsi_dev_alloc(). This function is used to define the ->slave_alloc callback of the scsi host template of most drivers.

Fixes a19a93e4c6a9 ("scsi: core: pm: Rely on the device driver core for

async power management")

---
  drivers/ata/libata-scsi.c | 46 ++++++++++++++++++++++++++++++++++----- drivers/ata/libata.h      |  1 +
  drivers/ata/pata_macio.c  |  1 +
  drivers/ata/sata_mv.c     |  1 +
  drivers/ata/sata_nv.c     |  2 ++
  drivers/ata/sata_sil24.c  |  1 +
  include/linux/libata.h    |  3 +++

  7 files changed, 50 insertions(+), 5 deletions(-)
drivers/ata/libata-scsi.c +38 -0
index d3f28b82c97b..f63cf6e7332e 100644
@@ -1089,6 +1089,45 @@ int ata_scsi_dev_config(struct scsi_device *sdev,
 struct ata_device *dev)
       return 0;
   }
   +int ata_scsi_dev_alloc(struct scsi_device *sdev, struct ata_port *ap)
+{
+    struct device_link *link;
+
+    ata_scsi_sdev_config(sdev);
+
+    /*
+     * Create a link from the ata_port device to the scsi device to ensure
+     * that PM does suspend/resume in the correct order: the scsi device is
+     * consumer (child) and the ata port the supplier (parent).
+     */
+    link = device_link_add(&sdev->sdev_gendev, &ap->tdev,
+                   DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE);
+    if (!link) {
+        ata_port_err(ap, "Failed to create link to scsi device %s\n",
+                 dev_name(&sdev->sdev_gendev));
+        return -ENODEV;
+    }
+
+    return 0;
+}
+
+/**
+ *    ata_scsi_slave_alloc - Early setup of SCSI device
+ *    @sdev: SCSI device to examine
+ *
+ *    This is called from scsi_alloc_sdev() when the scsi device
+ *    associated with an ATA device is scanned on a port.
+ *
+ *    LOCKING:
+ *    Defined by SCSI layer.  We don't really care.
+ */
+
+int ata_scsi_slave_alloc(struct scsi_device *sdev)
+{
+    return ata_scsi_dev_alloc(sdev, ata_shost_to_port(sdev->host));
+}
+EXPORT_SYMBOL_GPL(ata_scsi_slave_alloc);
+
   /**
    *    ata_scsi_slave_config - Set SCSI device attributes

   *    @sdev: SCSI device to examine

drivers/ata/libata-scsi.c +2 -4
@@ -1105,14 +1144,11 @@ int ata_scsi_slave_config(struct scsi_device *sdev)
   {
       struct ata_port *ap = ata_shost_to_port(sdev->host);
       struct ata_device *dev = __ata_scsi_find_dev(ap, sdev);
-    int rc = 0;
-
-    ata_scsi_sdev_config(sdev);
         if (dev)
-        rc = ata_scsi_dev_config(sdev, dev);
+        return ata_scsi_dev_config(sdev, dev);
   -    return rc;
+    return 0;
   }
   EXPORT_SYMBOL_GPL(ata_scsi_slave_config);

  diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h index 6e7d352803bd..079981e7156a 100644

drivers/ata/libata.h +1 -0
@@ -111,6 +111,7 @@ extern struct ata_device *ata_scsi_find_dev(struct
 ata_port *ap,
   extern int ata_scsi_add_hosts(struct ata_host *host,
                     const struct scsi_host_template *sht);
   extern void ata_scsi_scan_host(struct ata_port *ap, int sync);
+extern int ata_scsi_dev_alloc(struct scsi_device *sdev, struct ata_port *ap);
   extern int ata_scsi_offline_dev(struct ata_device *dev);
   extern bool ata_scsi_sense_is_valid(u8 sk, u8 asc, u8 ascq);

  extern void ata_scsi_set_sense(struct ata_device *dev,

drivers/ata/pata_macio.c +1 -0
index 17f6ccee53c7..32968b4cf8e4 100644
@@ -918,6 +918,7 @@ static const struct scsi_host_template pata_macio_sht = {
        * use 64K minus 256
        */
       .max_segment_size    = MAX_DBDMA_SEG,
+    .slave_alloc        = ata_scsi_slave_alloc,
       .slave_configure    = pata_macio_slave_config,
       .sdev_groups        = ata_common_sdev_groups,
       .can_queue        = ATA_DEF_QUEUE,
drivers/ata/sata_mv.c +1 -0
index d105db5c7d81..353ac7b2f14a 100644
@@ -673,6 +673,7 @@ static const struct scsi_host_template mv6_sht = {
       .sdev_groups        = ata_ncq_sdev_groups,
       .change_queue_depth    = ata_scsi_change_queue_depth,
       .tag_alloc_policy    = BLK_TAG_ALLOC_RR,
+    .slave_alloc        = ata_scsi_slave_alloc,

      .slave_configure    = ata_scsi_slave_config };
  diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c index 0a0cee755bde..5428dc2ec5e3 100644

drivers/ata/sata_nv.c +2 -0
@@ -380,6 +380,7 @@ static const struct scsi_host_template nv_adma_sht = {
       .can_queue        = NV_ADMA_MAX_CPBS,
       .sg_tablesize        = NV_ADMA_SGTBL_TOTAL_LEN,
       .dma_boundary        = NV_ADMA_DMA_BOUNDARY,
+    .slave_alloc        = ata_scsi_slave_alloc,
       .slave_configure    = nv_adma_slave_config,
       .sdev_groups        = ata_ncq_sdev_groups,
       .change_queue_depth     = ata_scsi_change_queue_depth,
@@ -391,6 +392,7 @@ static const struct scsi_host_template nv_swncq_sht = {
       .can_queue        = ATA_MAX_QUEUE - 1,
       .sg_tablesize        = LIBATA_MAX_PRD,
       .dma_boundary        = ATA_DMA_BOUNDARY,
+    .slave_alloc        = ata_scsi_slave_alloc,
       .slave_configure    = nv_swncq_slave_config,
       .sdev_groups        = ata_ncq_sdev_groups,
       .change_queue_depth     = ata_scsi_change_queue_depth,
drivers/ata/sata_sil24.c +1 -0
index 142e70bfc498..e0b1b3625031 100644
@@ -381,6 +381,7 @@ static const struct scsi_host_template sil24_sht = {
       .tag_alloc_policy    = BLK_TAG_ALLOC_FIFO,
       .sdev_groups        = ata_ncq_sdev_groups,
       .change_queue_depth    = ata_scsi_change_queue_depth,
+    .slave_alloc        = ata_scsi_slave_alloc,

      .slave_configure    = ata_scsi_slave_config
  };
  diff --git a/include/linux/libata.h b/include/linux/libata.h index 52d58b13e5ee..c8cfea386c16 100644

include/linux/libata.h +2 -0
@@ -1144,6 +1144,7 @@ extern int ata_std_bios_param(struct scsi_device *sdev,
                     struct block_device *bdev,
                     sector_t capacity, int geom[]);
   extern void ata_scsi_unlock_native_capacity(struct scsi_device *sdev);
+extern int ata_scsi_slave_alloc(struct scsi_device *sdev);
   extern int ata_scsi_slave_config(struct scsi_device *sdev);
   extern void ata_scsi_slave_destroy(struct scsi_device *sdev);
   extern int ata_scsi_change_queue_depth(struct scsi_device *sdev,
@@ -1401,12 +1402,14 @@ extern const struct attribute_group
 *ata_common_sdev_groups[];
       __ATA_BASE_SHT(drv_name),                \
       .can_queue        = ATA_DEF_QUEUE,        \
       .tag_alloc_policy    = BLK_TAG_ALLOC_RR,        \
+    .slave_alloc        = ata_scsi_slave_alloc,        \

      .slave_configure    = ata_scsi_slave_config #define ATA_SUBBASE_SHT_QD(drv_name, drv_qd)            \
      __ATA_BASE_SHT(drv_name),                \
      .can_queue        = drv_qd,            \
      .tag_alloc_policy    = BLK_TAG_ALLOC_RR,        \
+    .slave_alloc        = ata_scsi_slave_alloc,        \

Quoted 8 lines
Quoted 2 lines

      .slave_configure    = ata_scsi_slave_config #define ATA_BASE_SHT(drv_name)                    \

I do understand the rationale here, as the relationship between ata port and scsi devices are blurred. Question is: blurred by what?
Is it the libata/SAS duality where SCSI devices will have a different ancestry for libata and libsas?
If so, why don't we need the 'link' mechanism for SAS?
Or is it something else?

On scsi, ancestry from Scsi_Host is clear: host->target->device->disk. For ATA, it is also clear: port->link->device

The relationship is that ata port has its own Scsi_Host, but no "family" relationship here. They are just "friends", so scsi disk and ata_port have no direct ancestry. Adding the link creates that.

libsas does not need the link because it does its own PM management of the ports, not relying on PM to order things.

Quoted 4 lines

Cheers,

Hannes

Signature · 2 lines
Damien Le Moal
Western Digital Research
Reviewed-by
#12 lore raw

Re: [PATCH 03/19] ata: libata-scsi: link ata port and scsi device

Quoted 200 lines · Damien Le Moal
Quoted 187 lines · Hannes Reinecke
Quoted 186 lines · Damien Le Moal

There is no direct device ancestry defined between an ata_device and its scsi device which prevents the power management code from correctly ordering suspend and resume operations. Create such ancestry with the ata device as the parent to ensure that the scsi device (child) is suspended before the ata device and that resume handles the ata device before the scsi device.

The parent-child (supplier-consumer) relationship is established between the ata_port (parent) and the scsi device (child) with the function device_add_link(). The parent used is not the ata_device as the PM operations are defined per port and the status of all devices connected through that port is controlled from the port operations.

The device link is established with the new function ata_scsi_dev_alloc(). This function is used to define the ->slave_alloc callback of the scsi host template of most drivers.

Fixes a19a93e4c6a9 ("scsi: core: pm: Rely on the device driver core for

async power management")

---
   drivers/ata/libata-scsi.c | 46 ++++++++++++++++++++++++++++++++++-----
   drivers/ata/libata.h      |  1 +
   drivers/ata/pata_macio.c  |  1 +
   drivers/ata/sata_mv.c     |  1 +
   drivers/ata/sata_nv.c     |  2 ++
   drivers/ata/sata_sil24.c  |  1 +
   include/linux/libata.h    |  3 +++
   7 files changed, 50 insertions(+), 5 deletions(-)
drivers/ata/libata-scsi.c +40 -4
index d3f28b82c97b..f63cf6e7332e 100644
@@ -1089,6 +1089,45 @@ int ata_scsi_dev_config(struct scsi_device *sdev,
 struct ata_device *dev)
       return 0;
   }
   +int ata_scsi_dev_alloc(struct scsi_device *sdev, struct ata_port *ap)
+{
+    struct device_link *link;
+
+    ata_scsi_sdev_config(sdev);
+
+    /*
+     * Create a link from the ata_port device to the scsi device to ensure
+     * that PM does suspend/resume in the correct order: the scsi device is
+     * consumer (child) and the ata port the supplier (parent).
+     */
+    link = device_link_add(&sdev->sdev_gendev, &ap->tdev,
+                   DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE);
+    if (!link) {
+        ata_port_err(ap, "Failed to create link to scsi device %s\n",
+                 dev_name(&sdev->sdev_gendev));
+        return -ENODEV;
+    }
+
+    return 0;
+}
+
+/**
+ *    ata_scsi_slave_alloc - Early setup of SCSI device
+ *    @sdev: SCSI device to examine
+ *
+ *    This is called from scsi_alloc_sdev() when the scsi device
+ *    associated with an ATA device is scanned on a port.
+ *
+ *    LOCKING:
+ *    Defined by SCSI layer.  We don't really care.
+ */
+
+int ata_scsi_slave_alloc(struct scsi_device *sdev)
+{
+    return ata_scsi_dev_alloc(sdev, ata_shost_to_port(sdev->host));
+}
+EXPORT_SYMBOL_GPL(ata_scsi_slave_alloc);
+
   /**
    *    ata_scsi_slave_config - Set SCSI device attributes
    *    @sdev: SCSI device to examine
@@ -1105,14 +1144,11 @@ int ata_scsi_slave_config(struct scsi_device *sdev)
   {
       struct ata_port *ap = ata_shost_to_port(sdev->host);
       struct ata_device *dev = __ata_scsi_find_dev(ap, sdev);
-    int rc = 0;
-
-    ata_scsi_sdev_config(sdev);
         if (dev)
-        rc = ata_scsi_dev_config(sdev, dev);
+        return ata_scsi_dev_config(sdev, dev);
   -    return rc;
+    return 0;
   }
   EXPORT_SYMBOL_GPL(ata_scsi_slave_config);
   diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h

index 6e7d352803bd..079981e7156a 100644

drivers/ata/libata.h +1 -0
@@ -111,6 +111,7 @@ extern struct ata_device *ata_scsi_find_dev(struct
 ata_port *ap,
   extern int ata_scsi_add_hosts(struct ata_host *host,
                     const struct scsi_host_template *sht);
   extern void ata_scsi_scan_host(struct ata_port *ap, int sync);
+extern int ata_scsi_dev_alloc(struct scsi_device *sdev, struct ata_port *ap);
   extern int ata_scsi_offline_dev(struct ata_device *dev);
   extern bool ata_scsi_sense_is_valid(u8 sk, u8 asc, u8 ascq);

  extern void ata_scsi_set_sense(struct ata_device *dev,

drivers/ata/pata_macio.c +1 -0
index 17f6ccee53c7..32968b4cf8e4 100644
@@ -918,6 +918,7 @@ static const struct scsi_host_template pata_macio_sht = {
        * use 64K minus 256
        */
       .max_segment_size    = MAX_DBDMA_SEG,
+    .slave_alloc        = ata_scsi_slave_alloc,
       .slave_configure    = pata_macio_slave_config,
       .sdev_groups        = ata_common_sdev_groups,
       .can_queue        = ATA_DEF_QUEUE,
drivers/ata/sata_mv.c +1 -0
index d105db5c7d81..353ac7b2f14a 100644
@@ -673,6 +673,7 @@ static const struct scsi_host_template mv6_sht = {
       .sdev_groups        = ata_ncq_sdev_groups,
       .change_queue_depth    = ata_scsi_change_queue_depth,
       .tag_alloc_policy    = BLK_TAG_ALLOC_RR,
+    .slave_alloc        = ata_scsi_slave_alloc,
       .slave_configure    = ata_scsi_slave_config
   };
   diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c

index 0a0cee755bde..5428dc2ec5e3 100644

drivers/ata/sata_nv.c +2 -0
@@ -380,6 +380,7 @@ static const struct scsi_host_template nv_adma_sht = {
       .can_queue        = NV_ADMA_MAX_CPBS,
       .sg_tablesize        = NV_ADMA_SGTBL_TOTAL_LEN,
       .dma_boundary        = NV_ADMA_DMA_BOUNDARY,
+    .slave_alloc        = ata_scsi_slave_alloc,
       .slave_configure    = nv_adma_slave_config,
       .sdev_groups        = ata_ncq_sdev_groups,
       .change_queue_depth     = ata_scsi_change_queue_depth,
@@ -391,6 +392,7 @@ static const struct scsi_host_template nv_swncq_sht = {
       .can_queue        = ATA_MAX_QUEUE - 1,
       .sg_tablesize        = LIBATA_MAX_PRD,
       .dma_boundary        = ATA_DMA_BOUNDARY,
+    .slave_alloc        = ata_scsi_slave_alloc,
       .slave_configure    = nv_swncq_slave_config,
       .sdev_groups        = ata_ncq_sdev_groups,
       .change_queue_depth     = ata_scsi_change_queue_depth,
drivers/ata/sata_sil24.c +1 -0
index 142e70bfc498..e0b1b3625031 100644
@@ -381,6 +381,7 @@ static const struct scsi_host_template sil24_sht = {
       .tag_alloc_policy    = BLK_TAG_ALLOC_FIFO,
       .sdev_groups        = ata_ncq_sdev_groups,
       .change_queue_depth    = ata_scsi_change_queue_depth,
+    .slave_alloc        = ata_scsi_slave_alloc,
       .slave_configure    = ata_scsi_slave_config
   };
   diff --git a/include/linux/libata.h b/include/linux/libata.h

index 52d58b13e5ee..c8cfea386c16 100644

include/linux/libata.h +3 -0
@@ -1144,6 +1144,7 @@ extern int ata_std_bios_param(struct scsi_device *sdev,
                     struct block_device *bdev,
                     sector_t capacity, int geom[]);
   extern void ata_scsi_unlock_native_capacity(struct scsi_device *sdev);
+extern int ata_scsi_slave_alloc(struct scsi_device *sdev);
   extern int ata_scsi_slave_config(struct scsi_device *sdev);
   extern void ata_scsi_slave_destroy(struct scsi_device *sdev);
   extern int ata_scsi_change_queue_depth(struct scsi_device *sdev,
@@ -1401,12 +1402,14 @@ extern const struct attribute_group
 *ata_common_sdev_groups[];
       __ATA_BASE_SHT(drv_name),                \
       .can_queue        = ATA_DEF_QUEUE,        \
       .tag_alloc_policy    = BLK_TAG_ALLOC_RR,        \
+    .slave_alloc        = ata_scsi_slave_alloc,        \
       .slave_configure    = ata_scsi_slave_config
     #define ATA_SUBBASE_SHT_QD(drv_name, drv_qd)            \
       __ATA_BASE_SHT(drv_name),                \
       .can_queue        = drv_qd,            \
       .tag_alloc_policy    = BLK_TAG_ALLOC_RR,        \
+    .slave_alloc        = ata_scsi_slave_alloc,        \
Quoted 8 lines

include/linux/libata.h extern const struct attribute_group · line 1414

Quoted 2 lines
include/linux/libata.h +0 -0
       .slave_configure    = ata_scsi_slave_config
     #define ATA_BASE_SHT(drv_name)                    \

I do understand the rationale here, as the relationship between ata port and scsi devices are blurred. Question is: blurred by what?
Is it the libata/SAS duality where SCSI devices will have a different ancestry for libata and libsas?
If so, why don't we need the 'link' mechanism for SAS?
Or is it something else?

On scsi, ancestry from Scsi_Host is clear: host->target->device->disk. For ATA, it is also clear: port->link->device

Quoted 7 lines

The relationship is that ata port has its own Scsi_Host, but no "family" relationship here. They are just "friends", so scsi disk and ata_port have no direct ancestry. Adding the link creates that.

libsas does not need the link because it does its own PM management of the ports, not relying on PM to order things.

Okay.

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
reviewer
#13 lore raw

Re: [PATCH 03/19] ata: libata-scsi: link ata port and scsi device

In reply to Damien Le Moal
Quoted 10 lines · Damien Le Moal
Quoted 7 lines
Quoted 1 line
  #define ATA_BASE_SHT(drv_name)                    \

I do understand the rationale here, as the relationship between ata port and scsi devices are blurred. Question is: blurred by what?
Is it the libata/SAS duality where SCSI devices will have a different ancestry for libata and libsas?
If so, why don't we need the 'link' mechanism for SAS?
Or is it something else?

On scsi, ancestry from Scsi_Host is clear: host->target->device->disk. For ATA, it is also clear: port->link->device

Quoted 7 lines

The relationship is that ata port has its own Scsi_Host, but no "family" relationship here. They are just "friends", so scsi disk and ata_port have no direct ancestry. Adding the link creates that.

libsas does not need the link because it does its own PM management of the ports, not relying on PM to order things.

For hisi_sas v3, RPM is supported and a link was required to be added in that driver between the sdev and those host controller device - see 16fd4a7c59. And that is for a similar reason here. I see that we already have an ancestry for libata between ata_dev -> ata_link -> ata_port -> ata_host dev, so it seems reasonable to add ata_dev <-> sdev dependency here.

I think that this should really be added for all libsas drivers which support RPM - pm8001 is missing, IIRC.

Thanks,
John

author maintainer
#14 lore raw

Re: [PATCH 03/19] ata: libata-scsi: link ata port and scsi device

Quoted 19 lines · John Garry
Quoted 10 lines · Damien Le Moal
Quoted 7 lines
Quoted 1 line
  #define ATA_BASE_SHT(drv_name)                    \

I do understand the rationale here, as the relationship between ata port and scsi devices are blurred. Question is: blurred by what?
Is it the libata/SAS duality where SCSI devices will have a different ancestry for libata and libsas?
If so, why don't we need the 'link' mechanism for SAS?
Or is it something else?

On scsi, ancestry from Scsi_Host is clear: host->target->device->disk. For ATA, it is also clear: port->link->device

Quoted 7 lines

The relationship is that ata port has its own Scsi_Host, but no "family" relationship here. They are just "friends", so scsi disk and ata_port have no direct ancestry. Adding the link creates that.

libsas does not need the link because it does its own PM management of the ports, not relying on PM to order things.

Quoted 6 lines

For hisi_sas v3, RPM is supported and a link was required to be added in that driver between the sdev and those host controller device - see 16fd4a7c59. And that is for a similar reason here. I see that we already have an ancestry for libata between ata_dev -> ata_link -> ata_port -> ata_host dev, so it seems reasonable to add ata_dev <-> sdev dependency here.

libata creates a link between ata_port and sdev. This is not very natural, but as I said in the cover letter, the power management model is weird as everything is per port, even if a port has multiple devices. Nevertheless, I tried to apply the same for libsas but failed: if I add the link creation in the slave_alloc method, I fail to get the scsi disks to show up (no /dev/sdX device files). Not sure why. That was with my pm8001 adapter, which is the only libsas one I have.

Side note: having an ata_debug (ata equivalent of scsi_debug) driver that could act as a pure libata driver or as a libsas adapter would really be awesome for this kind of tests :)

Quoted 2 lines

I think that this should really be added for all libsas drivers which support RPM - pm8001 is missing, IIRC.

Will try again tomorrow looking at hisi driver for inspiration. The other thing I need to better understand is how the suspend & resume operations of libsas get triggered. For suspend, it is easy-ish, but for resume, it seems to be tightly coupled with discovery, so the the adapter resume (scsi host class resume ?). If that is the case, then that is done *before* the scsi_dev resume because of the existing device link dependencies. So I am not yet entirely convinced that adding a link between ata_port and sdev will serve any purpose now that the asynchronous libata resume is fixed and synchronized with the scsi disk resume... Will dig further.

That said, it may be good to move libsas suspend/resume fixes to another patch series. There is an outstanding regression/problem for pure libata devices that this series fixes. So would like to get the fixes patches in Linus tree ASAP.

Cheers.

Signature · 2 lines
Damien Le Moal
Western Digital Research
reviewer
#15 lore raw

Re: [PATCH 03/19] ata: libata-scsi: link ata port and scsi device

Quoted 13 lines · Damien Le Moal
Quoted 6 lines

For hisi_sas v3, RPM is supported and a link was required to be added in that driver between the sdev and those host controller device - see 16fd4a7c59. And that is for a similar reason here. I see that we already have an ancestry for libata between ata_dev -> ata_link -> ata_port -> ata_host dev, so it seems reasonable to add ata_dev <-> sdev dependency here.

libata creates a link between ata_port and sdev. This is not very natural, but as I said in the cover letter, the power management model is weird as everything is per port, even if a port has multiple devices. Nevertheless, I tried to apply the same for libsas but failed: if I add the link creation in the slave_alloc method, I fail to get the scsi disks to show up (no /dev/sdX device files). Not sure why. That was with my pm8001 adapter, which is the only libsas one I have.

Quoted 3 lines

Side note: having an ata_debug (ata equivalent of scsi_debug) driver that could act as a pure libata driver or as a libsas adapter would really be awesome for this kind of tests 😄

hmm... maybe scsi_debug could be modified to support ATA devices (with libata).

Regardless, I'm happy to check the code change which you attempted if shared.

Quoted 13 lines
Quoted 3 lines

I think that this should really be added for all libsas drivers which support RPM - pm8001 is missing, IIRC.

Will try again tomorrow looking at hisi driver for inspiration. The other thing I need to better understand is how the suspend & resume operations of libsas get triggered. For suspend, it is easy-ish, but for resume, it seems to be tightly coupled with discovery, so the the adapter resume (scsi host class resume ?). If that is the case, then that is done*before* the scsi_dev resume because of the existing device link dependencies. So I am not yet entirely convinced that adding a link between ata_port and sdev will serve any purpose now that the asynchronous libata resume is fixed and synchronized with the scsi disk resume... Will dig further.

Quoted 3 lines

That said, it may be good to move libsas suspend/resume fixes to another patch series. There is an outstanding regression/problem for pure libata devices that this series fixes. So would like to get the fixes patches in Linus tree ASAP.

Sure, and I do realize that libsas is outside the scope of this series.

Cheers,
John

author maintainer
#16 lore raw

Re: [PATCH 03/19] ata: libata-scsi: link ata port and scsi device

Quoted 18 lines · John Garry
Quoted 13 lines · Damien Le Moal
Quoted 6 lines

For hisi_sas v3, RPM is supported and a link was required to be added in that driver between the sdev and those host controller device - see 16fd4a7c59. And that is for a similar reason here. I see that we already have an ancestry for libata between ata_dev -> ata_link -> ata_port -> ata_host dev, so it seems reasonable to add ata_dev <-> sdev dependency here.

libata creates a link between ata_port and sdev. This is not very natural, but as I said in the cover letter, the power management model is weird as everything is per port, even if a port has multiple devices. Nevertheless, I tried to apply the same for libsas but failed: if I add the link creation in the slave_alloc method, I fail to get the scsi disks to show up (no /dev/sdX device files). Not sure why. That was with my pm8001 adapter, which is the only libsas one I have.

Quoted 3 lines

Side note: having an ata_debug (ata equivalent of scsi_debug) driver that could act as a pure libata driver or as a libsas adapter would really be awesome for this kind of tests 😄

Quoted 5 lines

hmm... maybe scsi_debug could be modified to support ATA devices (with libata).

Regardless, I'm happy to check the code change which you attempted if shared.

I had a closer look at what the hisi_sas driver is doing. The link it creates with device_add_link() is between the scsi device gendev and the hisi_hba->dev, that is, the HBA device. So nothing to do with libata port of device.

Trying to mimic what libata is now doing, I tried this:

drivers/scsi/libsas/sas_ata.c +14 -0
index 12e2653846e3..91d76258e6ea 100644
@@ -624,6 +624,26 @@ int sas_ata_init(struct domain_device *found_dev)
        return rc;
 }
 
+int sas_ata_slave_configure(struct scsi_device *sdev)
+{
+       struct domain_device *dev = sdev_to_domain_dev(sdev);
+       struct ata_port *ap = dev->sata_dev.ap;
+       struct device *sdev_dev = &sdev->sdev_gendev;
+       struct device_link *link;
+
+       ata_sas_slave_configure(sdev, ap);
+
+       link = device_link_add(sdev_dev, &ap->tdev,
+                              DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE);
+       if (!link && pm_runtime_enabled(sdev_dev)) {
+               dev_err(sdev_dev,
+                       "add device link failed, disabling runtime PM for the

host\n");
+ pm_runtime_disable(sdev_dev);
+ }
+
+ return 0;
+}
+
void sas_ata_task_abort(struct sas_task *task)

 {
        struct ata_queued_cmd *qc = task->uldd_task;
drivers/scsi/libsas/sas_internal.h +2 -0
index a6dc7dc07fce..82c6d8e1e8c7 100644
@@ -58,6 +58,8 @@ 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);
 
+int sas_ata_slave_configure(struct scsi_device *sdev);
+
 void sas_deform_port(struct asd_sas_phy *phy, int gone);
 
 void sas_porte_bytes_dmaed(struct work_struct *work);
(quoted diff) +0 -0

b/drivers/scsi/libsas/sas_scsi_host.c
index 9047cfcd1072..b6c419aa563e 100644

drivers/scsi/libsas/sas_scsi_host.c +2 -4
@@ -810,10 +810,8 @@ int sas_slave_configure(struct scsi_device *scsi_dev)
 
        BUG_ON(dev->rphy->identify.device_type != SAS_END_DEVICE);
 
-       if (dev_is_sata(dev)) {
-               ata_sas_slave_configure(scsi_dev, dev->sata_dev.ap);
-               return 0;
-       }
+       if (dev_is_sata(dev))
+               return sas_ata_slave_configure(scsi_dev);
 
        sas_read_port_mode_page(scsi_dev);
 

This does not change the drive discovery, all drives connected to the HBA are found and identified:

[   52.974154] scsi host12: pm80xx
[   54.445986] sas: phy-12:4 added to port-12:0, phy_mask:0x10 (50010860002f5644)
[   54.449019] sas: DOING DISCOVERY on port 0, pid:423
[   54.461108] sas: Enter sas_scsi_recover_host busy: 0 failed: 0
[   54.465601] sas: ata13: end_device-12:0: dev error handler
[   54.618588] ata13.00: ATA-11: WDC  WUH721818ALN604, PCGNW232, max UDMA/133
[   56.295486] ata13.00: 4394582016 sectors, multi 0: LBA48 NCQ (depth 32)
[   56.306849] ata13.00: Features: NCQ-sndrcv NCQ-prio
[   56.320849] ata13.00: configured for UDMA/133
[   56.324504] sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 0 tries: 1
[   56.340346] scsi 12:0:0:0: Direct-Access     ATA      WDC  WUH721818AL W232
PQ: 0 ANSI: 5
[   56.347160] sas: DONE DISCOVERY on port 0, pid:423, result:0
[   56.348161] sas: phy-12:5 added to port-12:1, phy_mask:0x20 (50010860002f5645)
[   56.351291] sas: DOING DISCOVERY on port 1, pid:423
[   56.363285] sas: Enter sas_scsi_recover_host busy: 0 failed: 0
[   56.368788] sas: ata13: end_device-12:0: dev error handler
[   56.368859] sas: ata14: end_device-12:1: dev error handler
[   56.522903] ata14.00: ATA-11: WDC  WUH721818ALN604, PCGNWTW2, max UDMA/133
[   58.402015] ata14.00: 4394582016 sectors, multi 0: LBA48 NCQ (depth 32)
[   58.406207] ata14.00: Features: NCQ-sndrcv NCQ-prio
[   58.420254] ata14.00: configured for UDMA/133
[   58.423222] sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 0 tries: 1
[   58.438869] scsi 12:0:1:0: Direct-Access     ATA      WDC  WUH721818AL WTW2
PQ: 0 ANSI: 5
[   58.445619] sas: DONE DISCOVERY on port 1, pid:423, result:0
[   58.446687] sas: phy-12:6 added to port-12:2, phy_mask:0x40 (50010860002f5646)
[   58.449803] sas: DOING DISCOVERY on port 2, pid:423
[   58.461732] sas: Enter sas_scsi_recover_host busy: 0 failed: 0
[   58.465369] sas: ata13: end_device-12:0: dev error handler
[   58.465426] sas: ata14: end_device-12:1: dev error handler
[   58.465432] sas: ata15: end_device-12:2: dev error handler
[   58.622345] ata15.00: ATA-12: WDC  WUH722222ALN604, LNGNW1TZ, max UDMA/133
[   59.540468] ata15.00: 5371330560 sectors, multi 0: LBA48 NCQ (depth 32)
[   59.588217] ata15.00: Features: NCQ-sndrcv NCQ-prio CDL
[   59.697588] ata15.00: configured for UDMA/133
[   59.700514] sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 0 tries: 1
[   59.716067] scsi 12:0:2:0: Direct-Access     ATA      WDC  WUH722222AL W1TZ
PQ: 0 ANSI: 5
[   59.723838] sas: DONE DISCOVERY on port 2, pid:423, result:0
[   59.724893] sas: phy-12:7 added to port-12:3, phy_mask:0x80 (50010860002f5647)
[   59.727852] sas: DOING DISCOVERY on port 3, pid:423
[   59.739699] sas: Enter sas_scsi_recover_host busy: 0 failed: 0
[   59.743344] sas: ata13: end_device-12:0: dev error handler
[   59.743402] sas: ata14: end_device-12:1: dev error handler
[   59.743405] sas: ata15: end_device-12:2: dev error handler
[   59.743429] sas: ata16: end_device-12:3: dev error handler
[   59.898839] ata16.00: ATA-11: WDC  WSH722020ALN604, PCGMW803, max UDMA/133
[   62.348080] ata16.00: 4882956288 sectors, multi 0: LBA48 NCQ (depth 32)
[   62.353731] ata16.00: Features: NCQ-sndrcv NCQ-prio
[   62.370011] ata16.00: configured for UDMA/133
[   62.373532] sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 0 tries: 1
[   62.389238] scsi 12:0:3:0: Direct-Access-ZBC ATA      WDC  WSH722020AL W803
PQ: 0 ANSI: 7
[   62.396642] sas: DONE DISCOVERY on port 3, pid:423, result:0

But no scsi disk added:

# lsscsi
[12:0:0:0]   disk    ATA      WDC  WUH721818AL W232  -
[12:0:1:0]   disk    ATA      WDC  WUH721818AL WTW2  -
[12:0:2:0]   disk    ATA      WDC  WUH722222AL W1TZ  -
[12:0:3:0]   zbc     ATA      WDC  WSH722020AL W803  -

(no /dev/sdX device).
While not printed on this run, I often see messages like

scsi 12:0:0:0: deferred probe error

No idea why... I do not see where that comes from.

This is all with pm8001 driver.

Signature · 2 lines
Damien Le Moal
Western Digital Research
reviewer
#17 lore raw

Re: [PATCH 03/19] ata: libata-scsi: link ata port and scsi device

Quoted 11 lines · Damien Le Moal
Quoted 5 lines

hmm... maybe scsi_debug could be modified to support ATA devices (with libata).

Regardless, I'm happy to check the code change which you attempted if shared.

I had a closer look at what the hisi_sas driver is doing. The link it creates with device_add_link() is between the scsi device gendev and the hisi_hba->dev, that is, the HBA device.

So nothing to do with libata port of device.

Correct, it's just for the sdev -> host dev link

Quoted 12 lines

Trying to mimic what libata is now doing, I tried this:

drivers/scsi/libsas/sas_ata.c +1 -0
index 12e2653846e3..91d76258e6ea 100644
@@ -624,6 +624,26 @@ int sas_ata_init(struct domain_device *found_dev)
         return rc;
  }
 
+int sas_ata_slave_configure(struct scsi_device *sdev)

drivers/scsi/libsas/sas_ata.c int sas_ata_init(struct domain_device *found_dev) · line 637

Quoted 10 lines
drivers/scsi/libsas/sas_ata.c +10 -0
+{
+       struct domain_device *dev = sdev_to_domain_dev(sdev);
+       struct ata_port *ap = dev->sata_dev.ap;
+       struct device *sdev_dev = &sdev->sdev_gendev;
+       struct device_link *link;
+
+       ata_sas_slave_configure(sdev, ap);
+
+       link = device_link_add(sdev_dev, &ap->tdev,
+                              DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE);

I am not sure what is the point of this. For libsas/pm8001 driver, there is no ata_port -> .. -> host dev link, right? If so, it just seems that you are are adding a dependency to a device (&ap->tdev) which has no dependency to a real device itself.

Quoted 110 lines
drivers/scsi/libsas/sas_ata.c +9 -0
+       if (!link && pm_runtime_enabled(sdev_dev)) {
+               dev_err(sdev_dev,
+                       "add device link failed, disabling runtime PM for the
 host\n");
+               pm_runtime_disable(sdev_dev);
+       }
+
+       return 0;
+}
+
  void sas_ata_task_abort(struct sas_task *task)
  {
         struct ata_queued_cmd *qc = task->uldd_task;
drivers/scsi/libsas/sas_internal.h +2 -0
index a6dc7dc07fce..82c6d8e1e8c7 100644
@@ -58,6 +58,8 @@ 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);
 
+int sas_ata_slave_configure(struct scsi_device *sdev);
+
  void sas_deform_port(struct asd_sas_phy *phy, int gone);
 
  void sas_porte_bytes_dmaed(struct work_struct *work);
drivers/scsi/libsas/sas_scsi_host.c +2 -4
index 9047cfcd1072..b6c419aa563e 100644
@@ -810,10 +810,8 @@ int sas_slave_configure(struct scsi_device *scsi_dev)
 
         BUG_ON(dev->rphy->identify.device_type != SAS_END_DEVICE);
 
-       if (dev_is_sata(dev)) {
-               ata_sas_slave_configure(scsi_dev, dev->sata_dev.ap);
-               return 0;
-       }
+       if (dev_is_sata(dev))
+               return sas_ata_slave_configure(scsi_dev);
 
         sas_read_port_mode_page(scsi_dev);
 

This does not change the drive discovery, all drives connected to the HBA are found and identified:

[   52.974154] scsi host12: pm80xx
[   54.445986] sas: phy-12:4 added to port-12:0, phy_mask:0x10 (50010860002f5644)
[   54.449019] sas: DOING DISCOVERY on port 0, pid:423
[   54.461108] sas: Enter sas_scsi_recover_host busy: 0 failed: 0
[   54.465601] sas: ata13: end_device-12:0: dev error handler
[   54.618588] ata13.00: ATA-11: WDC  WUH721818ALN604, PCGNW232, max UDMA/133
[   56.295486] ata13.00: 4394582016 sectors, multi 0: LBA48 NCQ (depth 32)
[   56.306849] ata13.00: Features: NCQ-sndrcv NCQ-prio
[   56.320849] ata13.00: configured for UDMA/133
[   56.324504] sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 0 tries: 1
[   56.340346] scsi 12:0:0:0: Direct-Access     ATA      WDC  WUH721818AL W232
PQ: 0 ANSI: 5
[   56.347160] sas: DONE DISCOVERY on port 0, pid:423, result:0
[   56.348161] sas: phy-12:5 added to port-12:1, phy_mask:0x20 (50010860002f5645)
[   56.351291] sas: DOING DISCOVERY on port 1, pid:423
[   56.363285] sas: Enter sas_scsi_recover_host busy: 0 failed: 0
[   56.368788] sas: ata13: end_device-12:0: dev error handler
[   56.368859] sas: ata14: end_device-12:1: dev error handler
[   56.522903] ata14.00: ATA-11: WDC  WUH721818ALN604, PCGNWTW2, max UDMA/133
[   58.402015] ata14.00: 4394582016 sectors, multi 0: LBA48 NCQ (depth 32)
[   58.406207] ata14.00: Features: NCQ-sndrcv NCQ-prio
[   58.420254] ata14.00: configured for UDMA/133
[   58.423222] sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 0 tries: 1
[   58.438869] scsi 12:0:1:0: Direct-Access     ATA      WDC  WUH721818AL WTW2
PQ: 0 ANSI: 5
[   58.445619] sas: DONE DISCOVERY on port 1, pid:423, result:0
[   58.446687] sas: phy-12:6 added to port-12:2, phy_mask:0x40 (50010860002f5646)
[   58.449803] sas: DOING DISCOVERY on port 2, pid:423
[   58.461732] sas: Enter sas_scsi_recover_host busy: 0 failed: 0
[   58.465369] sas: ata13: end_device-12:0: dev error handler
[   58.465426] sas: ata14: end_device-12:1: dev error handler
[   58.465432] sas: ata15: end_device-12:2: dev error handler
[   58.622345] ata15.00: ATA-12: WDC  WUH722222ALN604, LNGNW1TZ, max UDMA/133
[   59.540468] ata15.00: 5371330560 sectors, multi 0: LBA48 NCQ (depth 32)
[   59.588217] ata15.00: Features: NCQ-sndrcv NCQ-prio CDL
[   59.697588] ata15.00: configured for UDMA/133
[   59.700514] sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 0 tries: 1
[   59.716067] scsi 12:0:2:0: Direct-Access     ATA      WDC  WUH722222AL W1TZ
PQ: 0 ANSI: 5
[   59.723838] sas: DONE DISCOVERY on port 2, pid:423, result:0
[   59.724893] sas: phy-12:7 added to port-12:3, phy_mask:0x80 (50010860002f5647)
[   59.727852] sas: DOING DISCOVERY on port 3, pid:423
[   59.739699] sas: Enter sas_scsi_recover_host busy: 0 failed: 0
[   59.743344] sas: ata13: end_device-12:0: dev error handler
[   59.743402] sas: ata14: end_device-12:1: dev error handler
[   59.743405] sas: ata15: end_device-12:2: dev error handler
[   59.743429] sas: ata16: end_device-12:3: dev error handler
[   59.898839] ata16.00: ATA-11: WDC  WSH722020ALN604, PCGMW803, max UDMA/133
[   62.348080] ata16.00: 4882956288 sectors, multi 0: LBA48 NCQ (depth 32)
[   62.353731] ata16.00: Features: NCQ-sndrcv NCQ-prio
[   62.370011] ata16.00: configured for UDMA/133
[   62.373532] sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 0 tries: 1
[   62.389238] scsi 12:0:3:0: Direct-Access-ZBC ATA      WDC  WSH722020AL W803
PQ: 0 ANSI: 7
[   62.396642] sas: DONE DISCOVERY on port 3, pid:423, result:0

But no scsi disk added: >

# lsscsi
[12:0:0:0]   disk    ATA      WDC  WUH721818AL W232  -
[12:0:1:0]   disk    ATA      WDC  WUH721818AL WTW2  -
[12:0:2:0]   disk    ATA      WDC  WUH722222AL W1TZ  -
[12:0:3:0]   zbc     ATA      WDC  WSH722020AL W803  -
Quoted 6 lines

(no /dev/sdX device).
While not printed on this run, I often see messages like

scsi 12:0:0:0: deferred probe error

No idea why... I do not see where that comes from.

It might be related to what I mention, above. Anyway, I would rather you don't get this series bogged down in libsas issues.

Quoted 2 lines

This is all with pm8001 driver.

Thanks,
John

author maintainer
#18 lore raw

Re: [PATCH 03/19] ata: libata-scsi: link ata port and scsi device

Quoted 8 lines · John Garry
Quoted 8 lines
drivers/scsi/libsas/sas_ata.c +8 -0
+int sas_ata_slave_configure(struct scsi_device *sdev)
+{
+       struct domain_device *dev = sdev_to_domain_dev(sdev);
+       struct ata_port *ap = dev->sata_dev.ap;
+       struct device *sdev_dev = &sdev->sdev_gendev;
+       struct device_link *link;
+
+       ata_sas_slave_configure(sdev, ap);
Quoted 8 lines

drivers/scsi/libsas/sas_ata.c int sas_ata_init(struct domain_device *found_dev) · line 637

Quoted 3 lines
drivers/scsi/libsas/sas_ata.c +3 -0
+
+       link = device_link_add(sdev_dev, &ap->tdev,
+                              DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE);

I am not sure what is the point of this. For libsas/pm8001 driver, there is no ata_port -> .. -> host dev link, right? If so, it just seems that you are are adding a dependency to a device (&ap->tdev) which has no dependency to a real device itself.

For libata, the point is to have PM order suspend resume operations correctly:
suspend: scsi dev first, then ata_port
resume: ata_port first then scsi dev

Strictly speaking, we should do that with ata_dev and scsi dev, but libata PM operations are defined for ata_port, not ata_dev.

For libsas, the PM operations come from scsi, so scsi bus operations for devices. And given that:
1) we should already have the dev chain:
hba dev -> scsi host ->scsi target -> scsi dev
2) libsas does its own ata PM control when suspending & resuming the HBA (if I understand things correctly)

we should thus not need anything special for libsas. ata devices suspend/resume will be done in the right order without an extra link. I do not really understand why hisi_sas need to create that link. If it is indeed needed, then we probably should have it created always by libsas generic code...

Signature · 2 lines
Damien Le Moal
Western Digital Research
reviewer
#19 lore raw

Re: [PATCH 03/19] ata: libata-scsi: link ata port and scsi device

+

Quoted 19 lines · Damien Le Moal
Quoted 7 lines
Quoted 7 lines
drivers/scsi/libsas/sas_ata.c +7 -0
+int sas_ata_slave_configure(struct scsi_device *sdev)
+{
+       struct domain_device *dev = sdev_to_domain_dev(sdev);
+       struct ata_port *ap = dev->sata_dev.ap;
+       struct device *sdev_dev = &sdev->sdev_gendev;
+       struct device_link *link;
+
Quoted 8 lines

drivers/scsi/libsas/sas_ata.c int sas_ata_init(struct domain_device *found_dev) · line 637

Quoted 4 lines
drivers/scsi/libsas/sas_ata.c +4 -0
+       ata_sas_slave_configure(sdev, ap);
+
+       link = device_link_add(sdev_dev, &ap->tdev,
+                              DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE);

I am not sure what is the point of this. For libsas/pm8001 driver, there is no ata_port -> .. -> host dev link, right? If so, it just seems that you are are adding a dependency to a device (&ap->tdev) which has no dependency to a real device itself.

For libata, the point is to have PM order suspend resume operations correctly:
suspend: scsi dev first, then ata_port
resume: ata_port first then scsi dev

Quoted 7 lines

Strictly speaking, we should do that with ata_dev and scsi dev, but libata PM operations are defined for ata_port, not ata_dev.

For libsas, the PM operations come from scsi, so scsi bus operations for devices. And given that:
1) we should already have the dev chain:
hba dev -> scsi host ->scsi target -> scsi dev

That would seem right, since scsi_add_host() is passed the host dev and the ancestry would be created naturally, but for hisi_sas a link was still required. Let me check why that is again ...

Quoted 7 lines

2) libsas does its own ata PM control when suspending & resuming the HBA (if I understand things correctly)

we should thus not need anything special for libsas. ata devices suspend/resume will be done in the right order without an extra link. I do not really understand why hisi_sas need to create that link. If it is indeed needed, then we probably should have it created always by libsas generic code...

Thanks,
John

Tested-by on v1 3/19
#20 lore raw

Re: [PATCH 03/19] ata: libata-scsi: link ata port and scsi device

In reply to Damien Le Moal
Quoted 13 lines · Damien Le Moal

There is no direct device ancestry defined between an ata_device and its scsi device which prevents the power management code from correctly ordering suspend and resume operations. Create such ancestry with the ata device as the parent to ensure that the scsi device (child) is suspended before the ata device and that resume handles the ata device before the scsi device.

The parent-child (supplier-consumer) relationship is established between the ata_port (parent) and the scsi device (child) with the function device_add_link(). The parent used is not the ata_device as the PM operations are defined per port and the status of all devices connected through that port is controlled from the port operations.

Quoted 7 lines

The device link is established with the new function ata_scsi_dev_alloc(). This function is used to define the ->slave_alloc callback of the scsi host template of most drivers.

Fixes a19a93e4c6a9 ("scsi: core: pm: Rely on the device driver core for async power management")
author maintainer
#21 lore raw

PATCH 4/19 ata: libata-scsi: Disable scsi device manage_start_stop

In reply to Damien Le Moal

The introduction of a device link to create a consumer/supplier relationship between the scsi device of an ATA device and the ATA port of the ATA device fixed the ordering of the suspend and resume operations. For suspend, the scsi device is suspended first and the ata port after it. This is fine as this allows the synchronize cache and START STOP UNIT commands issued by the scsi disk driver to be executed before the ata port is disabled.

For resume operations, the ata port is resumed first, followed by the scsi device. This allows having the request queue of the scsi device to be unfrozen after the ata port restart is scheduled in EH, thus avoiding to see new requests issued to the ATA device prematurely. However, since libata sets manage_start_stop to 1, the scsi disk resume operation also results in issuing a START STOP UNIT command to wakeup the device. This is too late and that must be done before libata EH resume handling starts revalidating the drive with IDENTIFY etc commands. Commit 0a8589055936 ("ata,scsi: do not issue START STOP UNIT on resume") disabled issuing the START STOP UNIT command to avoid issues with it. However, this is incorrect as transitioning a device to the active power mode from the standby power mode set on suspend requires a media access command. The device link reset and subsequent SET FEATURES, IDENTIFY and READ LOG commands executed in libata EH context triggered by the ata port resume operation may thus fail.

Fix this by handling a device power mode transitions for suspend and resume in libata EH context without relying on the scsi disk management triggered with the manage_start_stop flag.

To do this, the following libata helper functions are introduced:

1) ata_dev_power_set_standby():

This function issues a STANDBY IMMEDIATE command to transitiom a device to the standby power mode. For HDDs, this spins down the disks. This function applies only to ATA and ZAC devices and does nothing otherwise. This function also does nothing for devices that have the ATA_FLAG_NO_POWEROFF_SPINDOWN or ATA_FLAG_NO_HIBERNATE_SPINDOWN flag set.

For suspend, call ata_dev_power_set_standby() in ata_eh_handle_port_suspend() before the port is disabled and frozen. ata_eh_unload() is also modified to transition all enabled devices to the standby power mode when the system is shutdown or devices removed.

2) ata_dev_power_set_active() and

This function applies to ATA or ZAC devices and issues a VERIFY command for 1 sector at LBA 0 to transition the device to the active power mode. For HDDs, since this function will complete only once the disk spin up. Its execution uses the same timeouts as for reset, to give the drive enough time to complete spinup without triggering a command timeout.

For resume, call ata_dev_power_set_active() in ata_eh_revalidate_and_attach() after the port has been enabled and before any other command is issued to the device.

With these changes, the manage_start_stop flag does not need to be set in ata_scsi_dev_config().

Fixes 0a8589055936 ("ata,scsi: do not issue START STOP UNIT on resume")
Fixes a19a93e4c6a9 ("scsi: core: pm: Rely on the device driver core for async power management")
---
 drivers/ata/libata-core.c | 90 +++++++++++++++++++++++++++++++++++++++
 drivers/ata/libata-eh.c   | 46 +++++++++++++++++++-
 drivers/ata/libata-scsi.c | 14 ++----
 drivers/ata/libata.h      |  2 +
 include/linux/libata.h    |  6 ++-
 5 files changed, 144 insertions(+), 14 deletions(-)
drivers/ata/libata-core.c +90 -0
index 693cb3cd70cd..0cf0caf77907 100644
@@ -1972,6 +1972,96 @@ int ata_dev_read_id(struct ata_device *dev, unsigned int *p_class,
 	return rc;
 }
 
+/**
+ *	ata_dev_power_set_standby - Set a device power mode to standby
+ *	@dev: target device
+ *
+ *	Issue a STANDBY IMMEDIATE command to set a device power mode to standby.
+ *	For an HDD device, this spins down the disks.
+ *
+ *	LOCKING:
+ *	Kernel thread context (may sleep).
+ */
+void ata_dev_power_set_standby(struct ata_device *dev)
+{
+	unsigned long ap_flags = dev->link->ap->flags;
+	struct ata_taskfile tf;
+	unsigned int err_mask;
+
+	/* Issue STANDBY IMMEDIATE command only if supported by the device */
+	if (dev->class != ATA_DEV_ATA && dev->class != ATA_DEV_ZAC)
+		return;
+
+	/*
+	 * Some odd clown BIOSes issue spindown on power off (ACPI S4 or S5)
+	 * causing some drives to spin up and down again. For these, do nothing
+	 * if we are being called on shutdown.
+	 */
+	if ((ap_flags & ATA_FLAG_NO_POWEROFF_SPINDOWN) &&
+	    system_state == SYSTEM_POWER_OFF)
+		return;
+
+	if ((ap_flags & ATA_FLAG_NO_HIBERNATE_SPINDOWN) &&
+	    system_entering_hibernation())
+		return;
+
+	ata_tf_init(dev, &tf);
+	tf.flags |= ATA_TFLAG_DEVICE | ATA_TFLAG_ISADDR;
+	tf.protocol = ATA_PROT_NODATA;
+	tf.command = ATA_CMD_STANDBYNOW1;
+
+	ata_dev_notice(dev, "Entering standby power mode\n");
+
+	err_mask = ata_exec_internal(dev, &tf, NULL, DMA_NONE, NULL, 0, 0);
+	if (err_mask)
+		ata_dev_err(dev, "STANDBY IMMEDIATE failed (err_mask=0x%x)\n",
+			    err_mask);
+}
+
+/**
+ *	ata_dev_power_set_active -  Set a device power mode to active
+ *	@dev: target device
+ *
+ *	Issue a VERIFY command to enter to ensure that the device is in the
+ *	active power mode. For a spun-down HDD (standby or idle power mode),
+ *	the VERIFY command will complete after the disk spins up.
+ *
+ *	LOCKING:
+ *	Kernel thread context (may sleep).
+ */
+void ata_dev_power_set_active(struct ata_device *dev)
+{
+	struct ata_taskfile tf;
+	unsigned int err_mask;
+
+	/*
+	 * Issue READ VERIFY SECTORS command for 1 sector at lba=0 only
+	 * if supported by the device.
+	 */
+	if (dev->class != ATA_DEV_ATA && dev->class != ATA_DEV_ZAC)
+		return;
+
+	ata_tf_init(dev, &tf);
+	tf.flags |= ATA_TFLAG_DEVICE | ATA_TFLAG_ISADDR;
+	tf.protocol = ATA_PROT_NODATA;
+	tf.command = ATA_CMD_VERIFY;
+	tf.nsect = 1;
+	if (dev->flags & ATA_DFLAG_LBA) {
+		tf.flags |= ATA_TFLAG_LBA;
+		tf.device |= ATA_LBA;
+	} else {
+		/* CHS */
+		tf.lbal = 0x1; /* sect */
+	}
+
+	ata_dev_notice(dev, "Entering active power mode\n");
+
+	err_mask = ata_exec_internal(dev, &tf, NULL, DMA_NONE, NULL, 0, 0);
+	if (err_mask)
+		ata_dev_err(dev, "VERIFY failed (err_mask=0x%x)\n",
+			    err_mask);
+}
+
 /**
  *	ata_read_log_page - read a specific log page
  *	@dev: target device
drivers/ata/libata-eh.c +45 -1

100 lines not on this page · raw message

drivers/ata/libata-scsi.c +3 -11
index f63cf6e7332e..9bb1ace8bf79 100644
@@ -1050,14 +1050,6 @@ int ata_scsi_dev_config(struct scsi_device *sdev, struct ata_device *dev)
 		}
 	} else {
 		sdev->sector_size = ata_id_logical_sector_size(dev->id);
-		/*
-		 * Stop the drive on suspend but do not issue START STOP UNIT
-		 * on resume as this is not necessary and may fail: the device
-		 * will be woken up by ata_port_pm_resume() with a port reset
-		 * and device revalidation.
-		 */
-		sdev->manage_start_stop = 1;
-		sdev->no_start_on_resume = 1;
 	}
 
 	/*
@@ -1231,7 +1223,7 @@ static unsigned int ata_scsi_start_stop_xlat(struct ata_queued_cmd *qc)
 	}
 
 	if (cdb[4] & 0x1) {
-		tf->nsect = 1;	/* 1 sector, lba=0 */
+		tf->nsect = 1;  /* 1 sector, lba=0 */
 
 		if (qc->dev->flags & ATA_DFLAG_LBA) {
 			tf->flags |= ATA_TFLAG_LBA;
@@ -1247,7 +1239,7 @@ static unsigned int ata_scsi_start_stop_xlat(struct ata_queued_cmd *qc)
 			tf->lbah = 0x0; /* cyl high */
 		}
 
-		tf->command = ATA_CMD_VERIFY;	/* READ VERIFY */
+		tf->command = ATA_CMD_VERIFY;   /* READ VERIFY */
 	} else {
 		/* Some odd clown BIOSen issue spindown on power off (ACPI S4
 		 * or S5) causing some drives to spin up and down again.
@@ -1257,7 +1249,7 @@ static unsigned int ata_scsi_start_stop_xlat(struct ata_queued_cmd *qc)
 			goto skip;
 
 		if ((qc->ap->flags & ATA_FLAG_NO_HIBERNATE_SPINDOWN) &&
-		     system_entering_hibernation())
+		    system_entering_hibernation())
 			goto skip;
 
 		/* Issue ATA STANDBY IMMEDIATE command */
drivers/ata/libata.h +2 -0
index 079981e7156a..a5ee06f0234a 100644
@@ -60,6 +60,8 @@ extern int ata_dev_reread_id(struct ata_device *dev, unsigned int readid_flags);
 extern int ata_dev_revalidate(struct ata_device *dev, unsigned int new_class,
 			      unsigned int readid_flags);
 extern int ata_dev_configure(struct ata_device *dev);
+extern void ata_dev_power_set_standby(struct ata_device *dev);
+extern void ata_dev_power_set_active(struct ata_device *dev);
 extern int sata_down_spd_limit(struct ata_link *link, u32 spd_limit);
 extern int ata_down_xfermask_limit(struct ata_device *dev, unsigned int sel);
 extern unsigned int ata_dev_set_feature(struct ata_device *dev,
include/linux/libata.h +4 -2
index c8cfea386c16..6593c79b7290 100644
@@ -192,6 +192,7 @@ enum {
 	ATA_PFLAG_UNLOADING	= (1 << 9), /* driver is being unloaded */
 	ATA_PFLAG_UNLOADED	= (1 << 10), /* driver is unloaded */
 
+	ATA_PFLAG_RESUMING	= (1 << 16),  /* port is being resumed */
 	ATA_PFLAG_SUSPENDED	= (1 << 17), /* port is suspended (power) */
 	ATA_PFLAG_PM_PENDING	= (1 << 18), /* PM operation pending */
 	ATA_PFLAG_INIT_GTM_VALID = (1 << 19), /* initial gtm data valid */
@@ -314,9 +315,10 @@ enum {
 	ATA_EH_ENABLE_LINK	= (1 << 3),
 	ATA_EH_PARK		= (1 << 5), /* unload heads and stop I/O */
 	ATA_EH_GET_SUCCESS_SENSE = (1 << 6), /* Get sense data for successful cmd */
+	ATA_EH_SET_ACTIVE	= (1 << 7), /* Set a device to active power mode */
 
 	ATA_EH_PERDEV_MASK	= ATA_EH_REVALIDATE | ATA_EH_PARK |
-				  ATA_EH_GET_SUCCESS_SENSE,
+				  ATA_EH_GET_SUCCESS_SENSE | ATA_EH_SET_ACTIVE,
 	ATA_EH_ALL_ACTIONS	= ATA_EH_REVALIDATE | ATA_EH_RESET |
 				  ATA_EH_ENABLE_LINK,
 
@@ -353,7 +355,7 @@ enum {
 	/* This should match the actual table size of
 	 * ata_eh_cmd_timeout_table in libata-eh.c.
 	 */
-	ATA_EH_CMD_TIMEOUT_TABLE_SIZE = 7,
+	ATA_EH_CMD_TIMEOUT_TABLE_SIZE = 8,
 
 	/* Horkage types. May be set by libata or controller on drives
 	   (some horkage may be drive/controller pair dependent */
Signature · 1 line
2.41.0
#22 lore raw

Re: [PATCH 04/19] ata: libata-scsi: Disable scsi device manage_start_stop

Quoted 44 lines · Damien Le Moal

The introduction of a device link to create a consumer/supplier relationship between the scsi device of an ATA device and the ATA port of the ATA device fixed the ordering of the suspend and resume operations. For suspend, the scsi device is suspended first and the ata port after it. This is fine as this allows the synchronize cache and START STOP UNIT commands issued by the scsi disk driver to be executed before the ata port is disabled.

For resume operations, the ata port is resumed first, followed by the scsi device. This allows having the request queue of the scsi device to be unfrozen after the ata port restart is scheduled in EH, thus avoiding to see new requests issued to the ATA device prematurely. However, since libata sets manage_start_stop to 1, the scsi disk resume operation also results in issuing a START STOP UNIT command to wakeup the device. This is too late and that must be done before libata EH resume handling starts revalidating the drive with IDENTIFY etc commands. Commit 0a8589055936 ("ata,scsi: do not issue START STOP UNIT on resume") disabled issuing the START STOP UNIT command to avoid issues with it. However, this is incorrect as transitioning a device to the active power mode from the standby power mode set on suspend requires a media access command. The device link reset and subsequent SET FEATURES, IDENTIFY and READ LOG commands executed in libata EH context triggered by the ata port resume operation may thus fail.

Fix this by handling a device power mode transitions for suspend and resume in libata EH context without relying on the scsi disk management triggered with the manage_start_stop flag.

To do this, the following libata helper functions are introduced:

1) ata_dev_power_set_standby():

This function issues a STANDBY IMMEDIATE command to transitiom a device to the standby power mode. For HDDs, this spins down the disks. This function applies only to ATA and ZAC devices and does nothing otherwise. This function also does nothing for devices that have the ATA_FLAG_NO_POWEROFF_SPINDOWN or ATA_FLAG_NO_HIBERNATE_SPINDOWN flag set.

For suspend, call ata_dev_power_set_standby() in ata_eh_handle_port_suspend() before the port is disabled and frozen. ata_eh_unload() is also modified to transition all enabled devices to the standby power mode when the system is shutdown or devices removed.

Quoted 8 lines

2) ata_dev_power_set_active() and

This function applies to ATA or ZAC devices and issues a VERIFY command for 1 sector at LBA 0 to transition the device to the active power mode. For HDDs, since this function will complete only once the disk spin up. Its execution uses the same timeouts as for reset, to give the drive enough time to complete spinup without triggering a command timeout.

Neat. But why VERIFY?
Isn't there a dedicated command (ie the opposite of STANDBY IMMEDIATE)? And can we be sure that VERIFY is implemented everywhere? It's not that this command had been in active use until now ...

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
author maintainer
#23 lore raw

Re: [PATCH 04/19] ata: libata-scsi: Disable scsi device manage_start_stop

Quoted 47 lines · Hannes Reinecke
Quoted 46 lines · Damien Le Moal

The introduction of a device link to create a consumer/supplier relationship between the scsi device of an ATA device and the ATA port of the ATA device fixed the ordering of the suspend and resume operations. For suspend, the scsi device is suspended first and the ata port after it. This is fine as this allows the synchronize cache and START STOP UNIT commands issued by the scsi disk driver to be executed before the ata port is disabled.

For resume operations, the ata port is resumed first, followed by the scsi device. This allows having the request queue of the scsi device to be unfrozen after the ata port restart is scheduled in EH, thus avoiding to see new requests issued to the ATA device prematurely. However, since libata sets manage_start_stop to 1, the scsi disk resume operation also results in issuing a START STOP UNIT command to wakeup the device. This is too late and that must be done before libata EH resume handling starts revalidating the drive with IDENTIFY etc commands. Commit 0a8589055936 ("ata,scsi: do not issue START STOP UNIT on resume") disabled issuing the START STOP UNIT command to avoid issues with it. However, this is incorrect as transitioning a device to the active power mode from the standby power mode set on suspend requires a media access command. The device link reset and subsequent SET FEATURES, IDENTIFY and READ LOG commands executed in libata EH context triggered by the ata port resume operation may thus fail.

Fix this by handling a device power mode transitions for suspend and resume in libata EH context without relying on the scsi disk management triggered with the manage_start_stop flag.

To do this, the following libata helper functions are introduced:

1) ata_dev_power_set_standby():

This function issues a STANDBY IMMEDIATE command to transitiom a device to the standby power mode. For HDDs, this spins down the disks. This function applies only to ATA and ZAC devices and does nothing otherwise. This function also does nothing for devices that have the ATA_FLAG_NO_POWEROFF_SPINDOWN or ATA_FLAG_NO_HIBERNATE_SPINDOWN flag set.

For suspend, call ata_dev_power_set_standby() in ata_eh_handle_port_suspend() before the port is disabled and frozen. ata_eh_unload() is also modified to transition all enabled devices to the standby power mode when the system is shutdown or devices removed.

2) ata_dev_power_set_active() and

Quoted 7 lines
Quoted 6 lines

This function applies to ATA or ZAC devices and issues a VERIFY command for 1 sector at LBA 0 to transition the device to the active power mode. For HDDs, since this function will complete only once the disk spin up. Its execution uses the same timeouts as for reset, to give the drive enough time to complete spinup without triggering a command timeout.

Neat. But why VERIFY?

Ask that to T13 :) Need a media access command to get out of sleep state... Could use a read, but then need a buffer, which is silly for just waking up a drive. VERIFY is a mandatory command.

Quoted 3 lines

Isn't there a dedicated command (ie the opposite of STANDBY IMMEDIATE)? And can we be sure that VERIFY is implemented everywhere? It's not that this command had been in active use until now ...

START STOP UNIT with start == 1 has been translated to a VERIFY command since forever. This is according to SAT specs. There is no command to explicitly get out of standby-mode. Even a reset should not change the drive power state (though I do see a lot of drive waking up on COMRESET). A media access command does that. See ACS specs "Power Management states and transitions". The only exception is that you can use SET FEATURE command to wake up a drive, but only from PUIS state (Power-Up in Standby), which is a different feature that is not necessarilly supported by a device.

Quoted 4 lines

Cheers,

Hannes

Signature · 2 lines
Damien Le Moal
Western Digital Research
Reviewed-by
#24 lore raw

Re: [PATCH 04/19] ata: libata-scsi: Disable scsi device manage_start_stop

Quoted 64 lines · Damien Le Moal
Quoted 47 lines · Hannes Reinecke
Quoted 46 lines · Damien Le Moal

The introduction of a device link to create a consumer/supplier relationship between the scsi device of an ATA device and the ATA port of the ATA device fixed the ordering of the suspend and resume operations. For suspend, the scsi device is suspended first and the ata port after it. This is fine as this allows the synchronize cache and START STOP UNIT commands issued by the scsi disk driver to be executed before the ata port is disabled.

For resume operations, the ata port is resumed first, followed by the scsi device. This allows having the request queue of the scsi device to be unfrozen after the ata port restart is scheduled in EH, thus avoiding to see new requests issued to the ATA device prematurely. However, since libata sets manage_start_stop to 1, the scsi disk resume operation also results in issuing a START STOP UNIT command to wakeup the device. This is too late and that must be done before libata EH resume handling starts revalidating the drive with IDENTIFY etc commands. Commit 0a8589055936 ("ata,scsi: do not issue START STOP UNIT on resume") disabled issuing the START STOP UNIT command to avoid issues with it. However, this is incorrect as transitioning a device to the active power mode from the standby power mode set on suspend requires a media access command. The device link reset and subsequent SET FEATURES, IDENTIFY and READ LOG commands executed in libata EH context triggered by the ata port resume operation may thus fail.

Fix this by handling a device power mode transitions for suspend and resume in libata EH context without relying on the scsi disk management triggered with the manage_start_stop flag.

To do this, the following libata helper functions are introduced:

1) ata_dev_power_set_standby():

This function issues a STANDBY IMMEDIATE command to transitiom a device to the standby power mode. For HDDs, this spins down the disks. This function applies only to ATA and ZAC devices and does nothing otherwise. This function also does nothing for devices that have the ATA_FLAG_NO_POWEROFF_SPINDOWN or ATA_FLAG_NO_HIBERNATE_SPINDOWN flag set.

For suspend, call ata_dev_power_set_standby() in ata_eh_handle_port_suspend() before the port is disabled and frozen. ata_eh_unload() is also modified to transition all enabled devices to the standby power mode when the system is shutdown or devices removed.

2) ata_dev_power_set_active() and

Quoted 7 lines
Quoted 6 lines

This function applies to ATA or ZAC devices and issues a VERIFY command for 1 sector at LBA 0 to transition the device to the active power mode. For HDDs, since this function will complete only once the disk spin up. Its execution uses the same timeouts as for reset, to give the drive enough time to complete spinup without triggering a command timeout.

Neat. But why VERIFY?

Ask that to T13 :) Need a media access command to get out of sleep state... Could use a read, but then need a buffer, which is silly for just waking up a drive. VERIFY is a mandatory command.

Quoted 3 lines

Isn't there a dedicated command (ie the opposite of STANDBY IMMEDIATE)? And can we be sure that VERIFY is implemented everywhere? It's not that this command had been in active use until now ...

Quoted 9 lines

START STOP UNIT with start == 1 has been translated to a VERIFY command since forever. This is according to SAT specs. There is no command to explicitly get out of standby-mode. Even a reset should not change the drive power state (though I do see a lot of drive waking up on COMRESET). A media access command does that. See ACS specs "Power Management states and transitions". The only exception is that you can use SET FEATURE command to wake up a drive, but only from PUIS state (Power-Up in Standby), which is a different feature that is not necessarilly supported by a device.

Sheesh. Why make life simple if you can also make it complicated... But if we've been using VERIFY already I'm fine with this change.

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
#25 lore raw

Re: [PATCH 04/19] ata: libata-scsi: Disable scsi device manage_start_stop

Quoted 7 lines · Hannes Reinecke
Quoted 7 lines

forever. This is according to SAT specs. There is no command to explicitly get out of standby-mode. Even a reset should not change the drive power state (though I do see a lot of drive waking up on COMRESET). A media access command does that. See ACS specs "Power Management states and transitions". The only exception is that you can use SET FEATURE command to wake up a drive, but only from PUIS state (Power-Up in Standby), which is a different feature that is not necessarilly supported by a device.

IIRC, there was a bit in the IDENTIFY block that if set, meant that the drive WOULD NOT automatically spin up on access, and you HAD to use SET FEATURE to wake it. I seem to recall that my Western Digital drives did this and my bios would not recognize the drives as bootable because it did not know how to issue SET FEATURE to force it to spin up.

Tested-by on v1 4/19
#26 lore raw

Re: [PATCH 04/19] ata: libata-scsi: Disable scsi device manage_start_stop

In reply to Damien Le Moal
Quoted 56 lines · Damien Le Moal

The introduction of a device link to create a consumer/supplier relationship between the scsi device of an ATA device and the ATA port of the ATA device fixed the ordering of the suspend and resume operations. For suspend, the scsi device is suspended first and the ata port after it. This is fine as this allows the synchronize cache and START STOP UNIT commands issued by the scsi disk driver to be executed before the ata port is disabled.

For resume operations, the ata port is resumed first, followed by the scsi device. This allows having the request queue of the scsi device to be unfrozen after the ata port restart is scheduled in EH, thus avoiding to see new requests issued to the ATA device prematurely. However, since libata sets manage_start_stop to 1, the scsi disk resume operation also results in issuing a START STOP UNIT command to wakeup the device. This is too late and that must be done before libata EH resume handling starts revalidating the drive with IDENTIFY etc commands. Commit 0a8589055936 ("ata,scsi: do not issue START STOP UNIT on resume") disabled issuing the START STOP UNIT command to avoid issues with it. However, this is incorrect as transitioning a device to the active power mode from the standby power mode set on suspend requires a media access command. The device link reset and subsequent SET FEATURES, IDENTIFY and READ LOG commands executed in libata EH context triggered by the ata port resume operation may thus fail.

Fix this by handling a device power mode transitions for suspend and resume in libata EH context without relying on the scsi disk management triggered with the manage_start_stop flag.

To do this, the following libata helper functions are introduced:

1) ata_dev_power_set_standby():

This function issues a STANDBY IMMEDIATE command to transitiom a device to the standby power mode. For HDDs, this spins down the disks. This function applies only to ATA and ZAC devices and does nothing otherwise. This function also does nothing for devices that have the ATA_FLAG_NO_POWEROFF_SPINDOWN or ATA_FLAG_NO_HIBERNATE_SPINDOWN flag set.

For suspend, call ata_dev_power_set_standby() in ata_eh_handle_port_suspend() before the port is disabled and frozen. ata_eh_unload() is also modified to transition all enabled devices to the standby power mode when the system is shutdown or devices removed.

2) ata_dev_power_set_active() and

This function applies to ATA or ZAC devices and issues a VERIFY command for 1 sector at LBA 0 to transition the device to the active power mode. For HDDs, since this function will complete only once the disk spin up. Its execution uses the same timeouts as for reset, to give the drive enough time to complete spinup without triggering a command timeout.

For resume, call ata_dev_power_set_active() in ata_eh_revalidate_and_attach() after the port has been enabled and before any other command is issued to the device.

Quoted 7 lines

With these changes, the manage_start_stop flag does not need to be set in ata_scsi_dev_config().

Fixes 0a8589055936 ("ata,scsi: do not issue START STOP UNIT on resume")
Fixes a19a93e4c6a9 ("scsi: core: pm: Rely on the device driver core for async power management")
author maintainer
#27 lore raw

PATCH 5/19 ata: libata-scsi: Fix delayed scsi_rescan_device() execution

In reply to Damien Le Moal

Commit 6aa0365a3c85 ("ata: libata-scsi: Avoid deadlock on rescan after device resume") modified ata_scsi_dev_rescan() to check the scsi device "is_suspended" power field to ensure that the scsi device associated with an ATA device is fully resumed when scsi_rescan_device() is executed. However, this fix is problematic as:
1) it relies on a PM internal field that should not be used without PM device locking protection.
2) The check for is_suspended and the call to ata_scsi_dev_rescan() are not atomic and a suspend PM even may be triggered between them, casuing ata_scsi_dev_rescan() to be called on a suspended device, resulting in that function blocking while holding the scsi device lock, which would deadlock a following resume operation.
These problems can trigger PM deadlocks on resume, especially with resume operations triggered quickly after or during suspend operations. E.g., a simple bash script like:

for (( i=0; i<10; i++ )); do
echo "+2 > /sys/class/rtc/rtc0/wakealarm
echo mem > /sys/power/state
done

that triggers a resume 2 seconds after starting suspending a system can quickly lead to a PM deadlock preventing the system from correctly resuming.

Fix this by replacing the check on is_suspended with a check on the scsi device state inside ata_scsi_dev_rescan(), while holding the scsi device lock, thus making the device rescan atomic with regard to PM operations. Additionnly, make sure that scheduled rescan tasks are first cancelled before suspending an ata port.

Fixes 6aa0365a3c85 ("ata: libata-scsi: Avoid deadlock on rescan after device resume")
---
 drivers/ata/libata-core.c | 16 ++++++++++++++++
 drivers/ata/libata-scsi.c | 36 ++++++++++++++++++------------------
 drivers/scsi/scsi_scan.c  | 12 +++++++++++-
 include/scsi/scsi_host.h  |  2 +-
 4 files changed, 46 insertions(+), 20 deletions(-)
drivers/ata/libata-core.c +16 -0
index 0cf0caf77907..0479493e54bd 100644
@@ -5172,11 +5172,27 @@ static const unsigned int ata_port_suspend_ehi = ATA_EHI_QUIET
 
 static void ata_port_suspend(struct ata_port *ap, pm_message_t mesg)
 {
+	/*
+	 * We are about to suspend the port, so we do not care about
+	 * scsi_rescan_device() calls scheduled by previous resume operations.
+	 * The next resume will schedule the rescan again. So cancel any rescan
+	 * that is not done yet.
+	 */
+	cancel_delayed_work_sync(&ap->scsi_rescan_task);
+
 	ata_port_request_pm(ap, mesg, 0, ata_port_suspend_ehi, false);
 }
 
 static void ata_port_suspend_async(struct ata_port *ap, pm_message_t mesg)
 {
+	/*
+	 * We are about to suspend the port, so we do not care about
+	 * scsi_rescan_device() calls scheduled by previous resume operations.
+	 * The next resume will schedule the rescan again. So cancel any rescan
+	 * that is not done yet.
+	 */
+	cancel_delayed_work_sync(&ap->scsi_rescan_task);
+
 	ata_port_request_pm(ap, mesg, 0, ata_port_suspend_ehi, true);
 }
 
drivers/ata/libata-scsi.c +18 -18
index 9bb1ace8bf79..f2d4460ab450 100644
@@ -4750,7 +4750,7 @@ void ata_scsi_dev_rescan(struct work_struct *work)
 	struct ata_link *link;
 	struct ata_device *dev;
 	unsigned long flags;
-	bool delay_rescan = false;
+	int ret = 0;
 
 	mutex_lock(&ap->scsi_scan_mutex);
 	spin_lock_irqsave(ap->lock, flags);
@@ -4759,37 +4759,37 @@ void ata_scsi_dev_rescan(struct work_struct *work)
 		ata_for_each_dev(dev, link, ENABLED) {
 			struct scsi_device *sdev = dev->sdev;
 
+			/*
+			 * If the port was suspended before this was scheduled,
+			 * bail out.
+			 */
+			if (ap->pflags & ATA_PFLAG_SUSPENDED)
+				goto unlock;
+
 			if (!sdev)
 				continue;
 			if (scsi_device_get(sdev))
 				continue;
 
-			/*
-			 * If the rescan work was scheduled because of a resume
-			 * event, the port is already fully resumed, but the
-			 * SCSI device may not yet be fully resumed. In such
-			 * case, executing scsi_rescan_device() may cause a
-			 * deadlock with the PM code on device_lock(). Prevent
-			 * this by giving up and retrying rescan after a short
-			 * delay.
-			 */
-			delay_rescan = sdev->sdev_gendev.power.is_suspended;
-			if (delay_rescan) {
-				scsi_device_put(sdev);
-				break;
-			}
-
 			spin_unlock_irqrestore(ap->lock, flags);
-			scsi_rescan_device(sdev);
+			ret = scsi_rescan_device(sdev);
 			scsi_device_put(sdev);
 			spin_lock_irqsave(ap->lock, flags);
+
+			if (ret)
+				goto unlock;
 		}
 	}
 
+unlock:
 	spin_unlock_irqrestore(ap->lock, flags);
 	mutex_unlock(&ap->scsi_scan_mutex);
 
-	if (delay_rescan)
+	/*
+	 * Reschedule with a delay if scsi_rescan_device() returned an error
+	 * and the port has not been suspended.
+	 */
+	if (ret && !(ap->pflags & ATA_PFLAG_SUSPENDED))
 		schedule_delayed_work(&ap->scsi_rescan_task,
 				      msecs_to_jiffies(5));
 }
drivers/scsi/scsi_scan.c +11 -1
index 52014b2d39e1..6650f63afec9 100644
@@ -1619,12 +1619,18 @@ int scsi_add_device(struct Scsi_Host *host, uint channel,
 }
 EXPORT_SYMBOL(scsi_add_device);
 
-void scsi_rescan_device(struct scsi_device *sdev)
+int scsi_rescan_device(struct scsi_device *sdev)
 {
 	struct device *dev = &sdev->sdev_gendev;
+	int ret = 0;
 
 	device_lock(dev);
 
+	if (sdev->sdev_state != SDEV_RUNNING) {
+		ret = -ENXIO;
+		goto unlock;
+	}
+
 	scsi_attach_vpd(sdev);
 	scsi_cdl_check(sdev);
 
@@ -1638,7 +1644,11 @@ void scsi_rescan_device(struct scsi_device *sdev)
 			drv->rescan(dev);
 		module_put(dev->driver->owner);
 	}
+
+unlock:
 	device_unlock(dev);
+
+	return ret;
 }
 EXPORT_SYMBOL(scsi_rescan_device);
 
include/scsi/scsi_host.h +1 -1
index 49f768d0ff37..4c2dc8150c6d 100644
@@ -764,7 +764,7 @@ scsi_template_proc_dir(const struct scsi_host_template *sht);
 #define scsi_template_proc_dir(sht) NULL
 #endif
 extern void scsi_scan_host(struct Scsi_Host *);
-extern void scsi_rescan_device(struct scsi_device *);
+extern int scsi_rescan_device(struct scsi_device *sdev);
 extern void scsi_remove_host(struct Scsi_Host *);
 extern struct Scsi_Host *scsi_host_get(struct Scsi_Host *);
 extern int scsi_host_busy(struct Scsi_Host *shost);
Signature · 1 line
2.41.0
Reviewed-by on v1 5/19
#28 lore raw

Re: [PATCH 05/19] ata: libata-scsi: Fix delayed scsi_rescan_device() execution

Quoted 34 lines · Damien Le Moal

Commit 6aa0365a3c85 ("ata: libata-scsi: Avoid deadlock on rescan after device resume") modified ata_scsi_dev_rescan() to check the scsi device "is_suspended" power field to ensure that the scsi device associated with an ATA device is fully resumed when scsi_rescan_device() is executed. However, this fix is problematic as:
1) it relies on a PM internal field that should not be used without PM device locking protection.
2) The check for is_suspended and the call to ata_scsi_dev_rescan() are not atomic and a suspend PM even may be triggered between them, casuing ata_scsi_dev_rescan() to be called on a suspended device, resulting in that function blocking while holding the scsi device lock, which would deadlock a following resume operation.
These problems can trigger PM deadlocks on resume, especially with resume operations triggered quickly after or during suspend operations. E.g., a simple bash script like:

for (( i=0; i<10; i++ )); do
echo "+2 > /sys/class/rtc/rtc0/wakealarm
echo mem > /sys/power/state
done

that triggers a resume 2 seconds after starting suspending a system can quickly lead to a PM deadlock preventing the system from correctly resuming.

Fix this by replacing the check on is_suspended with a check on the scsi device state inside ata_scsi_dev_rescan(), while holding the scsi device lock, thus making the device rescan atomic with regard to PM operations. Additionnly, make sure that scheduled rescan tasks are first cancelled before suspending an ata port.

Fixes 6aa0365a3c85 ("ata: libata-scsi: Avoid deadlock on rescan after device resume")
Quoted 7 lines
---
  drivers/ata/libata-core.c | 16 ++++++++++++++++
  drivers/ata/libata-scsi.c | 36 ++++++++++++++++++------------------
  drivers/scsi/scsi_scan.c  | 12 +++++++++++-
  include/scsi/scsi_host.h  |  2 +-
  4 files changed, 46 insertions(+), 20 deletions(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
Tested-by on v1 5/19
#29 lore raw

Re: [PATCH 05/19] ata: libata-scsi: Fix delayed scsi_rescan_device() execution

In reply to Damien Le Moal
Quoted 31 lines · Damien Le Moal

Commit 6aa0365a3c85 ("ata: libata-scsi: Avoid deadlock on rescan after device resume") modified ata_scsi_dev_rescan() to check the scsi device "is_suspended" power field to ensure that the scsi device associated with an ATA device is fully resumed when scsi_rescan_device() is executed. However, this fix is problematic as:
1) it relies on a PM internal field that should not be used without PM device locking protection.
2) The check for is_suspended and the call to ata_scsi_dev_rescan() are not atomic and a suspend PM even may be triggered between them, casuing ata_scsi_dev_rescan() to be called on a suspended device, resulting in that function blocking while holding the scsi device lock, which would deadlock a following resume operation.
These problems can trigger PM deadlocks on resume, especially with resume operations triggered quickly after or during suspend operations. E.g., a simple bash script like:

for (( i=0; i<10; i++ )); do
echo "+2 > /sys/class/rtc/rtc0/wakealarm
echo mem > /sys/power/state
done

that triggers a resume 2 seconds after starting suspending a system can quickly lead to a PM deadlock preventing the system from correctly resuming.

Fix this by replacing the check on is_suspended with a check on the scsi device state inside ata_scsi_dev_rescan(), while holding the scsi device lock, thus making the device rescan atomic with regard to PM operations. Additionnly, make sure that scheduled rescan tasks are first cancelled before suspending an ata port.

Quoted 3 lines
Fixes 6aa0365a3c85 ("ata: libata-scsi: Avoid deadlock on rescan after device resume")
#30 lore raw

Re: [PATCH 05/19] ata: libata-scsi: Fix delayed scsi_rescan_device() execution

In reply to Damien Le Moal
Quoted 25 lines · Damien Le Moal

Commit 6aa0365a3c85 ("ata: libata-scsi: Avoid deadlock on rescan after device resume") modified ata_scsi_dev_rescan() to check the scsi device "is_suspended" power field to ensure that the scsi device associated with an ATA device is fully resumed when scsi_rescan_device() is executed. However, this fix is problematic as:
1) it relies on a PM internal field that should not be used without PM device locking protection.
2) The check for is_suspended and the call to ata_scsi_dev_rescan() are not atomic and a suspend PM even may be triggered between them, casuing ata_scsi_dev_rescan() to be called on a suspended device, resulting in that function blocking while holding the scsi device lock, which would deadlock a following resume operation.
These problems can trigger PM deadlocks on resume, especially with resume operations triggered quickly after or during suspend operations. E.g., a simple bash script like:

for (( i=0; i<10; i++ )); do
echo "+2 > /sys/class/rtc/rtc0/wakealarm
echo mem > /sys/power/state
done

that triggers a resume 2 seconds after starting suspending a system can quickly lead to a PM deadlock preventing the system from correctly resuming.

Quoted 5 lines

Fix this by replacing the check on is_suspended with a check on the scsi device state inside ata_scsi_dev_rescan(), while holding the scsi device lock, thus making the device rescan atomic with regard to PM operations. Additionnly, make sure that scheduled rescan tasks are first cancelled before suspending an ata port.

One patch per subsystem please. I think this patch can be split easily into an ATA patch and a SCSI core patch.

Thanks,

Bart.

author maintainer
#31 lore raw

Re: [PATCH 05/19] ata: libata-scsi: Fix delayed scsi_rescan_device() execution

Quoted 26 lines · Bart Van Assche
Quoted 25 lines · Damien Le Moal

Commit 6aa0365a3c85 ("ata: libata-scsi: Avoid deadlock on rescan after device resume") modified ata_scsi_dev_rescan() to check the scsi device "is_suspended" power field to ensure that the scsi device associated with an ATA device is fully resumed when scsi_rescan_device() is executed. However, this fix is problematic as:
1) it relies on a PM internal field that should not be used without PM device locking protection.
2) The check for is_suspended and the call to ata_scsi_dev_rescan() are not atomic and a suspend PM even may be triggered between them, casuing ata_scsi_dev_rescan() to be called on a suspended device, resulting in that function blocking while holding the scsi device lock, which would deadlock a following resume operation.
These problems can trigger PM deadlocks on resume, especially with resume operations triggered quickly after or during suspend operations. E.g., a simple bash script like:

for (( i=0; i<10; i++ )); do
echo "+2 > /sys/class/rtc/rtc0/wakealarm
echo mem > /sys/power/state
done

that triggers a resume 2 seconds after starting suspending a system can quickly lead to a PM deadlock preventing the system from correctly resuming.

Quoted 8 lines
Quoted 5 lines

Fix this by replacing the check on is_suspended with a check on the scsi device state inside ata_scsi_dev_rescan(), while holding the scsi device lock, thus making the device rescan atomic with regard to PM operations. Additionnly, make sure that scheduled rescan tasks are first cancelled before suspending an ata port.

One patch per subsystem please. I think this patch can be split easily into an ATA patch and a SCSI core patch.

In general, I agree that should be done. But this is a bug fix and having it split in 2 risks breaking something if only one is reverted and also potentially give bad bisect results. So I would rather not do that.

Quoted 4 lines

Thanks,

Bart.

Signature · 2 lines
Damien Le Moal
Western Digital Research
author maintainer
#32 lore raw

PATCH 6/19 ata: libata-core: Do not register PM operations for SAS ports

In reply to Damien Le Moal

libsas does its own domain based power management of ports. For such ports, libata should not use a device type defining power management operations as executing these operations for suspend/resume in addition to libsas calls to ata_sas_port_suspend() and ata_sas_port_resume() is not necessary (and likely dangerous to do, even though problems are not seen currently).

Introduce the new ata_port_sas_type device_type for ports managed by libsas. This new device type is used in ata_tport_add() and is defined without power management operations.

Fixes 2fcbdcb4c802 ("[SCSI] libata: export ata_port suspend/resume infrastructure for sas")
---
 drivers/ata/libata-core.c      | 2 +-
 drivers/ata/libata-transport.c | 9 ++++++++-
 drivers/ata/libata.h           | 2 ++
 3 files changed, 11 insertions(+), 2 deletions(-)
drivers/ata/libata-core.c +1 -1
index 0479493e54bd..18b2a0da9e54 100644
@@ -5339,7 +5339,7 @@ EXPORT_SYMBOL_GPL(ata_host_resume);
 #endif
 
 const struct device_type ata_port_type = {
-	.name = "ata_port",
+	.name = ATA_PORT_TYPE_NAME,
 #ifdef CONFIG_PM
 	.pm = &ata_port_pm_ops,
 #endif
drivers/ata/libata-transport.c +8 -1
index e4fb9d1b9b39..3e49a877500e 100644
@@ -266,6 +266,10 @@ void ata_tport_delete(struct ata_port *ap)
 	put_device(dev);
 }
 
+static const struct device_type ata_port_sas_type = {
+	.name = ATA_PORT_TYPE_NAME,
+};
+
 /** ata_tport_add - initialize a transport ATA port structure
  *
  * @parent:	parent device
@@ -283,7 +287,10 @@ int ata_tport_add(struct device *parent,
 	struct device *dev = &ap->tdev;
 
 	device_initialize(dev);
-	dev->type = &ata_port_type;
+	if (ap->flags & ATA_FLAG_SAS_HOST)
+		dev->type = &ata_port_sas_type;
+	else
+		dev->type = &ata_port_type;
 
 	dev->parent = parent;
 	ata_host_get(ap->host);
drivers/ata/libata.h +2 -0
index a5ee06f0234a..c57e094c3af9 100644
@@ -30,6 +30,8 @@ enum {
 	ATA_DNXFER_QUIET	= (1 << 31),
 };
 
+#define ATA_PORT_TYPE_NAME	"ata_port"
+
 extern atomic_t ata_print_id;
 extern int atapi_passthru16;
 extern int libata_fua;
Signature · 1 line
2.41.0
Reviewed-by on v1 6/19
#33 lore raw

Re: [PATCH 06/19] ata: libata-core: Do not register PM operations for SAS ports

Quoted 11 lines · Damien Le Moal

libsas does its own domain based power management of ports. For such ports, libata should not use a device type defining power management operations as executing these operations for suspend/resume in addition to libsas calls to ata_sas_port_suspend() and ata_sas_port_resume() is not necessary (and likely dangerous to do, even though problems are not seen currently).

Introduce the new ata_port_sas_type device_type for ports managed by libsas. This new device type is used in ata_tport_add() and is defined without power management operations.

Quoted 9 lines
Fixes 2fcbdcb4c802 ("[SCSI] libata: export ata_port suspend/resume infrastructure for sas")
---
  drivers/ata/libata-core.c      | 2 +-
  drivers/ata/libata-transport.c | 9 ++++++++-
  drivers/ata/libata.h           | 2 ++
  3 files changed, 11 insertions(+), 2 deletions(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
Tested-by on v1 6/19
#34 lore raw

Re: [PATCH 06/19] ata: libata-core: Do not register PM operations for SAS ports

In reply to Damien Le Moal
Quoted 7 lines · Damien Le Moal

libsas does its own domain based power management of ports. For such ports, libata should not use a device type defining power management operations as executing these operations for suspend/resume in addition to libsas calls to ata_sas_port_suspend() and ata_sas_port_resume() is not necessary (and likely dangerous to do, even though problems are not seen currently).

Quoted 7 lines

Introduce the new ata_port_sas_type device_type for ports managed by libsas. This new device type is used in ata_tport_add() and is defined without power management operations.

Fixes 2fcbdcb4c802 ("[SCSI] libata: export ata_port suspend/resume infrastructure for sas")
author maintainer
#35 lore raw

PATCH 7/19 scsi: sd: Do not issue commands to suspended disks on remove

In reply to Damien Le Moal

If an error occurs when resuming a host adapter before the devices attached to the adapter are resumed, the adapter low level driver may remove the scsi host, resulting in a call to sd_remove() for the disks of the host. However, since this function calls sd_shutdown(), a synchronize cache command and a start stop unit may be issued with the drive still sleeping and the HBA non-functional. This causes PM resume to hang, forcing a reset of the machine to recover.

Fix this by checking a device host state in sd_shutdown() and by returning early doing nothing if the host state is not SHOST_RUNNING.

---
 drivers/scsi/sd.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/sd.c +2 -1
index c92a317ba547..a415abb721d3 100644
@@ -3763,7 +3763,8 @@ static void sd_shutdown(struct device *dev)
 	if (!sdkp)
 		return;         /* this can happen */
 
-	if (pm_runtime_suspended(dev))
+	if (pm_runtime_suspended(dev) ||
+	    sdkp->device->host->shost_state != SHOST_RUNNING)
 		return;
 
 	if (sdkp->WCE && sdkp->media_present) {
Signature · 1 line
2.41.0
Reviewed-by on v1 7/19
#36 lore raw

Re: [PATCH 07/19] scsi: sd: Do not issue commands to suspended disks on remove

Quoted 22 lines · Damien Le Moal

If an error occurs when resuming a host adapter before the devices attached to the adapter are resumed, the adapter low level driver may remove the scsi host, resulting in a call to sd_remove() for the disks of the host. However, since this function calls sd_shutdown(), a synchronize cache command and a start stop unit may be issued with the drive still sleeping and the HBA non-functional. This causes PM resume to hang, forcing a reset of the machine to recover.

Fix this by checking a device host state in sd_shutdown() and by returning early doing nothing if the host state is not SHOST_RUNNING.

---
  drivers/scsi/sd.c | 3 ++-
  1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/sd.c +0 -0
index c92a317ba547..a415abb721d3 100644

drivers/scsi/sd.c static void sd_shutdown(struct device *dev) · line 3770

Quoted 9 lines
drivers/scsi/sd.c +2 -1
@@ -3763,7 +3763,8 @@ static void sd_shutdown(struct device *dev)
  	if (!sdkp)
  		return;         /* this can happen */
 
-	if (pm_runtime_suspended(dev))
+	if (pm_runtime_suspended(dev) ||
+	    sdkp->device->host->shost_state != SHOST_RUNNING)
  		return;
 
  	if (sdkp->WCE && sdkp->media_present) {

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
McDonald, Martje Boudien Moerman
Tested-by on v1 7/19
#37 lore raw

Re: [PATCH 07/19] scsi: sd: Do not issue commands to suspended disks on remove

In reply to Damien Le Moal
Quoted 8 lines · Damien Le Moal

If an error occurs when resuming a host adapter before the devices attached to the adapter are resumed, the adapter low level driver may remove the scsi host, resulting in a call to sd_remove() for the disks of the host. However, since this function calls sd_shutdown(), a synchronize cache command and a start stop unit may be issued with the drive still sleeping and the HBA non-functional. This causes PM resume to hang, forcing a reset of the machine to recover.

Quoted 5 lines

Fix this by checking a device host state in sd_shutdown() and by returning early doing nothing if the host state is not SHOST_RUNNING.

#38 lore raw

Re: [PATCH 07/19] scsi: sd: Do not issue commands to suspended disks on remove

In reply to Damien Le Moal
Quoted 22 lines · Damien Le Moal

If an error occurs when resuming a host adapter before the devices attached to the adapter are resumed, the adapter low level driver may remove the scsi host, resulting in a call to sd_remove() for the disks of the host. However, since this function calls sd_shutdown(), a synchronize cache command and a start stop unit may be issued with the drive still sleeping and the HBA non-functional. This causes PM resume to hang, forcing a reset of the machine to recover.

Fix this by checking a device host state in sd_shutdown() and by returning early doing nothing if the host state is not SHOST_RUNNING.

---
  drivers/scsi/sd.c | 3 ++-
  1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/sd.c +0 -0
index c92a317ba547..a415abb721d3 100644

drivers/scsi/sd.c static void sd_shutdown(struct device *dev) · line 3770

Quoted 9 lines
drivers/scsi/sd.c +2 -1
@@ -3763,7 +3763,8 @@ static void sd_shutdown(struct device *dev)
  	if (!sdkp)
  		return;         /* this can happen */
 
-	if (pm_runtime_suspended(dev))
+	if (pm_runtime_suspended(dev) ||
+	    sdkp->device->host->shost_state != SHOST_RUNNING)
  		return;
 
  	if (sdkp->WCE && sdkp->media_present) {

Why to test the host state instead of dev->power.runtime_status? I don't think that it is safe to skip shutdown if the error handler is active. If the error handler can recover the device a SYNCHRONIZE CACHE command should be submitted.

Thanks,

Bart.

author maintainer
#39 lore raw

Re: [PATCH 07/19] scsi: sd: Do not issue commands to suspended disks on remove

Quoted 31 lines · Bart Van Assche
Quoted 30 lines · Damien Le Moal

If an error occurs when resuming a host adapter before the devices attached to the adapter are resumed, the adapter low level driver may remove the scsi host, resulting in a call to sd_remove() for the disks of the host. However, since this function calls sd_shutdown(), a synchronize cache command and a start stop unit may be issued with the drive still sleeping and the HBA non-functional. This causes PM resume to hang, forcing a reset of the machine to recover.

Fix this by checking a device host state in sd_shutdown() and by returning early doing nothing if the host state is not SHOST_RUNNING.

---
  drivers/scsi/sd.c | 3 ++-
  1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/sd.c +2 -1
index c92a317ba547..a415abb721d3 100644
@@ -3763,7 +3763,8 @@ static void sd_shutdown(struct device *dev)
  	if (!sdkp)
  		return;         /* this can happen */
 
-	if (pm_runtime_suspended(dev))
+	if (pm_runtime_suspended(dev) ||
+	    sdkp->device->host->shost_state != SHOST_RUNNING)
  		return;
 
Quoted 6 lines

drivers/scsi/sd.c static void sd_shutdown(struct device *dev) · line 3770

Quoted 1 line
drivers/scsi/sd.c +0 -0
  	if (sdkp->WCE && sdkp->media_present) {

Why to test the host state instead of dev->power.runtime_status? I don't think that it is safe to skip shutdown if the error handler is active. If the error handler can recover the device a SYNCHRONIZE CACHE command should be submitted.

But there is no synchronization with EH that I can see anyway. At least for sd_remove(), I would assume that this is called only once the device references were all dropped, so presumably EH is not doing anything with the drive when that happen, no ?

In any case, looking at dev->power.runtime_status is not correct as this is set to RPM_ACTIVE when the device is suspended through system suspend. We could replace the test "sdkp->device->host->shost_state != SHOST_RUNNING" with "dev->power.is_suspended", as that indicates true (1) for a suspended device. However, I really do not like that as that is a PM internal field and should not be accessing it directly. The PM code comments say as much. Any better idea ?

Signature · 2 lines
Damien Le Moal
Western Digital Research
#40 lore raw

Re: [PATCH 07/19] scsi: sd: Do not issue commands to suspended disks on remove

Quoted 44 lines · Damien Le Moal
Quoted 31 lines · Bart Van Assche
Quoted 30 lines · Damien Le Moal

If an error occurs when resuming a host adapter before the devices attached to the adapter are resumed, the adapter low level driver may remove the scsi host, resulting in a call to sd_remove() for the disks of the host. However, since this function calls sd_shutdown(), a synchronize cache command and a start stop unit may be issued with the drive still sleeping and the HBA non-functional. This causes PM resume to hang, forcing a reset of the machine to recover.

Fix this by checking a device host state in sd_shutdown() and by returning early doing nothing if the host state is not SHOST_RUNNING.

---
   drivers/scsi/sd.c | 3 ++-
   1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/sd.c +2 -1
index c92a317ba547..a415abb721d3 100644
@@ -3763,7 +3763,8 @@ static void sd_shutdown(struct device *dev)
   	if (!sdkp)
   		return;         /* this can happen */
 
-	if (pm_runtime_suspended(dev))
+	if (pm_runtime_suspended(dev) ||
+	    sdkp->device->host->shost_state != SHOST_RUNNING)
   		return;
 
Quoted 6 lines

drivers/scsi/sd.c static void sd_shutdown(struct device *dev) · line 3770

Quoted 1 line
drivers/scsi/sd.c +0 -0
   	if (sdkp->WCE && sdkp->media_present) {

Why to test the host state instead of dev->power.runtime_status? I don't think that it is safe to skip shutdown if the error handler is active. If the error handler can recover the device a SYNCHRONIZE CACHE command should be submitted.

But there is no synchronization with EH that I can see anyway. At least for sd_remove(), I would assume that this is called only once the device references were all dropped, so presumably EH is not doing anything with the drive when that happen, no ?

Quoted 6 lines

In any case, looking at dev->power.runtime_status is not correct as this is set to RPM_ACTIVE when the device is suspended through system suspend. We could replace the test "sdkp->device->host->shost_state != SHOST_RUNNING" with "dev->power.is_suspended", as that indicates true (1) for a suspended device. However, I really do not like that as that is a PM internal field and should not be accessing it directly. The PM code comments say as much. Any better idea ?

I will reply to the above question on v2 of this patch.

Bart.

author maintainer
#41 lore raw

PATCH 8/19 scsi: Remove scsi device no_start_on_resume flag

In reply to Damien Le Moal

The scsi device flag no_start_on_resume is not set by any scsi low level driver. Remove it.

---
 drivers/scsi/sd.c          | 7 ++-----
 include/scsi/scsi_device.h | 1 -
 2 files changed, 2 insertions(+), 6 deletions(-)
drivers/scsi/sd.c +2 -5
index a415abb721d3..b9504bb3cf4d 100644
@@ -3846,11 +3846,8 @@ static int sd_resume(struct device *dev)
 	if (!sdkp->device->manage_start_stop)
 		return 0;
 
-	if (!sdkp->device->no_start_on_resume) {
-		sd_printk(KERN_NOTICE, sdkp, "Starting disk\n");
-		ret = sd_start_stop_device(sdkp, 1);
-	}
-
+	sd_printk(KERN_NOTICE, sdkp, "Starting disk\n");
+	ret = sd_start_stop_device(sdkp, 1);
 	if (!ret)
 		opal_unlock_from_suspend(sdkp->opal_dev);
 	return ret;
include/scsi/scsi_device.h +0 -1
index b9230b6add04..75b2235b99e2 100644
@@ -194,7 +194,6 @@ struct scsi_device {
 	unsigned no_start_on_add:1;	/* do not issue start on add */
 	unsigned allow_restart:1; /* issue START_UNIT in error handler */
 	unsigned manage_start_stop:1;	/* Let HLD (sd) manage start/stop */
-	unsigned no_start_on_resume:1; /* Do not issue START_STOP_UNIT on resume */
 	unsigned start_stop_pwr_cond:1;	/* Set power cond. in START_STOP_UNIT */
 	unsigned no_uld_attach:1; /* disable connecting to upper level drivers */
 	unsigned select_no_atn:1;
Signature · 1 line
2.41.0
Reviewed-by on v1 8/19
#42 lore raw

Re: [PATCH 08/19] scsi: Remove scsi device no_start_on_resume flag

Quoted 9 lines · Damien Le Moal

The scsi device flag no_start_on_resume is not set by any scsi low level driver. Remove it.

---
  drivers/scsi/sd.c          | 7 ++-----
  include/scsi/scsi_device.h | 1 -
  2 files changed, 2 insertions(+), 6 deletions(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
#44 lore raw

Re: [PATCH 08/19] scsi: Remove scsi device no_start_on_resume flag

In reply to Damien Le Moal
Quoted 2 lines · Damien Le Moal

The scsi device flag no_start_on_resume is not set by any scsi low level driver. Remove it.

Please consider mentioning that this patch reverts commit 0a8589055936 ("ata,scsi: do not issue START STOP UNIT on resume").

Thanks,

Bart.

author maintainer
#45 lore raw

PATCH 9/19 ata: libata-scsi: Cleanup ata_scsi_start_stop_xlat()

In reply to Damien Le Moal

Now that libata does its own internal device power mode management through libata EH, the scsi disk driver will not issue START STOP UNIT commands anymore. We can receive this command only from user passthrough operations. So there is no need to consider the system state and ATA port flags for suspend to translate the command.

Since setting up the taskfile for the verify and standby immediate commands is the same as done in ata_dev_power_set_active() and ata_dev_power_set_standby(), factor out this code into the helper function ata_dev_power_init_tf() to simplify ata_scsi_start_stop_xlat() as well as ata_dev_power_set_active() and ata_dev_power_set_standby().

---
 drivers/ata/libata-core.c | 55 +++++++++++++++++++++++----------------
 drivers/ata/libata-scsi.c | 53 +++++++------------------------------
 drivers/ata/libata.h      |  2 ++
 3 files changed, 44 insertions(+), 66 deletions(-)
drivers/ata/libata-core.c +33 -22
index 18b2a0da9e54..d20ad3fc71cf 100644
@@ -1972,6 +1972,35 @@ int ata_dev_read_id(struct ata_device *dev, unsigned int *p_class,
 	return rc;
 }
 
+bool ata_dev_power_init_tf(struct ata_device *dev, struct ata_taskfile *tf,
+			   bool set_active)
+{
+	/* Only applies to ATA and ZAC devices */
+	if (dev->class != ATA_DEV_ATA && dev->class != ATA_DEV_ZAC)
+		return false;
+
+	ata_tf_init(dev, tf);
+	tf->flags |= ATA_TFLAG_DEVICE | ATA_TFLAG_ISADDR;
+	tf->protocol = ATA_PROT_NODATA;
+
+	if (set_active) {
+		/* VERIFY for 1 sector at lba=0 */
+		tf->command = ATA_CMD_VERIFY;
+		tf->nsect = 1;
+		if (dev->flags & ATA_DFLAG_LBA) {
+			tf->flags |= ATA_TFLAG_LBA;
+			tf->device |= ATA_LBA;
+		} else {
+			/* CHS */
+			tf->lbal = 0x1; /* sect */
+		}
+	} else {
+		tf->command = ATA_CMD_STANDBYNOW1;
+	}
+
+	return true;
+}
+
 /**
  *	ata_dev_power_set_standby - Set a device power mode to standby
  *	@dev: target device
@@ -1988,10 +2017,6 @@ void ata_dev_power_set_standby(struct ata_device *dev)
 	struct ata_taskfile tf;
 	unsigned int err_mask;
 
-	/* Issue STANDBY IMMEDIATE command only if supported by the device */
-	if (dev->class != ATA_DEV_ATA && dev->class != ATA_DEV_ZAC)
-		return;
-
 	/*
 	 * Some odd clown BIOSes issue spindown on power off (ACPI S4 or S5)
 	 * causing some drives to spin up and down again. For these, do nothing
@@ -2005,10 +2030,9 @@ void ata_dev_power_set_standby(struct ata_device *dev)
 	    system_entering_hibernation())
 		return;
 
-	ata_tf_init(dev, &tf);
-	tf.flags |= ATA_TFLAG_DEVICE | ATA_TFLAG_ISADDR;
-	tf.protocol = ATA_PROT_NODATA;
-	tf.command = ATA_CMD_STANDBYNOW1;
+	/* Issue STANDBY IMMEDIATE command only if supported by the device */
+	if (!ata_dev_power_init_tf(dev, &tf, false))
+		return;
 
 	ata_dev_notice(dev, "Entering standby power mode\n");
 
@@ -2038,22 +2062,9 @@ void ata_dev_power_set_active(struct ata_device *dev)
 	 * Issue READ VERIFY SECTORS command for 1 sector at lba=0 only
 	 * if supported by the device.
 	 */
-	if (dev->class != ATA_DEV_ATA && dev->class != ATA_DEV_ZAC)
+	if (!ata_dev_power_init_tf(dev, &tf, true))
 		return;
 
-	ata_tf_init(dev, &tf);
-	tf.flags |= ATA_TFLAG_DEVICE | ATA_TFLAG_ISADDR;
-	tf.protocol = ATA_PROT_NODATA;
-	tf.command = ATA_CMD_VERIFY;
-	tf.nsect = 1;
-	if (dev->flags & ATA_DFLAG_LBA) {
-		tf.flags |= ATA_TFLAG_LBA;
-		tf.device |= ATA_LBA;
-	} else {
-		/* CHS */
-		tf.lbal = 0x1; /* sect */
-	}
-
 	ata_dev_notice(dev, "Entering active power mode\n");
 
 	err_mask = ata_exec_internal(dev, &tf, NULL, DMA_NONE, NULL, 0, 0);
drivers/ata/libata-scsi.c +9 -44

82 lines not on this page · raw message

drivers/ata/libata.h +2 -0
index c57e094c3af9..af0e718f2b72 100644
@@ -62,6 +62,8 @@ extern int ata_dev_reread_id(struct ata_device *dev, unsigned int readid_flags);
 extern int ata_dev_revalidate(struct ata_device *dev, unsigned int new_class,
 			      unsigned int readid_flags);
 extern int ata_dev_configure(struct ata_device *dev);
+extern bool ata_dev_power_init_tf(struct ata_device *dev,
+				  struct ata_taskfile *tf, bool set_active);
 extern void ata_dev_power_set_standby(struct ata_device *dev);
 extern void ata_dev_power_set_active(struct ata_device *dev);
 extern int sata_down_spd_limit(struct ata_link *link, u32 spd_limit);
Signature · 1 line
2.41.0
Reviewed-by on v1 9/19
#46 lore raw

Re: [PATCH 09/19] ata: libata-scsi: Cleanup ata_scsi_start_stop_xlat()

Quoted 12 lines · Damien Le Moal

Now that libata does its own internal device power mode management through libata EH, the scsi disk driver will not issue START STOP UNIT commands anymore. We can receive this command only from user passthrough operations. So there is no need to consider the system state and ATA port flags for suspend to translate the command.

Since setting up the taskfile for the verify and standby immediate commands is the same as done in ata_dev_power_set_active() and ata_dev_power_set_standby(), factor out this code into the helper function ata_dev_power_init_tf() to simplify ata_scsi_start_stop_xlat() as well as ata_dev_power_set_active() and ata_dev_power_set_standby().

Quoted 7 lines
---
  drivers/ata/libata-core.c | 55 +++++++++++++++++++++++----------------
  drivers/ata/libata-scsi.c | 53 +++++++------------------------------
  drivers/ata/libata.h      |  2 ++
  3 files changed, 44 insertions(+), 66 deletions(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
Tested-by on v1 9/19
#47 lore raw

Re: [PATCH 09/19] ata: libata-scsi: Cleanup ata_scsi_start_stop_xlat()

In reply to Damien Le Moal
Quoted 6 lines · Damien Le Moal

Now that libata does its own internal device power mode management through libata EH, the scsi disk driver will not issue START STOP UNIT commands anymore. We can receive this command only from user passthrough operations. So there is no need to consider the system state and ATA port flags for suspend to translate the command.

Quoted 7 lines

Since setting up the taskfile for the verify and standby immediate commands is the same as done in ata_dev_power_set_active() and ata_dev_power_set_standby(), factor out this code into the helper function ata_dev_power_init_tf() to simplify ata_scsi_start_stop_xlat() as well as ata_dev_power_set_active() and ata_dev_power_set_standby().

author maintainer
#48 lore raw

PATCH 10/19 ata: libata-core: Synchronize ata_port_detach() with hotplug

In reply to Damien Le Moal

The call to async_synchronize_cookie() to synchronize a port removal and hotplug probe is done in ata_host_detach() right before calling ata_port_detach(). Move this call at the beginning of ata_port_detach() to ensure that this operation is always synchronized with probe.

---
 drivers/ata/libata-core.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/ata/libata-core.c +4 -4
index d20ad3fc71cf..fafd5a82c5ea 100644
@@ -6069,6 +6069,9 @@ static void ata_port_detach(struct ata_port *ap)
 	struct ata_link *link;
 	struct ata_device *dev;
 
+	/* Ensure ata_port probe has completed */
+	async_synchronize_cookie(ap->cookie + 1);
+
 	/* Wait for any ongoing EH */
 	ata_port_wait_eh(ap);
 
@@ -6133,11 +6136,8 @@ void ata_host_detach(struct ata_host *host)
 {
 	int i;
 
-	for (i = 0; i < host->n_ports; i++) {
-		/* Ensure ata_port probe has completed */
-		async_synchronize_cookie(host->ports[i]->cookie + 1);
+	for (i = 0; i < host->n_ports; i++)
 		ata_port_detach(host->ports[i]);
-	}
 
 	/* the host is dead now, dissociate ACPI */
 	ata_acpi_dissociate(host);
Signature · 1 line
2.41.0
Reviewed-by on v1 10/19
#49 lore raw

Re: [PATCH 10/19] ata: libata-core: Synchronize ata_port_detach() with hotplug

Quoted 10 lines · Damien Le Moal

The call to async_synchronize_cookie() to synchronize a port removal and hotplug probe is done in ata_host_detach() right before calling ata_port_detach(). Move this call at the beginning of ata_port_detach() to ensure that this operation is always synchronized with probe.

---
  drivers/ata/libata-core.c | 8 ++++----
  1 file changed, 4 insertions(+), 4 deletions(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
Tested-by on v1 10/19
#50 lore raw

Re: [PATCH 10/19] ata: libata-core: Synchronize ata_port_detach() with hotplug

In reply to Damien Le Moal
Quoted 6 lines · Damien Le Moal

The call to async_synchronize_cookie() to synchronize a port removal and hotplug probe is done in ata_host_detach() right before calling ata_port_detach(). Move this call at the beginning of ata_port_detach() to ensure that this operation is always synchronized with probe.

author maintainer
#51 lore raw

PATCH 11/19 ata: libata-core: Detach a port devices on shutdown

In reply to Damien Le Moal

Modify ata_pci_shutdown_one() to schedule EH to unload a port devices before freezing and thawing the port. This ensures that drives are cleanly disabled and transitioned to standby power mode when a PCI adapter is removed or the system is powered off.

---
 drivers/ata/libata-core.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)
drivers/ata/libata-core.c +15 -1
index fafd5a82c5ea..fa70d0ef92e5 100644
@@ -6168,10 +6168,24 @@ EXPORT_SYMBOL_GPL(ata_pci_remove_one);
 void ata_pci_shutdown_one(struct pci_dev *pdev)
 {
 	struct ata_host *host = pci_get_drvdata(pdev);
+	struct ata_port *ap;
+	unsigned long flags;
 	int i;
 
+	/* Tell EH to disable all devices */
 	for (i = 0; i < host->n_ports; i++) {
-		struct ata_port *ap = host->ports[i];
+		ap = host->ports[i];
+		spin_lock_irqsave(ap->lock, flags);
+		ap->pflags |= ATA_PFLAG_UNLOADING;
+		ata_port_schedule_eh(ap);
+		spin_unlock_irqrestore(ap->lock, flags);
+	}
+
+	for (i = 0; i < host->n_ports; i++) {
+		ap = host->ports[i];
+
+		/* Wait for EH to complete before freezing the port */
+		ata_port_wait_eh(ap);
 
 		ap->pflags |= ATA_PFLAG_FROZEN;
 
Signature · 1 line
2.41.0
Reviewed-by on v1 11/19
#52 lore raw

Re: [PATCH 11/19] ata: libata-core: Detach a port devices on shutdown

Quoted 10 lines · Damien Le Moal

Modify ata_pci_shutdown_one() to schedule EH to unload a port devices before freezing and thawing the port. This ensures that drives are cleanly disabled and transitioned to standby power mode when a PCI adapter is removed or the system is powered off.

---
  drivers/ata/libata-core.c | 16 +++++++++++++++-
  1 file changed, 15 insertions(+), 1 deletion(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
Tested-by on v1 11/19
#53 lore raw

Re: [PATCH 11/19] ata: libata-core: Detach a port devices on shutdown

In reply to Damien Le Moal
Quoted 6 lines · Damien Le Moal

Modify ata_pci_shutdown_one() to schedule EH to unload a port devices before freezing and thawing the port. This ensures that drives are cleanly disabled and transitioned to standby power mode when a PCI adapter is removed or the system is powered off.

author maintainer
#54 lore raw

PATCH 12/19 ata: libata-core: Remove ata_port_suspend_async()

In reply to Damien Le Moal

ata_port_suspend_async() is only called by ata_sas_port_suspend(). Modify ata_port_suspend() with an additional bool argument indicating an asynchronous or synchronous suspend to allow removing that helper function. With this change, the variable ata_port_resume_ehi can also be removed and its value (ATA_EHI_XXX flags passed directly to ata_port_request_pm().

---
 drivers/ata/libata-core.c | 46 +++++++++++++++------------------------
 1 file changed, 17 insertions(+), 29 deletions(-)
drivers/ata/libata-core.c +17 -29
index fa70d0ef92e5..6fd9ea2b210d 100644
@@ -5170,18 +5170,8 @@ static void ata_port_request_pm(struct ata_port *ap, pm_message_t mesg,
 		ata_port_wait_eh(ap);
 }
 
-/*
- * On some hardware, device fails to respond after spun down for suspend.  As
- * the device won't be used before being resumed, we don't need to touch the
- * device.  Ask EH to skip the usual stuff and proceed directly to suspend.
- *
- * http://thread.gmane.org/gmane.linux.ide/46764
- */
-static const unsigned int ata_port_suspend_ehi = ATA_EHI_QUIET
-						 | ATA_EHI_NO_AUTOPSY
-						 | ATA_EHI_NO_RECOVERY;
-
-static void ata_port_suspend(struct ata_port *ap, pm_message_t mesg)
+static void ata_port_suspend(struct ata_port *ap, pm_message_t mesg,
+			     bool async)
 {
 	/*
 	 * We are about to suspend the port, so we do not care about
@@ -5191,20 +5181,18 @@ static void ata_port_suspend(struct ata_port *ap, pm_message_t mesg)
 	 */
 	cancel_delayed_work_sync(&ap->scsi_rescan_task);
 
-	ata_port_request_pm(ap, mesg, 0, ata_port_suspend_ehi, false);
-}
-
-static void ata_port_suspend_async(struct ata_port *ap, pm_message_t mesg)
-{
 	/*
-	 * We are about to suspend the port, so we do not care about
-	 * scsi_rescan_device() calls scheduled by previous resume operations.
-	 * The next resume will schedule the rescan again. So cancel any rescan
-	 * that is not done yet.
+	 * On some hardware, device fails to respond after spun down for
+	 * suspend. As the device wil not t be used until being resumed, we
+	 * do not need to touch the device. Ask EH to skip the usual stuff
+	 * and proceed directly to suspend.
+	 *
+	 * http://thread.gmane.org/gmane.linux.ide/46764
 	 */
-	cancel_delayed_work_sync(&ap->scsi_rescan_task);
-
-	ata_port_request_pm(ap, mesg, 0, ata_port_suspend_ehi, true);
+	ata_port_request_pm(ap, mesg, 0,
+			    ATA_EHI_QUIET | ATA_EHI_NO_AUTOPSY |
+			    ATA_EHI_NO_RECOVERY,
+			    async);
 }
 
 static int ata_port_pm_suspend(struct device *dev)
@@ -5214,7 +5202,7 @@ static int ata_port_pm_suspend(struct device *dev)
 	if (pm_runtime_suspended(dev))
 		return 0;
 
-	ata_port_suspend(ap, PMSG_SUSPEND);
+	ata_port_suspend(ap, PMSG_SUSPEND, false);
 	return 0;
 }
 
@@ -5225,13 +5213,13 @@ static int ata_port_pm_freeze(struct device *dev)
 	if (pm_runtime_suspended(dev))
 		return 0;
 
-	ata_port_suspend(ap, PMSG_FREEZE);
+	ata_port_suspend(ap, PMSG_FREEZE, false);
 	return 0;
 }
 
 static int ata_port_pm_poweroff(struct device *dev)
 {
-	ata_port_suspend(to_ata_port(dev), PMSG_HIBERNATE);
+	ata_port_suspend(to_ata_port(dev), PMSG_HIBERNATE, false);
 	return 0;
 }
 
@@ -5283,7 +5271,7 @@ static int ata_port_runtime_idle(struct device *dev)
 
 static int ata_port_runtime_suspend(struct device *dev)
 {
-	ata_port_suspend(to_ata_port(dev), PMSG_AUTO_SUSPEND);
+	ata_port_suspend(to_ata_port(dev), PMSG_AUTO_SUSPEND, false);
 	return 0;
 }
 
@@ -5313,7 +5301,7 @@ static const struct dev_pm_ops ata_port_pm_ops = {
  */
 void ata_sas_port_suspend(struct ata_port *ap)
 {
-	ata_port_suspend_async(ap, PMSG_SUSPEND);
+	ata_port_suspend(ap, PMSG_SUSPEND, true);
 }
 EXPORT_SYMBOL_GPL(ata_sas_port_suspend);
 
Signature · 1 line
2.41.0
Reviewed-by on v1 12/19
#55 lore raw

Re: [PATCH 12/19] ata: libata-core: Remove ata_port_suspend_async()

Quoted 7 lines · Damien Le Moal

ata_port_suspend_async() is only called by ata_sas_port_suspend(). Modify ata_port_suspend() with an additional bool argument indicating an asynchronous or synchronous suspend to allow removing that helper function. With this change, the variable ata_port_resume_ehi can also be removed and its value (ATA_EHI_XXX flags passed directly to ata_port_request_pm().

Quoted 5 lines
---
  drivers/ata/libata-core.c | 46 +++++++++++++++------------------------
  1 file changed, 17 insertions(+), 29 deletions(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
Tested-by on v1 12/19
#56 lore raw

Re: [PATCH 12/19] ata: libata-core: Remove ata_port_suspend_async()

In reply to Damien Le Moal
Quoted 8 lines · Damien Le Moal

ata_port_suspend_async() is only called by ata_sas_port_suspend(). Modify ata_port_suspend() with an additional bool argument indicating an asynchronous or synchronous suspend to allow removing that helper function. With this change, the variable ata_port_resume_ehi can also be removed and its value (ATA_EHI_XXX flags passed directly to ata_port_request_pm().

author maintainer
#57 lore raw

PATCH 13/19 ata: libata-core: Remove ata_port_resume_async()

In reply to Damien Le Moal

Remove ata_port_resume_async() and replace it with a modified ata_port_resume() taking an additional bool argument indicating if ata EH resume operation should be executed synchronously or asynchronously. With this change, the variable ata_port_resume_ehi is not longer necessary and its value (ATA_EHI_XXX flags) passed directly to ata_port_request_pm().

---
 drivers/ata/libata-core.c | 21 ++++++++-------------
 1 file changed, 8 insertions(+), 13 deletions(-)
drivers/ata/libata-core.c +8 -13
index 6fd9ea2b210d..8fa5fbae14f3 100644
@@ -5223,22 +5223,17 @@ static int ata_port_pm_poweroff(struct device *dev)
 	return 0;
 }
 
-static const unsigned int ata_port_resume_ehi = ATA_EHI_NO_AUTOPSY
-						| ATA_EHI_QUIET;
-
-static void ata_port_resume(struct ata_port *ap, pm_message_t mesg)
+static void ata_port_resume(struct ata_port *ap, pm_message_t mesg,
+			    bool async)
 {
-	ata_port_request_pm(ap, mesg, ATA_EH_RESET, ata_port_resume_ehi, false);
-}
-
-static void ata_port_resume_async(struct ata_port *ap, pm_message_t mesg)
-{
-	ata_port_request_pm(ap, mesg, ATA_EH_RESET, ata_port_resume_ehi, true);
+	ata_port_request_pm(ap, mesg, ATA_EH_RESET,
+			    ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET,
+			    async);
 }
 
 static int ata_port_pm_resume(struct device *dev)
 {
-	ata_port_resume_async(to_ata_port(dev), PMSG_RESUME);
+	ata_port_resume(to_ata_port(dev), PMSG_RESUME, true);
 	pm_runtime_disable(dev);
 	pm_runtime_set_active(dev);
 	pm_runtime_enable(dev);
@@ -5277,7 +5272,7 @@ static int ata_port_runtime_suspend(struct device *dev)
 
 static int ata_port_runtime_resume(struct device *dev)
 {
-	ata_port_resume(to_ata_port(dev), PMSG_AUTO_RESUME);
+	ata_port_resume(to_ata_port(dev), PMSG_AUTO_RESUME, false);
 	return 0;
 }
 
@@ -5307,7 +5302,7 @@ EXPORT_SYMBOL_GPL(ata_sas_port_suspend);
 
 void ata_sas_port_resume(struct ata_port *ap)
 {
-	ata_port_resume_async(ap, PMSG_RESUME);
+	ata_port_resume(ap, PMSG_RESUME, true);
 }
 EXPORT_SYMBOL_GPL(ata_sas_port_resume);
 
Signature · 1 line
2.41.0
Reviewed-by on v1 13/19
#58 lore raw

Re: [PATCH 13/19] ata: libata-core: Remove ata_port_resume_async()

Quoted 7 lines · Damien Le Moal

Remove ata_port_resume_async() and replace it with a modified ata_port_resume() taking an additional bool argument indicating if ata EH resume operation should be executed synchronously or asynchronously. With this change, the variable ata_port_resume_ehi is not longer necessary and its value (ATA_EHI_XXX flags) passed directly to ata_port_request_pm().

Quoted 5 lines
---
  drivers/ata/libata-core.c | 21 ++++++++-------------
  1 file changed, 8 insertions(+), 13 deletions(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
Tested-by on v1 13/19
#59 lore raw

Re: [PATCH 13/19] ata: libata-core: Remove ata_port_resume_async()

In reply to Damien Le Moal
Quoted 8 lines · Damien Le Moal

Remove ata_port_resume_async() and replace it with a modified ata_port_resume() taking an additional bool argument indicating if ata EH resume operation should be executed synchronously or asynchronously. With this change, the variable ata_port_resume_ehi is not longer necessary and its value (ATA_EHI_XXX flags) passed directly to ata_port_request_pm().

author maintainer
#60 lore raw

PATCH 14/19 ata: libata-core: skip poweroff for devices that are runtime suspended

In reply to Damien Le Moal

When powering off, there is no need to suspend a port that has already been runtime suspended. Skip the EH PM request in ata_port_pm_poweroff() in this case.

---
 drivers/ata/libata-core.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/ata/libata-core.c +2 -1
index 8fa5fbae14f3..c4a32abc2e29 100644
@@ -5219,7 +5219,8 @@ static int ata_port_pm_freeze(struct device *dev)
 
 static int ata_port_pm_poweroff(struct device *dev)
 {
-	ata_port_suspend(to_ata_port(dev), PMSG_HIBERNATE, false);
+	if (!pm_runtime_suspended(dev))
+		ata_port_suspend(to_ata_port(dev), PMSG_HIBERNATE, false);
 	return 0;
 }
 
Signature · 1 line
2.41.0
Reviewed-by on v1 14/19
#61 lore raw

Re: [PATCH 14/19] ata: libata-core: skip poweroff for devices that are runtime suspended

Quoted 9 lines · Damien Le Moal

When powering off, there is no need to suspend a port that has already been runtime suspended. Skip the EH PM request in ata_port_pm_poweroff() in this case.

---
  drivers/ata/libata-core.c | 3 ++-
  1 file changed, 2 insertions(+), 1 deletion(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
author maintainer
#63 lore raw

PATCH 15/19 ata: libata-core: Do not resume ports that have been runtime suspended

In reply to Damien Le Moal

The scsi disk driver does not resume disks that have been runtime suspended by the user. To be consistent with this behavior, do the same for ata ports and skip the PM request in ata_port_pm_resume() if the port was already runtime suspended. With this change, it is no longer necessary to for the PM state of the port to ACTIVE as the PM core code will take care of that when handling runtime resume.

---
 drivers/ata/libata-core.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
drivers/ata/libata-core.c +2 -4
index c4a32abc2e29..26e94edf5103 100644
@@ -5234,10 +5234,8 @@ static void ata_port_resume(struct ata_port *ap, pm_message_t mesg,
 
 static int ata_port_pm_resume(struct device *dev)
 {
-	ata_port_resume(to_ata_port(dev), PMSG_RESUME, true);
-	pm_runtime_disable(dev);
-	pm_runtime_set_active(dev);
-	pm_runtime_enable(dev);
+	if (!pm_runtime_suspended(dev))
+		ata_port_resume(to_ata_port(dev), PMSG_RESUME, true);
 	return 0;
 }
 
Signature · 1 line
2.41.0
Reviewed-by on v1 15/19
#64 lore raw

Re: [PATCH 15/19] ata: libata-core: Do not resume ports that have been runtime suspended

Quoted 7 lines · Damien Le Moal

The scsi disk driver does not resume disks that have been runtime suspended by the user. To be consistent with this behavior, do the same for ata ports and skip the PM request in ata_port_pm_resume() if the port was already runtime suspended. With this change, it is no longer necessary to for the PM state of the port to ACTIVE as the PM core code will take care of that when handling runtime resume.

Quoted 5 lines
---
  drivers/ata/libata-core.c | 6 ++----
  1 file changed, 2 insertions(+), 4 deletions(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
Tested-by on v1 15/19
#65 lore raw

Re: [PATCH 15/19] ata: libata-core: Do not resume ports that have been runtime suspended

In reply to Damien Le Moal
Quoted 8 lines · Damien Le Moal

The scsi disk driver does not resume disks that have been runtime suspended by the user. To be consistent with this behavior, do the same for ata ports and skip the PM request in ata_port_pm_resume() if the port was already runtime suspended. With this change, it is no longer necessary to for the PM state of the port to ACTIVE as the PM core code will take care of that when handling runtime resume.

author maintainer
#66 lore raw

PATCH 16/19 ata: libata-sata: Improve ata_sas_slave_configure()

In reply to Damien Le Moal

Change ata_sas_slave_configure() to return the return value of ata_scsi_dev_config() to ensure that any error from that function is propagated to libsas.

---
 drivers/ata/libata-sata.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/ata/libata-sata.c +2 -2
index 5d31c08be013..0748e9ea4f5f 100644
@@ -1169,8 +1169,8 @@ EXPORT_SYMBOL_GPL(ata_sas_tport_delete);
 int ata_sas_slave_configure(struct scsi_device *sdev, struct ata_port *ap)
 {
 	ata_scsi_sdev_config(sdev);
-	ata_scsi_dev_config(sdev, ap->link.device);
-	return 0;
+
+	return ata_scsi_dev_config(sdev, ap->link.device);
 }
 EXPORT_SYMBOL_GPL(ata_sas_slave_configure);
 
Signature · 1 line
2.41.0
Reviewed-by on v1 16/19
#67 lore raw

Re: [PATCH 16/19] ata: libata-sata: Improve ata_sas_slave_configure()

Quoted 14 lines · Damien Le Moal

Change ata_sas_slave_configure() to return the return value of ata_scsi_dev_config() to ensure that any error from that function is propagated to libsas.

---
  drivers/ata/libata-sata.c | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)
drivers/ata/libata-sata.c +0 -0
index 5d31c08be013..0748e9ea4f5f 100644

drivers/ata/libata-sata.c EXPORT_SYMBOL_GPL(ata_sas_tport_delete); · line 1176

Quoted 10 lines
drivers/ata/libata-sata.c +2 -2
@@ -1169,8 +1169,8 @@ EXPORT_SYMBOL_GPL(ata_sas_tport_delete);
  int ata_sas_slave_configure(struct scsi_device *sdev, struct ata_port *ap)
  {
  	ata_scsi_sdev_config(sdev);
-	ata_scsi_dev_config(sdev, ap->link.device);
-	return 0;
+
+	return ata_scsi_dev_config(sdev, ap->link.device);
  }
  EXPORT_SYMBOL_GPL(ata_sas_slave_configure);
 

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
author maintainer
#69 lore raw

PATCH 17/19 ata: libata-eh: Improve reset error messages

In reply to Damien Le Moal

Some drives are really slow to spinup on resume, resulting is a very slow response to COMRESET and to error messages such as:

ata1: COMRESET failed (errno=-16)
ata1: link is slow to respond, please be patient (ready=0)
ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300) ata1.00: configured for UDMA/133

Given that the slowness of the response is indicated with the message "link is slow to respond..." and that resets are retried until the device is detected as online after up to 1min (ata_eh_reset_timeouts), there is no point in printing the "COMRESET failed" error message. Let's not scare the user with non fatal errors and only warn about reset failures in ata_eh_reset() when all reset retries have been exhausted.

---
 drivers/ata/libata-eh.c   | 2 ++
 drivers/ata/libata-sata.c | 1 -
 2 files changed, 2 insertions(+), 1 deletion(-)
drivers/ata/libata-eh.c +2 -0
index 43b0a1509548..bbc522d16f44 100644
@@ -2919,6 +2919,8 @@ int ata_eh_reset(struct ata_link *link, int classify,
 		 */
 		if (ata_is_host_link(link))
 			ata_eh_thaw_port(ap);
+		ata_link_warn(link, "%sreset failed\n",
+			      reset == hardreset ? "hard" : "soft");
 		goto out;
 	}
 
drivers/ata/libata-sata.c +0 -1
index 0748e9ea4f5f..00674aae1696 100644
@@ -608,7 +608,6 @@ int sata_link_hardreset(struct ata_link *link, const unsigned int *timing,
 		/* online is set iff link is online && reset succeeded */
 		if (online)
 			*online = false;
-		ata_link_err(link, "COMRESET failed (errno=%d)\n", rc);
 	}
 	return rc;
 }
Signature · 1 line
2.41.0
Reviewed-by on v1 17/19
#70 lore raw

Re: [PATCH 17/19] ata: libata-eh: Improve reset error messages

Quoted 39 lines · Damien Le Moal

Some drives are really slow to spinup on resume, resulting is a very slow response to COMRESET and to error messages such as:

ata1: COMRESET failed (errno=-16)
ata1: link is slow to respond, please be patient (ready=0)
ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300) ata1.00: configured for UDMA/133

Given that the slowness of the response is indicated with the message "link is slow to respond..." and that resets are retried until the device is detected as online after up to 1min (ata_eh_reset_timeouts), there is no point in printing the "COMRESET failed" error message. Let's not scare the user with non fatal errors and only warn about reset failures in ata_eh_reset() when all reset retries have been exhausted.

---
  drivers/ata/libata-eh.c   | 2 ++
  drivers/ata/libata-sata.c | 1 -
  2 files changed, 2 insertions(+), 1 deletion(-)
drivers/ata/libata-eh.c +2 -0
index 43b0a1509548..bbc522d16f44 100644
@@ -2919,6 +2919,8 @@ int ata_eh_reset(struct ata_link *link, int classify,
  		 */
  		if (ata_is_host_link(link))
  			ata_eh_thaw_port(ap);
+		ata_link_warn(link, "%sreset failed\n",
+			      reset == hardreset ? "hard" : "soft");
  		goto out;
  	}
 
drivers/ata/libata-sata.c +0 -0
index 0748e9ea4f5f..00674aae1696 100644

drivers/ata/libata-sata.c int sata_link_hardreset(struct ata_link *link, const unsigned int *timing, · line 613

Quoted 7 lines
drivers/ata/libata-sata.c +0 -1
@@ -608,7 +608,6 @@ int sata_link_hardreset(struct ata_link *link, const unsigned int *timing,
  		/* online is set iff link is online && reset succeeded */
  		if (online)
  			*online = false;
-		ata_link_err(link, "COMRESET failed (errno=%d)\n", rc);
  	}
  	return rc;
  }

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
reviewer
#71 lore raw

Re: [PATCH 17/19] ata: libata-eh: Improve reset error messages

In reply to Damien Le Moal
Quoted 26 lines · Damien Le Moal

Some drives are really slow to spinup on resume, resulting is a very slow response to COMRESET and to error messages such as:

ata1: COMRESET failed (errno=-16)
ata1: link is slow to respond, please be patient (ready=0)
ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300) ata1.00: configured for UDMA/133

Given that the slowness of the response is indicated with the message "link is slow to respond..." and that resets are retried until the device is detected as online after up to 1min (ata_eh_reset_timeouts), there is no point in printing the "COMRESET failed" error message. Let's not scare the user with non fatal errors and only warn about reset failures in ata_eh_reset() when all reset retries have been exhausted.

---
  drivers/ata/libata-eh.c   | 2 ++
  drivers/ata/libata-sata.c | 1 -
  2 files changed, 2 insertions(+), 1 deletion(-)
drivers/ata/libata-eh.c +0 -0
index 43b0a1509548..bbc522d16f44 100644

drivers/ata/libata-eh.c int ata_eh_reset(struct ata_link *link, int classify, · line 2923

Quoted 5 lines
drivers/ata/libata-eh.c +2 -0
@@ -2919,6 +2919,8 @@ int ata_eh_reset(struct ata_link *link, int classify,
  		 */
  		if (ata_is_host_link(link))
  			ata_eh_thaw_port(ap);
+		ata_link_warn(link, "%sreset failed\n",
+			      reset == hardreset ? "hard" : "soft");

nit: how about spell out the full reset type string

ata_link_warn(link, "%s failed\n",

drivers/ata/libata-eh.c int ata_eh_reset(struct ata_link *link, int classify, · line 2924

Quoted 1 line
drivers/ata/libata-eh.c +1 -0
+			      reset == hardreset ? "hardreset" : "softreset");

which may make grep'ing slightly easier.

Quoted 15 lines
drivers/ata/libata-eh.c +0 -0
  		goto out;
  	}
drivers/ata/libata-sata.c +0 -1
index 0748e9ea4f5f..00674aae1696 100644
@@ -608,7 +608,6 @@ int sata_link_hardreset(struct ata_link *link, const unsigned int *timing,
  		/* online is set iff link is online && reset succeeded */
  		if (online)
  			*online = false;
-		ata_link_err(link, "COMRESET failed (errno=%d)\n", rc);
  	}
  	return rc;
  }
Tested-by on v1 17/19
#72 lore raw

Re: [PATCH 17/19] ata: libata-eh: Improve reset error messages

In reply to Damien Le Moal
Quoted 8 lines · Damien Le Moal

Some drives are really slow to spinup on resume, resulting is a very slow response to COMRESET and to error messages such as:

ata1: COMRESET failed (errno=-16)
ata1: link is slow to respond, please be patient (ready=0)
ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300) ata1.00: configured for UDMA/133

Quoted 8 lines

Given that the slowness of the response is indicated with the message "link is slow to respond..." and that resets are retried until the device is detected as online after up to 1min (ata_eh_reset_timeouts), there is no point in printing the "COMRESET failed" error message. Let's not scare the user with non fatal errors and only warn about reset failures in ata_eh_reset() when all reset retries have been exhausted.

author maintainer
#73 lore raw

PATCH 18/19 ata: libata-eh: Reduce "disable device" message verbosity

In reply to Damien Le Moal

There is no point in warning about a device being diabled when we expect it to be, that is, on suspend, shutdown or when detaching a device. Suppress this message for these cases by introducing the EH static function ata_eh_dev_disable() and by using it in ata_eh_unload() and ata_eh_detach_dev(). ata_dev_disable() code is modified to call this new function after printing the "disable device" message.

---
 drivers/ata/libata-eh.c | 32 +++++++++++++++++++-------------
 1 file changed, 19 insertions(+), 13 deletions(-)
drivers/ata/libata-eh.c +19 -13
index bbc522d16f44..5bad5dabffe0 100644
@@ -494,6 +494,18 @@ void ata_eh_release(struct ata_port *ap)
 	mutex_unlock(&ap->host->eh_mutex);
 }
 
+static void ata_eh_dev_disable(struct ata_device *dev)
+{
+	ata_acpi_on_disable(dev);
+	ata_down_xfermask_limit(dev, ATA_DNXFER_FORCE_PIO0 | ATA_DNXFER_QUIET);
+	dev->class++;
+
+	/* From now till the next successful probe, ering is used to
+	 * track probe failures.  Clear accumulated device error info.
+	 */
+	ata_ering_clear(&dev->ering);
+}
+
 static void ata_eh_unload(struct ata_port *ap)
 {
 	struct ata_link *link;
@@ -517,8 +529,8 @@ static void ata_eh_unload(struct ata_port *ap)
 	 */
 	ata_for_each_link(link, ap, PMP_FIRST) {
 		sata_scr_write(link, SCR_CONTROL, link->saved_scontrol & 0xff0);
-		ata_for_each_dev(dev, link, ALL)
-			ata_dev_disable(dev);
+		ata_for_each_dev(dev, link, ENABLED)
+			ata_eh_dev_disable(dev);
 	}
 
 	/* freeze and set UNLOADED */
@@ -1211,14 +1223,8 @@ void ata_dev_disable(struct ata_device *dev)
 		return;
 
 	ata_dev_warn(dev, "disable device\n");
-	ata_acpi_on_disable(dev);
-	ata_down_xfermask_limit(dev, ATA_DNXFER_FORCE_PIO0 | ATA_DNXFER_QUIET);
-	dev->class++;
 
-	/* From now till the next successful probe, ering is used to
-	 * track probe failures.  Clear accumulated device error info.
-	 */
-	ata_ering_clear(&dev->ering);
+	ata_eh_dev_disable(dev);
 }
 EXPORT_SYMBOL_GPL(ata_dev_disable);
 
@@ -1240,12 +1246,12 @@ void ata_eh_detach_dev(struct ata_device *dev)
 
 	/*
 	 * If the device is still enabled, transition it to standby power mode
-	 * (i.e. spin down HDDs).
+	 * (i.e. spin down HDDs) and disable it.
 	 */
-	if (ata_dev_enabled(dev))
+	if (ata_dev_enabled(dev)) {
 		ata_dev_power_set_standby(dev);
-
-	ata_dev_disable(dev);
+		ata_eh_dev_disable(dev);
+	}
 
 	spin_lock_irqsave(ap->lock, flags);
 
Signature · 1 line
2.41.0
Reviewed-by on v1 18/19
#74 lore raw

Re: [PATCH 18/19] ata: libata-eh: Reduce "disable device" message verbosity

Quoted 1 line · Damien Le Moal

There is no point in warning about a device being diabled when we expect

disabled

Quoted 11 lines

it to be, that is, on suspend, shutdown or when detaching a device. Suppress this message for these cases by introducing the EH static function ata_eh_dev_disable() and by using it in ata_eh_unload() and ata_eh_detach_dev(). ata_dev_disable() code is modified to call this new function after printing the "disable device" message.

---
  drivers/ata/libata-eh.c | 32 +++++++++++++++++++-------------
  1 file changed, 19 insertions(+), 13 deletions(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman
#75 lore raw

Re: [PATCH 18/19] ata: libata-eh: Reduce "disable device" message verbosity

In reply to Damien Le Moal

Hello!

Quoted 1 line · Damien Le Moal

There is no point in warning about a device being diabled when we expect

Disabled. :-)

Quoted 7 lines

it to be, that is, on suspend, shutdown or when detaching a device. Suppress this message for these cases by introducing the EH static function ata_eh_dev_disable() and by using it in ata_eh_unload() and ata_eh_detach_dev(). ata_dev_disable() code is modified to call this new function after printing the "disable device" message.

[...]

MBR, Sergey

Tested-by on v1 18/19
#76 lore raw

Re: [PATCH 18/19] ata: libata-eh: Reduce "disable device" message verbosity

In reply to Damien Le Moal
Quoted 8 lines · Damien Le Moal

There is no point in warning about a device being diabled when we expect it to be, that is, on suspend, shutdown or when detaching a device. Suppress this message for these cases by introducing the EH static function ata_eh_dev_disable() and by using it in ata_eh_unload() and ata_eh_detach_dev(). ata_dev_disable() code is modified to call this new function after printing the "disable device" message.

author maintainer
#77 lore raw

PATCH 19/19 ata: libata: Cleanup inline DMA helper functions

In reply to Damien Le Moal

Simplify the inline DMA helper functions ata_using_mwdma(), ata_using_udma() and ata_dma_enabled() to directly return as a boolean the result of their test condition.

---
 include/linux/libata.h | 18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)
include/linux/libata.h +8 -10
index 6593c79b7290..f48fe27dae94 100644
@@ -1878,23 +1878,21 @@ static inline unsigned long ata_deadline(unsigned long from_jiffies,
    change in future hardware and specs, secondly 0xFF means 'no DMA' but is
    > UDMA_0. Dyma ddreigiau */
 
-static inline int ata_using_mwdma(struct ata_device *adev)
+static inline bool ata_using_mwdma(struct ata_device *adev)
 {
-	if (adev->dma_mode >= XFER_MW_DMA_0 && adev->dma_mode <= XFER_MW_DMA_4)
-		return 1;
-	return 0;
+	return adev->dma_mode >= XFER_MW_DMA_0 &&
+		adev->dma_mode <= XFER_MW_DMA_4;
 }
 
-static inline int ata_using_udma(struct ata_device *adev)
+static inline bool ata_using_udma(struct ata_device *adev)
 {
-	if (adev->dma_mode >= XFER_UDMA_0 && adev->dma_mode <= XFER_UDMA_7)
-		return 1;
-	return 0;
+	return adev->dma_mode >= XFER_UDMA_0 &&
+		adev->dma_mode <= XFER_UDMA_7;
 }
 
-static inline int ata_dma_enabled(struct ata_device *adev)
+static inline bool ata_dma_enabled(struct ata_device *adev)
 {
-	return (adev->dma_mode == 0xFF ? 0 : 1);
+	return adev->dma_mode != 0xFF;
 }
 
 /**************************************************************************
Signature · 1 line
2.41.0
Reviewed-by on v1 19/19
#78 lore raw

Re: [PATCH 19/19] ata: libata: Cleanup inline DMA helper functions

Quoted 9 lines · Damien Le Moal

Simplify the inline DMA helper functions ata_using_mwdma(), ata_using_udma() and ata_dma_enabled() to directly return as a boolean the result of their test condition.

---
  include/linux/libata.h | 18 ++++++++----------
  1 file changed, 8 insertions(+), 10 deletions(-)

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke                Kernel Storage Architect
[email protected]                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman