[PATCH 00/28] SCSI EH argument reshuffle part I

[PATCH 00/28] SCSI EH argument reshuffle part I

[Hannes Reinecke]

Hi all,
finally here's the patchset to revamp the SCSI EH callback arguments
which I promised to do (some years ago ...).

The overall idea is to match the scope of the eh_XXX callbacks with
the appropriate argument, eg eh_device_reset_handler() should have a
scsi device as argument etc.
Relying on the scsi command has the problem that
a) we're holding a reference on that command for the entire lifetime
of the error handling and
b) it leads to some 'interesting' driver implementations; some
higher-level EH callback implementations go through the list of
outstanding commands and try to abort this particular command only;
makes one wonder what'll happen to the other commands ...

However, this patchset has the nice side-effect that we don't need to
allocate an out-of-order scsi command in scsi_ioctl_reset(), but can
call the function directly.

As per request from hch the patchset has been split in two parts;
this is the first part which just does a general EH function cleanup
without introducing any functional differences.

As usual, comments and reviews are welcome.

Hannes Reinecke (28):
  scsi: fix comment in scsi_device_set_state()
  mptfc: Do not call fc_block_scsi_eh() on host reset
  ibmvfc: Do not call fc_block_scsi_eh() on host reset
  fc_fcp: do not call fc_block_scsi_eh() from host reset
  fnic: do not call host reset from command abort
  uas: move eh_bus_reset_handler to eh_device_reset_handler
  libsas: move bus_reset_handler() to target_reset_handler()
  bfa: move bus reset to target reset
  hptiop: Simplify reset handling
  fdomain: move bus reset to host reset
  scsi: drop bus reset for wd33c93-compatible boards
  rtsx: drop bus reset function
  qlogicpti: move bus reset to host reset
  acornscsi: move bus reset to host reset
  NCR5380: Move bus reset to host reset
  qlogicfas: move bus_reset to host_reset
  imm: drop duplicate bus_reset handler
  ppa: drop duplicate bus_reset handler
  qedf: drop bus reset handler
  nsp32: drop bus reset
  aha152x: drop bus reset
  53c700: move bus reset to host reset
  bnx2fc: remove obsolete bnx2fc_eh_host_reset() definition
  megaraid_mbox: drop duplicate bus reset and device reset function
  lpfc: drop lpfc_no_handler and lpfc_bus_reset_handler()
  visorhba: sanitze private device data allocation
  eata: remove 'arg_done' from eata2x_eh_host_reset()
  NCR5380: Clear all unissued commands on host reset

 drivers/message/fusion/mptfc.c                  |  10 +-
 drivers/scsi/53c700.c                           |  22 +----
 drivers/scsi/NCR5380.c                          |  11 ++-
 drivers/scsi/a2091.c                            |  17 ----
 drivers/scsi/a3000.c                            |  17 ----
 drivers/scsi/aha152x.c                          |  10 --
 drivers/scsi/aic94xx/aic94xx_init.c             |   2 +-
 drivers/scsi/arm/acornscsi.c                    |  11 ++-
 drivers/scsi/arm/cumana_1.c                     |   2 +-
 drivers/scsi/arm/oak.c                          |   2 +-
 drivers/scsi/atari_scsi.c                       |   6 +-
 drivers/scsi/bfa/bfad_im.c                      |  37 +++----
 drivers/scsi/bnx2fc/bnx2fc.h                    |   1 -
 drivers/scsi/dmx3191d.c                         |   2 +-
 drivers/scsi/eata.c                             |   9 +-
 drivers/scsi/fdomain.c                          |   6 +-
 drivers/scsi/fdomain.h                          |   2 +-
 drivers/scsi/fnic/fnic_scsi.c                   |   4 -
 drivers/scsi/g_NCR5380.c                        |   4 +-
 drivers/scsi/gvp11.c                            |  18 ----
 drivers/scsi/hisi_sas/hisi_sas_main.c           |   2 +-
 drivers/scsi/hptiop.c                           |  11 +--
 drivers/scsi/ibmvscsi/ibmvfc.c                  |   6 +-
 drivers/scsi/imm.c                              |   1 -
 drivers/scsi/isci/init.c                        |   2 +-
 drivers/scsi/libfc/fc_fcp.c                     |   2 -
 drivers/scsi/libsas/sas_scsi_host.c             |  12 +--
 drivers/scsi/lpfc/lpfc_scsi.c                   |  99 -------------------
 drivers/scsi/mac_scsi.c                         |   4 +-
 drivers/scsi/megaraid/megaraid_mbox.c           |   2 -
 drivers/scsi/mvme147.c                          |  16 ---
 drivers/scsi/mvsas/mv_init.c                    |   2 +-
 drivers/scsi/nsp32.c                            |  22 +----
 drivers/scsi/pcmcia/fdomain_stub.c              |   2 +-
 drivers/scsi/pcmcia/qlogic_stub.c               |   4 +-
 drivers/scsi/pm8001/pm8001_init.c               |   2 +-
 drivers/scsi/ppa.c                              |   1 -
 drivers/scsi/qedf/qedf_main.c                   |  11 ---
 drivers/scsi/qlogicfas.c                        |   2 +-
 drivers/scsi/qlogicfas408.c                     |   6 +-
 drivers/scsi/qlogicfas408.h                     |   2 +-
 drivers/scsi/qlogicpti.c                        |   5 +-
 drivers/scsi/scsi_lib.c                         |   2 +-
 drivers/scsi/sgiwd93.c                          |  15 ---
 drivers/scsi/sun3_scsi.c                        |   4 +-
 drivers/scsi/wd33c93.c                          |   2 +
 drivers/staging/rts5208/rtsx.c                  |  11 ---
 drivers/staging/unisys/visorhba/visorhba_main.c | 123 ++++++++++--------------
 drivers/usb/storage/uas.c                       |   4 +-
 include/scsi/libsas.h                           |   2 +-
 50 files changed, 135 insertions(+), 437 deletions(-)

-- 
1.8.5.6

[PATCH 01/28] scsi: fix comment in scsi_device_set_state()

[Hannes Reinecke]

The function returns '0' if successful; with the original comment
the function doesn't have a way to indicate success ...

Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Bart van Assche <bvanassche@wdc.com>
---
 drivers/scsi/scsi_lib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 41c19c7..1ff7d72 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -2547,7 +2547,7 @@ void scsi_exit_queue(void)
  *	@sdev:	scsi device to change the state of.
  *	@state:	state to change to.
  *
- *	Returns zero if unsuccessful or an error if the requested 
+ *	Returns zero if successful or an error if the requested
  *	transition is illegal.
  */
 int
-- 
1.8.5.6

Re: [PATCH 01/28] scsi: fix comment in scsi_device_set_state()

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

Re: [PATCH 01/28] scsi: fix comment in scsi_device_set_state()

[Christoph Hellwig]

Looks fine,

Reviewed-by: Christoph Hellwig <hch@lst.de>

On Wed, Jun 28, 2017 at 10:24:49AM +0200, Hannes Reinecke wrote:
> The function returns '0' if successful; with the original comment
> the function doesn't have a way to indicate success ...
> 
> Signed-off-by: Hannes Reinecke <hare@suse.de>
> Reviewed-by: Bart van Assche <bvanassche@wdc.com>
> ---
>  drivers/scsi/scsi_lib.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
> index 41c19c7..1ff7d72 100644
> --- a/drivers/scsi/scsi_lib.c
> +++ b/drivers/scsi/scsi_lib.c
> @@ -2547,7 +2547,7 @@ void scsi_exit_queue(void)
>   *	@sdev:	scsi device to change the state of.
>   *	@state:	state to change to.
>   *
> - *	Returns zero if unsuccessful or an error if the requested 
> + *	Returns zero if successful or an error if the requested
>   *	transition is illegal.
>   */
>  int
> -- 
> 1.8.5.6
---end quoted text---

[PATCH 02/28] mptfc: Do not call fc_block_scsi_eh() on host reset

[Hannes Reinecke]

When we're resetting the host any remote port states will
be reset anyway, so it's pointless to wait for dev_loss_tmo
during host reset.

Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/message/fusion/mptfc.c | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/drivers/message/fusion/mptfc.c b/drivers/message/fusion/mptfc.c
index d065062..6d461ca 100644
--- a/drivers/message/fusion/mptfc.c
+++ b/drivers/message/fusion/mptfc.c
@@ -104,7 +104,6 @@
 static int mptfc_abort(struct scsi_cmnd *SCpnt);
 static int mptfc_dev_reset(struct scsi_cmnd *SCpnt);
 static int mptfc_bus_reset(struct scsi_cmnd *SCpnt);
-static int mptfc_host_reset(struct scsi_cmnd *SCpnt);
 
 static struct scsi_host_template mptfc_driver_template = {
 	.module				= THIS_MODULE,
@@ -123,7 +122,7 @@
 	.eh_abort_handler		= mptfc_abort,
 	.eh_device_reset_handler	= mptfc_dev_reset,
 	.eh_bus_reset_handler		= mptfc_bus_reset,
-	.eh_host_reset_handler		= mptfc_host_reset,
+	.eh_host_reset_handler		= mptscsih_host_reset,
 	.bios_param			= mptscsih_bios_param,
 	.can_queue			= MPT_FC_CAN_QUEUE,
 	.this_id			= -1,
@@ -254,13 +253,6 @@
 	    mptfc_block_error_handler(SCpnt, mptscsih_bus_reset, __func__);
 }
 
-static int
-mptfc_host_reset(struct scsi_cmnd *SCpnt)
-{
-	return
-	    mptfc_block_error_handler(SCpnt, mptscsih_host_reset, __func__);
-}
-
 static void
 mptfc_set_rport_loss_tmo(struct fc_rport *rport, uint32_t timeout)
 {
-- 
1.8.5.6

Re: [PATCH 02/28] mptfc: Do not call fc_block_scsi_eh() on host reset

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 03/28] ibmvfc: Do not call fc_block_scsi_eh() on host reset

[Hannes Reinecke]

When we're resetting the host any remote port states will
be reset anyway, so it's pointless to wait for dev_loss_tmo
during host reset.

Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/scsi/ibmvscsi/ibmvfc.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/scsi/ibmvscsi/ibmvfc.c b/drivers/scsi/ibmvscsi/ibmvfc.c
index 26cd3c2..bb033ce 100644
--- a/drivers/scsi/ibmvscsi/ibmvfc.c
+++ b/drivers/scsi/ibmvscsi/ibmvfc.c
@@ -2528,16 +2528,12 @@ static int ibmvfc_eh_target_reset_handler(struct scsi_cmnd *cmd)
  **/
 static int ibmvfc_eh_host_reset_handler(struct scsi_cmnd *cmd)
 {
-	int rc, block_rc;
+	int rc;
 	struct ibmvfc_host *vhost = shost_priv(cmd->device->host);
 
-	block_rc = fc_block_scsi_eh(cmd);
 	dev_err(vhost->dev, "Resetting connection due to error recovery\n");
 	rc = ibmvfc_issue_fc_host_lip(vhost->host);
 
-	if (block_rc == FAST_IO_FAIL)
-		return FAST_IO_FAIL;
-
 	return rc ? FAILED : SUCCESS;
 }
 
-- 
1.8.5.6

Re: [PATCH 03/28] ibmvfc: Do not call fc_block_scsi_eh() on host reset

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 04/28] fc_fcp: do not call fc_block_scsi_eh() from host reset

[Hannes Reinecke]

When calling host reset we're resetting all ports anyway, so
there is no point in waiting for the ports to become unblocked.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/scsi/libfc/fc_fcp.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/scsi/libfc/fc_fcp.c b/drivers/scsi/libfc/fc_fcp.c
index 234352da..772c35a 100644
--- a/drivers/scsi/libfc/fc_fcp.c
+++ b/drivers/scsi/libfc/fc_fcp.c
@@ -2222,8 +2222,6 @@ int fc_eh_host_reset(struct scsi_cmnd *sc_cmd)
 
 	FC_SCSI_DBG(lport, "Resetting host\n");
 
-	fc_block_scsi_eh(sc_cmd);
-
 	fc_lport_reset(lport);
 	wait_tmo = jiffies + FC_HOST_RESET_TIMEOUT;
 	while (!fc_fcp_lport_queue_ready(lport) && time_before(jiffies,
-- 
1.8.5.6

Re: [PATCH 04/28] fc_fcp: do not call fc_block_scsi_eh() from host reset

[Johannes Thumshirn]

Acked-by: Johannes Thumshirn <jth@kernel.org>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 05/28] fnic: do not call host reset from command abort

[Hannes Reinecke]

command abort already returns FAILED, which will then be escalated
to a host reset. So no need to call host_reset directly.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/scsi/fnic/fnic_scsi.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/scsi/fnic/fnic_scsi.c b/drivers/scsi/fnic/fnic_scsi.c
index 6c0646d..242e2ee 100644
--- a/drivers/scsi/fnic/fnic_scsi.c
+++ b/drivers/scsi/fnic/fnic_scsi.c
@@ -1990,10 +1990,6 @@ int fnic_abort_cmd(struct scsi_cmnd *sc)
 		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
 			"Issuing Host reset due to out of order IO\n");
 
-		if (fnic_host_reset(sc) == FAILED) {
-			FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
-				"fnic_host_reset failed.\n");
-		}
 		ret = FAILED;
 		goto fnic_abort_cmd_end;
 	}
-- 
1.8.5.6

Re: [PATCH 05/28] fnic: do not call host reset from command abort

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 06/28] uas: move eh_bus_reset_handler to eh_device_reset_handler

[Hannes Reinecke]

The bus_reset handler is really a device reset, so move it
to eh_device_reset_handler().

Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/usb/storage/uas.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/storage/uas.c b/drivers/usb/storage/uas.c
index 5ef014b..cfb1e3b 100644
--- a/drivers/usb/storage/uas.c
+++ b/drivers/usb/storage/uas.c
@@ -737,7 +737,7 @@ static int uas_eh_abort_handler(struct scsi_cmnd *cmnd)
 	return FAILED;
 }
 
-static int uas_eh_bus_reset_handler(struct scsi_cmnd *cmnd)
+static int uas_eh_device_reset_handler(struct scsi_cmnd *cmnd)
 {
 	struct scsi_device *sdev = cmnd->device;
 	struct uas_dev_info *devinfo = sdev->hostdata;
@@ -848,7 +848,7 @@ static int uas_slave_configure(struct scsi_device *sdev)
 	.slave_alloc = uas_slave_alloc,
 	.slave_configure = uas_slave_configure,
 	.eh_abort_handler = uas_eh_abort_handler,
-	.eh_bus_reset_handler = uas_eh_bus_reset_handler,
+	.eh_device_reset_handler = uas_eh_device_reset_handler,
 	.this_id = -1,
 	.sg_tablesize = SG_NONE,
 	.skip_settle_delay = 1,
-- 
1.8.5.6

Re: [PATCH 06/28] uas: move eh_bus_reset_handler to eh_device_reset_handler

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 07/28] libsas: move bus_reset_handler() to target_reset_handler()

[Hannes Reinecke]

The bus reset handler is calling I_T Nexus reset, which logically
is a target reset as it need to specify both the initiator and
the target.
So move it to target reset.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/scsi/aic94xx/aic94xx_init.c   |  2 +-
 drivers/scsi/hisi_sas/hisi_sas_main.c |  2 +-
 drivers/scsi/isci/init.c              |  2 +-
 drivers/scsi/libsas/sas_scsi_host.c   | 12 ++++++------
 drivers/scsi/mvsas/mv_init.c          |  2 +-
 drivers/scsi/pm8001/pm8001_init.c     |  2 +-
 include/scsi/libsas.h                 |  2 +-
 7 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/drivers/scsi/aic94xx/aic94xx_init.c b/drivers/scsi/aic94xx/aic94xx_init.c
index a14ba7a..d5b8d0b 100644
--- a/drivers/scsi/aic94xx/aic94xx_init.c
+++ b/drivers/scsi/aic94xx/aic94xx_init.c
@@ -70,7 +70,7 @@
 	.max_sectors		= SCSI_DEFAULT_MAX_SECTORS,
 	.use_clustering		= ENABLE_CLUSTERING,
 	.eh_device_reset_handler	= sas_eh_device_reset_handler,
-	.eh_bus_reset_handler	= sas_eh_bus_reset_handler,
+	.eh_target_reset_handler	= sas_eh_target_reset_handler,
 	.target_destroy		= sas_target_destroy,
 	.ioctl			= sas_ioctl,
 	.track_queue_depth	= 1,
diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index e2f8d92..801edce 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -1510,7 +1510,7 @@ void hisi_sas_rescan_topology(struct hisi_hba *hisi_hba, u32 old_state,
 	.max_sectors		= SCSI_DEFAULT_MAX_SECTORS,
 	.use_clustering		= ENABLE_CLUSTERING,
 	.eh_device_reset_handler = sas_eh_device_reset_handler,
-	.eh_bus_reset_handler	= sas_eh_bus_reset_handler,
+	.eh_target_reset_handler = sas_eh_target_reset_handler,
 	.target_destroy		= sas_target_destroy,
 	.ioctl			= sas_ioctl,
 };
diff --git a/drivers/scsi/isci/init.c b/drivers/scsi/isci/init.c
index 4537117..922e3e5 100644
--- a/drivers/scsi/isci/init.c
+++ b/drivers/scsi/isci/init.c
@@ -166,7 +166,7 @@ struct device_attribute *isci_host_attrs[] = {
 	.use_clustering			= ENABLE_CLUSTERING,
 	.eh_abort_handler		= sas_eh_abort_handler,
 	.eh_device_reset_handler        = sas_eh_device_reset_handler,
-	.eh_bus_reset_handler           = sas_eh_bus_reset_handler,
+	.eh_target_reset_handler        = sas_eh_target_reset_handler,
 	.target_destroy			= sas_target_destroy,
 	.ioctl				= sas_ioctl,
 	.shost_attrs			= isci_host_attrs,
diff --git a/drivers/scsi/libsas/sas_scsi_host.c b/drivers/scsi/libsas/sas_scsi_host.c
index 87e5079..3c51da8 100644
--- a/drivers/scsi/libsas/sas_scsi_host.c
+++ b/drivers/scsi/libsas/sas_scsi_host.c
@@ -526,7 +526,7 @@ int sas_eh_device_reset_handler(struct scsi_cmnd *cmd)
 	return FAILED;
 }
 
-int sas_eh_bus_reset_handler(struct scsi_cmnd *cmd)
+int sas_eh_target_reset_handler(struct scsi_cmnd *cmd)
 {
 	int res;
 	struct Scsi_Host *host = cmd->device->host;
@@ -554,15 +554,15 @@ static int try_to_reset_cmd_device(struct scsi_cmnd *cmd)
 	struct Scsi_Host *shost = cmd->device->host;
 
 	if (!shost->hostt->eh_device_reset_handler)
-		goto try_bus_reset;
+		goto try_target_reset;
 
 	res = shost->hostt->eh_device_reset_handler(cmd);
 	if (res == SUCCESS)
 		return res;
 
-try_bus_reset:
-	if (shost->hostt->eh_bus_reset_handler)
-		return shost->hostt->eh_bus_reset_handler(cmd);
+try_target_reset:
+	if (shost->hostt->eh_target_reset_handler)
+		return shost->hostt->eh_target_reset_handler(cmd);
 
 	return FAILED;
 }
@@ -996,6 +996,6 @@ int sas_request_addr(struct Scsi_Host *shost, u8 *addr)
 EXPORT_SYMBOL_GPL(sas_task_abort);
 EXPORT_SYMBOL_GPL(sas_phy_reset);
 EXPORT_SYMBOL_GPL(sas_eh_device_reset_handler);
-EXPORT_SYMBOL_GPL(sas_eh_bus_reset_handler);
+EXPORT_SYMBOL_GPL(sas_eh_target_reset_handler);
 EXPORT_SYMBOL_GPL(sas_target_destroy);
 EXPORT_SYMBOL_GPL(sas_ioctl);
diff --git a/drivers/scsi/mvsas/mv_init.c b/drivers/scsi/mvsas/mv_init.c
index 4e047b5..a3533ea 100644
--- a/drivers/scsi/mvsas/mv_init.c
+++ b/drivers/scsi/mvsas/mv_init.c
@@ -61,7 +61,7 @@
 	.max_sectors		= SCSI_DEFAULT_MAX_SECTORS,
 	.use_clustering		= ENABLE_CLUSTERING,
 	.eh_device_reset_handler = sas_eh_device_reset_handler,
-	.eh_bus_reset_handler	= sas_eh_bus_reset_handler,
+	.eh_target_reset_handler = sas_eh_target_reset_handler,
 	.target_destroy		= sas_target_destroy,
 	.ioctl			= sas_ioctl,
 	.shost_attrs		= mvst_host_attrs,
diff --git a/drivers/scsi/pm8001/pm8001_init.c b/drivers/scsi/pm8001/pm8001_init.c
index 034b2f7..4318f33 100644
--- a/drivers/scsi/pm8001/pm8001_init.c
+++ b/drivers/scsi/pm8001/pm8001_init.c
@@ -86,7 +86,7 @@
 	.max_sectors		= SCSI_DEFAULT_MAX_SECTORS,
 	.use_clustering		= ENABLE_CLUSTERING,
 	.eh_device_reset_handler = sas_eh_device_reset_handler,
-	.eh_bus_reset_handler	= sas_eh_bus_reset_handler,
+	.eh_target_reset_handler = sas_eh_target_reset_handler,
 	.target_destroy		= sas_target_destroy,
 	.ioctl			= sas_ioctl,
 	.shost_attrs		= pm8001_host_attrs,
diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
index cfaeed2..e7c012c 100644
--- a/include/scsi/libsas.h
+++ b/include/scsi/libsas.h
@@ -714,7 +714,7 @@ extern int sas_bios_param(struct scsi_device *,
 void sas_task_abort(struct sas_task *);
 int sas_eh_abort_handler(struct scsi_cmnd *cmd);
 int sas_eh_device_reset_handler(struct scsi_cmnd *cmd);
-int sas_eh_bus_reset_handler(struct scsi_cmnd *cmd);
+int sas_eh_target_reset_handler(struct scsi_cmnd *cmd);
 
 extern void sas_target_destroy(struct scsi_target *);
 extern int sas_slave_alloc(struct scsi_device *);
-- 
1.8.5.6

Re: [PATCH 07/28] libsas: move bus_reset_handler() to target_reset_handler()

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 08/28] bfa: move bus reset to target reset

[Hannes Reinecke]

The bus reset handler is just calling target reset on all targets,
which is exactly what SCSI EH will be doing anyway.
So move the bus reset function to target reset and drop the loop.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/scsi/bfa/bfad_im.c | 37 +++++++++++++++----------------------
 1 file changed, 15 insertions(+), 22 deletions(-)

diff --git a/drivers/scsi/bfa/bfad_im.c b/drivers/scsi/bfa/bfad_im.c
index 7eb0eef..24e657a 100644
--- a/drivers/scsi/bfa/bfad_im.c
+++ b/drivers/scsi/bfa/bfad_im.c
@@ -373,32 +373,28 @@ static void bfad_im_fc_rport_add(struct bfad_im_port_s  *im_port,
 }
 
 /*
- * Scsi_Host template entry, resets the bus and abort all commands.
+ * Scsi_Host template entry, resets the target and abort all commands.
  */
 static int
-bfad_im_reset_bus_handler(struct scsi_cmnd *cmnd)
+bfad_im_reset_target_handler(struct scsi_cmnd *cmnd)
 {
 	struct Scsi_Host *shost = cmnd->device->host;
+	struct scsi_target *starget = scsi_target(cmnd->device);
 	struct bfad_im_port_s *im_port =
 				(struct bfad_im_port_s *) shost->hostdata[0];
 	struct bfad_s         *bfad = im_port->bfad;
 	struct bfad_itnim_s   *itnim;
 	unsigned long   flags;
-	u32        i, rc, err_cnt = 0;
+	u32        rc, rtn = FAILED;
 	DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq);
 	enum bfi_tskim_status task_status;
 
 	spin_lock_irqsave(&bfad->bfad_lock, flags);
-	for (i = 0; i < MAX_FCP_TARGET; i++) {
-		itnim = bfad_get_itnim(im_port, i);
-		if (itnim) {
-			cmnd->SCp.ptr = (char *)&wq;
-			rc = bfad_im_target_reset_send(bfad, cmnd, itnim);
-			if (rc != BFA_STATUS_OK) {
-				err_cnt++;
-				continue;
-			}
-
+	itnim = bfad_get_itnim(im_port, starget->id);
+	if (itnim) {
+		cmnd->SCp.ptr = (char *)&wq;
+		rc = bfad_im_target_reset_send(bfad, cmnd, itnim);
+		if (rc == BFA_STATUS_OK) {
 			/* wait target reset to complete */
 			spin_unlock_irqrestore(&bfad->bfad_lock, flags);
 			wait_event(wq, test_bit(IO_DONE_BIT,
@@ -406,20 +402,17 @@ static void bfad_im_fc_rport_add(struct bfad_im_port_s  *im_port,
 			spin_lock_irqsave(&bfad->bfad_lock, flags);
 
 			task_status = cmnd->SCp.Status >> 1;
-			if (task_status != BFI_TSKIM_STS_OK) {
+			if (task_status != BFI_TSKIM_STS_OK)
 				BFA_LOG(KERN_ERR, bfad, bfa_log_level,
 					"target reset failure,"
 					" status: %d\n", task_status);
-				err_cnt++;
-			}
+			else
+				rtn = SUCCESS;
 		}
 	}
 	spin_unlock_irqrestore(&bfad->bfad_lock, flags);
 
-	if (err_cnt)
-		return FAILED;
-
-	return SUCCESS;
+	return rtn;
 }
 
 /*
@@ -816,7 +809,7 @@ struct scsi_host_template bfad_im_scsi_host_template = {
 	.eh_timed_out = fc_eh_timed_out,
 	.eh_abort_handler = bfad_im_abort_handler,
 	.eh_device_reset_handler = bfad_im_reset_lun_handler,
-	.eh_bus_reset_handler = bfad_im_reset_bus_handler,
+	.eh_target_reset_handler = bfad_im_reset_target_handler,
 
 	.slave_alloc = bfad_im_slave_alloc,
 	.slave_configure = bfad_im_slave_configure,
@@ -839,7 +832,7 @@ struct scsi_host_template bfad_im_vport_template = {
 	.eh_timed_out = fc_eh_timed_out,
 	.eh_abort_handler = bfad_im_abort_handler,
 	.eh_device_reset_handler = bfad_im_reset_lun_handler,
-	.eh_bus_reset_handler = bfad_im_reset_bus_handler,
+	.eh_target_reset_handler = bfad_im_reset_target_handler,
 
 	.slave_alloc = bfad_im_slave_alloc,
 	.slave_configure = bfad_im_slave_configure,
-- 
1.8.5.6

Re: [PATCH 08/28] bfa: move bus reset to target reset

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 09/28] hptiop: Simplify reset handling

[Hannes Reinecke]

The Highpoint driver only has one reset function, and that is
a host reset. So stop pretending we're doing anything else.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/scsi/hptiop.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/drivers/scsi/hptiop.c b/drivers/scsi/hptiop.c
index 7226226..2fad7f0 100644
--- a/drivers/scsi/hptiop.c
+++ b/drivers/scsi/hptiop.c
@@ -1106,12 +1106,10 @@ static int hptiop_reset_hba(struct hptiop_hba *hba)
 
 static int hptiop_reset(struct scsi_cmnd *scp)
 {
-	struct Scsi_Host * host = scp->device->host;
-	struct hptiop_hba * hba = (struct hptiop_hba *)host->hostdata;
+	struct hptiop_hba * hba = (struct hptiop_hba *)scp->device->host->hostdata;
 
-	printk(KERN_WARNING "hptiop_reset(%d/%d/%d) scp=%p\n",
-			scp->device->host->host_no, scp->device->channel,
-			scp->device->id, scp);
+	printk(KERN_WARNING "hptiop_reset(%d/%d/%d)\n",
+	       scp->device->host->host_no, -1, -1);
 
 	return hptiop_reset_hba(hba)? FAILED : SUCCESS;
 }
@@ -1179,8 +1177,7 @@ static int hptiop_slave_config(struct scsi_device *sdev)
 	.module                     = THIS_MODULE,
 	.name                       = driver_name,
 	.queuecommand               = hptiop_queuecommand,
-	.eh_device_reset_handler    = hptiop_reset,
-	.eh_bus_reset_handler       = hptiop_reset,
+	.eh_host_reset_handler      = hptiop_reset,
 	.info                       = hptiop_info,
 	.emulated                   = 0,
 	.use_clustering             = ENABLE_CLUSTERING,
-- 
1.8.5.6

Re: [PATCH 09/28] hptiop: Simplify reset handling

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 10/28] fdomain: move bus reset to host reset

[Hannes Reinecke]

The bus reset function really is a host reset, so move it to
eh_host_reset_handler().

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/fdomain.c             | 6 +++---
 drivers/scsi/fdomain.h             | 2 +-
 drivers/scsi/pcmcia/fdomain_stub.c | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/scsi/fdomain.c b/drivers/scsi/fdomain.c
index b87ab38..ebbe5a3 100644
--- a/drivers/scsi/fdomain.c
+++ b/drivers/scsi/fdomain.c
@@ -933,7 +933,7 @@ struct Scsi_Host *__fdomain_16x0_detect(struct scsi_host_template *tpnt )
       }
    }
 
-   fdomain_16x0_bus_reset(NULL);
+   fdomain_16x0_host_reset(NULL);
 
    if (fdomain_test_loopback()) {
       printk(KERN_ERR  "scsi: <fdomain> Detection failed (loopback test failed at port base 0x%x)\n", port_base);
@@ -1568,7 +1568,7 @@ static int fdomain_16x0_abort(struct scsi_cmnd *SCpnt)
    return SUCCESS;
 }
 
-int fdomain_16x0_bus_reset(struct scsi_cmnd *SCpnt)
+int fdomain_16x0_host_reset(struct scsi_cmnd *SCpnt)
 {
    unsigned long flags;
 
@@ -1758,7 +1758,7 @@ struct scsi_host_template fdomain_driver_template = {
 	.info			= fdomain_16x0_info,
 	.queuecommand		= fdomain_16x0_queue,
 	.eh_abort_handler	= fdomain_16x0_abort,
-	.eh_bus_reset_handler	= fdomain_16x0_bus_reset,
+	.eh_host_reset_handler	= fdomain_16x0_host_reset,
 	.bios_param		= fdomain_16x0_biosparam,
 	.release		= fdomain_16x0_release,
 	.can_queue		= 1,
diff --git a/drivers/scsi/fdomain.h b/drivers/scsi/fdomain.h
index 47021d9..5cbe86b 100644
--- a/drivers/scsi/fdomain.h
+++ b/drivers/scsi/fdomain.h
@@ -21,4 +21,4 @@
 extern struct scsi_host_template fdomain_driver_template;
 extern int fdomain_setup(char *str);
 extern struct Scsi_Host *__fdomain_16x0_detect(struct  scsi_host_template *tpnt );
-extern int fdomain_16x0_bus_reset(struct scsi_cmnd *SCpnt);
+extern int fdomain_16x0_host_reset(struct scsi_cmnd *SCpnt);
diff --git a/drivers/scsi/pcmcia/fdomain_stub.c b/drivers/scsi/pcmcia/fdomain_stub.c
index 714b248..953a792 100644
--- a/drivers/scsi/pcmcia/fdomain_stub.c
+++ b/drivers/scsi/pcmcia/fdomain_stub.c
@@ -173,7 +173,7 @@ static void fdomain_release(struct pcmcia_device *link)
 
 static int fdomain_resume(struct pcmcia_device *link)
 {
-	fdomain_16x0_bus_reset(NULL);
+	fdomain_16x0_host_reset(NULL);
 
 	return 0;
 }
-- 
1.8.5.6

Re: [PATCH 10/28] fdomain: move bus reset to host reset

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

Re: [PATCH 10/28] fdomain: move bus reset to host reset

[Christoph Hellwig]

Looks fine,

Reviewed-by: Christoph Hellwig <hch@lst.de>

[PATCH 11/28] scsi: drop bus reset for wd33c93-compatible boards

[Hannes Reinecke]

The bus reset function is just a wrapper calling host reset
under the host lock. So move taking of the host lock into
the host reset function and drop bus reset.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/a2091.c   | 17 -----------------
 drivers/scsi/a3000.c   | 17 -----------------
 drivers/scsi/gvp11.c   | 18 ------------------
 drivers/scsi/mvme147.c | 16 ----------------
 drivers/scsi/sgiwd93.c | 15 ---------------
 drivers/scsi/wd33c93.c |  2 ++
 6 files changed, 2 insertions(+), 83 deletions(-)

diff --git a/drivers/scsi/a2091.c b/drivers/scsi/a2091.c
index 9176bfb..61aadc7 100644
--- a/drivers/scsi/a2091.c
+++ b/drivers/scsi/a2091.c
@@ -147,22 +147,6 @@ static void dma_stop(struct Scsi_Host *instance, struct scsi_cmnd *SCpnt,
 	}
 }
 
-static int a2091_bus_reset(struct scsi_cmnd *cmd)
-{
-	struct Scsi_Host *instance = cmd->device->host;
-
-	/* FIXME perform bus-specific reset */
-
-	/* FIXME 2: kill this function, and let midlayer fall back
-	   to the same action, calling wd33c93_host_reset() */
-
-	spin_lock_irq(instance->host_lock);
-	wd33c93_host_reset(cmd);
-	spin_unlock_irq(instance->host_lock);
-
-	return SUCCESS;
-}
-
 static struct scsi_host_template a2091_scsi_template = {
 	.module			= THIS_MODULE,
 	.name			= "Commodore A2091/A590 SCSI",
@@ -171,7 +155,6 @@ static int a2091_bus_reset(struct scsi_cmnd *cmd)
 	.proc_name		= "A2901",
 	.queuecommand		= wd33c93_queuecommand,
 	.eh_abort_handler	= wd33c93_abort,
-	.eh_bus_reset_handler	= a2091_bus_reset,
 	.eh_host_reset_handler	= wd33c93_host_reset,
 	.can_queue		= CAN_QUEUE,
 	.this_id		= 7,
diff --git a/drivers/scsi/a3000.c b/drivers/scsi/a3000.c
index e6375b4..2427a85 100644
--- a/drivers/scsi/a3000.c
+++ b/drivers/scsi/a3000.c
@@ -162,22 +162,6 @@ static void dma_stop(struct Scsi_Host *instance, struct scsi_cmnd *SCpnt,
 	}
 }
 
-static int a3000_bus_reset(struct scsi_cmnd *cmd)
-{
-	struct Scsi_Host *instance = cmd->device->host;
-
-	/* FIXME perform bus-specific reset */
-
-	/* FIXME 2: kill this entire function, which should
-	   cause mid-layer to call wd33c93_host_reset anyway? */
-
-	spin_lock_irq(instance->host_lock);
-	wd33c93_host_reset(cmd);
-	spin_unlock_irq(instance->host_lock);
-
-	return SUCCESS;
-}
-
 static struct scsi_host_template amiga_a3000_scsi_template = {
 	.module			= THIS_MODULE,
 	.name			= "Amiga 3000 built-in SCSI",
@@ -186,7 +170,6 @@ static int a3000_bus_reset(struct scsi_cmnd *cmd)
 	.proc_name		= "A3000",
 	.queuecommand		= wd33c93_queuecommand,
 	.eh_abort_handler	= wd33c93_abort,
-	.eh_bus_reset_handler	= a3000_bus_reset,
 	.eh_host_reset_handler	= wd33c93_host_reset,
 	.can_queue		= CAN_QUEUE,
 	.this_id		= 7,
diff --git a/drivers/scsi/gvp11.c b/drivers/scsi/gvp11.c
index 3b6f83f..a27fc49 100644
--- a/drivers/scsi/gvp11.c
+++ b/drivers/scsi/gvp11.c
@@ -171,23 +171,6 @@ static void dma_stop(struct Scsi_Host *instance, struct scsi_cmnd *SCpnt,
 	}
 }
 
-static int gvp11_bus_reset(struct scsi_cmnd *cmd)
-{
-	struct Scsi_Host *instance = cmd->device->host;
-
-	/* FIXME perform bus-specific reset */
-
-	/* FIXME 2: shouldn't we no-op this function (return
-	   FAILED), and fall back to host reset function,
-	   wd33c93_host_reset ? */
-
-	spin_lock_irq(instance->host_lock);
-	wd33c93_host_reset(cmd);
-	spin_unlock_irq(instance->host_lock);
-
-	return SUCCESS;
-}
-
 static struct scsi_host_template gvp11_scsi_template = {
 	.module			= THIS_MODULE,
 	.name			= "GVP Series II SCSI",
@@ -196,7 +179,6 @@ static int gvp11_bus_reset(struct scsi_cmnd *cmd)
 	.proc_name		= "GVP11",
 	.queuecommand		= wd33c93_queuecommand,
 	.eh_abort_handler	= wd33c93_abort,
-	.eh_bus_reset_handler	= gvp11_bus_reset,
 	.eh_host_reset_handler	= wd33c93_host_reset,
 	.can_queue		= CAN_QUEUE,
 	.this_id		= 7,
diff --git a/drivers/scsi/mvme147.c b/drivers/scsi/mvme147.c
index e7f6661..4f51570 100644
--- a/drivers/scsi/mvme147.c
+++ b/drivers/scsi/mvme147.c
@@ -121,21 +121,6 @@ int mvme147_detect(struct scsi_host_template *tpnt)
 	return 0;
 }
 
-static int mvme147_bus_reset(struct scsi_cmnd *cmd)
-{
-	/* FIXME perform bus-specific reset */
-
-	/* FIXME 2: kill this function, and let midlayer fallback to
-	   the same result, calling wd33c93_host_reset() */
-
-	spin_lock_irq(cmd->device->host->host_lock);
-	wd33c93_host_reset(cmd);
-	spin_unlock_irq(cmd->device->host->host_lock);
-
-	return SUCCESS;
-}
-
-
 static struct scsi_host_template driver_template = {
 	.proc_name		= "MVME147",
 	.name			= "MVME147 built-in SCSI",
@@ -143,7 +128,6 @@ static int mvme147_bus_reset(struct scsi_cmnd *cmd)
 	.release		= mvme147_release,
 	.queuecommand		= wd33c93_queuecommand,
 	.eh_abort_handler	= wd33c93_abort,
-	.eh_bus_reset_handler	= mvme147_bus_reset,
 	.eh_host_reset_handler	= wd33c93_host_reset,
 	.can_queue		= CAN_QUEUE,
 	.this_id		= 7,
diff --git a/drivers/scsi/sgiwd93.c b/drivers/scsi/sgiwd93.c
index 80cfa93..5ed696d 100644
--- a/drivers/scsi/sgiwd93.c
+++ b/drivers/scsi/sgiwd93.c
@@ -192,20 +192,6 @@ static inline void init_hpc_chain(struct ip22_hostdata *hdata)
 	hcp->desc.pnext = hdata->dma;
 }
 
-static int sgiwd93_bus_reset(struct scsi_cmnd *cmd)
-{
-	/* FIXME perform bus-specific reset */
-
-	/* FIXME 2: kill this function, and let midlayer fallback
-	   to the same result, calling wd33c93_host_reset() */
-
-	spin_lock_irq(cmd->device->host->host_lock);
-	wd33c93_host_reset(cmd);
-	spin_unlock_irq(cmd->device->host->host_lock);
-
-	return SUCCESS;
-}
-
 /*
  * Kludge alert - the SCSI code calls the abort and reset method with int
  * arguments not with pointers.  So this is going to blow up beautyfully
@@ -217,7 +203,6 @@ static int sgiwd93_bus_reset(struct scsi_cmnd *cmd)
 	.name			= "SGI WD93",
 	.queuecommand		= wd33c93_queuecommand,
 	.eh_abort_handler	= wd33c93_abort,
-	.eh_bus_reset_handler	= sgiwd93_bus_reset,
 	.eh_host_reset_handler	= wd33c93_host_reset,
 	.can_queue		= 16,
 	.this_id		= 7,
diff --git a/drivers/scsi/wd33c93.c b/drivers/scsi/wd33c93.c
index 9e09da4..74be04f 100644
--- a/drivers/scsi/wd33c93.c
+++ b/drivers/scsi/wd33c93.c
@@ -1578,6 +1578,7 @@
 	int i;
 
 	instance = SCpnt->device->host;
+	spin_lock_irq(instance->host_lock);
 	hostdata = (struct WD33C93_hostdata *) instance->hostdata;
 
 	printk("scsi%d: reset. ", instance->host_no);
@@ -1603,6 +1604,7 @@
 	reset_wd33c93(instance);
 	SCpnt->result = DID_RESET << 16;
 	enable_irq(instance->irq);
+	spin_unlock_irq(instance->host_lock);
 	return SUCCESS;
 }
 
-- 
1.8.5.6

Re: [PATCH 11/28] scsi: drop bus reset for wd33c93-compatible boards

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

Re: [PATCH 11/28] scsi: drop bus reset for wd33c93-compatible boards

[Christoph Hellwig]

On Wed, Jun 28, 2017 at 10:24:59AM +0200, Hannes Reinecke wrote:
> The bus reset function is just a wrapper calling host reset
> under the host lock. So move taking of the host lock into
> the host reset function and drop bus reset.

The changelog could be expanded a bit, especially about the
weird locking state before this patch, but except for that:

Reviewed-by: Christoph Hellwig <hch@lst.de>

[PATCH 12/28] rtsx: drop bus reset function

[Hannes Reinecke]

Function is a stub, so can as well be dropped.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/staging/rts5208/rtsx.c | 11 -----------
 1 file changed, 11 deletions(-)

diff --git a/drivers/staging/rts5208/rtsx.c b/drivers/staging/rts5208/rtsx.c
index b8177f5..e719b1e 100644
--- a/drivers/staging/rts5208/rtsx.c
+++ b/drivers/staging/rts5208/rtsx.c
@@ -205,16 +205,6 @@ static int device_reset(struct scsi_cmnd *srb)
 	return SUCCESS;
 }
 
-/* Simulate a SCSI bus reset by resetting the device's USB port. */
-static int bus_reset(struct scsi_cmnd *srb)
-{
-	struct rtsx_dev *dev = host_to_rtsx(srb->device->host);
-
-	dev_info(&dev->pci->dev, "%s called\n", __func__);
-
-	return SUCCESS;
-}
-
 /*
  * this defines our host template, with which we'll allocate hosts
  */
@@ -231,7 +221,6 @@ static int bus_reset(struct scsi_cmnd *srb)
 	/* error and abort handlers */
 	.eh_abort_handler =		command_abort,
 	.eh_device_reset_handler =	device_reset,
-	.eh_bus_reset_handler =		bus_reset,
 
 	/* queue commands only, only one command per LUN */
 	.can_queue =			1,
-- 
1.8.5.6

Re: [PATCH 12/28] rtsx: drop bus reset function

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 13/28] qlogicpti: move bus reset to host reset

[Hannes Reinecke]

The bus reset function really is a host reset, so move it to
eh_host_reset_handler().

Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/scsi/qlogicpti.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/qlogicpti.c b/drivers/scsi/qlogicpti.c
index 69bfc0a..274ee1cc 100644
--- a/drivers/scsi/qlogicpti.c
+++ b/drivers/scsi/qlogicpti.c
@@ -1250,10 +1250,9 @@ static int qlogicpti_abort(struct scsi_cmnd *Cmnd)
 	return return_status;
 }
 
-static int qlogicpti_reset(struct scsi_cmnd *Cmnd)
+static int qlogicpti_reset(struct Scsi_Host *host)
 {
 	u_short param[6];
-	struct Scsi_Host *host = Cmnd->device->host;
 	struct qlogicpti *qpti = (struct qlogicpti *) host->hostdata;
 	int return_status = SUCCESS;
 
@@ -1283,7 +1282,7 @@ static int qlogicpti_reset(struct scsi_cmnd *Cmnd)
 	.queuecommand		= qlogicpti_queuecommand,
 	.slave_configure	= qlogicpti_slave_configure,
 	.eh_abort_handler	= qlogicpti_abort,
-	.eh_bus_reset_handler	= qlogicpti_reset,
+	.eh_host_reset_handler	= qlogicpti_reset,
 	.can_queue		= QLOGICPTI_REQ_QUEUE_LEN,
 	.this_id		= 7,
 	.sg_tablesize		= QLOGICPTI_MAX_SG(QLOGICPTI_REQ_QUEUE_LEN),
-- 
1.8.5.6

Re: [PATCH 13/28] qlogicpti: move bus reset to host reset

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 14/28] acornscsi: move bus reset to host reset

[Hannes Reinecke]

The bus reset function is really a host reset, so move it to
eh_host_reset_handler.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/scsi/arm/acornscsi.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/scsi/arm/acornscsi.c b/drivers/scsi/arm/acornscsi.c
index 12b8829..690816f 100644
--- a/drivers/scsi/arm/acornscsi.c
+++ b/drivers/scsi/arm/acornscsi.c
@@ -2725,23 +2725,24 @@ int acornscsi_abort(struct scsi_cmnd *SCpnt)
  * Params   : SCpnt  - command causing reset
  * Returns  : one of SCSI_RESET_ macros
  */
-int acornscsi_bus_reset(struct scsi_cmnd *SCpnt)
+int acornscsi_host_reset(struct Scsi_Host *shpnt)
 {
-	AS_Host *host = (AS_Host *)SCpnt->device->host->hostdata;
+	AS_Host *host = (AS_Host *)shpnt->hostdata;
 	struct scsi_cmnd *SCptr;
     
     host->stats.resets += 1;
 
 #if (DEBUG & DEBUG_RESET)
     {
-	int asr, ssr;
+	int asr, ssr, devidx;
 
 	asr = sbic_arm_read(host, SBIC_ASR);
 	ssr = sbic_arm_read(host, SBIC_SSR);
 
 	printk(KERN_WARNING "acornscsi_reset: ");
 	print_sbic_status(asr, ssr, host->scsi.phase);
-	acornscsi_dumplog(host, SCpnt->device->id);
+	for (devidx = 0; devidx < 9; devidx ++) {
+	    acornscsi_dumplog(host, devidx);
     }
 #endif
 
@@ -2884,7 +2885,7 @@ static int acornscsi_show_info(struct seq_file *m, struct Scsi_Host *instance)
 	.info			= acornscsi_info,
 	.queuecommand		= acornscsi_queuecmd,
 	.eh_abort_handler	= acornscsi_abort,
-	.eh_bus_reset_handler	= acornscsi_bus_reset,
+	.eh_host_reset_handler	= acornscsi_host_reset,
 	.can_queue		= 16,
 	.this_id		= 7,
 	.sg_tablesize		= SG_ALL,
-- 
1.8.5.6

Re: [PATCH 14/28] acornscsi: move bus reset to host reset

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 15/28] NCR5380: Move bus reset to host reset

[Hannes Reinecke]

The bus reset handler really is a host reset, so move it to
eh_bus_reset_handler.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Acked-by: Finn Thain <fthain@telegraphics.com.au>
---
 drivers/scsi/NCR5380.c      | 4 ++--
 drivers/scsi/arm/cumana_1.c | 2 +-
 drivers/scsi/arm/oak.c      | 2 +-
 drivers/scsi/atari_scsi.c   | 6 +++---
 drivers/scsi/dmx3191d.c     | 2 +-
 drivers/scsi/g_NCR5380.c    | 4 ++--
 drivers/scsi/mac_scsi.c     | 4 ++--
 drivers/scsi/sun3_scsi.c    | 4 ++--
 8 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/drivers/scsi/NCR5380.c b/drivers/scsi/NCR5380.c
index acc3344..8a08122 100644
--- a/drivers/scsi/NCR5380.c
+++ b/drivers/scsi/NCR5380.c
@@ -2296,13 +2296,13 @@ static int NCR5380_abort(struct scsi_cmnd *cmd)
 
 
 /**
- * NCR5380_bus_reset - reset the SCSI bus
+ * NCR5380_host_reset - reset the SCSI host
  * @cmd: SCSI command undergoing EH
  *
  * Returns SUCCESS
  */
 
-static int NCR5380_bus_reset(struct scsi_cmnd *cmd)
+static int NCR5380_host_reset(struct scsi_cmnd *cmd)
 {
 	struct Scsi_Host *instance = cmd->device->host;
 	struct NCR5380_hostdata *hostdata = shost_priv(instance);
diff --git a/drivers/scsi/arm/cumana_1.c b/drivers/scsi/arm/cumana_1.c
index a87b99c..ae1d809 100644
--- a/drivers/scsi/arm/cumana_1.c
+++ b/drivers/scsi/arm/cumana_1.c
@@ -216,7 +216,7 @@ static void cumanascsi_write(struct NCR5380_hostdata *hostdata,
 	.info			= cumanascsi_info,
 	.queuecommand		= cumanascsi_queue_command,
 	.eh_abort_handler	= NCR5380_abort,
-	.eh_bus_reset_handler	= NCR5380_bus_reset,
+	.eh_host_reset_handler	= NCR5380_host_reset,
 	.can_queue		= 16,
 	.this_id		= 7,
 	.sg_tablesize		= SG_ALL,
diff --git a/drivers/scsi/arm/oak.c b/drivers/scsi/arm/oak.c
index 6be6666..05b7f75 100644
--- a/drivers/scsi/arm/oak.c
+++ b/drivers/scsi/arm/oak.c
@@ -105,7 +105,7 @@ static inline int oakscsi_pread(struct NCR5380_hostdata *hostdata,
 	.info			= oakscsi_info,
 	.queuecommand		= oakscsi_queue_command,
 	.eh_abort_handler	= NCR5380_abort,
-	.eh_bus_reset_handler	= NCR5380_bus_reset,
+	.eh_host_reset_handler	= NCR5380_host_reset,
 	.can_queue		= 16,
 	.this_id		= 7,
 	.sg_tablesize		= SG_ALL,
diff --git a/drivers/scsi/atari_scsi.c b/drivers/scsi/atari_scsi.c
index a75feeb..89f5154 100644
--- a/drivers/scsi/atari_scsi.c
+++ b/drivers/scsi/atari_scsi.c
@@ -671,7 +671,7 @@ static void atari_scsi_falcon_reg_write(unsigned int reg, u8 value)
 
 #include "NCR5380.c"
 
-static int atari_scsi_bus_reset(struct scsi_cmnd *cmd)
+static int atari_scsi_host_reset(struct scsi_cmnd *cmd)
 {
 	int rv;
 	unsigned long flags;
@@ -688,7 +688,7 @@ static int atari_scsi_bus_reset(struct scsi_cmnd *cmd)
 		atari_dma_orig_addr = NULL;
 	}
 
-	rv = NCR5380_bus_reset(cmd);
+	rv = NCR5380_host_reset(cmd);
 
 	/* The 5380 raises its IRQ line while _RST is active but the ST DMA
 	 * "lock" has been released so this interrupt may end up handled by
@@ -711,7 +711,7 @@ static int atari_scsi_bus_reset(struct scsi_cmnd *cmd)
 	.info			= atari_scsi_info,
 	.queuecommand		= atari_scsi_queue_command,
 	.eh_abort_handler	= atari_scsi_abort,
-	.eh_bus_reset_handler	= atari_scsi_bus_reset,
+	.eh_host_reset_handler	= atari_scsi_host_reset,
 	.this_id		= 7,
 	.cmd_per_lun		= 2,
 	.use_clustering		= DISABLE_CLUSTERING,
diff --git a/drivers/scsi/dmx3191d.c b/drivers/scsi/dmx3191d.c
index 6af3394..003c3d7 100644
--- a/drivers/scsi/dmx3191d.c
+++ b/drivers/scsi/dmx3191d.c
@@ -58,7 +58,7 @@
 	.info			= NCR5380_info,
 	.queuecommand		= NCR5380_queue_command,
 	.eh_abort_handler	= NCR5380_abort,
-	.eh_bus_reset_handler	= NCR5380_bus_reset,
+	.eh_host_reset_handler	= NCR5380_host_reset,
 	.can_queue		= 32,
 	.this_id		= 7,
 	.sg_tablesize		= SG_ALL,
diff --git a/drivers/scsi/g_NCR5380.c b/drivers/scsi/g_NCR5380.c
index c34fc91..4965a46 100644
--- a/drivers/scsi/g_NCR5380.c
+++ b/drivers/scsi/g_NCR5380.c
@@ -54,7 +54,7 @@
 #define NCR5380_intr                    generic_NCR5380_intr
 #define NCR5380_queue_command           generic_NCR5380_queue_command
 #define NCR5380_abort                   generic_NCR5380_abort
-#define NCR5380_bus_reset               generic_NCR5380_bus_reset
+#define NCR5380_host_reset              generic_NCR5380_host_reset
 #define NCR5380_info                    generic_NCR5380_info
 
 #define NCR5380_io_delay(x)             udelay(x)
@@ -661,7 +661,7 @@ static int generic_NCR5380_dma_xfer_len(struct NCR5380_hostdata *hostdata,
 	.info			= generic_NCR5380_info,
 	.queuecommand		= generic_NCR5380_queue_command,
 	.eh_abort_handler	= generic_NCR5380_abort,
-	.eh_bus_reset_handler	= generic_NCR5380_bus_reset,
+	.eh_host_reset_handler	= generic_NCR5380_host_reset,
 	.can_queue		= 16,
 	.this_id		= 7,
 	.sg_tablesize		= SG_ALL,
diff --git a/drivers/scsi/mac_scsi.c b/drivers/scsi/mac_scsi.c
index 196acc7..dd60573 100644
--- a/drivers/scsi/mac_scsi.c
+++ b/drivers/scsi/mac_scsi.c
@@ -41,7 +41,7 @@
 #define NCR5380_intr                    macscsi_intr
 #define NCR5380_queue_command           macscsi_queue_command
 #define NCR5380_abort                   macscsi_abort
-#define NCR5380_bus_reset               macscsi_bus_reset
+#define NCR5380_host_reset              macscsi_host_reset
 #define NCR5380_info                    macscsi_info
 
 #include "NCR5380.h"
@@ -328,7 +328,7 @@ static int macscsi_dma_residual(struct NCR5380_hostdata *hostdata)
 	.info			= macscsi_info,
 	.queuecommand		= macscsi_queue_command,
 	.eh_abort_handler	= macscsi_abort,
-	.eh_bus_reset_handler	= macscsi_bus_reset,
+	.eh_host_reset_handler	= macscsi_host_reset,
 	.can_queue		= 16,
 	.this_id		= 7,
 	.sg_tablesize		= 1,
diff --git a/drivers/scsi/sun3_scsi.c b/drivers/scsi/sun3_scsi.c
index e64b0c5..9492638 100644
--- a/drivers/scsi/sun3_scsi.c
+++ b/drivers/scsi/sun3_scsi.c
@@ -46,7 +46,7 @@
 #define NCR5380_write(reg, value)       out_8(hostdata->io + (reg), value)
 
 #define NCR5380_queue_command           sun3scsi_queue_command
-#define NCR5380_bus_reset               sun3scsi_bus_reset
+#define NCR5380_host_reset              sun3scsi_host_reset
 #define NCR5380_abort                   sun3scsi_abort
 #define NCR5380_info                    sun3scsi_info
 
@@ -495,7 +495,7 @@ static int sun3scsi_dma_finish(int write_flag)
 	.info			= sun3scsi_info,
 	.queuecommand		= sun3scsi_queue_command,
 	.eh_abort_handler	= sun3scsi_abort,
-	.eh_bus_reset_handler	= sun3scsi_bus_reset,
+	.eh_host_reset_handler	= sun3scsi_host_reset,
 	.can_queue		= 16,
 	.this_id		= 7,
 	.sg_tablesize		= SG_NONE,
-- 
1.8.5.6

Re: [PATCH 15/28] NCR5380: Move bus reset to host reset

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

Re: [PATCH 15/28] NCR5380: Move bus reset to host reset

[Christoph Hellwig]

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

[PATCH 16/28] qlogicfas: move bus_reset to host_reset

[Hannes Reinecke]

From: Hannes Reinecke <hare@suse.com>

The bus reset handler is really a host reset.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/scsi/pcmcia/qlogic_stub.c | 4 ++--
 drivers/scsi/qlogicfas.c          | 2 +-
 drivers/scsi/qlogicfas408.c       | 6 +++---
 drivers/scsi/qlogicfas408.h       | 2 +-
 4 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/scsi/pcmcia/qlogic_stub.c b/drivers/scsi/pcmcia/qlogic_stub.c
index c670dc7..0556054 100644
--- a/drivers/scsi/pcmcia/qlogic_stub.c
+++ b/drivers/scsi/pcmcia/qlogic_stub.c
@@ -67,7 +67,7 @@
 	.info			= qlogicfas408_info,
 	.queuecommand		= qlogicfas408_queuecommand,
 	.eh_abort_handler	= qlogicfas408_abort,
-	.eh_bus_reset_handler	= qlogicfas408_bus_reset,
+	.eh_host_reset_handler	= qlogicfas408_host_reset,
 	.bios_param		= qlogicfas408_biosparam,
 	.can_queue		= 1,
 	.this_id		= -1,
@@ -264,7 +264,7 @@ static int qlogic_resume(struct pcmcia_device *link)
 		outb(0x04, link->resource[0]->start + 0xd);
 	}
 	/* Ugggglllyyyy!!! */
-	qlogicfas408_bus_reset(NULL);
+	qlogicfas408_host_reset(NULL);
 
 	return 0;
 }
diff --git a/drivers/scsi/qlogicfas.c b/drivers/scsi/qlogicfas.c
index 840823b..95431d6 100644
--- a/drivers/scsi/qlogicfas.c
+++ b/drivers/scsi/qlogicfas.c
@@ -188,7 +188,7 @@ static int qlogicfas_release(struct Scsi_Host *shost)
 	.info			= qlogicfas408_info,
 	.queuecommand		= qlogicfas408_queuecommand,
 	.eh_abort_handler	= qlogicfas408_abort,
-	.eh_bus_reset_handler	= qlogicfas408_bus_reset,
+	.eh_host_reset_handler	= qlogicfas408_host_reset,
 	.bios_param		= qlogicfas408_biosparam,
 	.can_queue		= 1,
 	.this_id		= -1,
diff --git a/drivers/scsi/qlogicfas408.c b/drivers/scsi/qlogicfas408.c
index c3a9151..8b471a9 100644
--- a/drivers/scsi/qlogicfas408.c
+++ b/drivers/scsi/qlogicfas408.c
@@ -496,13 +496,13 @@ int qlogicfas408_abort(struct scsi_cmnd *cmd)
 	return SUCCESS;
 }
 
-/* 
+/*
  *	Reset SCSI bus
  *	FIXME: This function is invoked with cmd = NULL directly by
  *	the PCMCIA qlogic_stub code. This wants fixing
  */
 
-int qlogicfas408_bus_reset(struct scsi_cmnd *cmd)
+int qlogicfas408_host_reset(struct scsi_cmnd *cmd)
 {
 	struct qlogicfas408_priv *priv = get_priv_by_cmd(cmd);
 	unsigned long flags;
@@ -607,7 +607,7 @@ static void __exit qlogicfas408_exit(void)
 EXPORT_SYMBOL(qlogicfas408_info);
 EXPORT_SYMBOL(qlogicfas408_queuecommand);
 EXPORT_SYMBOL(qlogicfas408_abort);
-EXPORT_SYMBOL(qlogicfas408_bus_reset);
+EXPORT_SYMBOL(qlogicfas408_host_reset);
 EXPORT_SYMBOL(qlogicfas408_biosparam);
 EXPORT_SYMBOL(qlogicfas408_ihandl);
 EXPORT_SYMBOL(qlogicfas408_get_chip_type);
diff --git a/drivers/scsi/qlogicfas408.h b/drivers/scsi/qlogicfas408.h
index 2f6c0a1..f6b1216 100644
--- a/drivers/scsi/qlogicfas408.h
+++ b/drivers/scsi/qlogicfas408.h
@@ -108,7 +108,7 @@ int qlogicfas408_biosparam(struct scsi_device * disk,
 			   struct block_device *dev,
 			   sector_t capacity, int ip[]);
 int qlogicfas408_abort(struct scsi_cmnd * cmd);
-int qlogicfas408_bus_reset(struct scsi_cmnd * cmd);
+extern int qlogicfas408_host_reset(struct scsi_cmnd *cmd);
 const char *qlogicfas408_info(struct Scsi_Host *host);
 int qlogicfas408_get_chip_type(int qbase, int int_type);
 void qlogicfas408_setup(int qbase, int id, int int_type);
-- 
1.8.5.6

Re: [PATCH 16/28] qlogicfas: move bus_reset to host_reset

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 17/28] imm: drop duplicate bus_reset handler

[Hannes Reinecke]

host_reset and bus_reset is the same function, so drop bus reset.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/imm.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/scsi/imm.c b/drivers/scsi/imm.c
index 9164ce12..87c9419 100644
--- a/drivers/scsi/imm.c
+++ b/drivers/scsi/imm.c
@@ -1106,7 +1106,6 @@ static int imm_adjust_queue(struct scsi_device *device)
 	.name			= "Iomega VPI2 (imm) interface",
 	.queuecommand		= imm_queuecommand,
 	.eh_abort_handler	= imm_abort,
-	.eh_bus_reset_handler	= imm_reset,
 	.eh_host_reset_handler	= imm_reset,
 	.bios_param		= imm_biosparam,
 	.this_id		= 7,
-- 
1.8.5.6

Re: [PATCH 17/28] imm: drop duplicate bus_reset handler

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

Re: [PATCH 17/28] imm: drop duplicate bus_reset handler

[Christoph Hellwig]

On Wed, Jun 28, 2017 at 10:25:05AM +0200, Hannes Reinecke wrote:
> host_reset and bus_reset is the same function, so drop bus reset.
> 
> Signed-off-by: Hannes Reinecke <hare@suse.com>

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

[PATCH 18/28] ppa: drop duplicate bus_reset handler

[Hannes Reinecke]

bus_reset and host_reset are the same functions, so drop bus_reset.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/ppa.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/scsi/ppa.c b/drivers/scsi/ppa.c
index f6ad579..7be5823 100644
--- a/drivers/scsi/ppa.c
+++ b/drivers/scsi/ppa.c
@@ -970,7 +970,6 @@ static int ppa_adjust_queue(struct scsi_device *device)
 	.name			= "Iomega VPI0 (ppa) interface",
 	.queuecommand		= ppa_queuecommand,
 	.eh_abort_handler	= ppa_abort,
-	.eh_bus_reset_handler	= ppa_reset,
 	.eh_host_reset_handler	= ppa_reset,
 	.bios_param		= ppa_biosparam,
 	.this_id		= -1,
-- 
1.8.5.6

Re: [PATCH 18/28] ppa: drop duplicate bus_reset handler

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

Re: [PATCH 18/28] ppa: drop duplicate bus_reset handler

[Christoph Hellwig]

On Wed, Jun 28, 2017 at 10:25:06AM +0200, Hannes Reinecke wrote:
> bus_reset and host_reset are the same functions, so drop bus_reset.
> 
> Signed-off-by: Hannes Reinecke <hare@suse.com>

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

[PATCH 19/28] qedf: drop bus reset handler

[Hannes Reinecke]

qedf has a host reset handler, but as the bus reset handler is
a stub always returning SUCCESS the host reset is never invoked.
So drop the bus reset handler.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Cc: Chad Dupuis <chad.dupuis@cavium.com>
---
 drivers/scsi/qedf/qedf_main.c | 11 -----------
 1 file changed, 11 deletions(-)

diff --git a/drivers/scsi/qedf/qedf_main.c b/drivers/scsi/qedf/qedf_main.c
index 6149ea0..d48a13a 100644
--- a/drivers/scsi/qedf/qedf_main.c
+++ b/drivers/scsi/qedf/qedf_main.c
@@ -629,16 +629,6 @@ static int qedf_eh_device_reset(struct scsi_cmnd *sc_cmd)
 	return qedf_initiate_tmf(sc_cmd, FCP_TMF_LUN_RESET);
 }
 
-static int qedf_eh_bus_reset(struct scsi_cmnd *sc_cmd)
-{
-	QEDF_ERR(NULL, "BUS RESET Issued...\n");
-	/*
-	 * Essentially a no-op but return SUCCESS to prevent
-	 * unnecessary escalation to the host reset handler.
-	 */
-	return SUCCESS;
-}
-
 void qedf_wait_for_upload(struct qedf_ctx *qedf)
 {
 	while (1) {
@@ -716,7 +706,6 @@ static int qedf_slave_configure(struct scsi_device *sdev)
 	.eh_abort_handler	= qedf_eh_abort,
 	.eh_device_reset_handler = qedf_eh_device_reset, /* lun reset */
 	.eh_target_reset_handler = qedf_eh_target_reset, /* target reset */
-	.eh_bus_reset_handler = qedf_eh_bus_reset,
 	.eh_host_reset_handler  = qedf_eh_host_reset,
 	.slave_configure	= qedf_slave_configure,
 	.dma_boundary = QED_HW_DMA_BOUNDARY,
-- 
1.8.5.6

Re: [PATCH 19/28] qedf: drop bus reset handler

[Christoph Hellwig]

On Wed, Jun 28, 2017 at 10:25:07AM +0200, Hannes Reinecke wrote:
> qedf has a host reset handler, but as the bus reset handler is
> a stub always returning SUCCESS the host reset is never invoked.
> So drop the bus reset handler.

Given that comment this needs a lot more explanation, as the current
behavior seems to be an odd/buggy workaround.

I hope this patch won't block the rest of the series, which all seems
nice and simple material for this late merge window :(

Re: [PATCH 19/28] qedf: drop bus reset handler

[Chad Dupuis]

On Wed, 28 Jun 2017, 9:58am, Christoph Hellwig wrote:

> On Wed, Jun 28, 2017 at 10:25:07AM +0200, Hannes Reinecke wrote:
> > qedf has a host reset handler, but as the bus reset handler is
> > a stub always returning SUCCESS the host reset is never invoked.
> > So drop the bus reset handler.
> 
> Given that comment this needs a lot more explanation, as the current
> behavior seems to be an odd/buggy workaround.
> 
> I hope this patch won't block the rest of the series, which all seems
> nice and simple material for this late merge window :(
> 

The test case this was meant to fix as using the sg_reset utility with the 
-b option which forces a bus reset with NPIV ports attached.  What we were 
finding is that some of the devices attached to the vport were being 
brought offline after 5 seconds as escalating to the host reset handler in 
this case was taking too long.

For this specific test case, returning SUCCESS allowed us to bypass the 
host reset escalation and force the SCSI error handler to send a TUR to 
the devices which of course would succeed since the device weren't 
physically offline keeping them from going offline.

Re: [PATCH 19/28] qedf: drop bus reset handler

[Christoph Hellwig]

On Wed, Jun 28, 2017 at 10:07:50AM -0400, Chad Dupuis wrote:
> The test case this was meant to fix as using the sg_reset utility with the 
> -b option which forces a bus reset with NPIV ports attached.  What we were 
> finding is that some of the devices attached to the vport were being 
> brought offline after 5 seconds as escalating to the host reset handler in 
> this case was taking too long.

sg_reset / SG_SCSI_RESET are designed to escalatae to to the bigger
hammers if the method isn't implemented or doesn't succeed.

Use the -N / --no-esc option to disable that.  I wish that behavior had
been the default from the beginning, but we can't fix that anymore.

Re: [PATCH 19/28] qedf: drop bus reset handler

[Chad Dupuis]

On Wed, 28 Jun 2017, 10:37am, Christoph Hellwig wrote:

> 
> sg_reset / SG_SCSI_RESET are designed to escalatae to to the bigger
> hammers if the method isn't implemented or doesn't succeed.
> 
> Use the -N / --no-esc option to disable that.  I wish that behavior had
> been the default from the beginning, but we can't fix that anymore.
> 

Yeah, for a sg_reset prespective that would be the better behavior.  I 
removed the bus_reset handler and used the --no-esc parameter which causes 
sg_reset to behave more reasonably:

# sg_reset -b --no-esc /dev/sdj
sg_reset: reset (for value=0x102) may not be available

Given this I don't have an objection to this patch.

Tested-by: Chad Dupuis <chad.dupuis@cavium.com>

[PATCH 20/28] nsp32: drop bus reset

[Hannes Reinecke]

From: Hannes Reinecke <hare@suse.com>

bus reset is essentially identical to host reset, and will always
return SUCCESS, thus disabling the use of host reset.
So drop bus reset in favour of host reset.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/nsp32.c | 22 +---------------------
 1 file changed, 1 insertion(+), 21 deletions(-)

diff --git a/drivers/scsi/nsp32.c b/drivers/scsi/nsp32.c
index 53c8477..107e191 100644
--- a/drivers/scsi/nsp32.c
+++ b/drivers/scsi/nsp32.c
@@ -201,7 +201,6 @@
 
 /* SCSI error handler */
 static int         nsp32_eh_abort     (struct scsi_cmnd *);
-static int         nsp32_eh_bus_reset (struct scsi_cmnd *);
 static int         nsp32_eh_host_reset(struct scsi_cmnd *);
 
 /* generate SCSI message */
@@ -276,8 +275,7 @@
 	.max_sectors			= 128,
 	.this_id			= NSP32_HOST_SCSIID,
 	.use_clustering			= DISABLE_CLUSTERING,
-	.eh_abort_handler       	= nsp32_eh_abort,
-	.eh_bus_reset_handler		= nsp32_eh_bus_reset,
+	.eh_abort_handler		= nsp32_eh_abort,
 	.eh_host_reset_handler		= nsp32_eh_host_reset,
 /*	.highmem_io			= 1, */
 };
@@ -2845,24 +2843,6 @@ static int nsp32_eh_abort(struct scsi_cmnd *SCpnt)
 	return SUCCESS;
 }
 
-static int nsp32_eh_bus_reset(struct scsi_cmnd *SCpnt)
-{
-	nsp32_hw_data *data = (nsp32_hw_data *)SCpnt->device->host->hostdata;
-	unsigned int   base = SCpnt->device->host->io_port;
-
-	spin_lock_irq(SCpnt->device->host->host_lock);
-
-	nsp32_msg(KERN_INFO, "Bus Reset");	
-	nsp32_dbg(NSP32_DEBUG_BUSRESET, "SCpnt=0x%x", SCpnt);
-
-	nsp32_write2(base, IRQ_CONTROL, IRQ_CONTROL_ALL_IRQ_MASK);
-	nsp32_do_bus_reset(data);
-	nsp32_write2(base, IRQ_CONTROL, 0);
-
-	spin_unlock_irq(SCpnt->device->host->host_lock);
-	return SUCCESS;	/* SCSI bus reset is succeeded at any time. */
-}
-
 static void nsp32_do_bus_reset(nsp32_hw_data *data)
 {
 	unsigned int   base = data->BaseAddress;
-- 
1.8.5.6

Re: [PATCH 20/28] nsp32: drop bus reset

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

Re: [PATCH 20/28] nsp32: drop bus reset

[Christoph Hellwig]

On Wed, Jun 28, 2017 at 10:25:08AM +0200, Hannes Reinecke wrote:
> From: Hannes Reinecke <hare@suse.com>
> 
> bus reset is essentially identical to host reset, and will always
> return SUCCESS, thus disabling the use of host reset.
> So drop bus reset in favour of host reset.
> 
> Signed-off-by: Hannes Reinecke <hare@suse.com>

nsp32_eh_host_reset additionally calls nsp32hw_init, so they
don't really look "mostly" the same.  That being said I think
the bus_reset as-is is probably rather useless.

[PATCH 21/28] aha152x: drop bus reset

[Hannes Reinecke]

From: Hannes Reinecke <hare@suse.com>

The bus reset function is a subset of host reset, but will always
return SUCCESS. But as the host reset also invokes a reset for
all ports there is a possibility that the bus reset will not be
able to fully reset the HBA, leading to a blocked system.
So drop bus reset completely.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/aha152x.c | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/drivers/scsi/aha152x.c b/drivers/scsi/aha152x.c
index ce5dc73..21e9d7c 100644
--- a/drivers/scsi/aha152x.c
+++ b/drivers/scsi/aha152x.c
@@ -1165,15 +1165,6 @@ static int aha152x_bus_reset_host(struct Scsi_Host *shpnt)
 }
 
 /*
- * Reset the bus
- *
- */
-static int aha152x_bus_reset(Scsi_Cmnd *SCpnt)
-{
-	return aha152x_bus_reset_host(SCpnt->device->host);
-}
-
-/*
  *  Restore default values to the AIC-6260 registers and reset the fifos
  *
  */
@@ -2916,7 +2907,6 @@ static int aha152x_adjust_queue(struct scsi_device *device)
 	.queuecommand			= aha152x_queue,
 	.eh_abort_handler		= aha152x_abort,
 	.eh_device_reset_handler	= aha152x_device_reset,
-	.eh_bus_reset_handler		= aha152x_bus_reset,
 	.eh_host_reset_handler		= aha152x_host_reset,
 	.bios_param			= aha152x_biosparam,
 	.can_queue			= 1,
-- 
1.8.5.6

Re: [PATCH 21/28] aha152x: drop bus reset

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

Re: [PATCH 21/28] aha152x: drop bus reset

[Christoph Hellwig]

On Wed, Jun 28, 2017 at 10:25:09AM +0200, Hannes Reinecke wrote:
> From: Hannes Reinecke <hare@suse.com>
> 
> The bus reset function is a subset of host reset, but will always
> return SUCCESS. But as the host reset also invokes a reset for
> all ports there is a possibility that the bus reset will not be
> able to fully reset the HBA, leading to a blocked system.
> So drop bus reset completely.

Wouldn't the more conservative option be to remove the host_reset
handler?  This driver is old and little used, and the existing
code will never have used the host_reset handler due to the fact
that the bus_reset handler always returns SUCCESS?

[PATCH 22/28] 53c700: move bus reset to host reset

[Hannes Reinecke]

bus reset always returns SUCCESS, meaning host reset was never
tested. At the same time the only difference to the HBA is a
missing call to NCR_700_chip_reset().
So add the missing call to bus reset, drop host reset, and move
bus reset to host reset.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/53c700.c | 22 +++-------------------
 1 file changed, 3 insertions(+), 19 deletions(-)

diff --git a/drivers/scsi/53c700.c b/drivers/scsi/53c700.c
index 4b3b080..2b8cc32 100644
--- a/drivers/scsi/53c700.c
+++ b/drivers/scsi/53c700.c
@@ -168,7 +168,6 @@
 
 STATIC int NCR_700_queuecommand(struct Scsi_Host *h, struct scsi_cmnd *);
 STATIC int NCR_700_abort(struct scsi_cmnd * SCpnt);
-STATIC int NCR_700_bus_reset(struct scsi_cmnd * SCpnt);
 STATIC int NCR_700_host_reset(struct scsi_cmnd * SCpnt);
 STATIC void NCR_700_chip_setup(struct Scsi_Host *host);
 STATIC void NCR_700_chip_reset(struct Scsi_Host *host);
@@ -1938,14 +1937,14 @@ STATIC DEF_SCSI_QCMD(NCR_700_queuecommand)
 }
 
 STATIC int
-NCR_700_bus_reset(struct scsi_cmnd * SCp)
+NCR_700_host_reset(struct scsi_cmnd * SCp)
 {
 	DECLARE_COMPLETION_ONSTACK(complete);
 	struct NCR_700_Host_Parameters *hostdata = 
 		(struct NCR_700_Host_Parameters *)SCp->device->host->hostdata[0];
 
 	scmd_printk(KERN_INFO, SCp,
-		"New error handler wants BUS reset, cmd %p\n\t", SCp);
+		"New error handler wants HOST reset, cmd %p\n\t", SCp);
 	scsi_print_command(SCp);
 
 	/* In theory, eh_complete should always be null because the
@@ -1960,6 +1959,7 @@ STATIC DEF_SCSI_QCMD(NCR_700_queuecommand)
 
 	hostdata->eh_complete = &complete;
 	NCR_700_internal_bus_reset(SCp->device->host);
+	NCR_700_chip_reset(SCp->device->host);
 
 	spin_unlock_irq(SCp->device->host->host_lock);
 	wait_for_completion(&complete);
@@ -1974,22 +1974,6 @@ STATIC DEF_SCSI_QCMD(NCR_700_queuecommand)
 	return SUCCESS;
 }
 
-STATIC int
-NCR_700_host_reset(struct scsi_cmnd * SCp)
-{
-	scmd_printk(KERN_INFO, SCp, "New error handler wants HOST reset\n\t");
-	scsi_print_command(SCp);
-
-	spin_lock_irq(SCp->device->host->host_lock);
-
-	NCR_700_internal_bus_reset(SCp->device->host);
-	NCR_700_chip_reset(SCp->device->host);
-
-	spin_unlock_irq(SCp->device->host->host_lock);
-
-	return SUCCESS;
-}
-
 STATIC void
 NCR_700_set_period(struct scsi_target *STp, int period)
 {
-- 
1.8.5.6

Re: [PATCH 22/28] 53c700: move bus reset to host reset

[Christoph Hellwig]

On Wed, Jun 28, 2017 at 10:25:10AM +0200, Hannes Reinecke wrote:
> bus reset always returns SUCCESS, meaning host reset was never
> tested. At the same time the only difference to the HBA is a
> missing call to NCR_700_chip_reset().
> So add the missing call to bus reset, drop host reset, and move
> bus reset to host reset.

Given that all the 53c700-based drivers only support a single channel
this looks correct:

Reviewed-by: Christoph Hellwig <hch@lst.de>

[PATCH 23/28] bnx2fc: remove obsolete bnx2fc_eh_host_reset() definition

[Hannes Reinecke]

Never used anywhere.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 drivers/scsi/bnx2fc/bnx2fc.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/scsi/bnx2fc/bnx2fc.h b/drivers/scsi/bnx2fc/bnx2fc.h
index 7e007e1..901a316 100644
--- a/drivers/scsi/bnx2fc/bnx2fc.h
+++ b/drivers/scsi/bnx2fc/bnx2fc.h
@@ -539,7 +539,6 @@ void bnx2fc_init_task(struct bnx2fc_cmd *io_req,
 void bnx2fc_add_2_sq(struct bnx2fc_rport *tgt, u16 xid);
 void bnx2fc_ring_doorbell(struct bnx2fc_rport *tgt);
 int bnx2fc_eh_abort(struct scsi_cmnd *sc_cmd);
-int bnx2fc_eh_host_reset(struct scsi_cmnd *sc_cmd);
 int bnx2fc_eh_target_reset(struct scsi_cmnd *sc_cmd);
 int bnx2fc_eh_device_reset(struct scsi_cmnd *sc_cmd);
 void bnx2fc_rport_event_handler(struct fc_lport *lport,
-- 
1.8.5.6

Re: [PATCH 23/28] bnx2fc: remove obsolete bnx2fc_eh_host_reset() definition

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH 24/28] megaraid_mbox: drop duplicate bus reset and device reset function

[Hannes Reinecke]

megaraid_mbox only has one reset function, and that is a host reset.
So drop the duplicate bus reset and device reset functions.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/megaraid/megaraid_mbox.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/scsi/megaraid/megaraid_mbox.c b/drivers/scsi/megaraid/megaraid_mbox.c
index f0987f2..3f914b1 100644
--- a/drivers/scsi/megaraid/megaraid_mbox.c
+++ b/drivers/scsi/megaraid/megaraid_mbox.c
@@ -341,8 +341,6 @@ static void megaraid_mbox_prepare_epthru(adapter_t *, scb_t *,
 	.proc_name			= "megaraid",
 	.queuecommand			= megaraid_queue_command,
 	.eh_abort_handler		= megaraid_abort_handler,
-	.eh_device_reset_handler	= megaraid_reset_handler,
-	.eh_bus_reset_handler		= megaraid_reset_handler,
 	.eh_host_reset_handler		= megaraid_reset_handler,
 	.change_queue_depth		= scsi_change_queue_depth,
 	.use_clustering			= ENABLE_CLUSTERING,
-- 
1.8.5.6

Re: [PATCH 24/28] megaraid_mbox: drop duplicate bus reset and device reset function

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

Re: [PATCH 24/28] megaraid_mbox: drop duplicate bus reset and device reset function

[Christoph Hellwig]

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

[PATCH 25/28] lpfc: drop lpfc_no_handler and lpfc_bus_reset_handler()

[Hannes Reinecke]

lpfc_bus_reset_handler() is really just a loop calling
lpfc_target_reset_handler() over all targets, which is what
the error handler will be doing anyway.
And lpfc_no_handler() will return FAILED, just like SCSI EH
does if no handler is present.

So drop both.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Cc: James Smart <james.smart@broadcom.com>
---
 drivers/scsi/lpfc/lpfc_scsi.c | 99 -------------------------------------------
 1 file changed, 99 deletions(-)

diff --git a/drivers/scsi/lpfc/lpfc_scsi.c b/drivers/scsi/lpfc/lpfc_scsi.c
index cfe1d01..19b1202 100644
--- a/drivers/scsi/lpfc/lpfc_scsi.c
+++ b/drivers/scsi/lpfc/lpfc_scsi.c
@@ -5294,93 +5294,6 @@ void lpfc_poll_timeout(unsigned long ptr)
 }
 
 /**
- * lpfc_bus_reset_handler - scsi_host_template eh_bus_reset_handler entry point
- * @cmnd: Pointer to scsi_cmnd data structure.
- *
- * This routine does target reset to all targets on @cmnd->device->host.
- * This emulates Parallel SCSI Bus Reset Semantics.
- *
- * Return code :
- *  0x2003 - Error
- *  0x2002 - Success
- **/
-static int
-lpfc_bus_reset_handler(struct scsi_cmnd *cmnd)
-{
-	struct Scsi_Host  *shost = cmnd->device->host;
-	struct lpfc_vport *vport = (struct lpfc_vport *) shost->hostdata;
-	struct lpfc_nodelist *ndlp = NULL;
-	struct lpfc_scsi_event_header scsi_event;
-	int match;
-	int ret = SUCCESS, status, i;
-
-	scsi_event.event_type = FC_REG_SCSI_EVENT;
-	scsi_event.subcategory = LPFC_EVENT_BUSRESET;
-	scsi_event.lun = 0;
-	memcpy(scsi_event.wwpn, &vport->fc_portname, sizeof(struct lpfc_name));
-	memcpy(scsi_event.wwnn, &vport->fc_nodename, sizeof(struct lpfc_name));
-
-	fc_host_post_vendor_event(shost, fc_get_event_number(),
-		sizeof(scsi_event), (char *)&scsi_event, LPFC_NL_VENDOR_ID);
-
-	status = fc_block_scsi_eh(cmnd);
-	if (status != 0 && status != SUCCESS)
-		return status;
-
-	/*
-	 * Since the driver manages a single bus device, reset all
-	 * targets known to the driver.  Should any target reset
-	 * fail, this routine returns failure to the midlayer.
-	 */
-	for (i = 0; i < LPFC_MAX_TARGET; i++) {
-		/* Search for mapped node by target ID */
-		match = 0;
-		spin_lock_irq(shost->host_lock);
-		list_for_each_entry(ndlp, &vport->fc_nodes, nlp_listp) {
-			if (!NLP_CHK_NODE_ACT(ndlp))
-				continue;
-			if (vport->phba->cfg_fcp2_no_tgt_reset &&
-			    (ndlp->nlp_fcp_info & NLP_FCP_2_DEVICE))
-				continue;
-			if (ndlp->nlp_state == NLP_STE_MAPPED_NODE &&
-			    ndlp->nlp_sid == i &&
-			    ndlp->rport &&
-			    ndlp->nlp_type & NLP_FCP_TARGET) {
-				match = 1;
-				break;
-			}
-		}
-		spin_unlock_irq(shost->host_lock);
-		if (!match)
-			continue;
-
-		status = lpfc_send_taskmgmt(vport, cmnd,
-					i, 0, FCP_TARGET_RESET);
-
-		if (status != SUCCESS) {
-			lpfc_printf_vlog(vport, KERN_ERR, LOG_FCP,
-					 "0700 Bus Reset on target %d failed\n",
-					 i);
-			ret = FAILED;
-		}
-	}
-	/*
-	 * We have to clean up i/o as : they may be orphaned by the TMFs
-	 * above; or if any of the TMFs failed, they may be in an
-	 * indeterminate state.
-	 * We will report success if all the i/o aborts successfully.
-	 */
-
-	status = lpfc_reset_flush_io_context(vport, 0, 0, LPFC_CTX_HOST);
-	if (status != SUCCESS)
-		ret = FAILED;
-
-	lpfc_printf_vlog(vport, KERN_ERR, LOG_FCP,
-			 "0714 SCSI layer issued Bus Reset Data: x%x\n", ret);
-	return ret;
-}
-
-/**
  * lpfc_host_reset_handler - scsi_host_template eh_host_reset_handler entry pt
  * @cmnd: Pointer to scsi_cmnd data structure.
  *
@@ -5926,11 +5839,6 @@ struct lpfc_device_data*
 	return SCSI_MLQUEUE_HOST_BUSY;
 }
 
-static int
-lpfc_no_handler(struct scsi_cmnd *cmnd)
-{
-	return FAILED;
-}
 
 static int
 lpfc_no_slave(struct scsi_device *sdev)
@@ -5944,11 +5852,6 @@ struct scsi_host_template lpfc_template_nvme = {
 	.proc_name		= LPFC_DRIVER_NAME,
 	.info			= lpfc_info,
 	.queuecommand		= lpfc_no_command,
-	.eh_abort_handler	= lpfc_no_handler,
-	.eh_device_reset_handler = lpfc_no_handler,
-	.eh_target_reset_handler = lpfc_no_handler,
-	.eh_bus_reset_handler	= lpfc_no_handler,
-	.eh_host_reset_handler  = lpfc_no_handler,
 	.slave_alloc		= lpfc_no_slave,
 	.slave_configure	= lpfc_no_slave,
 	.scan_finished		= lpfc_scan_finished,
@@ -5972,7 +5875,6 @@ struct scsi_host_template lpfc_template_no_hr = {
 	.eh_abort_handler	= lpfc_abort_handler,
 	.eh_device_reset_handler = lpfc_device_reset_handler,
 	.eh_target_reset_handler = lpfc_target_reset_handler,
-	.eh_bus_reset_handler	= lpfc_bus_reset_handler,
 	.slave_alloc		= lpfc_slave_alloc,
 	.slave_configure	= lpfc_slave_configure,
 	.slave_destroy		= lpfc_slave_destroy,
@@ -5998,7 +5900,6 @@ struct scsi_host_template lpfc_template = {
 	.eh_abort_handler	= lpfc_abort_handler,
 	.eh_device_reset_handler = lpfc_device_reset_handler,
 	.eh_target_reset_handler = lpfc_target_reset_handler,
-	.eh_bus_reset_handler	= lpfc_bus_reset_handler,
 	.eh_host_reset_handler  = lpfc_host_reset_handler,
 	.slave_alloc		= lpfc_slave_alloc,
 	.slave_configure	= lpfc_slave_configure,
-- 
1.8.5.6

[PATCH 26/28] visorhba: sanitze private device data allocation

[Hannes Reinecke]

There's no need to keep the private data for a device in a separate
list; better to store it in ->hostdata and do away with the additional
list.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Cc: David Kershner <david.kershner@unisys.com>
---
 drivers/staging/unisys/visorhba/visorhba_main.c | 123 ++++++++++--------------
 1 file changed, 53 insertions(+), 70 deletions(-)

diff --git a/drivers/staging/unisys/visorhba/visorhba_main.c b/drivers/staging/unisys/visorhba/visorhba_main.c
index 6997b16..811abfc 100644
--- a/drivers/staging/unisys/visorhba/visorhba_main.c
+++ b/drivers/staging/unisys/visorhba/visorhba_main.c
@@ -47,8 +47,8 @@
 MODULE_ALIAS("visorbus:" SPAR_VHBA_CHANNEL_PROTOCOL_UUID_STR);
 
 struct visordisk_info {
+	struct scsi_device *sdev;
 	u32 valid;
-	u32 channel, id, lun;	/* Disk Path */
 	atomic_t ios_threshold;
 	atomic_t error_count;
 	struct visordisk_info *next;
@@ -101,12 +101,6 @@ struct visorhba_devices_open {
 	struct visorhba_devdata *devdata;
 };
 
-#define for_each_vdisk_match(iter, list, match) \
-	for (iter = &list->head; iter->next; iter = iter->next) \
-		if ((iter->channel == match->channel) && \
-		    (iter->id == match->id) && \
-		    (iter->lun == match->lun))
-
 /*
  *	visor_thread_start - starts a thread for the device
  *	@threadfn: Function the thread starts
@@ -296,10 +290,9 @@ static void cleanup_scsitaskmgmt_handles(struct idr *idrtable,
  *	Returns whether the command was queued successfully or not.
  */
 static int forward_taskmgmt_command(enum task_mgmt_types tasktype,
-				    struct scsi_cmnd *scsicmd)
+				    struct scsi_device *scsidev)
 {
 	struct uiscmdrsp *cmdrsp;
-	struct scsi_device *scsidev = scsicmd->device;
 	struct visorhba_devdata *devdata =
 		(struct visorhba_devdata *)scsidev->host->hostdata;
 	int notifyresult = 0xffff;
@@ -347,12 +340,6 @@ static int forward_taskmgmt_command(enum task_mgmt_types tasktype,
 	dev_dbg(&scsidev->sdev_gendev,
 		"visorhba: taskmgmt type=%d success; result=0x%x\n",
 		 tasktype, notifyresult);
-	if (tasktype == TASK_MGMT_ABORT_TASK)
-		scsicmd->result = DID_ABORT << 16;
-	else
-		scsicmd->result = DID_RESET << 16;
-
-	scsicmd->scsi_done(scsicmd);
 	cleanup_scsitaskmgmt_handles(&devdata->idr, cmdrsp);
 	return SUCCESS;
 
@@ -376,17 +363,20 @@ static int visorhba_abort_handler(struct scsi_cmnd *scsicmd)
 	/* issue TASK_MGMT_ABORT_TASK */
 	struct scsi_device *scsidev;
 	struct visordisk_info *vdisk;
-	struct visorhba_devdata *devdata;
+	int rtn;
 
 	scsidev = scsicmd->device;
-	devdata = (struct visorhba_devdata *)scsidev->host->hostdata;
-	for_each_vdisk_match(vdisk, devdata, scsidev) {
-		if (atomic_read(&vdisk->error_count) < VISORHBA_ERROR_COUNT)
-			atomic_inc(&vdisk->error_count);
-		else
-			atomic_set(&vdisk->ios_threshold, IOS_ERROR_THRESHOLD);
+	vdisk = scsidev->hostdata;
+	if (atomic_read(&vdisk->error_count) < VISORHBA_ERROR_COUNT)
+		atomic_inc(&vdisk->error_count);
+	else
+		atomic_set(&vdisk->ios_threshold, IOS_ERROR_THRESHOLD);
+	rtn = forward_taskmgmt_command(TASK_MGMT_ABORT_TASK, scsidev);
+	if (rtn == SUCCESS) {
+		scsicmd->result = DID_ABORT << 16;
+		scsicmd->scsi_done(scsicmd);
 	}
-	return forward_taskmgmt_command(TASK_MGMT_ABORT_TASK, scsicmd);
+	return rtn;
 }
 
 /*
@@ -400,17 +390,20 @@ static int visorhba_device_reset_handler(struct scsi_cmnd *scsicmd)
 	/* issue TASK_MGMT_LUN_RESET */
 	struct scsi_device *scsidev;
 	struct visordisk_info *vdisk;
-	struct visorhba_devdata *devdata;
+	int rtn;
 
 	scsidev = scsicmd->device;
-	devdata = (struct visorhba_devdata *)scsidev->host->hostdata;
-	for_each_vdisk_match(vdisk, devdata, scsidev) {
-		if (atomic_read(&vdisk->error_count) < VISORHBA_ERROR_COUNT)
-			atomic_inc(&vdisk->error_count);
-		else
-			atomic_set(&vdisk->ios_threshold, IOS_ERROR_THRESHOLD);
+	vdisk = scsidev->hostdata;
+	if (atomic_read(&vdisk->error_count) < VISORHBA_ERROR_COUNT)
+		atomic_inc(&vdisk->error_count);
+	else
+		atomic_set(&vdisk->ios_threshold, IOS_ERROR_THRESHOLD);
+	rtn = forward_taskmgmt_command(TASK_MGMT_LUN_RESET, scsidev);
+	if (rtn == SUCCESS) {
+		scsicmd->result = DID_RESET << 16;
+		scsicmd->scsi_done(scsicmd);
 	}
-	return forward_taskmgmt_command(TASK_MGMT_LUN_RESET, scsicmd);
+	return rtn;
 }
 
 /*
@@ -424,17 +417,22 @@ static int visorhba_bus_reset_handler(struct scsi_cmnd *scsicmd)
 {
 	struct scsi_device *scsidev;
 	struct visordisk_info *vdisk;
-	struct visorhba_devdata *devdata;
+	int rtn;
 
 	scsidev = scsicmd->device;
-	devdata = (struct visorhba_devdata *)scsidev->host->hostdata;
-	for_each_vdisk_match(vdisk, devdata, scsidev) {
+	shost_for_each_device(scsidev, scsidev->host) {
+		vdisk = scsidev->hostdata;
 		if (atomic_read(&vdisk->error_count) < VISORHBA_ERROR_COUNT)
 			atomic_inc(&vdisk->error_count);
 		else
 			atomic_set(&vdisk->ios_threshold, IOS_ERROR_THRESHOLD);
 	}
-	return forward_taskmgmt_command(TASK_MGMT_BUS_RESET, scsicmd);
+	rtn = forward_taskmgmt_command(TASK_MGMT_BUS_RESET, scsidev);
+	if (rtn == SUCCESS) {
+		scsicmd->result = DID_RESET << 16;
+		scsicmd->scsi_done(scsicmd);
+	}
+	return rtn;
 }
 
 /*
@@ -569,25 +567,22 @@ static int visorhba_slave_alloc(struct scsi_device *scsidev)
 	 * LLD can alloc any struct & do init if needed.
 	 */
 	struct visordisk_info *vdisk;
-	struct visordisk_info *tmpvdisk;
 	struct visorhba_devdata *devdata;
 	struct Scsi_Host *scsihost = (struct Scsi_Host *)scsidev->host;
 
+	if (scsidev->hostdata)
+		return 0; /* already allocated return success */
+
 	devdata = (struct visorhba_devdata *)scsihost->hostdata;
 	if (!devdata)
 		return 0; /* even though we errored, treat as success */
 
-	for_each_vdisk_match(vdisk, devdata, scsidev)
-		return 0; /* already allocated return success */
-
-	tmpvdisk = kzalloc(sizeof(*tmpvdisk), GFP_ATOMIC);
-	if (!tmpvdisk)
+	vdisk = kzalloc(sizeof(*vdisk), GFP_ATOMIC);
+	if (!vdisk)
 		return -ENOMEM;
 
-	tmpvdisk->channel = scsidev->channel;
-	tmpvdisk->id = scsidev->id;
-	tmpvdisk->lun = scsidev->lun;
-	vdisk->next = tmpvdisk;
+	vdisk->sdev = scsidev;
+	scsidev->hostdata = vdisk;
 	return 0;
 }
 
@@ -603,17 +598,11 @@ static void visorhba_slave_destroy(struct scsi_device *scsidev)
 	/* midlevel calls this after device has been quiesced and
 	 * before it is to be deleted.
 	 */
-	struct visordisk_info *vdisk, *delvdisk;
-	struct visorhba_devdata *devdata;
-	struct Scsi_Host *scsihost = (struct Scsi_Host *)scsidev->host;
+	struct visordisk_info *vdisk;
 
-	devdata = (struct visorhba_devdata *)scsihost->hostdata;
-	for_each_vdisk_match(vdisk, devdata, scsidev) {
-		delvdisk = vdisk->next;
-		vdisk->next = delvdisk->next;
-		kfree(delvdisk);
-		return;
-	}
+	vdisk = scsidev->hostdata;
+	scsidev->hostdata = NULL;
+	kfree(vdisk);
 }
 
 static struct scsi_host_template visorhba_driver_template = {
@@ -787,7 +776,6 @@ static int visorhba_serverdown(struct visorhba_devdata *devdata)
 static void
 do_scsi_linuxstat(struct uiscmdrsp *cmdrsp, struct scsi_cmnd *scsicmd)
 {
-	struct visorhba_devdata *devdata;
 	struct visordisk_info *vdisk;
 	struct scsi_device *scsidev;
 
@@ -800,12 +788,10 @@ static int visorhba_serverdown(struct visorhba_devdata *devdata)
 	    (cmdrsp->scsi.addlstat == ADDL_SEL_TIMEOUT))
 		return;
 	/* Okay see what our error_count is here.... */
-	devdata = (struct visorhba_devdata *)scsidev->host->hostdata;
-	for_each_vdisk_match(vdisk, devdata, scsidev) {
-		if (atomic_read(&vdisk->error_count) < VISORHBA_ERROR_COUNT) {
-			atomic_inc(&vdisk->error_count);
-			atomic_set(&vdisk->ios_threshold, IOS_ERROR_THRESHOLD);
-		}
+	vdisk = scsidev->hostdata;
+	if (atomic_read(&vdisk->error_count) < VISORHBA_ERROR_COUNT) {
+		atomic_inc(&vdisk->error_count);
+		atomic_set(&vdisk->ios_threshold, IOS_ERROR_THRESHOLD);
 	}
 }
 
@@ -846,7 +832,6 @@ static int set_no_disk_inquiry_result(unsigned char *buf,
 	char *this_page_orig;
 	int bufind = 0;
 	struct visordisk_info *vdisk;
-	struct visorhba_devdata *devdata;
 
 	scsidev = scsicmd->device;
 	if ((cmdrsp->scsi.cmnd[0] == INQUIRY) &&
@@ -883,13 +868,11 @@ static int set_no_disk_inquiry_result(unsigned char *buf,
 		}
 		kfree(buf);
 	} else {
-		devdata = (struct visorhba_devdata *)scsidev->host->hostdata;
-		for_each_vdisk_match(vdisk, devdata, scsidev) {
-			if (atomic_read(&vdisk->ios_threshold) > 0) {
-				atomic_dec(&vdisk->ios_threshold);
-				if (atomic_read(&vdisk->ios_threshold) == 0)
-					atomic_set(&vdisk->error_count, 0);
-			}
+		vdisk = scsidev->hostdata;
+		if (atomic_read(&vdisk->ios_threshold) > 0) {
+			atomic_dec(&vdisk->ios_threshold);
+			if (atomic_read(&vdisk->ios_threshold) == 0)
+				atomic_set(&vdisk->error_count, 0);
 		}
 	}
 }
-- 
1.8.5.6

RE: [PATCH 26/28] visorhba: sanitze private device data allocation

[Kershner, David A]

> -----Original Message-----
> From: Hannes Reinecke [mailto:hare@suse.de]
> Sent: Wednesday, June 28, 2017 4:25 AM
> To: Christoph Hellwig <hch@lst.de>
> Cc: Martin K. Petersen <martin.petersen@oracle.com>; James Bottomley
> <james.bottomley@hansenpartnership.com>; linux-scsi@vger.kernel.org;
> Hannes Reinecke <hare@suse.de>; Hannes Reinecke <hare@suse.com>;
> Kershner, David A <David.Kershner@unisys.com>
> Subject: [PATCH 26/28] visorhba: sanitze private device data allocation
> 
> There's no need to keep the private data for a device in a separate
> list; better to store it in ->hostdata and do away with the additional
> list.
> 
> Signed-off-by: Hannes Reinecke <hare@suse.com>
> Cc: David Kershner <david.kershner@unisys.com>

Thanks for the patch, works great!  Tested it on s-Par. 

Acked-by: David Kershner <david.kershner@unisys.com>

Thanks, 
David Kershner

> ---
>  drivers/staging/unisys/visorhba/visorhba_main.c | 123 ++++++++++----------
> ----
>  1 file changed, 53 insertions(+), 70 deletions(-)
> 
> diff --git a/drivers/staging/unisys/visorhba/visorhba_main.c
> b/drivers/staging/unisys/visorhba/visorhba_main.c
> index 6997b16..811abfc 100644
> --- a/drivers/staging/unisys/visorhba/visorhba_main.c
> +++ b/drivers/staging/unisys/visorhba/visorhba_main.c
> @@ -47,8 +47,8 @@
>  MODULE_ALIAS("visorbus:"
> SPAR_VHBA_CHANNEL_PROTOCOL_UUID_STR);
> 
>  struct visordisk_info {
> +	struct scsi_device *sdev;
>  	u32 valid;
> -	u32 channel, id, lun;	/* Disk Path */
>  	atomic_t ios_threshold;
>  	atomic_t error_count;
>  	struct visordisk_info *next;
> @@ -101,12 +101,6 @@ struct visorhba_devices_open {
>  	struct visorhba_devdata *devdata;
>  };
> 
> -#define for_each_vdisk_match(iter, list, match) \
> -	for (iter = &list->head; iter->next; iter = iter->next) \
> -		if ((iter->channel == match->channel) && \
> -		    (iter->id == match->id) && \
> -		    (iter->lun == match->lun))
> -
>  /*
>   *	visor_thread_start - starts a thread for the device
>   *	@threadfn: Function the thread starts
> @@ -296,10 +290,9 @@ static void cleanup_scsitaskmgmt_handles(struct idr
> *idrtable,
>   *	Returns whether the command was queued successfully or not.
>   */
>  static int forward_taskmgmt_command(enum task_mgmt_types tasktype,
> -				    struct scsi_cmnd *scsicmd)
> +				    struct scsi_device *scsidev)
>  {
>  	struct uiscmdrsp *cmdrsp;
> -	struct scsi_device *scsidev = scsicmd->device;
>  	struct visorhba_devdata *devdata =
>  		(struct visorhba_devdata *)scsidev->host->hostdata;
>  	int notifyresult = 0xffff;
> @@ -347,12 +340,6 @@ static int forward_taskmgmt_command(enum
> task_mgmt_types tasktype,
>  	dev_dbg(&scsidev->sdev_gendev,
>  		"visorhba: taskmgmt type=%d success; result=0x%x\n",
>  		 tasktype, notifyresult);
> -	if (tasktype == TASK_MGMT_ABORT_TASK)
> -		scsicmd->result = DID_ABORT << 16;
> -	else
> -		scsicmd->result = DID_RESET << 16;
> -
> -	scsicmd->scsi_done(scsicmd);
>  	cleanup_scsitaskmgmt_handles(&devdata->idr, cmdrsp);
>  	return SUCCESS;
> 
> @@ -376,17 +363,20 @@ static int visorhba_abort_handler(struct scsi_cmnd
> *scsicmd)
>  	/* issue TASK_MGMT_ABORT_TASK */
>  	struct scsi_device *scsidev;
>  	struct visordisk_info *vdisk;
> -	struct visorhba_devdata *devdata;
> +	int rtn;
> 
>  	scsidev = scsicmd->device;
> -	devdata = (struct visorhba_devdata *)scsidev->host->hostdata;
> -	for_each_vdisk_match(vdisk, devdata, scsidev) {
> -		if (atomic_read(&vdisk->error_count) <
> VISORHBA_ERROR_COUNT)
> -			atomic_inc(&vdisk->error_count);
> -		else
> -			atomic_set(&vdisk->ios_threshold,
> IOS_ERROR_THRESHOLD);
> +	vdisk = scsidev->hostdata;
> +	if (atomic_read(&vdisk->error_count) < VISORHBA_ERROR_COUNT)
> +		atomic_inc(&vdisk->error_count);
> +	else
> +		atomic_set(&vdisk->ios_threshold,
> IOS_ERROR_THRESHOLD);
> +	rtn = forward_taskmgmt_command(TASK_MGMT_ABORT_TASK,
> scsidev);
> +	if (rtn == SUCCESS) {
> +		scsicmd->result = DID_ABORT << 16;
> +		scsicmd->scsi_done(scsicmd);
>  	}
> -	return forward_taskmgmt_command(TASK_MGMT_ABORT_TASK,
> scsicmd);
> +	return rtn;
>  }
> 
>  /*
> @@ -400,17 +390,20 @@ static int visorhba_device_reset_handler(struct
> scsi_cmnd *scsicmd)
>  	/* issue TASK_MGMT_LUN_RESET */
>  	struct scsi_device *scsidev;
>  	struct visordisk_info *vdisk;
> -	struct visorhba_devdata *devdata;
> +	int rtn;
> 
>  	scsidev = scsicmd->device;
> -	devdata = (struct visorhba_devdata *)scsidev->host->hostdata;
> -	for_each_vdisk_match(vdisk, devdata, scsidev) {
> -		if (atomic_read(&vdisk->error_count) <
> VISORHBA_ERROR_COUNT)
> -			atomic_inc(&vdisk->error_count);
> -		else
> -			atomic_set(&vdisk->ios_threshold,
> IOS_ERROR_THRESHOLD);
> +	vdisk = scsidev->hostdata;
> +	if (atomic_read(&vdisk->error_count) < VISORHBA_ERROR_COUNT)
> +		atomic_inc(&vdisk->error_count);
> +	else
> +		atomic_set(&vdisk->ios_threshold,
> IOS_ERROR_THRESHOLD);
> +	rtn = forward_taskmgmt_command(TASK_MGMT_LUN_RESET,
> scsidev);
> +	if (rtn == SUCCESS) {
> +		scsicmd->result = DID_RESET << 16;
> +		scsicmd->scsi_done(scsicmd);
>  	}
> -	return forward_taskmgmt_command(TASK_MGMT_LUN_RESET,
> scsicmd);
> +	return rtn;
>  }
> 
>  /*
> @@ -424,17 +417,22 @@ static int visorhba_bus_reset_handler(struct
> scsi_cmnd *scsicmd)
>  {
>  	struct scsi_device *scsidev;
>  	struct visordisk_info *vdisk;
> -	struct visorhba_devdata *devdata;
> +	int rtn;
> 
>  	scsidev = scsicmd->device;
> -	devdata = (struct visorhba_devdata *)scsidev->host->hostdata;
> -	for_each_vdisk_match(vdisk, devdata, scsidev) {
> +	shost_for_each_device(scsidev, scsidev->host) {
> +		vdisk = scsidev->hostdata;
>  		if (atomic_read(&vdisk->error_count) <
> VISORHBA_ERROR_COUNT)
>  			atomic_inc(&vdisk->error_count);
>  		else
>  			atomic_set(&vdisk->ios_threshold,
> IOS_ERROR_THRESHOLD);
>  	}
> -	return forward_taskmgmt_command(TASK_MGMT_BUS_RESET,
> scsicmd);
> +	rtn = forward_taskmgmt_command(TASK_MGMT_BUS_RESET,
> scsidev);
> +	if (rtn == SUCCESS) {
> +		scsicmd->result = DID_RESET << 16;
> +		scsicmd->scsi_done(scsicmd);
> +	}
> +	return rtn;
>  }
> 
>  /*
> @@ -569,25 +567,22 @@ static int visorhba_slave_alloc(struct scsi_device
> *scsidev)
>  	 * LLD can alloc any struct & do init if needed.
>  	 */
>  	struct visordisk_info *vdisk;
> -	struct visordisk_info *tmpvdisk;
>  	struct visorhba_devdata *devdata;
>  	struct Scsi_Host *scsihost = (struct Scsi_Host *)scsidev->host;
> 
> +	if (scsidev->hostdata)
> +		return 0; /* already allocated return success */
> +
>  	devdata = (struct visorhba_devdata *)scsihost->hostdata;
>  	if (!devdata)
>  		return 0; /* even though we errored, treat as success */
> 
> -	for_each_vdisk_match(vdisk, devdata, scsidev)
> -		return 0; /* already allocated return success */
> -
> -	tmpvdisk = kzalloc(sizeof(*tmpvdisk), GFP_ATOMIC);
> -	if (!tmpvdisk)
> +	vdisk = kzalloc(sizeof(*vdisk), GFP_ATOMIC);
> +	if (!vdisk)
>  		return -ENOMEM;
> 
> -	tmpvdisk->channel = scsidev->channel;
> -	tmpvdisk->id = scsidev->id;
> -	tmpvdisk->lun = scsidev->lun;
> -	vdisk->next = tmpvdisk;
> +	vdisk->sdev = scsidev;
> +	scsidev->hostdata = vdisk;
>  	return 0;
>  }
> 
> @@ -603,17 +598,11 @@ static void visorhba_slave_destroy(struct
> scsi_device *scsidev)
>  	/* midlevel calls this after device has been quiesced and
>  	 * before it is to be deleted.
>  	 */
> -	struct visordisk_info *vdisk, *delvdisk;
> -	struct visorhba_devdata *devdata;
> -	struct Scsi_Host *scsihost = (struct Scsi_Host *)scsidev->host;
> +	struct visordisk_info *vdisk;
> 
> -	devdata = (struct visorhba_devdata *)scsihost->hostdata;
> -	for_each_vdisk_match(vdisk, devdata, scsidev) {
> -		delvdisk = vdisk->next;
> -		vdisk->next = delvdisk->next;
> -		kfree(delvdisk);
> -		return;
> -	}
> +	vdisk = scsidev->hostdata;
> +	scsidev->hostdata = NULL;
> +	kfree(vdisk);
>  }
> 
>  static struct scsi_host_template visorhba_driver_template = {
> @@ -787,7 +776,6 @@ static int visorhba_serverdown(struct
> visorhba_devdata *devdata)
>  static void
>  do_scsi_linuxstat(struct uiscmdrsp *cmdrsp, struct scsi_cmnd *scsicmd)
>  {
> -	struct visorhba_devdata *devdata;
>  	struct visordisk_info *vdisk;
>  	struct scsi_device *scsidev;
> 
> @@ -800,12 +788,10 @@ static int visorhba_serverdown(struct
> visorhba_devdata *devdata)
>  	    (cmdrsp->scsi.addlstat == ADDL_SEL_TIMEOUT))
>  		return;
>  	/* Okay see what our error_count is here.... */
> -	devdata = (struct visorhba_devdata *)scsidev->host->hostdata;
> -	for_each_vdisk_match(vdisk, devdata, scsidev) {
> -		if (atomic_read(&vdisk->error_count) <
> VISORHBA_ERROR_COUNT) {
> -			atomic_inc(&vdisk->error_count);
> -			atomic_set(&vdisk->ios_threshold,
> IOS_ERROR_THRESHOLD);
> -		}
> +	vdisk = scsidev->hostdata;
> +	if (atomic_read(&vdisk->error_count) < VISORHBA_ERROR_COUNT)
> {
> +		atomic_inc(&vdisk->error_count);
> +		atomic_set(&vdisk->ios_threshold,
> IOS_ERROR_THRESHOLD);
>  	}
>  }
> 
> @@ -846,7 +832,6 @@ static int set_no_disk_inquiry_result(unsigned char
> *buf,
>  	char *this_page_orig;
>  	int bufind = 0;
>  	struct visordisk_info *vdisk;
> -	struct visorhba_devdata *devdata;
> 
>  	scsidev = scsicmd->device;
>  	if ((cmdrsp->scsi.cmnd[0] == INQUIRY) &&
> @@ -883,13 +868,11 @@ static int set_no_disk_inquiry_result(unsigned char
> *buf,
>  		}
>  		kfree(buf);
>  	} else {
> -		devdata = (struct visorhba_devdata *)scsidev->host-
> >hostdata;
> -		for_each_vdisk_match(vdisk, devdata, scsidev) {
> -			if (atomic_read(&vdisk->ios_threshold) > 0) {
> -				atomic_dec(&vdisk->ios_threshold);
> -				if (atomic_read(&vdisk->ios_threshold) == 0)
> -					atomic_set(&vdisk->error_count, 0);
> -			}
> +		vdisk = scsidev->hostdata;
> +		if (atomic_read(&vdisk->ios_threshold) > 0) {
> +			atomic_dec(&vdisk->ios_threshold);
> +			if (atomic_read(&vdisk->ios_threshold) == 0)
> +				atomic_set(&vdisk->error_count, 0);
>  		}
>  	}
>  }
> --
> 1.8.5.6

Re: [PATCH 26/28] visorhba: sanitze private device data allocation

[Hannes Reinecke]

On 07/19/2017 07:45 PM, Kershner, David A wrote:
> 
> 
>> -----Original Message-----
>> From: Hannes Reinecke [mailto:hare@suse.de]
>> Sent: Wednesday, June 28, 2017 4:25 AM
>> To: Christoph Hellwig <hch@lst.de>
>> Subject: [PATCH 26/28] visorhba: sanitze private device data allocation
>>
>> There's no need to keep the private data for a device in a separate
>> list; better to store it in ->hostdata and do away with the additional
>> list.
>>
>> Signed-off-by: Hannes Reinecke <hare@suse.com>
>> Cc: David Kershner <david.kershner@unisys.com>
> 
> I'm not sure what the process for this patch was to get into the upstream, do
> you know the forecast for it? I.e. when it will be available in Linus's tree? 
> 
> We have several more changes that I would like to make to this driver and
> to simplify them, I would like to put them on top of this change, would you
> mind if I submitted it to GregKH for the staging branch if it is going to take
> some time?
> 
> Sorry for the newbie questions.
> 
typically the maintainer (that'll be you :-) does a review, and replies
with something like:

Patch is okay;

Reviewed-by: David Kershner <XXX@YYY.com>

Then the subsystem maintainer (ie Martin Petersen) will be picking it
up, and queueing it in his subsystem tree for eventual merge with linux
upstream.

The first step has been completed, so we're waiting for Martin to pick
it up. It might be some time, though; depending on the merge window and
his overall workload.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		   Teamlead Storage & Networking
hare@suse.de			               +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)

[PATCH 27/28] eata: remove 'arg_done' from eata2x_eh_host_reset()

[Hannes Reinecke]

Just displaying some different information; drop it.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/eata.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/drivers/scsi/eata.c b/drivers/scsi/eata.c
index 227dd2c..6501c33 100644
--- a/drivers/scsi/eata.c
+++ b/drivers/scsi/eata.c
@@ -1899,7 +1899,6 @@ static int eata2x_eh_abort(struct scsi_cmnd *SCarg)
 static int eata2x_eh_host_reset(struct scsi_cmnd *SCarg)
 {
 	unsigned int i, time, k, c, limit = 0;
-	int arg_done = 0;
 	struct scsi_cmnd *SCpnt;
 	struct Scsi_Host *shost = SCarg->device->host;
 	struct hostdata *ha = (struct hostdata *)shost->hostdata;
@@ -1967,9 +1966,6 @@ static int eata2x_eh_host_reset(struct scsi_cmnd *SCarg)
 		if (SCpnt->scsi_done == NULL)
 			panic("%s: reset, mbox %d, SCpnt->scsi_done == NULL.\n",
 			      ha->board_name, i);
-
-		if (SCpnt == SCarg)
-			arg_done = 1;
 	}
 
 	if (do_dma(shost->io_port, 0, RESET_PIO)) {
@@ -2037,10 +2033,7 @@ static int eata2x_eh_host_reset(struct scsi_cmnd *SCarg)
 	ha->in_reset = 0;
 	do_trace = 0;
 
-	if (arg_done)
-		printk("%s: reset, exit, done.\n", ha->board_name);
-	else
-		printk("%s: reset, exit.\n", ha->board_name);
+	printk("%s: reset, exit.\n", ha->board_name);
 
 	spin_unlock_irq(shost->host_lock);
 	return SUCCESS;
-- 
1.8.5.6

Re: [PATCH 27/28] eata: remove 'arg_done' from eata2x_eh_host_reset()

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

Re: [PATCH 27/28] eata: remove 'arg_done' from eata2x_eh_host_reset()

[Christoph Hellwig]

Looks fine,

Reviewed-by: Christoph Hellwig <hch@lst.de>

[PATCH 28/28] NCR5380: Clear all unissued commands on host reset

[Hannes Reinecke]

When doing a host reset we should be clearing all outstanding
commands, not just the command triggering the reset.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/NCR5380.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/NCR5380.c b/drivers/scsi/NCR5380.c
index 8a08122..429cb00 100644
--- a/drivers/scsi/NCR5380.c
+++ b/drivers/scsi/NCR5380.c
@@ -2313,7 +2313,7 @@ static int NCR5380_host_reset(struct scsi_cmnd *cmd)
 	spin_lock_irqsave(&hostdata->lock, flags);
 
 #if (NDEBUG & NDEBUG_ANY)
-	scmd_printk(KERN_INFO, cmd, __func__);
+	shost_printk(KERN_INFO, instance, __func__);
 #endif
 	NCR5380_dprint(NDEBUG_ANY, instance);
 	NCR5380_dprint_phase(NDEBUG_ANY, instance);
@@ -2331,10 +2331,13 @@ static int NCR5380_host_reset(struct scsi_cmnd *cmd)
 	 * commands!
 	 */
 
-	if (list_del_cmd(&hostdata->unissued, cmd)) {
+	list_for_each_entry(ncmd, &hostdata->unissued, list) {
+		struct scsi_cmnd *cmd = NCR5380_to_scmd(ncmd);
+
 		cmd->result = DID_RESET << 16;
 		cmd->scsi_done(cmd);
 	}
+	INIT_LIST_HEAD(&hostdata->unissued);
 
 	if (hostdata->selecting) {
 		hostdata->selecting->result = DID_RESET << 16;
-- 
1.8.5.6

Re: [PATCH 28/28] NCR5380: Clear all unissued commands on host reset

[Johannes Thumshirn]

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850