The legacy API wrappers in include/linux/pci-dma-compat.h should go away as
they create unnecessary midlayering for include/linux/dma-mapping.h API.
Instead use dma-mapping.h API directly.

The patch has been generated with the coccinelle script below.
Compile-tested.

@@@@
- PCI_DMA_BIDIRECTIONAL
+ DMA_BIDIRECTIONAL

@@@@
- PCI_DMA_TODEVICE
+ DMA_TO_DEVICE

@@@@
- PCI_DMA_FROMDEVICE
+ DMA_FROM_DEVICE

@@@@
- PCI_DMA_NONE
+ DMA_NONE

@@ expression E1, E2, E3; @@
- pci_alloc_consistent(E1, E2, E3)
+ dma_alloc_coherent(&E1->dev, E2, E3, GFP_)

@@ expression E1, E2, E3; @@
- pci_zalloc_consistent(E1, E2, E3)
+ dma_alloc_coherent(&E1->dev, E2, E3, GFP_)

@@ expression E1, E2, E3, E4; @@
- pci_free_consistent(E1, E2, E3, E4)
+ dma_free_coherent(&E1->dev, E2, E3, E4)

@@ expression E1, E2, E3, E4; @@
- pci_map_single(E1, E2, E3, E4)
+ dma_map_single(&E1->dev, E2, E3, E4)

@@ expression E1, E2, E3, E4; @@
- pci_unmap_single(E1, E2, E3, E4)
+ dma_unmap_single(&E1->dev, E2, E3, E4)

@@ expression E1, E2, E3, E4, E5; @@
- pci_map_page(E1, E2, E3, E4, E5)
+ dma_map_page(&E1->dev, E2, E3, E4, E5)

@@ expression E1, E2, E3, E4; @@
- pci_unmap_page(E1, E2, E3, E4)
+ dma_unmap_page(&E1->dev, E2, E3, E4)

@@ expression E1, E2, E3, E4; @@
- pci_map_sg(E1, E2, E3, E4)
+ dma_map_sg(&E1->dev, E2, E3, E4)

@@ expression E1, E2, E3, E4; @@
- pci_unmap_sg(E1, E2, E3, E4)
+ dma_unmap_sg(&E1->dev, E2, E3, E4)

@@ expression E1, E2, E3, E4; @@
- pci_dma_sync_single_for_cpu(E1, E2, E3, E4)
+ dma_sync_single_for_cpu(&E1->dev, E2, E3, E4)

@@ expression E1, E2, E3, E4; @@
- pci_dma_sync_single_for_device(E1, E2, E3, E4)
+ dma_sync_single_for_device(&E1->dev, E2, E3, E4)

@@ expression E1, E2, E3, E4; @@
- pci_dma_sync_sg_for_cpu(E1, E2, E3, E4)
+ dma_sync_sg_for_cpu(&E1->dev, E2, E3, E4)

@@ expression E1, E2, E3, E4; @@
- pci_dma_sync_sg_for_device(E1, E2, E3, E4)
+ dma_sync_sg_for_device(&E1->dev, E2, E3, E4)

@@ expression E1, E2; @@
- pci_dma_mapping_error(E1, E2)
+ dma_mapping_error(&E1->dev, E2)

@@ expression E1, E2; @@
- pci_set_consistent_dma_mask(E1, E2)
+ dma_set_coherent_mask(&E1->dev, E2)

@@ expression E1, E2; @@
- pci_set_dma_mask(E1, E2)
+ dma_set_mask(&E1->dev, E2)

Link: https://lore.kernel.org/r/f8d4778440d55ba26c04eef0f7d63fb211a39443.1596045683.git.usuraj35@gmail.comSigned-off-by: NSuraj Upadhyay <usuraj35@gmail.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Remove cxgbi_alloc_big_mem(), cxgbi_free_big_mem() functions and use
kvzalloc/kvfree instead. __GFP_NOWARN added to kvzalloc() call because we
already print a warning in case of allocation fail.

Link: https://lore.kernel.org/r/20200801133123.61834-1-efremov@linux.comSigned-off-by: NDenis Efremov <efremov@linux.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Use %*ph format to print small buffer as hex string.

Link: https://lore.kernel.org/r/20200730153535.39691-1-andriy.shevchenko@linux.intel.comSigned-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Fixes coccicheck warning:

./drivers/scsi/mpt3sas/mpt3sas_base.c:5247:16-34: WARNING: dma_alloc_coherent use in ioc -> request already zeroes out memory,  so memset is not needed

dma_alloc_coherent() already zeroes out memory so memset() is not needed.

Link: https://lore.kernel.org/r/1596079918-41115-4-git-send-email-liheng40@huawei.comSigned-off-by: NLi Heng <liheng40@huawei.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Fixes coccicheck warning:

./drivers/scsi/qla2xxx/qla_mbx.c:4928:15-33: WARNING: dma_alloc_coherent use in els_cmd_map already zeroes out memory,  so memset is not needed

dma_alloc_coherent() already zeroes out memory so memset() is not needed.

Link: https://lore.kernel.org/r/1596079918-41115-3-git-send-email-liheng40@huawei.comSigned-off-by: NLi Heng <liheng40@huawei.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Fixes coccicheck warning:

./drivers/scsi/pmcraid.c:4709:3-21: WARNING: dma_alloc_coherent use in pinstance -> hrrq_start [ i ] already zeroes out memory,  so memset is not needed

dma_alloc_coherent() already zeroes out memory so memset() is not needed.

Link: https://lore.kernel.org/r/1596079918-41115-2-git-send-email-liheng40@huawei.comSigned-off-by: NLi Heng <liheng40@huawei.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Fixes coccicheck warning:

./drivers/scsi/mvsas/mv_init.c:244:11-29: WARNING: dma_alloc_coherent use in mvi -> tx already zeroes out memory,  so memset is not needed
./drivers/scsi/mvsas/mv_init.c:250:15-33: WARNING: dma_alloc_coherent use in mvi -> rx_fis already zeroes out memory,  so memset is not needed
./drivers/scsi/mvsas/mv_init.c:256:11-29: WARNING: dma_alloc_coherent use in mvi -> rx already zeroes out memory,  so memset is not needed
./drivers/scsi/mvsas/mv_init.c:265:13-31: WARNING: dma_alloc_coherent use in mvi -> slot already zeroes out memory,  so memset is not needed

dma_alloc_coherent() already zeroes out memory so memset() is not needed.

Link: https://lore.kernel.org/r/1596078235-54002-1-git-send-email-liheng40@huawei.comSigned-off-by: NLi Heng <liheng40@huawei.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Remove casting the values returned by memory allocation function.

Coccinelle emits WARNING:

./drivers/message/fusion/mptctl.c:2596:14-31: WARNING: casting value returned by memory allocation function to (SCSIDevicePage0_t *) is useless.
./drivers/message/fusion/mptctl.c:2660:15-32: WARNING: casting value returned by memory allocation function to (SCSIDevicePage3_t *) is useless.

Link: https://lore.kernel.org/r/1596014390-18605-1-git-send-email-liheng40@huawei.comSigned-off-by: NLi Heng <liheng40@huawei.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Remove casting the values returned by memory allocation function.

Coccinelle emits WARNING:

./drivers/message/fusion/mptfc.c:766:17-30: WARNING: casting value returned by memory allocation function to (FCPortPage0_t *) is useless.
./drivers/message/fusion/mptfc.c:907:17-30: WARNING: casting value returned by memory allocation function to (FCPortPage1_t *) is useless.

[mkp: memset()]

Link: https://lore.kernel.org/r/1596014354-59935-1-git-send-email-liheng40@huawei.comSigned-off-by: NLi Heng <liheng40@huawei.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

MediaTek UFS host now supports 2 lanes. Modify the lane count to 2.

This modification shall not impact old 1-lane host because
PA_CONNECTEDRXDATALANES and PA_CONNECTEDTXDATALANES will limit the target
lanes properly during power mode change. So we could relax the limitation
in ufs_dev_params.

Link: https://lore.kernel.org/r/20200819084340.7021-1-stanley.chu@mediatek.comReviewed-by: NStanley Chu <stanley.chu@mediatek.com>
Reviewed-by: NAvri Altman <avri.altman@wdc.com>
Signed-off-by: NAndy Teng <andy.teng@mediatek.com>
Signed-off-by: NStanley Chu <stanley.chu@mediatek.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Commit 5586dd8e ("scsi: ufs: Fix a race condition between error handler
and runtime PM ops") moves the ufshcd_scsi_block_requests() inside
err_handler() but forgets to remove the ufshcd_scsi_unblock_requests() in
the early return path. Correct the mistake.

Link: https://lore.kernel.org/r/1597798958-24322-1-git-send-email-cang@codeaurora.org
Fixes: 5586dd8e ("scsi: ufs: Fix a race condition between error handler and runtime PM ops")
Reviewed-by: NAsutosh Das <asutoshd@codeaurora.org>
Reviewed-by: Hongwu Su<hongwus@codeaurora.org>
Signed-off-by: NCan Guo <cang@codeaurora.org>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Currently, the UFS driver busy waits for fDeviceInit to be cleared. Provide
an upper bound and sleep between attempts instead of busy waiting.

Link: https://lore.kernel.org/r/1597053747-75171-1-git-send-email-kwmad.kim@samsung.comTested-by: NKiwoong Kim <kwmad.kim@samsung.com>
Reviewed-by: NAvri Altman <avri.altman@wdc.com>
Signed-off-by: NKiwoong Kim <kwmad.kim@samsung.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Link: https://lore.kernel.org/r/20200814095034.20709-3-huobean@gmail.comReviewed-by: NStanley Chu <stanley.chu@mediatek.com>
Reviewed-by: NAsutosh Das <asutoshd@codeaurora.org>
Signed-off-by: NBean Huo <beanhuo@micron.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

ufshcd_comp_devman_upiu() was poorly named leading people to think it was a
completion function. Rename it to ufshcd_compose_devman_upiu().

Link: https://lore.kernel.org/r/20200814095034.20709-2-huobean@gmail.comReviewed-by: NStanley Chu <stanley.chu@mediatek.com>
Reviewed-by: NAsutosh Das <asutoshd@codeaurora.org>
Acked-by: NAvri Altman <avri.altman@wdc.com>
Signed-off-by: NBean Huo <beanhuo@micron.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Fix race between ELS completion and flushing ELS request.

Link: https://lore.kernel.org/r/20200807110656.19965-8-jhasan@marvell.comSigned-off-by: NSaurav Kashyap <skashyap@marvell.com>
Signed-off-by: NJaved Hasan <jhasan@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Don't process ELS completion if event is flushed or cleaned up.

Link: https://lore.kernel.org/r/20200807110656.19965-7-jhasan@marvell.comSigned-off-by: NSaurav Kashyap <skashyap@marvell.com>
Signed-off-by: NJaved Hasan <jhasan@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Initiate cleanup for ELS commands as well.

Link: https://lore.kernel.org/r/20200807110656.19965-6-jhasan@marvell.comSigned-off-by: NSaurav Kashyap <skashyap@marvell.com>
Signed-off-by: NJaved Hasan <jhasan@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Send cleanup even for RRQ on timeout.

Link: https://lore.kernel.org/r/20200807110656.19965-5-jhasan@marvell.comSigned-off-by: NSaurav Kashyap <skashyap@marvell.com>
Signed-off-by: NJaved Hasan <jhasan@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The timer is already cancelled when abort is completed, hence no need to
cancel it again.

Link: https://lore.kernel.org/r/20200807110656.19965-4-jhasan@marvell.comSigned-off-by: NSaurav Kashyap <skashyap@marvell.com>
Signed-off-by: NJaved Hasan <jhasan@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

This is reported by Klockwork.

Link: https://lore.kernel.org/r/20200807110656.19965-3-jhasan@marvell.comSigned-off-by: NSaurav Kashyap <skashyap@marvell.com>
Signed-off-by: NJaved Hasan <jhasan@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The rport lock gets initialized during offload. If a non-FCP or non-target
rport got logout then this rport will be uninitialized. KASAN was
complaining because of it.

=========
[   14.384434] the code is fine but needs lockdep annotation.
[   14.384482] turning off the locking correctness validator.
========

Link: https://lore.kernel.org/r/20200807110656.19965-2-jhasan@marvell.comSigned-off-by: NSaurav Kashyap <skashyap@marvell.com>
Signed-off-by: NJaved Hasan <jhasan@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

MSM bus scaling has moved on to use interconnect framework and downstream
bus scaling APIs like msm_bus_scale*() do not exist anymore in the
kernel. Currently they are guarded by a config which also does not exist
and hence there are no build failures reported. Remove these unused
interfaces as they are currently no-ops and the scaling support that may be
added in future will use interconnect API.

Link: https://lore.kernel.org/r/20200804161033.15586-1-saiprakash.ranjan@codeaurora.orgSigned-off-by: NSai Prakash Ranjan <saiprakash.ranjan@codeaurora.org>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Link: https://lore.kernel.org/r/159622931040.30579.9167901134341507088.stgit@brunhildaReviewed-by: NScott Teel <scott.teel@microsemi.com>
Reviewed-by: NScott Benesh <scott.benesh@microsemi.com>
Reviewed-by: NGerry Morong <gerry.morong@microsemi.com>
Reviewed-by: NKevin Barnett <kevin.barnett@microsemi.com>
Reviewed-by: NMartin Wilck <mwilck@suse.com>
Signed-off-by: NDon Brace <don.brace@microsemi.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Add a counter to assist in verifying when RAID bypass is being used.

Link: https://lore.kernel.org/r/159622930468.30579.13153724465552773544.stgit@brunhildaReviewed-by: NScott Teel <scott.teel@microsemi.com>
Reviewed-by: NScott Benesh <scott.benesh@microsemi.com>
Reviewed-by: NMartin Wilck <mwilck@suse.com>
Signed-off-by: NKevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: NDon Brace <don.brace@microsemi.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Support device deletion via sysfs.

I.e: echo 1 > /sys/block/sd<X>/device/delete

Link: https://lore.kernel.org/r/159622929885.30579.2727491506675011534.stgit@brunhildaReviewed-by: NScott Teel <scott.teel@microsemi.com>
Reviewed-by: NScott Benesh <scott.benesh@microsemi.com>
Reviewed-by: NMartin Wilck <mwilck@suse.com>
Signed-off-by: NKevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: NDon Brace <don.brace@microsemi.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Eliminate kernel panics when getting invalid responses from controller.
Take controller offline instead of causing kernel panics.

Link: https://lore.kernel.org/r/159622929306.30579.16523318707596752828.stgit@brunhildaReviewed-by: NScott Teel <scott.teel@microsemi.com>
Reviewed-by: NScott Benesh <scott.benesh@microsemi.com>
Reviewed-by: NPrasad Munirathnam <Prasad.Munirathnam@microsemi.com>
Reviewed-by: NMartin Wilck <mwilck@suse.com>
Signed-off-by: NKevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: NDon Brace <don.brace@microsemi.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Have OS rescan after logical volume expansion to reflect new size.

Link: https://lore.kernel.org/r/159622928727.30579.298277463169866711.stgit@brunhildaReviewed-by: NScott Teel <scott.teel@microsemi.com>
Reviewed-by: NScott Benesh <scott.benesh@microsemi.com>
Reviewed-by: NMartin Wilck <mwilck@suse.com>
Signed-off-by: NMahesh Rajashekhara <mahesh.rajashekhara@microsemi.com>
Signed-off-by: NDon Brace <don.brace@microsemi.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

VID_9005, DID_028F, SVID_9005 and SDID_080A.

Link: https://lore.kernel.org/r/159622928143.30579.14769183842894725454.stgit@brunhildaReviewed-by: NScott Teel <scott.teel@microsemi.com>
Reviewed-by: NScott Benesh <scott.benesh@microsemi.com>
Reviewed-by: NKevin Barnett <kevin.barnett@microsemi.com>
Reviewed-by: NMartin Wilck <mwilck@suse.com>
Signed-off-by: NMahesh Rajashekhara <mahesh.rajashekhara@microsemi.com>
Signed-off-by: NDon Brace <don.brace@microsemi.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Eliminate issuing INQUIRYs to problematic devices by using information
provided by controller.

Link: https://lore.kernel.org/r/159622927172.30579.3960527536810532094.stgit@brunhildaReviewed-by: NScott Teel <scott.teel@microsemi.com>
Reviewed-by: NScott Benesh <scott.benesh@microsemi.com>
Reviewed-by: NMartin Wilck <mwilck@suse.com>
Signed-off-by: NKevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: NDon Brace <don.brace@microsemi.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Updated driver version to 35.100.00.00

Link: https://lore.kernel.org/r/1596096229-3341-8-git-send-email-suganath-prabu.subramani@broadcom.comSigned-off-by: NSuganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

If driver has not received the interrupt for the aborted SCSI command
before processing the TM reply, driver polls all the reply descriptor pools
looking for the reply for the aborted SCSI command before marking TM as
FAILED. If it finds the reply, then it marks the TM as SUCCESS otherwise it
marks it FAILED.

scsih_tm_cmd_map_status() checks whether TM has aborted the timed out SCSI
command or not. If TM has aborted the IO, then it returns SUCCESS else it
returns FAILED.

Link: https://lore.kernel.org/r/1596096229-3341-7-git-send-email-suganath-prabu.subramani@broadcom.comSigned-off-by: NSuganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Add helper functions to check whether any SCSI command is outstanding on
particular Target, LUN device.

Also add function parameters 'channel', 'id' to function
mpt3sas_scsih_issue_tm().

Link: https://lore.kernel.org/r/1596096229-3341-6-git-send-email-suganath-prabu.subramani@broadcom.comSigned-off-by: NSuganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Rename Function _base_unmask_interrupts() to
mpt3sas_base_unmask_interrupts() and _base_mask_interrupts() to
mpt3sas_base_mask_interrupts(). Also add function declarion to
mpt3sas_base.h

Link: https://lore.kernel.org/r/1596096229-3341-5-git-send-email-suganath-prabu.subramani@broadcom.comSigned-off-by: NSuganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

It is not recommended to issue back-to-back host reset without any delay.
However, if someone issues back-to-back host reset then we observe that
target devices get unregistered and re-register with SML.  And if OS drive
is behind the HBA when it gets unregistered, then file-system goes into
read-only mode.

Normally during host reset, driver marks accessible target devices as
responding and triggers the event MPT3SAS_REMOVE_UNRESPONDING_DEVICES to
remove any non-responding devices through FW worker thread. While
processing this event, driver unregisters the non-responding devices and
clears the responding flag for all the devices.

Currently, during host reset, driver is cancelling only those Firmware
event works which are pending in Firmware event workqueue. It is not
cancelling work which is currently running. Change the driver to cancel all
events.

Link: https://lore.kernel.org/r/1596096229-3341-4-git-send-email-suganath-prabu.subramani@broadcom.comSigned-off-by: NSuganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

When controller fails to transition to READY state during driver probe,
dump the system interface register set. This will give snapshot of the
firmware status for debugging driver load issues.

Link: https://lore.kernel.org/r/1596096229-3341-3-git-send-email-suganath-prabu.subramani@broadcom.comSigned-off-by: NSuganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Currently config_cmds.reply buffer is not memset to zero before posting
config page request message.  In some cases, for the current config
request, the previous config reply is getting processed and we will observe
PageType mismatch between request to reply buffer. It will be difficult to
debug this type of issue and it confuses by thinking that HBA Firmware
itself posted the wrong config reply.  So it is better to memset the
config_cmds.reply buffer with zeros before issuing the config request.

Link: https://lore.kernel.org/r/1596096229-3341-2-git-send-email-suganath-prabu.subramani@broadcom.comSigned-off-by: NSuganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

In current UFS task abort hook, namely ufshcd_abort(), if one task is
aborted successfully, clk_gating.active_reqs held by this task is not
decreased, which makes clk_gating.active_reqs stay above zero forever, thus
clock gating would never happen. Instead of releasing resources of one task
"manually", use the existing func __ufshcd_transfer_req_compl().  This
change also eliminates a possible race of scsi_dma_unmap() from the real
completion in IRQ handler path.

Link: https://lore.kernel.org/r/1596975355-39813-10-git-send-email-cang@codeaurora.org
Fixes: 1ab27c9c ("ufs: Add support for clock gating")
CC: Stanley Chu <stanley.chu@mediatek.com>
Reviewed-by: NStanley Chu <stanley.chu@mediatek.com>
Reviewed-by: NAsutosh Das <asutoshd@codeaurora.org>
Signed-off-by: NCan Guo <cang@codeaurora.org>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The current IRQ handler blocks SCSI requests before scheduling eh_work,
when error handler calls pm_runtime_get_sync, if ufshcd_suspend/resume
sends a SCSI cmd, most likely the SSU cmd, since SCSI requests are blocked,
pm_runtime_get_sync() will never return because ufshcd_suspend/resume is
blocked by the SCSI cmd.

 - In queuecommand path, hba->ufshcd_state check and ufshcd_send_command
   should stay under the same spin lock. This is to make sure that no more
   commands leak into doorbell after hba->ufshcd_state is changed.

 - Don't block SCSI requests before error handler starts to run, let error
   handler block SCSI requests when it is ready to start error recovery.

 - Don't let SCSI layer keep requeuing the SCSI cmds sent from HBA runtime
   PM ops, let them pass or fail them. Let them pass if eh_work is
   scheduled due to non-fatal errors. Fail them if eh_work is scheduled due
   to fatal errors, otherwise the cmds may eventually time out since UFS is
   in bad state, which gets error handler blocked for too long. If we fail
   the SCSI cmds sent from HBA runtime PM ops, HBA runtime PM ops fails
   too, but it does not hurt since error handler can recover HBA runtime PM
   error.

Link: https://lore.kernel.org/r/1596975355-39813-9-git-send-email-cang@codeaurora.orgReviewed-by: NBean Huo <beanhuo@micron.com>
Reviewed-by: NAsutosh Das <asutoshd@codeaurora.org>
Signed-off-by: NCan Guo <cang@codeaurora.org>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Performing dumps in the IRQ handler causes system stability issues. Move
dumps to the error handler and only print basic host registers here.

Link: https://lore.kernel.org/r/1596975355-39813-8-git-send-email-cang@codeaurora.orgReviewed-by: NBean Huo <beanhuo@micron.com>
Reviewed-by: NAsutosh Das <asutoshd@codeaurora.org>
Signed-off-by: NCan Guo <cang@codeaurora.org>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The current error handler can not recover HBA runtime PM error if
ufshcd_suspend/resume has failed due to UFS errors, e.g. hibern8 enter/exit
error or SSU cmd error. When this happens, error handler may fail
performing a full reset and restore because error handler always assumes
that power, IRQs and clocks are ready after pm_runtime_get_sync returns,
but actually they are not if ufshcd_resume fails[1].

If ufschd_suspend/resume fails due to UFS errors, runtime PM framework
saves the error value to dev.power.runtime_error. After that, HBA dev
runtime suspend/resume would not be invoked anymore unless runtime_error is
cleared[2].

In case of ufshcd_suspend/resume fails due to UFS errors, for scenario [1],
error handler cannot assume anything of pm_runtime_get_sync, meaning error
handler should explicitly turn ON powers, IRQs and clocks again. To get the
HBA runtime PM work as regard for scenario [2], error handler can clear the
runtime_error by calling pm_runtime_set_active() if full reset and restore
succeeds. And, more important, if pm_runtime_set_active() returns no error,
which means runtime_error has been cleared, we also need to resume those
scsi devices under HBA in case any of them has failed to be resumed due to
HBA runtime resume failure. This is to unblock blk_queue_enter in case
there are bios waiting inside it.

Link: https://lore.kernel.org/r/1596975355-39813-7-git-send-email-cang@codeaurora.orgReviewed-by: NBean Huo <beanhuo@micron.com>
Reviewed-by: NAsutosh Das <asutoshd@codeaurora.org>
Signed-off-by: NCan Guo <cang@codeaurora.org>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>