Increase the default ring buffer size as this can significantly
improve performance especially on high latency storage back-ends.
Signed-off-by: K. Y. Srinivasan k...@microsoft.com
Reviewed-by: Long Li lon...@microsoft.com
---
drivers/scsi/storvsc_drv.c |2 +-
1 files changed, 1
The current code assumes that the scatterlists presented are not chained.
Fix the code to not make this assumption.
Signed-off-by: K. Y. Srinivasan k...@microsoft.com
Reviewed-by: Long Li lon...@microsoft.com
---
V2: Got rid of double assignment. Olaf Hering o...@aepfle.de
The current code always sent packets without data on the primary channel.
Properly distribute sending of packets with no data amongst all available
channels. I would like to thank Long Li for noticing this problem.
Signed-off-by: K. Y. Srinivasan k...@microsoft.com
Reviewed-by: Long Li
We may exit this function without properly freeing up the maapings
we may have acquired. Fix the bug.
Signed-off-by: K. Y. Srinivasan k...@microsoft.com
Reviewed-by: Long Li lon...@microsoft.com
Cc: sta...@vger.kernel.org
---
drivers/scsi/storvsc_drv.c | 15 ---
1 files changed, 8
On systems with shared interrupts the interrupt routine might
be called as soon as the interrupt is enabled.
As this might happen before pci_set_drvdata() is called the
system would crash.
Reported-by: Andreas Brogle an...@ok.de
Tested-by: Andreas Brogle an...@ok.de
Signed-off-by: Hannes Reinecke
Set the tablesize based on the information given by the host.
Signed-off-by: K. Y. Srinivasan k...@microsoft.com
Reviewed-by: Long Li lon...@microsoft.com
Tested-by: Long Li lon...@microsoft.com
---
drivers/scsi/storvsc_drv.c | 89 +++
1 files changed,
The storage protocol informs the guest of the I/O capabilities of the storage
stack. Retrieve this information and use it in the guest.
Signed-off-by: K. Y. Srinivasan k...@microsoft.com
Reviewed-by: Long Li lon...@microsoft.com
---
drivers/scsi/storvsc_drv.c |6 ++
1 files changed, 6
Size the queue depth based on the ringbuffer size. Also accomodate for the
fact that we could have multiple channels (ringbuffers) per adaptor.
Signed-off-by: K. Y. Srinivasan k...@microsoft.com
Reviewed-by: Long Li lon...@microsoft.com
---
v2: Fixed a bug in computing queue depth:
This patch-set addresses perf issues discovered on the Azure storage stack.
These patches also fix a couple of bugs.
As in the first version of this patch-set, some of the patches are simply a
resend.
I have bumped up the version number of all patches though. In this version, I
have
addressed
On Fri, 2015-03-27 at 08:29 +0100, Hannes Reinecke wrote:
On systems with shared interrupts the interrupt routine might
be called as soon as the interrupt is enabled.
As this might happen before pci_set_drvdata() is called the
system would crash.
Reported-by: Andreas Brogle an...@ok.de
From: Nicholas Bellinger n...@linux-iscsi.org
This includes:
- target_add_lun
- target_remove_lun
- target_lookup_cmd_lun
- target_lookup_tmr_lun
- target_clear_lun_from_tpg
Also, update fabric drivers for loopback, iscsi/iser-target,
xen-scsiback.
Signed-off-by: Nicholas Bellinger
From: Nicholas Bellinger n...@linux-iscsi.org
This includes:
- target_add_node_to_devs
- target_get_deve_from_rtpi
- target_create_device_list_for_node
- target_free_device_list_for_node
- target_enable_device_list_for_node
- target_disable_device_list_for_node
Signed-off-by:
From: Nicholas Bellinger n...@linux-iscsi.org
Signed-off-by: Nicholas Bellinger n...@linux-iscsi.org
---
drivers/target/target_core_device.c | 13 -
drivers/target/target_core_tpg.c| 4
include/target/target_core_base.h | 3 ---
3 files changed, 20 deletions(-)
diff
From: Nicholas Bellinger n...@linux-iscsi.org
Signed-off-by: Nicholas Bellinger n...@linux-iscsi.org
---
drivers/target/iscsi/iscsi_target_tpg.c | 2 --
drivers/target/target_core_tpg.c| 21 -
include/target/target_core_fabric.h | 1 -
3 files changed, 24
From: Nicholas Bellinger n...@linux-iscsi.org
This includes + fabric driver changes:
- target_check_initiator_node_acl
- target_add_initiator_node_acl
- target_del_initiator_node_acl
- target_set_initiator_node_queue_depth
- target_del_initiator_node_acl
-
From: Nicholas Bellinger n...@linux-iscsi.org
This includes:
- target_alloc_port
- target_export_port
- target_release_port
- target_dev_export
- target_dev_unexport
Signed-off-by: Nicholas Bellinger n...@linux-iscsi.org
---
drivers/target/target_core_device.c | 16
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts core_tpg_deregister() to perform a list splice
for any remaining dynamically generated se_node_acls attached to
se_tpg, before calling tfo-tpg_release_fabric_acl() to free memory.
Cc: Hannes Reinecke h...@suse.de
Cc: Christoph
From: Nicholas Bellinger n...@linux-iscsi.org
This includes:
- target_register_tpg
- target_deregister_tpg
This includes updates for loopback, tcm_fc, iscsi/iser-target,
ib_srpt, sbp-target, usb-gadget, tcm_qla2xxx, vhost-scsi and
xen-scsiback.
Signed-off-by: Nicholas Bellinger
On 03/26/2015 03:41 PM, Mahesh Rajashekhara wrote:
Add new AIF support for SES device add/remove
Signed-off-by: Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com
Reviewed-by: Hannes Reinecke h...@suse.de
Cheers,
Hannes
--
Dr. Hannes ReineckezSeries Storage
On 03/26/2015 03:41 PM, Mahesh Rajashekhara wrote:
The Linux aacriad driver fails to detect the case of SG list count=0 on IOCTL
pass-through command and cause intermittent fault.
The result is the Linux aacriad driver send down IOCTL pass-through command
with one not initialized SG list to
On 03/26/2015 03:41 PM, Mahesh Rajashekhara wrote:
Add MSI-x interrupt mode support.
Signed-off-by: Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com
Reviewed-by: Hannes Reinecke h...@suse.de
Cheers,
Hannes
--
Dr. Hannes ReineckezSeries Storage
h...@suse.de
Thanks Hannes,
When we created new 4K based logical drive/container with the name length 16
but we were seeing only first 15 characters in Linux OS by sg_map -i, cat
/proc/scsi/scsi.
This change fixes logical drive name display issue.
-Original Message-
From: Hannes Reinecke
On 03/27/2015 11:59 AM, Mahesh Rajashekhara wrote:
Thanks Hannes,
When we created new 4K based logical drive/container with the name length
16 but we were seeing only first 15 characters in Linux OS by
sg_map -i,
cat /proc/scsi/scsi.
This change fixes logical drive name display issue.
On 03/26/2015 03:41 PM, Mahesh Rajashekhara wrote:
Add 4KB sector support
Signed-off-by: Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com
---
drivers/scsi/aacraid/aachba.c | 240 +++
drivers/scsi/aacraid/aacraid.h | 11 ++-
2 files changed, 200
On 03/26/2015 03:41 PM, Mahesh Rajashekhara wrote:
Add cfg IOCTL command fix
Signed-off-by: Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com
Same here; this patch could do with some more information.
Otherwise:
Reviewed-by: Hannes Reinecke h...@suse.de
Cheers,
Hannes
--
Dr. Hannes
On 03/26/2015 03:41 PM, Mahesh Rajashekhara wrote:
Add performance improvement changes.
Signed-off-by: Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com
Reviewed-by: Hannes Reinecke h...@suse.de
Cheers,
Hannes
--
Dr. Hannes ReineckezSeries Storage
h...@suse.de
On 03/26/2015 03:41 PM, Mahesh Rajashekhara wrote:
Add vpd page code 0x83 support
Signed-off-by: Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com
Reviewed-by: Hannes Reinecke h...@suse.de
Cheers,
Hannes
--
Dr. Hannes ReineckezSeries Storage
h...@suse.de
On 03/26/2015 03:41 PM, Mahesh Rajashekhara wrote:
Add 240 simple volume support
Signed-off-by: Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com
Reviewed-by: Hannes Reinecke h...@suse.de
Cheers,
Hannes
--
Dr. Hannes ReineckezSeries Storage
h...@suse.de
Thanks Hannes,
This patch fixes the IOP_RESET issue. Sending IOP_RESET command need to wait
for only 10 sec instead of 5 minutes in case of firmware does not response
IOP_RESET command.
Disable interrupt before setup interrupt routine to prevent spurious interrupts.
-Original Message-
On 03/26/2015 03:41 PM, Mahesh Rajashekhara wrote:
Add IOP RESET command handling changes
Signed-off-by: Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com
The message is very terse; some more information would've been good.
Otherwise:
Reviewed-by: Hannes Reinecke h...@suse.de
Cheers,
On 03/26/2015 03:41 PM, Mahesh Rajashekhara wrote:
Add driver version and build change
Signed-off-by: Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com
Reviewed-by: Hannes Reinecke h...@suse.de
Cheers,
Hannes
--
Dr. Hannes ReineckezSeries Storage
h...@suse.de
On 03/26/2015 03:41 PM, Mahesh Rajashekhara wrote:
Add AIF raw device remove support
Signed-off-by: Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com
Reviewed-by: Hannes Reinecke h...@suse.de
Cheers,
Hannes
--
Dr. Hannes ReineckezSeries Storage
h...@suse.de
Thanks Hannes,
After getting the platform shutdown command VM_CloseAll response from the
firmware, driver was getting configuration IOCTL request from the upper layers
and it sends down to firmware. This causes firmware assert issue.
This patch fixes the firmware assert issue. During the
From: Nicholas Bellinger n...@linux-iscsi.org
Now that se_node_acl-lun_entry_hlist[] fast path access uses RCU,
go ahead and convert se_node_acl-lun_entry_lock to struct mutex,
for non fast-path RCU updater code.
Cc: Hannes Reinecke h...@suse.de
Cc: Christoph Hellwig h...@lst.de
Cc: Sagi
From: Nicholas Bellinger n...@linux-iscsi.org
Now that se_node_acl-lun_entry_hlist[] is using RCU, convert existing
se_tpg-tpg_lun_lock to struct mutex so core_tpg_add_node_to_devs()
can perform RCU updater logic without releasing -tpg_lun_mutex.
Cc: Hannes Reinecke h...@suse.de
Cc: Christoph
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts se_dev_entry-pr_ref_count access to use modern
percpu reference counting.
It updates core_enable_device_list_for_node() to percpu_ref_init()
when se_dev_entry is being enabled.
It also updates core_disable_device_list_for_node()
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts SCSI MIB configfs show attribute code to use
RCU read path macros for se_node_acl-lun_entry_hlist[] access.
Cc: Hannes Reinecke h...@suse.de
Cc: Christoph Hellwig h...@lst.de
Cc: Sagi Grimberg sa...@mellanox.com
Signed-off-by:
From: Nicholas Bellinger n...@linux-iscsi.org
Hi Hannes, HCH, Sagi,
Here is an initial pass at the conversion of se_node_acl-device_list[]
to use RCU protected pointers for se_lun fast-path lookup code.
The big advantage with RCU is that transport_lookup_cmd_lun() can now
run completely
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts transport_lookup_*_lun() fast-path code to
use RCU read path primitives when looking up se_dev_entry. It
adds a new array of pointers in se_node_acl-lun_entry_hlist
for this purpose.
For transport_lookup_cmd_lun() code, it works
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts core_[enable,disable]_device_list_for_node() to RCU
updater path code when modifying se_dev_entry pointers.
It includes protected rcu_assign_pointer() and invokes synchronize_rcu()
to wait for RCU read paths to finish.
Required
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts various target_core_device.c to RCU reader path
to extract deve-mapped_lun before rcu_read_unlock() ahead of invoking
core_disable_device_list_for_node() to trigger subsequent RCU updater
code.
It includes
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts fabric independent configfs link/unlink to use
RCU read path macros for se_node_acl-lun_entry_hlist[] access.
Also, update show_write_protect attribute access as well.
Cc: Hannes Reinecke h...@suse.de
Cc: Christoph Hellwig
From: Nicholas Bellinger n...@linux-iscsi.org
This patch changes __core_scsi3_do_alloc_registration() code to
drop pr_reg-pr_reg_tg_pt_lun pointer usage in favor of a new
pr_reg RPTI + existing target_lun.
It also includes changes to REGISTER, REGISTER_AND_MOVE and APTPL
feature bit codepaths.
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts SPC emulation for REPORT_LUN + MODE_SENSE to use
RCU read path macros for se_node_acl-lun_entry_hlist[] access.
Cc: Hannes Reinecke h...@suse.de
Cc: Christoph Hellwig h...@lst.de
Cc: Sagi Grimberg sa...@mellanox.com
Signed-off-by:
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts target_core_ua.c code to use RCU read path macros
for se_node_acl-lun_entry_hlist[] access.
Cc: Hannes Reinecke h...@suse.de
Cc: Christoph Hellwig h...@lst.de
Cc: Sagi Grimberg sa...@mellanox.com
Signed-off-by: Nicholas Bellinger
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts se_tpg-acl_node_lock to struct mutex, so that
-acl_node_acl walkers in core_clear_lun_from_tpg() can block when
calling core_disable_device_list_for_node().
It also updates core_dev_add_lun() to hold -acl_node_mutex when
calling
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts the MODE_SENSE special case in pscsi_transport_complete()
to use RCU read path macros for se_node_acl-lun_entry_hlist[] access.
Cc: Hannes Reinecke h...@suse.de
Cc: Christoph Hellwig h...@lst.de
Cc: Sagi Grimberg
From: Nicholas Bellinger n...@linux-iscsi.org
This patch converts core_scsi3_pr_seq_non_holder() check for non
reservation holding registrations to use se_deve-pr_reg as an RCU
protected pointer.
It also includes associated rcu_assign_pointer() + synchronize_rcu()
in
This patch has code optimization and refactoring in below areas-
1) Created separate functions for sysPD IOs and non Read Write LDIOs.
2) Added function for refiring MFI commands in controller reset path.
3) Enabling interrupt immediately after adapter's initialization.
4) Added function for
megaraid_sas driver changes.
Signed-off-by: Sumit Saxena sumit.sax...@avagotech.com
---
[PATCH 0/5] megaraid_sas : Description Patch.
[PATCH 1/5] megaraid_sas : Code optimization and refactoring.
[PATCH 2/5] megaraid_sas : Use Block layer tag support for internal command
indexing.
[PATCH 3/5]
This patch will add support for Single Server High Availability(SSHA) cluster
support.
Here is the short decsription of changes done to add support for SSHA-
1) Host will send system's Unique ID based on DMI_PRODUCT_UUID to firmware.
2) Toggle the devhandle in LDIO path for Remote LDs.
This patch will update module parameters and driver name inside SCSI host
template to reflect Avago.
Signed-off-by: Sumit Saxena sumit.sax...@avagotech.com
---
drivers/scsi/megaraid/megaraid_sas_base.c |6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
diff --git
This patch will upgrade the driver version and add back the release date and
sysfs hook for the same.
Some Avago's internal applications uses sysfs parameter of release date, so
they were broken because of removal
of release date from sysfs.
Signed-off-by: Sumit Saxena
megaraid_sas driver will use block layer provided tag for indexing internal MPT
frames to get any unique MPT frame tied with tag.
Each IO request submitted from SCSI mid layer will get associated MPT frame
from MPT framepool (retrieved and return back using spinlock
inside megaraid_sas driver's
On Thu, 2015-03-12 at 17:54 +0200, Gilad Broner wrote:
From: Lee Susman lsus...@codeaurora.org
Adding debugfs capability for ufshcd.
This patch fails to build as a module because you aren't building
corectly:
ERROR: ufsdbg_add_debugfs [drivers/scsi/ufs/ufshcd.ko] undefined!
ERROR:
Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com writes:
Add MSI-x interrupt mode support.
Signed-off-by: Mahesh Rajashekhara mahesh.rajashekh...@pmcs.com
---
drivers/scsi/aacraid/aacraid.h | 80 -
drivers/scsi/aacraid/comminit.c | 95 +-
On systems with shared interrupts the interrupt routine might
be called as soon as the interrupt is enabled.
As this might happen before pci_set_drvdata() is called the
system would crash.
Reported-by: Andreas Brogle an...@ok.de
Tested-by: Andreas Brogle an...@ok.de
Signed-off-by: Hannes Reinecke
This patch series enables RSS for FCoE frames and adds
DDP support for FCoE target.
Varun Prakash (3):
cxgb4: add structure and macro definitions for FCoE DDP
cxgb4: add DDP support for FCoE target
cxgb4: enable RSS for FCoE frames
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h |1 +
This patch adds code for ndo_fcoe_ddp_target and
ndo_fcoe_ddp_done.
Signed-off-by: Varun Prakash va...@chelsio.com
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h |1 +
drivers/net/ethernet/chelsio/cxgb4/cxgb4_fcoe.c | 605 +++
This patch adds new header file t4_tcb.h and
structure, macro definitions for FCoE DDP support
in cxgb4 driver.
Signed-off-by: Varun Prakash va...@chelsio.com
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4_fcoe.h | 54 +-
drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h |7 ++
Signed-off-by: Varun Prakash va...@chelsio.com
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c |8
drivers/net/ethernet/chelsio/cxgb4/t4_regs.h|8
2 files changed, 16 insertions(+), 0 deletions(-)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
On 03/17/2015 09:02 PM, Don Brace wrote:
From: Webb Scales webb.sca...@hp.com
Allow driver initiated commands to have a timeout. It does not
yet try to do anything with timeouts on such commands.
We are sending a reset in order to get rid of a command we want to abort.
If we make it return
Reviewed-by: Wen Xiongwenxi...@linux.vnet.ibm.com
Thanks,
Wendy
Quoting Brian King brk...@linux.vnet.ibm.com:
The pci_set_pcie_reset_state has changed semantics to not be callable
from interrupt context, so change ipr's usage of the API to comply with
this change by ensuring this occurs from
Reviewed-by: Wen Xiongwenxi...@linux.vnet.ibm.com
Thanks,
Wendy
Quoting Brian King brk...@linux.vnet.ibm.com:
Re-enable write same support for ipr RAID adapters.
Signed-off-by: Brian King brk...@linux.vnet.ibm.com
---
drivers/scsi/ipr.c |1 -
1 file changed, 1 deletion(-)
diff -puN
Reviewed-by: Wen Xiongwenxi...@linux.vnet.ibm.com
Thanks,
Wendy
Quoting Brian King brk...@linux.vnet.ibm.com:
Bump driver version.
Signed-off-by: Brian King brk...@linux.vnet.ibm.com
---
drivers/scsi/ipr.h |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff -puN
msi_enabled field is used in per Adapter structure (struct aac_dev). If the
driver uses MSIx or MSI interrupt mode, this will be set.
Before posting the commands (for each and every command requests) to the
Adapter and vector number assignments, this will be checked.
Thanks,
Mahesh
On 03/26/2015 07:43 PM, Kashyap Desai wrote:
-Original Message-
From: Hannes Reinecke [mailto:h...@suse.de]
Sent: Thursday, March 26, 2015 9:28 PM
To: Kashyap Desai; linux-scsi@vger.kernel.org
Subject: Re: Scsi Error handling query
On 03/26/2015 02:38 PM, Kashyap Desai wrote:
Hi
Reviewed-by: Wen Xiongwenxi...@linux.vnet.ibm.com
Thanks,
Wendy
Quoting Brian King brk...@linux.vnet.ibm.com:
Fixes a possible oops during adapter initialization in some
memory allocation failure error paths scenarios.
Reported-by: Dan Carpenter dan.carpen...@oracle.com
Signed-off-by: Brian
Reviewed-by: Wen Xiongwenxi...@linux.vnet.ibm.com
Thanks,
Wendy
Quoting Brian King brk...@linux.vnet.ibm.com:
Currently when performing a reboot with an ipr adapter,
the adapter gets shutdown completely, flushing all write
cache, as well as performing a full hardware reset of the card
during
Noted.
Thanks for your review.
Don
-Original Message-
From: Tomas Henzl [mailto:the...@redhat.com]
Sent: Friday, March 27, 2015 10:11 AM
To: Don Brace; Scott Teel; Kevin Barnett; james.bottom...@parallels.com;
h...@infradead.org; Justin Lindley; brace
Cc:
I'll send up another patch to fix this issue.
-Original Message-
From: Tomas Henzl [mailto:the...@redhat.com]
Sent: Monday, March 23, 2015 11:58 AM
To: Don Brace; Scott Teel; Kevin Barnett; james.bottom...@parallels.com;
h...@infradead.org; Justin Lindley; brace
Cc:
This patch set addresses several issues caused by driver reloading in
ufs driver, although the first patch also fixes error path in driver
probe.
Akinobu Mita (3):
scsi: ufs: avoid using hostdata after scsi_host_put()
scsi: ufs: fix unbalanced power.usage_count after reloading driver
scsi:
The hostdata array, which is denoted by 'hba' in ufs driver, should
not be accessed after calling scsi_host_put().
Signed-off-by: Akinobu Mita akinobu.m...@gmail.com
Cc: Vinayak Holikatti vinholika...@gmail.com
Cc: James E.J. Bottomley jbottom...@parallels.com
Cc: Christoph Hellwig h...@lst.de
Every time the driver is reloaded, the warning message
Unbalanced pm_runtime_enable! is triggered due to unbalanced
power.disable_depth. This is because pm_runtime_enable() is called
during driver probe but pm_runtime_disable() is missed on driver remove.
This also restores the device's runtime
On driver removal, pm_runtime_get_sync() is called, but
pm_runtime_put_sync() is missed. So once the driver is reloaded, the
device's power.usage_count is unbalanced and the idle callback for the
device will never be called.
Signed-off-by: Akinobu Mita akinobu.m...@gmail.com
Cc: Vinayak
75 matches
Mail list logo