with 1 expander, connect 8 HDD, the write performance will be
improved by 80%.
Signed-off-by: Xiangliang Yu <yuxiangl@marvell.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
A recent conversion has introduced references to &pdev->dev, which does
not actually exist in all the contexts it's used in.
Replace this with card->dev where necessary, in order to let
the driver build again.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@kernel.org
-- change connection behavior
-- set bit8 to 1 for performance tuning
-- set bit0 to 0 to enable retry for no_dest reject case.
Signed-off-by: Xiangliang Yu <yuxiangl@marvell.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
spin up issue: some direct attached SAS device can't spin up
Signed-off-by: Xiangliang Yu <yuxiangl@marvell.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Add support for Marvell 88SE9480 SAS/SATA HBA
Signed-off-by: Xiangliang Yu <yuxiangl@marvell.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Expanders fail to link when the phy rates are mismatched.
Signed-off-by: Xiangliang Yu <yuxiangl@marvell.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Needed to jump to scic_lock unlock.
Also spotted by coccicheck.
Signed-off-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Cc: <stable@kernel.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Kill the local smp response buffer.
Besides being unnecessary, it is too small (currently truncates
responses to 60 bytes). The mid-layer will have already allocated a
sufficiently sized buffer, just kmap and copy into it directly.
Cc: <stable@kernel.org>
Reported-by: Derick Marks <derick.w.marks@intel.com>
Tested-by: Derick Marks <derick.w.marks@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Code Inspection: found two missing break directives. First one will
result in not retrying an a task that report
IO_OPEN_CNX_ERROR_HW_RESOURCE_BUSY, the second will result in cosmetic
debug printk conflicting statement stutter. Because checkpatch.pl came
up with a warning regarding unnecessary space before a newline on one of
the fragments associated with the diff context, I took the liberty of
fixing all the cases of this issue in the pair of files touched by this
defect. These cosmetic changes hide the break changes :-(
To help focus, break changes are in pm8001_hwi.c fragment line 1649 for
the IO_OPEN_CNX_ERROR_HW_RESOURCE_BUSY case statement and pm8001_sas.c
line 1000 deals with the conflicting debug print stutter.
Signed-off-by: Mark Salyzyn <mark_salyzyn@us.xyratex.com>
Acked-by: Jack Wang <jack_wang@usish.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
On the pm8001, when a device is in the process of going away (device
power off or hot plug), depending on the timing, the driver would return
SAS_PHY_DOWN as the return value to the queuecommand DEV_IS_GONE logic.
The net result is an near infinite retry (especially if SAS debugging is
enabled), the logs will fill with:
kernel: mpi_ssp_completion 2119:e21:SSP IO status 0x13 tag 0xcc1c0000
dlen=90 param=0xe
kernel: wwn=5000c50034069e86 cdb=12 00 00 00 5a 00 00 00 00 00 00 00 00
00 00 00
kernel: sas: lldd_execute_task returned: 138
kernel: sas: lldd_execute_task returned: 138
kernel: sas: lldd_execute_task returned: 138
kernel: sas: lldd_execute_task returned: 138
kernel: sas: lldd_execute_task returned: 138
kernel: sas: lldd_execute_task returned: 138
kernel: sas: lldd_execute_task returned: 138
. . .
This patch changes to leverage the port_attached logic to complete the
command with a status of PHY_DOWN so that the disposition can be handled
immediately and correctly.
Signed-off-by: Mark Salyzyn <mark_salyzyn@us.xyratex.com>
Acked-by: Jack Wang <jack_wang@usish.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Currently fcoe_ddp_min doesn't have default value
so by default not used, so setting up default value
as 4k as this works better by avoiding overhead
of programing DDP for small IOs.
Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Use real dev in case it has HW vlan acceleration
support since in this case the real dev would
do needed vlan processing, this way unnecessary
vlan layer processing avoided and it gives
slightly better IOPS with 512B size IOs.
Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
fix holes and better cache aligned fields.
Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
cache aligned xid and ex_lock beside
removing holes.
Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Re-arrange its fields to avoid padding and have better
cacheline alignments.
Removed not used start_time, end_time and last_pkt_time
fields.
This all reduced this struct size to 448 from 480 and
that also reduced one cacheline on x86_64 beside
eliminating 8 pads. However kept logical fields together.
Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Since fcoe_percpu_thread_create() creates percpu kthread, it makes sense
to use kthread_create_on_node() to get proper NUMA affinity for kthread
stack.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Several sas drivers legitimately check the protocol against the union of
SAS_PROTOCOL_SATA and SAS_PROTOCOL_STP. Provide a SAS_PROTOCOL_STP_ALL
to silence warnings like:
drivers/scsi/pm8001/pm8001_sas.c:438:3: warning: case value ‘5’ not in enumerated type ‘enum sas_protocol’ [-Wswitch]
drivers/scsi/mvsas/mv_sas.c:798:2: warning: case value ‘5’ not in enumerated type ‘enum sas_protocol’ [-Wswitch]
drivers/scsi/mvsas/mv_sas.c:1783:2: warning: case value ‘5’ not in enumerated type ‘enum sas_protocol’ [-Wswitch]
drivers/scsi/mvsas/mv_sas.c:1886:2: warning: case value ‘5’ not in enumerated type ‘enum sas_protocol’ [-Wswitch]
drivers/scsi/isci/request.c:3565:2: warning: case value ‘5’ not in enumerated type ‘enum sas_protocol’ [-Wswitch]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Libsas forget to set the sas_address and device type of rphy lead to file
under /sys/class/sas_x show wrong value, fix that.
Signed-off-by: Jack Wang <jack_wang@usish.com>
Tested-by: Crystal Yu <crystal_yu@usish.com>
Cc: stable@kernel.org
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
If the user has disabled CONFIG_SCSI_SAS_HOST_SMP then libsas drivers
will not be receiving smp-gpio frames and do not need this lookup code.
Reported-by: Randy Dunlap <rdunlap@xenotime.net>
Tested-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
libsas handles:
1/ limiting ata scanning to lun0
2/ changes to /sys/block/<sdX>/device/queue_depth for ata devices
libata handles turning off ncq globally via kernel command line
(libata.force=noncq) or sysfs (echo 1 >
/sys/block/<sdX>/device/queue_depth). A lldd specific compile option is
not necessary.
Cc: Jack Wang <jack_wang@usish.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
libsas now handles:
1/ limiting ata scanning to lun0
2/ maximizing the queue_depth of sas devices (up to 256, mvsas only
supports 64)
3/ changes to /sys/block/<sdX>/device/queue_depth for ata devices
Acked-by: Xiangliang Yu <yuxiangl@marvell.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
The queue-depth for libsas-attached devices initializes to 32 and can
only be increased manually via sysfs to a max of 64, while mpt2sas
attached devices initialize to 254 and dynamically float via the
midlayer ->change_queue_depth interface.
No performance regression was observed with this change on the isci
driver.
Tested-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Pass queue_depth change requests to libata, and prevent queue_type
changes for ATA devices.
Otherwise:
1/ we do not honor the libata specific restrictions on the queue depth
2/ libsas drivers that do not set sdev->tagged_supported are unable to
change the queue_depth of ata devices via sysfs
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Acked-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Currently mvsas and pm8001 have custom ->slave_alloc implementations to
achieve this. Uplevel it for all libsas drivers as isci encounters problems
with atapi devices when scanning past lun0.
Just do what Darrick suggested [1], and limit the scan for ata devices.
[1] http://marc.info/?l=linux-scsi&m=116604101119861&w=2
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Allow expander table-to-table attachments for
expanders that support it.
Signed-off-by: Luben Tuikov <ltuikov@yahoo.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
I hit a crash in qla2x00_abort_all_cmds() if the qla2xxx module is
unloaded right after it is loaded. I debugged this down to the abort
handling improperly treating a command of type SRB_ADISC_CMD as if it
had a bsg_job to complete when that command actually uses the iocb_cmd
part of the union. (I guess to hit this one has to unload the module
while the async FC initialization is still in progress)
It seems we should only look for a bsg_job if type is SRB_ELS_CMD_RPT,
SRB_ELS_CMD_HST or SRB_CT_CMD, so switch the test to make that explicit.
Signed-off-by: Roland Dreier <roland@purestorage.com>
Acked-by: Chad Dupuis <chad.dupuis@qlogic.com>
Cc: stable@kernel.org
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Scrub uses a simple tree-enumeration to bring the relevant portions
of the extent- and csum-tree into the page cache before starting the
scrub-I/O. This is now replaced by using the new readahead-API.
During readahead the scrub is being accounted as paused, so it won't
hold off transaction commits.
This change raises the average disk bandwith utilisation on my test
volume from 70% to 90%. On another volume, the time for a test run
went down from 89s to 43s.
Changes v5:
- reada1/2 are now of type struct reada_control *
Signed-off-by: Arne Jansen <sensille@gmx.net>
This adds the hooks needed for readahead. In the readpage_end_io_hook,
the extent state is checked for the EXTENT_READAHEAD flag. Only in this
case the readahead hook is called, to keep the impact on non-ra as low
as possible.
Additionally, a hook for a failed IO is added, otherwise readahead would
wait indefinitely for the extent to finish.
Changes for v2:
- eliminate race condition
Signed-off-by: Arne Jansen <sensille@gmx.net>
This is the implementation for the generic read ahead framework.
To trigger a readahead, btrfs_reada_add must be called. It will start
a read ahead for the given range [start, end) on tree root. The returned
handle can either be used to wait on the readahead to finish
(btrfs_reada_wait), or to send it to the background (btrfs_reada_detach).
The read ahead works as follows:
On btrfs_reada_add, the root of the tree is inserted into a radix_tree.
reada_start_machine will then search for extents to prefetch and trigger
some reads. When a read finishes for a node, all contained node/leaf
pointers that lie in the given range will also be enqueued. The reads will
be triggered in sequential order, thus giving a big win over a naive
enumeration. It will also make use of multi-device layouts. Each disk
will have its on read pointer and all disks will by utilized in parallel.
Also will no two disks read both sides of a mirror simultaneously, as this
would waste seeking capacity. Instead both disks will read different parts
of the filesystem.
Any number of readaheads can be started in parallel. The read order will be
determined globally, i.e. 2 parallel readaheads will normally finish faster
than the 2 started one after another.
Changes v2:
- protect root->node by transaction instead of node_lock
- fix missed branches:
The readahead had a too simple check to determine if a branch from
a node should be checked or not. It now also records the upper bound
of each node to see if the requested RA range lies within.
- use KERN_CONT to debug output, to avoid line breaks
- defer reada_start_machine to worker to avoid deadlock
Changes v3:
- protect root->node by rcu
Changes v5:
- changed EIO-semantics of reada_tree_block_flagged
- remove spin_lock from reada_control and make elems an atomic_t
- remove unused read_total from reada_control
- kill reada_key_cmp, use btrfs_comp_cpu_keys instead
- use kref-style release functions where possible
- return struct reada_control * instead of void * from btrfs_reada_add
Signed-off-by: Arne Jansen <sensille@gmx.net>
Add state information for readahead to btrfs_fs_info and btrfs_device
Changes v2:
- don't wait in radix_trees
- add own set of workers for readahead
Reviewed-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Arne Jansen <sensille@gmx.net>
Add a READAHEAD extent buffer flag.
Add a function to trigger a read with this flag set.
Changes v2:
- use extent buffer flags instead of extent state flags
Changes v5:
- adapt to changed read_extent_buffer_pages interface
- don't return eb from reada_tree_block_flagged if it has CORRUPT flag set
Signed-off-by: Arne Jansen <sensille@gmx.net>
read_extent_buffer_pages currently has two modes, either trigger a read
without waiting for anything, or wait for the I/O to finish. The former
also bails when it's unable to lock the page. This patch now adds an
additional parameter to allow it to block on page lock, but don't wait
for completion.
Changes v5:
- merge the 2 wait parameters into one and define WAIT_NONE, WAIT_COMPLETE and
WAIT_PAGE_LOCK
Change v6:
- fix bug introduced in v5
Signed-off-by: Arne Jansen <sensille@gmx.net>
The <mach/gpio.h> file is included from upper directories
and deal with generic GPIO and gpiolib stuff. Break out the
platform and driver specific defines and functions into its own
header file.
Cc: Eric Miao <eric.y.miao@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
As per example from the other ARM boards, push the PXA
GPIO driver down to the GPIO subsystem so it can be consolidated.
Acked-by: Eric Miao <eric.y.miao@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The <mach/gpio.h> file is included from upper directories
and deal with generic GPIO and gpiolib stuff. Break out the
platform and driver specific defines and functions into its own
header file.
Cc: Ryan Mallon <rmallon@gmail.com>
Acked-by: Hartley Sweeten <hsweeten@visionengravers.com>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Four cpufreq-like governors are provided as examples.
powersave: use the lowest frequency possible. The user (device) should
set the polling_ms as 0 because polling is useless for this governor.
performance: use the highest freqeuncy possible. The user (device)
should set the polling_ms as 0 because polling is useless for this
governor.
userspace: use the user specified frequency stored at
devfreq.user_set_freq. With sysfs support in the following patch, a user
may set the value with the sysfs interface.
simple_ondemand: simplified version of cpufreq's ondemand governor.
When a user updates OPP entries (enable/disable/add), OPP framework
automatically notifies devfreq to update operating frequency
accordingly. Thus, devfreq users (device drivers) do not need to update
devfreq manually with OPP entry updates or set polling_ms for powersave
, performance, userspace, or any other "static" governors.
Note that these are given only as basic examples for governors and any
devices with devfreq may implement their own governors with the drivers
and use them.
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Reviewed-by: Mike Turquette <mturquette@ti.com>
Acked-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Device specific sysfs interface /sys/devices/.../power/devfreq_*
- governor R: name of governor
- cur_freq R: current frequency
- polling_interval R: polling interval in ms given with devfreq profile
W: update polling interval.
- central_polling R: 1 if polling is managed by devfreq framework
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Reviewed-by: Mike Turquette <mturquette@ti.com>
Acked-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
--
Documentation/ABI/testing/sysfs-class-devfreq | 44 ++++++++++++++++
drivers/devfreq/devfreq.c | 69 ++++++++++++++++++++++++++
2 files changed, 113 insertions(+)
create mode 100644 Documentation/ABI/testing/sysfs-class-devfreq
With OPPs, a device may have multiple operable frequency and voltage
sets. However, there can be multiple possible operable sets and a system
will need to choose one from them. In order to reduce the power
consumption (by reducing frequency and voltage) without affecting the
performance too much, a Dynamic Voltage and Frequency Scaling (DVFS)
scheme may be used.
This patch introduces the DVFS capability to non-CPU devices with OPPs.
DVFS is a techique whereby the frequency and supplied voltage of a
device is adjusted on-the-fly. DVFS usually sets the frequency as low
as possible with given conditions (such as QoS assurance) and adjusts
voltage according to the chosen frequency in order to reduce power
consumption and heat dissipation.
The generic DVFS for devices, devfreq, may appear quite similar with
/drivers/cpufreq. However, cpufreq does not allow to have multiple
devices registered and is not suitable to have multiple heterogenous
devices with different (but simple) governors.
Normally, DVFS mechanism controls frequency based on the demand for
the device, and then, chooses voltage based on the chosen frequency.
devfreq also controls the frequency based on the governor's frequency
recommendation and let OPP pick up the pair of frequency and voltage
based on the recommended frequency. Then, the chosen OPP is passed to
device driver's "target" callback.
When PM QoS is going to be used with the devfreq device, the device
driver should enable OPPs that are appropriate with the current PM QoS
requests. In order to do so, the device driver may call opp_enable and
opp_disable at the notifier callback of PM QoS so that PM QoS's
update_target() call enables the appropriate OPPs. Note that at least
one of OPPs should be enabled at any time; be careful when there is a
transition.
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Reviewed-by: Mike Turquette <mturquette@ti.com>
Acked-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
The CONFIG_ARM_AMBA symbol is selected by the platforms that support
AMBA, and the other platforms (e.g. orion) might not be able to
build that code at all when they do not support the clk API.
Instead, make sure that OC_ETM is only built on platforms that
do support the amba subsystem already.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Some PCI drivers call pcibios_bus_to_resource directly,
but it is only exported when CONFIG_HOTPLUG is set,
because it was initially mean for pccard support.
Moving the export out of the #ifdef lets us avoid these
build errors:
ERROR: "pcibios_bus_to_resource" [drivers/video/vt8623fb.ko] undefined!
ERROR: "pcibios_bus_to_resource" [drivers/video/arkfb.ko] undefined!
ERROR: "pcibios_bus_to_resource" [drivers/video/s3fb.ko] undefined!
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
These functions are used in some PCI drivers with big-endian
MMIO space, and they are trivial to add here.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Support for IDE drivers should not be automatic, since most platforms
cannot actually support any IDE low-level drivers. This partly
reverts 2064c946e "ARM: always select HAVE_IDE" to set this symbol
only when either a PC-style bus (PCI, ISA, PCMCIA) is enabled or
a platform is used that is known to have an existing driver in
drivers/ide.
New platforms should not need this option and just use CONFIG_ATA
with drivers/ata/.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Support for the cpu_suspend functions is only built-in
when CONFIG_PM_SLEEP is enabled, but omap3/4, exynos4
and pxa always call cpu_suspend when CONFIG_PM is enabled.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Distros are starting to ship with toolchains defaulting to
hardfloat. Using such a compiler to build the kernel fails
in the VFP directory with
arch/arm/vfp/entry.S:1:0: sorry, unimplemented: -mfloat-abi=hard and VFP
Adding -mfloat-abi=soft to the gcc command line fixes this.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
The two functions cpu_is_v6_unaligned and safe_usermode
are only defined when CONFIG_PROC_FS is enabled, but
are used outside of the #ifdef.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Dave Martin <dave.martin@linaro.org>
