Commit graph

534,192 commits

Author SHA1 Message Date
Antti Palosaari
a6f62431a0 [media] rtl28xxu: simplify FE callback handling
Logic is so simple that there is no idea to separate tuner selection to
own function, instead do it in a callback and get rid of one function.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:21:19 -02:00
Antti Palosaari
1dbbdceeb6 [media] rtl28xxu: fix DVB FE callback
DVB FE callback functionality went broken after I moved tuners to
demod muxed I2C adapter. That happens because driver state was
carried by I2C adapter and when mux is used there is one adapter
more in a chain.
USB adapter <-> I2C adapter <-> muxed I2C adapter

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:20:46 -02:00
Antti Palosaari
a08c378119 [media] rtl2832: implement sleep
Put demod to soft reset in order to save power when sleep. That drops
power usage ~30mA @5V on USB dongle I tested. In real life it does
not matter much as USB IF powers off demod too, but now it is done
twice - demod and USB IF.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:20:02 -02:00
Antti Palosaari
75c24005ee [media] rtl2832: claim copyright and module author
I have implemented tons of things for that driver, more than anyone
else, so lets claim copyright and module authorship.

Cc: Thomas Mair <thomas.mair86@gmail.com>
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:19:24 -02:00
Antti Palosaari
cd559e0b13 [media] rtl2832: cleanups and minor changes
Remove all the stuff that is not needed anymore. Rename variable.
Remove extra new lines.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:18:51 -02:00
Antti Palosaari
3bda2bf1f8 [media] rtl2832_sdr: cleanups
Small cleanups. Remove unneeded variables. Some checkpatch issues.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:18:03 -02:00
Antti Palosaari
e4e78693a1 [media] rtl2832_sdr: fix logging
Pass correct device, platform device, for logging system in order
to format printouts correctly.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:17:24 -02:00
Antti Palosaari
488be13a5e [media] rtl28xxu: use master I2C adapter for slave demods
Both mn88472 and mn88473 slave demods are connected to master I2C
bus, not the bus behind master demod I2C gate like tuners. Use
correct bus.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:16:47 -02:00
Antti Palosaari
a2f7f220df [media] rtl28xxu: switch SDR module to platform driver
RTL2832 SDR module implements kernel platform driver. Change old
binding to that one.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:16:17 -02:00
Antti Palosaari
63bdab5d31 [media] rtl2832_sdr: convert to platform driver
That SDR driver module was abusing DVB frontend SEC (satellite
equipment controller) device and due to that it was also using
legacy DVB binding. Platform bus is pseudo-bus provided by kernel
driver model and it fits cases like that, where any other busses
are not suitable.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:15:43 -02:00
Antti Palosaari
725b7f3046 [media] rtl2832_sdr: rename state variable from 's' to 'dev'
'dev' sounds better than 's' for such variable.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:13:17 -02:00
Antti Palosaari
0aa32ef9b3 [media] rtl2832: provide register IO callbacks
Provide register read and write callbacks for SDR module.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:12:34 -02:00
Antti Palosaari
eec21beb35 [media] rtl2832: merge reg page as a part of reg address
Chips uses 8-bit register addresses with 5 pages. Extend register
address by using register page as a first byte of address, defining
virtual register addresses. That is common method of handling
register pages and regmap also uses it. Remove page + address
conversion glue which was there for regmap.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:12:07 -02:00
Antti Palosaari
5e5d52a00f [media] rtl2832: remove unneeded software reset from init()
There is no need to do software reset on init() as it is done a bit
later on end of set_frontend(). Software reset usually means
restarting (resetting to starting point) chip internal state machine
(FSM). Naturally it is done after all parameters are programmed.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:11:17 -02:00
Antti Palosaari
dcadb82a7a [media] rtl2832: use regmap reg cache
Enable regmap register cache in order to reduce IO.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:10:54 -02:00
Antti Palosaari
e20b0cf2f3 [media] rtl28xxu: add support for RTL2832U/RTL2832 PID filter
RTL2832 demod integrated into RTL2832U has PID filter. PID filtering
is provided by rtl2832 demod driver. Add support for it.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:10:19 -02:00
Antti Palosaari
4b01e01a81 [media] rtl2832: implement PID filter
Implement PID filter. This demod has PID filter size of 32 PIDs.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:09:56 -02:00
Antti Palosaari
1c7da405c6 [media] rtl2832: define more demod lock statuses
Demod lock flags are derived from demod state machine states. States
are running from 1 to 11, where highest state 11 means demod is
fully locked and streaming. Naturally smaller state numbers means
there is some partial locks.

Define now state 10 as missing synch and lock.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:09:19 -02:00
Antti Palosaari
c65dbf652d [media] rtl2832: drop FE i2c gate control support
We don't need it anymore as all users are using muxed I2C adapter.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:08:37 -02:00
Antti Palosaari
164f3d2a83 [media] rtl28xxu: use demod mux I2C adapter for every tuner
Tuners are connected to demod I2C adapter. Use that muxed adapter
for each tuner. That allows us to get rid of hackish FE gate control
solution.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:08:16 -02:00
Antti Palosaari
25ef9f5547 [media] rtl2832: implement DVBv5 signal strength statistics
Estimate signal strength from IF digital AGC.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:07:43 -02:00
Antti Palosaari
084330b746 [media] rtl2832: wrap DVBv5 BER to DVBv3
Change legacy DVBv3 read BER to return values calculated by DVBv5
statistics.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:07:22 -02:00
Antti Palosaari
f7caf93fb8 [media] rtl2832: wrap DVBv5 CNR to DVBv3 SNR
Change legacy DVBv3 read SNR to return values calculated by DVBv5
statistics.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:07:01 -02:00
Antti Palosaari
6b4fd01804 [media] rtl2832: implement DVBv5 BER statistic
DVBv5 BER.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:06:30 -02:00
Antti Palosaari
19d273d635 [media] rtl2832: implement DVBv5 CNR statistic
DVBv5 CNR.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:06:07 -02:00
Antti Palosaari
d10165815b [media] rtl2832: convert to regmap API
Use regmap to cover register access routines.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:05:33 -02:00
Antti Palosaari
de0a5f1132 [media] rtl28xxu: use platform data config for rtl2832 demod
Use platform data configuration for rtl2832 demod driver. Old
configuration are still left as it is used for rtl2832_sdr driver.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:04:13 -02:00
Antti Palosaari
e1174d788a [media] rtl2832: move all configuration to platform data struct
Move all needed configuration values to platform data structure
and remove old configuration code where possible.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:03:41 -02:00
Antti Palosaari
6e6aac6525 [media] rtl2832: enhance / fix logging
Pass correct device pointer to dev_* logging functions in order
print logs correctly.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:03:04 -02:00
Antti Palosaari
038c6f240e [media] rtl2832: rename driver state variable from 'priv' to 'dev'
Rename it device state variable to dev. Both priv and dev are very
common terms for such variable in kernel, but I like use dev in
order to keep drivers consistent.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:02:19 -02:00
Antti Palosaari
ac32ee42cc [media] rtl2832: remove exported resources
Exported resources are not needed anymore as all users are using
callbacks carried via platform data. Due to that we will remove
those.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:01:25 -02:00
Antti Palosaari
77a2e76b85 [media] rtl28xxu: use rtl2832 demod callbacks accessing its resources
Switch demod resource use from exported symbols to callbacks its
provides.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:01:04 -02:00
Antti Palosaari
6f5f6ee24e [media] rtl2832: add platform data callbacks for exported resources
Add callback for all of those functions which are currently
exported using EXPORT_SYMBOL. That allows us convert every user to
callbacks and eventually all exported symbols could be removed.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 16:00:05 -02:00
Antti Palosaari
15d37f3824 [media] rtl2830: convert to regmap API
Use regmap to cover register access routines.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 15:59:19 -02:00
Antti Palosaari
fd4cfa8bb1 [media] rtl2830: implement own I2C locking
Own I2C locking is needed due to two special reasons:
1) Chips uses multiple register pages/banks on single I2C slave.
Page is changed via I2C register access.
2) Chip offers muxed/gated I2C adapter for tuner. Gate/mux is
controlled by I2C register access.

Due to these reasons, I2C locking did not fit very well.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-02-03 15:54:59 -02:00
Arnd Bergmann
a4188beee5 iommu/arm-smmu: Avoid build warning
ARM allmodconfig gained a new warning when dma_addr_t is 32-bit wide:

drivers/iommu/arm-smmu.c: In function 'arm_smmu_iova_to_phys_hard':
drivers/iommu/arm-smmu.c:1255:3: warning: right shift count >= width of type

This changes the calculation so that the effective type is always
64-bit.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 859a732e4f ("iommu/arm-smmu: add support for iova_to_phys through ATS1PR")
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-02-03 18:53:27 +01:00
Steven Rostedt (Red Hat)
eae473581c tracing: Have mkdir and rmdir be part of tracefs
The tracing "instances" directory can create sub tracing buffers
with mkdir, and remove them with rmdir. As a mkdir will also create
all the files and directories that control the sub buffer the inode
mutexes need to be released before this is done, to avoid deadlocks.
It is better to let the tracing system unlock the inode mutexes before
calling the functions that create the files within the new directory
(or deletes the files from the one being destroyed).

Now that tracing has been converted over to tracefs, the tracefs file
system can be modified to accommodate this feature. It still releases
the locks, but the filesystem itself can take care of the ugly
business and let the user just do what it needs.

The tracing system now attaches a descriptor to the directory dentry
that can have userspace create or remove sub directories. If this
descriptor does not exist for a dentry, then that dentry can not be
used to create other directories. This descriptor holds a mkdir and
rmdir method that only takes a character string as an argument.

The tracefs file system will first make a copy of the dentry name
before releasing the locks. Then it will pass the copied name to the
methods. It is up to the tracing system that supplied the methods to
handle races with duplicate names and such as all the inode mutexes
would be released when the functions are called.

Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2015-02-03 12:48:43 -05:00
Steven Rostedt (Red Hat)
cc31004a4a tracefs: Add directory /sys/kernel/tracing
When tracefs is configured, have the directory /sys/kernel/tracing appear
just like /sys/kernel/debug appears when debugfs is configured.

This will give a consistent place for system admins to mount tracefs.

Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2015-02-03 12:48:42 -05:00
Steven Rostedt (Red Hat)
f76180bc07 tracing: Automatically mount tracefs on debugfs/tracing
As tools currently rely on the tracing directory in debugfs, we can not
just created a tracefs infrastructure and expect sysadmins to mount
the new tracefs to have their old tools work.

Instead, the debugfs tracing directory is still created and the tracefs
file system is mounted there when the debugfs filesystem is mounted.

No longer does the tracing infrastructure update the debugfs file system,
but instead interacts with the tracefs file system. But now, it still
appears to the user like nothing changed, except you also have the feature
of mounting just the tracing system without needing all of debugfs!

Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2015-02-03 12:48:42 -05:00
Steven Rostedt (Red Hat)
8434dc9340 tracing: Convert the tracing facility over to use tracefs
debugfs was fine for the tracing facility as a quick way to get
an interface. Now that tracing has matured, it should separate itself
from debugfs such that it can be mounted separately without needing
to mount all of debugfs with it. That is, users resist using tracing
because it requires mounting debugfs. Having tracing have its own file
system lets users get the features of tracing without needing to bring
in the rest of the kernel's debug infrastructure.

Another reason for tracefs is that debubfs does not support mkdir.
Currently, to create instances, one does a mkdir in the tracing/instance
directory. This is implemented via a hack that forces debugfs to do
something it is not intended on doing. By converting over to tracefs, this
hack can be removed and mkdir can be properly implemented. This patch does
not address this yet, but it lays the ground work for that to be done.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2015-02-03 12:48:41 -05:00
Steven Rostedt (Red Hat)
4282d60689 tracefs: Add new tracefs file system
Add a separate file system to handle the tracing directory. Currently it
is part of debugfs, but that is starting to show its limits.

One thing is that in order to access the tracing infrastructure, you need
to mount debugfs. As that includes debugging from all sorts of sub systems
in the kernel, it is not considered advisable to mount such an all
encompassing debugging system.

Having the tracing system in its own file systems gives access to the
tracing sub system without needing to include all other systems.

Another problem with tracing using the debugfs system is that the
instances use mkdir to create sub buffers. debugfs does not support mkdir
from userspace so to implement it, special hacks were used. By controlling
the file system that the tracing infrastructure uses, this can be properly
done without hacks.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2015-02-03 12:48:40 -05:00
Steven Rostedt (Red Hat)
09d23a1d8a tracing: Create cmdline tracer options on tracing fs init
The options for cmdline tracers are not created if the debugfs system
is not ready yet. If tracing has started before debugfs is up, then the
option files for the tracer are not created. Create them when creating
the tracing directory if the current tracer requires option files.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2015-02-03 12:48:39 -05:00
Steven Rostedt (Red Hat)
0f67f04ffc tracing: Only create tracer options files if directory exists
Do not bother creating tracer options if no tracing directory
exists. If a tracer is enabled via the command line, and is
started before the tracing directory is created, then it wont have
its tracer specific options created.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2015-02-03 12:48:38 -05:00
Emil Medve
cd70d4659f iommu/fsl: Various cleanups
Currently a PAMU driver patch is very likely to receive some
checkpatch complaints about the code in the context of the
patch. This patch is an attempt to fix most of that and make
the driver more readable

Also fixed a subset of the sparse and coccinelle reported
issues.

Signed-off-by: Emil Medve <Emilian.Medve@Freescale.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-02-03 18:47:18 +01:00
Roland Dreier
ecb7b1233c Merge branches 'ipoib' and 'odp' into for-next 2015-02-03 09:29:25 -08:00
Haggai Eran
7e36ef8205 IB/core: Temporarily disable ex_query_device uverb
Commit 5a77abf9a9 ("IB/core: Add support for extended query device caps")
added a new extended verb to query the capabilities of RDMA devices, but the
semantics of this verb are still under debate [1].

Don't expose this verb to userspace until the ABI is nailed down.

[1] [PATCH v1 0/5] IB/core: extended query device caps cleanup for v3.19
    http://www.spinics.net/lists/linux-rdma/msg22904.html

Signed-off-by: Haggai Eran <haggaie@mellanox.com>
Reviewed-by: Yann Droneaud <ydroneaud@opteya.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2015-02-03 09:29:11 -08:00
Takashi Iwai
3fe9cf390f Merge branch 'topic/snd-device' into for-next 2015-02-03 17:57:16 +01:00
Trond Myklebust
cc3ea893cb NFS: Client side changes for RDMA
These patches improve the scalability of the NFSoRDMA client and take large
 variables off of the stack.  Additionally, the GFP_* flags are updated to
 match what TCP uses.
 
 Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2
 
 iQIcBAABCAAGBQJUy/21AAoJENfLVL+wpUDrDR4QAJ+aZ1LqO5muJG5LDIReh0p+
 IyvAO3ekWKaaXZd5e/wOviHRRSBCs3oRBm1sNyjOOZYf7nl67RrndjGq2Ccp6tGw
 6V6ligIlhu3QAApFiQmLfpDXQeDBysMfhEQMXDj8kVhAvn3SWwbT6q1Bb8tMZvnu
 lwSPYCuLAdtPzDs3j3Pt3jSrOD7gS06Vtumco1yS82YiLK86E7JiOLb9w5Ltk0Wo
 DT9hZ7JXtflBy3trBsNYQ7GTjoYEp792Ca9DT/nie8/Tuuy3DhooLJKdwVErq/0/
 ycI033mw//RneY+/eF2wLeS4PlzaiSq7DmXKpE5MnkpVuNAdPbHbCuVzuXa4/LLD
 NhKyw/NTAE9ucFJ8e3apj3m42O00bMV0rx0rOUvMMdbNPBKzXk3rrb4uRYfT4k+D
 yss7aEZg0EcDbe9KqfuDdwXVJIZnvAvXD+V1iL4X3QmVM0JyJZouNZPoG7fCwA1g
 uoqGJK9zu7osr59YMomf2okWQFUQYCea5ombXYOo3ZtANw+OYaHDcMKTXaiCtMdA
 CZUEOND8mB/EFtgTMA2TZmDzch8iOOuD7wmtifb8h1DqN/3GxgecBcT5XMCx5zDR
 bMpt9vbCfe4YP6idpl5XEnVbcPG9CZF2W2Hg2gI7axX8R+ZMvwQQlWJjOqxoR6+F
 bpD/jjWFYaS6U7NGyylM
 =YCZe
 -----END PGP SIGNATURE-----

Merge tag 'nfs-rdma-for-3.20' of git://git.linux-nfs.org/projects/anna/nfs-rdma

NFS: Client side changes for RDMA

These patches improve the scalability of the NFSoRDMA client and take large
variables off of the stack.  Additionally, the GFP_* flags are updated to
match what TCP uses.

Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>

* tag 'nfs-rdma-for-3.20' of git://git.linux-nfs.org/projects/anna/nfs-rdma: (21 commits)
  xprtrdma: Update the GFP flags used in xprt_rdma_allocate()
  xprtrdma: Clean up after adding regbuf management
  xprtrdma: Allocate zero pad separately from rpcrdma_buffer
  xprtrdma: Allocate RPC/RDMA receive buffer separately from struct rpcrdma_rep
  xprtrdma: Allocate RPC/RDMA send buffer separately from struct rpcrdma_req
  xprtrdma: Allocate RPC send buffer separately from struct rpcrdma_req
  xprtrdma: Add struct rpcrdma_regbuf and helpers
  xprtrdma: Refactor rpcrdma_buffer_create() and rpcrdma_buffer_destroy()
  xprtrdma: Simplify synopsis of rpcrdma_buffer_create()
  xprtrdma: Take struct ib_qp_attr and ib_qp_init_attr off the stack
  xprtrdma: Take struct ib_device_attr off the stack
  xprtrdma: Free the pd if ib_query_qp() fails
  xprtrdma: Remove rpcrdma_ep::rep_func and ::rep_xprt
  xprtrdma: Move credit update to RPC reply handler
  xprtrdma: Remove rl_mr field, and the mr_chunk union
  xprtrdma: Remove rpcrdma_ep::rep_ia
  xprtrdma: Rename "xprt" and "rdma_connect" fields in struct rpcrdma_xprt
  xprtrdma: Clean up hdrlen
  xprtrdma: Display XIDs in host byte order
  xprtrdma: Modernize htonl and ntohl
  ...
2015-02-03 11:54:58 -05:00
Suman Tripathi
09c32aaa36 ahci_xgene: Fix the dma state machine lockup for the ATA_CMD_SMART PIO mode command.
This patch addresses the issue with ATA_CMD_SMART pio mode command for
enumeration and device detection with ATA devices.  The X-Gene AHCI
controller has an errata in which it cannot clear the BSY bit after
the PIO setup FIS. The dma state machine enters CMFatalErrorUpdate
state and locks up. It is the same issue as in the commit 2a0bdff6b9
("ahci-xgene: fix the dma state machine lockup for the IDENTIFY DEVICE
PIO mode command").

For example :  without this patch it results in READ DMA command failure
as shown below :

 [  126.700072] ata2.00: exception Emask 0x0 SAct 0x0
		SErr 0x0 action 0x6 frozen
 [  126.707089] ata2.00: failed command: READ DMA
 [  126.711426] ata2.00: cmd c8/00:08:00:55:57/00:00:00:00:00/e1 tag 1
                dma 4096 in
 [  126.711426]  res 40/00:ff:00:00:00/00:00:00:00:00/40 Emask
		0x4 (timeout)
 [  126.725956] ata2.00: status: { DRDY }

Signed-off-by: Suman Tripathi <stripathi@apm.com>
Reported-by:   Mark Langsdorf <mlangsdo@redhat.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2015-02-03 11:16:47 -05:00
Wincy Van
705699a139 KVM: nVMX: Enable nested posted interrupt processing
If vcpu has a interrupt in vmx non-root mode, injecting that interrupt
requires a vmexit.  With posted interrupt processing, the vmexit
is not needed, and interrupts are fully taken care of by hardware.
In nested vmx, this feature avoids much more vmexits than non-nested vmx.

When L1 asks L0 to deliver L1's posted interrupt vector, and the target
VCPU is in non-root mode, we use a physical ipi to deliver POSTED_INTR_NV
to the target vCPU.  Using POSTED_INTR_NV avoids unexpected interrupts
if a concurrent vmexit happens and L1's vector is different with L0's.
The IPI triggers posted interrupt processing in the target physical CPU.

In case the target vCPU was not in guest mode, complete the posted
interrupt delivery on the next entry to L2.

Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2015-02-03 17:15:08 +01:00