Dexuan Cui [Thu, 23 Apr 2015 04:31:30 +0000 (21:31 -0700)]
hv: vmbus_free_channels(): remove the redundant free_channel()
free_channel() has been invoked in
vmbus_remove() -> hv_process_channel_removal(), or vmbus_remove() ->
... -> vmbus_close_internal() -> hv_process_channel_removal().
We also change to use list_for_each_entry_safe(), because the entry
is removed in hv_process_channel_removal().
This patch fixes a bug in the vmbus unload path.
Thank Dan Carpenter for finding the issue!
Signed-off-by: Dexuan Cui <decui@microsoft.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Cc: K. Y. Srinivasan <kys@microsoft.com> Cc: Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Drivers: hv: vmbus: unregister panic notifier on module unload
Commit 96c1d0581d00f7abe033350edb021a9d947d8d81 ("Drivers: hv: vmbus: Add
support for VMBus panic notifier handler") introduced
atomic_notifier_chain_register() call on module load. We also need to call
atomic_notifier_chain_unregister() on module unload as otherwise the following
crash is observed when we bring hv_vmbus back:
In case we do request_resource() in vmbus_acpi_add() we need to tear it down
to be able to load the driver again. Otherwise the following crash in observed
when hv_vmbus unload/load sequence is performed on a Generation2 instance:
Unify driver registration reporting and move it to debug level as normally daemons write to syslog themselves
and these kernel messages are useless.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Introduce FCOPY_VERSION_1 to support kernel replying to the negotiation
message with its own version.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Introduce VSS_OP_REGISTER1 to support kernel replying to the negotiation
message with its own version.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Tools: hv: vss: use misc char device to communicate with kernel
Use /dev/vmbus/hv_vss instead of netlink.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Tools: hv: kvp: use misc char device to communicate with kernel
Use /dev/vmbus/hv_kvp instead of netlink.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Convert to hv_utils_transport to support both netlink and /dev/vmbus/hv_kvp communication methods.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unify the code with the recently introduced hv_utils_transport. Netlink
communication is disabled for fcopy.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Convert to hv_utils_transport to support both netlink and /dev/vmbus/hv_vss communication methods.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The intention is to make KVP/VSS drivers work through misc char devices.
Introduce an abstraction for kernel/userspace communication to make the
migration smoother. Transport operational mode (netlink or char device)
is determined by the first received message. To support driver upgrades
the switch from netlink to chardev operational mode is supported.
Every hv_util daemon is supposed to register 2 callbacks:
1) on_msg() to get notified when the userspace daemon sent a message;
2) on_reset() to get notified when the userspace daemon drops the connection.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Drivers: hv: fcopy: set .owner reference for file operations
Get an additional reference otherwise a crash is observed when hv_utils module is being unloaded while
fcopy daemon is still running. .owner gives us an additional reference when
someone holds a descriptor for the device.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Drivers: hv: fcopy: switch to using the hvutil_device_state state machine
Switch to using the hvutil_device_state state machine from using 3 different state variables:
fcopy_transaction.active, opened, and in_hand_shake.
State transitions are:
-> HVUTIL_DEVICE_INIT when driver loads or on device release
-> HVUTIL_READY if the handshake was successful
-> HVUTIL_HOSTMSG_RECEIVED when there is a non-negotiation message from the host
-> HVUTIL_USERSPACE_REQ after userspace daemon read the message
-> HVUTIL_USERSPACE_RECV after/if userspace has replied
-> HVUTIL_READY after we respond to the host
-> HVUTIL_DEVICE_DYING on driver unload
In hv_fcopy_onchannelcallback() process ICMSGTYPE_NEGOTIATE messages even when
the userspace daemon is disconnected, otherwise we can make the host think
we don't support FCOPY and disable the service completely.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Drivers: hv: vss: switch to using the hvutil_device_state state machine
Switch to using the hvutil_device_state state machine from using kvp_transaction.active.
State transitions are:
-> HVUTIL_DEVICE_INIT when driver loads or on device release
-> HVUTIL_READY if the handshake was successful
-> HVUTIL_HOSTMSG_RECEIVED when there is a non-negotiation message from the host
-> HVUTIL_USERSPACE_REQ after we sent the message to the userspace daemon
-> HVUTIL_USERSPACE_RECV after/if the userspace daemon has replied
-> HVUTIL_READY after we respond to the host
-> HVUTIL_DEVICE_DYING on driver unload
In hv_vss_onchannelcallback() process ICMSGTYPE_NEGOTIATE messages even when
the userspace daemon is disconnected, otherwise we can make the host think
we don't support VSS and disable the service completely.
Unfortunately there is no good way we can figure out that the userspace daemon
has died (unless we start treating all timeouts as such), add a protection
against processing new VSS_OP_REGISTER messages while being in the middle of a
transaction (HVUTIL_USERSPACE_REQ or HVUTIL_USERSPACE_RECV state).
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Drivers: hv: kvp: switch to using the hvutil_device_state state machine
Switch to using the hvutil_device_state state machine from using 2 different state variables: kvp_transaction.active and
in_hand_shake.
State transitions are:
-> HVUTIL_DEVICE_INIT when driver loads or on device release
-> HVUTIL_READY if the handshake was successful
-> HVUTIL_HOSTMSG_RECEIVED when there is a non-negotiation message from the host
-> HVUTIL_USERSPACE_REQ after we sent the message to the userspace daemon
-> HVUTIL_USERSPACE_RECV after/if the userspace daemon has replied
-> HVUTIL_READY after we respond to the host
-> HVUTIL_DEVICE_DYING on driver unload
In hv_kvp_onchannelcallback() process ICMSGTYPE_NEGOTIATE messages even when
the userspace daemon is disconnected, otherwise we can make the host think
we don't support KVP and disable the service completely.
Unfortunately there is no good way we can figure out that the userspace daemon
has died (unless we start treating all timeouts as such). In case the daemon
restarts we skip the negotiation procedure (so the daemon is supposed to has
the same version). This behavior is unchanged from in_handshake approach.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Drivers: hv: util: introduce state machine for util drivers
KVP/VSS/FCOPY drivers work in fully serialized mode: we wait till userspace
daemon registers, wait for a message from the host, send this message to the
daemon, get the reply, send it back to host, wait for another message.
Introduce enum hvutil_device_state to represend this state in all 3 drivers.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
'fcopy_work' (and fcopy_work_func) is a misnomer as it sounds like we expect
this useful work to happen and in reality it is just an emergency escape when
timeout happens.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
'kvp_work' (and kvp_work_func) is a misnomer as it sounds like we expect
this useful work to happen and in reality it is just an emergency escape when
timeout happens.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Drivers: hv: vss: process deferred messages when we complete the transaction
In theory, the host is not supposed to issue any requests before be reply to
the previous one. In KVP we, however, support the following scenarios:
1) A message was received before userspace daemon registered;
2) A message was received while the previous one is still being processed.
In VSS we support only the former. Add support for the later, use
hv_poll_channel() to do the job.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Drivers: hv: fcopy: process deferred messages when we complete the transaction
In theory, the host is not supposed to issue any requests before be reply to
the previous one. In KVP we, however, support the following scenarios:
1) A message was received before userspace daemon registered;
2) A message was received while the previous one is still being processed.
In FCOPY we support only the former. Add support for the later, use
hv_poll_channel() to do the job.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Drivers: hv: kvp: move poll_channel() to hyperv_vmbus.h
Move poll_channel() to hyperv_vmbus.h and make it inline and rename it to hv_poll_channel() so it can be reused
in other hv_util modules.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We set kvp_context when we want to postpone receiving a packet from vmbus due
to the previous transaction being unfinished. We, however, never reset this
state, all consequent kvp_respond_to_host() calls will result in poll_channel()
calling hv_kvp_onchannelcallback(). This doesn't cause real issues as:
1) Host is supposed to serialize transactions as well
2) If no message is pending vmbus_recvpacket() will return 0 recvlen.
This is just a cleanup.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Drivers: hv: util: move kvp/vss function declarations to hyperv_vmbus.h
These declarations are internal to hv_util module and hv_fcopy_* declarations
already reside there.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by: Alex Ng <alexng@microsoft.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Instead, the code should return NULL to avoid the following null
pointer deference. Also, remove the error message as it is
redundant, the caller emits an error message to alert of a
failure anyhow.
Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
checkkconfigsymbols.py: add option -i to ignore files
Sometimes a user might be interested to filter certain reports (e.g.,
the many defconfigs). Now, this can be achieved by specifying a Python
regex with -i / --ignore.
misc: mic: MIC card driver specific changes to enable SCIF
MIC card driver specific changes to enable SCIF. This patch implements
the SCIF hardware bus operations and registers a SCIF device on the
SCIF hardware bus.
misc: mic: MIC host driver specific changes to enable SCIF
MIC host driver specific changes to enable SCIF. This patch implements
the SCIF hardware bus operations and registers a SCIF device on the
SCIF hardware bus.
misc: mic: SCIF messaging and node enumeration APIs
SCIF messaging APIs which allow sending messages between the SCIF
endpoints via a byte stream based ring buffer which has been
optimized to avoid reads across PCIe. The SCIF messaging APIs
are typically used for short < 1024 byte messages for best
performance while the RDMA APIs which will be submitted in a future
patch series is recommended for larger transfers. The node
enumeration API enables a user to query for the number of nodes
online in the SCIF network and their node ids.
misc: mic: SCIF connections APIs i.e. accept and connect
SCIF connection APIs which establish a SCIF connection between
a pair of SCIF endpoints. A SCIF connection consists of a
dedicated queue-pair between the endpoints. Client messages are
sent over the queue-pair whereas the signaling associated with the
message is multiplexed over the node queue-pair. Similarly other
control messages such as exposing registered memory are also sent
over the node queue-pair. The SCIF endpoints must be in connected
state to exchange messages, register memory, map remote memory and
trigger DMA transfers. SCIF connections can be set up
asynchronously or synchronously.
Thanks to Johnnie S Peters for authoring parts of this patch during
early bring up of the SCIF driver.
SCIF character device file operations and kernel APIs for opening and
closing a user and kernel mode SCIF endpoint. This patch also enables
binding to a SCIF port and listening for incoming SCIF connections.
SCIF node queue pair setup creates the SCIF driver kernel
mode private node queue pairs between all the nodes to enable
internal control message communication once SCIF gets probed
by the SCIF hardware bus. Peer to peer communication between
MIC Coprocessor nodes is supported.
SCIF module initialization, DMA mapping, ioremap wrapper APIs
and debugfs hooks. SCIF gets probed by the SCIF hardware bus
if SCIF devices were registered by base drivers. A MISC device
is registered to provide the SCIF character device interface.
The SCIF peer bus is used to register and unregister SCIF peer devices
internally by the SCIF driver to signify the addition and removal of
peer nodes respectively from the SCIF network. This simplifies remote node
handling within SCIF and will also be used to support device probe/remove
for SCIF client drivers (e.g. netdev over SCIF)
The SCIF hardware bus abstracts the low level hardware driver details
like interrupts and mapping remote memory so that the same SCIF driver
can work without any changes with the MIC host or card driver as long
as the hardware bus operations are implemented. The SCIF hardware
device is registered by the host and card drivers on the SCIF hardware
bus resulting in probing the SCIF driver.
SCIF ring buffer is a single producer, single consumer byte stream
ring buffer optimized for avoiding reads across the PCIe bus while
adding the required barriers and hardware workarounds for the MIC
Coprocessor. The ring buffer is used to implement a receive queue for
SCIF driver messaging between two nodes and for byte stream messaging
between SCIF endpoints. The existing in-kernel ring buffer was not
reused since it has not been designed for our use across the PCIe bus
where each node runs an independent OS. Each SCIF node has a receive
queue for every other SCIF node, and each connected endpoint has a
receive queue for messages from its peer. This pair of receive
queues is referred to as a SCIF queue pair.
This patch introduces the SCIF documentation in the header file
and describes the IOCTL interface for user mode. mic_overview.txt
is updated with documentation on SCIF and a new document
describing SCIF in more details is available in scif_overview.txt.
Matt Campbell [Tue, 28 Apr 2015 11:44:17 +0000 (07:44 -0400)]
w1: Add support for DS28EA00 sequence to w1-therm
This patch provides support for the DS28EA00 digital thermometer.
The DS28EA00 provides an additional two pins for implementing a sequence
detection algorithm. This feature allows you to determine the physical
location of the chip in the 1-wire bus without needing pre-existing
knowledge of the bus ordering. Support is provided through the sysfs
w1_seq file. The file will contain a single line with an integer value
representing the device index in the bus starting at 0.
David Fries [Sat, 9 May 2015 00:51:50 +0000 (19:51 -0500)]
w1_therm reference count family data
A temperature conversion can take 750 ms and when possible the
w1_therm slave driver drops the bus_mutex to allow other bus
operations, but that includes operations such as a periodic slave
search, which can remove this slave when it is no longer detected.
If that happens the sl->family_data will be freed and set to NULL
causing w1_slave_show to crash when it wakes up.
Dmitry Khromov [Tue, 12 May 2015 19:29:44 +0000 (22:29 +0300)]
w1: introduce an ability to specify microseconds bus scanning intervals
Some of 1-Wire devices commonly associated with physical access control
systems are attached/generate presence for as short as 100 ms - hence
the tens-to-hundreds milliseconds scan intervals are required.
Tomas Winkler [Thu, 7 May 2015 12:54:06 +0000 (15:54 +0300)]
NFC: mei_phy: adjust mei nfc header according the spec
1. mei_nfc_hci_hdr and mei_nfc_hdr are just the same thing so drop one
2. use mei_nfc_hdr structure as the member of the command and the reply
instead of replicating all header fields.
3. dump the header for easier debugging
Cc: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Tomas Winkler [Thu, 7 May 2015 12:54:02 +0000 (15:54 +0300)]
mei: bus: report also uuid in module alias
In order to automate modules matching add device uuid
which is reported in client enumeration, keep also
the name that is needed in for nfc distinguishing radio vendor
Report mei:name:uuid
Cc: linux-api@vger.kernel.org Cc: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Tomas Winkler [Thu, 7 May 2015 12:54:00 +0000 (15:54 +0300)]
mei: add also write waiting list to runtime pm blockers
The io callback is clear from write_waitling_list after
we receive interrupt from the hw to ack the write completion.
We need to wait for this interrupt deliver before we try
to enter low power state
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
mei: consume flow control on the first chunk of writing
Consume the write flow control on the first chunk of the write instead
of on the buffer completion.
We can safely assume that the consequent chunks have the flow control
granted.
This addresses two issues:
1. Blocks other callbacks from the same client riding on the client's
flow control and prevents interleaving of messages as FW cannot distinguish
between two messages from the same client.
2. Fixes single buffer flow control arbitration in a clean way, without
connection/disconnection book keeping
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fixed address is simplified FW client that doesn't require
connection and doesn't support flow control.
So it can be only one host client per fixed FW client.
Fixed client access is available only for drivers on mei bus,
connection from user-space is blocked.
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For ME clients that use single receiving buffer
the driver tracks credentials on mei_me_clients structure
for all connections. The driver needs to book keep the shared
resource correctly and track the connections, particularly
the credit has to be cleaned when there is no active connection
to a particular me client. This solves issue when subsequent
connection will not get an ill impression that it can write.
We add active connection counter the particular ME client and
when the counter reach zero, we clear the credits.
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
mei: add a reference from the host client to the me client
Keep a pointer to associated me client in the host client object to
eliminate me client searches. Check if the me client is active in the
firmware by checking if its is linked on the me clients list
Add accessors for the me client properties from host client.
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Tomas Winkler [Mon, 4 May 2015 06:43:53 +0000 (09:43 +0300)]
mei: revamp client connection
Simplify connect state machine by changing the logic around
Connection request in progress - only check if we have a callback in
relevant queue.
Extract common code into mei_cl_send_connect() function
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Tomas Winkler [Mon, 4 May 2015 06:43:52 +0000 (09:43 +0300)]
mei: revamp client disconnection flow
Split disconnected state into two parts first reception disconnect
response from the firmware and second actually setting of disconnected
state. Book keeping data are needed for processing and after firmware
disconnected the client and are cleaned when setting the disconnected
state in mei_cl_set_disconneted() function.
Add mei_cl_send_disconnect to reduce code duplication.
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Tomas Winkler [Tue, 14 Apr 2015 07:27:26 +0000 (10:27 +0300)]
mei: txe: reduce suspend/resume time
HW has to be in known state before the initialisation
sequence is started. The polling step for settling aliveness
was set to 200ms while in practise this can be done in up to 30msecs.
This driver manages Qualcomm CoreSight Replicator device, which
resides on the AMBA bus. Replicator has been made programmable to
allow software to turn of the replicator branch to sink that is not
being used. This avoids trace traffic to the unused/non-current sink
from causing back pressure that results in overflows at the source.
Signed-off-by: Pratik Patel <pratikp@codeaurora.org> Signed-off-by: Ivan T. Ivanov <ivan.ivanov@linaro.org> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Linus Walleij [Tue, 19 May 2015 16:55:18 +0000 (10:55 -0600)]
coresight: replicator: retrieve and handle atclk
As can be seen from the datasheet of the CoreSight
Components, DDI0314 table A-4 the funnel has a clock signal
apart from the AHB interconnect ("amba_pclk", that we're
already handling) called ATCLK, ARM Trace Clock, that SoC
implementers may provide from an entirely different clock
source. So to model this correctly create an optional
path for handling ATCLK alongside the PCLK so we don't
break old platforms that only define PCLK ("amba_pclk") but
still makes it possible for SoCs that have both clock signals
(such as the DB8500) to fetch and prepare/enable/disable/
unprepare both clocks.
The ATCLK is enabled and disabled using the runtime PM
callbacks. As the replicator is a platform device, the
code is a bit different from the other CoreSight components
and the bus core does not activate runtime PM by default,
so we need a few extra calls.
Linus Walleij [Tue, 19 May 2015 16:55:17 +0000 (10:55 -0600)]
coresight: funnel: retrieve and handle atclk
As can be seen from the datasheet of the CoreSight
Components, DDI0314 table A-6 the funnel has a clock signal
apart from the AHB interconnect ("amba_pclk", that we're
already handling) called ATCLK, ARM Trace Clock, that SoC
implementers may provide from an entirely different clock
source. So to model this correctly create an optional
path for handling ATCLK alongside the PCLK so we don't
break old platforms that only define PCLK ("amba_pclk") but
still makes it possible for SoCs that have both clock signals
(such as the DB8500) to fetch and prepare/enable/disable/
unprepare both clocks.
The ATCLK is enabled and disabled using the runtime PM
callbacks.
Linus Walleij [Tue, 19 May 2015 16:55:16 +0000 (10:55 -0600)]
coresight: etb: retrieve and handle atclk
As can be seen from the datasheet of the CoreSight
Components, DDI0314 table A-8 the ETB has a clock signal
apart from the AHB interconnect ("amba_pclk", that we're
already handling) called ATCLK, ARM Trace Clock, that SoC
implementers may provide from an entirely different clock
source. So to model this correctly create an optional
path for handling ATCLK alongside the PCLK so we don't
break old platforms that only define PCLK ("amba_pclk") but
still makes it possible for SoCs that have both clock signals
(such as the DB8500) to fetch and prepare/enable/disable/
unprepare both clocks.
The ATCLK is enabled and disabled using the runtime PM
callbacks.
Linus Walleij [Tue, 19 May 2015 16:55:15 +0000 (10:55 -0600)]
coresight: tpiu: retrieve and handle atclk
As can be seen from the datasheet of the CoreSight
Components, DDI0314H page A-19 the TPIU has a clock signal
apart from the AHB interconnect ("amba_pclk", that we're
already handling) called ATCLK, ARM Trace Clock, that SoC
implementers may provide from an entirely different clock
source. So to model this correctly create an optional
path for handling ATCLK alongside the PCLK so we don't
break old platforms that only define PCLK ("amba_pclk") but
still makes it possible for SoCs that have both clock signals
(such as the DB8500) to fetch and prepare/enable/disable/
unprepare both clocks in conjunction.
The ATCLK is enabled and disabled using the runtime PM
callbacks.
Linus Walleij [Tue, 19 May 2015 16:55:14 +0000 (10:55 -0600)]
coresight: etm: retrieve and handle atclk
As can be seen from the datasheet of the CoreSight
Components, DDI0401C A.1.1 the ETM has a clock signal
apart from the AHB interconnect ("amba_pclk", that we're
already handling) called ATCLK, ARM Trace Clock, that SoC
implementers may provide from an entirely different clock
source. So to model this correctly create an optional
path for handling ATCLK alongside the PCLK so we don't
break old platforms that only define PCLK ("amba_pclk") but
still makes it possible for SoCs that have both clock signals
(such as the DB8500) to fetch and prepare/enable/disable/
unprepare both clocks.
The ATCLK is enabled and disabled using the runtime PM
callbacks.
Linus Walleij [Tue, 19 May 2015 16:55:13 +0000 (10:55 -0600)]
coresight: tmc: let runtime PM handle core clock
This uses runtime PM to manage the PCLK ("amba_pclk") instead
of screwing around with the framework by going in and taking
a copy from the amba device. The amba bus core will unprepare
and disable the clock when the device is unused when
CONFIG_PM is selected, else the clock will be always on.
Prior to this patch, as the AMBA primecell bus code enables
the PCLK, it would be left on after probe as
the clk_prepare_enable() and clk_disable_unprepare() was
called and thus just increase and decreas the refcount by
one, without it reaching zero and actually disabling the
clock. Now the runtime PM callbacks will make sure the PCLK
is properly disabled after probe.
Linus Walleij [Tue, 19 May 2015 16:55:12 +0000 (10:55 -0600)]
coresight: funnel: let runtime PM handle core clock
This uses runtime PM to manage the PCLK ("amba_pclk") instead
of screwing around with the framework by going in and taking
a copy from the amba device. The amba bus core will unprepare
and disable the clock when the device is unused when
CONFIG_PM is selected, else the clock will be always on.
Prior to this patch, as the AMBA primecell bus code enables
the PCLK, it would be left on after probe as
clk_disable_unprepare() was not called. Now the runtime PM
callbacks will make sure the PCLK is properly disabled
after probe.
Linus Walleij [Tue, 19 May 2015 16:55:11 +0000 (10:55 -0600)]
coresight: etb: let runtime PM handle core clock
This uses runtime PM to manage the PCLK ("amba_pclk") instead
of screwing around with the framework by going in and taking
a copy from the amba device. The amba bus core will unprepare
and disable the clock when the device is unused when
CONFIG_PM is selected, else the clock will be always on.
Prior to this patch, as the AMBA primecell bus code enables
the PCLK, it would be left on after probe as
the clk_prepare_enable() and clk_disable_unprepare() was
called and thus just increase and decreas the refcount by
one, without it reaching zero and actually disabling the
clock. Now the runtime PM callbacks will make sure the PCLK
is properly disabled after probe.
Linus Walleij [Tue, 19 May 2015 16:55:10 +0000 (10:55 -0600)]
coresight: tpiu: let runtime PM handle core clock
This uses runtime PM to manage the PCLK ("amba_pclk") instead
of screwing around with the framework by going in and taking
a copy from the amba device. The amba bus core will unprepare
and disable the clock when the device is unused when
CONFIG_PM is selected, else the clock will be always on.
Prior to this patch, as the AMBA primecell bus code enables
the PCLK, it would be left on after probe as
the clk_prepare_enable() and clk_disable_unprepare() was
called and thus just increase and decreas the refcount by
one, without it reaching zero and actually disabling the
clock. Now the runtime PM callbacks will make sure the PCLK
is properly disabled after probe.
Linus Walleij [Tue, 19 May 2015 16:55:09 +0000 (10:55 -0600)]
coresight: etm: let runtime PM handle core clock
This uses runtime PM to manage the PCLK ("amba_pclk") instead
of screwing around with the framework by going in and taking
a copy from the amba device. The amba bus core will unprepare
and disable the clock when the device is unused when
CONFIG_PM is selected, else the clock will be always on.
Prior to this patch, as the AMBA primecell bus code enables
the PCLK, it would be left on after probe as
the clk_prepare_enable() and clk_disable_unprepare() was
called and thus just increase and decreas the refcount by
one, without it reaching zero and actually disabling the
clock. Now the runtime PM callbacks will make sure the PCLK
is properly disabled after probe.
Linus Walleij [Tue, 19 May 2015 16:55:07 +0000 (10:55 -0600)]
coresight: etm: print what version of ETM/PTM is detected
Helpfully report a bit more about the hardware found in the
silicon when matching the AMBA device IDs by using the associated
.data pointer in the AMBA match.
Mathieu Poirier [Wed, 13 May 2015 16:34:21 +0000 (10:34 -0600)]
coresight-etm4x: Read only access to the tracer's ID registers
ETM ID registers contain valuable information about the capabilities
of the implementation and are very useful when configuring the device for
various trace scenarios.
Mathieu Poirier [Wed, 13 May 2015 16:34:20 +0000 (10:34 -0600)]
coresight-etm4x: Read only access to the main management registers
Having access to the ETMv4 management registers is very useful as they
give meaningful information on how the IP block has been configured at
synthesis time.
Pratik Patel [Wed, 13 May 2015 16:34:15 +0000 (10:34 -0600)]
coresight-etm4x: Controls pertaining to the sequencer functions
Adding sysfs entries to access the sequencers related registers,
more specifically the sequencer state, the sequencer state
transition and the sequencer reset control registers.
Pratik Patel [Wed, 13 May 2015 16:34:14 +0000 (10:34 -0600)]
coresight-etm4x: Controls pertaining to the address comparator functions
Adding sysfs entries to control the various mode the address comparator
registers can enact, i.e, start/top, single, and range. Also supplementing
with address comparator types configuration registers access, mandatory
to complete the configuration of the comparator functions.
Pratik Patel [Wed, 13 May 2015 16:34:12 +0000 (10:34 -0600)]
coresight-etm4x: Controls pertaining to various configuration options
Adding sysfs entries to configure:
. global timestamp.
. how often trace synchronisation occur.
. the threashold value for cycle counting.
. branch and broadcasting regions.
Pratik Patel [Wed, 13 May 2015 16:34:11 +0000 (10:34 -0600)]
coresight-etm4x: Controls pertaining to the reset, mode, pe and events
Adding sysfs entries to:
. set the tracing entity with default values.
. set various mode associated to the tracing entity.
. select the processing entity the tracing entity relates to.
. select various events of interest.
Mark Brown [Fri, 10 Apr 2015 15:25:37 +0000 (09:25 -0600)]
coresight: etb10: Fix check for bogus buffer depth
We attempt to sanity check the buffer depth reported by the hardware by
making sure it is not less than zero however this check will never be true
since the buffer depth is stored in an unsigned integer. Instead change
the check to look for the top bit being set which was the intention.
Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Mark Brown [Fri, 10 Apr 2015 15:25:36 +0000 (09:25 -0600)]
coresight: etb10: Print size of buffer we fail to allocate
When we initialise the ETB driver we attempt to allocate a buffer suitable
for storing the data buffered in the hardware based on sizing information
reported by the hardware. Unfortunately if the hardware is not properly
configured (for example if power domains are not set up correctly) then we
may read back a nonsensically large value and therefore the allocation will
be too big to succeed. Print an error message showing the amount of memory
we tried to allocate if the buffer allocation fails to help users diagnose
such problems.
Normally it is bad practice to print an error message on memory allocation
failures since there are verbose core messages reported for this but in
this case where the allocation size might be incorrect it is a useful hint.
Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / linux-beck.git / log