Commit graph

661,541 commits

Author SHA1 Message Date
Lv Zheng
0fc5e8f4e4 ACPICA: Hardware: Add sleep register hooks
ACPICA commit ba665dc8e20d9f7730466a659564dd6c557a6cbc

In Linux, para-virtualization implmentation hooks critical register
writes to prevent real hardware operations. This increases divergences
when the sleep registers are cracked in Linux resident ACPICA.

This patch tries to introduce a single OSL to reduce the divergences.

Link: ba665dc8
Signed-off-by: Lv Zheng <lv.zheng@intel.com>
Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-01-02 23:18:41 +01:00
Bob Moore
fcfb45531d ACPICA: Macro header: Fix some typos in comments
ACPICA commit efc97d1d209947d6990ec81a192c6b2589d3e368

No functional change.

Link: efc97d1
Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Lv Zheng <lv.zheng@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-01-02 23:18:40 +01:00
Lv Zheng
04cf053799 ACPICA: Hardware: Sort access bit width algorithm
ACPICA commit 365b321a31cb701957c055cae2d2161577147252

GAS can be in register or register region format, so we need to
improve our "register" format detection code in order not to
regress.

Such detection may be still experimental, and is generated according
to the current known facts.

Link: 365b321a
Link: https://bugzilla.kernel.org/show_bug.cgi?id=151501
Reported-and-tested-by: Andrey Skvortsov <andrej.skvortzov@gmail.com>
Signed-off-by: Lv Zheng <lv.zheng@intel.com>
Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-01-02 23:18:38 +01:00
Lv Zheng
cc573b97c8 ACPICA: Utilities: Add power of two rounding support
ACPICA commit cbb0294649cbd7e8bd6107e4329461a6a7a0d967

This patch adds power of two rounding support up to 32 bits.

The result of the shift operations rearching to the boundary of the cpu
word is unpredicatable, so 64-bit roundings are not supported in order to
make sure no rounded shift-overs.

This support may not be performance friendly, so the APIs might be
overridden by the hosts implementations with ACPI_USE_NATIVE_BIT_FINDER
defined.

Link: cbb02946
Signed-off-by: Lv Zheng <lv.zheng@intel.com>
Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-01-02 23:18:38 +01:00
Lv Zheng
dc4c737665 ACPICA: Hardware: Add access_width/bit_offset support in acpi_hw_write()
ACPICA commit 1ecab20bbe69a176dfb6da7210fe77aa6b3ad680

This patch adds access_width/bit_offset support in acpi_hw_write().

Link: 1ecab20b
Signed-off-by: Lv Zheng <lv.zheng@intel.com>
Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-01-02 23:18:37 +01:00
Lv Zheng
123a1577b7 ACPICA: Hardware: Remove bit_offset masking support
ACPICA commit bc7c5291865e099ce01f345d0265f0eba6997e23

This linuxized ACPICA commit is a back port result of the following
Linux commit:

  Commit c3bc26d4b4
  Subject: ACPICA: ACPI 2.0, Hardware: Add access_width/bit_offset
           support in acpi_hw_read()

The commit was in ACPICA and Linux upstream, after reversion and
re-integration, it is designed not to do bit_offset masking (bit_offset is
only used to determine the boundary of the register) inside of the ACPICA
APIs, but let the callers to do that as:

 1. Register can have different masking schemes (W1C, W0C);
 2. Normally a mask value will be provided for region format GAS.

So actually the callers are the only ones having the knowledge of masking
the register values. Suggested by Bob Moore, Fixed by Lv Zheng.

Link: bc7c5291
Signed-off-by: Lv Zheng <lv.zheng@intel.com>
Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-01-02 23:18:35 +01:00
Lv Zheng
703ecd220d ACPICA: Debugger: Rename debugger OSL names
ACPICA commit e76eb8b36ace880e4d475880db1128a206e57b6f

This linuxized ACPICA commit is a back port result of the following
linux commit:

  Commit: f8d3148962
  Subject: ACPICA: Debugger: Convert some mechanisms to OSPM specific

During the back porting, it is requested by ACPICA to use expected OSL
names. Suggested by Bob Moore, Fixed by Lv Zheng.

Linux is not affected by this patch.

Link: e76eb8b3
Signed-off-by: Lv Zheng <lv.zheng@intel.com>
Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-01-02 23:18:34 +01:00
Santosh Shilimkar
3289025aed RDS: add receive message trace used by application
Socket option to tap receive path latency in various stages
in nano seconds. It can be enabled on selective sockets using
using SO_RDS_MSG_RXPATH_LATENCY socket option. RDS will return
the data to application with RDS_CMSG_RXPATH_LATENCY in defined
format. Scope is left to add more trace points for future
without need of change in the interface.

Reviewed-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:59 -08:00
Avinash Repaka
f9fb69adb6 RDS: make message size limit compliant with spec
RDS support max message size as 1M but the code doesn't check this
in all cases. Patch fixes it for RDMA & non-RDMA and RDS MR size
and its enforced irrespective of underlying transport.

Signed-off-by: Avinash Repaka <avinash.repaka@oracle.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:57 -08:00
Venkat Venkatsubra
192a798f52 RDS: add stat for socket recv memory usage
Tracks the receive side memory added to scokets and removed from sockets.

Signed-off-by: Venkat Venkatsubra <venkat.x.venkatsubra@oracle.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:56 -08:00
Santosh Shilimkar
cf657269d3 RDS: IB: fix panic due to handlers running post teardown
Shutdown code reaping loop takes care of emptying the
CQ's before they being destroyed. And once tasklets are
killed, the hanlders are not expected to run.

But because of core tasklet code issues, tasklet handler could
still run even after tasklet_kill,
RDS IB shutdown code already reaps the CQs before freeing
cq/qp resources so as such the handlers have nothing left
to do post shutdown.

On other hand any handler running after teardown and trying
to access already freed qp/cq resources causes issues
Patch fixes this race by  makes sure that handlers returns
without any action post teardown.

Reviewed-by: Wengang <wen.gang.wang@oracle.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:55 -08:00
Santosh Shilimkar
941f8d55f6 RDS: RDMA: Fix the composite message user notification
When application sends an RDS RDMA composite message consist of
RDMA transfer to be followed up by non RDMA payload, it expect to
be notified *only* when the full message gets delivered. RDS RDMA
notification doesn't behave this way though.

Thanks to Venkat for debug and root casuing the issue
where only first part of the message(RDMA) was
successfully delivered but remainder payload delivery failed.
In that case, application should not be notified with
a false positive of message delivery success.

Fix this case by making sure the user gets notified only after
the full message delivery.

Reviewed-by: Venkat Venkatsubra <venkat.x.venkatsubra@oracle.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:54 -08:00
Santosh Shilimkar
be2f76eacc RDS: IB: Add vector spreading for cqs
Based on available device vectors, allocate cqs accordingly to
get better spread of completion vectors which helps performace
great deal..

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:52 -08:00
Santosh Shilimkar
09b2b8f528 RDS: IB: add few useful cache stasts
Tracks the ib receive cache total, incoming and frag allocations.

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:51 -08:00
Santosh Shilimkar
581d53c91c RDS: IB: track and log active side endpoint in connection
Useful to know the active and passive end points in a
RDS IB connection.

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:50 -08:00
Santosh Shilimkar
c536a06887 RDS: RDMA: silence the use_once mr log flood
In absence of extension headers, message log will keep
flooding the console. As such even without use_once we can
clean up the MRs so its not really an error case message
so make it debug message

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:49 -08:00
Santosh Shilimkar
5601245931 RDS: IB: split the mr registration and invalidation path
MR invalidation in RDS is done in background thread and not in
data path like registration. So break the dependency between them
which helps to remove the performance bottleneck.

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:47 -08:00
Santosh Shilimkar
584a8279a4 RDS: RDMA: return appropriate error on rdma map failures
The first message to a remote node should prompt a new
connection even if it is RDMA operation. For RDMA operation
the MR mapping can fail because connections is not yet up.

Since the connection establishment is asynchronous,
we make sure the map failure because of unavailable
connection reach to the user by appropriate error code.
Before returning to the user, lets trigger the connection
so that its ready for the next retry.

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:46 -08:00
Qing Huang
8d5d8a5fd7 RDS: RDMA: start rdma listening after init
This prevents RDS from handling incoming rdma packets before RDS
completes initializing its recv/send components.

Signed-off-by: Qing Huang <qing.huang@oracle.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:45 -08:00
Santosh Shilimkar
3e56c2f856 RDS: RDMA: fix the ib_map_mr_sg_zbva() argument
Fixes warning: Using plain integer as NULL pointer

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:44 -08:00
Santosh Shilimkar
fab8688d71 RDS: IB: make the transport retry count smallest
Transport retry is not much useful since it indicate packet loss
in fabric so its better to failover fast rather than longer retry.

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:43 -08:00
Santosh Shilimkar
ff3f19a2f6 RDS: IB: include faddr in connection log
Also use pr_* for it.

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:41 -08:00
Santosh Shilimkar
bb7897631d RDS: mark few internal functions static to make sparse build happy
Fixes below warnings:
warning: symbol 'rds_send_probe' was not declared. Should it be static?
warning: symbol 'rds_send_ping' was not declared. Should it be static?
warning: symbol 'rds_tcp_accept_one_path' was not declared. Should it be static?
warning: symbol 'rds_walk_conn_path_info' was not declared. Should it be static?

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:40 -08:00
Santosh Shilimkar
f69b22e65e RDS: log the address on bind failure
It's useful to know the IP address when RDS fails to bind a
connection. Thus, adding it to the error message.

Orabug: 21894138
Reviewed-by: Wei Lin Guay <wei.lin.guay@oracle.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
2017-01-02 14:02:39 -08:00
Philippe Reynes
0a0a8d6b0e net: fealnx: use new api ethtool_{get|set}_link_ksettings
The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 16:59:10 -05:00
Philippe Reynes
9da12b6454 net: faraday: ftmac100: use new api ethtool_{get|set}_link_ksettings
The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 16:59:10 -05:00
Philippe Reynes
fd90095db5 net: emulex: benet: use new api ethtool_{get|set}_link_ksettings
The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 16:59:10 -05:00
Philippe Reynes
88c513d478 net: dlink: sundance: use new api ethtool_{get|set}_link_ksettings
The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 16:59:10 -05:00
Philippe Reynes
9022255006 net: dlink: dl2k: use new api ethtool_{get|set}_link_ksettings
The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

The previous implementation of set_settings was modifying
the value of speed and duplex, but with the new API, it's not
possible. The structure ethtool_link_ksettings is defined
as const.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 16:59:10 -05:00
Philippe Reynes
453e5f3dfc net: dec: winbond-840: use new api ethtool_{get|set}_link_ksettings
The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 16:59:10 -05:00
Philippe Reynes
6711a87a9b net: dec: uli526x: use new api ethtool_{get|set}_link_ksettings
The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 16:59:10 -05:00
Philippe Reynes
41a65f7040 net: dec: de2104x: use new api ethtool_{get|set}_link_ksettings
The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 16:59:10 -05:00
Philippe Reynes
e938ed150f net: sfc: falcon: use new api ethtool_{get|set}_link_ksettings
The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 16:59:10 -05:00
Philippe Reynes
8e4881aa1d net: mdio: add mdio45_ethtool_ksettings_get
There is a function in mdio for the old ethtool api gset.
We add a new function mdio45_ethtool_ksettings_get for the
new ethtool api glinksettings.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 16:59:10 -05:00
Bjorn Helgaas
977509f7c5 PCI: Apply _HPX settings only to relevant devices
Previously we didn't check the type of device before trying to apply Type 1
(PCI-X) or Type 2 (PCIe) Setting Records from _HPX.

We don't support PCI-X Setting Records, so this was harmless, but the
warning was useless.

We do support PCIe Setting Records, and we didn't check whether a device
was PCIe before applying settings.  I don't think anything bad happened on
non-PCIe devices because pcie_capability_clear_and_set_word(),
pcie_cap_has_lnkctl(), etc., would fail before doing any harm.  But it's
ugly to depend on those internals.

Check the device type before attempting to apply Type 1 and Type 2 Setting
Records (Type 0 records are applicable to PCI, PCI-X, and PCIe devices).

A side benefit is that this prevents useless "not supported" warnings when
a BIOS supplies a Type 1 (PCI-X) Setting Record and we try to apply it to
every single device:

  pci 0000:00:00.0: PCI-X settings not supported

After this patch, we'll get the warning only when a BIOS supplies a Type 1
record and we have a PCI-X device to which it should be applied.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=187731
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2017-01-02 15:56:25 -06:00
Christophe JAILLET
0f0861e31e drm/sun4i: Fix a return value in case of error
If 'sun4i_backend_drm_format_to_layer()' does not return 0, then 'val' is
left unmodified.
As it is not initialized either, the return value can be anything.

It is likely that returning the error code was expected here.

As the only caller of 'sun4i_backend_update_layer_formats()' does not check
the return value, this fix is purely theorical.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
2017-01-02 22:27:58 +01:00
Quentin Schulz
11937f69b6 ARM: sunxi: Add CONFIG_THERMAL_OF
This enables CONFIG_THERMAL_OF by default for sunxi_defconfig. It is
required to get Allwinner SoCs' temperature from the GPADC driver.

The Allwinner SoCs all have an ADC that can also act as a touchscreen
controller and a thermal sensor. The first four channels can be used
either for the ADC or the touchscreen and the fifth channel is used for
the thermal sensor.

The thermal sensor requires the IP to be in touchscreen mode to return
correct values. Therefore, if the user is continuously reading the ADC
channel(s), the thermal framework in which the thermal sensor is
registered will switch the IP in touchscreen mode to get a temperature
value and requires a delay of 100ms (because of the mode switching),
then the ADC will switch back to ADC mode and requires also a delay of
100ms. If the ADC readings are critical to user and the SoC temperature
is not, the GPADC driver is capable of not registering the thermal
sensor in the thermal framework and thus, "quicken" the ADC readings. In
most use cases, the SoC temperature is more critical (for cpu throttling
for example or activating cooling devices) than ADC readings, thus it is
now enabled by default.

Signed-off-by: Quentin Schulz <quentin.schulz@free-electrons.com>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
2017-01-02 22:27:19 +01:00
Icenowy Zheng
16fa9629c4 ARM: sunxi: add support for H2+ SoC
Allwinner H2+ is a quad-core Cortex-A7 SoC.

It is very like H3, that they share the same SoC ID (0x1680), and H3
memory maps as well as drivers works well on the SoC.

Signed-off-by: Icenowy Zheng <icenowy@aosc.xyz>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
2017-01-02 22:25:32 +01:00
Marcus Cooper
70421257c0 clk: sunxi-ng: A31: Fix spdif clock register
As the SPDIF was rarely documented on the earlier Allwinner SoCs
it was assumed that it had a similar clock register to the one
described in the H3 User Manual.

However this is not the case and it looks to shares the same setup
as the I2S clock registers.

Signed-off-by: Marcus Cooper <codekipper@gmail.com>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
2017-01-02 22:24:55 +01:00
Icenowy Zheng
bb021cda2c clk: sunxi-ng: set the parent rate when adjustin CPUX clock on A33
The CPUX clock on A33, which is for the Cortex-A7 cores, is designed to
be changeable by changing the rate of PLL_CPUX.

Add CLK_SET_RATE_PARENT flag to this clock.

Signed-off-by: Icenowy Zheng <icenowy@aosc.xyz>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
2017-01-02 22:24:55 +01:00
Icenowy Zheng
790d929b54 clk: sunxi-ng: fix PLL_CPUX adjusting on A33
When adjusting PLL_CPUX on A33, the PLL is temporarily driven too high,
and the system hangs.

Add a notifier to avoid this situation by temporarily switching to a
known stable 24 MHz oscillator.

Signed-off-by: Icenowy Zheng <icenowy@aosc.xyz>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
2017-01-02 22:24:55 +01:00
Ondrej Jirman
a43c96427e clk: sunxi-ng: fix PLL_CPUX adjusting on H3
When adjusting PLL_CPUX on H3, the PLL is temporarily driven
too high, and the system becomes unstable (oopses or hangs).

Add a notifier to avoid this situation by temporarily switching
to a known stable 24 MHz oscillator.

Signed-off-by: Ondrej Jirman <megous@megous.com>
Tested-by: Lutz Sammer <johns98@gmx.net>
Acked-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
2017-01-02 22:24:55 +01:00
Rafael J. Wysocki
c2a6bbaf0c ACPI / scan: Prefer devices without _HID/_CID for _ADR matching
The way acpi_find_child_device() works currently is that, if there
are two (or more) devices with the same _ADR value in the same
namespace scope (which is not specifically allowed by the spec and
the OS behavior in that case is not defined), the first one of them
found to be present (with the help of _STA) will be returned.

This covers the majority of cases, but is not sufficient if some of
the devices in question have a _HID (or _CID) returning some valid
ACPI/PNP device IDs (which is disallowed by the spec) and the
ASL writers' expectation appears to be that the OS will match
devices without a valid ACPI/PNP device ID against a given bus
address first.

To cover this special case as well, modify find_child_checks()
to prefer devices without ACPI/PNP device IDs over devices that
have them.

Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Tested-by: Hans de Goede <hdegoede@redhat.com>
2017-01-02 22:21:59 +01:00
David S. Miller
525dfa2cdc Merge branch 'mlx5-odp'
Saeed Mahameed says:

====================
Mellanox mlx5 core and ODP updates 2017-01-01

The following eleven patches mainly come from Artemy Kovalyov
who expanded mlx5 on-demand-paging (ODP) support. In addition
there are three cleanup patches which don't change any functionality,
but are needed to align codebase prior accepting other patches.

Memory region (MR) in IB can be huge and ODP (on-demand paging)
technique allows to use unpinned memory, which can be consumed and
released on demand. This allows to applications do not pin down
the underlying physical pages of the address space, and save from them
need to track the validity of the mappings.

Rather, the HCA requests the latest translations from the OS when pages
are not present, and the OS invalidates translations which are no longer
valid due to either non-present pages or mapping changes.

In existing ODP implementation applications is needed to register
memory buffers for communication, though registered memory regions
need not have valid mappings at registration time.

This patch set performs the following steps to expand
current ODP implementation:

1. It refactors UMR to support large regions, by introducing generic
   function to perform HCA translation table modifications. This
   function supports both atomic and process contexts and is not limited
   by number of modified entries.

   This function allows to enable reallocated memory regions of
   arbitrary size, so adding MR cache buckets to support up to 16GB MRs.

2. It changes page fault event format and refactor page faults logic
   together with addition of atomic support.

3. It prepares mlx5 core code to support implicit registration with
   simplified and relaxed semantics.

   Implicit ODP semantics allows to applications provide special memory
   key that represents their complete address space. Thus all IO accesses
   referencing to this key (with proper access rights associated with the key)
   wouldn't need not register any virtual address range.

Thanks,
        Artemy, Ilya and Leon

v1->v2:
  - Don't use 'inline' in .c files
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 15:51:21 -05:00
Artemy Kovalyov
aa8e08d2f5 IB/mlx5: Improve MR check
Add "type" field to mlx5_core MKEY struct.
Check whether page fault happens on MKEY corresponding to MR.

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 15:51:20 -05:00
Artemy Kovalyov
17d2f88f92 IB/mlx5: Add ODP atomics support
Handle ODP atomic operations. When initiator of RDMA atomic
operation use ODP MR to provide source data handle pagefault properly.

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 15:51:20 -05:00
Artemy Kovalyov
d9aaed8387 {net,IB}/mlx5: Refactor page fault handling
* Update page fault event according to last specification.
* Separate code path for page fault EQ, completion EQ and async EQ.
* Move page fault handling work queue from mlx5_ib static variable
  into mlx5_core page fault EQ.
* Allocate memory to store ODP event dynamically as the
  events arrive, since in atomic context - use mempool.
* Make mlx5_ib page fault handler run in process context.

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 15:51:20 -05:00
Artemy Kovalyov
223cdc7242 net/mlx5: Update PAGE_FAULT_RESUME layout
Update PAGE_FAULT_RESUME command layout.

Three bit fields describing page fault: rdma, rdma_write, req_res gave 8
possible combinations, while only a few were legal. Now they
are interpreted as three-bit type field, where former legal
combinations turns into corresponding types and unused were added as new
types.

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 15:51:20 -05:00
Artemy Kovalyov
7d0cc6edcc IB/mlx5: Add MR cache for large UMR regions
In this change we turn mlx5_ib_update_mtt() into generic
mlx5_ib_update_xlt() to perfrom HCA translation table modifiactions
supporting both atomic and process contexts and not limited by number
of modified entries.
Using this function we increase preallocated MRs up to 16GB.

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 15:51:20 -05:00
Artemy Kovalyov
c438fde1c2 IB/mlx5: Add support for big MRs
Make use of extended UMR translation offset.

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-02 15:51:20 -05:00