In the device, VxLAN encapsulation takes place in the FDB table where
certain {MAC, FID} entries are programmed with an underlay unicast IP.
MAC addresses that are not programmed in the FDB are flooded to the
relevant local ports and also to a list of underlay unicast IPs that are
programmed using the all zeros MAC address in the VxLAN driver.
One difference between the hardware and software data paths is the fact
that in the software data path there are two FDB lookups prior to the
encapsulation of the packet. First in the bridge's FDB table using {MAC,
VID} and another in the VxLAN's FDB table using {MAC, VNI}.
Therefore, when a new VxLAN FDB entry is notified, it is only programmed
to the device if there is a corresponding entry in the bridge's FDB
table. Similarly, when a new bridge FDB entry pointing to the VxLAN
device is notified, it is only programmed to the device if there is a
corresponding entry in the VxLAN's FDB table.
Note that the above scheme will result in a discrepancy between both
data paths if only one FDB table is populated in the software data path.
For example, if only the bridge's FDB is populated with an entry
pointing to a VxLAN device, then a packet hitting the entry will only be
flooded by the kernel to remote VTEPs whereas the device will also flood
the packets to other local ports member in the VLAN.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Enslavement of VxLAN devices to offloaded bridges was never forbidden by
mlxsw, but this patch makes sure the required configuration is performed
in order to allow VxLAN encapsulation and decapsulation to take place in
the device.
The patch handles both the case where a VxLAN device is enslaved to an
already offloaded bridge and the case where the first mlxsw port is
enslaved to a bridge that already has VxLAN device configured.
Invalid configurations are sanitized and an error string is returned via
extack.
Since encapsulation and decapsulation do not occur when the VxLAN device
is down, the driver makes sure to enable / disable these functionalities
based on NETDEV_PRE_UP and NETDEV_DOWN events.
Note that NETDEV_PRE_UP is used in favor of NETDEV_UP, as the former
allows to veto the operation, if necessary.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, an FDB entry only ceases being offloaded when it is deleted.
This changes with VxLAN encapsulation.
Devices capable of performing VxLAN encapsulation usually have only one
FDB table, unlike the software data path which has two - one in the
bridge driver and another in the VxLAN driver.
Therefore, bridge FDB entries pointing to a VxLAN device are only
offloaded if there is a corresponding entry in the VxLAN FDB.
Allow clearing the offload indication in case the corresponding entry
was deleted from the VxLAN FDB.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When notifications are sent about FDB activity, and an FDB entry with
several remotes is removed, the notification is sent only for the first
destination. That makes it impossible to distinguish between the case
where only this first remote is removed, and the one where the FDB entry
is removed as a whole.
Therefore send one notification for each remote of a removed FDB entry.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Offloaded bridge FDB entries are marked with NTF_OFFLOADED. Implement a
similar mechanism for VXLAN, where a given remote destination can be
marked as offloaded.
To that end, introduce a new event, SWITCHDEV_VXLAN_FDB_OFFLOADED,
through which the marking is communicated to the vxlan driver. To
identify which RDST should be marked as offloaded, an
switchdev_notifier_vxlan_fdb_info is passed to the listeners. The
"offloaded" flag in that object determines whether the offloaded mark
should be set or cleared.
When sending offloaded FDB entries over netlink, mark them with
NTF_OFFLOADED.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A switchdev-capable driver that is aware of VXLAN may need to query
VXLAN FDB. In the particular case of mlxsw, this functionality is
limited to querying UC FDBs. Those being easier to deal with than the
general case of RDST chain traversal, introduce an interface to query
specifically UC FDBs: vxlan_fdb_find_uc().
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When offloading VXLAN devices, drivers need to know about events in
VXLAN FDB database. Since VXLAN models a bridge, it is natural to
distribute the VXLAN FDB notifications using the pre-existing switchdev
notification mechanism.
To that end, introduce two new notification types:
SWITCHDEV_VXLAN_FDB_ADD_TO_DEVICE and SWITCHDEV_VXLAN_FDB_DEL_TO_DEVICE.
Introduce a new function, vxlan_fdb_switchdev_call_notifiers() to send
the new notifier types, and a struct switchdev_notifier_vxlan_fdb_info
to communicate the details of the FDB entry under consideration.
Invoke the new function from vxlan_fdb_notify().
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the ability to determine whether a netdev is a VxLAN netdev by
calling the above mentioned function that checks the netdev's
rtnl_link_ops.
This will allow modules to identify netdev events involving a VxLAN
netdev and act accordingly. For example, drivers capable of VxLAN
offload will need to configure the underlying device when a VxLAN netdev
is being enslaved to an offloaded bridge.
Convert nfp to use the newly introduced helper.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a local route that matches the source IP of an offloaded NVE tunnel
is notified, the driver needs to program it to perform NVE decapsulation
instead of merely trapping packets to the CPU.
This patch complements "mlxsw: spectrum_router: Enable local routes
promotion to perform NVE decap" where existing local routes were
promoted to perform NVE decapsulation.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
802.1D FIDs are used to represent VLAN-unaware bridges and currently
this is the only type of FID that supports NVE configuration.
Since the NVE tunnel device does not take a reference on the FID, it is
possible for the FID to be destroyed when it still has NVE
configuration.
Therefore, when destroying the FID make sure to disable its NVE
configuration.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The common NVE core expects each encapsulation type to implement a
certain set of operations that are specific to this type and the
currently used ASIC. These operations include things such as the ability
to determine whether a certain NVE configuration can be offloaded and
ASIC-specific initialization for this type.
Implement these operations for VxLAN on the Spectrum ASIC. Spectrum-2
support will be added by a future patchset.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The Spectrum ASIC supports different types of NVE encapsulations (e.g.,
VxLAN, NVGRE) with more types to be supported by future ASICs.
Despite being different, all these encapsulations share some common
functionality such as the enablement of NVE encapsulation on a given
filtering identifier (FID) and the addition of remote VTEPs to the
linked-list of VTEPs that traffic should be flooded to.
Implement this common core and allow different ASICs to register
different operations for different encapsulation types.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Drivers that support tunnel decapsulation (IPinIP or NVE) need to
configure the underlying device to conform to the behavior outlined in
RFC 6040 with respect to the ECN bits.
This behavior is implemented by INET_ECN_decapsulate() which requires an
skb to be passed where the ECN CE bit can be potentially set. Since
these drivers do not need to mark an skb, but only configure the device
to do so, factor out the business logic to __INET_ECN_decapsulate() and
potentially perform the marking in INET_ECN_decapsulate().
This allows drivers to invoke __INET_ECN_decapsulate() and configure the
device.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Suggested-by: Petr Machata <petrm@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Drivers that support VxLAN offload need to be able to sanitize the
configuration of the VxLAN device and accept / reject its offload.
For example, mlxsw requires that the local IP of the VxLAN device be set
and that packets be flooded to unicast IP(s) and not to a multicast
group.
Expose the functions that perform such checks.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the device, different VRFs (routing tables) are represented using
different virtual routers (VRs) and thus the kernel's table IDs are
mapped to VR IDs.
Allow internal users of the IP router to query the VR ID based on a
kernel table ID.
This is needed - for example - when configuring the underlay VR where
VxLAN encapsulated packets will undergo an L3 lookup. In this case, the
kernel's table ID is derived from the VxLAN device's configuration.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When an NVE tunnel with an IP underlay (e.g., VxLAN) is configured the
local route to the tunnel's source IP needs to be promoted to perform
NVE decapsulation.
Expose an API in the unicast IP router to promote / demote local routes.
The case where a local route is configured after the creation of the NVE
tunnel will be handled in a subsequent patch in the set.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Current APIs only allow looking for a FID and creating it in case it
does not exist.
With VxLAN, in case the bridge to which the VxLAN device was enslaved
does not already have a corresponding FID, then it means that something
went wrong that we need to be aware of.
Add an API to look up a FID, but without creating it in order to catch
above-mentioned situation.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the device, the VNI and the list of remote VTEPs a packet should be
flooded to is a property of the filtering identifier (FID).
During encapsulation, the VNI is taken from the FID the packet was
classified to. During decapsulation, the overlay packet is injected into
a bridge and classified to a FID based on the VNI it came with.
Allow NVE configuration for a FID. Currently, this is only supported
with 802.1D FIDs which are used for VLAN-unaware bridges. However, NVE
configuration is going to be supported with 802.1Q FIDs which is why the
related fields are placed in the common FID struct.
Since the device requires a 1:1 mapping between FID and VNI, the driver
maintains a hashtable keyed by VNI and checks if the VNI is already
associated with an existing FID.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently we fail when user specify a non-zero chain, this patch adds the
support for it and tc priorities. To get to a new chain, use the tc
goto action.
Currently we support a fixed prio range 1-16, and chain range 0-3.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
When adding a vxlan tc rule, and a neighbour isn't available, we
don't insert any rule to hardware. Once we enable offloading flows
with multiple priorities, a packet that should have matched this rule
will continue in hardware pipeline and might match a wrong one.
This is unlike in tc software path where it will be matched and
forwarded to the vxlan device (which will cause a ARP lookup
eventually) and stop processing further tc filters.
To address that, when when a neighbour isn't available (EAGAIN from
attach_encap), or gets deleted, change the original action to be a
forward to slow path instead. Neighbour update will restore the original
action once the neighbour becomes available. This will be done atomically
so at any given time we will have a the correct match.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
A pre-step for the tc offloads code to use this when a neigh is
not available for encap rules.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
The code for adding/deleting fdb flow is repeated when
user-space does flow add/del and when we add/del from
the neigh update path - unify them to avoid the duplication.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
When replacing a tc flower rule, flower first requests to add the
new rule (new action), then deletes the old one.
But currently when asked to add a new tc flower flow, we append the
actions (and counters to it).
This can result in a fte with two flow counters or conflicting
actions (drop and encap action) which firmware complains/errs
about and isn't achieving what the user aimed for.
Instead, insert the flow using the new no-append flag which will add a
new HW rule, the old flow and rule will be deleted later by flower
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanmox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
If no-append flag is set, we will add a new FTE, instead of appending
the actions of the inserted rule when the same match already exists.
While here, move the has_flow_tag boolean indicator to be a flag too.
This patch doesn't change any functionality.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanmox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
A chain is a group of priorities, so use the fdb parallel
sub namespaces to implement chains, and a flow table for each
priority in them.
Because these namespaces are parallel and in series to the slow path
fdb, the chains aren't connected to one another (but to the slow path),
and one must use a explicit goto action to reach a different chain.
Flow tables for the priorities will be created on demand and destroyed
once not used.
The Firmware has four pools of tables for sizes S/XS/M/L (4k, 64k, 1m, 4m).
We maintain ghost copies of the pools occupancy.
When a new table is to be created, we scan the pools from large to small
and find the 1st table size which can be now created. When a table is
destroyed, we update the relevant pool.
Multi chain/prio isn't enabled yet by this patch, for now all flows
will use the default chain 0, and prio 1.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Be symmetric with the e-switch API to add rules which has a
specific function to add fwd rules which are used as part of
vport mirroring.
This patch doesn't change any functionality.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Paul Blakey <paulb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Towards supporting multi-chains and priorities, split the FDB fast path
to multiple namespaces (sub namespaces), each with multiple priorities.
This patch adds a new flow steering type, FS_TYPE_PRIO_CHAINS, which is
like current FS_TYPE_PRIO, but may contain only namespaces, and those
will be in parallel to one another in terms of managing of the flow
tables connections inside them. Meaning, while searching for the next
or previous flow table to connect for a new table inside such namespace
we skip the parallel namespaces in the same level under the
FS_TYPE_PRIO_CHAINS prio we originated from.
We use this new type for splitting the fast path prio into multiple
parallel namespaces, each containing normal prios.
The prios inside them (and their tables) will be connected to one
another, but not from one parallel namespace to another, instead the
last prio in each namespace will be connected to the next prio in
the containing FDB namespace, which is the slow path prio.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Acked-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
If set, the firmware supports creating of flow tables with encap
enabled while VFs are configured, if we already created one
(restriction still applies on the first creation).
Signed-off-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Move to have clear separation on the code path to add nic vs e-switch
flows. While here we break the code that deals with adding offloaded
TC tool to few smaller stages, each on helper function.
Besides getting us simpler and readable code, these are pre-steps
for being able to have two HW flows serving one SW TC flow for some
e-switch use cases.
Signed-off-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Refactor the flow add utility functions to return err code instead of rule
pointers. This will allow for simpler logic when one tc rule is
duplicated to two HW rules in downstream patches.
Signed-off-by: Rabie Loulou <rabiel@mellanox.com>
Signed-off-by: Shahar Klein <shahark@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Currently, when a flow rule is created using the FS core layer, the caller
has to pass the entire flow counter object and not just the counter HW
handle (ID). This requires both the FS core and the caller to have
knowledge about the inner implementation of the FS layer flow counters
cache and limits the possible users.
Move to use the counter ID across the place when dealing with flows.
Doing this decoupling, now can we privatize the inner implementation
of the flow counters.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
There's no real reason for the e-switch logic to manage the creation of
counters for offloaded flows. The API already has the directive for the
caller to denote they want to attach a counter to the created flow.
As such, we go and move the management of flow counters to the mlx5e
tc offload logic. This also lets us remove an inelegant interface where
the FS layer had to provide a way to retrieve a counter from a flow rule.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
mlx5 updates for both net-next and rdma-next
* 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux: (21 commits)
net/mlx5: Expose DC scatter to CQE capability bit
net/mlx5: Update mlx5_ifc with DEVX UID bits
net/mlx5: Set uid as part of DCT commands
net/mlx5: Set uid as part of SRQ commands
net/mlx5: Set uid as part of SQ commands
net/mlx5: Set uid as part of RQ commands
net/mlx5: Set uid as part of QP commands
net/mlx5: Set uid as part of CQ commands
net/mlx5: Rename incorrect naming in IFC file
net/mlx5: Export packet reformat alloc/dealloc functions
net/mlx5: Pass a namespace for packet reformat ID allocation
net/mlx5: Expose new packet reformat capabilities
{net, RDMA}/mlx5: Rename encap to reformat packet
net/mlx5: Move header encap type to IFC header file
net/mlx5: Break encap/decap into two separated flow table creation flags
net/mlx5: Add support for more namespaces when allocating modify header
net/mlx5: Export modify header alloc/dealloc functions
net/mlx5: Add proper NIC TX steering flow tables support
net/mlx5: Cleanup flow namespace getter switch logic
net/mlx5: Add memic command opcode to command checker
...
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Add link establishment methods
Add auto negotiation methods
Add read MAC address method
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Add PHY's ID support
Add support for initialization, acquire and release of PHY
Enable register access
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Add code for NVM support and get MAC address, complete probe
method.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Add code for hardware initialization and reset
Add code for semaphore handling
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
This patch adds support for allocating, configuring, and freeing Tx/Rx ring
resources. With these changes in place the descriptor queues are in a
state where they are ready to transmit or receive if provided buffers.
This also adds the transmit and receive fastpath and interrupt handlers.
With this code in place the network device is now able to send and receive
frames over the network interface using a single queue.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
This change adds the defines and structures necessary to support both Tx
and Rx descriptor rings.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
This patch set adds interrupt support for the igc interfaces.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Now that we have the ability to configure the basic settings on the device
we can start allocating and configuring a netdev for the interface.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
This patch adds the basic defines and structures needed by the PF for
operation. With this it is possible to bring up the interface,
but without being able to configure any of the filters on
the interface itself.
Add skeleton for a function pointers.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
The preemptirq_delay_test module is used for the ftrace selftest code that
tests the latency tracers. The problem is that it uses ktime for the delay
loop, and then checks the tracer to see if the delay loop is caught, but the
tracer uses trace_clock_local() which uses various different other clocks to
measure the latency. As ktime uses the clock cycles, and the code then
converts that to nanoseconds, it causes rounding errors, and the preemptirq
latency tests are failing due to being off by 1 (it expects to see a delay
of 500000 us, but the delay is only 499999 us). This is happening due to a
rounding error in the ktime (which is totally legit). The purpose of the
test is to see if it can catch the delay, not to test the accuracy between
trace_clock_local() and ktime_get(). Best to use apples to apples, and have
the delay loop use the same clock as the latency tracer does.
Cc: stable@vger.kernel.org
Fixes: f96e8577da ("lib: Add module for testing preemptoff/irqsoff latency tracers")
Acked-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
commit 46e0c9be20 ("kernel: tracepoints: add support for relative
references") changes the layout of the __tracepoint_ptrs section on
architectures supporting relative references. However, it does so
without turning struct tracepoint * const into const int elsewhere in
the tracepoint code, which has the following side-effect:
Setting mod->num_tracepoints is done in by module.c:
mod->tracepoints_ptrs = section_objs(info, "__tracepoints_ptrs",
sizeof(*mod->tracepoints_ptrs),
&mod->num_tracepoints);
Basically, since sizeof(*mod->tracepoints_ptrs) is a pointer size
(rather than sizeof(int)), num_tracepoints is erroneously set to half the
size it should be on 64-bit arch. So a module with an odd number of
tracepoints misses the last tracepoint due to effect of integer
division.
So in the module going notifier:
for_each_tracepoint_range(mod->tracepoints_ptrs,
mod->tracepoints_ptrs + mod->num_tracepoints,
tp_module_going_check_quiescent, NULL);
the expression (mod->tracepoints_ptrs + mod->num_tracepoints) actually
evaluates to something within the bounds of the array, but miss the
last tracepoint if the number of tracepoints is odd on 64-bit arch.
Fix this by introducing a new typedef: tracepoint_ptr_t, which
is either "const int" on architectures that have PREL32 relocations,
or "struct tracepoint * const" on architectures that does not have
this feature.
Also provide a new tracepoint_ptr_defer() static inline to
encapsulate deferencing this type rather than duplicate code and
ugly idefs within the for_each_tracepoint_range() implementation.
This issue appears in 4.19-rc kernels, and should ideally be fixed
before the end of the rc cycle.
Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: Jessica Yu <jeyu@kernel.org>
Link: http://lkml.kernel.org/r/20181013191050.22389-1-mathieu.desnoyers@efficios.com
Link: http://lkml.kernel.org/r/20180704083651.24360-7-ard.biesheuvel@linaro.org
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morris <james.morris@microsoft.com>
Cc: James Morris <jmorris@namei.org>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Nicolas Pitre <nico@linaro.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Russell King <linux@armlinux.org.uk>
Cc: "Serge E. Hallyn" <serge@hallyn.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Thomas Garnier <thgarnie@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
This patch adds the beginning framework onto which I am going to add
the igc driver which supports the Intel(R) I225-LM/I225-V 2.5G
Ethernet Controller.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
num can be indirectly controlled by user-space, hence leading to
a potential exploitation of the Spectre variant 1 vulnerability.
This issue was detected with the help of Smatch:
drivers/usb/gadget/function/f_mass_storage.c:3177 fsg_lun_make() warn:
potential spectre issue 'fsg_opts->common->luns' [r] (local cap)
Fix this by sanitizing num before using it to index
fsg_opts->common->luns
Notice that given that speculation windows are large, the policy is
to kill the speculation on the first load and not worry if it can be
completed with a dependent load/store [1].
[1] https://marc.info/?l=linux-kernel&m=152449131114778&w=2
Cc: stable@vger.kernel.org
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Acked-by: Felipe Balbi <felipe.balbi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
David reports that:
<quote>
Perf has this hack where it uses the kernel symbol map as a backup when
a symbol can't be found in the user's symbol table(s).
This causes problems because the tests driving this code path use
machine__kernel_ip(), and that is completely meaningless on Sparc. On
sparc64 the kernel and user live in physically separate virtual address
spaces, rather than a shared one. And the kernel lives at a virtual
address that overlaps common userspace addresses. So this test passes
almost all the time when a user symbol lookup fails.
The consequence of this is that, if the unfound user virtual address in
the sample doesn't match up to a kernel symbol either, we trigger things
like this code in builtin-top.c:
if (al.sym == NULL && al.map != NULL) {
const char *msg = "Kernel samples will not be resolved.\n";
/*
* As we do lazy loading of symtabs we only will know if the
* specified vmlinux file is invalid when we actually have a
* hit in kernel space and then try to load it. So if we get
* here and there are _no_ symbols in the DSO backing the
* kernel map, bail out.
*
* We may never get here, for instance, if we use -K/
* --hide-kernel-symbols, even if the user specifies an
* invalid --vmlinux ;-)
*/
if (!machine->kptr_restrict_warned && !top->vmlinux_warned &&
__map__is_kernel(al.map) && map__has_symbols(al.map)) {
if (symbol_conf.vmlinux_name) {
char serr[256];
dso__strerror_load(al.map->dso, serr, sizeof(serr));
ui__warning("The %s file can't be used: %s\n%s",
symbol_conf.vmlinux_name, serr, msg);
} else {
ui__warning("A vmlinux file was not found.\n%s",
msg);
}
if (use_browser <= 0)
sleep(5);
top->vmlinux_warned = true;
}
}
When I fire up a compilation on sparc, this triggers immediately.
I'm trying to figure out what the "backup to kernel map" code is
accomplishing.
I see some language in the current code and in the changes that have
happened in this area talking about vdso. Does that really happen?
The vdso is mapped into userspace virtual addresses, not kernel ones.
More history. This didn't cause problems on sparc some time ago,
because the kernel IP check used to be "ip < 0" :-) Sparc kernel
addresses are not negative. But now with machine__kernel_ip(), which
works using the symbol table determined kernel address range, it does
trigger.
What it all boils down to is that on architectures like sparc,
machine__kernel_ip() should always return false in this scenerio, and
therefore this kind of logic:
if (cpumode == PERF_RECORD_MISC_USER && machine &&
mg != &machine->kmaps &&
machine__kernel_ip(machine, al->addr)) {
is basically invalid. PERF_RECORD_MISC_USER implies no kernel address
can possibly match for the sample/event in question (no matter how
hard you try!) :-)
</>
So, I thought something had changed and in the past we would somehow
find that address in the kallsyms, but I couldn't find anything to back
that up, the patch introducing this is over a decade old, lots of things
changed, so I was just thinking I was missing something.
I tried a gtod busy loop to generate vdso activity and added a 'perf
probe' at that branch, on x86_64 to see if it ever gets hit:
Made thread__find_map() noinline, as 'perf probe' in lines of inline
functions seems to not be working, only at function start. (Masami?)
# perf probe -x ~/bin/perf -L thread__find_map:57
<thread__find_map@/home/acme/git/perf/tools/perf/util/event.c:57>
57 if (cpumode == PERF_RECORD_MISC_USER && machine &&
58 mg != &machine->kmaps &&
59 machine__kernel_ip(machine, al->addr)) {
60 mg = &machine->kmaps;
61 load_map = true;
62 goto try_again;
}
} else {
/*
* Kernel maps might be changed when loading
* symbols so loading
* must be done prior to using kernel maps.
*/
69 if (load_map)
70 map__load(al->map);
71 al->addr = al->map->map_ip(al->map, al->addr);
# perf probe -x ~/bin/perf thread__find_map:60
Added new event:
probe_perf:thread__find_map (on thread__find_map:60 in /home/acme/bin/perf)
You can now use it in all perf tools, such as:
perf record -e probe_perf:thread__find_map -aR sleep 1
#
Then used this to see if, system wide, those probe points were being hit:
# perf trace -e *perf:thread*/max-stack=8/
^C[root@jouet ~]#
No hits when running 'perf top' and:
# cat gtod.c
#include <sys/time.h>
int main(void)
{
struct timeval tv;
while (1)
gettimeofday(&tv, 0);
return 0;
}
[root@jouet c]# ./gtod
^C
Pressed 'P' in 'perf top' and the [vdso] samples are there:
62.84% [vdso] [.] __vdso_gettimeofday
8.13% gtod [.] main
7.51% [vdso] [.] 0x0000000000000914
5.78% [vdso] [.] 0x0000000000000917
5.43% gtod [.] _init
2.71% [vdso] [.] 0x000000000000092d
0.35% [kernel] [k] native_io_delay
0.33% libc-2.26.so [.] __memmove_avx_unaligned_erms
0.20% [vdso] [.] 0x000000000000091d
0.17% [i2c_i801] [k] i801_access
0.06% firefox [.] free
0.06% libglib-2.0.so.0.5400.3 [.] g_source_iter_next
0.05% [vdso] [.] 0x0000000000000919
0.05% libpthread-2.26.so [.] __pthread_mutex_lock
0.05% libpixman-1.so.0.34.0 [.] 0x000000000006d3a7
0.04% [kernel] [k] entry_SYSCALL_64_trampoline
0.04% libxul.so [.] style::dom_apis::query_selector_slow
0.04% [kernel] [k] module_get_kallsym
0.04% firefox [.] malloc
0.04% [vdso] [.] 0x0000000000000910
I added a 'perf probe' to thread__find_map:69, and that surely got tons
of hits, i.e. for every map found, just to make sure the 'perf probe'
command was really working.
In the process I noticed a bug, we're only have records for '[vdso]' for
pre-existing commands, i.e. ones that are running when we start 'perf top',
when we will generate the PERF_RECORD_MMAP by looking at /perf/PID/maps.
I.e. like this, for preexisting processes with a vdso map, again,
tracing for all the system, only pre-existing processes get a [vdso] map
(when having one):
[root@jouet ~]# perf probe -x ~/bin/perf __machine__addnew_vdso
Added new event:
probe_perf:__machine__addnew_vdso (on __machine__addnew_vdso in /home/acme/bin/perf)
You can now use it in all perf tools, such as:
perf record -e probe_perf:__machine__addnew_vdso -aR sleep 1
[root@jouet ~]# perf trace -e probe_perf:__machine__addnew_vdso/max-stack=8/
0.000 probe_perf:__machine__addnew_vdso:(568eb3)
__machine__addnew_vdso (/home/acme/bin/perf)
map__new (/home/acme/bin/perf)
machine__process_mmap2_event (/home/acme/bin/perf)
machine__process_event (/home/acme/bin/perf)
perf_event__process (/home/acme/bin/perf)
perf_tool__process_synth_event (/home/acme/bin/perf)
perf_event__synthesize_mmap_events (/home/acme/bin/perf)
__event__synthesize_thread (/home/acme/bin/perf)
The kernel is generating a PERF_RECORD_MMAP for vDSOs, but somehow
'perf top' is not getting those records while 'perf record' is:
# perf record ~acme/c/gtod
^C[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.076 MB perf.data (1499 samples) ]
# perf report -D | grep PERF_RECORD_MMAP2
71293612401913 0x11b48 [0x70]: PERF_RECORD_MMAP2 25484/25484: [0x400000(0x1000) @ 0 fd:02 1137 541179306]: r-xp /home/acme/c/gtod
71293612419012 0x11be0 [0x70]: PERF_RECORD_MMAP2 25484/25484: [0x7fa4a2783000(0x227000) @ 0 fd:00 3146370 854107250]: r-xp /usr/lib64/ld-2.26.so
71293612432110 0x11c50 [0x60]: PERF_RECORD_MMAP2 25484/25484: [0x7ffcdb53a000(0x2000) @ 0 00:00 0 0]: r-xp [vdso]
71293612509944 0x11cb0 [0x70]: PERF_RECORD_MMAP2 25484/25484: [0x7fa4a23cd000(0x3b6000) @ 0 fd:00 3149723 262067164]: r-xp /usr/lib64/libc-2.26.so
#
# perf script | grep vdso | head
gtod 25484 71293.612768: 2485554 cycles:ppp: 7ffcdb53a914 [unknown] ([vdso])
gtod 25484 71293.613576: 2149343 cycles:ppp: 7ffcdb53a917 [unknown] ([vdso])
gtod 25484 71293.614274: 1814652 cycles:ppp: 7ffcdb53aca8 __vdso_gettimeofday+0x98 ([vdso])
gtod 25484 71293.614862: 1669070 cycles:ppp: 7ffcdb53acc5 __vdso_gettimeofday+0xb5 ([vdso])
gtod 25484 71293.615404: 1451589 cycles:ppp: 7ffcdb53acc5 __vdso_gettimeofday+0xb5 ([vdso])
gtod 25484 71293.615999: 1269941 cycles:ppp: 7ffcdb53ace6 __vdso_gettimeofday+0xd6 ([vdso])
gtod 25484 71293.616405: 1177946 cycles:ppp: 7ffcdb53a914 [unknown] ([vdso])
gtod 25484 71293.616775: 1121290 cycles:ppp: 7ffcdb53ac47 __vdso_gettimeofday+0x37 ([vdso])
gtod 25484 71293.617150: 1037721 cycles:ppp: 7ffcdb53ace6 __vdso_gettimeofday+0xd6 ([vdso])
gtod 25484 71293.617478: 994526 cycles:ppp: 7ffcdb53ace6 __vdso_gettimeofday+0xd6 ([vdso])
#
The patch is the obvious one and with it we also continue to resolve
vdso symbols for pre-existing processes in 'perf top' and for all
processes in 'perf record' + 'perf report/script'.
Suggested-by: David Miller <davem@davemloft.net>
Acked-by: David Miller <davem@davemloft.net>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-cs7skq9pp0kjypiju6o7trse@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Helge writes:
"parisc fix:
Fix an unitialized variable usage in the parisc unwind code."
* 'parisc-4.19-3' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
parisc: Fix uninitialized variable usage in unwind.c
Ido Schimmel