xref: /illumos-gate/usr/src/man/man9e/mac.9e (revision 72d3dbb9ab4481606cb93caca98ba3b3a8eb6ce2)

.\"
.\" This file and its contents are supplied under the terms of the
.\" Common Development and Distribution License ("CDDL"), version 1.0.
.\" You may only use this file in accordance with the terms of version
.\" 1.0 of the CDDL.
.\"
.\" A full copy of the text of the CDDL should have accompanied this
.\" source.  A copy of the CDDL is also available via the Internet at
.\" http://www.illumos.org/license/CDDL.
.\"
.\"
.\" Copyright 2016 Joyent, Inc.
.\"
.Dd March 26, 2017
.Dt MAC 9E
.Os
.Sh NAME
.Nm mac ,
.Nm GLDv3
.Nd MAC networking device driver overview
.Sh SYNOPSIS
.In sys/mac_provider.h
.In sys/mac_ether.h
.Sh INTERFACE LEVEL
illumos DDI specific
.Sh DESCRIPTION
The
.Sy MAC
framework provides a means for implementing high-performance networking
device drivers.
It is the successor to the GLD interfaces and is sometimes referred to as the
GLDv3.
The remainder of this manual introduces the aspects of writing devices drivers
that leverage the MAC framework.
While both the GLDv3 and MAC framework refer to the same thing, in this manual
page we use the term the
.Em MAC framework
to refer to the device driver interface.
.Pp
MAC device drivers are character devices.
They define the standard
.Xr _init 9E ,
.Xr _fini 9E ,
and
.Xr _info 9E
entry points to initialize the module, as well as
.Xr dev_ops 9S
and
.Xr cb_ops 9S
structures.
.Pp
The main interface with MAC is through a series of callbacks defined in
a
.Xr mac_callbacks 9S
structure.
These callbacks control all the aspects of the device.
They range from sending data, getting and setting of properties, controlling mac
address filters, and also managing promiscuous mode.
.Pp
The MAC framework takes care of many aspects of the device driver's
management.
A device that uses the MAC framework does not have to worry about creating
device nodes or implementing
.Xr open 9E
or
.Xr close 9E
routines.
In addition, all of the work to interact with
.Xr dlpi 7P
is taken care of automatically and transparently.
.Ss Initializing MAC Support
For a device to be used in the framework, it must register with the
framework and take specific actions during
.Xr _init 9E ,
.Xr attach 9E ,
.Xr detach 9E ,
and
.Xr _fini 9E .
.Pp
All device drivers have to define a
.Xr dev_ops 9S
structure which is pointed to by a
.Xr modldrv 9S
structure and the corresponding NULL-terminated
.Xr modlinkage 9S
structure.
The
.Xr dev_ops 9S
structure should have a
.Xr cb_ops 9S
structure defined for it; however, it does not need to implement any of
the standard
.Xr cb_ops 9S
entry points.
.Pp
Normally, in a driver's
.Xr _init 9E
entry point, it passes its
.Sy modlinkage
structure directly to
.Xr mod_install 9F .
To properly register with MAC, the driver must call
.Xr mac_init_ops 9F
before it calls
.Xr mod_install 9F .
If for some reason the
.Xr mod_install 9F
function fails, then the driver must be removed by a call to
.Xr mac_fini_ops 9F .
.Pp
Conversely, in the driver's
.Xr _fini 9F
routine, it should call
.Xr mac_fini_ops 9F
after it successfully calls
.Xr mod_remove 9F .
For an example of how to use the
.Xr mac_init_ops 9F
and
.Xr mac_fini_ops 9F
functions, see the examples section in
.Xr mac_init_ops 9F .
.Ss Registering with MAC
Every instance of a device should register separately with MAC.
To register with MAC, a driver must allocate a
.Xr mac_register 9S
structure, fill it in, and then call
.Xr mac_register 9F .
The
.Sy mac_register_t
structure contains information about the device and all of the required
function pointers that will be used as callbacks by the framework.
.Pp
These steps should all be taken during a device's
.Xr attach 9E
entry point.
It is recommended that the driver perform this sequence of steps after the
device has finished its initialization of the chipset and interrupts, though
interrupts should not be enabled at that point.
After it calls
.Xr mac_register 9F
it will start receiving callbacks from the MAC framework.
.Pp
To allocate the registration structure, the driver should call
.Xr mac_alloc 9F .
Device drivers should generally always pass the symbol
.Sy MAC_VERSION
as the argument to
.Xr mac_alloc 9F .
Upon successful completion, the driver will receive a
.Sy mac_register_t
structure which it should fill in.
The structure and its members are documented in
.Xr mac_register 9S .
.Pp
The
.Xr mac_callbacks 9S
structure is not allocated as a part of the
.Xr mac_register 9S
structure.
In general, device drivers declare this statically.
See the
.Sx MAC Callbacks
section for more information on how to fill it out.
.Pp
Once the structure has been filled in, the driver should call
.Xr mac_register 9F
to register itself with MAC.
The handle that it uses to register with should be part of the driver's soft
state.
It will be used in various other support functions and callbacks.
.Pp
If the call is successful, then the device driver
should enable interrupts and finish any other initialization required.
If the call to
.Xr mac_register 9F
failed, then it should unwind its initialization and should return
.Sy DDI_FAILURE
from its
.Xr attach 9E
routine.
.Ss MAC Callbacks
The MAC framework interacts with a device driver through a series of
callbacks.
These callbacks are described in their individual manual pages and the
collection of callbacks is indicated in the
.Xr mac_callbacks 9S
manual page.
This section does not focus on the specific functions, but rather on
interactions between them and the rest of the device driver framework.
.Pp
A device driver should make no assumptions about when the various
callbacks will be called and whether or not they will be called
simultaneously.
For example, a device driver may be asked to transmit data through a call to its
.Xr mc_tx 9F
entry point while it is being asked to get a device property through a
call to its
.Xr mc_getprop 9F
entry point.
As such, while some calls may be serialized to the device, such as setting
properties, the device driver should always presume that all of its data needs
to be protected with locks.
While the device is holding locks, it is safe for it call the following MAC
routines:
.Bl -bullet -offset indent -compact
.It
.Xr mac_hcksum_get 9F
.It
.Xr mac_hcksum_set 9F
.It
.Xr mac_lso_get 9F
.It
.Xr mac_maxsdu_update 9F
.It
.Xr mac_prop_info_set_default_link_flowctrl 9F
.It
.Xr mac_prop_info_set_default_str 9F
.It
.Xr mac_prop_info_set_default_uint8 9F
.It
.Xr mac_prop_info_set_default_uint32 9F
.It
.Xr mac_prop_info_set_default_uint64 9F
.It
.Xr mac_prop_info_set_perm 9F
.It
.Xr mac_prop_info_set_range_uint32 9F
.El
.Pp
Any other MAC related routines should not be called with locks held,
such as
.Xr mac_link_update 9F
or
.Xr mac_rx 9F .
Other routines in the DDI may be called while locks are held; however,
device driver writers should be careful about calling blocking routines
while locks are held or in interrupt context, though it is generally
legal to do so.
.Ss Receiving Data
A device driver will often receive data through the means of an
interrupt.
When that interrupt occurs, the device driver will receive one or more frames
with optional metadata.
Often each frame has a corresponding descriptor which has information about
whether or not there were errors or whether or not the device successfully
checksummed the packet.
.Pp
During a single interrupt, a device driver should process a fixed number
of frames.
For each frame the device driver should:
.Bl -enum -offset indent
.It
First check whether or not the frame has errors.
If errors were detected, then the frame should not be sent to the operating
system.
It is recommended that devices keep kstats (see
.Xr kstat_create 9S
for more information) and bump the counter whenever such an error is
detected.
If the device distinguishes between the types of errors, then separate kstats
for each class of error are recommended.
See the
.Sx STATISTICS
section for more information on the various error cases that should be
considered.
.It
Once the frame has been determined to be valid, the device driver should
transform the frame into a
.Xr mblk 9S .
See the section
.Sx MBLKS AND DMA
for more information on how to transform and prepare a message block.
.It
If the device supports hardware checksumming (see the
.Sx CAPABILITIES
section for more information on checksumming), then the device driver
should set the corresponding checksumming information with a call to
.Xr mac_hcksum_set 9F .
.It
It should then append this new message block to the
.Em end
of the message block chain, linking it to the
.Sy b_next
pointer.
It is vitally important that all the frames be chained in the order that they
were received.
If the device driver mistakenly reorders frames, then it may cause performance
impacts in the TCP stack and potentially impact application correctness.
.El
.Pp
Once all the frames have been processed and assembled, the device driver
should deliver them to the rest of the operating system by calling
.Xr mac_rx 9F .
The device driver should try to give as many mblk_t structures to the
system at once.
It
.Em should not
call
.Xr mac_rx 9F
once for every assembled mblk_t.
.Pp
The device driver must not hold any locks across the call to
.Xr mac_rx 9F .
When this function is called, received data will be pushed through the
networking stack and some replies may be generated and given to the
driver to send out.
.Pp
It is not the device driver's responsibility to determine whether or not
the system can keep up with a driver's delivery rate of frames.
The rest of the networking stack will handle issues related to keeping up
appropriately and ensure that kernel memory is not exhausted by packets
that are not being processed.
.Pp
Finally, the device driver should make sure that any other housekeeping
activities required for the ring are taken care of such that more data
can be received.
.Ss Transmitting Data and Back Pressure
A device driver will be asked to transmit a message block chain by
having it's
.Xr mc_tx 9E
entry point called.
While the driver is processing the message blocks, it may run out of resources.
For example, a transmit descriptor ring may become full.
At that point, the device driver should return the remaining unprocessed frames.
The act of returning frames indicates that the device has asserted flow control.
Once this has been done, no additional calls will be made to the
driver's transmit entry point and the back pressure will be propagated
throughout the rest of the networking stack.
.Pp
At some point in the future when resources have become available again,
for example after an interrupt indicating that some portion of the
transmit ring has been sent, then the device driver must notify the
system that it can continue transmission.
To do this, the driver should call
.Xr mac_tx_update 9F .
After that point, the driver will receive calls to its
.Xr mc_tx 9E
entry point again.
As mentioned in the section on callbacks, the device driver should avoid holding
any particular locks across the call to
.Xr mac_tx_update 9F .
.Ss Interrupt Coalescing
For devices operating at higher data rates, interrupt coalescing is an
important part of a well functioning device and may impact the
performance of the device.
Not all devices support interrupt coalescing.
If interrupt coalescing is supported on the device, it is recommended that
device driver writers provide private properties for their device to control the
interrupt coalescing rate.
This will make it much easier to perform experiments and observe the impact of
different interrupt rates on the rest of the system.
.Ss MAC Address Filter Management
The MAC framework will attempt to use as many MAC address filters as a
device has.
To program a multicast address filter, the driver's
.Xr mc_multicst 9E
entry point will be called.
If the device driver runs out of filters, it should not take any special action
and just return the appropriate error as documented in the corresponding manual
pages for the entry points.
The framework will ensure that the device is placed in promiscuous mode
if it needs to.
.Ss Link Updates
It is the responsibility of the device driver to keep track of the
data link's state.
Many devices provide a means of receiving an interrupt when the state of the
link changes.
When such a change happens, the driver should update its internal data
structures and then call
.Xr mac_link_update 9F
to inform the MAC layer that this has occurred.
If the device driver does not properly inform the system about link changes,
then various features like link aggregations and other mechanisms that leverage
the link state will not work correctly.
.Ss Link Speed and Auto-negotiation
Many networking devices support more than one possible speed that they
can operate at.
The selection of a speed is often performed through
.Em auto-negotiation ,
though some devices allow the user to control what speeds are advertised
and used.
.Pp
Logically, there are two different sets of things that the device driver
needs to keep track of while it's operating:
.Bl -enum
.It
The supported speeds in hardware.
.It
The enabled speeds from the user.
.El
.Pp
By default, when a link first comes up, the device driver should
generally configure the link to support the common set of speeds and
perform auto-negotiation.
.Pp
A user can control what speeds a device advertises via auto-negotiation
and whether or not it performs auto-negotiation at all by using a series
of properties that have
.Sy _EN_
in the name.
These are read/write properties and there is one for each speed supported in the
operating system.
For a full list of them, see the
.Sx PROPERTIES
section.
.Pp
In addition to these properties, there is a corresponding set of
properties with
.Sy _ADV_
in the name.
These are similar to the
.Sy _EN_
family of properties, but they are read-only and indicate what the
device has actually negotiated.
While they are generally similar to the
.Sy _EN_
family of properties, they may change depending on power settings.
See the
.Sy Ethernet Link Properties
section in
.Xr dladm 1M
for more information.
.Pp
It's worth discussing how these different values get used throughout the
different entry points.
The first entry point to consider is the
.Xr mc_propinfo 9E
entry point.
For a given speed, the driver should consult whether or not the hardware
supports this speed.
If it does, it should fill in the default value that the hardware takes and
whether or not the property is writable.
The properties should also be updated to indicate whether or not it is writable.
This holds for both the
.Sy _EN_
and
.Sy _ADV_
family of properties.
.Pp
The next entry point is
.Xr mc_getprop 9E .
Here, the device should first consult whether the given speed is
supported.
If it is not, then the driver should return
.Er ENOTSUP .
If it does, then it should return the current value of the property.
.Pp
The last property endpoint is the
.Xr mc_setprop 9E
entry point.
Here, the same logic applies.
Before the driver considers whether or not the property is writable, it should
first check whether or not it's a supported property.
If it's not, then it should return
.Er ENOTSUP .
Otherwise, it should proceed to check whether the property is writable,
and if it is and a valid value, then it should update the property and
restart the link's negotiation.
.Pp
Finally, there is the
.Xr mc_getstat 9E
entry point.
Several of the statistics that are queried relate to auto-negotiation and
hardware capabilities.
When a statistic relates to the hardware supporting a given speed, the
.Sy _EN_
properties should be ignored.
The only thing that should be consulted is what the hardware itself supports.
Otherwise, the statistics should look at what is currently being advertised by
the device.
.Ss Unregistering from MAC
During a driver's
.Xr detach 9E
routine, it should unregister the device instance from MAC by calling
.Xr mac_unregister 9F
on the handle that it originally called it on.
If the call to
.Xr mac_unregister 9F
failed, then the device is likely still in use and the driver should
fail the call to
.Xr detach 9E .
.Ss Interacting with Devices
Administrators always interact with devices through the
.Xr dladm 1M
command line interface.
The state of devices such as whether the link is considered
.Sy up
or
.Sy down ,
various link properties such as the
.Sy MTU ,
.Sy auto-negotiation
state,
and
.Sy flow control
state,
are all exposed.
It is also the preferred way that these properties are set and configured.
.Pp
While device tunables may be presented in a
.Xr driver.conf 4
file, it is recommended instead to expose such things through
.Xr dladm 1M
private properties, whether explicitly documented or not.
.Sh CAPABILITIES
Capabilities in the MAC Framework are optional features that a device
supports which indicate various hardware features that the device
supports.
The two current capabilities that the system supports are related to being able
to hardware perform large send offloads (LSO), often also known as TCP
segmentation and the ability for hardware to calculate and verify the checksums
present in IPv4, IPV6, and protocol headers such as TCP and UDP.
.Pp
The MAC framework will query a device for support of a capability
through the
.Xr mc_getcapab 9E
function.
Each capability has its own constant and may have corresponding data that goes
along with it and a specific structure that the device is required to fill in.
Note, the set of capabilities changes over time and there are also private
capabilities in the system.
Several of the capabilities are used in the implementation of the MAC framework.
Others, like
.Sy MAC_CAPAB_RINGS ,
represent feature that have not been stabilized and thus both API and binary
compatibility for them is not guaranteed.
It is important that the device driver handles unknown capabilities correctly.
For more information, see
.Xr mc_getcapab 9E .
.Pp
The following capabilities are
stable and defined in the system:
.Ss MAC_CAPAB_HCKSUM
The
.Sy MAC_CAPAB_HCKSUM
capability indicates to the system that the device driver supports some
amount of checksumming.
The specific data for this capability is a pointer to a
.Sy uint32_t .
To indicate no support for any kind of checksumming, the driver should
either set this value to zero or simply return that it doesn't support
the capability.
.Pp
Note, the values that the driver declares in this capability indicate
what it can do when it transmits data.
If the driver can only verify checksums when receiving data, then it should not
indicate that it supports this capability.
The following set of flags may be combined through a bitwise inclusive OR:
.Bl -tag -width Ds
.It Sy HCKSUM_INET_PARTIAL
This indicates that the hardware can calculate a partial checksum for
both IPv4 and IPv6; however, it requires the pseudo-header checksum be
calculated for it.
The pseudo-header checksum will be available for the mblk_t when calling
.Xr mac_hcksum_get 9F .
Note this does not imply that the hardware is capable of calculating the
IPv4 header checksum.
That should be indicated with the
.Sy HCKSUM_IPHDRCKSUM flag.
.It Sy HCKSUM_INET_FULL_V4
This indicates that the hardware will fully calculate the L4 checksum
for outgoing IPv4 packets and does not require a pseudo-header checksum.
Note this does not imply that the hardware is capable of calculating the
IPv4 header checksum.
That should be indicated with the
.Sy HCKSUM_IPHDRCKSUM .
.It Sy HCKSUM_INET_FULL_V6
This indicates that the hardware will fully calculate the L4 checksum
for outgoing IPv6 packets and does not require a pseudo-header checksum.
.It Sy HCKSUM_IPHDRCKSUM
This indicates that the hardware supports calculating the checksum for
the IPv4 header itself.
.El
.Pp
When in a driver's transmit function, the driver will be processing a
single frame.
It should call
.Xr mac_hcksum_get 9F
to see what checksum flags are set on it.
Note that the flags that are set on it are different from the ones described
above and are documented in its manual page.
These flags indicate how the driver is expected to program the hardware and what
checksumming is required.
Not all frames will require hardware checksumming or will ask the hardware to
checksum it.
.Pp
If a driver supports offloading the receive checksum and verification,
it should check to see what the hardware indicated was verified.
The driver should then call
.Xr mac_hcksum_set 9F .
The flags used are different from the ones above and are discussed in
detail in the
.Xr mac_hcksum_set 9F
manual page.
If there is no checksum information available or the driver does not support
checksumming, then it should simply not call
.Xr mac_hcksum_set 9F .
.Pp
Note that the checksum flags should be set on the first
mblk_t that makes up a given message.
In other words, if multiple mblk_t structures are linked together by the
.Sy b_cont
member to describe a single frame, then it should only be called on the
first mblk_t of that set.
However, each distinct message should have the checksum bits set on it, if
applicable.
In other words, each mblk_t that is linked together by the
.Sy b_next
pointer may have checksum flags set.
.Pp
It is recommended that device drivers provide a private property or
.Xr driver.conf 4
property to control whether or not checksumming is enabled for both rx
and tx; however, the default disposition is recommended to be enabled
for both.
This way if hardware bugs are found in the checksumming implementation, they can
be disabled without requiring software updates.
The transmit property should be checked when determining how to reply to
.Xr mc_getcapab 9E
and the receive property should be checked in the context of the receive
function.
.Ss MAC_CAPAB_LSO
The
.Sy MAC_CAPAB_LSO
capability indicates that the driver supports various forms of large
send offload (LSO).
The private data is a pointer to a
.Sy mac_capab_lso_t
structure.
At the moment, LSO support is limited to TCP inside of IPv4.
This structure has the following members which are used to indicate
various types of LSO support.
.Bd -literal -offset indent
t_uscalar_t		lso_flags;
lso_basic_tcp_ivr4_t	lso_basic_tcp_ipv4;
.Ed
.Pp
The
.Sy lso_flags
member is used to indicate which members are valid and should be
considered.
Each flag represents a different form of LSO.
The member should be set to the bitwise inclusive OR of the following values:
.Bl -tag -width Dv -offset indent
.It Sy LSO_TX_BASIC_TCP_IPV4
This indicates hardware support for performing TCP segmentation
offloading over IPv4.
When this flag is set, the
.Sy lso_basic_tcp_ipv4
member must be filled in.
.El
.Pp
The
.Sy lso_basic_tcp_ipv4
member is a structure with the following members:
.Bd -literal -offset indent
t_uscalar_t	lso_max
.Ed
.Bd -filled -offset indent
The
.Sy lso_max
member should be set to the maximum size of the TCP data
payload that can be offloaded to the hardware.
.Ed
.Pp
Like with checksumming, it is recommended that driver writers provide a
means for disabling the support of LSO even if it is enabled by default.
This deals with the case where issues that pop up for LSO may be worked
around without requiring additional driver work.
.Sh PROPERTIES
Properties in the MAC framework represent aspects of a link.
These include things like the link's current state and MTU.
Many of the properties in the system are focused around auto-negotiation and
controlling what link speeds are advertised.
Information about properties is covered by three different device entry points.
The
.Xr mc_propinfo 9E
entry point obtains metadata about the property.
The
.Xr mc_getprop 9E
entry point obtains the property.
The
.Xr mc_setprop 9E
entry point updates the property to a new value.
.Pp
Many of the properties listed below are read-only.
Each property indicates whether it's read-only or it's read/write.
However, driver writers may not implement the ability to set all writable
properties.
Many of these depend on the card itself.
In particular, all properties that relate to auto-negotiation and are read/write
may not be updated if the hardware in question does not support toggling what
link speeds are auto-negotiated.
While copper Ethernet often does not have this restriction, it often exists with
various fiber standards and phys.
.Pp
The following properties are the subset of MAC framework properties that
driver writers should be aware of and handle.
While other properties exist in the system, driver writers should always return
an error when a property not listed below is encountered.
See
.Xr mc_getprop 9E
and
.Xr mc_setprop 9E
for more information on how to handle them.
.Bl -hang -width Ds
.It Sy MAC_PROP_DUPLEX
.Bd -filled -compact
Type:
.Sy link_duplex_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_DUPLEX
property is used to indicate whether or not the link is duplex.
A duplex link may have traffic flowing in both directions at the same time.
The
.Sy link_duplex_t
is an enumeration which may be set to any of the following values:
.Bl -tag -width Ds
.It Sy LINK_DUPLEX_UNKNOWN
The current state of the link is unknown.
This may be because the link has not negotiated to a specific speed or it is
down.
.It Sy LINK_DUPLEX_HALF
The link is running at half duplex.
Communication may travel in only one direction on the link at a given time.
.It Sy LINK_DUPLEX_FULL
The link is running at full duplex.
Communication may travel in both directions on the link simultaneously.
.El
.It Sy MAC_PROP_SPEED
.Bd -filled -compact
Type:
.Sy uint64_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_SPEED
property stores the current link speed in bits per second.
A link that is running at 100 MBit/s would store the value 100000000ULL.
A link that is running at 40 Gbit/s would store the value 40000000000ULL.
.It Sy MAC_PROP_STATUS
.Bd -filled -compact
Type:
.Sy link_state_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_STATUS
property is used to indicate the current state of the link.
It indicates whether the link is up or down.
The
.Sy link_state_t
is an enumeration which may be set to any of the following values:
.Bl -tag -width Ds
.It Sy LINK_STATE_UNKNOWN
The current state of the link is unknown.
This may be because the driver's
.Xr mc_start 9E
endpoint has not been called so it has not attempted to start the link.
.It Sy LINK_STATE_DOWN
The link is down.
This may be because of a negotiation problem, a cable problem, or some other
device specific issue.
.It Sy LINK_STATE_UP
The link is up.
If auto-negotiation is in use, it should have completed.
Traffic should be able to flow over the link, barring other issues.
.El
.It Sy MAC_PROP_AUTONEG
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_AUTONEG
property indicates whether or not the device is currently configured to
perform auto-negotiation.
A value of
.Sy 0
indicates that auto-negotiation is disabled.
A
.Sy non-zero
value indicates that auto-negotiation is enabled.
Devices should generally default to enabling auto-negotiation.
.Pp
When getting this property, the device driver should return the current
state.
When setting this property, if the device supports operating in the requested
mode, then the device driver should reset the link to negotiate to the new speed
after updating any internal registers.
.It Sy MAC_PROP_MTU
.Bd -filled -compact
Type:
.Sy uint32_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_MTU
property determines the maximum transmission unit (MTU).
This indicates the maximum size packet that the device can transmit, ignoring
its own headers.
For an Ethernet device, this would exclude the size of the Ethernet header and
any VLAN headers that would be placed.
It is up to the driver to ensure that any MTU values that it accepts when adding
in its margin and header sizes does not exceed its maximum frame size.
.Pp
By default, drivers for Ethernet should initialize this value and the
MTU to
.Sy 1500 .
When getting this property, the driver should return its current
recorded MTU.
When setting this property, the driver should first validate that it is within
the device's valid range and then it must call
.Xr mac_maxsdu_update 9F .
Note that the call may fail.
If the call completes successfully, the driver should update the hardware with
the new value of the MTU and perform any other work needed to handle it.
.Pp
If the device does not support changing the MTU after the device's
.Xr mc_start 9E
entry point has been called, then driver writers should return
.Er EBUSY .
.It Sy MAC_PROP_FLOWCTRL
.Bd -filled -compact
Type:
.Sy link_flowctrl_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_FLOWCTRL
property manages the configuration of pause frames as part of Ethernet
flow control.
Note, this only describes what this device will advertise.
What is actually enabled may be different and is subject to the rules of
auto-negotiation.
The
.Sy link_flowctrl_t
is an enumeration that may be set to one of the following values:
.Bl -tag -width Ds
.It Sy LINK_FLOWCTRL_NONE
Flow control is disabled.
No pause frames should be generated or honored.
.It Sy LINK_FLOWCTRL_RX
The device can receive pause frames; however, it should not generate
them.
.It Sy LINK_FLOWCTRL_TX
The device can generate pause frames; however, it does not support
receiving them.
.It Sy LINK_FLOWCTRL_BI
The device supports both sending and receiving pause frames.
.El
.Pp
When getting this property, the device driver should return the way that
it has configured the device, not what the device has actually
negotiated.
When setting the property, it should update the hardware and allow the link to
potentially perform auto-negotiation again.
.El
.Pp
The remaining properties are all about various auto-negotiation link
speeds.
They fall into two different buckets: properties with
.Sy _ADV_
in the name and properties with
.Sy _EN_
in the name.
For any given supported speed, there is one of each.
The
.Sy _EN_
set of properties are read/write properties that control what should be
advertised by the device.
When these are retrieved, they should return the current value of the property.
When they are set, they should change how the hardware advertises the specific
speed and trigger any kind of link reset and auto-negotiation, if enabled, to
occur.
.Pp
The
.Sy _ADV_
set of properties are read-only properties.
They are meant to reflect what has actually been negotiated.
These may be different from the
.Sy _EN_
family of properties, especially when different power management
settings are at play.
.Pp
See the
.Sx Link Speed and Auto-negotiation
section for more information.
.Pp
The properties are ordered in increasing link speed:
.Bl -hang -width Ds
.It Sy MAC_PROP_ADV_10HDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_10HDX_CAP
property describes whether or not 10 Mbit/s half-duplex support is
advertised.
.It Sy MAC_PROP_EN_10HDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_EN_10HDX_CAP
property describes whether or not 10 Mbit/s half-duplex support is
enabled.
.It Sy MAC_PROP_ADV_10FDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_10FDX_CAP
property describes whether or not 10 Mbit/s full-duplex support is
advertised.
.It Sy MAC_PROP_EN_10FDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_EN_10FDX_CAP
property describes whether or not 10 Mbit/s full-duplex support is
enabled.
.It Sy MAC_PROP_ADV_100HDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_100HDX_CAP
property describes whether or not 100 Mbit/s half-duplex support is
advertised.
.It Sy MAC_PROP_EN_100HDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_EN_100HDX_CAP
property describes whether or not 100 Mbit/s half-duplex support is
enabled.
.It Sy MAC_PROP_ADV_100FDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_100FDX_CAP
property describes whether or not 100 Mbit/s full-duplex support is
advertised.
.It Sy MAC_PROP_EN_100FDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_EN_100FDX_CAP
property describes whether or not 100 Mbit/s full-duplex support is
enabled.
.It Sy MAC_PROP_ADV_100T4_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_100T4_CAP
property describes whether or not 100 Mbit/s Ethernet using the
100BASE-T4 standard is
advertised.
.It Sy MAC_PROP_EN_100T4_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_ADV_100T4_CAP
property describes whether or not 100 Mbit/s Ethernet using the
100BASE-T4 standard is
enabled.
.It Sy MAC_PROP_ADV_1000HDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_1000HDX_CAP
property describes whether or not 1 Gbit/s half-duplex support is
advertised.
.It Sy MAC_PROP_EN_1000HDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_EN_1000HDX_CAP
property describes whether or not 1 Gbit/s half-duplex support is
enabled.
.It Sy MAC_PROP_ADV_1000FDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_1000FDX_CAP
property describes whether or not 1 Gbit/s full-duplex support is
advertised.
.It Sy MAC_PROP_EN_1000FDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_EN_1000FDX_CAP
property describes whether or not 1 Gbit/s full-duplex support is
enabled.
.It Sy MAC_PROP_ADV_2500FDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_2500FDX_CAP
property describes whether or not 2.5 Gbit/s full-duplex support is
advertised.
.It Sy MAC_PROP_EN_2500FDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_EN_2500FDX_CAP
property describes whether or not 2.5 Gbit/s full-duplex support is
enabled.
.It Sy MAC_PROP_ADV_5000FDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_5000FDX_CAP
property describes whether or not 5.0 Gbit/s full-duplex support is
advertised.
.It Sy MAC_PROP_EN_5000FDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_EN_5000FDX_CAP
property describes whether or not 5.0 Gbit/s full-duplex support is
enabled.
.It Sy MAC_PROP_ADV_10GFDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_10GFDX_CAP
property describes whether or not 10 Gbit/s full-duplex support is
advertised.
.It Sy MAC_PROP_EN_10GFDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_EN_10GFDX_CAP
property describes whether or not 10 Gbit/s full-duplex support is
enabled.
.It Sy MAC_PROP_ADV_40GFDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_40GFDX_CAP
property describes whether or not 40 Gbit/s full-duplex support is
advertised.
.It Sy MAC_PROP_EN_40GFDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_EN_40GFDX_CAP
property describes whether or not 40 Gbit/s full-duplex support is
enabled.
.It Sy MAC_PROP_ADV_100GFDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read-Only
.Ed
.Pp
The
.Sy MAC_PROP_ADV_100GFDX_CAP
property describes whether or not 100 Gbit/s full-duplex support is
advertised.
.It Sy MAC_PROP_EN_100GFDX_CAP
.Bd -filled -compact
Type:
.Sy uint8_t |
Permissions:
.Sy Read/Write
.Ed
.Pp
The
.Sy MAC_PROP_EN_100GFDX_CAP
property describes whether or not 100 Gbit/s full-duplex support is
enabled.
.El
.Ss Private Properties
In addition to the defined properties above, drivers are allowed to
define private properties.
These private properties are device-specific properties.
All private properties share the same constant,
.Sy MAC_PROP_PRIVATE .
Properties are distinguished by a name, which is a character string.
The list of such private properties is defined when registering with mac in the
.Sy m_priv_props
member of the
.Xr mac_register 9S
structure.
.Pp
The driver may define whatever semantics it wants for these private
properties.
They will not be listed when running
.Xr dladm 1M ,
unless explicitly requested by name.
All such properties should start with a leading underscore character and then
consist of alphanumeric ASCII characters and additional underscores or hyphens.
.Pp
Properties of type
.Sy MAC_PROP_PRIVATE
may show up in all three property related entry points:
.Xr mc_propinfo 9E ,
.Xr mc_getprop 9E ,
and
.Xr mc_setprop 9E .
Device drivers should tell the different properties apart by using the
.Xr strcmp 9F
function to compare it to the set of properties that it knows about.
When encountering properties that it doesn't know, it should treat them
like all other unknown properties.
.Sh STATISTICS
The MAC framework defines a couple different sets of statistics which
are based on various standards for devices to implement.
Statistics are retrieved through the
.Xr mc_getstat 9E
entry point.
There are both statistics that are required for all devices and then there is a
separate set of Ethernet specific statistics.
Not all devices will support every statistic.
In many cases, several device registers will need to be combined to create the
proper stat.
.Pp
In general, if the device is not keeping track of these statistics, then
it is recommended that the driver store these values as a
.Sy uint64_t
to ensure that overflow does not occur.
.Pp
If a device does not support a specific statistic, then it is fine to
return that it is not supported.
The same should be used for unrecognized statistics.
See
.Xr mc_getstat 9E
for more information on the proper way to handle these.
.Ss General Device Statistics
The following statistics are based on MIB-II statistics from both RFC
1213 and RFC 1573.
.Bl -tag -width Ds
.It Sy MAC_STAT_IFSPEED
The device's current speed in bits per second.
.It Sy MAC_STAT_MULTIRCV
The total number of received multicast packets.
.It Sy MAC_STAT_BRDCSTRCV
The total number of received broadcast packets.
.It Sy MAC_STAT_MULTIXMT
The total number of transmitted multicast packets.
.It Sy MAC_STAT_BRDCSTXMT
The total number of received broadcast packets.
.It Sy MAC_STAT_NORCVBUF
The total number of packets discarded by the hardware due to a lack of
receive buffers.
.It Sy MAC_STAT_IERRORS
The total number of errors detected on input.
.It Sy MAC_STAT_UNKNOWNS
The total number of received packets that were discarded because they
were of an unknown protocol.
.It Sy MAC_STAT_NOXMTBUF
The total number of outgoing packets dropped due to a lack of transmit
buffers.
.It Sy MAC_STAT_OERRORS
The total number of outgoing packets that resulted in errors.
.It Sy MAC_STAT_COLLISIONS
Total number of collisions encountered by the transmitter.
.It Sy MAC_STAT_RBYTES
The total number of
.Sy bytes
received by the device, regardless of packet type.
.It Sy MAC_STAT_IPACKETS
The total number of
.Sy packets
received by the device, regardless of packet type.
.It Sy MAC_STAT_OBYTES
The total number of
.Sy bytes
transmitted by the device, regardless of packet type.
.It Sy MAC_STAT_OPACKETS
The total number of
.Sy packets
sent by the device, regardless of packet type.
.It Sy MAC_STAT_UNDERFLOWS
The total number of packets that were smaller than the minimum sized
packet for the device and were therefore dropped.
.It Sy MAC_STAT_OVERFLOWS
The total number of packets that were larger than the maximum sized
packet for the device and were therefore dropped.
.El
.Ss Ethernet Specific Statistics
The following statistics are specific to Ethernet devices.
They refer to values from RFC 1643 and include various MII/GMII specific stats.
Many of these are also defined in IEEE 802.3.
.Bl -tag -width Ds
.It Sy ETHER_STAT_ADV_CAP_1000FDX
Indicates that the device is advertising support for 1 Gbit/s
full-duplex operation.
.It Sy ETHER_STAT_ADV_CAP_1000HDX
Indicates that the device is advertising support for 1 Gbit/s
half-duplex operation.
.It Sy ETHER_STAT_ADV_CAP_100FDX
Indicates that the device is advertising support for 100 Mbit/s
full-duplex operation.
.It Sy ETHER_STAT_ADV_CAP_100GFDX
Indicates that the device is advertising support for 100 Gbit/s
full-duplex operation.
.It Sy ETHER_STAT_ADV_CAP_100HDX
Indicates that the device is advertising support for 100 Mbit/s
half-duplex operation.
.It Sy ETHER_STAT_ADV_CAP_100T4
Indicates that the device is advertising support for 100 Mbit/s
100BASE-T4 operation.
.It Sy ETHER_STAT_ADV_CAP_10FDX
Indicates that the device is advertising support for 10 Mbit/s
full-duplex operation.
.It Sy ETHER_STAT_ADV_CAP_10GFDX
Indicates that the device is advertising support for 10 Gbit/s
full-duplex operation.
.It Sy ETHER_STAT_ADV_CAP_10HDX
Indicates that the device is advertising support for 10 Mbit/s
half-duplex operation.
.It Sy ETHER_STAT_ADV_CAP_2500FDX
Indicates that the device is advertising support for 2.5 Gbit/s
full-duplex operation.
.It Sy ETHER_STAT_ADV_CAP_40GFDX
Indicates that the device is advertising support for 40 Gbit/s
full-duplex operation.
.It Sy ETHER_STAT_ADV_CAP_5000FDX
Indicates that the device is advertising support for 5.0 Gbit/s
full-duplex operation.
.It Sy ETHER_STAT_ADV_CAP_ASMPAUSE
Indicates that the device is advertising support for receiving pause
frames.
.It Sy ETHER_STAT_ADV_CAP_AUTONEG
Indicates that the device is advertising support for auto-negotiation.
.It Sy ETHER_STAT_ADV_CAP_PAUSE
Indicates that the device is advertising support for generating pause
frames.
.It Sy ETHER_STAT_ADV_REMFAULT
Indicates that the device is advertising support for detecting faults in
the remote link peer.
.It Sy ETHER_STAT_ALIGN_ERRORS
Indicates the number of times an alignment error was generated by the
Ethernet device.
This is a count of packets that were not an integral number of octets and failed
the FCS check.
.It Sy ETHER_STAT_CAP_1000FDX
Indicates the device supports 1 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_CAP_1000HDX
Indicates the device supports 1 Gbit/s half-duplex operation.
.It Sy ETHER_STAT_CAP_100FDX
Indicates the device supports 100 Mbit/s full-duplex operation.
.It Sy ETHER_STAT_CAP_100GFDX
Indicates the device supports 100 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_CAP_100HDX
Indicates the device supports 100 Mbit/s half-duplex operation.
.It Sy ETHER_STAT_CAP_100T4
Indicates the device supports 100 Mbit/s 100BASE-T4 operation.
.It Sy ETHER_STAT_CAP_10FDX
Indicates the device supports 10 Mbit/s full-duplex operation.
.It Sy ETHER_STAT_CAP_10GFDX
Indicates the device supports 10 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_CAP_10HDX
Indicates the device supports 10 Mbit/s half-duplex operation.
.It Sy ETHER_STAT_CAP_2500FDX
Indicates the device supports 2.5 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_CAP_40GFDX
Indicates the device supports 40 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_CAP_5000FDX
Indicates the device supports 5.0 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_CAP_ASMPAUSE
Indicates that the device supports the ability to receive pause frames.
.It Sy ETHER_STAT_CAP_AUTONEG
Indicates that the device supports the ability to perform link
auto-negotiation.
.It Sy ETHER_STAT_CAP_PAUSE
Indicates that the device supports the ability to transmit pause frames.
.It Sy ETHER_STAT_CAP_REMFAULT
Indicates that the device supports the ability of detecting a remote
fault in a link peer.
.It Sy ETHER_STAT_CARRIER_ERRORS
Indicates the number of times that the Ethernet carrier sense condition
was lost or not asserted.
.It Sy ETHER_STAT_DEFER_XMTS
Indicates the number of frames for which the device was unable to
transmit the frame due to being busy and had to try again.
.It Sy ETHER_STAT_EX_COLLISIONS
Indicates the number of frames that failed to send due to an excessive
number of collisions.
.It Sy ETHER_STAT_FCS_ERRORS
Indicates the number of times that a frame check sequence failed.
.It Sy ETHER_STAT_FIRST_COLLISIONS
Indicates the number of times that a frame was eventually transmitted
successfully, but only after a single collision.
.It Sy ETHER_STAT_JABBER_ERRORS
Indicates the number of frames that were received that were both larger
than the maximum packet size and failed the frame check sequence.
.It Sy ETHER_STAT_LINK_ASMPAUSE
Indicates whether the link is currently configured to accept pause
frames.
.It Sy ETHER_STAT_LINK_AUTONEG
Indicates whether the current link state is a result of
auto-negotiation.
.It Sy ETHER_STAT_LINK_DUPLEX
Indicates the current duplex state of the link.
The values used here should be the same as documented for
.Sy MAC_PROP_DUPLEX .
.It Sy ETHER_STAT_LINK_PAUSE
Indicates whether the link is currently configured to generate pause
frames.
.It Sy ETHER_STAT_LP_CAP_1000FDX
Indicates the remote device supports 1 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_LP_CAP_1000HDX
Indicates the remote device supports 1 Gbit/s half-duplex operation.
.It Sy ETHER_STAT_LP_CAP_100FDX
Indicates the remote device supports 100 Mbit/s full-duplex operation.
.It Sy ETHER_STAT_LP_CAP_100GFDX
Indicates the remote device supports 100 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_LP_CAP_100HDX
Indicates the remote device supports 100 Mbit/s half-duplex operation.
.It Sy ETHER_STAT_LP_CAP_100T4
Indicates the remote device supports 100 Mbit/s 100BASE-T4 operation.
.It Sy ETHER_STAT_LP_CAP_10FDX
Indicates the remote device supports 10 Mbit/s full-duplex operation.
.It Sy ETHER_STAT_LP_CAP_10GFDX
Indicates the remote device supports 10 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_LP_CAP_10HDX
Indicates the remote device supports 10 Mbit/s half-duplex operation.
.It Sy ETHER_STAT_LP_CAP_2500FDX
Indicates the remote device supports 2.5 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_LP_CAP_40GFDX
Indicates the remote device supports 40 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_LP_CAP_5000FDX
Indicates the remote device supports 5.0 Gbit/s full-duplex operation.
.It Sy ETHER_STAT_LP_CAP_ASMPAUSE
Indicates that the remote device supports the ability to receive pause
frames.
.It Sy ETHER_STAT_LP_CAP_AUTONEG
Indicates that the remote device supports the ability to perform link
auto-negotiation.
.It Sy ETHER_STAT_LP_CAP_PAUSE
Indicates that the remote device supports the ability to transmit pause
frames.
.It Sy ETHER_STAT_LP_CAP_REMFAULT
Indicates that the remote device supports the ability of detecting a
remote fault in a link peer.
.It Sy ETHER_STAT_MACRCV_ERRORS
Indicates the number of times that the internal MAC layer encountered an
error when attempting to receive and process a frame.
.It Sy ETHER_STAT_MACXMT_ERRORS
Indicates the number of times that the internal MAC layer encountered an
error when attempting to process and transmit a frame.
.It Sy ETHER_STAT_MULTI_COLLISIONS
Indicates the number of times that a frame was eventually transmitted
successfully, but only after more than one collision.
.It Sy ETHER_STAT_SQE_ERRORS
Indicates the number of times that an SQE error occurred.
The specific conditions for this error are documented in IEEE 802.3.
.It Sy ETHER_STAT_TOOLONG_ERRORS
Indicates the number of frames that were received that were longer than
the maximum frame size supported by the device.
.It Sy ETHER_STAT_TOOSHORT_ERRORS
Indicates the number of frames that were received that were shorter than
the minimum frame size supported by the device.
.It Sy ETHER_STAT_TX_LATE_COLLISIONS
Indicates the number of times a collision was detected late on the
device.
.It Sy ETHER_STAT_XCVR_ADDR
Indicates the address of the MII/GMII receiver address.
.It Sy ETHER_STAT_XCVR_ID
Indicates the id of the MII/GMII receiver address.
.It Sy ETHER_STAT_XCVR_INUSE
Indicates what kind of receiver is in use.
The following values may be used:
.Bl -tag -width Ds
.It Sy XCVR_UNDEFINED
The receiver type is undefined by the hardware.
.It Sy XCVR_NONE
There is no receiver in use by the hardware.
.It Sy XCVR_10
The receiver supports 10BASE-T operation.
.It Sy XCVR_100T4
The receiver supports 100BASE-T4 operation.
.It Sy XCVR_100X
The receiver supports 100BASE-TX operation.
.It Sy XCVR_100T2
The receiver supports 100BASE-T2 operation.
.It Sy XCVR_1000X
The receiver supports 1000BASE-X operation.
This is used for all fiber receivers.
.It Sy XCVR_1000T
The receiver supports 1000BASE-T operation.
This is used for all copper receivers.
.El
.El
.Ss Device Specific kstats
In addition to the defined statistics above, if the device driver
maintains additional statistics or the device provides additional
statistics, it should create its own kstats through the
.Xr kstat_create 9F
function to allow operators to observe them.
.Sh TX STALL DETECTION, DEVICE RESETS, AND FAULT MANAGEMENT
Device drivers are the first line of defense for dealing with broken
devices and bugs in their firmware.
While most devices will rarely fail, it is important that when designing and
implementing the device driver that particular attention is paid in the design
with respect to RAS (Reliability, Availability, and Serviceability).
While everything described in this section is optional, it is highly recommended
that all new device drivers follow these guidelines.
.Pp
The Fault Management Architecture (FMA) provides facilities for
detecting and reporting various classes of defects and faults.
Specifically for networking device drivers, issues that should be
detected and reported include:
.Bl -bullet -offset indent
.It
Device internal uncorrectable errors
.It
Device internal correctable errors
.It
PCI and PCI Express transport errors
.It
Device temperature alarms
.It
Device transmission stalls
.It
Device communication timeouts
.It
High invalid interrupts
.El
.Pp
All such errors fall into three primary categories:
.Bl -enum -offset indent
.It
Errors detected by the Fault Management Architecture
.It
Errors detected by the device and indicated to the device driver
.It
Errors detected by the device driver
.El
.Ss Fault Management Setup and Teardown
Drivers should initialize support for the fault management framework by
calling
.Xr ddi_fm_init 9F
from their
.Xr attach 9E
routine.
By registering with the fault management framework, a device driver is given the
chance to detect and notice transport errors as well as report other errors that
exist.
While a device driver does not need to indicate that it is capable of all such
capabilities described in
.Xr ddi_fm_init 9F ,
we suggest that device drivers at least register the
.Sy DDI_FM_EREPORT_CAPABLE
so as to allow the driver to report issues that it detects.
.Pp
If the driver registers with the fault management framework during its
.Xr attach 9E
entry point, it must call
.Xr ddi_fm_fini 9E
during its
.Xr detach 9E
entry point.
.Ss Transport Errors
Many modern networking devices leverage PCI or PCI Express.
As such, there are two primary ways that device drivers access data: they either
memory map device registers and use routines like
.Xr ddi_get8 9F
and
.Xr ddi_put8 9F
or they use direct memory access (DMA).
New device drivers should always enable checking of the transport layer by
marking their support in the
.Xr ddi_device_acc_attr_t 9S
structure and using routines like
.Xr ddi_fm_acc_err_get 9F
and
.Xr ddi_fm_dma_err_get 9F
to detect if errors have occurred.
.Ss Device Indicated Errors
Many devices have capabilities to announce to a device driver that a
fatal correctable error or uncorrectable error has occurred.
Other devices have the ability to indicate that various physical issues have
occurred such as a fan failing or a temperature sensor having fired.
.Pp
Drivers should wire themselves to receive notifications when these
events occur.
The means and capabilities will vary from device to device.
For example, some devices will generate information about these notifications
through special interrupts.
Other devices may have a register that software can poll.
In the cases where polling is required, driver writers should try not to poll
too frequently and should generally only poll when the device is actively being
used, e.g. between calls to the
.Xr mc_start 9E
and
.Xr mc_stop 9E
entry points.
.Ss Driver Transmit Stall Detection
One of the primary responsibilities of a hardened device driver is to
perform transmit stall detection.
The core idea behind tx stall detection is that the driver should record when
it's getting activity related to when data has been successfully transmitted.
Most devices should be transmitting data on a regular basis as long as the link
is up.
If it is not, then this may indicate that the device is stuck and needs to be
reset.
At this time, the MAC framework does not provide any resources for performing
these checks; however, polling on each individual transmit ring for the last
completion time while something is actively being transmitted through the use of
routines such as
.Xr timeout 9F
may be a reasonable starting point.
.Ss Driver Command Timeout Detection
Each device is programmed in different ways.
Some devices are programmed through asynchronous commands while others are
programmed by writing directly to memory mapped registers.
If a device receives asynchronous replies to commands, then the device driver
should set reasonable timeouts for all such commands and plan on detecting them.
If a timeout occurs, the driver should presume that there is an issue with the
hardware and proceed to abort the command or reset the device.
.Pp
Many devices do not have such a communication mechanism.
However, whenever there is some activity where the device driver must wait, then
it should be prepared for the fact that the device may never get back to
it and react appropriately by performing some kind of device reset.
.Ss Reacting to Errors
When any of the above categories of errors has been triggered, the
behavior that the device driver should take depends on the kind of
error.
If a fatal error, for example, a transport error, a transmit stall was detected,
or the device indicated an uncorrectable error was detected, then it is
important that the driver take the following steps:
.Bl -enum -offset indent
.It
Set a flag in the device driver's state that indicates that it has hit
an error condition.
When this error condition flag is asserted, transmitted packets should be
accepted and dropped and actions that would require writing to the device state
should fail with an error.
This flag should remain until the device has been successfully restarted.
.It
If the error was not a transport error that was indicated by the fault
management architecture, e.g. a transport error that was detected, then
the device driver should post an
.Sy ereport
indicating what has occurred with the
.Xr ddi_fm_ereport_post 9F
function.
.It
The device driver should indicate that the device's service was lost
with a call to
.Xr ddi_fm_service_impact 9F
using the symbol
.Sy DDI_SERVICE_LOST .
.It
At this point the device driver should issue a device reset through some
device-specific means.
.It
When the device reset has been completed, then the device driver should
restore all of the programmed state to the device.
This includes things like the current MTU, advertised auto-negotiation speeds,
MAC address filters, and more.
.It
Finally, when service has been restored, the device driver should call
.Xr ddi_fm_service_impact 9F
using the symbol
.Sy DDI_SERVICE_RESTORED .
.El
.Pp
When a non-fatal error occurs, then the device driver should submit an
ereport and should optionally mark the device degraded using
.Xr ddi_fm_service_impact 9F
with the
.Sy DDI_SERVICE_DEGRADED
value depending on the nature of the problem that has occurred.
.Pp
Device drivers should never make the decision to remove a device from
service based on errors that have occurred nor should they panic the
system.
Rather, the device driver should always try to notify the operating system with
various ereports and allow its policy decisions to occur.
The decision to retire a device lies in the hands of the fault management
architecture.
It knows more about the operator's intent and the surrounding system's state
than the device driver itself does and it will make the call to offline and
retire the device if it is required.
.Ss Device Resets
When resetting a device, a device driver must exercise caution.
If a device driver has not been written to plan for a device reset, then it
may not correctly restore the device's state after such a reset.
Such state should be stored in the instance's private state data as the MAC
framework does not know about device resets and will not inform the
device again about the expected, programmed state.
.Pp
One wrinkle with device resets is that many networking cards show up as
multiple PCI functions on a single device, for example, each port may
show up as a separate function and thus have a separate instance of the
device driver attached.
When resetting a function, device driver writers should carefully read the
device programming manuals and verify whether or not a reset impacts only the
stalled function or if it impacts all function across the device.
.Pp
If the only way to reset a given function is through the device, then
this may require more coordination and work on the part of the device
driver to ensure that all the other instances are correctly restored.
In cases where this occurs, some devices offer ways of injecting
interrupts onto those other functions to notify them that this is
occurring.
.Sh MBLKS AND DMA
The networking stack manages framed data through the use of the
.Xr mblk 9S
structure.
The mblk allows for a single message to be made up of individual blocks.
Each part is linked together through its
.Sy b_cont
member.
However, it also allows for multiple messages to be chained together through the
use of the
.Sy b_next
member.
While the networking stack works with these structures, device drivers generally
work with DMA regions.
There are two different strategies that device drivers use for handling these
two different cases: copying and binding.
.Ss Copying Data
The first way that device drivers handle interfacing between the two is
by having two separate regions of memory.
One part is memory which has been allocated for DMA through a call to
.Xr ddi_dma_alloc 9F
and the other is memory associated with the memory block.
.Pp
In this case, a driver will use
.Xr bcopy 9F
to copy memory between the two distinct regions.
When transmitting a packet, it will copy the memory from the mblk_t to the DMA
region.
When receiving memory, it will allocate a mblk_t through the
.Xr allocb 9F
routine, copy the memory across with
.Xr bcopy 9F ,
and then increment the mblk_t's
.Sy w_ptr
structure.
.Pp
If, when receiving, memory is not available for a new message block,
then the frame should be skipped and effectively dropped.
A kstat should be bumped when such an occasion occurs.
.Ss Binding Data
An alternative approach to copying data is to use DMA binding.
When using DMA binding, the OS takes care of mapping between DMA memory and
normal device memory.
The exact process is a bit different between transmit and receive.
.Pp
When transmitting a device driver has an mblk_t and needs to call the
.Xr ddi_dma_addr_bind_handle 9F
function to bind it to an already existing DMA handle.
At that point, it will receive various DMA cookies that it can use to obtain the
addresses to program the device with for transmitting data.
Once the transmit is done, the driver must then make sure to call
.Xr freemsg 9F
to release the data.
It must not call
.Xr freemsg 9F
before it receives an interrupt from the device indicating that the data
has been transmitted, otherwise it risks sending arbitrary kernel
memory.
.Pp
When receiving data, the device can perform a similar operation.
First, it must bind the DMA memory into the kernel's virtual memory address
space through a call to the
.Xr ddi_dma_addr_bind_handle 9F
function if it has not already.
Once it has, it must then call
.Xr desballoc 9F
to try and create a new mblk_t which leverages the associated memory.
It can then pass that mblk_t up to the stack.
.Ss Considerations
When deciding which of these options to use, there are many different
considerations that must be made.
The answer as to whether to bind memory or to copy data is not always simpler.
.Pp
The first thing to remember is that DMA resources may be finite on a
given platform.
Consider the case of receiving data.
A device driver that binds one of its receive descriptors may not get it back
for quite some time as it may be used by the kernel until an application
actually consumes it.
Device drivers that try to bind memory for receive, often work with the
constraint that they must be able to replace that DMA memory with another DMA
descriptor.
If they were not replaced, then eventually the device would not be able to
receive additional data into the ring.
.Pp
On the other hand, particularly for larger frames, copying every packet
from one buffer to another can be a source of additional latency and
memory waste in the system.
For larger copies, the cost of copying may dwarf any potential cost of
performing DMA binding.
.Pp
For device driver authors that are unsure of what to do, they should
first employ the copying method to simplify the act of writing the
device driver.
The copying method is simpler and also allows the device driver author not to
worry about allocated DMA memory that is still outstanding when it is asked to
unload.
.Pp
If device driver writers are worried about the cost, it is recommended
to make the decision as to whether or not to copy or bind DMA data
a separate private property for both transmitting and receiving.
That private property should indicate the size of the received frame at which
to switch from one format to the other.
This way, data can be gathered to determine what the impact of each method is on
a given platform.
.Sh SEE ALSO
.Xr dladm 1M ,
.Xr driver.conf 4 ,
.Xr ieee802.3 5 ,
.Xr dlpi 7P ,
.Xr _fini 9E ,
.Xr _info 9E ,
.Xr _init 9E ,
.Xr attach 9E ,
.Xr close 9E ,
.Xr detach 9E ,
.Xr mc_close 9E ,
.Xr mc_getcapab 9E ,
.Xr mc_getprop 9E ,
.Xr mc_getstat 9E ,
.Xr mc_multicst 9E  ,
.Xr mc_open 9E ,
.Xr mc_propinfo 9E  ,
.Xr mc_setpromisc 9E  ,
.Xr mc_setprop 9E ,
.Xr mc_start 9E ,
.Xr mc_stop 9E ,
.Xr mc_tx 9E ,
.Xr mc_unicst 9E  ,
.Xr open 9E ,
.Xr allocb 9F ,
.Xr bcopy 9F ,
.Xr ddi_dma_addr_bind_handle 9F ,
.Xr ddi_dma_alloc 9F ,
.Xr ddi_fm_acc_err_get 9F ,
.Xr ddi_fm_dma_err_get 9F ,
.Xr ddi_fm_ereport_post 9F ,
.Xr ddi_fm_fini 9F ,
.Xr ddi_fm_init 9F ,
.Xr ddi_fm_service_impact 9F ,
.Xr ddi_get8 9F ,
.Xr ddi_put8 9F ,
.Xr desballoc 9F ,
.Xr freemsg 9F ,
.Xr kstat_create 9F ,
.Xr mac_alloc 9F ,
.Xr mac_fini_ops 9F ,
.Xr mac_hcksum_get 9F ,
.Xr mac_hcksum_set 9F ,
.Xr mac_init_ops 9F ,
.Xr mac_link_update 9F ,
.Xr mac_lso_get 9F ,
.Xr mac_maxsdu_update 9F ,
.Xr mac_prop_info_set_default_link_flowctrl 9F ,
.Xr mac_prop_info_set_default_str 9F ,
.Xr mac_prop_info_set_default_uint32 9F ,
.Xr mac_prop_info_set_default_uint64 9F ,
.Xr mac_prop_info_set_default_uint8 9F ,
.Xr mac_prop_info_set_perm 9F ,
.Xr mac_prop_info_set_range_uint32 9F ,
.Xr mac_register 9F ,
.Xr mac_rx 9F ,
.Xr mac_unregister 9F ,
.Xr mc_getprop 9F ,
.Xr mc_tx 9F ,
.Xr mod_install 9F ,
.Xr mod_remove 9F ,
.Xr strcmp 9F ,
.Xr timeout 9F ,
.Xr cb_ops 9S ,
.Xr ddi_device_acc_attr_t 9S ,
.Xr dev_ops 9S ,
.Xr kstat_create 9S ,
.Xr mac_callbacks 9S ,
.Xr mac_register 9S ,
.Xr mblk 9S ,
.Xr modldrv 9S ,
.Xr modlinkage 9S
.Rs
.%A McCloghrie, K.
.%A Rose, M.
.%T RFC 1213 Management Information Base for Network Management of
.%T TCP/IP-based internets: MIB-II
.%D March 1991
.Re
.Rs
.%A McCloghrie, K.
.%A Kastenholz, F.
.%T RFC 1573 Evolution of the Interfaces Group of MIB-II
.%D January 1994
.Re
.Rs
.%A Kastenholz, F.
.%T RFC 1643 Definitions of Managed Objects for the Ethernet-like
.%T Interface Types
.Re