This is the Ethernet part of the driver for the Mellanox ConnectX(R)-4
Single/Dual-Port Adapter supporting 100Gb/s with VPI. The driver
extends the existing mlx5 driver with Ethernet functionality.
This patch contains the driver entry points but does not include
transmit and receive (see the previous patch in the series) routines.
It also adds the option MLX5_CORE_EN to Kconfig to enable/disable the
Ethernet functionality. Currently, Kconfig is programmed to make
Ethernet and Infiniband functionality mutally exclusive.
Also changed MLX5_INFINIBAND to be depandant on MLX5_CORE instead of
selecting it, since MLX5_CORE could be selected without MLX5_INFINIBAND
being selected.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch contains the resource handling files:
- flow_table.c: This file contains the code to handle the low level API
to configure hardware flow table. It is separated from
the flow_table_en.c, because it will be used in the
future by Raw Ethernet QP in mlx5_ib too.
- en_flow_table.[ch]: Ethernet flow steering handling. The flow table
object contain a mapping between flow specs and TIRs.
This mechanism will be used also to configure e-switch
in the future, when SR-IOV support will be added.
- transobj.[ch] - Low level functions to create/modify/destroy the
transport objects: RQ/SQ/TIR/TIS
- vport.[ch] - Handle attributes of a virtual port (vPort) in the
embedded switch. Currently this switch is a passthrough, until SR-IOV
support will be added.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
en_[rt]x.c contains the data path related code specific to tx or rx.
en_txrx.c contains data path code which is common for both the rx and
tx, this is mainly napi related code.
Below are the objects that are being used by the hardware and the driver
in the data path:
Channel - one channel per IRQ. Every channel object contains:
RQ - describes the rx queue
TIR - One TIR (Transport Interface Receive) object per flow type. TIR
contains attributes for a type of rx flow (e.g IPv4, IPv6 etc).
A flow is defined in the Flow Table.
Currently TIR describes the RSS hash parameters if exists and LRO
attributes.
SQ - describes the a tx queue. There is one SQ (Send Queue) per
TC (traffic class).
TIS - There is one TIS (Transport Interface Send) per TC. It
describes the TC and may later be extended to describe more
transport properties.
Both RQ and SQ inherit from the object WQ (work queue). This common code
to describe the layout of CQE's WQE's in memory is in the files wq.[cj]
For every channel there is one NAPI context that is used for RX and
for TX.
Driver is using netdev_alloc_skb() to allocate skb's.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce set/Query low level functions to access MTU in hardware. To be
used by the netdev.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce mlx5_core_modify_cq_moderation() to be used by the netdev, to
set hardware coalescing.
Signed-off-by: Rana Shahout <ranas@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implemet get/set port status low level functions to be exposed by the
netdev.
Signed-off-by: Rana Shahout <ranas@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Those registers will be used by the ethtool to set/get settings.
Signed-off-by: Rana Shahout <ranas@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- Query all supported types of dev caps on driver load.
- Store the Cap data outbox per cap type into driver private data.
- Introduce new Macros to access/dump stored caps (using the auto
generated data types).
- Obsolete SW representation of dev caps (no need for SW copy for each
cap).
- Modify IB driver to use new macros for checking caps.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlx5_ifc.h was heavily modified here since it is now generated by a
script from the device specification (PRM rev 0.25). This specification
is backward compatible to existing hardware.
Some structures/fields were added here in order to enable the Ethernet
functionality of the driver.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Preparation for upcoming ethernet driver.
- Move msix array from eq_table struct to priv since its not related to
eq_table
- Intorduce irq_info struct to hold all irq information
- Move name from mlx5_eq to irq_info struct since it is irq property.
- Set IRQ affinity hints
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Rana Shahout <ranas@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As David Daney pointed in mlx4_core driver [1], mlx5_core is also
misusing the DMA-API.
This patch is removing the code that vmap() memory allocated by
dma_alloc_coherent().
After this patch, users of this drivers might fail allocating resources
on memory fragmeneted systems. This will be fixed later on.
[1] - https://patchwork.ozlabs.org/patch/458531/
CC: David Daney <david.daney@cavium.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jeff Kirsher says:
====================
Intel Wired LAN Driver Updates 2015-05-28
This series contains updates to ethtool, ixgbe, i40e and i40evf.
John adds helper routines for ethtool to pass VF to rx_flow_spec. Since
the ring_cookie is 64 bits wide which is much larger than what could be
used for actual queue index values, provide helper routines to pack a VF
index into the cookie. Then John provides a ixgbe patch to allow flow
director to use the entire queue space.
Neerav provides a i40e patch to collect XOFF Rx stats, where it was not
being collected before.
Anjali provides ATR support for tunneled packets, as well as stats to
count tunnel ATR hits. Cleaned up PF struct members which are
unnecessary, since we can use the stat index macro directly. Cleaned
up flow director ATR/SB messages to a higher debug level since they
are not useful unless silicon validation is happening.
Greg provides a patch to disable offline diagnostics if VFs are enabled
since ethtool offline diagnostic tests are not designed (out of scope)
to disable VF functions for testing and re-enable afterward. Also cleans
up TODO comment that is no longer needed.
Vasu provides a fix an FCoE EOF case where i40e_fcoe_ctxt_eof() maybe
called before i40e_fcoe_eof_is_supported() is called.
Jesse adds skb->xmit_more support for i40evf. Then provides a performance
enhancement for i40evf by inlining some functions which provides a 15%
gain in small packet performance. Also cleans up the use of time_stamp
since it is no longer used to determine if there is a tx_hang and was
a part of a previous tx_hang design which is no longer used.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
As sock refcnt is taken when sock timer is started in
sk_reset_timer(), the sock refcnt should be put when sock timer
to be deleted is in pending state no matter what "probing_state"
value of tipc sock is.
Reviewed-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
pcl812.c (resend of earlier patch)
Fixed lines exceeding 80 columns, correcting some spelling in process
Signed-off-by: Matthew Needes <mneedes@gmail.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Changed "register 0x%x" to "register=0x%x" to keep the consistency
of this file.
Signed-off-by: Geliang Tang <geliangtang@163.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fixed an error found by checkpatch.pl.
ERROR: space required after that ',' (ctx:VxV)
./drivers/ni_mio_common.c:3764
Signed-off-by: Geliang Tang <geliangtang@163.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Comedi "comedi_isadma.h" header is included by the source for the
"comedi_isadma" helper module and other modules that use it. It does
not compile cleanly when it is the first header file included. It uses
the `dma_addr_t` type, so include <linux/types.h> to declare it. (Also,
that indirectly takes care of the use of `NULL`.) It uses `struct
comedi_device *` in various function prototypes, so add an incomplete
declaration of `struct comedi_device`.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Comedi "comedi_8254.h" header file is included by various Comedi
drivers with timer/counters based on the 8254 chip. The drivers do not
compile cleanly if this header file is included first. It uses pointers
to the `struct comedi_device`, `struct comedi_subdevice`, and `struct
comedi_insn` structures in various function prototypes, so declare those
as incomplete types. It use the `bool` type, so include
<linux/types.h>. It also uses the `__iomem` tag, but that seems to be
taken care of indirectly by including <linux/types.h>.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Comedi "addi_watchdog.h" header doesn't use anything form
"comedidev.h" apart from `struct comedi_subdevice`, which it only uses
to construct a corresponding pointer type within the parameter list of a
function prototype. Just declare the structure type incompletely and
don't bother including the header file.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reformat the copyright comment at the top of the file to use the
preferred block comment style.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use braces when the single statement following an `if` (or `else`)
spans more than one line (including any preceding comments).
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reformat the copyright comment at the top of the file to use the
preferred block comment style.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reformat the copyright comment at the top of the file to use the
preferred block comment style.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Comedi "amplc_dio200.h" header file included by drivers for Amplicon
DIO200 series cards does not compile cleanly when it is the first header
included by the ".c" file. It uses `struct comedi_device *` in the
parameter lists of some function prototypes, so just declare `struct
comedi_device` as an incomplete type. It also uses `bool`, so include
<linux/types.h> to declare it.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reformat the copyright comment at the top of the file to use the
preferred block comment style.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Comedi "8255" module is both a standalone Comedi device driver
module for simple devices with one or more 8255 "Programmable Peripheral
Interface" chips at known I/O base addresses (configured at run-time),
and a helper module to configure a 8255-based digital I/O subdevice for
other Comedi drivers.
Split the "8255 subdevice helper" functionality into a new module:
"comedi_8255", leaving the standalone 8255 Comedi driver in the "8255"
module.
The Comedi "detach" routine of the standalone "8255" driver needs to
retrieve the I/O base address passed to the "comedi_8255" module to set
up each subdevice in order to release the I/O port regions it requested
in its "attach" routine. The "comedi_8255" module stores it in a
"subdevice private" data structure that is no longer known to the "8255"
module, so add a new, exported function `subdev_8255_regbase()` to
retrieve it.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Several Comedi driver modules call `subdev_8255_init()` or
`subdev_8255_mm_init()` to set up a digital I/O subdevice based on the
"8255" chip. One of the parameters to these functions is an optional
pointer to an I/O callback function to perform the actual register
accesses (an internal default callback function is used if NULL).
The kerneldoc for `subdev_8255_init()` and `subdev_8255_mm_init()`
describe the prototype of the optional I/O callback function
incorrectly (my fault), adding a non-existent parameter of type `struct
comedi_subdevice *`. Fix the kerneldoc. Also add parameter names to
the callback function pointer type wherever it occurs to make the usage
clearer.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Comedi "8255.h" header doesn't use anything from "comedidev.h" apart
from `struct comedi_device` and `struct comedi_subdevice`, which are
only used to construct corresponding pointer types within the parameter
lists of function prototypes. Just declare those structure types
incompletely and don't bother including "comedidev.h".
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Comedi "8255" driver does not clean up properly on failure. It can
leave requested I/O port regions unreleased. Specifically, the Comedi
"attach" handler (`dev_8255_attach()`) requests a specified I/O port
region before calling `subdev_8255_init()` to set up the subdevice. If
that fails, the "attach" handler returns an error and the Comedi core
will call the "detach" handler (`dev_8255_detach()`) to clean up. The
"detach" handler is responsible for releasing the I/O port regions
successfully requested by the "attach" handler. Unfortunately, it is
unable to obtain the base address of the region if the call to
`subdev_8255_init()` failed.
Fix the I/O region leak by releasing the region in the "attach" handler
directly if the call to `subdev_8255_init()` fails.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is just one configfs subsystem in the target code, so we might as
well add two helpers to reference / unreference it from the core code
instead of passing pointers to it around.
This fixes a regression introduced for v4.1-rc1 with commit 9ac8928e6,
where configfs_depend_item() callers using se_tpg_tfo->tf_subsys would
fail, because the assignment from the original target_core_subsystem[]
is no longer happening at target_register_template() time.
(Fix target_core_exit_configfs pointer dereference - Sagi)
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reported-by: Himanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
clean up checkpatch.pl in prism2sta.c
WARNING : line over 80 characters
Signed-off-by: chaehyun lim <chaehyun.lim@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
check return value of kmalloc before accessing the memory pointer and
return -ENOMEM if allocation fails.
Signed-off-by: Hari Prasath Gujulan Elango <hgujulan@visteon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fixes "vlaue" for "value" in include/linux/if_vlan.h.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
classic BPF already exposes skb->dev->ifindex via SKF_AD_IFINDEX extension.
Allow eBPF program to access it as well. Note that classic aborts execution
of the program if 'skb->dev == NULL' (which is inconvenient for program
writers), whereas eBPF returns zero in such case.
Also expose the 'skb_iif' field, since programs triggered by redirected
packet need to known the original interface index.
Summary:
__skb->ifindex -> skb->dev->ifindex
__skb->ingress_ifindex -> skb->skb_iif
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
A recent patch to simplify the lustre large memory allocation
causes new warnings as an unintended side-effect:
lustre/lov/lov_request.c: In function 'lov_finish_set':
lustre/lov/lov_request.c:78:7: warning: unused variable 'len' [-Wunused-variable]
int len = set->set_oabufs * sizeof(*set->set_pga);
^
lustre/obdclass/acl.c: In function 'lustre_ext_acl_xattr_reduce_space':
lustre/obdclass/acl.c:123:6: warning: unused variable 'old_size' [-Wunused-variable]
int old_size = CFS_ACL_XATTR_SIZE(old_count, ext_acl_xattr);
^
The reason is that the 'size' argument to OBD_FREE_LARGE()
is never needed, which was previously hidden by the extra
abstractions.
This avoids the warnings by adding a cast to void, to tell
the compiler that the argument is intentionally unused.
A better fix is probably to remove the entire set of allocation
macros and open-code the normal kernel interface calls.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 99d56ff7c1 ("staging/lustre: Always try kmalloc first for OBD_ALLOC_LARGE")
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current algorithm for deciding on the number of VRSS channels is
not optimal since we open up the min of number of CPUs online and the
number of VRSS channels the host is offering. So on a 32 VCPU guest
we could potentially open 32 VRSS subchannels. Experimentation has
shown that it is best to limit the number of VRSS channels to the number
of CPUs within a NUMA node.
Here is the new algorithm for deciding on the number of sub-channels we
would open up:
1) Pick the minimum of what the host is offering and what the driver
in the guest is specifying as the default value.
2) Pick the minimum of (1) and the numbers of CPUs in the NUMA
node the primary channel is bound to.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is similar to b1cb59cf2efe(net: sysctl_net_core: check SNDBUF
and RCVBUF for min length). I don't think too small values can cause
crashes in the case of udp and tcp, but I've seen this set to too
small values which triggered awful performance. It also makes the
setting consistent across all the wmem/rmem sysctls.
Signed-off-by: Sorin Dumitru <sdumitru@ixiacom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
LNET_MUTEX_LOCK and LNET_MUTEX_UNLOCK are verbose wrappers to mutex_lock and
mutex_unlock. Get rid of these.
Signed-off-by: Mike Shuey <shuey@purdue.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unify spacing in variable declarations, and align indentation in headers.
General whitespace cleanups.
Signed-off-by: Mike Shuey <shuey@purdue.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unify variable declarations to use a single space. Also include several
miscellaneous whitespace cleanups, particularly in socklnd.h.
Signed-off-by: Mike Shuey <shuey@purdue.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unify variable declarations to use a single whitespace. Also line up
declarations and comments in o2iblnd.h.
Signed-off-by: Mike Shuey <shuey@purdue.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix the module version to match upstream development.
Signed-off-by: Mike Shuey <shuey@purdue.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
LNetSetAsync() returns 0, and is never called. Doesn't exist in the Intel
tree, either. Remove it.
Signed-off-by: Mike Shuey <shuey@purdue.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unify variable declarations to use a single space, and any other obvious
spacing flaws.
Signed-off-by: Mike Shuey <shuey@purdue.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In the explanation of the function the name of the function was incorrect
Signed-off-by: Adrian Remonda <adrianremonda@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In lov_unpackmd() there's this strange bit of code where we first try
to look inside of lmm striping pattern for it's type, and then
we check if the pattern is NULL which cannot be right.
Move the check under if (lmm) branch so that it's safe.
Found by Coverity version 6.6.1
Signed-off-by: Sebastien Buisson <sebastien.buisson@bull.net>
Reviewed-on: http://review.whamcloud.com/7827
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4049
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: jacques-Charles Lafoucriere <jacques-charles.lafoucriere@cea.fr>
Signed-off: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
smatch highlighted a wrongly indented bit of code that almost
hides the extra assignment.
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
