Skip to content

R
raspberrypi-kernel

openeuler / raspberrypi-kernel

通知 13
Star 1
Fork 0
R
raspberrypi-kernel
切换分支/标签
  1. 15 1月, 2014 1 次提交
    • M
      IB/core: Ethernet L2 attributes in verbs/cm structures · dd5f03be
      由 Matan Barak 提交于 
      This patch add the support for Ethernet L2 attributes in the
verbs/cm/cma structures.

When dealing with L2 Ethernet, we should use smac, dmac, vlan ID and priority
in a similar manner that the IB L2 (and the L4 PKEY) attributes are used.

Thus, those attributes were added to the following structures:

* ib_ah_attr - added dmac
* ib_qp_attr - added smac and vlan_id, (sl remains vlan priority)
* ib_wc - added smac, vlan_id
* ib_sa_path_rec - added smac, dmac, vlan_id
* cm_av - added smac and vlan_id

For the path record structure, extra care was taken to avoid the new
fields when packing it into wire format, so we don't break the IB CM
and SA wire protocol.

On the active side, the CM fills. its internal structures from the
path provided by the ULP.  We add there taking the ETH L2 attributes
and placing them into the CM Address Handle (struct cm_av).

On the passive side, the CM fills its internal structures from the WC
associated with the REQ message.  We add there taking the ETH L2
attributes from the WC.

When the HW driver provides the required ETH L2 attributes in the WC,
they set the IB_WC_WITH_SMAC and IB_WC_WITH_VLAN flags. The IB core
code checks for the presence of these flags, and in their absence does
address resolution from the ib_init_ah_from_wc() helper function.

ib_modify_qp_is_ok is also updated to consider the link layer. Some
parameters are mandatory for Ethernet link layer, while they are
irrelevant for IB.  Vendor drivers are modified to support the new
function signature.
Signed-off-by: NMatan Barak <matanb@mellanox.com>
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      dd5f03be
  3. 21 12月, 2013 7 次提交
  5. 17 12月, 2013 1 次提交
  7. 16 12月, 2013 1 次提交
    • S
      RDMA/iwcm: Don't touch cm_id after deref in rem_ref · 6b59ba60
      由 Steve Wise 提交于 
      rem_ref() calls iwcm_deref_id(), which will wake up any blockers on
cm_id_priv->destroy_comp if the refcnt hits 0.  That will unblock
someone in iw_destroy_cm_id() which will free the cmid.  If that
happens before rem_ref() calls test_bit(IWCM_F_CALLBACK_DESTROY,
&cm_id_priv->flags), then the test_bit() will touch freed memory.

The fix is to read the bit first, then deref.  We should never be in
iw_destroy_cm_id() with IWCM_F_CALLBACK_DESTROY set, and there is a
BUG_ON() to make sure of that.
Signed-off-by: NSteve Wise <swise@opengridcomputing.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      6b59ba60
  9. 18 11月, 2013 6 次提交
    • M
      IB/core: Re-enable create_flow/destroy_flow uverbs · 69ad5da4
      由 Matan Barak 提交于 
      This commit reverts commit 7afbddfa ("IB/core: Temporarily disable
create_flow/destroy_flow uverbs").  Since the uverbs extensions
functionality was experimental for v3.12, this patch re-enables the
support for them and flow-steering for v3.13.
Signed-off-by: NMatan Barak <matanb@mellanox.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      69ad5da4
    • Y
      IB/core: extended command: an improved infrastructure for uverbs commands · f21519b2
      由 Yann Droneaud 提交于 
      Commit 400dbc96 ("IB/core: Infrastructure for extensible uverbs
commands") added an infrastructure for extensible uverbs commands
while later commit 436f2ad0 ("IB/core: Export ib_create/destroy_flow
through uverbs") exported ib_create_flow()/ib_destroy_flow() functions
using this new infrastructure.

According to the commit 400dbc96, the purpose of this
infrastructure is to support passing around provider (eg. hardware)
specific buffers when userspace issue commands to the kernel, so that
it would be possible to extend uverbs (eg. core) buffers independently
from the provider buffers.

But the new kernel command function prototypes were not modified to
take advantage of this extension. This issue was exposed by Roland
Dreier in a previous review[1].

So the following patch is an attempt to a revised extensible command
infrastructure.

This improved extensible command infrastructure distinguish between
core (eg. legacy)'s command/response buffers from provider
(eg. hardware)'s command/response buffers: each extended command
implementing function is given a struct ib_udata to hold core
(eg. uverbs) input and output buffers, and another struct ib_udata to
hold the hw (eg. provider) input and output buffers.

Having those buffers identified separately make it easier to increase
one buffer to support extension without having to add some code to
guess the exact size of each command/response parts: This should make
the extended functions more reliable.

Additionally, instead of relying on command identifier being greater
than IB_USER_VERBS_CMD_THRESHOLD, the proposed infrastructure rely on
unused bits in command field: on the 32 bits provided by command
field, only 6 bits are really needed to encode the identifier of
commands currently supported by the kernel. (Even using only 6 bits
leaves room for about 23 new commands).

So this patch makes use of some high order bits in command field to
store flags, leaving enough room for more command identifiers than one
will ever need (eg. 256).

The new flags are used to specify if the command should be processed
as an extended one or a legacy one. While designing the new command
format, care was taken to make usage of flags itself extensible.

Using high order bits of the commands field ensure that newer
libibverbs on older kernel will properly fail when trying to call
extended commands. On the other hand, older libibverbs on newer kernel
will never be able to issue calls to extended commands.

The extended command header includes the optional response pointer so
that output buffer length and output buffer pointer are located
together in the command, allowing proper parameters checking. This
should make implementing functions easier and safer.

Additionally the extended header ensure 64bits alignment, while making
all sizes multiple of 8 bytes, extending the maximum buffer size:

                             legacy      extended

   Maximum command buffer:  256KBytes   1024KBytes (512KBytes + 512KBytes)
  Maximum response buffer:  256KBytes   1024KBytes (512KBytes + 512KBytes)

For the purpose of doing proper buffer size accounting, the headers
size are no more taken in account in "in_words".

One of the odds of the current extensible infrastructure, reading
twice the "legacy" command header, is fixed by removing the "legacy"
command header from the extended command header: they are processed as
two different parts of the command: memory is read once and
information are not duplicated: it's making clear that's an extended
command scheme and not a different command scheme.

The proposed scheme will format input (command) and output (response)
buffers this way:

- command:

  legacy header +
  extended header +
  command data (core + hw):

    +----------------------------------------+
    | flags     |   00      00    |  command |
    |        in_words    |   out_words       |
    +----------------------------------------+
    |                 response               |
    |                 response               |
    | provider_in_words | provider_out_words |
    |                 padding                |
    +----------------------------------------+
    |                                        |
    .              <uverbs input>            .
    .              (in_words * 8)            .
    |                                        |
    +----------------------------------------+
    |                                        |
    .             <provider input>           .
    .          (provider_in_words * 8)       .
    |                                        |
    +----------------------------------------+

- response, if present:

    +----------------------------------------+
    |                                        |
    .          <uverbs output space>         .
    .             (out_words * 8)            .
    |                                        |
    +----------------------------------------+
    |                                        |
    .         <provider output space>        .
    .         (provider_out_words * 8)       .
    |                                        |
    +----------------------------------------+

The overall design is to ensure that the extensible infrastructure is
itself extensible while begin more reliable with more input and bound
checking.

Note:

The unused field in the extended header would be perfect candidate to
hold the command "comp_mask" (eg. bit field used to handle
compatibility).  This was suggested by Roland Dreier in a previous
review[2].  But "comp_mask" field is likely to be present in the uverb
input and/or provider input, likewise for the response, as noted by
Matan Barak[3], so it doesn't make sense to put "comp_mask" in the
header.

[1]:
http://marc.info/?i=CAL1RGDWxmM17W2o_era24A-TTDeKyoL6u3NRu_=t_dhV_ZA9MA@mail.gmail.com

[2]:
http://marc.info/?i=CAL1RGDXJtrc849M6_XNZT5xO1+ybKtLWGq6yg6LhoSsKpsmkYA@mail.gmail.com

[3]:
http://marc.info/?i=525C1149.6000701@mellanox.comSigned-off-by: NYann Droneaud <ydroneaud@opteya.com>
Link: http://marc.info/?i=cover.1383773832.git.ydroneaud@opteya.com

[ Convert "ret ? ret : 0" to the equivalent "ret".  - Roland ]
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      f21519b2
    • Y
      IB/core: Remove ib_uverbs_flow_spec structure from userspace · 2490f20b
      由 Yann Droneaud 提交于 
      The structure holding any types of flow_spec is of no use to
userspace.  It would be wrong for userspace to do:

  struct ib_uverbs_flow_spec flow_spec;

  flow_spec.type = IB_FLOW_SPEC_TCP;
  flow_spec.size = sizeof(flow_spec);

Instead, userspace should use the dedicated flow_spec structure for
  - Ethernet : struct ib_uverbs_flow_spec_eth,
  - IPv4     : struct ib_uverbs_flow_spec_ipv4,
  - TCP/UDP  : struct ib_uverbs_flow_spec_tcp_udp.

In other words, struct ib_uverbs_flow_spec is a "virtual" data
structure that can only be use by the kernel as an alias to the other.
Signed-off-by: NYann Droneaud <ydroneaud@opteya.com>
Link: http://marc.info/?i=cover.1383773832.git.ydroneaud@opteya.comSigned-off-by: NRoland Dreier <roland@purestorage.com>
      2490f20b
    • Y
      IB/core: Make uverbs flow structure use names like verbs ones · b68c9560
      由 Yann Droneaud 提交于 
      This patch adds "flow" prefix to most of data structure added as part
of commit 436f2ad0 ("IB/core: Export ib_create/destroy_flow through
uverbs") to keep those names in sync with the data structures added in
commit 319a441d ("IB/core: Add receive flow steering support").

It's just a matter of translating 'ib_flow' to 'ib_uverbs_flow'.
Signed-off-by: NYann Droneaud <ydroneaud@opteya.com>
Link: http://marc.info/?i=cover.1383773832.git.ydroneaud@opteya.comSigned-off-by: NRoland Dreier <roland@purestorage.com>
      b68c9560
    • Y
      IB/core: Rename 'flow' structs to match other uverbs structs · d82693da
      由 Yann Droneaud 提交于 
      Commit 436f2ad0 ("IB/core: Export ib_create/destroy_flow through
uverbs") added public data structures to support receive flow
steering.  The new structs are not following the 'uverbs' pattern:
they're lacking the common prefix 'ib_uverbs'.

This patch replaces ib_kern prefix by ib_uverbs.
Signed-off-by: NYann Droneaud <ydroneaud@opteya.com>
Link: http://marc.info/?i=cover.1383773832.git.ydroneaud@opteya.comSigned-off-by: NRoland Dreier <roland@purestorage.com>
      d82693da
    • M
      IB/core: clarify overflow/underflow checks on ib_create/destroy_flow · f8848274
      由 Matan Barak 提交于 
      This patch fixes the following issues:

1. Unneeded checks were removed

2. Removed the fixed size out of flow_attr.size, thus simplifying the checks.

3. Remove a 32bit hole on 64bit systems with strict alignment in
   struct ib_kern_flow_att by adding a reserved field.
Signed-off-by: NMatan Barak <matanb@mellanox.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      f8848274
  11. 17 11月, 2013 2 次提交
  13. 16 11月, 2013 1 次提交
  15. 12 11月, 2013 2 次提交
    • M
      RDMA/cma: Remove unused argument and minor dead code · 352b9056
      由 Michal Nazarewicz 提交于 
      The dev variable is never assigned after being initialised.
Signed-off-by: NMichal Nazarewicz <mina86@mina86.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      352b9056
    • S
      RDMA/ucma: Discard events for IDs not yet claimed by user space · c6b21824
      由 Sean Hefty 提交于 
      Problem reported by Avneesh Pant <avneesh.pant@oracle.com>:

    It looks like we are triggering a bug in RDMA CM/UCM interaction.
    The bug specifically hits when we have an incoming connection
    request and the connecting process dies BEFORE the passive end of
    the connection can process the request i.e. it does not call
    rdma_get_cm_event() to retrieve the initial connection event.  We
    were able to triage this further and have some additional
    information now.

    In the example below when P1 dies after issuing a connect request
    as the CM id is being destroyed all outstanding connects (to P2)
    are sent a reject message. We see this reject message being
    received on the passive end and the appropriate CM ID created for
    the initial connection message being retrieved in cm_match_req().
    The problem is in the ucma_event_handler() code when this reject
    message is delivered to it and the initial connect message itself
    HAS NOT been delivered to the client. In fact the client has not
    even called rdma_cm_get_event() at this stage so we haven't
    allocated a new ctx in ucma_get_event() and updated the new
    connection CM_ID to point to the new UCMA context.

    This results in the reject message not being dropped in
    ucma_event_handler() for the new connection request as the
    (if (!ctx->uid)) block is skipped since the ctx it refers to is
    the listen CM id context which does have a valid UID associated
    with it (I believe the new CMID for the connection initially
    uses the listen CMID -> context when it is created in
    cma_new_conn_id). Thus the assumption that new events for a
    connection can get dropped in ucma_event_handler() is incorrect
    IF the initial connect request has not been retrieved in the
    first case. We end up getting a CM Reject event on the listen CM
    ID and our upper layer code asserts (in fact this event does not
    even have the listen_id set as that only gets set up librdmacm
    for connect requests).

The solution is to verify that the cm_id being reported in the event
is the same as the cm_id referenced by the ucma context.  A mismatch
indicates that the ucma context corresponds to the listen.  This fix
was validated by using a modified version of librdmacm that was able
to verify the problem and see that the reject message was indeed
dropped after this patch was applied.
Signed-off-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      c6b21824
  17. 09 11月, 2013 5 次提交
    • U
      IB/core: Add Cisco usNIC rdma node and transport types · 180771a3
      由 Upinder Malhi \(umalhi\) 提交于 
      This patch adds new rdma node and new rdma transport, and supporting
code used by Cisco's low latency driver called usNIC.  usNIC uses its
own transport, distinct from IB and iWARP.
Signed-off-by: NUpinder Malhi <umalhi@cisco.com>
Signed-off-by: NJeff Squyres <jsquyres@cisco.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      180771a3
    • M
      IB/netlink: Remove superfluous RDMA_NL_GET_OP() masking · 5476781b
      由 Mathias Krause 提交于 
      'op' is the already RDMA_NL_GET_OP() masked 'type'.  No need to mask it again.
Signed-off-by: NMathias Krause <minipli@googlemail.com>
Reviewed-by: NYann Droneaud <ydroneaud@opteya.com>
Acked-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      5476781b
    • L
      IB/core: Pass imm_data from ib_uverbs_send_wr to ib_send_wr correctly · 6b7d103c
      由 Latchesar Ionkov 提交于 
      Currently, we don't copy the immediate data from the userspace struct
to the kernel one when UD messages are being sent.

This patch makes sure that the immediate data is set correctly.
Signed-off-by: NLatchesar Ionkov <lucho@ionkov.net>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      6b7d103c
    • D
      IB/cma: Check for GID on listening device first · be9130cc
      由 Doug Ledford 提交于 
      As a simple optimization that should speed up the vast majority of
connect attemps on IB devices, when we are searching for the GID of an
incoming connection in the cached GID lists of devices, search the
device that received the incoming connection request first.  If we
don't find it there, then move on to other devices.

This reduces the time to perform 10,000 connections considerably.
Prior to this patch, a bad run of cmtime would look like this:

connect      :    12399.26   12351.10    8609.00    1239.93

With this patch, it looks more like this:

connect      :     5864.86    5799.80    8876.00     586.49
Signed-off-by: NDoug Ledford <dledford@redhat.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      be9130cc
    • D
      IB/cma: Use cached gids · 29f27e84
      由 Doug Ledford 提交于 
      The cma_acquire_dev function was changed by commit 3c86aa70
("RDMA/cm: Add RDMA CM support for IBoE devices") to use find_gid_port()
because multiport devices might have either IB or IBoE formatted gids.
The old function assumed that all ports on the same device used the
same GID format.

However, when it was changed to use find_gid_port(), we inadvertently
lost usage of the GID cache.  This turned out to be a very costly
change.  In our testing, each iteration through each index of the GID
table takes roughly 35us.  When you have multiple devices in a system,
and the GID you are looking for is on one of the later devices, the
code loops through all of the GID indexes on all of the early devices
before it finally succeeds on the target device.  This pathological
search behavior combined with 35us per GID table index retrieval
results in results such as the following from the cmtime application
that's part of the latest librdmacm git repo:

ib1:
step              total ms     max ms     min us  us / conn
create id    :       29.42       0.04       1.00       2.94
bind addr    :   186705.66      19.00   18556.00   18670.57
resolve addr :       41.93       9.68     619.00       4.19
resolve route:      486.93       0.48     101.00      48.69
create qp    :     4021.95       6.18     330.00     402.20
connect      :    68350.39   68588.17   24632.00    6835.04
disconnect   :     1460.43     252.65-1862269.00     146.04
destroy      :       41.16       0.04       2.00       4.12

ib0:
step              total ms     max ms     min us  us / conn
create id    :       28.61       0.68       1.00       2.86
bind addr    :     2178.86       2.95     201.00     217.89
resolve addr :       51.26      16.85     845.00       5.13
resolve route:      620.08       0.43      92.00      62.01
create qp    :     3344.40       6.36     273.00     334.44
connect      :     6435.99    6368.53    7844.00     643.60
disconnect   :     5095.38     321.90     757.00     509.54
destroy      :       37.13       0.02       2.00       3.71

Clearly, both the bind address and connect operations suffer
a huge penalty for being anything other than the default
GID on the first port in the system.

After applying this patch, the numbers now look like this:

ib1:
step              total ms     max ms     min us  us / conn
create id    :       30.15       0.03       1.00       3.01
bind addr    :       80.27       0.04       7.00       8.03
resolve addr :       43.02      13.53     589.00       4.30
resolve route:      482.90       0.45     100.00      48.29
create qp    :     3986.55       5.80     330.00     398.66
connect      :     7141.53    7051.29    5005.00     714.15
disconnect   :     5038.85     193.63     918.00     503.88
destroy      :       37.02       0.04       2.00       3.70

ib0:
step              total ms     max ms     min us  us / conn
create id    :       34.27       0.05       1.00       3.43
bind addr    :       26.45       0.04       1.00       2.64
resolve addr :       38.25      10.54     760.00       3.82
resolve route:      604.79       0.43      97.00      60.48
create qp    :     3314.95       6.34     273.00     331.49
connect      :    12399.26   12351.10    8609.00    1239.93
disconnect   :     5096.76     270.72    1015.00     509.68
destroy      :       37.10       0.03       2.00       3.71

It's worth noting that we still suffer a bit of a penalty on
connect to the wrong device, but the penalty is much less than
it used to be.  Follow on patches deal with this penalty.

Many thanks to Neil Horman for helping to track the source of
slow function that allowed us to track down the fact that
the original patch I mentioned above backed out cache usage
and identify just how much that impacted the system.
Signed-off-by: NDoug Ledford <dledford@redhat.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      29f27e84
  19. 08 11月, 2013 1 次提交
  21. 22 10月, 2013 1 次提交
  23. 01 10月, 2013 1 次提交
  25. 03 9月, 2013 1 次提交
  27. 29 8月, 2013 3 次提交
    • H
      IB/core: Export ib_create/destroy_flow through uverbs · 436f2ad0
      由 Hadar Hen Zion 提交于 
      Implement ib_uverbs_create_flow() and ib_uverbs_destroy_flow() to
support flow steering for user space applications.
Signed-off-by: NHadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      436f2ad0
    • I
      IB/core: Infrastructure for extensible uverbs commands · 400dbc96
      由 Igor Ivanov 提交于 
      Add infrastructure to support extended uverbs capabilities in a
forward/backward manner.  Uverbs command opcodes which are based on
the verbs extensions approach should be greater or equal to
IB_USER_VERBS_CMD_THRESHOLD.  They have new header format and
processed a bit differently.

Whenever a specific IB_USER_VERBS_CMD_XXX is extended, which practically means
it needs to have additional arguments, we will be able to add them without creating
a completely new IB_USER_VERBS_CMD_YYY command or bumping the uverbs ABI version.

This patch for itself doesn't provide the whole scheme which is also dependent
on adding a comp_mask field to each extended uverbs command struct.

The new header framework allows for future extension of the CMD arguments
(ib_uverbs_cmd_hdr.in_words, ib_uverbs_cmd_hdr.out_words) for an existing
new command (that is a command that supports the new uverbs command header format
suggested in this patch) w/o bumping ABI version and with maintaining backward
and formward compatibility to new and old libibverbs versions.

In the uverbs command we are passing both uverbs arguments and the provider arguments.
We split the ib_uverbs_cmd_hdr.in_words to ib_uverbs_cmd_hdr.in_words which will now carry only
uverbs input argument struct size and  ib_uverbs_cmd_hdr.provider_in_words that will carry
the provider input argument size. Same goes for the response (the uverbs CMD output argument).

For example take the create_cq call and the mlx4_ib provider:

The uverbs layer gets libibverb's struct ibv_create_cq (named struct ib_uverbs_create_cq
in the kernel), mlx4_ib gets libmlx4's struct mlx4_create_cq (which includes struct
ibv_create_cq and is named struct mlx4_ib_create_cq in the kernel) and
in_words = sizeof(mlx4_create_cq)/4 .

Thus ib_uverbs_cmd_hdr.in_words carry both uverbs plus mlx4_ib input argument sizes,
where uverbs assumes it knows the size of its input argument - struct ibv_create_cq.

Now, if we wish to add a variable to struct ibv_create_cq, we can add a comp_mask field
to the struct which is basically bit field indicating which fields exists in the struct
(as done for the libibverbs API extension), but we need a way to tell what is the total
size of the struct and not assume the struct size is predefined (since we may get different
struct sizes from different user libibverbs versions). So we know at which point the
provider input argument (struct mlx4_create_cq) begins. Same goes for extending the
provider struct mlx4_create_cq. Thus we split the ib_uverbs_cmd_hdr.in_words to
ib_uverbs_cmd_hdr.in_words which will now carry only uverbs input argument struct size and
ib_uverbs_cmd_hdr.provider_in_words that will carry the provider (mlx4_ib) input argument size.
Signed-off-by: NIgor Ivanov <Igor.Ivanov@itseez.com>
Signed-off-by: NHadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      400dbc96
    • H
      IB/core: Add receive flow steering support · 319a441d
      由 Hadar Hen Zion 提交于 
      The RDMA stack allows for applications to create IB_QPT_RAW_PACKET
QPs, which receive plain Ethernet packets, specifically packets that
don't carry any QPN to be matched by the receiving side.  Applications
using these QPs must be provided with a method to program some
steering rule with the HW so packets arriving at the local port can be
routed to them.

This patch adds ib_create_flow(), which allow providing a flow
specification for a QP.  When there's a match between the
specification and a received packet, the packet is forwarded to that
QP, in a the same way one uses ib_attach_multicast() for IB UD
multicast handling.

Flow specifications are provided as instances of struct ib_flow_spec_yyy,
which describe L2, L3 and L4 headers.  Currently specs for Ethernet, IPv4,
TCP and UDP are defined.  Flow specs are made of values and masks.

The input to ib_create_flow() is a struct ib_flow_attr, which contains
a few mandatory control elements and optional flow specs.

    struct ib_flow_attr {
            enum ib_flow_attr_type type;
            u16      size;
            u16      priority;
            u32      flags;
            u8       num_of_specs;
            u8       port;
            /* Following are the optional layers according to user request
             * struct ib_flow_spec_yyy
             * struct ib_flow_spec_zzz
             */
    };

As these specs are eventually coming from user space, they are defined and
used in a way which allows adding new spec types without kernel/user ABI
change, just with a little API enhancement which defines the newly added spec.

The flow spec structures are defined with TLV (Type-Length-Value)
entries, which allows calling ib_create_flow() with a list of variable
length of optional specs.

For the actual processing of ib_flow_attr the driver uses the number
of specs and the size mandatory fields along with the TLV nature of
the specs.

Steering rules processing order is according to the domain over which
the rule is set and the rule priority.  All rules set by user space
applicatations fall into the IB_FLOW_DOMAIN_USER domain, other domains
could be used by future IPoIB RFS and Ethetool flow-steering interface
implementation.  Lower numerical value for the priority field means
higher priority.

The returned value from ib_create_flow() is a struct ib_flow, which
contains a database pointer (handle) provided by the HW driver to be
used when calling ib_destroy_flow().

Applications that offload TCP/IP traffic can also be written over IB
UD QPs.  The ib_create_flow() / ib_destroy_flow() API is designed to
support UD QPs too.  A HW driver can set IB_DEVICE_MANAGED_FLOW_STEERING
to denote support for flow steering.

The ib_flow_attr enum type supports usage of flow steering for promiscuous
and sniffer purposes:

    IB_FLOW_ATTR_NORMAL - "regular" rule, steering according to rule specification

    IB_FLOW_ATTR_ALL_DEFAULT - default unicast and multicast rule, receive
        all Ethernet traffic which isn't steered to any QP

    IB_FLOW_ATTR_MC_DEFAULT - same as IB_FLOW_ATTR_ALL_DEFAULT but only for multicast

    IB_FLOW_ATTR_SNIFFER - sniffer rule, receive all port traffic

ALL_DEFAULT and MC_DEFAULT rules options are valid only for Ethernet link type.
Signed-off-by: NHadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      319a441d
  29. 14 8月, 2013 2 次提交
  31. 13 8月, 2013 1 次提交
  33. 01 8月, 2013 1 次提交
    • J
      IB/core: Create QP1 using the pkey index which contains the default pkey · ef5ed416
      由 Jack Morgenstein 提交于 
      Currently, QP1 is created using pkey index 0. This patch simply looks
for the index containing the default pkey, rather than hard-coding
pkey index 0.

This change will have no effect in native mode, since QP0 and QP1 are
created before the SM configures the port, so pkey table will still be
the default table defined by the IB Spec, in C10-123: "If non-volatile
storage is not used to hold P_Key Table contents, then if a PM
(Partition Manager) is not present, and prior to PM initialization of
the P_Key Table, the P_Key Table must act as if it contains a single
valid entry, at P_Key_ix = 0, containing the default partition
key. All other entries in the P_Key Table must be invalid."

Thus, in the native mode case, the driver will find the default pkey
at index 0 (so it will be no different than the hard-coding).

However, in SR-IOV mode, for VFs, the pkey table may be
paravirtualized, so that the VF's pkey index zero may not necessarily
be mapped to the real pkey index 0. For VFs, therefore, it is
important to find the virtual index which maps to the real default
pkey.

This commit does the following for QP1 creation:

1. Find the pkey index containing the default pkey, and use that index
   if found.  ib_find_pkey() returns the index of the
   limited-membership default pkey (0x7FFF) if the full-member default
   pkey is not in the table.

2. If neither form of the default pkey is found, use pkey index 0
   (previous behavior).
Signed-off-by: NJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: NSean Hefty <sean.hefty@intel.com>
Signed-off-by: NRoland Dreier <roland@purestorage.com>
      ef5ed416
  35. 31 7月, 2013 3 次提交