This commit allows setting the RSS hash generation parameters from
ethtool. When setting parameters for a given flow type from ethtool
(e.g. tcp4), all the corresponding flows in the flow table are updated,
according to the supported hash parameters.

For example, when configuring TCP over IPv4 hash parameters to be
src/dst IP  + src/dst port ("ethtool -N eth0 rx-flow-hash tcp4 sdfn"),
we only set the "src/dst port" hash parameters on the non-fragmented TCP
over IPv4 flows.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

One of the classification action that can be performed is to compute a
hash of the packet header based on some header fields, and lookup a RSS
table based on this hash to determine the final RxQ.

This is done by adding one lookup entry per flow per port, so that we
can configure the hash generation parameters for each flow and each
port.

There are 2 possible engines that can be used for RSS hash generation :

 - C3HA, that generates a hash based on up to 4 header-extracted fields
 - C3HB, that does the same as c3HA, but also includes L4 info in the hash

There are a lot of fields that can be extracted from the header. For now,
we only use the ones that we can configure using ethtool :
 - DST MAC address
 - L3 info
 - Source IP
 - Destination IP
 - Source port
 - Destination port

The C3HB engine is selected when we use L4 fields (src/dst port).

               Header parser          Dec table
 Ingress pkt  +-------------+ flow id +----------------------------+
------------->| TCAM + SRAM |-------->|TCP IPv4 w/ VLAN, not frag  |
              +-------------+         |TCP IPv4 w/o VLAN, not frag |
                                      |TCP IPv4 w/ VLAN, frag      |--+
                                      |etc.                        |  |
                                      +----------------------------+  |
                                                                      |
                                            Flow table                |
  +---------+   +------------+         +--------------------------+   |
  | RSS tbl |<--| Classifier |<--------| flow 0: C2 lookup        |   |
  +---------+   +------------+         |         C3 lookup port 0 |   |
                 |         |           |         C3 lookup port 1 |   |
         +-----------+ +-------------+ |         ...              |   |
         | C2 engine | | C3H engines | | flow 1: C2 lookup        |<--+
         +-----------+ +-------------+ |         C3 lookup port 0 |
                                       |         ...              |
                                       | ...                      |
                                       | flow 51 : C2 lookup      |
                                       |           ...            |
                                       +--------------------------+

The C2 engine also gains the role of enabling and disabling the RSS
table lookup for this packet.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The PPv2 classifier allows to perform classification operations on each
ingress packet, based on the flow the packet is assigned to.

The current code uses only 1 flow per port, and the only classification
action consists of assigning the rx queue to the packet, depending on the
port.

In preparation for adding RSS support, we have to split all incoming
traffic into different flows. Since RSS assigns a rx queue depending on
the hash of some header fields, we have to make sure that the hash is
generated in a consistent way for all packets in the same flow.

What we call a "flow" is actually a set of attributes attached to a
packet that depends on various L2/L3/L4 info.

This patch introduces 52 flows, wich are a combination of various L2, L3
and L4 attributes :
 - Whether or not the packet has a VLAN tag
 - Whether the packet is IPv4, IPv6 or something else
 - Whether the packet is TCP, UDP or something else
 - Whether or not the packet is fragmented at L3 level.

The flow is associated to a packet by the Header Parser. Each flow
corresponds to an entry in the decoding table. This entry then points to
the sequence of classification lookups to be performed by the
classifier, represented in the flow table.

For now, the only lookup we perform is a C2 lookup to set the default
rx queue.

               Header parser          Dec table
 Ingress pkt  +-------------+ flow id +----------------------------+
------------->| TCAM + SRAM |-------->|TCP IPv4 w/ VLAN, not frag  |
              +-------------+         |TCP IPv4 w/o VLAN, not frag |
                                      |TCP IPv4 w/ VLAN, frag      |--+
                                      |etc.                        |  |
                                      +----------------------------+  |
                                                                      |
                                           Flow table                 |
                +------------+        +---------------------+         |
     To RxQ <---| Classifier |<-------| flow 0: C2 lookup   |<--------+
                +------------+        | flow 1: C2 lookup   |
                       |              | ...                 |
                +------------+        | flow 51 : C2 lookup |
		| C2 engine  |        +---------------------+
                +------------+
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The PPv2 Controller has a classifier, that can perform multiple lookup
operations for each packet, using different engines.

One of these engines is the C2 engine, which performs TCAM based lookups
on data extracted from the packet header. When a packet matches an
entry, the engine sets various attributes, used to perform
classification operations.

One of these attributes is the rx queue in which the packet should be sent.
The current code uses the lookup_id table (also called decoding table)
to assign the rx queue. However, this only works if we use one entry per
port in the decoding table, which won't be the case once we add RSS
lookups.

This patch uses the C2 engine to assign the rx queue to each packet.

The C2 engine is used through the flow table, which dictates what
classification operations are done for a given flow.

Right now, we have one flow per port, which contains every ingress
packet for this port.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

mvpp22_init_rss function configures the RSS parameters for each port, so
rename it accordingly. Since this function relies on classifier
configuration, move its call right after the classifier config.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When filling the RSS table, we have to make sure that the rx queue is
attached to an online CPU.

This patch is not a full support for cpu_hotplug, but rather a way to
make sure that we don't break network on system booted with the maxcpus
parameter.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds an extra indirection when setting the indirection table
into the RSS hardware table to improve the packets distribution across
CPUs. For example, if 2 queues are used on a multi-core system this new
indirection will choose two queues on two different CPUs instead of the
two first queues which are on the same first CPU.
Signed-off-by: NAntoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds the RSS indirection table support, allowing to use the
ethtool -x and -X options to dump and set this table.
Signed-off-by: NAntoine Tenart <antoine.tenart@bootlin.com>
[Maxime: Small warning fixes, use one table per port]
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

PPv2 Controller has 8 RSS Tables, of 32 entries each. A lookup in the
RXQ2RSS_TABLE is performed for each incoming packet, and the RSS Table
to be used is chosen according to the default rx queue that would be
used for the packet.

This default rx queue is set in the Lookup_id Table (also called
Decoding Table), and is equal to the port->first_rxq.

Since the Classifier itself isn't active at any time for the moment,
this doesn't have a direct effect, the default rx queue at the moment is
the one where all packets end-up into.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is no RSS_TABLE register in PPv2 Controller. The register 0x1510
which was specified is actually named "RSS_HASH_SEL", but isn't used by
this driver at all.

Based on how this register was used, it should have been the
RXQ2RSS_TABLE register, which allows to select the RSS table that will
be used for the incoming packet.

The RSS_TABLE_POINTER is actually a field of this RXQ2RSS_TABLE
register.

Since RSS tables are actually not used by the driver for now, this
commit does not fix a runtime bug.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Cosmetic patch fixing a typo in one of the RSS comments.
Signed-off-by: NAntoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The number of receive queue per port is :
 - MVPP2_DEFAULT_RXQ if in single queue mode
 - MVPP2_DEFAULT_RXQ * num_possible_cpus if in multi queue mode

with MVPP2_DEFAULT_RXQ = 4.

However, we don't use the extra rx queues at the moment, we really only
need one per port per CPU, until some more advanced classification rules
are implemented.
Suggested-by: NStefan Chulski <stefanc@marvell.com>
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There's a dedicated #define that indicates the number of rx queues per
port per cpu, this commit removes a harcoded use of that value

This doesn't fix any runtime bugs since the harcoded value matches the
expected value.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since RSS only applies when we have per-cpu rx queues, it should only
be enabled when the driver is configured to make use of multi-queue
mode.
Signed-off-by: NYan Markman <ymarkman@marvell.com>
[Maxime: Commit message]
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The multi queue mode is needed to have RSS available, and offers some
nice advantages, being able to have one rx queue vector per CPU.

This mode has been usable through the use of a module parameter, this
commit makes it the default value.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The PPv2 driver defines 2 "queue_modes" :
 - QDIST_SINGLE_MODE, where each port share one rx queue vector
   between all CPUs
 - QDIST_MULTI_MODE, where each port has one rx queue vector per CPU.

Multi queue mode isn't available on PPv2.1, make sure we fallback to
single mode when running on this revision.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The size of the the RSS indirection tables should be defined in mvpp2.h,
so that we can use it in all files of the PPv2 driver.

This commit moves the define in mvpp2.h, and adds the missing #include
in mvpp2_cls.h.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Include guards should be put before #includes. This doesn't fix any bug,
but prevent future compilation issues when adding new files in the mvpp2
driver

The Header Parser init function needs the platform_device definition,
and with the fixed include guards we need to add the missing include.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Move the xmit of offload-eligible (ie IPv4) traffic on OSA over to the
new, copy-free path.
As with L2, we'll need to preserve the skb_orphan() behaviour of the
old code path until TX completion is sufficiently fast.
Signed-off-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This implements a new xmit path for L3 HiperSockets, which carves the
HW header from skb headroom instead of allocating it from the hdr cache.
It also adds NETIF_F_SG support.

The delta in qeth_l3_xmit() is all just removal of IQD-specific code and
some minor consolidation.
Signed-off-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In preparation for future work, move the high-level xmit work into a
separate wrapper. This matches the L2 xmit code.
Signed-off-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When a L3 device doesn't offer TSO, allow the stack to build full-size
GSO skbs.
Signed-off-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove some redundant EXPORTs. While at it, also move some L2-only
prototypes into the proper header file.
Signed-off-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Reshuffle the code a bit so that everything is in one place.
Signed-off-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Consolidate duplicated code, fix the misuse of RTN_UNSPEC and simplify
the handling of non-unicast traffic on IQD devices.
Signed-off-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Changing a device's address lists (or its promisc mode) already triggers
an RX modeset, there's no need to do it manually from the L2 driver's
ndo_vlan_rx_kill_vid() hook.

Also when setting a device online, dev_open() already calls
dev_set_rx_mode(). So a manual modeset is only necessary from the
recovery path.
Signed-off-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Except for tracing, the pointer is not used.

At the same time, accessing it from qeth_qdio_output_handler() is racy:
whenever qeth_qdio_cq_handler() gets control, its call to
qeth_qdio_handle_aob() frees the AOB.

So the AOB pointer that qeth_qdio_output_handler() stores into 'buffer'
can go stale at any time, and trigger a use-after-free.
Signed-off-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use the new qeth_scrub_qdio_buffer() helper, remove an extra parameter
from qeth_clear_output_buffer(), init the bufstates.user field just once
(in qeth_flush_buffers()) and remove some noisy trace messages.
Signed-off-by: NJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

getnstimeofday64 is deprecated in favor of the ktime_get() family of
functions. The direct replacement would be ktime_get_real_ts64(),
but I'm picking the basic ktime_get() instead:

- using a ktime_t simplifies the code compared to timespec64
- using monotonic time instead of real time avoids issues caused
  by a concurrent settimeofday() or during a leap second adjustment.
Acked-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The two do the same thing, but we want to have a consistent
naming in the kernel.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NFelix Manlunas <felix.manlunas@cavium.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Using get_seconds() for timestamps is deprecated since it can lead
to overflows on 32-bit systems. While the interface generally doesn't
overflow until year 2106, the specific implementation of the TCP PAWS
algorithm breaks in 2038 when the intermediate signed 32-bit timestamps
overflow.

A related problem is that the local timestamps in CLOCK_REALTIME form
lead to unexpected behavior when settimeofday is called to set the system
clock backwards or forwards by more than 24 days.

While the first problem could be solved by using an overflow-safe method
of comparing the timestamps, a nicer solution is to use a monotonic
clocksource with ktime_get_seconds() that simply doesn't overflow (at
least not until 136 years after boot) and that doesn't change during
settimeofday().

To make 32-bit and 64-bit architectures behave the same way here, and
also save a few bytes in the tcp_options_received structure, I'm changing
the type to a 32-bit integer, which is now safe on all architectures.

Finally, the ts_recent_stamp field also (confusingly) gets used to store
a jiffies value in tcp_synq_overflow()/tcp_synq_no_recent_overflow().
This is currently safe, but changing the type to 32-bit requires
some small changes there to keep it working.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current position of .rss_flags field in struct rss_info causes
that fields .rsstable and .rssqueue (both 128 bytes long) crosses
cache-line boundaries. Moving it at the end properly align all fields.

Before patch:
struct rss_info {
        u64                        rss_flags;            /*     0     8 */
        u8                         rsstable[128];        /*     8   128 */
        /* --- cacheline 2 boundary (128 bytes) was 8 bytes ago --- */
        u8                         rss_queue[128];       /*   136   128 */
        /* --- cacheline 4 boundary (256 bytes) was 8 bytes ago --- */
        u8                         rss_hkey[40];         /*   264    40 */
};

After patch:
struct rss_info {
        u8                         rsstable[128];        /*     0   128 */
        /* --- cacheline 2 boundary (128 bytes) --- */
        u8                         rss_queue[128];       /*   128   128 */
        /* --- cacheline 4 boundary (256 bytes) --- */
        u8                         rss_hkey[40];         /*   256    40 */
        u64                        rss_flags;            /*   296     8 */
};
Signed-off-by: NIvan Vecera <cera@cera.cz>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- Unionize two u8 fields where only one of them is used depending on NIC
chipset.
- Move recovery_supported field after that union

These changes eliminate 7-bytes hole in the struct and makes it smaller
by 8 bytes.
Signed-off-by: NIvan Vecera <cera@cera.cz>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NIvan Vecera <cera@cera.cz>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Before patch:
struct be_tx_obj {
        u32                        db_offset;            /*     0     4 */

        /* XXX 4 bytes hole, try to pack */

        struct be_queue_info       q;                    /*     8    56 */
        /* --- cacheline 1 boundary (64 bytes) --- */
        struct be_queue_info       cq;                   /*    64    56 */
        struct be_tx_compl_info    txcp;                 /*   120     4 */

        /* XXX 4 bytes hole, try to pack */

        /* --- cacheline 2 boundary (128 bytes) --- */
        struct sk_buff *           sent_skb_list[2048];  /*   128 16384 */
        ...
}:

After patch:
struct be_tx_obj {
        u32                        db_offset;            /*     0     4 */
        struct be_tx_compl_info    txcp;                 /*     4     4 */
        struct be_queue_info       q;                    /*     8    56 */
        /* --- cacheline 1 boundary (64 bytes) --- */
        struct be_queue_info       cq;                   /*    64    56 */
        struct sk_buff *           sent_skb_list[2048];  /*   120 16384 */
        ...
};
Signed-off-by: NIvan Vecera <cera@cera.cz>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Re-order fields in struct be_eq_obj to ensure that .napi field begins
at start of cache-line. Also the .adapter field is moved to the first
cache-line next to .q field and 3 fields (idx,msi_idx,spurious_intr)
and the 4-bytes hole to 3rd cache-line.
Signed-off-by: NIvan Vecera <cera@cera.cz>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The event queue description (be_eq_obj.desc) field is used only to format
string for IRQ name and it is not really needed to hold this value.
Remove it and use local variable to format string for IRQ name.
Signed-off-by: NIvan Vecera <cera@cera.cz>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The commit fb6113e6 ("be2net: get rid of custom busy poll code")
replaced custom busy-poll code by the generic one but left several
macros and fields in struct be_eq_obj that are currently unused.
Remove this stuff.

Fixes: fb6113e6 ("be2net: get rid of custom busy poll code")
Signed-off-by: NIvan Vecera <cera@cera.cz>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The commit 2632bafd ("be2net: fix adaptive interrupt coalescing")
introduced a separate struct be_aic_obj to hold AIC information but
unfortunately left the old stuff in be_eq_obj. So remove it.

Fixes: 2632bafd ("be2net: fix adaptive interrupt coalescing")
Signed-off-by: NIvan Vecera <cera@cera.cz>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The late ts queue can contain a bunch of skbs while hi rate testing,
no need to check all of them if timestamp is already matched.
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>