Skip to content
GitLab
  1. Jan 19, 2017
    • Josef Bacik's avatar
      inet: reset tb->fastreuseport when adding a reuseport sk · 637bc8bb
      Josef Bacik authored 
      

If we have non reuseport sockets on a tb we will set tb->fastreuseport to 0 and
never set it again.  Which means that in the future if we end up adding a bunch
of reuseport sk's to that tb we'll have to do the expensive scan every time.
Instead add the ipv4/ipv6 saddr fields to the bind bucket, as well as the family
so we know what comparison to make, and the ipv6 only setting so we can make
sure to compare with new sockets appropriately.  Once one sk has made it onto
the list we know that there are no potential bind conflicts on the owners list
that match that sk's rcv_addr.  So copy the sk's information into our bind
bucket and set tb->fastruseport to FASTREUSESOCK_STRICT so we know we have to do
an extra check for subsequent reuseport sockets and skip the expensive bind
conflict check.

Signed-off-by: default avatarJosef Bacik <jbacik@fb.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      637bc8bb
    • Josef Bacik's avatar
      inet: split inet_csk_get_port into two functions · 289141b7
      Josef Bacik authored 
      

inet_csk_get_port does two different things, it either scans for an open port,
or it tries to see if the specified port is available for use.  Since these two
operations have different rules and are basically independent lets split them
into two different functions to make them both more readable.

Signed-off-by: default avatarJosef Bacik <jbacik@fb.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      289141b7
    • Josef Bacik's avatar
      inet: don't check for bind conflicts twice when searching for a port · 6cd66616
      Josef Bacik authored 
      

This is just wasted time, we've already found a tb that doesn't have a bind
conflict, and we don't drop the head lock so scanning again isn't going to give
us a different answer.  Instead move the tb->reuse setting logic outside of the
found_tb path and put it in the success: path.  Then make it so that we don't
goto again if we find a bind conflict in the found_tb path as we won't reach
this anymore when we are scanning for an ephemeral port.

Signed-off-by: default avatarJosef Bacik <jbacik@fb.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      6cd66616
    • Josef Bacik's avatar
      inet: kill smallest_size and smallest_port · b9470c27
      Josef Bacik authored 
      

In inet_csk_get_port we seem to be using smallest_port to figure out where the
best place to look for a SO_REUSEPORT sk that matches with an existing set of
SO_REUSEPORT's.  However if we get to the logic

if (smallest_size != -1) {
	port = smallest_port;
	goto have_port;
}

we will do a useless search, because we would have already done the
inet_csk_bind_conflict for that port and it would have returned 1, otherwise we
would have gone to found_tb and succeeded.  Since this logic makes us do yet
another trip through inet_csk_bind_conflict for a port we know won't work just
delete this code and save us the time.

Signed-off-by: default avatarJosef Bacik <jbacik@fb.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      b9470c27
    • Josef Bacik's avatar
      inet: drop ->bind_conflict · aa078842
      Josef Bacik authored 
      

The only difference between inet6_csk_bind_conflict and inet_csk_bind_conflict
is how they check the rcv_saddr, so delete this call back and simply
change inet_csk_bind_conflict to call inet_rcv_saddr_equal.

Signed-off-by: default avatarJosef Bacik <jbacik@fb.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      aa078842
    • Josef Bacik's avatar
      inet: collapse ipv4/v6 rcv_saddr_equal functions into one · fe38d2a1
      Josef Bacik authored 
      

We pass these per-protocol equal functions around in various places, but
we can just have one function that checks the sk->sk_family and then do
the right comparison function.  I've also changed the ipv4 version to
not cast to inet_sock since it is unneeded.

Signed-off-by: default avatarJosef Bacik <jbacik@fb.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      fe38d2a1
    • jpinto's avatar
      stmicro: add more information to Kconfig · ab70e586
      jpinto authored 
      This patch adds more info to stmicro' Kconfig files in order to be clearer
that the driver can be used by ethernet cards based on 10/100/1000/EQOS
Synopsys IP Cores.

EQOS was also added stmmac/Kconfig Kconfig, since dwmac4 is in fact EQoS,
one of Synopsys Ethernet IPs. More info at:
https://www.synopsys.com/dw/ipdir.php?ds=dwc_ether_qos



Signed-off-by: default avatarJoao Pinto <jpinto@synopsys.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      ab70e586
    • Martin KaFai Lau's avatar
      net/mlx5e: Support bpf_xdp_adjust_head() · d8bec2b2
      Martin KaFai Lau authored 
      

This patch adds bpf_xdp_adjust_head() support to mlx5e.

1. rx_headroom is added to struct mlx5e_rq.  It uses
   an existing 4 byte hole in the struct.
2. The adjusted data length is checked against
   MLX5E_XDP_MIN_INLINE and MLX5E_SW2HW_MTU(rq->netdev->mtu).
3. The macro MLX5E_SW2HW_MTU is moved from en_main.c to en.h.
   MLX5E_HW2SW_MTU is also moved to en.h for symmetric reason
   but it is not a must.

v2:
- Keep the xdp specific logic in mlx5e_xdp_handle()
- Update dma_len after the sanity checks in mlx5e_xmit_xdp_frame()

Signed-off-by: default avatarMartin KaFai Lau <kafai@fb.com>
Acked-by: default avatarSaeed Mahameed <saeedm@mellanox.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      d8bec2b2
  3. Jan 18, 2017
  5. Jan 17, 2017
    • David S. Miller's avatar
      Merge branch 'mvneta-xmit_more-bql' · b8128c42
      David S. Miller authored 
      

Marcin Wojtas says:

====================
mvneta xmit_more and bql support

This is a delayed v2 of short patchset, which introduces xmit_more and BQL
to mvneta driver. The only one change was added in xmit_more support -
condition check preventing excessive descriptors concatenation before
flushing in HW.

Any comments or feedback would be welcome.

Changelog:
v1 -> v2:

* Add checking condition that ensures too much descriptors are not
  concatenated before flushing in HW.
====================

Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      b8128c42
    • Marcin Wojtas's avatar
      net: mvneta: add BQL support · a29b6235
      Marcin Wojtas authored 
      

Tests showed that when whole bandwidth is consumed, the latency for
various kind of traffic can reach high values. With saturated
link (e.g. with iperf from target to host) simple ping could take
significant amount of time. BQL proved to improve this situation
when implemented in mvneta driver. Measurements of ping latency
for 3 link speeds:
Speed | Latency w/o BQL | Latency with BQL
10    |      7-14 ms    |     3.5 ms
100   |      2-12 ms    |     0.6 ms
1000  |   often timeout |   up to 2ms

Decreasing latency as above result in sligt performance cost - 4kpps
(-1.4%) when pushing 64B packets via two bridged interfaces of Armada 38x.
For 1500B packets in the same setup, the mpstat tool showed +8% of
CPU occupation (default affinity, second CPU idle). Even though this
cost seems reasonable to take, considering other improvements.

This commit adds byte queue limit mechanism for the mvneta driver.

Signed-off-by: default avatarMarcin Wojtas <mw@semihalf.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      a29b6235
    • Simon Guinot's avatar
      net: mvneta: add xmit_more support · 2a90f7e1
      Simon Guinot authored 
      

Basing on xmit_more flag of the skb, TX descriptors can be concatenated
before flushing. This commit delay Tx descriptor flush if the queue is
running and if there is more skb's to send.

A maximum allowed number of descriptors for flushing at once due to
MVNETA_TXQ_UPDATE_REG(q) reqisters limitation, is 255. Because of that
a new macro was added (MVNETA_TXQ_DEC_SENT_MASK) in order to ensure that
concatenated amount of descriptor does not exceed that value.

Signed-off-by: default avatarSimon Guinot <simon.guinot@sequanux.org>
Signed-off-by: default avatarMarcin Wojtas <mw@semihalf.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      2a90f7e1
    • Jamal Hadi Salim's avatar
      net sched actions: fix refcnt when GETing of action after bind · 0faa9cb5
      Jamal Hadi Salim authored 
      Demonstrating the issue:

.. add a drop action
$sudo $TC actions add action drop index 10

.. retrieve it
$ sudo $TC -s actions get action gact index 10

	action order 1: gact action drop
	 random type none pass val 0
	 index 10 ref 2 bind 0 installed 29 sec used 29 sec
 	Action statistics:
	Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
	backlog 0b 0p requeues 0

... bug 1 above: reference is two.
    Reference is actually 1 but we forget to subtract 1.

... do a GET again and we see the same issue
    try a few times and nothing changes
~$ sudo $TC -s actions get action gact index 10

	action order 1: gact action drop
	 random type none pass val 0
	 index 10 ref 2 bind 0 installed 31 sec used 31 sec
 	Action statistics:
	Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
	backlog 0b 0p requeues 0

... lets try to bind the action to a filter..
$ sudo $TC qdisc add dev lo ingress
$ sudo $TC filter add dev lo parent ffff: protocol ip prio 1 \
  u32 match ip dst 127.0.0.1/32 flowid 1:1 action gact index 10

... and now a few GETs:
$ sudo $TC -s actions get action gact index 10

	action order 1: gact action drop
	 random type none pass val 0
	 index 10 ref 3 bind 1 installed 204 sec used 204 sec
 	Action statistics:
	Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
	backlog 0b 0p requeues 0

$ sudo $TC -s actions get action gact index 10

	action order 1: gact action drop
	 random type none pass val 0
	 index 10 ref 4 bind 1 installed 206 sec used 206 sec
 	Action statistics:
	Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
	backlog 0b 0p requeues 0

$ sudo $TC -s actions get action gact index 10

	action order 1: gact action drop
	 random type none pass val 0
	 index 10 ref 5 bind 1 installed 235 sec used 235 sec
 	Action statistics:
	Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
	backlog 0b 0p requeues 0

.... as can be observed the reference count keeps going up.

After the fix

$ sudo $TC actions add action drop index 10
$ sudo $TC -s actions get action gact index 10

	action order 1: gact action drop
	 random type none pass val 0
	 index 10 ref 1 bind 0 installed 4 sec used 4 sec
 	Action statistics:
	Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
	backlog 0b 0p requeues 0

$ sudo $TC -s actions get action gact index 10

	action order 1: gact action drop
	 random type none pass val 0
	 index 10 ref 1 bind 0 installed 6 sec used 6 sec
 	Action statistics:
	Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
	backlog 0b 0p requeues 0

$ sudo $TC qdisc add dev lo ingress
$ sudo $TC filter add dev lo parent ffff: protocol ip prio 1 \
  u32 match ip dst 127.0.0.1/32 flowid 1:1 action gact index 10

$ sudo $TC -s actions get action gact index 10

	action order 1: gact action drop
	 random type none pass val 0
	 index 10 ref 2 bind 1 installed 32 sec used 32 sec
 	Action statistics:
	Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
	backlog 0b 0p requeues 0

$ sudo $TC -s actions get action gact index 10

	action order 1: gact action drop
	 random type none pass val 0
	 index 10 ref 2 bind 1 installed 33 sec used 33 sec
 	Action statistics:
	Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
	backlog 0b 0p requeues 0

Fixes: aecc5cef

 ("net sched actions: fix GETing actions")
Signed-off-by: default avatarJamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      0faa9cb5
    • Linus Torvalds's avatar
      Merge tag 'nfs-for-4.10-3' of git://git.linux-nfs.org/projects/trondmy/linux-nfs · 5cf7a0f3
      Linus Torvalds authored 
      Pull NFS client bugfixes from Trond Myklebust:

 - fix invalid fget()/fput() calls when doing file locking

 - fix multiple directory cache invalidation issues due to the client
   failing to recognise that the directory wasn't changed

 - fix client recovery when server reboots multiple times

* tag 'nfs-for-4.10-3' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
  NFSv4: Fix client recovery when server reboots multiple times
  NFSv4: update_changeattr should update the attribute timestamp
  NFSv4: Don't call update_changeattr() unless the unlink is successful
  NFSv4: Don't apply change_info4 twice on rename within a directory
  NFSv4: Call update_changeattr() from _nfs4_proc_open only if a file was created
  nfs: Don't take a reference on fl->fl_file for LOCK operation
      5cf7a0f3
    • David S. Miller's avatar
      Merge branch 'mlx4-core-fixes' · 617125e7
      David S. Miller authored 
      Tariq Toukan says:

====================
mlx4 core fixes

This patchset contains bug fixes from Jack to the mlx4 Core driver.

Patch 1 solves a race in the flow of CQ free.
Patch 2 moves some qp context flags update to the correct qp transition.
Patch 3 eliminates warnings from the path of SRQ_LIMIT that flood the message log,
and keeps them only in the path of SRQ_CATAS_ERROR.

Series generated against net commit:
1a717fcf Merge tag 'mac80211-for-davem-2017-01-13' of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211


====================

Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      617125e7
    • Jack Morgenstein's avatar
      net/mlx4_core: Eliminate warning messages for SRQ_LIMIT under SRIOV · 9577b174
      Jack Morgenstein authored 
      When running SRIOV, warnings for SRQ LIMIT events flood the Hypervisor's
message log when (correct, normally operating) apps use SRQ LIMIT events
as a trigger to post WQEs to SRQs.

Add more information to the existing debug printout for SRQ_LIMIT, and
output the warning messages only for the SRQ CATAS ERROR event.

Fixes: acba2420 ("mlx4_core: Add wrapper functions and comm channel and slave event support to EQs")
Fixes: e0debf9c

 ("mlx4_core: Reduce warning message for SRQ_LIMIT event to debug level")
Signed-off-by: default avatarJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: default avatarTariq Toukan <tariqt@mellanox.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      9577b174
    • Jack Morgenstein's avatar
      net/mlx4_core: Fix when to save some qp context flags for dynamic VST to VGT transitions · 7c3945bc
      Jack Morgenstein authored 
      Save the qp context flags byte containing the flag disabling vlan stripping
in the RESET to INIT qp transition, rather than in the INIT to RTR
transition. Per the firmware spec, the flags in this byte are active
in the RESET to INIT transition.

As a result of saving the flags in the incorrect qp transition, when
switching dynamically from VGT to VST and back to VGT, the vlan
remained stripped (as is required for VST) and did not return to
not-stripped (as is required for VGT).

Fixes: f0f829bf

 ("net/mlx4_core: Add immediate activate for VGT->VST->VGT")
Signed-off-by: default avatarJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: default avatarTariq Toukan <tariqt@mellanox.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      7c3945bc
    • Jack Morgenstein's avatar
      net/mlx4_core: Fix racy CQ (Completion Queue) free · 291c566a
      Jack Morgenstein authored 
      In function mlx4_cq_completion() and mlx4_cq_event(), the
radix_tree_lookup requires a rcu_read_lock.
This is mandatory: if another core frees the CQ, it could
run the radix_tree_node_rcu_free() call_rcu() callback while
its being used by the radix tree lookup function.

Additionally, in function mlx4_cq_event(), since we are adding
the rcu lock around the radix-tree lookup, we no longer need to take
the spinlock. Also, the synchronize_irq() call for the async event
eliminates the need for incrementing the cq reference count in
mlx4_cq_event().

Other changes:
1. In function mlx4_cq_free(), replace spin_lock_irq with spin_lock:
   we no longer take this spinlock in the interrupt context.
   The spinlock here, therefore, simply protects against different
   threads simultaneously invoking mlx4_cq_free() for different cq's.

2. In function mlx4_cq_free(), we move the radix tree delete to before
   the synchronize_irq() calls. This guarantees that we will not
   access this cq during any subsequent interrupts, and therefore can
   safely free the CQ after the synchronize_irq calls. The rcu_read_lock
   in the interrupt handlers only needs to protect against corrupting the
   radix tree; the interrupt handlers may access the cq outside the
   rcu_read_lock due to the synchronize_irq calls which protect against
   premature freeing of the cq.

3. In function mlx4_cq_event(), we change the mlx_warn message to mlx4_dbg.

4. We leave the cq reference count mechanism in place, because it is
   still needed for the cq completion tasklet mechanism.

Fixes: 6d90aa5c ("net/mlx4_core: Make sure there are no pending async events when freeing CQ")
Fixes: 225c7b1f

 ("IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters")
Signed-off-by: default avatarJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: default avatarMatan Barak <matanb@mellanox.com>
Signed-off-by: default avatarTariq Toukan <tariqt@mellanox.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      291c566a