GitHub/moto-9609/android_kernel_motorola_exynos9610.git
9 years agotcp: do not lock listener to process SYN packets
Eric Dumazet [Fri, 2 Oct 2015 18:43:39 +0000 (11:43 -0700)]
tcp: do not lock listener to process SYN packets

Everything should now be ready to finally allow SYN
packets processing without holding listener lock.

Tested:

3.5 Mpps SYNFLOOD. Plenty of cpu cycles available.

Next bottleneck is the refcount taken on listener,
that could be avoided if we remove SLAB_DESTROY_BY_RCU
strict semantic for listeners, and use regular RCU.

    13.18%  [kernel]  [k] __inet_lookup_listener
     9.61%  [kernel]  [k] tcp_conn_request
     8.16%  [kernel]  [k] sha_transform
     5.30%  [kernel]  [k] inet_reqsk_alloc
     4.22%  [kernel]  [k] sock_put
     3.74%  [kernel]  [k] tcp_make_synack
     2.88%  [kernel]  [k] ipt_do_table
     2.56%  [kernel]  [k] memcpy_erms
     2.53%  [kernel]  [k] sock_wfree
     2.40%  [kernel]  [k] tcp_v4_rcv
     2.08%  [kernel]  [k] fib_table_lookup
     1.84%  [kernel]  [k] tcp_openreq_init_rwin

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp/dccp: add a reschedule point in inet_csk_listen_stop()
Eric Dumazet [Fri, 2 Oct 2015 18:43:38 +0000 (11:43 -0700)]
tcp/dccp: add a reschedule point in inet_csk_listen_stop()

If a listener with thousands of children in accept queue
is dismantled, it can take a while to close all of them.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: remove max_qlen_log
Eric Dumazet [Fri, 2 Oct 2015 18:43:37 +0000 (11:43 -0700)]
tcp: remove max_qlen_log

This control variable was set at first listen(fd, backlog)
call, but not updated if application tried to increase or decrease
backlog. It made sense at the time listener had a non resizeable
hash table.

Also rounding to powers of two was not very friendly.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp/dccp: remove struct listen_sock
Eric Dumazet [Fri, 2 Oct 2015 18:43:36 +0000 (11:43 -0700)]
tcp/dccp: remove struct listen_sock

It is enough to check listener sk_state, no need for an extra
condition.

max_qlen_log can be moved into struct request_sock_queue

We can remove syn_wait_lock and the alignment it enforced.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: attach SYNACK messages to request sockets instead of listener
Eric Dumazet [Fri, 2 Oct 2015 18:43:35 +0000 (11:43 -0700)]
tcp: attach SYNACK messages to request sockets instead of listener

If a listen backlog is very big (to avoid syncookies), then
the listener sk->sk_wmem_alloc is the main source of false
sharing, as we need to touch it twice per SYNACK re-transmit
and TX completion.

(One SYN packet takes listener lock once, but up to 6 SYNACK
are generated)

By attaching the skb to the request socket, we remove this
source of contention.

Tested:

 listen(fd, 10485760); // single listener (no SO_REUSEPORT)
 16 RX/TX queue NIC
 Sustain a SYNFLOOD attack of ~320,000 SYN per second,
 Sending ~1,400,000 SYNACK per second.
 Perf profiles now show listener spinlock being next bottleneck.

    20.29%  [kernel]  [k] queued_spin_lock_slowpath
    10.06%  [kernel]  [k] __inet_lookup_established
     5.12%  [kernel]  [k] reqsk_timer_handler
     3.22%  [kernel]  [k] get_next_timer_interrupt
     3.00%  [kernel]  [k] tcp_make_synack
     2.77%  [kernel]  [k] ipt_do_table
     2.70%  [kernel]  [k] run_timer_softirq
     2.50%  [kernel]  [k] ip_finish_output
     2.04%  [kernel]  [k] cascade

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoipv6: remove obsolete inet6 functions
Eric Dumazet [Fri, 2 Oct 2015 18:43:34 +0000 (11:43 -0700)]
ipv6: remove obsolete inet6 functions

inet6_csk_search_req() and inet6_csk_reqsk_queue_hash_add()
no longer exist.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp/dccp: shrink struct listen_sock
Eric Dumazet [Fri, 2 Oct 2015 18:43:33 +0000 (11:43 -0700)]
tcp/dccp: shrink struct listen_sock

We no longer use hash_rnd, nr_table_entries and syn_table[]

For a listener with a backlog of 10 millions sockets, this
saves 80 MBytes of vmalloced memory.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp/dccp: install syn_recv requests into ehash table
Eric Dumazet [Fri, 2 Oct 2015 18:43:32 +0000 (11:43 -0700)]
tcp/dccp: install syn_recv requests into ehash table

In this patch, we insert request sockets into TCP/DCCP
regular ehash table (where ESTABLISHED and TIMEWAIT sockets
are) instead of using the per listener hash table.

ACK packets find SYN_RECV pseudo sockets without having
to find and lock the listener.

In nominal conditions, this halves pressure on listener lock.

Note that this will allow for SO_REUSEPORT refinements,
so that we can select a listener using cpu/numa affinities instead
of the prior 'consistent hash', since only SYN packets will
apply this selection logic.

We will shrink listen_sock in the following patch to ease
code review.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Ying Cai <ycai@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp/dccp: remove inet_csk_reqsk_queue_added() timeout argument
Eric Dumazet [Fri, 2 Oct 2015 18:43:31 +0000 (11:43 -0700)]
tcp/dccp: remove inet_csk_reqsk_queue_added() timeout argument

This is no longer used.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: get_openreq[46]() changes
Eric Dumazet [Fri, 2 Oct 2015 18:43:30 +0000 (11:43 -0700)]
tcp: get_openreq[46]() changes

When request sockets are no longer in a per listener hash table
but on regular TCP ehash, we need to access listener uid
through req->rsk_listener

get_openreq6() also gets a const for its request socket argument.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: remove BUG_ON() in tcp_check_req()
Eric Dumazet [Fri, 2 Oct 2015 18:43:29 +0000 (11:43 -0700)]
tcp: remove BUG_ON() in tcp_check_req()

Once listener is lockless, its sk_state can change anytime.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: cleanup tcp_v[46]_inbound_md5_hash()
Eric Dumazet [Fri, 2 Oct 2015 18:43:28 +0000 (11:43 -0700)]
tcp: cleanup tcp_v[46]_inbound_md5_hash()

We'll soon have to call tcp_v[46]_inbound_md5_hash() twice.
Also add const attribute to the socket, as it might be the
unlocked listener for SYN packets.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp/dccp: init sk_prot and call sk_node_init() in reqsk_alloc()
Eric Dumazet [Fri, 2 Oct 2015 18:43:27 +0000 (11:43 -0700)]
tcp/dccp: init sk_prot and call sk_node_init() in reqsk_alloc()

We plan to use generic functions to insert request sockets
into ehash table.

sk_prot needs to be set (to retrieve sk_prot->h.hashinfo)
sk_node needs to be cleared.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: call sk_mark_napi_id() on the child, not the listener
Eric Dumazet [Fri, 2 Oct 2015 18:43:26 +0000 (11:43 -0700)]
tcp: call sk_mark_napi_id() on the child, not the listener

This fixes a typo : We want to store the NAPI id on child socket.
Presumably nobody really uses busy polling, on short lived flows.

Fixes: 3d97379a67486 ("tcp: move sk_mark_napi_id() at the right place")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: move synflood_warned into struct request_sock_queue
Eric Dumazet [Fri, 2 Oct 2015 18:43:25 +0000 (11:43 -0700)]
tcp: move synflood_warned into struct request_sock_queue

long term plan is to remove struct listen_sock when its hash
table is no longer there.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: move qlen/young out of struct listen_sock
Eric Dumazet [Fri, 2 Oct 2015 18:43:24 +0000 (11:43 -0700)]
tcp: move qlen/young out of struct listen_sock

qlen_inc & young_inc were protected by listener lock,
while qlen_dec & young_dec were atomic fields.

Everything needs to be atomic for upcoming lockless listener.

Also move qlen/young in request_sock_queue as we'll get rid
of struct listen_sock eventually.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: add a spinlock to protect struct request_sock_queue
Eric Dumazet [Fri, 2 Oct 2015 18:43:23 +0000 (11:43 -0700)]
tcp: add a spinlock to protect struct request_sock_queue

struct request_sock_queue fields are currently protected
by the listener 'lock' (not a real spinlock)

We need to add a private spinlock instead, so that softirq handlers
creating children do not have to worry with backlog notion
that the listener 'lock' carries.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
David S. Miller [Fri, 2 Oct 2015 14:21:25 +0000 (07:21 -0700)]
Merge git://git./linux/kernel/git/davem/net

Conflicts:
net/dsa/slave.c

net/dsa/slave.c simply had overlapping changes.

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge tag 'mmc-v4.3-rc3' of git://git.linaro.org/people/ulf.hansson/mmc
Linus Torvalds [Fri, 2 Oct 2015 12:03:04 +0000 (08:03 -0400)]
Merge tag 'mmc-v4.3-rc3' of git://git.linaro.org/people/ulf.hansson/mmc

Pull MMC fixes from Ulf Hansson:
 "Here are some mmc fixes intended for v4.3 rc4:

  MMC core:
   - Allow users of mmc_of_parse() to succeed when CONFIG_GPIOLIB is
     unset
   - Prevent infinite loop of re-tuning for CRC-errors for CMD19 and
     CMD21

   MMC host:
   - pxamci: Fix issues with card detect
   - sunxi: Fix clk-delay settings"

* tag 'mmc-v4.3-rc3' of git://git.linaro.org/people/ulf.hansson/mmc:
  mmc: core: fix dead loop of mmc_retune
  mmc: pxamci: fix card detect with slot-gpio API
  mmc: sunxi: Fix clk-delay settings
  mmc: core: Don't return an error for CD/WP GPIOs when GPIOLIB is unset

9 years agoMerge git://git.infradead.org/intel-iommu
Linus Torvalds [Fri, 2 Oct 2015 11:59:29 +0000 (07:59 -0400)]
Merge git://git.infradead.org/intel-iommu

Pull IOVA fixes from David Woodhouse:
 "The main fix here is the first one, fixing the over-allocation of
   size-aligned requests.  The other patches simply make the existing
  IOVA code available to users other than the Intel VT-d driver, with no
  functional change.

  I concede the latter really *should* have been submitted during the
  merge window, but since it's basically risk-free and people are
  waiting to build on top of it and it's my fault I didn't get it in, I
  (and they) would be grateful if you'd take it"

* git://git.infradead.org/intel-iommu:
  iommu: Make the iova library a module
  iommu: iova: Export symbols
  iommu: iova: Move iova cache management to the iova library
  iommu/iova: Avoid over-allocating when size-aligned

9 years agoMerge branch 'akpm' (patches from Andrew)
Linus Torvalds [Fri, 2 Oct 2015 02:20:11 +0000 (22:20 -0400)]
Merge branch 'akpm' (patches from Andrew)

Merge misc fixes from Andrew Morton:
 "12 fixes"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  dmapool: fix overflow condition in pool_find_page()
  thermal: avoid division by zero in power allocator
  memcg: remove pcp_counter_lock
  kprobes: use _do_fork() in samples to make them work again
  drivers/input/joystick/Kconfig: zhenhua.c needs BITREVERSE
  memcg: make mem_cgroup_read_stat() unsigned
  memcg: fix dirty page migration
  dax: fix NULL pointer in __dax_pmd_fault()
  mm: hugetlbfs: skip shared VMAs when unmapping private pages to satisfy a fault
  mm/slab: fix unexpected index mapping result of kmalloc_size(INDEX_NODE+1)
  userfaultfd: remove kernel header include from uapi header
  arch/x86/include/asm/efi.h: fix build failure

9 years agoMerge tag 'pm+acpi-4.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael...
Linus Torvalds [Fri, 2 Oct 2015 02:06:40 +0000 (22:06 -0400)]
Merge tag 'pm+acpi-4.3-rc4' of git://git./linux/kernel/git/rafael/linux-pm

Pull power management and ACPI fixes from Rafael Wysocki:
 "These are fixes mostly, for a few changes made in this cycle (the
  intel_idle driver, the OPP library, the ACPI EC driver, turbostat) and
  for some issues that have just been discovered (ACPI PCI IRQ
  management, PCI power management documentation, turbostat), with a
  couple of cleanups on top of them.

  Specifics:

   - intel_idle driver fixup for the recently added Skylake chips
     support (Len Brown).

   - Operating Performance Points (OPP) library fix related to the
     recently added support for new DT bindings and a fix for a typo in
     a comment (Viresh Kumar, Stephen Boyd).

   - ACPI EC driver fix for a recently introduced memory leak in an
     error code path (Lv Zheng).

   - ACPI PCI IRQ management fix for the issue where an ISA IRQ is
     shared with a PCI device which requires it to be configured in a
     different way and may cause an interrupt storm to happen as a
     result with an extra ACPI SCI IRQ handling simplification on top of
     it (Jiang Liu).

   - Update of the PCI power management documentation that became
     outdated and started to actively confuse the readers to make it
     actually reflect the code (Rafael J Wysocki).

   - turbostat fixes including an IVB Xeon regression fix (related to
     the --debug command line option), Skylake adjustment for the TSC
     running at a frequency that doesn't match the base one exactly, and
     a Knights Landing quirk to account for the fact that it only
     updates APERF and MPERF every 1024 clock cycles plus bumping up the
     turbostat version number (Len Brown, Hubert Chrzaniuk)"

* tag 'pm+acpi-4.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  tools/power turbosat: update version number
  tools/power turbostat: SKL: Adjust for TSC difference from base frequency
  tools/power turbostat: KNL workaround for %Busy and Avg_MHz
  tools/power turbostat: IVB Xeon: fix --debug regression
  ACPI / PCI: Remove duplicated penalty on SCI IRQ
  ACPI, PCI, irq: Do not share PCI IRQ with ISA IRQ
  ACPI / EC: Fix a memory leak issue in acpi_ec_query()
  PM / OPP: Fix typo modifcation -> modification
  PCI / PM: Update runtime PM documentation for PCI devices
  PM / OPP: of_property_count_u32_elems() can return errors
  intel_idle: Skylake Client Support - updated

9 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Fri, 2 Oct 2015 01:55:35 +0000 (21:55 -0400)]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David Miller:

1) Fix regression in SKB partial checksum handling, from Pravin B
   Shalar.

2) Fix VLAN inside of VXLAN handling in i40e driver, from Jesse
   Brandeburg.

3) Cure softlockups during accept() in SCTP, from Karl Heiss.

4) MSG_PEEK should return multiple SKBs worth of data in AF_UNIX, from
   Aaron Conole.

5) IPV6 erroneously ignores output interface specifier in lookup key for
   route lookups, fix from David Ahern.

6) In Marvell DSA driver, forward unknown frames to CPU port, from
   Andrew Lunn.

7) Mission flow flag initializations in some code paths, from David
   Ahern.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
  net: Initialize flow flags in input path
  net: dsa: fix preparation of a port STP update
  testptp: Silence compiler warnings on ppc64
  net/mlx4: Handle return codes in mlx4_qp_attach_common
  dsa: mv88e6xxx: Enable forwarding for unknown to the CPU port
  skbuff: Fix skb checksum partial check.
  net: ipv6: Add RT6_LOOKUP_F_IFACE flag if oif is set
  net sysfs: Print link speed as signed integer
  bna: fix error handling
  af_unix: return data from multiple SKBs on recv() with MSG_PEEK flag
  af_unix: Convert the unix_sk macro to an inline function for type safety
  net: sctp: Don't use 64 kilobyte lookup table for four elements
  l2tp: protect tunnel->del_work by ref_count
  net/ibm/emac: bump version numbers for correct work with ethtool
  sctp: Prevent soft lockup when sctp_accept() is called during a timeout event
  sctp: Whitespace fix
  i40e/i40evf: check for stopped admin queue
  i40e: fix VLAN inside VXLAN
  r8169: fix handling rtl_readphy result
  net: hisilicon: fix handling platform_get_irq result

9 years agodmapool: fix overflow condition in pool_find_page()
Robin Murphy [Thu, 1 Oct 2015 22:37:19 +0000 (15:37 -0700)]
dmapool: fix overflow condition in pool_find_page()

If a DMA pool lies at the very top of the dma_addr_t range (as may
happen with an IOMMU involved), the calculated end address of the pool
wraps around to zero, and page lookup always fails.

Tweak the relevant calculation to be overflow-proof.

Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: Sakari Ailus <sakari.ailus@iki.fi>
Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agothermal: avoid division by zero in power allocator
Andrea Arcangeli [Thu, 1 Oct 2015 22:37:16 +0000 (15:37 -0700)]
thermal: avoid division by zero in power allocator

During boot I get a div by zero Oops regression starting in v4.3-rc3.

Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Reviewed-by: Javi Merino <javi.merino@arm.com>
Cc: Zhang Rui <rui.zhang@intel.com>
Cc: Eduardo Valentin <edubezval@gmail.com>
Cc: Daniel Kurtz <djkurtz@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomemcg: remove pcp_counter_lock
Greg Thelen [Thu, 1 Oct 2015 22:37:13 +0000 (15:37 -0700)]
memcg: remove pcp_counter_lock

Commit 733a572e66d2 ("memcg: make mem_cgroup_read_{stat|event}() iterate
possible cpus instead of online") removed the last use of the per memcg
pcp_counter_lock but forgot to remove the variable.

Kill the vestigial variable.

Signed-off-by: Greg Thelen <gthelen@google.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agokprobes: use _do_fork() in samples to make them work again
Petr Mladek [Thu, 1 Oct 2015 22:37:11 +0000 (15:37 -0700)]
kprobes: use _do_fork() in samples to make them work again

Commit 3033f14ab78c ("clone: support passing tls argument via C rather
than pt_regs magic") introduced _do_fork() that allowed to pass @tls
parameter.

The old do_fork() is defined only for architectures that are not ready
to use this way and do not define HAVE_COPY_THREAD_TLS.

Let's use _do_fork() in the kprobe examples to make them work again on
all architectures.

Signed-off-by: Petr Mladek <pmladek@suse.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Thiago Macieira <thiago.macieira@intel.com>
Cc: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agodrivers/input/joystick/Kconfig: zhenhua.c needs BITREVERSE
Andrew Morton [Thu, 1 Oct 2015 22:37:08 +0000 (15:37 -0700)]
drivers/input/joystick/Kconfig: zhenhua.c needs BITREVERSE

It uses bitrev8(), so it must ensure that lib/bitrev.o gets included in
vmlinux.

Cc: Fengguang Wu <fengguang.wu@gmail.com>
Cc: yalin wang <yalin.wang2010@gmail.com>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomemcg: make mem_cgroup_read_stat() unsigned
Greg Thelen [Thu, 1 Oct 2015 22:37:05 +0000 (15:37 -0700)]
memcg: make mem_cgroup_read_stat() unsigned

mem_cgroup_read_stat() returns a page count by summing per cpu page
counters.  The summing is racy wrt.  updates, so a transient negative
sum is possible.  Callers don't want negative values:

 - mem_cgroup_wb_stats() doesn't want negative nr_dirty or nr_writeback.
   This could confuse dirty throttling.

 - oom reports and memory.stat shouldn't show confusing negative usage.

 - tree_usage() already avoids negatives.

Avoid returning negative page counts from mem_cgroup_read_stat() and
convert it to unsigned.

[akpm@linux-foundation.org: fix old typo while we're in there]
Signed-off-by: Greg Thelen <gthelen@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: <stable@vger.kernel.org> [4.2+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomemcg: fix dirty page migration
Greg Thelen [Thu, 1 Oct 2015 22:37:02 +0000 (15:37 -0700)]
memcg: fix dirty page migration

The problem starts with a file backed dirty page which is charged to a
memcg.  Then page migration is used to move oldpage to newpage.

Migration:
 - copies the oldpage's data to newpage
 - clears oldpage.PG_dirty
 - sets newpage.PG_dirty
 - uncharges oldpage from memcg
 - charges newpage to memcg

Clearing oldpage.PG_dirty decrements the charged memcg's dirty page
count.

However, because newpage is not yet charged, setting newpage.PG_dirty
does not increment the memcg's dirty page count.  After migration
completes newpage.PG_dirty is eventually cleared, often in
account_page_cleaned().  At this time newpage is charged to a memcg so
the memcg's dirty page count is decremented which causes underflow
because the count was not previously incremented by migration.  This
underflow causes balance_dirty_pages() to see a very large unsigned
number of dirty memcg pages which leads to aggressive throttling of
buffered writes by processes in non root memcg.

This issue:
 - can harm performance of non root memcg buffered writes.
 - can report too small (even negative) values in
   memory.stat[(total_)dirty] counters of all memcg, including the root.

To avoid polluting migrate.c with #ifdef CONFIG_MEMCG checks, introduce
page_memcg() and set_page_memcg() helpers.

Test:
    0) setup and enter limited memcg
    mkdir /sys/fs/cgroup/test
    echo 1G > /sys/fs/cgroup/test/memory.limit_in_bytes
    echo $$ > /sys/fs/cgroup/test/cgroup.procs

    1) buffered writes baseline
    dd if=/dev/zero of=/data/tmp/foo bs=1M count=1k
    sync
    grep ^dirty /sys/fs/cgroup/test/memory.stat

    2) buffered writes with compaction antagonist to induce migration
    yes 1 > /proc/sys/vm/compact_memory &
    rm -rf /data/tmp/foo
    dd if=/dev/zero of=/data/tmp/foo bs=1M count=1k
    kill %
    sync
    grep ^dirty /sys/fs/cgroup/test/memory.stat

    3) buffered writes without antagonist, should match baseline
    rm -rf /data/tmp/foo
    dd if=/dev/zero of=/data/tmp/foo bs=1M count=1k
    sync
    grep ^dirty /sys/fs/cgroup/test/memory.stat

                       (speed, dirty residue)
             unpatched                       patched
    1) 841 MB/s 0 dirty pages          886 MB/s 0 dirty pages
    2) 611 MB/s -33427456 dirty pages  793 MB/s 0 dirty pages
    3) 114 MB/s -33427456 dirty pages  891 MB/s 0 dirty pages

    Notice that unpatched baseline performance (1) fell after
    migration (3): 841 -> 114 MB/s.  In the patched kernel, post
    migration performance matches baseline.

Fixes: c4843a7593a9 ("memcg: add per cgroup dirty page accounting")
Signed-off-by: Greg Thelen <gthelen@google.com>
Reported-by: Dave Hansen <dave.hansen@intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: <stable@vger.kernel.org> [4.2+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agodax: fix NULL pointer in __dax_pmd_fault()
Ross Zwisler [Thu, 1 Oct 2015 22:36:59 +0000 (15:36 -0700)]
dax: fix NULL pointer in __dax_pmd_fault()

Commit 46c043ede471 ("mm: take i_mmap_lock in unmap_mapping_range() for
DAX") moved some code in __dax_pmd_fault() that was responsible for
zeroing newly allocated PMD pages.  The new location didn't properly set
up 'kaddr', so when run this code resulted in a NULL pointer BUG.

Fix this by getting the correct 'kaddr' via bdev_direct_access().

Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Reported-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Matthew Wilcox <willy@linux.intel.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Dave Chinner <david@fromorbit.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomm: hugetlbfs: skip shared VMAs when unmapping private pages to satisfy a fault
Mel Gorman [Thu, 1 Oct 2015 22:36:57 +0000 (15:36 -0700)]
mm: hugetlbfs: skip shared VMAs when unmapping private pages to satisfy a fault

SunDong reported the following on

  https://bugzilla.kernel.org/show_bug.cgi?id=103841

I think I find a linux bug, I have the test cases is constructed. I
can stable recurring problems in fedora22(4.0.4) kernel version,
arch for x86_64.  I construct transparent huge page, when the parent
and child process with MAP_SHARE, MAP_PRIVATE way to access the same
huge page area, it has the opportunity to lead to huge page copy on
write failure, and then it will munmap the child corresponding mmap
area, but then the child mmap area with VM_MAYSHARE attributes, child
process munmap this area can trigger VM_BUG_ON in set_vma_resv_flags
functions (vma - > vm_flags & VM_MAYSHARE).

There were a number of problems with the report (e.g.  it's hugetlbfs that
triggers this, not transparent huge pages) but it was fundamentally
correct in that a VM_BUG_ON in set_vma_resv_flags() can be triggered that
looks like this

 vma ffff8804651fd0d0 start 00007fc474e00000 end 00007fc475e00000
 next ffff8804651fd018 prev ffff8804651fd188 mm ffff88046b1b1800
 prot 8000000000000027 anon_vma           (null) vm_ops ffffffff8182a7a0
 pgoff 0 file ffff88106bdb9800 private_data           (null)
 flags: 0x84400fb(read|write|shared|mayread|maywrite|mayexec|mayshare|dontexpand|hugetlb)
 ------------
 kernel BUG at mm/hugetlb.c:462!
 SMP
 Modules linked in: xt_pkttype xt_LOG xt_limit [..]
 CPU: 38 PID: 26839 Comm: map Not tainted 4.0.4-default #1
 Hardware name: Dell Inc. PowerEdge R810/0TT6JF, BIOS 2.7.4 04/26/2012
 set_vma_resv_flags+0x2d/0x30

The VM_BUG_ON is correct because private and shared mappings have
different reservation accounting but the warning clearly shows that the
VMA is shared.

When a private COW fails to allocate a new page then only the process
that created the VMA gets the page -- all the children unmap the page.
If the children access that data in the future then they get killed.

The problem is that the same file is mapped shared and private.  During
the COW, the allocation fails, the VMAs are traversed to unmap the other
private pages but a shared VMA is found and the bug is triggered.  This
patch identifies such VMAs and skips them.

Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reported-by: SunDong <sund_sky@126.com>
Reviewed-by: Michal Hocko <mhocko@suse.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: David Rientjes <rientjes@google.com>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomm/slab: fix unexpected index mapping result of kmalloc_size(INDEX_NODE+1)
Joonsoo Kim [Thu, 1 Oct 2015 22:36:54 +0000 (15:36 -0700)]
mm/slab: fix unexpected index mapping result of kmalloc_size(INDEX_NODE+1)

Commit description is copied from the original post of this bug:

  http://comments.gmane.org/gmane.linux.kernel.mm/135349

Kernels after v3.9 use kmalloc_size(INDEX_NODE + 1) to get the next
larger cache size than the size index INDEX_NODE mapping.  In kernels
3.9 and earlier we used malloc_sizes[INDEX_L3 + 1].cs_size.

However, sometimes we can't get the right output we expected via
kmalloc_size(INDEX_NODE + 1), causing a BUG().

The mapping table in the latest kernel is like:
    index = {0,   1,  2 ,  3,  4,   5,   6,   n}
     size = {0,   96, 192, 8, 16,  32,  64,   2^n}
The mapping table before 3.10 is like this:
    index = {0 , 1 , 2,   3,  4 ,  5 ,  6,   n}
    size  = {32, 64, 96, 128, 192, 256, 512, 2^(n+3)}

The problem on my mips64 machine is as follows:

(1) When configured DEBUG_SLAB && DEBUG_PAGEALLOC && DEBUG_LOCK_ALLOC
    && DEBUG_SPINLOCK, the sizeof(struct kmem_cache_node) will be "150",
    and the macro INDEX_NODE turns out to be "2": #define INDEX_NODE
    kmalloc_index(sizeof(struct kmem_cache_node))

(2) Then the result of kmalloc_size(INDEX_NODE + 1) is 8.

(3) Then "if(size >= kmalloc_size(INDEX_NODE + 1)" will lead to "size
    = PAGE_SIZE".

(4) Then "if ((size >= (PAGE_SIZE >> 3))" test will be satisfied and
    "flags |= CFLGS_OFF_SLAB" will be covered.

(5) if (flags & CFLGS_OFF_SLAB)" test will be satisfied and will go to
    "cachep->slabp_cache = kmalloc_slab(slab_size, 0u)", and the result
    here may be NULL while kernel bootup.

(6) Finally,"BUG_ON(ZERO_OR_NULL_PTR(cachep->slabp_cache));" causes the
    BUG info as the following shows (may be only mips64 has this problem):

This patch fixes the problem of kmalloc_size(INDEX_NODE + 1) and removes
the BUG by adding 'size >= 256' check to guarantee that all necessary
small sized slabs are initialized regardless sequence of slab size in
mapping table.

Fixes: e33660165c90 ("slab: Use common kmalloc_index/kmalloc_size...")
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Reported-by: Liuhailong <liu.hailong6@zte.com.cn>
Acked-by: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agouserfaultfd: remove kernel header include from uapi header
Andre Przywara [Thu, 1 Oct 2015 22:36:51 +0000 (15:36 -0700)]
userfaultfd: remove kernel header include from uapi header

As include/uapi/linux/userfaultfd.h is a user visible header file, it
should not include kernel-exclusive header files.

So trying to build the userfaultfd test program from the selftests
directory fails, since it contains a reference to linux/compiler.h.  As
it turns out, that header is not really needed there, so we can simply
remove it to fix that issue.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Shuah Khan <shuahkh@osg.samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoarch/x86/include/asm/efi.h: fix build failure
Andrey Ryabinin [Thu, 1 Oct 2015 22:36:48 +0000 (15:36 -0700)]
arch/x86/include/asm/efi.h: fix build failure

With KMEMCHECK=y, KASAN=n:

  arch/x86/platform/efi/efi.c:673:3: error: implicit declaration of function `memcpy' [-Werror=implicit-function-declaration]
  arch/x86/platform/efi/efi_64.c:139:2: error: implicit declaration of function `memcpy' [-Werror=implicit-function-declaration]
  arch/x86/include/asm/desc.h:121:2: error: implicit declaration of function `memcpy' [-Werror=implicit-function-declaration]

Don't #undef memcpy if KASAN=n.

Fixes: 769a8089c1fd ("x86, efi, kasan: #undef memset/memcpy/memmove per arch")
Signed-off-by: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Reported-by: Ingo Molnar <mingo@kernel.org>
Reported-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoMerge branch 'bridge_vlan_cleanups_fixes'
David S. Miller [Fri, 2 Oct 2015 01:24:12 +0000 (18:24 -0700)]
Merge branch 'bridge_vlan_cleanups_fixes'

Nikolay Aleksandrov says:

====================
bridge: vlan: cleanups & fixes

This is the first follow-up set, patch 01 reduces the default rhashtable
size and the number of locks that can be allocated. Patch 02 and 04 fix
possible null pointer dereferences due to the new ordering and
initialization on port add/del, and patch 03 moves the "pvid" member in
the net_bridge_vlan_group struct in order to simplify code (similar to how
it was with the older struct). Patch 05 fixes adding a vlan on a port which
is pvid and doesn't have a global context yet.
Please review carefully, I think this is the first use of rhashtable's
"locks_mul" member in the tree and I'd like to make sure it's correct.
Another thing that needs special attention is the nbp_vlan_flush() move
after the rx_handler unregister.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agobridge: vlan: don't pass flags when creating context only
Nikolay Aleksandrov [Wed, 30 Sep 2015 18:16:55 +0000 (20:16 +0200)]
bridge: vlan: don't pass flags when creating context only

We should not pass the original flags when creating a context vlan only
because they may contain some flags that change behaviour in the bridge.
The new global context should be with minimal set of flags, so pass 0
and let br_vlan_add() set the master flag only.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agobridge: vlan: fix possible null ptr derefs on port init and deinit
Nikolay Aleksandrov [Wed, 30 Sep 2015 18:16:54 +0000 (20:16 +0200)]
bridge: vlan: fix possible null ptr derefs on port init and deinit

When a new port is being added we need to make vlgrp available after
rhashtable has been initialized and when removing a port we need to
flush the vlans and free the resources after we're sure noone can use
the port, i.e. after it's removed from the port list and synchronize_rcu
is executed.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agobridge: vlan: move pvid inside net_bridge_vlan_group
Nikolay Aleksandrov [Wed, 30 Sep 2015 18:16:53 +0000 (20:16 +0200)]
bridge: vlan: move pvid inside net_bridge_vlan_group

One obvious way to converge more code (which was also used by the
previous vlan code) is to move pvid inside net_bridge_vlan_group. This
allows us to simplify some and remove other port-specific functions.
Also gives us the ability to simply pass the vlan group and use all of the
contained information.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agobridge: vlan: fix possible null vlgrp deref while registering new port
Nikolay Aleksandrov [Wed, 30 Sep 2015 18:16:52 +0000 (20:16 +0200)]
bridge: vlan: fix possible null vlgrp deref while registering new port

While a new port is being initialized the rx_handler gets set, but the
vlans get initialized later in br_add_if() and in that window if we
receive a frame with a link-local address we can try to dereference
p->vlgrp in:
br_handle_frame() -> br_handle_local_finish() -> br_should_learn()

Fix this by checking vlgrp before using it.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agobridge: vlan: adjust rhashtable initial size and hash locks size
Nikolay Aleksandrov [Wed, 30 Sep 2015 18:16:51 +0000 (20:16 +0200)]
bridge: vlan: adjust rhashtable initial size and hash locks size

As Stephen pointed out the default initial size is more than we need, so
let's start small (4 elements, thus nelem_hint = 3). Also limit the hash
locks to the number of CPUs as we don't need any write-side scaling and
this looks like the minimum.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Linus Torvalds [Thu, 1 Oct 2015 20:43:25 +0000 (16:43 -0400)]
Merge tag 'for-linus' of git://git./virt/kvm/kvm

Pull KVM fixes from Paolo Bonzini:
 "(Relatively) a lot of reverts, mostly.

  Bugs have trickled in for a new feature in 4.2 (MTRR support in
  guests) so I'm reverting it all; let's not make this -rc period busier
  for KVM than it's been so far.  This covers the four reverts from me.

  The fifth patch is being reverted because Radim found a bug in the
  implementation of stable scheduler clock, *but* also managed to
  implement the feature entirely without hypervisor support.  So instead
  of fixing the hypervisor side we can remove it completely; 4.4 will
  get the new implementation"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  Use WARN_ON_ONCE for missing X86_FEATURE_NRIPS
  Update KVM homepage Url
  Revert "KVM: SVM: use NPT page attributes"
  Revert "KVM: svm: handle KVM_X86_QUIRK_CD_NW_CLEARED in svm_get_mt_mask"
  Revert "KVM: SVM: Sync g_pat with guest-written PAT value"
  Revert "KVM: x86: apply guest MTRR virtualization on host reserved pages"
  Revert "KVM: x86: zero kvmclock_offset when vcpu0 initializes kvmclock system MSR"

9 years agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma
Linus Torvalds [Thu, 1 Oct 2015 20:38:52 +0000 (16:38 -0400)]
Merge tag 'for-linus' of git://git./linux/kernel/git/dledford/rdma

Pull rdma fixes from Doug Ledford:
 - Fixes for mlx5 related issues
 - Fixes for ipoib multicast handling

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma:
  IB/ipoib: increase the max mcast backlog queue
  IB/ipoib: Make sendonly multicast joins create the mcast group
  IB/ipoib: Expire sendonly multicast joins
  IB/mlx5: Remove pa_lkey usages
  IB/mlx5: Remove support for IB_DEVICE_LOCAL_DMA_LKEY
  IB/iser: Add module parameter for always register memory
  xprtrdma: Replace global lkey with lkey local to PD

9 years agoMerge branches 'pm-cpuidle', 'pm-opp' and 'pm-tools'
Rafael J. Wysocki [Thu, 1 Oct 2015 20:30:47 +0000 (22:30 +0200)]
Merge branches 'pm-cpuidle', 'pm-opp' and 'pm-tools'

* pm-cpuidle:
  intel_idle: Skylake Client Support - updated

* pm-opp:
  PM / OPP: Fix typo modifcation -> modification
  PM / OPP: of_property_count_u32_elems() can return errors

* pm-tools:
  tools/power turbosat: update version number
  tools/power turbostat: SKL: Adjust for TSC difference from base frequency
  tools/power turbostat: KNL workaround for %Busy and Avg_MHz
  tools/power turbostat: IVB Xeon: fix --debug regression

9 years agoMerge branch 'acpi-ec'
Rafael J. Wysocki [Thu, 1 Oct 2015 20:30:35 +0000 (22:30 +0200)]
Merge branch 'acpi-ec'

* acpi-ec:
  ACPI / EC: Fix a memory leak issue in acpi_ec_query()

9 years agoMerge branches 'pm-pci' and 'acpi-pci'
Rafael J. Wysocki [Thu, 1 Oct 2015 20:30:12 +0000 (22:30 +0200)]
Merge branches 'pm-pci' and 'acpi-pci'

* pm-pci:
  PCI / PM: Update runtime PM documentation for PCI devices

* acpi-pci:
  ACPI / PCI: Remove duplicated penalty on SCI IRQ
  ACPI, PCI, irq: Do not share PCI IRQ with ISA IRQ

9 years agoUse WARN_ON_ONCE for missing X86_FEATURE_NRIPS
Dirk Müller [Thu, 1 Oct 2015 11:43:42 +0000 (13:43 +0200)]
Use WARN_ON_ONCE for missing X86_FEATURE_NRIPS

The cpu feature flags are not ever going to change, so warning
everytime can cause a lot of kernel log spam
(in our case more than 10GB/hour).

The warning seems to only occur when nested virtualization is
enabled, so it's probably triggered by a KVM bug.  This is a
sensible and safe change anyway, and the KVM bug fix might not
be suitable for stable releases anyway.

Cc: stable@vger.kernel.org
Signed-off-by: Dirk Mueller <dmueller@suse.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoUpdate KVM homepage Url
Dirk Müller [Thu, 1 Oct 2015 11:46:01 +0000 (13:46 +0200)]
Update KVM homepage Url

The old one appears to be a generic catch all page, which
is unhelpful.

Signed-off-by: Dirk Mueller <dmueller@suse.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoMerge tag 'upstream-4.3-rc4' of git://git.infradead.org/linux-ubifs
Linus Torvalds [Thu, 1 Oct 2015 11:57:27 +0000 (07:57 -0400)]
Merge tag 'upstream-4.3-rc4' of git://git.infradead.org/linux-ubifs

Pull UBI/UBIFS fixes from Richard Weinberger:
 "This contains three bug fixes for both UBI and UBIFS"

* tag 'upstream-4.3-rc4' of git://git.infradead.org/linux-ubifs:
  UBI: return ENOSPC if no enough space available
  UBI: Validate data_size
  UBIFS: Kill unneeded locking in ubifs_init_security

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris...
Linus Torvalds [Thu, 1 Oct 2015 11:50:08 +0000 (07:50 -0400)]
Merge branch 'for-linus' of git://git./linux/kernel/git/jmorris/linux-security

Pull key signing fixes from James Morris:
 "Keyrings and modsign fixes from David Howells"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:
  MODSIGN: Change from CMS to PKCS#7 signing if the openssl is too old
  X.509: Don't strip leading 00's from key ID when constructing key description
  KEYS: Remove unnecessary header #inclusions from extract-cert.c
  KEYS: Fix race between key destruction and finding a keyring by name

9 years agoRevert "KVM: SVM: use NPT page attributes"
Paolo Bonzini [Thu, 1 Oct 2015 11:20:22 +0000 (13:20 +0200)]
Revert "KVM: SVM: use NPT page attributes"

This reverts commit 3c2e7f7de3240216042b61073803b61b9b3cfb22.
Initializing the mapping from MTRR to PAT values was reported to
fail nondeterministically, and it also caused extremely slow boot
(due to caching getting disabled---bug 103321) with assigned devices.

Reported-by: Markus Trippelsdorf <markus@trippelsdorf.de>
Reported-by: Sebastian Schuette <dracon@ewetel.net>
Cc: stable@vger.kernel.org # 4.2+
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoRevert "KVM: svm: handle KVM_X86_QUIRK_CD_NW_CLEARED in svm_get_mt_mask"
Paolo Bonzini [Thu, 1 Oct 2015 11:19:55 +0000 (13:19 +0200)]
Revert "KVM: svm: handle KVM_X86_QUIRK_CD_NW_CLEARED in svm_get_mt_mask"

This reverts commit 5492830370171b6a4ede8a3bfba687a8d0f25fa5.
It builds on the commit that is being reverted next.

Cc: stable@vger.kernel.org # 4.2+
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoRevert "KVM: SVM: Sync g_pat with guest-written PAT value"
Paolo Bonzini [Thu, 1 Oct 2015 11:28:15 +0000 (13:28 +0200)]
Revert "KVM: SVM: Sync g_pat with guest-written PAT value"

This reverts commit e098223b789b4a618dacd79e5e0dad4a9d5018d1,
which has a dependency on other commits being reverted.

Cc: stable@vger.kernel.org # 4.2+
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoRevert "KVM: x86: apply guest MTRR virtualization on host reserved pages"
Paolo Bonzini [Thu, 1 Oct 2015 11:12:47 +0000 (13:12 +0200)]
Revert "KVM: x86: apply guest MTRR virtualization on host reserved pages"

This reverts commit fd717f11015f673487ffc826e59b2bad69d20fe5.
It was reported to cause Machine Check Exceptions (bug 104091).

Reported-by: harn-solo@gmx.de
Cc: stable@vger.kernel.org # 4.2+
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoMerge git://www.linux-watchdog.org/linux-watchdog
Linus Torvalds [Wed, 30 Sep 2015 17:13:55 +0000 (13:13 -0400)]
Merge git://www.linux-watchdog.org/linux-watchdog

Pull watchdog fixes from Wim Van Sebroeck:
 "This fixes:

   - module autoload for 3 OF platform drivers
   - poweroff behaviour on bcm2835 watchdog device
   - I2C dependencies for iTCO_wdt.c"

* git://www.linux-watchdog.org/linux-watchdog:
  watchdog: iTCO: Fix dependencies on I2C
  watchdog: bcm2835: Fix poweroff behaviour
  watchdog: Fix module autoload for OF platform driver

9 years agoMerge tag 'hwmon-for-linus-v4.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Wed, 30 Sep 2015 17:11:42 +0000 (13:11 -0400)]
Merge tag 'hwmon-for-linus-v4.3-rc4' of git://git./linux/kernel/git/groeck/linux-staging

Pull hwmin fixes from Guenter Roeck:
 "Fix module autoload for various drivers"

* tag 'hwmon-for-linus-v4.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
  hwmon: (pwm-fan) Fix module autoload for OF platform driver
  hwmon: (gpio-fan) Fix module autoload for OF platform driver
  hwmon: (abx500) Fix module autoload for OF platform driver

9 years agoMerge branch 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Wed, 30 Sep 2015 17:01:35 +0000 (13:01 -0400)]
Merge branch 'core-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull RCU fixes from Ingo Molnar:
 "Two RCU fixes:

   - work around bug with recent GCC versions.

   - fix false positive lockdep splat"

* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  rcu: Suppress lockdep false positive for rcp->exp_funnel_mutex
  rcu: Change _wait_rcu_gp() to work around GCC bug 67055

9 years agoInitialize msg/shm IPC objects before doing ipc_addid()
Linus Torvalds [Wed, 30 Sep 2015 16:48:40 +0000 (12:48 -0400)]
Initialize msg/shm IPC objects before doing ipc_addid()

As reported by Dmitry Vyukov, we really shouldn't do ipc_addid() before
having initialized the IPC object state.  Yes, we initialize the IPC
object in a locked state, but with all the lockless RCU lookup work,
that IPC object lock no longer means that the state cannot be seen.

We already did this for the IPC semaphore code (see commit e8577d1f0329:
"ipc/sem.c: fully initialize sem_array before making it visible") but we
clearly forgot about msg and shm.

Reported-by: Dmitry Vyukov <dvyukov@google.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Davidlohr Bueso <dbueso@suse.de>
Cc: stable@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agommc: core: fix dead loop of mmc_retune
Chaotian Jing [Wed, 30 Sep 2015 09:37:18 +0000 (17:37 +0800)]
mmc: core: fix dead loop of mmc_retune

When get a CRC error, start the mmc_retune, it will issue CMD19/CMD21
to do tune, assume there were 10 clock phase need to try, phase 0 to
phase 6 is ok, phase 7 to phase 9 is NG, we try it from 0 to 9, so
the last CMD19/CMD21 will get CRC error, host->need_retune was set and
cause mmc_retune was called, then dead loop of mmc_retune

Signed-off-by: Chaotian Jing <chaotian.jing@mediatek.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Fixes: bd11e8bd03ca ("mmc: core: Flag re-tuning is needed on CRC errors")
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
9 years agonet: Initialize flow flags in input path
David Ahern [Wed, 30 Sep 2015 02:07:07 +0000 (19:07 -0700)]
net: Initialize flow flags in input path

The fib_table_lookup tracepoint found 2 places where the flowi4_flags is
not initialized.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next
David S. Miller [Wed, 30 Sep 2015 04:46:21 +0000 (21:46 -0700)]
Merge git://git./linux/kernel/git/pablo/nf-next

Pablo Neira Ayuso says:

====================
Netfilter/IPVS updates for net-next

The following pull request contains Netfilter/IPVS updates for net-next
containing 90 patches from Eric Biederman.

The main goal of this batch is to avoid recurrent lookups for the netns
pointer, that happens over and over again in our Netfilter/IPVS code. The idea
consists of passing netns pointer from the hook state to the relevant functions
and objects where this may be needed.

You can find more information on the IPVS updates from Simon Horman's commit
merge message:

c3456026adc0 ("Merge tag 'ipvs2-for-v4.4' of https://git.kernel.org/pub/scm/linux/kernel/git/horms/ipvs-next").

Exceptionally, this time, I'm not posting the patches again on netdev, Eric
already Cc'ed this mailing list in the original submission. If you need me to
make, just let me know.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: dsa: fix preparation of a port STP update
Vivien Didelot [Tue, 29 Sep 2015 18:17:54 +0000 (14:17 -0400)]
net: dsa: fix preparation of a port STP update

Because of the default 0 value of ret in dsa_slave_port_attr_set, a
driver may return -EOPNOTSUPP from the commit phase of a STP state,
which triggers a WARN() from switchdev.

This happened on a 6185 switch which does not support hardware bridging.

Fixes: 3563606258cf ("switchdev: convert STP update to switchdev attr set")
Reported-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: dsa: fix preparation of a port STP update
Vivien Didelot [Tue, 29 Sep 2015 16:38:36 +0000 (12:38 -0400)]
net: dsa: fix preparation of a port STP update

Because of the default 0 value of ret in dsa_slave_port_attr_set, a
driver may return -EOPNOTSUPP from the commit phase of a STP state,
which triggers a WARN() from switchdev.

This happened on a 6185 switch which does not support hardware bridging.

Reported-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Acked-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: Add support for filtering neigh dump by master device
David Ahern [Tue, 29 Sep 2015 16:32:03 +0000 (09:32 -0700)]
net: Add support for filtering neigh dump by master device

Add support for filtering neighbor dumps by master device by adding
the NDA_MASTER attribute to the dump request. A new netlink flag,
NLM_F_DUMP_FILTERED, is added to indicate the kernel supports the
request and output is filtered as requested.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: fix tcp_v6_md5_do_lookup prototype
Eric Dumazet [Wed, 30 Sep 2015 04:24:05 +0000 (21:24 -0700)]
tcp: fix tcp_v6_md5_do_lookup prototype

tcp_v6_md5_do_lookup() now takes a const socket, even if
CONFIG_TCP_MD5SIG is not set.

Fixes: b83e3deb974c ("tcp: md5: constify tcp_md5_do_lookup() socket argument")
From: Eric Dumazet <edumazet@google.com>
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge branch 'switchdev-callback'
David S. Miller [Wed, 30 Sep 2015 04:32:00 +0000 (21:32 -0700)]
Merge branch 'switchdev-callback'

Vivien Didelot says:

====================
net: switchdev: use specific switchdev_obj_*

This patchset changes switchdev add, del, dump operations from this:

    int     (*switchdev_port_obj_add)(struct net_device *dev,
                                      struct switchdev_obj *obj,
                                      struct switchdev_trans *trans);
    int     (*switchdev_port_obj_del)(struct net_device *dev,
                                      struct switchdev_obj *obj);
    int     (*switchdev_port_obj_dump)(struct net_device *dev,
                                      struct switchdev_obj *obj);

to something similar to the notifier_call callback of a notifier_block:

    int     (*switchdev_port_obj_add)(struct net_device *dev,
                                      enum switchdev_obj_id id,
                                      const void *obj,
                                      struct switchdev_trans *trans);
    int     (*switchdev_port_obj_del)(struct net_device *dev,
                                      enum switchdev_obj_id id,
                                      const void *obj);
    int     (*switchdev_port_obj_dump)(struct net_device *dev,
                                       enum switchdev_obj_id id, void *obj,
                                       int (*cb)(void *obj));

This allows the caller to pass and expect back a specific switchdev_obj_*
structure (e.g. switchdev_obj_fdb) instead of the generic switchdev_obj one.

This will simplify pushing the callback function down to the drivers.

The first 3 patches get rid of the dev parameter of the dump callback, since it
is not always neeeded (e.g. vlan_dump) and some drivers (such as DSA drivers)
may not have easy access to it.

Patches 4 and 5 implement the change in the switchdev operations and its users.

Patch 6 extracts the inner switchdev_obj_* structures from switchdev_obj and
removes this last one.

v2: fix error spotted by kbuild (extra ';' inline switchdev_port_obj_dump).
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: switchdev: extract struct switchdev_obj_*
Vivien Didelot [Tue, 29 Sep 2015 16:07:18 +0000 (12:07 -0400)]
net: switchdev: extract struct switchdev_obj_*

Now that switchdev and its drivers directly use specific switchdev_obj_*
structures, move them out of the switchdev_obj union and get rif of this
outer structure.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: switchdev: abstract object in add/del ops
Vivien Didelot [Tue, 29 Sep 2015 16:07:17 +0000 (12:07 -0400)]
net: switchdev: abstract object in add/del ops

Similar to the notifier_call callback of a notifier_block, change the
function signature of switchdev add and del operations to:

    int switchdev_port_obj_add/del(struct net_device *dev,
                                   enum switchdev_obj_id id, void *obj);

This allows the caller to pass a specific switchdev_obj_* structure
instead of the generic switchdev_obj one.

Drivers implementation of these operations and switchdev have been
changed accordingly.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: switchdev: pass callback to dump operation
Vivien Didelot [Tue, 29 Sep 2015 16:07:16 +0000 (12:07 -0400)]
net: switchdev: pass callback to dump operation

Similar to the notifier_call callback of a notifier_block, change the
function signature of switchdev dump operation to:

    int switchdev_port_obj_dump(struct net_device *dev,
                                enum switchdev_obj_id id, void *obj,
                                int (*cb)(void *obj));

This allows the caller to pass and expect back a specific
switchdev_obj_* structure instead of the generic switchdev_obj one.

Drivers implementation of dump operation can now expect this specific
structure and call the callback with it. Drivers have been changed
accordingly.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: switchdev: remove dev from switchdev_obj cb
Vivien Didelot [Tue, 29 Sep 2015 16:07:15 +0000 (12:07 -0400)]
net: switchdev: remove dev from switchdev_obj cb

The net_device associated to a dump operation does not have to be passed
to the callback. switchdev stores it in a superset struct, if needed.

Also some drivers (such as DSA drivers) may not have easy access to it.

This will simplify pushing the callback function down to the drivers.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: switchdev: move dev in switchdev_fdb_dump
Vivien Didelot [Tue, 29 Sep 2015 16:07:14 +0000 (12:07 -0400)]
net: switchdev: move dev in switchdev_fdb_dump

The FDB dump callback requires the related net_device so move it to the
struct switchdev_fdb_dump superset instead of using a callback param.

With this done, it'll be simpler to change the dump function signature.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: switchdev: remove dev in port_vlan_dump_put
Vivien Didelot [Tue, 29 Sep 2015 16:07:13 +0000 (12:07 -0400)]
net: switchdev: remove dev in port_vlan_dump_put

The static switchdev_port_vlan_dump_put function does not need the
net_device parameter, so remove it.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotestptp: Silence compiler warnings on ppc64
Thomas Huth [Tue, 29 Sep 2015 15:45:28 +0000 (17:45 +0200)]
testptp: Silence compiler warnings on ppc64

When compiling Documentation/ptp/testptp.c the following compiler
warnings are printed out:

Documentation/ptp/testptp.c: In function ‘main’:
Documentation/ptp/testptp.c:367:11: warning: format ‘%lld’ expects argument
    of type ‘long long int’, but argument 3 has type ‘__s64’ [-Wformat=]
           event.t.sec, event.t.nsec);
           ^
Documentation/ptp/testptp.c:505:5: warning: format ‘%lld’ expects argument
    of type ‘long long int’, but argument 2 has type ‘__s64’ [-Wformat=]
     (pct+2*i)->sec, (pct+2*i)->nsec);
     ^
Documentation/ptp/testptp.c:507:5: warning: format ‘%lld’ expects argument
    of type ‘long long int’, but argument 2 has type ‘__s64’ [-Wformat=]
     (pct+2*i+1)->sec, (pct+2*i+1)->nsec);
     ^
Documentation/ptp/testptp.c:509:5: warning: format ‘%lld’ expects argument
    of type ‘long long int’, but argument 2 has type ‘__s64’ [-Wformat=]
     (pct+2*i+2)->sec, (pct+2*i+2)->nsec);

This happens because __s64 is by default defined as "long" on ppc64,
not as "long long". However, to fix these warnings, it's possible to
define the __SANE_USERSPACE_TYPES__ so that __s64 gets defined to
"long long" on ppc64, too.

Signed-off-by: Thomas Huth <thuth@redhat.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet/mlx4: Handle return codes in mlx4_qp_attach_common
Robb Manes [Tue, 29 Sep 2015 15:03:37 +0000 (11:03 -0400)]
net/mlx4: Handle return codes in mlx4_qp_attach_common

Both new_steering_entry() and existing_steering_entry() return values
based on their success or failure, but currently they fall through
silently.  This can make troubleshooting difficult, as we were unable
to tell which one of these two functions returned errors or
specifically what code was returned.  This patch remedies that
situation by passing the return codes to err, which is returned by
mlx4_qp_attach_common() itself.

This also addresses a leak in the call to mlx4_bitmap_free() as well.

Signed-off-by: Robb Manes <rmanes@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge branch 'm68k-netdev-modular'
David S. Miller [Wed, 30 Sep 2015 04:11:13 +0000 (21:11 -0700)]
Merge branch 'm68k-netdev-modular'

Geert Uytterhoeven says:

====================
net: m68k: Allow modular build

This patch series makes the remaining m68k Ethernet drivers modular.
It's an alternative to the last 3 patches of Paul Gortmaker's series
"[PATCH net-next 0/6] make non-modular code explicitly non-modular".

Note that "[PATCH 5/5] net: macmace: Allow modular build" depends on
"[PATCH 4/5] m68k/mac: Export Peripheral System Controller (PSC) base
address to modules". Feel free to take the dependency through the netdev
tree to avoid modular build breakage.

This was compile-tested only (mac_defconfig + allmodconfig) due to lack
of hardware.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: macmace: Allow modular build
Geert Uytterhoeven [Tue, 29 Sep 2015 08:24:06 +0000 (10:24 +0200)]
net: macmace: Allow modular build

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agom68k/mac: Export Peripheral System Controller (PSC) base address to modules
Geert Uytterhoeven [Tue, 29 Sep 2015 08:24:05 +0000 (10:24 +0200)]
m68k/mac: Export Peripheral System Controller (PSC) base address to modules

If CONFIG_MACMACE=m:

    ERROR: psc [drivers/net/ethernet/apple/macmace.ko] undefined!

Add the missing export to fix this.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: hplance: Allow modular build
Geert Uytterhoeven [Tue, 29 Sep 2015 08:24:04 +0000 (10:24 +0200)]
net: hplance: Allow modular build

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: 7990: Export lance_poll() to modules
Geert Uytterhoeven [Tue, 29 Sep 2015 08:24:03 +0000 (10:24 +0200)]
net: 7990: Export lance_poll() to modules

If CONFIG_HPLANCE=m and CONFIG_NET_POLL_CONTROLLER=y:

    ERROR: "lance_poll" [drivers/net/ethernet/amd/hplance.ko] undefined!

Add the missing export to fix this.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: mac8390: Allow modular build
Geert Uytterhoeven [Tue, 29 Sep 2015 08:24:02 +0000 (10:24 +0200)]
net: mac8390: Allow modular build

The modular driver supports only one card, just like the built-in
driver.

Note that this limitation is a problem which affects all Nubus card
drivers, because they have to do all their own bus matching, because
Nubus still lacks the necessary driver model support.

Suggested-by: Finn Thain <fthain@telegraphics.com.au>
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agodsa: mv88e6xxx: Fix unsigned/signed issue
Andrew Lunn [Mon, 28 Sep 2015 23:53:48 +0000 (01:53 +0200)]
dsa: mv88e6xxx: Fix unsigned/signed issue

commit dea870242a9c ("dsa: mv88e6xxx: Allow speed/duplex of port to be
configured") leads to the following static checker warning:

        drivers/net/dsa/mv88e6xxx.c:585 mv88e6xxx_adjust_link()
        warn: unsigned 'ret' is never less than zero.

drivers/net/dsa/mv88e6xxx.c
   573  void mv88e6xxx_adjust_link(struct dsa_switch *ds, int port,
   574                             struct phy_device *phydev)
   575  {
   576          struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
   577          u32 ret, reg;
   578
   579          if (!phy_is_pseudo_fixed_link(phydev))
   580                  return;
   581
   582          mutex_lock(&ps->smi_mutex);
   583
   584          ret = _mv88e6xxx_reg_read(ds, REG_PORT(port), PORT_PCS_CTRL);
   585          if (ret < 0)

Make ret an int, which is the return type for _mv88e6xxx_reg_read()

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agodsa: mv88e6xxx: Enable forwarding for unknown to the CPU port
Andrew Lunn [Mon, 28 Sep 2015 23:50:56 +0000 (01:50 +0200)]
dsa: mv88e6xxx: Enable forwarding for unknown to the CPU port

Frames destined to an unknown address must be forwarded to the CPU
port. Otherwise incoming ARP, dhcp leases, etc, do not work.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge branch 'L3_master_device'
David S. Miller [Wed, 30 Sep 2015 03:41:10 +0000 (20:41 -0700)]
Merge branch 'L3_master_device'

David Ahern says:

====================
net: L3 master device

The VRF device is essentially a Layer 3 master device used to associate
netdevices with a specific routing table and to influence FIB lookups
via 'ip rules' and controlling the oif/iif used for the lookup.

This series generalizes the VRF into L3 master device, l3mdev. Similar
to switchdev it has a Kconfig option and separate set of operations
in net_device allowing it to be completely compiled out if not wanted.
The l3mdev methods rely on the 'master' aspect and use of
netdev_master_upper_dev_get_rcu to retrieve the master device from a
given netdevice if it is enslaved to an L3_MASTER.

The VRF device is converted to use the l3mdev operations. At the end the
vrf_ptr is no longer and removed, as are all direct references to VRF.
The end result is a much simpler implementation for VRF.

Thanks to Nikolay for suggestions (eg., use of the master linkage which
is the key to making this work) and to Roopa, Andy and Shrijeet for
early reviews.

v3
- added license header to l3mdev.c

- export symbols in l3mdev.c for use with GPL modules

- removed netdevice header from l3mdev.h (not needed) and fixed
  typo in comment

v2
- rebased to top of net-next
- addressed Niks comments (checking master, removing extra lines, and
  flipping the order of patches 1 and 2)
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: Move netif_index_is_l3_master to l3mdev.h
David Ahern [Wed, 30 Sep 2015 03:07:18 +0000 (20:07 -0700)]
net: Move netif_index_is_l3_master to l3mdev.h

Change CONFIG dependency to CONFIG_NET_L3_MASTER_DEV as well.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: Remove vrf header file
David Ahern [Wed, 30 Sep 2015 03:07:17 +0000 (20:07 -0700)]
net: Remove vrf header file

Move remaining structs to VRF driver and delete the vrf header file.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: Remove the now unused vrf_ptr
David Ahern [Wed, 30 Sep 2015 03:07:16 +0000 (20:07 -0700)]
net: Remove the now unused vrf_ptr

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: Replace calls to vrf_dev_get_rth
David Ahern [Wed, 30 Sep 2015 03:07:15 +0000 (20:07 -0700)]
net: Replace calls to vrf_dev_get_rth

Replace calls to vrf_dev_get_rth with l3mdev_get_rtable.
The check on the flow flags is handled in the l3mdev operation.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: Replace vrf_dev_table and friends
David Ahern [Wed, 30 Sep 2015 03:07:14 +0000 (20:07 -0700)]
net: Replace vrf_dev_table and friends

Replace calls to vrf_dev_table and friends with l3mdev_fib_table
and kin.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: Replace vrf_master_ifindex{, _rcu} with l3mdev equivalents
David Ahern [Wed, 30 Sep 2015 03:07:13 +0000 (20:07 -0700)]
net: Replace vrf_master_ifindex{, _rcu} with l3mdev equivalents

Replace calls to vrf_master_ifindex_rcu and vrf_master_ifindex with either
l3mdev_master_ifindex_rcu or l3mdev_master_ifindex.

The pattern:
    oif = vrf_master_ifindex(dev) ? : dev->ifindex;
is replaced with
    oif = l3mdev_fib_oif(dev);

And remove the now unused vrf macros.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: Add support for l3mdev ops to VRF driver
David Ahern [Wed, 30 Sep 2015 03:07:12 +0000 (20:07 -0700)]
net: Add support for l3mdev ops to VRF driver

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: Introduce L3 Master device abstraction
David Ahern [Wed, 30 Sep 2015 03:07:11 +0000 (20:07 -0700)]
net: Introduce L3 Master device abstraction

L3 master devices allow users of the abstraction to influence FIB lookups
for enslaved devices. Current API provides a means for the master device
to return a specific FIB table for an enslaved device, to return an
rtable/custom dst and influence the OIF used for fib lookups.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: Rename IFF_VRF_MASTER to IFF_L3MDEV_MASTER
David Ahern [Wed, 30 Sep 2015 03:07:10 +0000 (20:07 -0700)]
net: Rename IFF_VRF_MASTER to IFF_L3MDEV_MASTER

Rename IFF_VRF_MASTER to IFF_L3MDEV_MASTER and update the name of the
netif_is_vrf and netif_index_is_vrf macros.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge branch 'listener-refactoring-preparations'
David S. Miller [Tue, 29 Sep 2015 23:53:10 +0000 (16:53 -0700)]
Merge branch 'listener-refactoring-preparations'

Eric Dumazet says:

====================
tcp: listener refactoring preparations

This patch series makes changes to TCP/DCCP stacks so that
we can switch listener code to lockless mode.

This is done by marking const the listener socket in all
appropriate paths.

FastOpen code had to be changed to not dynamically allocate
a very small structure to make code simpler for following changes.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: prepare fastopen code for upcoming listener changes
Eric Dumazet [Tue, 29 Sep 2015 14:42:52 +0000 (07:42 -0700)]
tcp: prepare fastopen code for upcoming listener changes

While auditing TCP stack for upcoming 'lockless' listener changes,
I found I had to change fastopen_init_queue() to properly init the object
before publishing it.

Otherwise an other cpu could try to lock the spinlock before it gets
properly initialized.

Instead of adding appropriate barriers, just remove dynamic memory
allocations :
- Structure is 28 bytes on 64bit arches. Using additional 8 bytes
  for holding a pointer seems overkill.
- Two listeners can share same cache line and performance would suffer.

If we really want to save few bytes, we would instead dynamically allocate
whole struct request_sock_queue in the future.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: constify tcp_syn_flood_action() socket argument
Eric Dumazet [Tue, 29 Sep 2015 14:42:51 +0000 (07:42 -0700)]
tcp: constify tcp_syn_flood_action() socket argument

tcp_syn_flood_action() will soon be called with unlocked socket.
In order to avoid SYN flood warning being emitted multiple times,
use xchg().
Extend max_qlen_log and synflood_warned fields in struct listen_sock
to u32

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: constify tcp_v{4|6}_route_req() sock argument
Eric Dumazet [Tue, 29 Sep 2015 14:42:50 +0000 (07:42 -0700)]
tcp: constify tcp_v{4|6}_route_req() sock argument

These functions do not change the listener socket.
Goal is to make sure tcp_conn_request() is not messing with
listener in a racy way.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: cookie_init_sequence() cleanups
Eric Dumazet [Tue, 29 Sep 2015 14:42:49 +0000 (07:42 -0700)]
tcp: cookie_init_sequence() cleanups

Some common IPv4/IPv6 code can be factorized.
Also constify cookie_init_sequence() socket argument.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp/dccp: constify syn_recv_sock() method sock argument
Eric Dumazet [Tue, 29 Sep 2015 14:42:48 +0000 (07:42 -0700)]
tcp/dccp: constify syn_recv_sock() method sock argument

We'll soon no longer hold listener socket lock, these
functions do not modify the socket in any way.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: constify tcp_create_openreq_child() socket argument
Eric Dumazet [Tue, 29 Sep 2015 14:42:47 +0000 (07:42 -0700)]
tcp: constify tcp_create_openreq_child() socket argument

This method does not touch the listener socket.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agodccp: constify dccp_create_openreq_child() sock argument
Eric Dumazet [Tue, 29 Sep 2015 14:42:46 +0000 (07:42 -0700)]
dccp: constify dccp_create_openreq_child() sock argument

socket no longer needs to be read/write

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>