Vivien Didelot says:
====================
net: dsa: mv88e6xxx: fix hardware cross-chip bridging
In order to accelerate cross-chip switching of frames with the hardware,
the DSA Tag ports, used to interconnect switch devices, must learn SA
and DA addresses, and share the same FDB with the user ports.
The two first patches restore address learning on DSA links. This fixes
hardware cross-chip bridging in a VLAN filtering enabled system, which
implements a bridge group as a 802.1Q VLAN and thus share an isolated
address database between DSA and user ports.
The third patch changes the distinct default databases used for each
port, to the same address database. This fixes the hardware cross-chip
bridging in a VLAN filtering disabled system, where a bridge group gets
implemented only as a port-based VLAN.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
For hardware cross-chip bridging to work, user ports *and* DSA ports
need to share a common address database, in order to switch a frame to
the correct interconnected device.
This is currently working for VLAN filtering aware systems, since Linux
will implement a bridge group as a 802.1Q VLAN, which has its own FDB,
including DSA and CPU links as members.
However when the system doesn't support VLAN filtering, Linux only
relies on the port-based VLAN to implement a bridge group.
To fix hardware cross-chip bridging for such systems, set the same
default address database 0 for user and DSA ports, instead of giving
them all a different default database.
Note that the bridging code prevents frames to egress between unbridged
ports, and flushes FDB entries of a port when changing its STP state.
Also note that the FID 0 is special and means "all" for ATU operations,
but it's OK since it is used as a default forwarding address database.
Fixes: 2db9ce1fd9 ("net: dsa: mv88e6xxx: assign default FDB to ports")
Fixes: 466dfa0770 ("net: dsa: mv88e6xxx: assign dynamic FDB to bridges")
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In multi-chip systems, DSA Tag ports must learn SA addresses in order to
correctly switch frames between interconnected chips.
This fixes cross-chip hardware bridging in a VLAN filtering aware
system, because a bridge group gets implemented as an hardware 802.1Q
VLAN and thus DSA and user ports share the same FDB.
Fixes: 4c7ea3c079 ("net: dsa: mv88e6xxx: disable SA learning for DSA and CPU ports")
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Locking a port generates an hardware interrupt when a new SA address is
received. This enables CPU directed learning, which is needed for 802.1X
MAC authentication.
To disable automatic learning on a port, the only configuration needed
is to set its Port Association Vector to all zero.
Clear PAV when SA learning should be disabled instead of locking a port.
Fixes: 4c7ea3c079 ("net: dsa: mv88e6xxx: disable SA learning for DSA and CPU ports")
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Two different threads with different rds sockets may be in
rds_recv_rcvbuf_delta() via receive path. If their ports
both map to the same word in the congestion map, then
using non-atomic ops to update it could cause the map to
be incorrect. Lets use atomics to avoid such an issue.
Full credit to Wengang <wen.gang.wang@oracle.com> for
finding the issue, analysing it and also pointing out
to offending code with spin lock based fix.
Reviewed-by: Leon Romanovsky <leon@leon.nu>
Signed-off-by: Wengang Wang <wen.gang.wang@oracle.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
dp->dp_ack_seq is used in big endian format. We need to do the
big endianness conversion when we assign a value in host format
to it.
Signed-off-by: Qing Huang <qing.huang@oracle.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Okay we some driver fixes piled up, so time to get them up.
This time we have some odd fixes in hsu, edma, omap and xilinx.
Usual fixes and nothing special
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=bSYQ
-----END PGP SIGNATURE-----
Merge tag 'dmaengine-fix-4.6-rc4' of git://git.infradead.org/users/vkoul/slave-dma
Pull dmaengine fixes from Vinod Koul:
"This time we have some odd fixes in hsu, edma, omap and xilinx.
Usual fixes and nothing special"
* tag 'dmaengine-fix-4.6-rc4' of git://git.infradead.org/users/vkoul/slave-dma:
dmaengine: dw: fix master selection
dmaengine: edma: special case slot limit workaround
dmaengine: edma: Remove dynamic TPTC power management feature
dmaengine: vdma: don't crash when bad channel is requested
dmaengine: omap-dma: Do not suppress interrupts for memcpy
dmaengine: omap-dma: Fix polled channel completion detection and handling
dmaengine: hsu: correct use of channel status register
dmaengine: hsu: correct residue calculation of active descriptor
dmaengine: hsu: set HSU_CH_MTSR to memory width
Pull locking fixlet from Ingo Molnar:
"Fixes a build warning on certain Kconfig combinations"
* 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking/lockdep: Fix print_collision() unused warning
Akinobu Mita says:
====================
net: w5100: add support W5100/W5200 for SPI interface
This series add support for Wiznet W5100 and W5200 for SPI interface.
We can easily find the ethernet modules and shield for Arduino with
these chips for purchase. I've tested them with BeagleBone.
Wiznet W5100 for mmio access has already supported by w5100 driver.
In order to share the code between mmio mode and SPI mode, this series
firstly adds ability to support another register access interface to
the existing w5100 driver. This ground work also requires to introduce
workqueue and threaded irq because SPI transfers are callable only from
contexts that can sleep unlike mmio access.
The latter part of this series adds w5100-spi driver which actually
support W5100 and W5200 for SPI interface. Supporting W5100 is
straight forward because it only required to add a register access
interface by the SPI transfer. W5100 and W5200 have similar memory
map which justifies adding W5200 support to w5100 driver.
* Changes from v2 to v3
- Add comment for reg_lock
- Add ability to allocate ops specific data structure
- Allocate w5200 ops specific data structure to put DMA-safe buffer
- Add missing chip_id assignment for w5100_*_ops
* Changes from v1 to v2
- Use a plain single pointer instead of SKB queue, spotted by David S. Miller
- Correct timeout period in w5100_command
- Use spi_write_then_read instead of spi_write which needs DMA-safe buffer
- Support W5200
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds support for W5200 chip.
W5100 and W5200 have similar memory map although some of their offsets
are different. The register access sequences between them are different
but w5100 driver has abstraction layer for difference bus interface
modes so it is easy to add W5200 support to w5100 and w5100-spi drivers.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Mike Sinkovsky <msink@permonline.ru>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds new w5100-spi driver which shares the bus interface
independent code with existing w5100 driver.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Mike Sinkovsky <msink@permonline.ru>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
SPI transfer routines are callable only from contexts that can sleep.
This adds ability to tell the core driver that the interface mode
cannot access w5100 register on atomic contexts. In this case,
workqueue and threaded irq are required.
This also corrects timeout period waiting for command register to be
automatically cleared because the latency of the register access with
SPI transfer can be interfered by other contexts.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Mike Sinkovsky <msink@permonline.ru>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The w5100 driver currently only supports direct and indirect bus
interface mode which use MMIO space for accessing w5100 registers.
In order to support SPI interface mode which is supported by W5100 chip,
this makes the bus interface abstraction layer more generic so that
separated w5100-spi driver can use w5100 driver as core module.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Mike Sinkovsky <msink@permonline.ru>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of sprinkle mmiowb over the driver code, move it into primary
register write callbacks. (w5100_write, w5100_write16, w5100_writebuf)
This is a preparation for supporting SPI interface which doesn't use
MMIO for accessing w5100 registers.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Mike Sinkovsky <msink@permonline.ru>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
We only need to do the synchronize_net dance once for both, ipv4 and
ipv6 sockets, thus removing one synchronize_net in case both sockets get
dismantled.
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Due to the fact that the udp socket is destructed asynchronously in a
work queue, we have some nondeterministic behavior during shutdown of
vxlan tunnels and creating new ones. Fix this by keeping the destruction
process synchronous in regards to the user space process so IFF_UP can
be reliably set.
udp_tunnel_sock_release destroys vs->sock->sk if reference counter
indicates so. We expect to have the same lifetime of vxlan_sock and
vxlan_sock->sock->sk even in fast paths with only rcu locks held. So
only destruct the whole socket after we can be sure it cannot be found
by searching vxlan_net->sock_list.
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Jiri Benc <jbenc@redhat.com>
Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
When __vlan_insert_tag() fails from skb_vlan_push() path due to the
skb_cow_head(), we need to undo the __skb_push() in the error path
as well that was done earlier to move skb->data pointer to mac header.
Moreover, I noticed that when in the non-error path the __skb_pull()
is done and the original offset to mac header was non-zero, we fixup
from a wrong skb->data offset in the checksum complete processing.
So the skb_postpush_rcsum() really needs to be done before __skb_pull()
where skb->data still points to the mac header start and thus operates
under the same conditions as in __vlan_insert_tag().
Fixes: 93515d53b1 ("net: move vlan pop/push functions into common code")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When tested the PHY SGMII Loopback:
1.set the LOOPBACK bit,
2.set the autoneg to AUTONEG_DISABLE, it calls the
genphy_setup_forced which will clear the bit.
The BMCR_LOOPBACK bit should be preserved.
As Florian pointed out that other bits should be preserved too.
So I make the BMCR_ISOLATE and BMCR_PDOWN as well.
Signed-off-by: Weidong Wang <wangweidong1@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull block fixes from Jens Axboe:
"A few fixes for the current series. This contains:
- Two fixes for NVMe:
One fixes a reset race that can be triggered by repeated
insert/removal of the module.
The other fixes an issue on some platforms, where we get probe
timeouts since legacy interrupts isn't working. This used not to
be a problem since we had the worker thread poll for completions,
but since that was killed off, it means those poor souls can't
successfully probe their NVMe device. Use a proper IRQ check and
probe (msi-x -> msi ->legacy), like most other drivers to work
around this. Both from Keith.
- A loop corruption issue with offset in iters, from Ming Lei.
- A fix for not having the partition stat per cpu ref count
initialized before sending out the KOBJ_ADD, which could cause user
space to access the counter prior to initialization. Also from
Ming Lei.
- A fix for using the wrong congestion state, from Kaixu Xia"
* 'for-linus' of git://git.kernel.dk/linux-block:
block: loop: fix filesystem corruption in case of aio/dio
NVMe: Always use MSI/MSI-x interrupts
NVMe: Fix reset/remove race
writeback: fix the wrong congested state variable definition
block: partition: initialize percpuref before sending out KOBJ_ADD
Pull libnvdimm fixes from Ross Zwisler:
"Two fixes:
- Fix memcpy_from_pmem() to fallback to memcpy() for architectures
where CONFIG_ARCH_HAS_PMEM_API=n.
- Add a comment explaining why we write data twice when clearing
poison in pmem_do_bvec().
This has passed a boot test on an X86_32 config, which was the
architecture where issue #1 above was first noticed"
Dan Williams adds:
"We're giving this multi-maintainer setup a shot, so expect libnvdimm
pull requests from either Ross or I going forward"
* 'libnvdimm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
libnvdimm, pmem: clarify the write+clear_poison+write flow
pmem: fix BUG() error in pmem.h:48 on X86_32
In the v4.4 cycle, we relaxed the requirement for assigning mtd->owner, but we
didn't remove this error case. It's hit only by drivers that are both:
(a) using nand_scan() directly and
(b) built as modules
We haven't seen explicit complaints about this (most use cases don't fit one or
both of the above), but we should definitely not be BUG()'ing here.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQIcBAABAgAGBQJXEWZUAAoJEFySrpd9RFgtuaYQAIjfmoefoDJuIw3QQE5tLNMc
xTj07f+V/Nzhats69zvjx4LNFF6R339IBFNq7WK+q7l7VbW6/XSa8XaQkgpO+wMI
Sqzkty4YIxQGucITVvauv82JW0K421XVUBAzeNt+4oT72RvMnXhOPka1LQXf3Ezh
vGb7ZGea7ZYuu060OQAGxJlrlDA5x7yhsvs3hpOT9hjx0jXr0JZXmohvFRqSiXqz
Rx60rmhpl37WJg46OX/LYRAF34StF8btDZsQVdpY75T9QtxD3vWvvlMHfJ4cQXRI
L3qUIGGj03tD0/4nf6/p+WMjxgctDqROVxWzLHOvTa+xIFR7iU8yra09fTbylOGL
C3I0B6kW05YJrY3TegL8EhNbotciC9SeCtbeRwbndyO1Jox88xGHovwWFN5++3Q+
4ahiP57cV/py1L2BgeYdPmHES6T4GLH/4y2MaMZr4hfGVxHbKN1/JkLp9cNqCSzZ
MtnEevbHYq4eN1rYULQCzLRxawhYjeY77cMYip6Ynu24UWUJmfdaroJncfmmBOhP
7NvWjtxpsjonrCVOdb4aq9UPTc3PxlyR4kZyzg7+Vx4eOByK8hFkfjckehJ64DCs
ac+r5d2LmaLqAMIvQt4ApwwzRzj4ST6CEWBEyKrvwwF4B4nHhkbf9Pf4IrzGmaao
W4dws70o9WUwGIv0Rw3D
=wxaC
-----END PGP SIGNATURE-----
Merge tag 'for-linus-20160415' of git://git.infradead.org/linux-mtd
Pull MTD fix from Brian Norris:
"One MTD fix for v4.6-rc4:
In the v4.4 cycle, we relaxed the requirement for assigning
mtd->owner, but we didn't remove this error case. It's hit only
by drivers that are both:
(a) using nand_scan() directly
and
(b) built as modules
We haven't seen explicit complaints about this (most use cases don't
fit one or both of the above), but we should definitely not be
BUG()'ing here"
* tag 'for-linus-20160415' of git://git.infradead.org/linux-mtd:
mtd: nand: Drop mtd.owner requirement in nand_scan
- Restore similar old behaviour when assigning mmcblk device indexes
MMC host:
- tegra: Disable UHS-I modes for Tegra124 to fix regression
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=nnxN
-----END PGP SIGNATURE-----
Merge tag 'mmc-v4.6-rc3' of git://git.linaro.org/people/ulf.hansson/mmc
Pull MMC fixes from Ulf Hansson:
"Here are a couple of mmc fixes intended for v4.6 rc4.
Regarding the fix for the regression about mmcblk device indexes. The
approach taken to solve the problem seems to be good enough. There
were some discussions around the solution, but it seems like people
were happy about it in the end.
MMC core:
- Restore similar old behaviour when assigning mmcblk device indexes
MMC host:
- tegra: Disable UHS-I modes for Tegra124 to fix regression"
* tag 'mmc-v4.6-rc3' of git://git.linaro.org/people/ulf.hansson/mmc:
mmc: tegra: Disable UHS-I modes for Tegra124
mmc: block: Use the mmc host device index as the mmcblk device index
Pull drm fixes from Dave Airlie:
"This contains fixes for exynos, amdgpu, radeon, i915 and qxl.
It also contains some fixes to the core drm edid parser.
qxl:
- fix for a cursor hotspot issue
radeon:
- some MST fixes that I've been running locally and make my monitor a
bit happier
exynos:
- fix some regressions and build fixes
amdgpu:
- a couple of small fixes
i915:
- two DP MST fixes and a couple of other regression fixes
Nothing too out of the ordinary or surprising at this point"
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
drm/exynos: Use VIDEO_SAMSUNG_S5P_G2D=n as G2D Kconfig dependency
drm/exynos: fix a warning message
drm/exynos: mic: fix an error code
drm/exynos: fimd: fix broken dp_clock control
drm/exynos: build fbdev code conditionally
drm/exynos: fix adjusted_mode pointer in exynos_plane_mode_set
drm/exynos: fix error handling in exynos_drm_subdrv_open
drm/amd/amdgpu: fix irq domain remove for tonga ih
drm/i915: fix deadlock on lid open
drm/radeon: use helper for mst connector dpms.
drm/radeon/mst: port some MST setup code from DAL.
drm/amdgpu: add invisible pin size statistic
drm/edid: Fix DMT 1024x768@43Hz (interlaced) timings
drm/i915: Exit cherryview_irq_handler() after one pass
drm/i915: Call intel_dp_mst_resume() before resuming displays
drm/i915: Fix race condition in intel_dp_destroy_mst_connector()
drm/edid: Fix parsing of EDID 1.4 Established Timings III descriptor
drm/edid: Fix EDID Established Timings I and II
drm/qxl: fix cursor position with non-zero hotspot
Pull parisc ftrace fixes from Helge Deller:
"This is (most likely) the last pull request for v4.6 for the parisc
architecture.
It fixes the FTRACE feature for parisc, which is horribly broken since
quite some time and doesn't even compile. This patch just fixes the
bare minimum (it actually removes more lines than it adds), so that
the function tracer works again on 32- and 64bit kernels.
I've queued up additional patches on top of this patch which e.g. add
the syscall tracer, but those have to wait for the merge window for
v4.7."
* 'parisc-4.6-4' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
parisc: Fix ftrace function tracer
Xin Long says:
====================
sctp: support sctp_diag in kernel
This patchset will add sctp_diag module to implement diag interface on
sctp in kernel.
For a listening sctp endpoint, we will just dump it's ep info.
For a sctp connection, we will the assoc info and it's ep info.
The ss dump will looks like:
[iproute2]# ./misc/ss --sctp -n -l
State Recv-Q Send-Q Local Address:Port Peer Address:Port
LISTEN 0 128 172.16.254.254:8888 *:*
LISTEN 0 5 127.0.0.1:1234 *:*
LISTEN 0 5 127.0.0.1:1234 *:*
- ESTAB 0 0 127.0.0.1%lo:1234 127.0.0.1:4321
LISTEN 0 128 172.16.254.254:8888 *:*
- ESTAB 0 0 172.16.254.254%eth1:8888 172.16.253.253:8888
- ESTAB 0 0 172.16.254.254%eth1:8888 172.16.1.1:8888
- ESTAB 0 0 172.16.254.254%eth1:8888 172.16.1.2:8888
- ESTAB 0 0 172.16.254.254%eth1:8888 172.16.2.1:8888
- ESTAB 0 0 172.16.254.254%eth1:8888 172.16.2.2:8888
- ESTAB 0 0 172.16.254.254%eth1:8888 172.16.3.1:8888
- ESTAB 0 0 172.16.254.254%eth1:8888 172.16.3.2:8888
LISTEN 0 0 127.0.0.1:4321 *:*
- ESTAB 0 0 127.0.0.1%lo:4321 127.0.0.1:1234
The entries with '- ESTAB' are the assocs, some of them may belong to
the same endpoint. So we will dump the parent endpoint first, like the
entry with 'LISTEN'. then dump the assocs. ep and assocs entries will
be dumped in right order so that ss can show them in tree format easily.
Besides, this patchset also simplifies sctp proc codes, cause it has
some similar codes with sctp diag in sctp transport traversal.
v1->v2:
1. inet_diag_get_handler needs to return it as const.
2. merge 5/7 into 2/7 of v1.
v2->v3:
do some improvements and fixes in patch 1-4, see the details in
each patch's comment.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
When rhashtable_walk_init return err, no release function should be
called, and when rhashtable_walk_start return err, we should only invoke
rhashtable_walk_exit to release the source.
But now when sctp_transport_walk_start return err, we just call
rhashtable_walk_stop/exit, and never care about if rhashtable_walk_init
or start return err, which is so bad.
We will fix it by calling rhashtable_walk_exit if rhashtable_walk_start
return err in sctp_transport_walk_start, and if sctp_transport_walk_start
return err, we do not need to call sctp_transport_walk_stop any more.
For sctp proc, we will use 'iter->start_fail' to decide if we will call
rhashtable_walk_stop/exit.
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In sctp proc, these three functions in remaddrs and assocs are the
same. we should merge them into one.
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This one will implement all the interface of inet_diag, inet_diag_handler.
which includes sctp_diag_dump, sctp_diag_dump_one and sctp_diag_get_info.
It will work as a module, and register inet_diag_handler when loading.
v2->v3:
- fix the mistake in inet_assoc_attr_size().
- change inet_diag_msg_laddrs_fill() name to inet_diag_msg_sctpladdrs_fill.
- change inet_diag_msg_paddrs_fill() name to inet_diag_msg_sctpaddrs_fill.
- add inet_diag_msg_sctpinfo_fill() to make asoc/ep fill code clearer.
- add inet_diag_msg_sctpasoc_fill() to make asoc fill code clearer.
- merge inet_asoc_diag_fill() and inet_ep_diag_fill() to
inet_sctp_diag_fill().
- call sctp_diag_get_info() directly, instead by handler, cause the caller
is in the same file with it.
- call lock_sock in sctp_tsp_dump_one() to make sure we call get sctp info
safely.
- after lock_sock(sk), we should check sk != assoc->base.sk.
- change mem[SK_MEMINFO_WMEM_ALLOC] to asoc->sndbuf_used for asoc dump when
asoc->ep->sndbuf_policy is set. don't use INET_DIAG_MEMINFO attr any more.
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
inet_diag_msg_common_fill is used to fill the diag msg common info,
we need to use it in sctp_diag as well, so export it.
inet_diag_msg_attrs_fill is used to fill some common attrs info between
sctp diag and tcp diag.
v2->v3:
- do not need to define and export inet_diag_get_handler any more.
cause all the functions in it are in sctp_diag.ko, we just call
them in sctp_diag.ko.
- add inet_diag_msg_attrs_fill to make codes clear.
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
For some main variables in sctp.ko, we couldn't export it to other modules,
so we have to define some api to access them.
It will include sctp transport and endpoint's traversal.
There are some transport traversal functions for sctp_diag, we can also
use it for sctp_proc. cause they have the similar situation to traversal
transport.
v2->v3:
- rhashtable_walk_init need the parameter gfp, because of recent upstrem
update
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sctp_diag will dump some important details of sctp's assoc or ep, we use
sctp_info to describe them, sctp_get_sctp_info to get them, and export
it to sctp_diag.ko.
v2->v3:
- we will not use list_for_each_safe in sctp_get_sctp_info, cause
all the callers of it will use lock_sock.
- fix the holes in struct sctp_info with __reserved* field.
because sctp_diag is a new feature, and sctp_info is just for now,
it may be changed in the future.
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
SCTP already serializes access to rcvbuf through its sock lock:
sctp_recvmsg takes it right in the start and release at the end, while
rx path will also take the lock before doing any socket processing. On
sctp_rcv() it will check if there is an user using the socket and, if
there is, it will queue incoming packets to the backlog. The backlog
processing will do the same. Even timers will do such check and
re-schedule if an user is using the socket.
Simplifying this will allow us to remove sctp_skb_list_tail and get ride
of some expensive lockings. The lists that it is used on are also
mangled with functions like __skb_queue_tail and __skb_unlink in the
same context, like on sctp_ulpq_tail_event() and sctp_clear_pd().
sctp_close() will also purge those while using only the sock lock.
Therefore the lockings performed by sctp_skb_list_tail() are not
necessary. This patch removes this function and replaces its calls with
just skb_queue_splice_tail_init() instead.
The biggest gain is at sctp_ulpq_tail_event(), because the events always
contain a list, even if it's queueing a single skb and this was
triggering expensive calls to spin_lock_irqsave/_irqrestore for every
data chunk received.
As SCTP will deliver each data chunk on a corresponding recvmsg, the
more effective the change will be.
Before this patch, with chunks with 30 bytes:
netperf -t SCTP_STREAM -H 192.168.1.2 -cC -l 60 -- -m 30 -S 400000
400000 -s 400000 400000
on a 10Gbit link with 1500 MTU:
SCTP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 () port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
425984 425984 30 60.00 137.45 7.34 7.36 52.504 52.608
With it:
SCTP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 () port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
425984 425984 30 60.00 179.10 7.97 6.70 43.740 36.788
Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mahameed says:
====================
mlx5_core: mlx5_ifc updates
This series include mlx5_core updates for both net-next and rdma
trees for 4.7 kernel cycle. This is the only shared code planned
for 4.7 between rdma and net trees. Hopefully, this will prevent
future conflicts when merging between ib-next and net-next once
4.7 cycle is over and merge window is opened.
Both Mellanox rdma and net submissions will proceed once this series
is applied into both trees.
Future shared code will be sent to both maintainers as pull requests
from Mellanox's kernel.org tree.
We have included all the maintainers of respective drivers.
Kindly review the change and let us know in case of any review comments.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Adding the needed mlx5_ifc hardware bits and structs
for the following feature:
* Add vport to steering commands for SRIOV ACL support
* Add mlcr, pcmr and mcia registers for dump module EEPROM
* Add support for FCS, baeacon led and disable_link bits to
hca caps
* Add CQE period mode bit in CQ context for CQE based CQ
moderation support
* Add umr SQ bit for fragmented memory registration
* Add needed bits and caps for Striding RQ support
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All reserved fields after early_vf_enable are off by 1, since
early_vf_enable was not explicitly declared as array of size 1.
Reserved field before cqe_zip had a wrong size, it should
be 0x80 + 0x3f.
Fixes: b084444459 ("net/mlx5_core: Introduce access function to read internal timer ")
Fixes: b4ff3a36d3 ("net/mlx5: Use offset based reserved field names in the IFC header file")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Manish Chopra says:
====================
qed/qede: Add tunneling support
This patch series adds support for VXLAN, GRE and GENEVE tunnels
to be used over this driver. With this support, adapter can perform
TSO offload, inner/outer checksums offloads on TX and RX for
encapsulated packets.
V1->V2 [ Comments from Jesse Gross incorporated ]
* Drop general infrastructure change patch.
"net: Make vxlan/geneve default udp ports public"
* Remove by default Linux default UDP ports configurations in driver.
Instead, use general registration APIs for UDP port configurations
* Removing .ndo_features_check - we will add it later with proper change.
Please consider applying this series to net-next.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch enables netdev tunneling features and adds
TX/RX fastpath support for tunneling in driver.
Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Ariel Elior <Ariel.Elior@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch enables GENEVE tunnel on the adapter and
add support for driver hooks to configure UDP ports
for GENEVE tunnel offload to be performed by the adapter.
Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Ariel Elior <Ariel.Elior@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch enables VXLAN tunnel on the adapter and
add support for driver hooks to configure UDP ports
for VXLAN tunnel offload to be performed by the adapter.
Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Ariel Elior <Ariel.Elior@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds various structure/APIs needed to configure/enable different
tunnel [VXLAN/GRE/GENEVE] parameters on the adapter.
Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Ariel Elior <Ariel.Elior@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds support for the newer version 1 of the HSR
networking standard. Version 0 is still default and the new
version has to be selected via iproute2.
Main changes are in the supervision frame handling and its
ethertype field.
Signed-off-by: Peter Heise <peter.heise@airbus.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ACPI specification does not specify the state of data after a clear
poison operation. Potential future libnvdimm bus implementations for
other architectures also might not specify or disagree on the state of
data after clear poison. Clarify why we write twice.
Reported-by: Jeff Moyer <jmoyer@redhat.com>
Reported-by: Vishal Verma <vishal.l.verma@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Eric Dumazet says:
====================
tcp: final work on SYNFLOOD behavior
In the first patch, I remove the costly association of SYNACK+COOKIES
to a listener. I believe other parts of the stack should be ready.
The second patch removes a useless write into listener socket
in tcp_rcv_state_process(), incurring false sharing in
tcp_conn_request()
Performance under SYNFLOOD goes from 3.2 Mpps to 6 Mpps.
Test was using a single TCP listener, on a host with 8 RX queues
on the NIC, and 24 cores (48 ht)
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller