Since commit 0ddcf43d5d ("ipv4: FIB Local/MAIN table collapse") the
local table uses the same trie allocated for the main table when custom
rules are not in use.
When a net namespace is dismantled, the main table is flushed and freed
(via an RCU callback) before the local table. In case the callback is
invoked before the local table is iterated, a use-after-free can occur.
Fix this by iterating over the FIB tables in reverse order, so that the
main table is always freed after the local table.
v3: Reworded comment according to Alex's suggestion.
v2: Add a comment to make the fix more explicit per Dave's and Alex's
feedback.
Fixes: 0ddcf43d5d ("ipv4: FIB Local/MAIN table collapse")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Acked-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make sure to check both return code fields before processing the
response. Otherwise we risk operating on invalid data.
Fixes: c9475369bd ("s390/qeth: rework RX/TX checksum offload")
Signed-off-by: Julian Wiedmann <jwi@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When we receive a JOIN message from a peer member, the message may
contain an advertised window value ADV_IDLE that permits removing the
member in question from the tipc_group::congested list. However, since
the removal has been made conditional on that the advertised window is
*not* ADV_IDLE, we miss this case. This has the effect that a sender
sometimes may enter a state of permanent, false, broadcast congestion.
We fix this by unconditinally removing the member from the congested
list before calling tipc_member_update(), which might potentially sort
it into the list again.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
kernel config fragement CONFIG_NUMA=y is need for reuseport_bpf_numa.
Signed-off-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Misc fixes for mlx5 core and mlx5 netdev driver.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEbBAABAgAGBQJaOYOtAAoJEEg/ir3gV/o+1/0H+I9yY/HYQIU0Cl08yNvYKBcS
KuhGpeJCH20rPQPrcPTG3zN0KF6DZKjwsQOwxdE5pUXqQNUuyuogUZCuLYB4OElL
P4b1o5G377sTcDQ7jH2YAWD5hO8c5vKxsDbvN+kJaUkK+SHvjhLdvC5teUPf58rx
tlDcWzdp3w1nBWeoLbASXTiqPYyA8vGkjOWWiQxBZoD4A4U7QpRKsEKaWd6g3mtH
mnKVd8sczIFCHoHCQ3e+efJMz478YvX2rzdKZ8eycAMkQHBIny0fZzc4IiFy5ZXN
L2CUGesr8x1CZ9dtK+OEw1STpalD0kpCfjRhYd2B7X0KgY/FN+7vgnpmBtxdzA==
=G//q
-----END PGP SIGNATURE-----
Merge tag 'mlx5-fixes-2017-12-19' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux
Saeed Mahameed says:
===================
Mellanox, mlx5 fixes 2017-12-19
The follwoing series includes some fixes for mlx5 core and etherent
driver.
Please pull and let me know if there is any problem.
This series doesn't introduce any conflict with the ongoing mlx5 for-next
submission.
For -stable:
kernels >= v4.7.y
("net/mlx5e: Fix possible deadlock of VXLAN lock")
("net/mlx5e: Add refcount to VXLAN structure")
("net/mlx5e: Prevent possible races in VXLAN control flow")
("net/mlx5e: Fix features check of IPv6 traffic")
kernels >= v4.9.y
("net/mlx5: Fix error flow in CREATE_QP command")
("net/mlx5: Fix rate limit packet pacing naming and struct")
kernels >= v4.13.y
("net/mlx5: FPGA, return -EINVAL if size is zero")
kernels >= v4.14.y
("Revert "mlx5: move affinity hints assignments to generic code")
All above patches apply and compile with no issues on corresponding -stable.
===================
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit f5775e0b61 ("x86/xen: discard RAM regions above the maximum
reservation") left host memory not assigned to dom0 as available for
memory hotplug.
Unfortunately this also meant that those regions could be used by
others. Specifically, commit fa564ad963 ("x86/PCI: Enable a 64bit BAR
on AMD Family 15h (Models 00-1f, 30-3f, 60-7f)") may try to map those
addresses as MMIO.
To prevent this mark unallocated host memory as E820_TYPE_UNUSABLE (thus
effectively reverting f5775e0b61) and keep track of that region as
a hostmem resource that can be used for the hotplug.
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
If a bio is throttled and split after throttling, the bio could be
resubmited and enters the throttling again. This will cause part of the
bio to be charged multiple times. If the cgroup has an IO limit, the
double charge will significantly harm the performance. The bio split
becomes quite common after arbitrary bio size change.
To fix this, we always set the BIO_THROTTLED flag if a bio is throttled.
If the bio is cloned/split, we copy the flag to new bio too to avoid a
double charge. However, cloned bio could be directed to a new disk,
keeping the flag be a problem. The observation is we always set new disk
for the bio in this case, so we can clear the flag in bio_set_dev().
This issue exists for a long time, arbitrary bio size change just makes
it worse, so this should go into stable at least since v4.2.
V1-> V2: Not add extra field in bio based on discussion with Tejun
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: stable@vger.kernel.org
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Shaohua Li <shli@fb.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jakub Kicinski says:
===================
cls_bpf: fix offload state tracking with block callbacks
After introduction of block callbacks classifiers can no longer track
offload state. cls_bpf used to do that in an attempt to move common
code from drivers to the core. Remove that functionality and fix
drivers.
The user-visible bug this is fixing is that trying to offload a second
filter would trigger a spurious DESTROY and in turn disable the already
installed one.
===================
Signed-off-by: David S. Miller <davem@davemloft.net>
After TC offloads were converted to callbacks we have no choice
but keep track of the offloaded filter in the driver.
The check for nn->dp.bpf_offload_xdp was a stop gap solution
to make sure failed TC offload won't disable XDP, it's no longer
necessary. nfp_net_bpf_offload() will return -EBUSY on
TC vs XDP conflicts.
Fixes: 3f7889c4c7 ("net: sched: cls_bpf: call block callbacks for offload")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
cls_bpf used to take care of tracking what offload state a filter
is in, i.e. it would track if offload request succeeded or not.
This information would then be used to issue correct requests to
the driver, e.g. requests for statistics only on offloaded filters,
removing only filters which were offloaded, using add instead of
replace if previous filter was not added etc.
This tracking of offload state no longer functions with the new
callback infrastructure. There could be multiple entities trying
to offload the same filter.
Throw out all the tracking and corresponding commands and simply
pass to the drivers both old and new bpf program. Drivers will
have to deal with offload state tracking by themselves.
Fixes: 3f7889c4c7 ("net: sched: cls_bpf: call block callbacks for offload")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
(I can trivially verify that that idr_remove in cleanup_net happens
after the network namespace count has dropped to zero --EWB)
Function get_net_ns_by_id() does not check for net::count
after it has found a peer in netns_ids idr.
It may dereference a peer, after its count has already been
finaly decremented. This leads to double free and memory
corruption:
put_net(peer) rtnl_lock()
atomic_dec_and_test(&peer->count) [count=0] ...
__put_net(peer) get_net_ns_by_id(net, id)
spin_lock(&cleanup_list_lock)
list_add(&net->cleanup_list, &cleanup_list)
spin_unlock(&cleanup_list_lock)
queue_work() peer = idr_find(&net->netns_ids, id)
| get_net(peer) [count=1]
| ...
| (use after final put)
v ...
cleanup_net() ...
spin_lock(&cleanup_list_lock) ...
list_replace_init(&cleanup_list, ..) ...
spin_unlock(&cleanup_list_lock) ...
... ...
... put_net(peer)
... atomic_dec_and_test(&peer->count) [count=0]
... spin_lock(&cleanup_list_lock)
... list_add(&net->cleanup_list, &cleanup_list)
... spin_unlock(&cleanup_list_lock)
... queue_work()
... rtnl_unlock()
rtnl_lock() ...
for_each_net(tmp) { ...
id = __peernet2id(tmp, peer) ...
spin_lock_irq(&tmp->nsid_lock) ...
idr_remove(&tmp->netns_ids, id) ...
... ...
net_drop_ns() ...
net_free(peer) ...
} ...
|
v
cleanup_net()
...
(Second free of peer)
Also, put_net() on the right cpu may reorder with left's cpu
list_replace_init(&cleanup_list, ..), and then cleanup_list
will be corrupted.
Since cleanup_net() is executed in worker thread, while
put_net(peer) can happen everywhere, there should be
enough time for concurrent get_net_ns_by_id() to pick
the peer up, and the race does not seem to be unlikely.
The patch fixes the problem in standard way.
(Also, there is possible problem in peernet2id_alloc(), which requires
check for net::count under nsid_lock and maybe_get_net(peer), but
in current stable kernel it's used under rtnl_lock() and it has to be
safe. Openswitch begun to use peernet2id_alloc(), and possibly it should
be fixed too. While this is not in stable kernel yet, so I'll send
a separate message to netdev@ later).
Cc: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Fixes: 0c7aecd4bd "netns: add rtnl cmd to add and get peer netns ids"
Reviewed-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Reviewed-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gregory CLEMENT says:
====================
Few mvneta fixes
here it is a small series of fixes found on the mvneta driver. They
had been already used in the vendor kernel and are now ported to
mainline.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
There are few reasons in mvneta_rx_swbm() function when received packet
is dropped. mvneta_rx_error() should be called only if error bit [16]
is set in rx descriptor.
[gregory.clement@free-electrons.com: add fixes tag]
Cc: stable@vger.kernel.org
Fixes: dc35a10f68 ("net: mvneta: bm: add support for hardware buffer management")
Signed-off-by: Yelena Krivosheev <yelena@marvell.com>
Tested-by: Dmitri Epshtein <dima@marvell.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When adding the RX queue association with each CPU, a typo was made in
the mvneta_cleanup_rxqs() function. This patch fixes it.
[gregory.clement@free-electrons.com: add commit log and fixes tag]
Cc: stable@vger.kernel.org
Fixes: 2dcf75e279 ("net: mvneta: Associate RX queues with each CPU")
Signed-off-by: Yelena Krivosheev <yelena@marvell.com>
Tested-by: Dmitri Epshtein <dima@marvell.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When port connect to PHY in polling mode (with poll interval 1 sec),
port and phy link status must be synchronize in order don't loss link
change event.
[gregory.clement@free-electrons.com: add fixes tag]
Cc: <stable@vger.kernel.org>
Fixes: c5aff18204 ("net: mvneta: driver for Marvell Armada 370/XP network unit")
Signed-off-by: Yelena Krivosheev <yelena@marvell.com>
Tested-by: Dmitri Epshtein <dima@marvell.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is a fairly large set of fixes, they've been delayed partly as more
and more keep coming in. Most of them are very small driver specific
fixes, the biggest individual thing is the revert of the rcar IOMMU
support - it was causing problems and there wasn't the confidence that
it could be resolved sensibly. There's also a relatively large change
in the Freescale SSI controller which resolves some issues with the
AC'97 mode, these aren't that large in the grand scheme of things and
reflect some fairly thorough review and testing.
-----BEGIN PGP SIGNATURE-----
iQFHBAABCgAxFiEEreZoqmdXGLWf4p/qJNaLcl1Uh9AFAlo6kHETHGJyb29uaWVA
a2VybmVsLm9yZwAKCRAk1otyXVSH0KWJB/42n+WPZlJQprXsvIVOP6nZ2DNCULVN
93/nBWF3Dtb/qdcFcUR84iLX8JySCnlqtoPw4OPkTvy47FNvha1e0j6W0KVRBeia
1Ugy2RQlFo9uN8UJkr7anDbE++7a0QTWl23jyblSSa56pK2uNlmuNl+KA1MgT9HC
EHloE3JbiG0x9f4IzxiOMNXk3zwXPSZhWrK6DVMIlBn2MYOAfSv2aT+aosZeSbGY
kuoNw1/PBgXlP7PVMI9Fin4pCSg7UUBRjTXIpoEupnRG7krANtUA9qjKZUYeTiDh
TNLyPawtFk8GCsEdxiqSrWwGiTpzdrhdfGtttuGq5OE6FL1bDmNdhGjs
=IV5j
-----END PGP SIGNATURE-----
Merge tag 'asoc-fix-v4.15-rc4' of https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus
ASoC: Fixes for v4.15
This is a fairly large set of fixes, they've been delayed partly as more
and more keep coming in. Most of them are very small driver specific
fixes, the biggest individual thing is the revert of the rcar IOMMU
support - it was causing problems and there wasn't the confidence that
it could be resolved sensibly. There's also a relatively large change
in the Freescale SSI controller which resolves some issues with the
AC'97 mode, these aren't that large in the grand scheme of things and
reflect some fairly thorough review and testing.
lock_all_ctx in setplane_internal may return -EINTR, and
__setplane_internal could return -EDEADLK. Making more
special cases for fb would make the code even harder to
read, so the easiest solution is not taking over the fb
refcount, and making callers responsible for dropping
the ref.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102707
Fixes: 13736ba3b3 ("drm/legacy: Convert setplane ioctl locking to interruptible.")
Testcase: kms_atomic_interruptible
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171220093545.613-2-maarten.lankhorst@linux.intel.com
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
When attached to the connector, the mode_valid callback will only filter
the modes provided by the connector itself as part of its probe.
However, it will not be doing it when the mode is provided by the
userspace, which still might result in a broken configuration.
In order to enforce these constraints, move our mode_valid callback to the
encoder which doesn't have this behaviour.
Acked-by: Daniel Vetter <daniel@ffwll.ch>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
[maxime: Wrote the commit log in order to update the patch from the merged
v3 to the v4 that was correct.]
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Link: https://patchwork.freedesktop.org/patch/msgid/0fa230a8-d01d-561a-f74f-6b4fd421255b@xs4all.nl
BRGCKR should use 0x80770000, instead of 0x80FF0000.
R-Car Gen2 xxx_TIMSEL should use 0x0F1F,
R-Car Gen3 xxx_TIMSEL should use 0x1F1F.
Here, Gen3 doesn't support AVD, thus, both case can use 0x0F1F.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
The hashing of %p was designed to restrict kernel addresses. There is
no reason to hash the userspace values seen during a segfault report,
so switch these to %px. (Some architectures already use %lx.)
Fixes: ad67b74d24 ("printk: hash addresses printed with %p")
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The following namespace configuration attempt:
# ndctl create-namespace -e namespace0.0 -m devdax -a 1G -f
libndctl: ndctl_dax_enable: dax0.1: failed to enable
Error: namespace0.0: failed to enable
failed to reconfigure namespace: No such device or address
...fails when the backing memory range is not physically aligned to 1G:
# cat /proc/iomem | grep Persistent
210000000-30fffffff : Persistent Memory (legacy)
In the above example the 4G persistent memory range starts and ends on a
256MB boundary.
We handle this case correctly when needing to handle cases that violate
section alignment (128MB) collisions against "System RAM", and we simply
need to extend that padding/truncation for the 1GB alignment use case.
Cc: <stable@vger.kernel.org>
Fixes: 315c562536 ("libnvdimm, pfn: add 'align' attribute...")
Reported-and-tested-by: Jane Chu <jane.chu@oracle.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
I'm getting various build failures on sparc64. The key is
usually that the userland tools get built 32-bit.
1) clock_gettime() is in librt, so that must be added to the link
libraries.
2) "sizeof(x)" must be printed with "%Z" printf prefix.
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
The alignment checks at pfn driver startup fail to properly account for
the 'start_pad' in the case where the namespace is misaligned relative
to its internal alignment. This is typically triggered in 1G aligned
namespace, but could theoretically trigger with small namespace
alignments. When this triggers the kernel reports messages of the form:
dax2.1: bad offset: 0x3c000000 dax disabled align: 0x40000000
Cc: <stable@vger.kernel.org>
Fixes: 1ee6667cd8 ("libnvdimm, pfn, dax: fix initialization vs autodetect...")
Reported-by: Jane Chu <jane.chu@oracle.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
During unload, on mlx5_stop_eqs we move command interface from events
mode to polling mode, but if command interface EQ destroy fail we move
back to events mode.
That's wrong since even if we fail to destroy command interface EQ, we
do release its irq, so no interrupts will be received.
Fixes: e126ba97db ("mlx5: Add driver for Mellanox Connect-IB adapters")
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
When mlx5_stop_eqs fails to destroy any of the eqs it returns with an error.
In such failure flow the function will return without
releasing all EQs irqs and then pci_free_irq_vectors will fail.
Fix by only warn on destroy EQ failure and continue to release other
EQs and their irqs.
It fixes the following kernel trace:
kernel: kernel BUG at drivers/pci/msi.c:352!
...
...
kernel: Call Trace:
kernel: pci_disable_msix+0xd3/0x100
kernel: pci_free_irq_vectors+0xe/0x20
kernel: mlx5_load_one.isra.17+0x9f5/0xec0 [mlx5_core]
Fixes: e126ba97db ("mlx5: Add driver for Mellanox Connect-IB adapters")
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Flow steering priority and namespace are software only objects that
didn't have the proper destructors and were not freed during steering
cleanup.
Fix it by adding destructor functions for these objects.
Fixes: bd71b08ec2 ("net/mlx5: Support multiple updates of steering rules in parallel")
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
When calling add/remove VXLAN port, a lock must be held in order to
prevent race scenarios when more than one add/remove happens at the
same time.
Fix by holding our state_lock (mutex) as done by all other parts of the
driver.
Note that the spinlock protecting the radix-tree is still needed in
order to synchronize radix-tree access from softirq context.
Fixes: b3f63c3d5e ("net/mlx5e: Add netdev support for VXLAN tunneling")
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
A refcount mechanism must be implemented in order to prevent unwanted
scenarios such as:
- Open an IPv4 VXLAN interface
- Open an IPv6 VXLAN interface (different socket)
- Remove one of the interfaces
With current implementation, the UDP port will be removed from our VXLAN
database and turn off the offloads for the other interface, which is
still active.
The reference count mechanism will only allow UDP port removals once all
consumers are gone.
Fixes: b3f63c3d5e ("net/mlx5e: Add netdev support for VXLAN tunneling")
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
mlx5e_vxlan_lookup_port is called both from mlx5e_add_vxlan_port (user
context) and mlx5e_features_check (softirq), but the lock acquired does
not disable bottom half and might result in deadlock. Fix it by simply
replacing spin_lock() with spin_lock_bh().
While at it, replace all unnecessary spin_lock_irq() to spin_lock_bh().
lockdep's WARNING: inconsistent lock state
[ 654.028136] inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage.
[ 654.028229] swapper/5/0 [HC0[0]:SC1[9]:HE1:SE0] takes:
[ 654.028321] (&(&vxlan_db->lock)->rlock){+.?.}, at: [<ffffffffa06e7f0e>] mlx5e_vxlan_lookup_port+0x1e/0x50 [mlx5_core]
[ 654.028528] {SOFTIRQ-ON-W} state was registered at:
[ 654.028607] _raw_spin_lock+0x3c/0x70
[ 654.028689] mlx5e_vxlan_lookup_port+0x1e/0x50 [mlx5_core]
[ 654.028794] mlx5e_vxlan_add_port+0x2e/0x120 [mlx5_core]
[ 654.028878] process_one_work+0x1e9/0x640
[ 654.028942] worker_thread+0x4a/0x3f0
[ 654.029002] kthread+0x141/0x180
[ 654.029056] ret_from_fork+0x24/0x30
[ 654.029114] irq event stamp: 579088
[ 654.029174] hardirqs last enabled at (579088): [<ffffffff818f475a>] ip6_finish_output2+0x49a/0x8c0
[ 654.029309] hardirqs last disabled at (579087): [<ffffffff818f470e>] ip6_finish_output2+0x44e/0x8c0
[ 654.029446] softirqs last enabled at (579030): [<ffffffff810b3b3d>] irq_enter+0x6d/0x80
[ 654.029567] softirqs last disabled at (579031): [<ffffffff810b3c05>] irq_exit+0xb5/0xc0
[ 654.029684] other info that might help us debug this:
[ 654.029781] Possible unsafe locking scenario:
[ 654.029868] CPU0
[ 654.029908] ----
[ 654.029947] lock(&(&vxlan_db->lock)->rlock);
[ 654.030045] <Interrupt>
[ 654.030090] lock(&(&vxlan_db->lock)->rlock);
[ 654.030162]
*** DEADLOCK ***
Fixes: b3f63c3d5e ("net/mlx5e: Add netdev support for VXLAN tunneling")
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
In error flow, when DESTROY_QP command should be executed, the wrong
mailbox was set with data, not the one that is written to hardware,
Fix that.
Fixes: 09a7d9eca1 '{net,IB}/mlx5: QP/XRCD commands via mlx5 ifc'
Signed-off-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Fixes the bug when turning on/off CQE compression mechanism
resets the RX rings size to default value when it is not
needed.
Fixes: 2fc4bfb725 ("net/mlx5e: Dynamic RQ type infrastructure")
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
The assumption that the next header field contains the transport
protocol is wrong for IPv6 packets with extension headers.
Instead, we should look the inner-most next header field in the buffer.
This will fix TSO offload for tunnels over IPv6 with extension headers.
Performance testing: 19.25x improvement, cool!
Measuring bandwidth of 16 threads TCP traffic over IPv6 GRE tap.
CPU: Intel(R) Xeon(R) CPU E5-2660 v2 @ 2.20GHz
NIC: Mellanox Technologies MT28800 Family [ConnectX-5 Ex]
TSO: Enabled
Before: 4,926.24 Mbps
Now : 94,827.91 Mbps
Fixes: b3f63c3d5e ("net/mlx5e: Add netdev support for VXLAN tunneling")
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Fix bug that allows ets bw sum to be 0% when ets tc type exists.
Fixes: 08fb1dacdd ('net/mlx5e: Support DCBNL IEEE ETS')
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Reviewed-by: Huy Nguyen <huyn@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
In mlx5_ifc, struct size was not complete, and thus driver was sending
garbage after the last defined field. Fixed it by adding reserved field
to complete the struct size.
In addition, rename all set_rate_limit to set_pp_rate_limit to be
compliant with the Firmware <-> Driver definition.
Fixes: 7486216b3a ("{net,IB}/mlx5: mlx5_ifc updates")
Fixes: 1466cc5b23 ("net/mlx5: Rate limit tables support")
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Before the offending commit, mlx5 core did the IRQ affinity itself,
and it seems that the new generic code have some drawbacks and one
of them is the lack for user ability to modify irq affinity after
the initial affinity values got assigned.
The issue is still being discussed and a solution in the new generic code
is required, until then we need to revert this patch.
This fixes the following issue:
echo <new affinity> > /proc/irq/<x>/smp_affinity
fails with -EIO
This reverts commit a435393aca.
Note: kept mlx5_get_vector_affinity in include/linux/mlx5/driver.h since
it is used in mlx5_ib driver.
Fixes: a435393aca ("mlx5: move affinity hints assignments to generic code")
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Jes Sorensen <jsorensen@fb.com>
Reported-by: Jes Sorensen <jsorensen@fb.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Currently, if a size of zero is passed to
mlx5_fpga_mem_{read|write}_i2c()
the "err" return value will not be initialized, which triggers gcc
warnings:
[..]/mlx5/core/fpga/sdk.c:87 mlx5_fpga_mem_read_i2c() error:
uninitialized symbol 'err'.
[..]/mlx5/core/fpga/sdk.c:115 mlx5_fpga_mem_write_i2c() error:
uninitialized symbol 'err'.
fix that.
Fixes: a9956d35d1 ('net/mlx5: FPGA, Add SBU infrastructure')
Signed-off-by: Kamal Heib <kamalh@mellanox.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Our MMC host driver now issues a reset, instead of just deasserting
the reset control, since commit c34eda69ad ("mmc: sunxi: Reset the
device at probe time"). The sun9i-mmc clock driver does not support
this, and will fail, which results in MMC not probing.
This patch implements the reset callback by asserting the reset control,
then deasserting it after a small delay.
Fixes: 7a6fca879f ("clk: sunxi: Add driver for A80 MMC config clocks/resets")
Cc: <stable@vger.kernel.org> # 4.14.x
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
Acked-by: Philipp Zabel <p.zabel@pengutronix.de>
Acked-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Michael Turquette <mturquette@baylibre.com>
Link: lkml.kernel.org/r/20171218035751.20661-1-wens@csie.org
The recently added fib_metrics_match() causes a regression for routes
with both RTAX_FEATURES and RTAX_CC_ALGO if the latter has
TCP_CONG_NEEDS_ECN flag set:
| # ip link add d0 type dummy
| # ip link set d0 up
| # ip route add 172.29.29.0/24 dev d0 features ecn congctl dctcp
| # ip route del 172.29.29.0/24 dev d0 features ecn congctl dctcp
| RTNETLINK answers: No such process
During route insertion, fib_convert_metrics() detects that the given CC
algo requires ECN and hence sets DST_FEATURE_ECN_CA bit in
RTAX_FEATURES.
During route deletion though, fib_metrics_match() compares stored
RTAX_FEATURES value with that from userspace (which obviously has no
knowledge about DST_FEATURE_ECN_CA) and fails.
Fixes: 5f9ae3d9e7 ("ipv4: do metrics match when looking up and deleting a route")
Signed-off-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>
As noted in dwmac4_wrback_get_rx_timestamp_status the timestamp is found
in the context descriptor following the current descriptor. However the
current code looks for the context descriptor in the current
descriptor, which will always fail.
Signed-off-by: Fredrik Hallenberg <megahallon@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When using GMAC4 the value written in PTP_SSIR should be shifted however
the shifted value is also used in subsequent calculations which results
in a bad timestamp value.
Signed-off-by: Fredrik Hallenberg <megahallon@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
