Daniel Borkmann [Mon, 6 Jun 2016 20:50:39 +0000 (22:50 +0200)]
net: sched: fix tc_should_offload for specific clsact classes
When offloading classifiers such as u32 or flower to hardware, and the
qdisc is clsact (TC_H_CLSACT), then we need to differentiate its classes,
since not all of them handle ingress, therefore we must leave those in
software path. Add a .tcf_cl_offload() callback, so we can generically
handle them, tested on ixgbe.
Fixes: 10cbc6843446 ("net/sched: cls_flower: Hardware offloaded filters statistics support") Fixes: 5b33f48842fa ("net/flower: Introduce hardware offload support") Fixes: a1b7c5fd7fe9 ("net: sched: add cls_u32 offload hooks for netdevs") Signed-off-by: Daniel Borkmann <[email protected]> Acked-by: John Fastabend <[email protected]> Signed-off-by: David S. Miller <[email protected]>
WANG Cong [Mon, 6 Jun 2016 16:54:30 +0000 (09:54 -0700)]
act_police: fix a crash during removal
The police action is using its own code to initialize tcf hash
info, which makes us to forgot to initialize a->hinfo correctly.
Fix this by calling the helper function tcf_hash_create() directly.
Fixes: ddf97ccdd7cb ("net_sched: add network namespace support for tc actions") Cc: Jamal Hadi Salim <[email protected]> Signed-off-by: Cong Wang <[email protected]> Signed-off-by: David S. Miller <[email protected]>
A while back, I sent one RFC patch using lockless stats gathering
on 64bit arches.
This patch series does it more cleanly, using a seqcount.
Since qdisc/class stats are written at dequeue() time,
we can ask the dequeue to change the seqcount, so that
stats readers can avoid taking the root qdisc lock,
and instead the typical read_seqcount_{begin|retry} guarded
loop.
This does not change fast path costs, as the seqcount
increments are not more expensive than the bit manipulation,
and allows readers to not freeze the fast path anymore.
====================
Eric Dumazet [Mon, 6 Jun 2016 16:37:16 +0000 (09:37 -0700)]
net: sched: do not acquire qdisc spinlock in qdisc/class stats dump
Large tc dumps (tc -s {qdisc|class} sh dev ethX) done by Google BwE host
agent [1] are problematic at scale :
For each qdisc/class found in the dump, we currently lock the root qdisc
spinlock in order to get stats. Sampling stats every 5 seconds from
thousands of HTB classes is a challenge when the root qdisc spinlock is
under high pressure. Not only the dumps take time, they also slow
down the fast path (queue/dequeue packets) by 10 % to 20 % in some cases.
An audit of existing qdiscs showed that sch_fq_codel is the only qdisc
that might need the qdisc lock in fq_codel_dump_stats() and
fq_codel_dump_class_stats()
In v2 of this patch, I now use the Qdisc running seqcount to provide
consistent reads of packets/bytes counters, regardless of 32/64 bit arches.
I also changed rate estimators to use the same infrastructure
so that they no longer need to lock root qdisc lock.
Eric Dumazet [Mon, 6 Jun 2016 16:12:39 +0000 (09:12 -0700)]
fq_codel: return non zero qlen in class dumps
We properly scan the flow list to count number of packets,
but John passed 0 to gnet_stats_copy_queue() so we report
a zero value to user space instead of the result.
This set fixes two small issues with error codes I noticed
in cls_u32. Second patch could be viewed as user space API
change but that portion of API is not part of any release,
yet.
Colin Ian King [Mon, 6 Jun 2016 15:08:41 +0000 (16:08 +0100)]
gtp: #define _UAPI_LINUX_GTP_H_ and not _UAPI_LINUX_GTP_H__
Fix clang build warning:
./include/uapi/linux/gtp.h:1:9: warning: '_UAPI_LINUX_GTP_H_' is
used as a header guard here, followed by #define of a different
macro [-Wheader-guard]
fix by defining _UAPI_LINUX_GTP_H_ and not _UAPI_LINUX_GTP_H__
Linus Torvalds [Tue, 7 Jun 2016 23:24:44 +0000 (16:24 -0700)]
Merge tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux
Pull clk fixes from Stephen Boyd:
"This finally removes the CLK_IS_ROOT flag by picking up the last few
stragglers that didn't get merged by anyone this time around.
Better to do it now than wait for another one to pop up. There's also
a minor maintainers update and a Kconfig fix"
* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
clk: nxp: Select MFD_SYSCON for creg driver
MAINTAINERS: Add file patterns for clock device tree bindings
clk: Remove CLK_IS_ROOT flag
clk: microchip: Remove CLK_IS_ROOT
powerpc/512x: clk: Remove CLK_IS_ROOT
vexpress/spc: Remove CLK_IS_ROOT
David S. Miller [Tue, 7 Jun 2016 23:18:20 +0000 (16:18 -0700)]
Merge branch 'be2net-noncrit-fixes'
Sathya Perla says:
====================
be2net: patch set
Hi David, the following patch set contains three non-critical fixes that
can go into the net-next tree.
Patch 1 fixes the logic for provisioning queue pairs on VFs to take into
account the limit on number of TXQs too as in some profiles the number
of TXQs is less than that of RXQs.
Patch 2 enables WoL support from shutdown on Skyhawk.
Patch 3 enhances the logic for provisioning queue pairs on VFs on
SR-IOV over multi-partition configs. Each PF (partition) on a port has to
compute the number of RSS tables it's VFs can use.
====================
Somnath Kotur [Mon, 6 Jun 2016 11:22:10 +0000 (07:22 -0400)]
be2net: Fix provisioning of RSS for VFs in multi-partition configurations
Currently, we do not distribute queue resources to enable RSS for VFs
in multi-channel/partition configurations.
Fix this by having each PF(SRIOV capable) calculate it's share of the
15 RSS Policy Tables available per port before provisioning resources for
all the VFs.
This proportional share calculation is done based on division of the
PF's MAX VFs with the Total MAX VFs on that port. It also needs to
learn about the no: of NIC PFs on the port and subtract that from
the 15 RSS Policy Tables on the port.
be2net: Enable Wake-On-LAN from shutdown for Skyhawk
Skyhawk does support wake-up from ACPI shutdown state - S5, provided the
platform supports it (like Auxiliary power source etc). The changes listed
below are done to fix this.
1) There's no need to defer the HW configuration of WOL to be_suspend().
Remove this in be_suspend() and move it to be_set_wol() ethtool function
so it is configured directly in the context of ethtool. This automatically
takes care of the shutdown case.
2) The driver incorrectly uses WOL_CAP field in the FW response to
get_acpi_wol_cap() command, to determine if WOL is enabled. Instead the
driver must rely on the macaddr field in the response to infer WOL state.
3) In be_get_config() during init, if we find that WOL is enabled in FW,
call pci_enable_wake() to enable pmcsr.pme_en bit. This is needed to
support persistent WOL configuration provided by the FW in some platforms.
4) Remove code in be_set_wol() that writes to PCICFG_PM_CONTROL_OFFSET
to set pme_en bit; pci_enable_wake() sets that.
Fixes: 028991e49 ("Enabling Wake-on-LAN is not supported in S5 state") Signed-off-by: Sriharsha Basavapatna <[email protected]> Signed-off-by: Sathya Perla <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Suresh Reddy [Mon, 6 Jun 2016 11:22:08 +0000 (07:22 -0400)]
be2net: use max-TXQs limit too while provisioning VF queue pairs
When the PF driver provisions resources for VFs, it currently only looks
at max RSS queues available to calculate the number of VF queue pairs.
This logic breaks when there are less number of TX-queues than RSS-queues.
This patch fixes this problem by using the max-TXQs available in the
PF-pool in the calculations. As a part of this change the
be_calculate_vf_qs() routine is renamed as be_calculate_vf_res() and the
code that calculates limits on other related resources is moved here to
contain all resource calculation code inside one routine.
Michael Chan [Mon, 6 Jun 2016 06:37:16 +0000 (02:37 -0400)]
bnxt_en: Simplify VLAN receive logic.
Since both CTAG and STAG rx acceleration must be enabled together, we
only need to check one feature flag (NETIF_F_HW_VLAN_CTAG_RX) before
calling __vlan_hwaccel_put_tag().
Michael Chan [Mon, 6 Jun 2016 06:37:15 +0000 (02:37 -0400)]
bnxt_en: Enable and disable RX CTAG and RX STAG VLAN acceleration together.
The hardware can only be set to strip or not strip both the VLAN CTAG and
STAG. It cannot strip one and not strip the other. Add logic to
bnxt_fix_features() to toggle both feature flags when the user is toggling
one of them.
Michael Chan [Mon, 6 Jun 2016 06:37:14 +0000 (02:37 -0400)]
bnxt_en: Fix tx push race condition.
Set the is_push flag in the software BD before the tx data is pushed to
the chip. It is possible to get the tx interrupt as soon as the tx data
is pushed. The tx handler will not handle the event properly if the
is_push flag is not set and it will crash.
Zhao Qiang [Mon, 6 Jun 2016 06:30:01 +0000 (14:30 +0800)]
fsl/qe: Add QE TDM lib
QE has module to support TDM, some other protocols
supported by QE are based on TDM.
add a qe-tdm lib, this lib provides functions to the protocols
using TDM to configurate QE-TDM.
Amir Vadai [Sun, 5 Jun 2016 14:11:18 +0000 (17:11 +0300)]
net/sched: cls_flower: Introduce support in SKIP SW flag
In order to make a filter processed only by hardware, skip_sw flag
should be supplied. This is an addition to the already existing skip_hw
flag (filter will be processed by software only). If no flag is
specified, filter will be processed by both software and hardware.
If only hardware offloaded filters exist, fl_classify() will return
without doing anything.
A following userspace patch will be sent once kernel patch is accepted.
Example:
tc filter add dev enp0s9 protocol ip prio 20 parent ffff: \
flower \
ip_proto 6 \
indev enp0s9 \
skip_sw \
action skbedit mark 0x1234
David S. Miller [Tue, 7 Jun 2016 22:40:12 +0000 (15:40 -0700)]
Merge branch 'qed-iov-fw-reqs'
Yuval Mintz says:
====================
qed: IOV series - relax firmware requirements
In order for VFs to work, current implementation demands that the VF's
requried storm firmware would be exactly the version that was loaded by
the PF, which is a very harsh requirement.
This patch series is intended to relax this -
the recently submitted firmware is intended to be forward/backward
compatible in its fastpath [slowpath is configured by PF on behalf of VF],
and so VFs would only be required of having the same major faspath HSI in
order to work.
Most of the other patches in this series extend current forward
compatibilty of driver to reduce chance of breaking PF/VF compatibility
in the future. A few are unrelated IOV changes.
====================
Yuval Mintz [Sun, 5 Jun 2016 10:11:16 +0000 (13:11 +0300)]
qed: PF to reply to unknown messages
If a future VF would send the PF an unknown message, the PF today would
not send a reply. This would have 2 bad effects:
a. VF would have to timeout on the request.
b. If VF were to send an additional message to PF, firmware would mark
it as malicious.
Instead, if there's some valid reply-address on the message - let the PF
answer and tell the VF it doesn't know the message.
Yuval Mintz [Sun, 5 Jun 2016 10:11:15 +0000 (13:11 +0300)]
qed: PF enforce MAC limitation of VFs
The only limitation relating to MACs the PF enforce today on its VFs
is in case it has a forced-unicast MAC address for them, in which case
they can't configure other unicast addresses.
Specifically, the PF isn't enforcing the number of MAC addresse a VF can
configure regardless of the nubmer of such filters agreed upon by PF and
VF during the acquisition process.
PF's shadow-config is now extended to also contain information about its
VFs' unicast addresses configuration, allowing such enforcement.
Yuval Mintz [Sun, 5 Jun 2016 10:11:14 +0000 (13:11 +0300)]
qed: Move doorbell calculation from VF to PF
Today, the VF is aware of its queues context-ids, and calculates the
doorbell address when opening its queues on its own.
The configuration of doorbells in HW can sometime in the future be changed
by the PF [hw has several configurable features that might affect doorbell
addresses, e.g., dpm support], this would break compatibility with older
VFs as their calculated doorbell addresses would be incorrect for such a
configuration.
In order to avoid such a backward compatibility failure, let the PF make
the calculation of the doorbell offset based on the context-id, and pass
that to the VF.
Yuval Mintz [Sun, 5 Jun 2016 10:11:13 +0000 (13:11 +0300)]
qed: Make PF more robust against malicious VF
There are several requests the VF can make toward the PF which the driver
would pass to firmware without checking the validity first - specifically,
opening queues and updating vports. Such configurations might cause the
firmware to assert.
This adds validation of the legality of said configurations on the PF side
before passing it onward via ramrod to firmware.
Yuval Mintz [Sun, 5 Jun 2016 10:11:12 +0000 (13:11 +0300)]
qed: PF-VF resource negotiation
One of the goals of the vf's first message to the PF [acquire]
is to learn about the number of resources available to it [macs, vlans,
etc.]. This is done via negotiation - the VF requires a set of resources,
which the PF either approves or disaproves and sends a smaller set of
resources as alternative. In this later case, the VF is then expected to
either abort the probe or re-send the acquire message with less
required resources.
While this infrastructure exists since the initial submision of qed
SRIOV support, it's in fact completely inoperational - PF isn't really
looking into the resources the VF has asked for and is never going to
reply to the VF that it lacks resources.
This patch addresses this flow, fixing it and allowing the PF and VF
to actually agree on a set of resources.
Yuval Mintz [Sun, 5 Jun 2016 10:11:11 +0000 (13:11 +0300)]
qed: Relax VF firmware requirements
Current driver require an exact match between VF and PF storm firmware;
Any difference would fail the VF acquire message, causing the VF probe
to be aborted.
While there's still dependencies between the two, the recent FW submission
has relaxed the match requirement - instead of an exact match, there's now
a 'fastpath' HSI major/minor scheme, where VFs and PFs that match in their
major number can co-exist even if their minor is different.
In order to accomadate this change some changes in the vf-start init flow
had to be made, as the VF start ramrod now has to be sent only after PF
learns which fastpath HSI its VF is requiring.
This three part patchset fixes bugs in synchronization between
rds_tcp_accept_one() and the rds-tcp send/recv path.
Patch 1 ensures that the lock_sock() is taken appropriately
and the RDS datagram reassembly state is reset to synchronize
with the receive path.
Patch 2 ensures that partially sent RDS datagrams will get
retransmitted after rds_tcp_accept_one() switches sockets.
Patch 3 fixes a race window which would prematurely re-enable
rds_send_xmit() before the rds_tcp_connection setup has been
completed in rds_tcp_accept_one().
====================
RDS: TCP: fix race windows in send-path quiescence by rds_tcp_accept_one()
The send path needs to be quiesced before resetting callbacks from
rds_tcp_accept_one(), and commit eb192840266f ("RDS:TCP: Synchronize
rds_tcp_accept_one with rds_send_xmit when resetting t_sock") achieves
this using the c_state and RDS_IN_XMIT bit following the pattern
used by rds_conn_shutdown(). However this leaves the possibility
of a race window as shown in the sequence below
take t_conn_lock in rds_tcp_conn_connect
send outgoing syn to peer
drop t_conn_lock in rds_tcp_conn_connect
incoming from peer triggers rds_tcp_accept_one, conn is
marked CONNECTING
wait for RDS_IN_XMIT to quiesce any rds_send_xmit threads
call rds_tcp_reset_callbacks
[.. race-window where incoming syn-ack can cause the conn
to be marked UP from rds_tcp_state_change ..]
lock_sock called from rds_tcp_reset_callbacks, and we set
t_sock to null
As soon as the conn is marked UP in the race-window above, rds_send_xmit()
threads will proceed to rds_tcp_xmit and may encounter a null-pointer
deref on the t_sock.
Given that rds_tcp_state_change() is invoked in softirq context, whereas
rds_tcp_reset_callbacks() is in workq context, and testing for RDS_IN_XMIT
after lock_sock could result in a deadlock with tcp_sendmsg, this
commit fixes the race by using a new c_state, RDS_TCP_RESETTING, which
will prevent a transition to RDS_CONN_UP from rds_tcp_state_change().
RDS: TCP: Retransmit half-sent datagrams when switching sockets in rds_tcp_reset_callbacks
When we switch a connection's sockets in rds_tcp_rest_callbacks,
any partially sent datagram must be retransmitted on the new
socket so that the receiver can correctly reassmble the RDS
datagram. Use rds_send_reset() which is designed for this purpose.
RDS: TCP: Add/use rds_tcp_reset_callbacks to reset tcp socket safely
When rds_tcp_accept_one() has to replace the existing tcp socket
with a newer tcp socket (duelling-syn resolution), it must lock_sock()
to suppress the rds_tcp_data_recv() path while callbacks are being
changed. Also, existing RDS datagram reassembly state must be reset,
so that the next datagram on the new socket does not have corrupted
state. Similarly when resetting the newly accepted socket, appropriate
locks and synchronization is needed.
This commit ensures correct synchronization by invoking
kernel_sock_shutdown to reset a newly accepted sock, and by taking
appropriate lock_sock()s (for old and new sockets) when resetting
existing callbacks.
Eric Dumazet [Sat, 4 Jun 2016 19:55:13 +0000 (12:55 -0700)]
fq_codel: fix NET_XMIT_CN behavior
My prior attempt to fix the backlogs of parents failed.
If we return NET_XMIT_CN, our parents wont increase their backlog,
so our qdisc_tree_reduce_backlog() should take this into account.
v2: Florian Westphal pointed out that we could drop the packet,
so we need to save qdisc_pkt_len(skb) in a temp variable before
calling fq_codel_drop()
Daniel Borkmann [Sat, 4 Jun 2016 18:50:59 +0000 (20:50 +0200)]
bpf, trace: use READ_ONCE for retrieving file ptr
In bpf_perf_event_read() and bpf_perf_event_output(), we must use
READ_ONCE() for fetching the struct file pointer, which could get
updated concurrently, so we must prevent the compiler from potential
refetching.
We already do this with tail calls for fetching the related bpf_prog,
but not so on stored perf events. Semantics for both are the same
with regards to updates.
Fixes: a43eec304259 ("bpf: introduce bpf_perf_event_output() helper") Fixes: 35578d798400 ("bpf: Implement function bpf_perf_event_read() that get the selected hardware PMU conuter") Signed-off-by: Daniel Borkmann <[email protected]> Acked-by: Alexei Starovoitov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Jason Wang [Wed, 1 Jun 2016 05:56:33 +0000 (01:56 -0400)]
vhost_net: stop polling socket during rx processing
We don't stop rx polling socket during rx processing, this will lead
unnecessary wakeups from under layer net devices (E.g
sock_def_readable() form tun). Rx will be slowed down in this
way. This patch avoids this by stop polling socket during rx
processing. A small drawback is that this introduces some overheads in
light load case because of the extra start/stop polling, but single
netperf TCP_RR does not notice any change. In a super heavy load case,
e.g using pktgen to inject packet to guest, we get about ~8.8%
improvement on pps:
Linus Torvalds [Tue, 7 Jun 2016 17:04:35 +0000 (10:04 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace
Pull userns fixes from Eric Biederman:
"This contains two small but significant fixes to fs/namespace.c.
The first adds a filesystem refcount drop on error. The second
corrects a test in fs_fully_visible which could be abused to allow
mounting of proc or sysfs, when that should not be allowed.
To keep myself honest I have tested to ensure the incorrect test in
fs_fully_visible actually allows improper mounting of proc before the
fix and that when fixed the improper mounting is not allowed"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace:
mnt: fs_fully_visible test the proper mount for MNT_LOCKED
mnt: If fs_fully_visible fails call put_filesystem.
Erez Shitrit [Sat, 4 Jun 2016 12:15:19 +0000 (15:15 +0300)]
IB/IPoIB: Don't update neigh validity for unresolved entries
ipoib_neigh_get unconditionally updates the "alive" variable member on
any packet send. This prevents the neighbor garbage collection from
cleaning out a dead neighbor entry if we are still queueing packets
for it. If the queue for this neighbor is full, then don't update the
alive timestamp. That way the neighbor can time out even if packets
are still being queued as long as none of them are being sent.
Fixes: b63b70d87741 ("IPoIB: Use a private hash table for path lookup in xmit path") Signed-off-by: Erez Shitrit <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Doug Ledford <[email protected]>
Achiad Shochat [Sat, 4 Jun 2016 12:15:37 +0000 (15:15 +0300)]
IB/mlx5: Fix alternate path code
Userspace flag IBV_QP_ALT_PATH is supposed to set the alternate path
including fields alt_pkey_index and alt_timeout.
Added IB_QP_PKEY_INDEX and IB_QP_TIMEOUT to the attribute mask when
calling mlx5_set_path for the alternate path to force setting the
alt_pkey_index and alt_timeout values.
Noa Osherovich [Sat, 4 Jun 2016 12:15:36 +0000 (15:15 +0300)]
IB/mlx5: Fix pkey_index length in the QP path record
Pkey index fields in the QP context path record are extended to 16
bits, as required by IB spec (version 1.3).
This change affects all QP commands which include path records.
To enable this change, moved the free adaptive routing flag bit
(free_ar) to the most significant byte of the QP path record.
Noa Osherovich [Sat, 4 Jun 2016 12:15:31 +0000 (15:15 +0300)]
IB/mlx5: Limit query HCA clock
When PAGE_SIZE is larger than 4K, the user shouldn't be able to query
the HCA core clock. This counter is within 4KB boundary and the
user-space shall not read information that's after this boundary.
Noa Osherovich [Sat, 4 Jun 2016 12:15:29 +0000 (15:15 +0300)]
IB/mlx5: Return PORT_ERR in Active to Initializing tranisition
FW port-change events are fired on Active <-> non Active port state
transitions only.
When the port state changes from Active to Initializing (Active ->
Down -> Initializing), a single event is fired.
The HCA transitions from Down to Initializing unless prevented from
doing so, hence the driver should also propagate events when the port
state is Initializing to consumers so they'll be aware that the port
is no longer Active and act accordingly.
Max Gurtovoy [Mon, 6 Jun 2016 16:34:39 +0000 (19:34 +0300)]
IB/core: Fix bit curruption in ib_device_cap_flags structure
ib_device_cap_flags 64-bit expansion caused caps overlapping
and made consumers read wrong device capabilities. For example
IB_DEVICE_SG_GAPS_REG was falsely read by the iser driver causing
it to use a non-existing capability. This happened because signed
int becomes sign extended when converted it to u64. Fix this by
casting IB_DEVICE_ON_DEMAND_PAGING enumeration to ULL.
Mark Bloch [Sat, 4 Jun 2016 12:15:24 +0000 (15:15 +0300)]
IB/core: Initialize sysfs attributes before sysfs create group
For dynamically allocated sysfs attributes there is a need to call
sysfs_attr_init in order to comply with lockdep, not calling it
will result in error complaining key is not in .data section.
Fixes: b40f4757daa1 ("IB/core: Make device counter infrastructure dynamic") Signed-off-by: Mark Bloch <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Doug Ledford <[email protected]>
Mark Bloch [Sat, 4 Jun 2016 12:15:22 +0000 (15:15 +0300)]
IB/IPoIB: Disable bottom half when dealing with device address
Align locking usage when touching device address with rest
of the kernel. Lock the bottom half when doing so using
netif_addr_lock_bh.
This also solves the following case as reported by lockdep:
CPU0 CPU1
---- ----
lock(_xmit_INFINIBAND);
local_irq_disable();
lock(&(&mc->mca_lock)->rlock);
lock(_xmit_INFINIBAND);
<Interrupt>
lock(&(&mc->mca_lock)->rlock);
*** DEADLOCK ***
Fixes: 492a7e67ff83 ("IB/IPoIB: Allow setting the device address") Signed-off-by: Mark Bloch <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Doug Ledford <[email protected]>
Aviv Heller [Sat, 4 Jun 2016 12:15:21 +0000 (15:15 +0300)]
IB/core: Fix removal of default GID cache entry
When deleting a default GID from the cache, its gid_type field is set
to 0.
This could set the gid_type to RoCE v1 for a RoCE v2 default GID,
essentially making it inaccessible to future modifications, since it
is no longer found by find_gid().
This fix preserves the gid_type value for default gids during cache
operations.
Erez Shitrit [Sat, 4 Jun 2016 12:15:20 +0000 (15:15 +0300)]
IB/IPoIB: Fix race between ipoib_remove_one to sysfs functions
In ipoib_remove_one the driver holds the rtnl_lock and tries to do some
operation like dev_change_flags or unregister_netdev, while sysfs
callback like ipoib_vlan_delete holds sysfs mutex and tries to hold the
rtnl_lock via rtnl_trylock() and restart_syscall() if the lock is not
free, meanwhile ipoib_remove_one tries to get the sysfs lock in order to
free its sysfs directory, and we will get a->b, b->a deadlock.
Eli Cohen [Sat, 4 Jun 2016 12:15:18 +0000 (15:15 +0300)]
IB/core: Fix query port failure in RoCE
Currently ib_query_port always attempts to to read the subnet prefix by
calling ib_query_gid(). For RoCE/iWARP there is no subnet manager and no
subnet prefix. Fix this by querying GID[0] only for IB networks.
Doug Ledford [Tue, 7 Jun 2016 11:43:46 +0000 (07:43 -0400)]
IB/core: fix error unwind in sysfs hw counters code
Between the initial and final versions of the function setup_hw_stats,
the order of variable initialization was changed. However, the unwind
flow on error did not properly keep up with the flow changes. Make
the unwind flow match a proper unwind of the allocation flow, then
remove no longer needed variable initializations.
mnt: fs_fully_visible test the proper mount for MNT_LOCKED
MNT_LOCKED implies on a child mount implies the child is locked to the
parent. So while looping through the children the children should be
tested (not their parent).
Typically an unshare of a mount namespace locks all mounts together
making both the parent and the slave as locked but there are a few
corner cases where other things work.
mnt: If fs_fully_visible fails call put_filesystem.
Add this trivial missing error handling.
Cc: [email protected] Fixes: 1b852bceb0d1 ("mnt: Refactor the logic for mounting sysfs and proc in a user namespace") Signed-off-by: "Eric W. Biederman" <[email protected]>
A dedicated workqueue has been used since the workitem viz
(&db_wq->wk.work which maps to check_db_timeout) is involved
in normal device operation. WQ_MEM_RECLAIM has been set to guarantee
forward progress under memory pressure, which is a requirement here.
Since there are only a fixed number of work items, explicit concurrency
limit is unnecessary.
flush_workqueue is unnecessary since destroy_workqueue() itself calls
drain_workqueue() which flushes repeatedly till the workqueue
becomes empty.
A dedicated workqueue has been used since the workitem viz
(&cwq->wk.work which maps to oct_poll_req_completion) is involved
in normal device operation. WQ_MEM_RECLAIM has been set to guarantee
forward progress under memory pressure, which is a requirement here.
Since there are only a fixed number of work items, explicit concurrency
limit is unnecessary.
flush_workqueue is unnecessary since destroy_workqueue() itself calls
drain_workqueue() which flushes repeatedly till the workqueue
becomes empty. Hence the call to flush_workqueue() has been dropped.
Aaron Conole [Fri, 3 Jun 2016 20:57:12 +0000 (16:57 -0400)]
virtio-net: Add initial MTU advice feature
This commit adds the feature bit and associated mtu device entry for the
virtio network device. When a virtio device comes up, it checks the
feature bit for the VIRTIO_NET_F_MTU feature. If such feature bit is
enabled, the driver will read the advised MTU and use it as the initial
value.
Lv Zheng [Fri, 3 Jun 2016 02:26:12 +0000 (10:26 +0800)]
ACPI / EC: Fix a boot EC regresion by restoring boot EC support for the DSDT EC
According to the Windows probing result, during the table loading, the EC
device described in the ECDT should be used. And the ECDT EC is also
effective during the period the namespace objects are initialized (we can
see a separate process executing _STA/_INI on Windows before executing
other device specific control methods, for example, EC._REG). During the
device enumration, the EC device described in the DSDT should be used. But
there are differences between Linux and Windows around the device probing
order. Thus in Linux, we should enable the DSDT EC as early as possible
before enumerating devices in order not to trigger issues related to the
device enumeration order differences.
This patch thus converts acpi_boot_ec_enable() into acpi_ec_dsdt_probe() to
fix the gap. This also fixes a user reported regression triggered after we
switched the "table loading"/"ECDT support" to be ACPI spec 2.0 compliant.
Bart Van Assche [Fri, 3 Jun 2016 19:10:37 +0000 (12:10 -0700)]
IB/hfi1: Use bit 0 instead of bit 1
The first argument of test_bit() and clear_bit() is a bit number and
not a bitmask. Hence change that first argument from (1 << 0) into 0.
This patch avoids that smatch reports the following warnings:
user_sdma.c:1059: sdma_cache_evict() warn: test_bit() takes a bit number
user_sdma.c:1590: sdma_rb_remove() warn: test_bit() takes a bit number
Bart Van Assche [Fri, 3 Jun 2016 18:40:24 +0000 (11:40 -0700)]
IB/srp: Fix srp_map_sg_dma()
Because patch "IB/srp: Move common code into the caller" was applied
partially srp_map_sg_dma() doesn't work properly. Fix this by
applying the remainder of that patch. See also
http://thread.gmane.org/gmane.linux.drivers.rdma/35803/focus=35811.
Bart Van Assche [Fri, 3 Jun 2016 18:39:35 +0000 (11:39 -0700)]
IB/srp: Always initialize use_fast_reg and use_fmr
Avoid that mapping fails due to use_fast_reg != 0 or use_fmr != 0
if both member variables should be zero (if never_register == 1 or
if neither FMR nor FR is supported). Remove an initialization that
became superfluous due to changing a kmalloc() into a kzalloc()
call.
Bart Van Assche [Fri, 3 Jun 2016 14:58:32 +0000 (07:58 -0700)]
IB/mlx4: Fix device managed flow steering support test
Perform the test for device managed flow steering support even if
memory windows are not supported. I noticed this because smatch
reported inconsistent indentation for the device managed flow
steering support test.
Colin Ian King [Wed, 1 Jun 2016 18:06:36 +0000 (19:06 +0100)]
IB/core: fix null pointer deref and mem leak in error handling
The current error handling in setup_hw_stats has a couple of issues.
It is possible to generate a null pointer deference on the
kfree of hsag->attrs[i] because two of the early error exit paths
jump to the kfree when hsags NULL and not allocated. Fix this by
moving the kfree on stats and jumping to that, avoiding the hsag
freeing.
Secondly, there is a memory leak of stats if the hsag allocation
fails; instead of returning, jump to the kfree on stats.
Leon Romanovsky [Tue, 31 May 2016 07:54:36 +0000 (10:54 +0300)]
IB/hfi1: Avoid large frame size warning
When CONFIG_FRAME_WARN is set to 1024 bytes, which is useful to find
stack consumers, we get a warning in hfi1 driver.
drivers/infiniband/hw/hfi1/affinity.c: In function
‘hfi1_get_proc_affinity’:
drivers/infiniband/hw/hfi1/affinity.c:415:1: warning: the frame size of
1056 bytes is larger than 1024 bytes [-Wframe-larger-than=]
This change removes unneeded buf[1024] declaration and usage.
Fixes: f48ad614c100 ("IB/hfi1: Move driver out of staging") Signed-off-by: Leon Romanovsky <[email protected]> Acked-by: Dennis Dalessandro <[email protected]> Signed-off-by: Doug Ledford <[email protected]>
David S. Miller [Mon, 6 Jun 2016 22:19:07 +0000 (15:19 -0700)]
Merge branch 'vrf-local'
David Ahern says:
====================
net: vrf: Add support for local traffic to local addresses
Add support for locally originated traffic to VRF-local addresses,
be it addresses on enslaved devices or addresses on the VRF device:
$ ip addr show dev red
33: red: <NOARP,MASTER,UP,LOWER_UP> mtu 65536 qdisc pfifo_fast state UP group default qlen 1000
link/ether be:00:53:b5:e4:25 brd ff:ff:ff:ff:ff:ff
inet 1.1.1.1/32 scope global red
valid_lft forever preferred_lft forever
inet6 1111:1::1/128 scope global
valid_lft forever preferred_lft forever
$ ip addr show dev eth1
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master red state UP group default qlen 1000
link/ether 02:e0:f9:79:34:bd brd ff:ff:ff:ff:ff:ff
inet 10.100.1.1/24 brd 10.100.1.255 scope global eth1
valid_lft forever preferred_lft forever
inet6 2100:1::1/120 scope global
valid_lft forever preferred_lft forever
inet6 fe80::e0:f9ff:fe79:34bd/64 scope link
valid_lft forever preferred_lft forever
$ ping -c1 -I red 10.100.1.1
ping: Warning: source address might be selected on device other than red.
PING 10.100.1.1 (10.100.1.1) from 10.100.1.1 red: 56(84) bytes of data.
64 bytes from 10.100.1.1: icmp_seq=1 ttl=64 time=0.057 ms
$ ping -c1 -I red 1.1.1.1
PING 1.1.1.1 (1.1.1.1) from 1.1.1.1 red: 56(84) bytes of data.
64 bytes from 1.1.1.1: icmp_seq=1 ttl=64 time=0.136 ms
--- 1.1.1.1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.136/0.136/0.136/0.000 ms
$ ping6 -c1 -I red 2100:1::1
ping6: Warning: source address might be selected on device other than red.
PING 2100:1::1(2100:1::1) from 2100:1::1 red: 56 data bytes
64 bytes from 2100:1::1: icmp_seq=1 ttl=64 time=0.167 ms
--- 2100:1::1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.167/0.167/0.167/0.000 ms
$ ping6 -c1 -I red 1111::1
PING 1111::1(1111::1) from 1111:1::1 red: 56 data bytes
64 bytes from 1111::1: icmp_seq=1 ttl=64 time=0.187 ms
--- 1111::1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.187/0.187/0.187/0.000 ms
This change also enables use of loopback address on the VRF device:
$ ip addr add dev red 127.0.0.1/8
$ ping -c1 -I red 127.0.0.1
PING 127.0.0.1 (127.0.0.1) from 127.0.0.1 red: 56(84) bytes of data.
64 bytes from 127.0.0.1: icmp_seq=1 ttl=64 time=0.058 ms
====================
David Ahern [Thu, 2 Jun 2016 20:15:12 +0000 (13:15 -0700)]
net: vrf: ipv6 support for local traffic to local addresses
Add support for locally originated traffic to VRF-local IPv6 addresses.
Similar to IPv4 a local dst is set on the skb and the packet is
reinserted with a call to netif_rx. With this patch, ping, tcp and udp
packets to a local IPv6 address are successfully routed:
$ ip addr show dev eth1
4: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master red state UP group default qlen 1000
link/ether 02:e0:f9:1c:b9:74 brd ff:ff:ff:ff:ff:ff
inet 10.100.1.1/24 brd 10.100.1.255 scope global eth1
valid_lft forever preferred_lft forever
inet6 2100:1::1/120 scope global
valid_lft forever preferred_lft forever
inet6 fe80::e0:f9ff:fe1c:b974/64 scope link
valid_lft forever preferred_lft forever
$ ping6 -c1 -I red 2100:1::1
ping6: Warning: source address might be selected on device other than red.
PING 2100:1::1(2100:1::1) from 2100:1::1 red: 56 data bytes
64 bytes from 2100:1::1: icmp_seq=1 ttl=64 time=0.098 ms
ip6_input is exported so the VRF driver can use it for the dst input
function. The dst_alloc function for IPv4 defaults to setting the input and
output functions; IPv6's does not. VRF does not need to duplicate the Rx path
so just export the ipv6 input function.
David Ahern [Thu, 2 Jun 2016 20:15:11 +0000 (13:15 -0700)]
net: vrf: ipv4 support for local traffic to local addresses
Add support for locally originated traffic to VRF-local addresses. If
destination device for an skb is the loopback or VRF device then set
its dst to a local version of the VRF cached dst_entry and call netif_rx
to insert the packet onto the rx queue - similar to what is done for
loopback. This patch handles IPv4 support; follow on patch handles IPv6.
With this patch, ping, tcp and udp packets to a local IPv4 address are
successfully routed:
$ ip addr show dev eth1
4: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master red state UP group default qlen 1000
link/ether 02:e0:f9:1c:b9:74 brd ff:ff:ff:ff:ff:ff
inet 10.100.1.1/24 brd 10.100.1.255 scope global eth1
valid_lft forever preferred_lft forever
inet6 2100:1::1/120 scope global
valid_lft forever preferred_lft forever
inet6 fe80::e0:f9ff:fe1c:b974/64 scope link
valid_lft forever preferred_lft forever
$ ping -c1 -I red 10.100.1.1
ping: Warning: source address might be selected on device other than red.
PING 10.100.1.1 (10.100.1.1) from 10.100.1.1 red: 56(84) bytes of data.
64 bytes from 10.100.1.1: icmp_seq=1 ttl=64 time=0.057 ms
This patch also enables use of IPv4 loopback address on the VRF device:
$ ip addr add dev red 127.0.0.1/8
$ ping -c1 -I red 127.0.0.1
PING 127.0.0.1 (127.0.0.1) from 127.0.0.1 red: 56(84) bytes of data.
64 bytes from 127.0.0.1: icmp_seq=1 ttl=64 time=0.058 ms
David Ahern [Thu, 2 Jun 2016 20:15:10 +0000 (13:15 -0700)]
net: vrf: Minor refactoring for local address patches
Move the stripping of the ethernet header from is_ip_tx_frame into the
ipv4 and ipv6 outbound functions. If the packet is destined to a local
address the header is retained since the packet is sent back to netif_rx.
Collapse vrf_send_v4_prep into vrf_process_v4_outbound.
Ben Skeggs [Fri, 3 Jun 2016 04:37:40 +0000 (14:37 +1000)]
drm/nouveau/disp/sor/gf119: both links use the same training register
It appears that, for whatever reason, both link A and B use the same
register to control the training pattern. It's a little odd, as the
GPUs before this (Tesla/Fermi1) have per-link registers, as do newer
GPUs (Maxwell).
projects / linux.git / log