Andrew Lunn says:
====================
mv88e6xxx: Poll when no interrupt defined
Not all boards using the mv88e6xxx switches have the interrupt output
connected to a GPIO. On these boards phylib has to poll the PHYs,
rather than use interrupts. Have the driver poll the interrupt status
register, which is more efficient than having phylib do it. And it
enables other switch interrupts to be services.
The Armada 370RD is such a board without a interrupt GPIO. Now that
interrupts work, wire up the PHYs to make use if them.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
The Ethernet switch has an embedded interrupt controller. Interrupts
from the embedded PHYs are part of this interrupt controller.
Explicitly list the MDIO bus the embedded PHYs are on, and wire up the
interrupts.
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Not all boards have the interrupt output from the switch connected to
a GPIO line. In such cases, phylib has to poll the internal PHYs,
rather than receive an interrupt when there is a change in the link
state. phylib polls once per second, and per PHY reads around 4
words. With a switch typically having 4 internal PHYs, this means 16
MDIO transactions per second.
Rather than performing this phylib level polling, have the driver poll
the interrupt status register. If the status register indicates an
interrupt condition processing of interrupts in the same way as if a
GPIO was used.
Polling 10 times a second places less load on the MDIO bus. But rather
than taking on average 0.5s to detect a link change, it takes less
than 0.05s. Additionally, other interrupts, such as the watchdog, ATU
and VTU violations will be reported.
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Up until now we only allowed VLAN devices to be put in a VLAN-unaware
bridge, but some users need the ability to enslave physical ports as
well.
This is achieved by mapping the port and VID 1 to the bridge's vFID,
instead of the port and the VID used by the VLAN device.
The above is valid because as long as the port is not enslaved to a
bridge, VID 1 is guaranteed to be configured as PVID and egress
untagged.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Tested-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann says:
====================
pull-request: bpf-next 2018-02-26
The following pull-request contains BPF updates for your *net-next* tree.
The main changes are:
1) Various improvements for BPF kselftests: i) skip unprivileged tests
when kernel.unprivileged_bpf_disabled sysctl knob is set, ii) count
the number of skipped tests from unprivileged, iii) when a test case
had an unexpected error then print the actual but also the unexpected
one for better comparison, from Joe.
2) Add a sample program for collecting CPU state statistics with regards
to how long the CPU resides in cstate and pstate levels. Based on
cpu_idle and cpu_frequency trace points, from Leo.
3) Various x64 BPF JIT optimizations to further shrink the generated
image size in order to make it more icache friendly. When tested on
the Cilium generated programs, image size reduced by approx 4-5% in
best case mainly due to how LLVM emits unsigned 32 bit constants,
from Daniel.
4) Improvements and fixes on the BPF sockmap sample programs: i) fix
the sockmap's Makefile to include nlattr.o for libbpf, ii) detach
the sock ops programs from the cgroup before exit, from Prashant.
5) Avoid including xdp.h in filter.h by just forward declaring the
struct xdp_rxq_info in filter.h, from Jesper.
6) Fix the BPF kselftests Makefile for cgroup_helpers.c by only declaring
it a dependency for test_dev_cgroup.c but not every other test case
where it is not needed, from Jesper.
7) Adjust rlimit RLIMIT_MEMLOCK for test_tcpbpf_user selftest since the
default is insufficient for creating the 'global_map' used in the
corresponding BPF program, from Yonghong.
8) Likewise, for the xdp_redirect sample, Tushar ran into the same when
invoking xdp_redirect and xdp_monitor at the same time, therefore
in order to have the sample generically work bump the limit here,
too. Fix from Tushar.
9) Avoid an unnecessary NULL check in BPF_CGROUP_RUN_PROG_INET_SOCK()
since sk is always guaranteed to be non-NULL, from Yafang.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
CPU is active when have running tasks on it and CPUFreq governor can
select different operating points (OPP) according to different workload;
we use 'pstate' to present CPU state which have running tasks with one
specific OPP. On the other hand, CPU is idle which only idle task on
it, CPUIdle governor can select one specific idle state to power off
hardware logics; we use 'cstate' to present CPU idle state.
Based on trace events 'cpu_idle' and 'cpu_frequency' we can accomplish
the duration statistics for every state. Every time when CPU enters
into or exits from idle states, the trace event 'cpu_idle' is recorded;
trace event 'cpu_frequency' records the event for CPU OPP changing, so
it's easily to know how long time the CPU stays in the specified OPP,
and the CPU must be not in any idle state.
This patch is to utilize the mentioned trace events for pstate and
cstate statistics. To achieve more accurate profiling data, the program
uses below sequence to insure CPU running/idle time aren't missed:
- Before profiling the user space program wakes up all CPUs for once, so
can avoid to missing account time for CPU staying in idle state for
long time; the program forces to set 'scaling_max_freq' to lowest
frequency and then restore 'scaling_max_freq' to highest frequency,
this can ensure the frequency to be set to lowest frequency and later
after start to run workload the frequency can be easily to be changed
to higher frequency;
- User space program reads map data and update statistics for every 5s,
so this is same with other sample bpf programs for avoiding big
overload introduced by bpf program self;
- When send signal to terminate program, the signal handler wakes up
all CPUs, set lowest frequency and restore highest frequency to
'scaling_max_freq'; this is exactly same with the first step so
avoid to missing account CPU pstate and cstate time during last
stage. Finally it reports the latest statistics.
The program has been tested on Hikey board with octa CA53 CPUs, below
is one example for statistics result, the format mainly follows up
Jesper Dangaard Brouer suggestion.
Jesper reminds to 'get printf to pretty print with thousands separators
use %' and setlocale(LC_NUMERIC, "en_US")', tried three different arm64
GCC toolchains (5.4.0 20160609, 6.2.1 20161016, 6.3.0 20170516) but all
of them cannot support printf flag character %' on arm64 platform, so go
back print number without grouping mode.
CPU states statistics:
state(ms) cstate-0 cstate-1 cstate-2 pstate-0 pstate-1 pstate-2 pstate-3 pstate-4
CPU-0 767 6111 111863 561 31 756 853 190
CPU-1 241 10606 107956 484 125 646 990 85
CPU-2 413 19721 98735 636 84 696 757 89
CPU-3 84 11711 79989 17516 909 4811 5773 341
CPU-4 152 19610 98229 444 53 649 708 1283
CPU-5 185 8781 108697 666 91 671 677 1365
CPU-6 157 21964 95825 581 67 566 684 1284
CPU-7 125 15238 102704 398 20 665 786 1197
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Daniel Borkmann says:
====================
Couple of minor improvements to the x64 JIT I had still around from
pre merge window in order to shrink the image size further. Added
test cases for kselftests too as well as running Cilium workloads on
them w/o issues.
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add few test cases that check the rnu-time results under JIT.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
While it's rather cumbersome to reduce prologue for cBPF->eBPF
migrations wrt spill/fill for r15 which is callee saved register
due to bpf_error path in bpf_jit.S that is both used by migrations
as well as native eBPF, we can still trivially save 5 bytes in
prologue for the former since tail calls can never be used there.
cBPF->eBPF migrations also have their own custom prologue in BPF
asm that xors A and X reg anyway, so it's fine we skip this here.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add a generic emit_mov_reg() helper in order to reuse it in BPF
multiplication to load the src into rax, we can save a few bytes
in alu32 while doing so.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Instead of unconditionally performing push/pop on rax/rdx
in case of multiplication, we can save a few bytes in case
of dest register being either BPF r0 (rax) or r3 (rdx)
since the result is written in there anyway.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
While analyzing some of the more complex BPF programs from Cilium,
I found that LLVM generally prefers to emit LD_IMM64 instead of MOV32
BPF instructions for loading unsigned 32-bit immediates into a
register. Given we cannot change the current/stable LLVM versions
that are already out there, lets optimize this case such that the
JIT prefers to emit 'mov %eax, imm32' over 'movabsq %rax, imm64'
whenever suitable in order to reduce the image size by 4-5 bytes per
such load in the typical case, reducing image size on some of the
bigger programs by up to 4%. emit_mov_imm32() and emit_mov_imm64()
have been added as helpers.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
When we shift by one, we can use a different encoding where imm
is not explicitly needed, which saves 1 byte per such op.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Pull networking fixes from David Miller:
1) Fix TTL offset calculation in mac80211 mesh code, from Peter Oh.
2) Fix races with procfs in ipt_CLUSTERIP, from Cong Wang.
3) Memory leak fix in lpm_trie BPF map code, from Yonghong Song.
4) Need to use GFP_ATOMIC in BPF cpumap allocations, from Jason Wang.
5) Fix potential deadlocks in netfilter getsockopt() code paths, from
Paolo Abeni.
6) Netfilter stackpointer size checks really are needed to validate
user input, from Florian Westphal.
7) Missing timer init in x_tables, from Paolo Abeni.
8) Don't use WQ_MEM_RECLAIM in mac80211 hwsim, from Johannes Berg.
9) When an ibmvnic device is brought down then back up again, it can be
sent queue entries from a previous session, handle this properly
instead of crashing. From Thomas Falcon.
10) Fix TCP checksum on LRO buffers in mlx5e, from Gal Pressman.
11) When we are dumping filters in cls_api, the output SKB is empty, and
the filter we are dumping is too large for the space in the SKB, we
should return -EMSGSIZE like other netlink dump operations do.
Otherwise userland has no signal that is needs to increase the size
of its read buffer. From Roman Kapl.
12) Several XDP fixes for virtio_net, from Jesper Dangaard Brouer.
13) Module refcount leak in netlink when a dump start fails, from Jason
Donenfeld.
14) Handle sub-optimal GSO sizes better in TCP BBR congestion control,
from Eric Dumazet.
15) Releasing bpf per-cpu arraymaps can take a long time, add a
condtional scheduling point. From Eric Dumazet.
16) Implement retpolines for tail calls in x64 and arm64 bpf JITs. From
Daniel Borkmann.
17) Fix page leak in gianfar driver, from Andy Spencer.
18) Missed clearing of estimator scratch buffer, from Eric Dumazet.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (76 commits)
net_sched: gen_estimator: fix broken estimators based on percpu stats
gianfar: simplify FCS handling and fix memory leak
ipv6 sit: work around bogus gcc-8 -Wrestrict warning
macvlan: fix use-after-free in macvlan_common_newlink()
bpf, arm64: fix out of bounds access in tail call
bpf, x64: implement retpoline for tail call
rxrpc: Fix send in rxrpc_send_data_packet()
net: aquantia: Fix error handling in aq_pci_probe()
bpf: fix rcu lockdep warning for lpm_trie map_free callback
bpf: add schedule points in percpu arrays management
regulatory: add NUL to request alpha2
ibmvnic: Fix early release of login buffer
net/smc9194: Remove bogus CONFIG_MAC reference
net: ipv4: Set addr_type in hash_keys for forwarded case
tcp_bbr: better deal with suboptimal GSO
smsc75xx: fix smsc75xx_set_features()
netlink: put module reference if dump start fails
selftests/bpf/test_maps: exit child process without error in ENOMEM case
selftests/bpf: update gitignore with test_libbpf_open
selftests/bpf: tcpbpf_kern: use in6_* macros from glibc
..
Pull security subsystem fixes from James Morris:
- keys fixes via David Howells:
"A collection of fixes for Linux keyrings, mostly thanks to Eric
Biggers:
- Fix some PKCS#7 verification issues.
- Fix handling of unsupported crypto in X.509.
- Fix too-large allocation in big_key"
- Seccomp updates via Kees Cook:
"These are fixes for the get_metadata interface that landed during
-rc1. While the new selftest is strictly not a bug fix, I think
it's in the same spirit of avoiding bugs"
- an IMA build fix from Randy Dunlap
* 'fixes-v4.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:
integrity/security: fix digsig.c build error with header file
KEYS: Use individual pages in big_key for crypto buffers
X.509: fix NULL dereference when restricting key with unsupported_sig
X.509: fix BUG_ON() when hash algorithm is unsupported
PKCS#7: fix direct verification of SignerInfo signature
PKCS#7: fix certificate blacklisting
PKCS#7: fix certificate chain verification
seccomp: add a selftest for get_metadata
ptrace, seccomp: tweak get_metadata behavior slightly
seccomp, ptrace: switch get_metadata types to arch independent
A single MIPS fix for mismatching struct compat_flock, resulting in bus
errors starting Firefox on Debian 8 since 4.13.
-----BEGIN PGP SIGNATURE-----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=h/p9
-----END PGP SIGNATURE-----
Merge tag 'mips_fixes_4.16_3' of git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/mips
Pull MIPS fix from James Hogan:
"A single MIPS fix for mismatching struct compat_flock, resulting in
bus errors starting Firefox on Debian 8 since 4.13"
* tag 'mips_fixes_4.16_3' of git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/mips:
MIPS: Drop spurious __unused in struct compat_flock
Pull printk fixlet from Petr Mladek:
"People expect to see the real pointer value for %px.
Let's substitute '(null)' only for the other %p? format modifiers that
need to deference the pointer"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/pmladek/printk:
vsprintf: avoid misleading "(null)" for %px
Pull i2c fixes from Wolfram Sang:
"Two bugfixes, one v4.16 regression fix, and two documentation fixes"
* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: designware: Consider SCL GPIO optional
i2c: busses: i2c-sirf: Fix spelling: "formular" -> "formula".
i2c: bcm2835: Set up the rising/falling edge delays
i2c: i801: Add missing documentation entries for Braswell and Kaby Lake
i2c: designware: must wait for enable
These are mostly fixes for problems with merge window code. In
addition we have one doc update (alua) and two dead code removals
(aiclib and octogon) a spurious assignment removal (csiostor) and a
performance improvement for storvsc involving better interrupt
spreading and increasing the command per lun handling.
Signed-off-by: James E.J. Bottomley <jejb@linux.vnet.ibm.com>
-----BEGIN PGP SIGNATURE-----
iJwEABMIAEQWIQTnYEDbdso9F2cI+arnQslM7pishQUCWo+H2yYcamFtZXMuYm90
dG9tbGV5QGhhbnNlbnBhcnRuZXJzaGlwLmNvbQAKCRDnQslM7pishe2eAQDyWfoK
Mfjbrl6cdPop+JIoED0VtBzAQyeXceJt8GYDQwEApXTIZon2HTdJqGawfUhaapBA
JnO6iOiC13/nZjl7C28=
=K3Pk
-----END PGP SIGNATURE-----
Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"These are mostly fixes for problems with merge window code.
In addition we have one doc update (alua) and two dead code removals
(aiclib and octogon) a spurious assignment removal (csiostor) and a
performance improvement for storvsc involving better interrupt
spreading and increasing the command per lun handling"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: qla4xxx: skip error recovery in case of register disconnect.
scsi: aacraid: fix shutdown crash when init fails
scsi: qedi: Cleanup local str variable
scsi: qedi: Fix truncation of CHAP name and secret
scsi: qla2xxx: Fix incorrect handle for abort IOCB
scsi: qla2xxx: Fix double free bug after firmware timeout
scsi: storvsc: Increase cmd_per_lun for higher speed devices
scsi: qla2xxx: Fix a locking imbalance in qlt_24xx_handle_els()
scsi: scsi_dh: Document alua_rtpg_queue() arguments
scsi: Remove Makefile entry for oktagon files
scsi: aic7xxx: remove aiclib.c
scsi: qla2xxx: Avoid triggering undefined behavior in qla2x00_mbx_completion()
scsi: mptfusion: Add bounds check in mptctl_hp_targetinfo()
scsi: sym53c8xx_2: iterator underflow in sym_getsync()
scsi: bnx2fc: Fix check in SCSI completion handler for timed out request
scsi: csiostor: remove redundant assignment to pointer 'ln'
scsi: ufs: Enable quirk to ignore sending WRITE_SAME command
scsi: ibmvfc: fix misdefined reserved field in ibmvfc_fcp_rsp_info
scsi: qla2xxx: Fix memory corruption during hba reset test
scsi: mpt3sas: fix an out of bound write
For ages iproute2 has used `struct rtmsg` as the ancillary header for
FIB rules and in the process set the protocol value to RTPROT_BOOT.
Until ca56209a66 ("net: Allow a rule to track originating protocol")
the kernel rules code ignored the protocol value sent from userspace
and always returned 0 in notifications. To avoid incompatibility with
existing iproute2, send the protocol as a new attribute.
Fixes: cac56209a6 ("net: Allow a rule to track originating protocol")
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When in switchdev mode, there is no need to do self loopback checks
as we can't receive those packets, we insert steering rules to the
eswitch that make sure packets can't be looped back.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Up until this point it wasn't possible to activate IB representors
when switching to switchdev mode, remove this limitation.
We trigger reload of the PF IB interface in order to make sure that
already allocated resources are invalid and new resources will be opened
correctly with all the limitations of switchdev mode applied (only raw
packet capabilities, without RoCE). We also move the remove/add to a
place where the E-Switch mode is set/unset to better control when to
trigger this action, this will allow the IB side to start in the correct
mode.
For better code reuse, create a function which reloads an interface and
export it.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
This commit adds full support for IB representor:
1) Representors profile, We add two new profiles:
nic_rep_profile - This profile will be used to create an IB device that
represents the PF/UPLINK.
rep_profile - This profile will be used to create an IB device that
represents VFs. Each VF will be its own representor.
2) Proper load/unload callbacks, Those are called by the E-Switch when
moving to/from switchdev mode.
3) Different flow DB handling for when we in switchdev mode.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
In order to forward traffic from representor's SQ to the right virtual
function, every time an SQ is created also add the corresponding flow rule
to the FDB.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
When enabling many VFs and switching to switchdev mode, the total amount
of mkeys we try to allocate when loading representors is very large and
may cause timeouts on allocations, the same issues was observed on VFs
and we employ the same fix that was done for them. We avoid allocating
the full MR cache on load but still allow it to be manipulated once the
IB device is loaded.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Currently in switchdev mode we allow only for raw packet QPs.
Expose the right capabilities and set the gid table length to 0, also
make sure we don't try to enable RoCE, so split the function
to enable RoCE so representors can enable only the notifier needed for
net device events.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Currently we listen to netdev register/unregister event based on PCI
device. When in switchdev mode PF and representors share the same PCI
device, so in order to pair ib device and netdev in switchdev mode
compare the netdev that triggered the event to that of the representor.
Expose a function that lets you receive the netdev associated what
a given representor.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
When we point to a representor, it means we are in switchdev mode.
The flow db is shared between PF and virtual function representors
so each rule created needs to have a match on its specific source port.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
A flow DB is a shared resource between PF and representors,
need to allocate it only when creating the PF IB device.
Once we add IB representors, they will use the flow db which was
created by the PF.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Create the basic infrastructure of registering and unregistering
IB representors. The load/unload callbacks are left empty and
proper implementation will be introduced in following patches.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Create a new representor type: REP_IB. which will be initialized by an IB
device that is used as a logical representor of a eswitch vport (VF or
uplink) just like we have a net device today in switchdev mode.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Under switchdev mode we insert an eswitch miss rule causing any
unmatched traffic to be sent towards the PF vport. This miss rule can
be optimized if we break it to two, one case is for multicast traffic and
the other for unicast.
Breaking the miss rule into two (unicast and multicast) allows the firmware
to program the hardware in a more efficient way.
Using ConncetX-5 Ex with IXIA and testpmd (which use IB representors):
IXIA -> NIC -> PF -> IB representor -> NIC -> VF:
- Without this optimization: 9.2 MPPS.
- With this optimization: 18 MPPS.
VF -> NIC -> IB representor-> PF -> NIC -> IXIA:
- Without this optimization: 17 MPPS.
- With this optimization: 23.4 MPPS.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
The max FTE number should be the max number of SQs that can be opened.
Ethernet representors open one SQ each. Once we add IB representor this
will increase (depends on the user). For now lets start with 31
per IB representor and if needed increase in the future.
This increase only affects the number of FTEs in the slow path FDB,
offloaded rules (done via TC on the fast path portion of the FDB)
aren't affected.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
In preparation for IB representors, move representors structs to a global
scope, also expose functions needed for registration, unregistration,
eswitch mode and creating a flow rule to direct traffic from SQs to the
right VF.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Add a callback interface to get a protocol device (per representor type).
The Ethernet representors will expose their netdev via this interface.
This functionality can be later used by IB representor in order to find the
corresponding net device representor.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
-----BEGIN PGP SIGNATURE-----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=LRdE
-----END PGP SIGNATURE-----
Merge tag 'drm-fixes-for-v4.16-rc3' of git://people.freedesktop.org/~airlied/linux
Pull drm fixes from Dave Airlie:
"A bunch of fixes for rc3:
Exynos:
- fixes for using monotonic timestamps
- register definitions
- removal of unused file
ipu-v3L
- minor changes
- make some register arrays const+static
- fix some leaks
meson:
- fix for vsync
atomic:
- fix for memory leak
EDID parser:
- add quirks for some more non-desktop devices
- 6-bit panel fix.
drm_mm:
- fix a bug in the core drm mm hole handling
cirrus:
- fix lut loading regression
Lastly there is a deadlock fix around runtime suspend for secondary
GPUs.
There was a deadlock between one thread trying to wait for a workqueue
job to finish in the runtime suspend path, and the workqueue job it
was waiting for in turn waiting for a runtime_get_sync to return.
The fixes avoids it by not doing the runtime sync in the workqueue as
then we always wait for all those tasks to complete before we runtime
suspend"
* tag 'drm-fixes-for-v4.16-rc3' of git://people.freedesktop.org/~airlied/linux: (25 commits)
drm/tve200: fix kernel-doc documentation comment include
drm/edid: quirk Sony PlayStation VR headset as non-desktop
drm/edid: quirk Windows Mixed Reality headsets as non-desktop
drm/edid: quirk Oculus Rift headsets as non-desktop
drm/meson: fix vsync buffer update
drm: Handle unexpected holes in color-eviction
drm: exynos: Use proper macro definition for HDMI_I2S_PIN_SEL_1
drm/exynos: remove exynos_drm_rotator.h
drm/exynos: g2d: Delete an error message for a failed memory allocation in two functions
drm/exynos: fix comparison to bitshift when dealing with a mask
drm/exynos: g2d: use monotonic timestamps
drm/edid: Add 6 bpc quirk for CPT panel in Asus UX303LA
gpu: ipu-csi: add 10/12-bit grayscale support to mbus_code_to_bus_cfg
gpu: ipu-cpmem: add 16-bit grayscale support to ipu_cpmem_set_image
gpu: ipu-v3: prg: fix device node leak in ipu_prg_lookup_by_phandle
gpu: ipu-v3: pre: fix device node leak in ipu_pre_lookup_by_phandle
drm/amdgpu: Fix deadlock on runtime suspend
drm/radeon: Fix deadlock on runtime suspend
drm/nouveau: Fix deadlock on runtime suspend
drm: Allow determining if current task is output poll worker
...
The packet fanout test generates UDP traffic and reads this with
a pair of packet sockets, testing the various fanout algorithms.
Avoid non-determinism from reading unrelated background traffic.
Fanout decisions are made before unrelated packets can be dropped with
a filter, so that is an insufficient strategy [*]. Run the packet
socket tests in a network namespace, similar to msg_zerocopy.
It it still good practice to install a filter on a packet socket
before accepting traffic. Because this is example code, demonstrate
that pattern. Open the socket initially bound to no protocol, install
a filter, and only then bind to ETH_P_IP.
Another source of non-determinism is hash collisions in FANOUT_HASH.
The hash function used to select a socket in the fanout group includes
the pseudorandom number hashrnd, which is not visible from userspace.
To work around this, the test tries to find a pair of UDP source ports
that do not collide. It gives up too soon (5 times, every 32 runs) and
output is confusing. Increase tries to 20 and revise the error msg.
[*] another approach would be to add a third socket to the fanout
group and direct all unexpected traffic here. This is possible
only when reimplementing methods like RR or HASH alongside this
extra catch-all bucket, using the BPF fanout method.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Zero is being bit-wise or'd in a calculation twice; these are redundant
and can be removed.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
pfifo_fast got percpu stats lately, uncovering a bug I introduced last
year in linux-4.10.
I missed the fact that we have to clear our temporary storage
before calling __gnet_stats_copy_basic() in the case of percpu stats.
Without this fix, rate estimators (tc qd replace dev xxx root est 1sec
4sec pfifo_fast) are utterly broken.
Fixes: 1c0d32fde5 ("net_sched: gen_estimator: complete rewrite of rate estimators")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Alexei Starovoitov says:
====================
pull-request: bpf 2018-02-22
The following pull-request contains BPF updates for your *net* tree.
The main changes are:
1) two urgent fixes for bpf_tail_call logic for x64 and arm64 JITs, from Daniel.
2) cond_resched points in percpu array alloc/free paths, from Eric.
3) lockdep and other minor fixes, from Yonghong, Arnd, Anders, Li.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
if either or both of MSG_ZEROCOPY and SOCK_ZEROCOPY have not been
specified, the rm->data.op_mmp_znotifier allocation will be skipped.
In this case, it is invalid ot pass down a cmsghdr with
RDS_CMSG_ZCOPY_COOKIE, so return EINVAL from rds_msg_zcopy for this
case.
Reported-by: syzbot+f893ae7bb2f6456dfbc3@syzkaller.appspotmail.com
Fixes: 0cebaccef3 ("rds: zerocopy Tx support.")
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
r8168_check_dash() returns false anyway for all chip versions not
supporting dash. So we can simplify the check conditions.
In addition change the check functions to return bool instead of int,
because they actually return a bool value.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, if BIOS enables WOL in the chip, settings are inconsistent
because the device isn't marked as wakeup-enabled (if not done
explicitly via userspace tools). This causes issues with suspend/
resume because mdio_bus_phy_may_suspend() checks whether device is
wakeup-enabled. In detail MDIO bus access in phy_suspend() can fail
because the MDIO bus is disabled.
In the history of the driver we find two competing approaches:
8f9d513803 "r8169: remember WOL preferences on driver load" prefers
to preserve what the BIOS may have set, whilst bde135a672
"r8169: only enable PCI wakeups when WOL is active" disabled PCI
wakeup per default to work around a bug on one platform.
Seems like nobody complained after the latter patch about non-working
WOL, what makes me think that nobody uses WOL w/o configuring it
explicitly.
My opinion:
Vast majority of users doesn't use WOL even if the BIOS enables it in
the chip. And having WOL being active keeps the PHY(s) from powering
down if being idle.
If somebody needs WOL, he can enable it during boot, e.g. by
configuring systemd.link/WakeOnLan.
Therefore, to make WOL consistent again, disable it per default.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Previously, buffer descriptors containing only the frame check sequence
(FCS) were skipped and not added to the skb. However, the page reference
count was still incremented, leading to a memory leak.
Fixing this inside gfar_add_rx_frag() is difficult due to reserved
memory handling and page reuse. Instead, move the FCS handling to
gfar_process_frame() and trim off the FCS before passing the skb up the
networking stack.
Signed-off-by: Andy Spencer <aspencer@spacex.com>
Signed-off-by: Jim Gruen <jgruen@spacex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
gcc-8 has a new warning that detects overlapping input and output arguments
in memcpy(). It triggers for sit_init_net() calling ipip6_tunnel_clone_6rd(),
which is actually correct:
net/ipv6/sit.c: In function 'sit_init_net':
net/ipv6/sit.c:192:3: error: 'memcpy' source argument is the same as destination [-Werror=restrict]
The problem here is that the logic detecting the memcpy() arguments finds them
to be the same, but the conditional that tests for the input and output of
ipip6_tunnel_clone_6rd() to be identical is not a compile-time constant.
We know that netdev_priv(t->dev) is the same as t for a tunnel device,
and comparing "dev" directly here lets the compiler figure out as well
that 'dev == sitn->fb_tunnel_dev' when called from sit_init_net(), so
it no longer warns.
This code is old, so Cc stable to make sure that we don't get the warning
for older kernels built with new gcc.
Cc: Martin Sebor <msebor@gmail.com>
Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83456
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
The following use-after-free was reported by KASan when running
LTP macvtap01 test on 4.16-rc2:
[10642.528443] BUG: KASAN: use-after-free in
macvlan_common_newlink+0x12ef/0x14a0 [macvlan]
[10642.626607] Read of size 8 at addr ffff880ba49f2100 by task ip/18450
...
[10642.963873] Call Trace:
[10642.994352] dump_stack+0x5c/0x7c
[10643.035325] print_address_description+0x75/0x290
[10643.092938] kasan_report+0x28d/0x390
[10643.137971] ? macvlan_common_newlink+0x12ef/0x14a0 [macvlan]
[10643.207963] macvlan_common_newlink+0x12ef/0x14a0 [macvlan]
[10643.275978] macvtap_newlink+0x171/0x260 [macvtap]
[10643.334532] rtnl_newlink+0xd4f/0x1300
...
[10646.256176] Allocated by task 18450:
[10646.299964] kasan_kmalloc+0xa6/0xd0
[10646.343746] kmem_cache_alloc_trace+0xf1/0x210
[10646.397826] macvlan_common_newlink+0x6de/0x14a0 [macvlan]
[10646.464386] macvtap_newlink+0x171/0x260 [macvtap]
[10646.522728] rtnl_newlink+0xd4f/0x1300
...
[10647.022028] Freed by task 18450:
[10647.061549] __kasan_slab_free+0x138/0x180
[10647.111468] kfree+0x9e/0x1c0
[10647.147869] macvlan_port_destroy+0x3db/0x650 [macvlan]
[10647.211411] rollback_registered_many+0x5b9/0xb10
[10647.268715] rollback_registered+0xd9/0x190
[10647.319675] register_netdevice+0x8eb/0xc70
[10647.370635] macvlan_common_newlink+0xe58/0x14a0 [macvlan]
[10647.437195] macvtap_newlink+0x171/0x260 [macvtap]
Commit d02fd6e7d2 ("macvlan: Fix one possible double free") handles
the case when register_netdevice() invokes ndo_uninit() on error and
as a result free the port. But 'macvlan_port_get_rtnl(dev))' check
(returns dev->rx_handler_data), which was added by this commit in order
to prevent double free, is not quite correct:
* for macvlan it always returns NULL because 'lowerdev' is the one that
was used to register rx handler (port) in macvlan_port_create() as
well as to unregister it in macvlan_port_destroy().
* for macvtap it always returns a valid pointer because macvtap registers
its own rx handler before macvlan_common_newlink().
Fixes: d02fd6e7d2 ("macvlan: Fix one possible double free")
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sk is already allocated in inet_create/inet6_create, hence when
BPF_CGROUP_RUN_PROG_INET_SOCK is executed sk will never be NULL.
The logic is as bellow,
sk = sk_alloc();
if (!sk)
goto out;
BPF_CGROUP_RUN_PROG_INET_SOCK(sk);
Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
do_task_stat() calls get_wchan(), which further does unwind_frame().
unwind_frame() restores frame->pc to original value in case function
graph tracer has modified a return address (LR) in a stack frame to hook
a function return. However, if function graph tracer has hit a filtered
function, then we can't unwind it as ftrace_push_return_trace() has
biased the index(frame->graph) with a 'huge negative'
offset(-FTRACE_NOTRACE_DEPTH).
Moreover, arm64 stack walker defines index(frame->graph) as unsigned
int, which can not compare a -ve number.
Similar problem we can have with calling of walk_stackframe() from
save_stack_trace_tsk() or dump_backtrace().
This patch fixes unwind_frame() to test the index for -ve value and
restore index accordingly before we can restore frame->pc.
Reproducer:
cd /sys/kernel/debug/tracing/
echo schedule > set_graph_notrace
echo 1 > options/display-graph
echo wakeup > current_tracer
ps -ef | grep -i agent
Above commands result in:
Unable to handle kernel paging request at virtual address ffff801bd3d1e000
pgd = ffff8003cbe97c00
[ffff801bd3d1e000] *pgd=0000000000000000, *pud=0000000000000000
Internal error: Oops: 96000006 [#1] SMP
[...]
CPU: 5 PID: 11696 Comm: ps Not tainted 4.11.0+ #33
[...]
task: ffff8003c21ba000 task.stack: ffff8003cc6c0000
PC is at unwind_frame+0x12c/0x180
LR is at get_wchan+0xd4/0x134
pc : [<ffff00000808892c>] lr : [<ffff0000080860b8>] pstate: 60000145
sp : ffff8003cc6c3ab0
x29: ffff8003cc6c3ab0 x28: 0000000000000001
x27: 0000000000000026 x26: 0000000000000026
x25: 00000000000012d8 x24: 0000000000000000
x23: ffff8003c1c04000 x22: ffff000008c83000
x21: ffff8003c1c00000 x20: 000000000000000f
x19: ffff8003c1bc0000 x18: 0000fffffc593690
x17: 0000000000000000 x16: 0000000000000001
x15: 0000b855670e2b60 x14: 0003e97f22cf1d0f
x13: 0000000000000001 x12: 0000000000000000
x11: 00000000e8f4883e x10: 0000000154f47ec8
x9 : 0000000070f367c0 x8 : 0000000000000000
x7 : 00008003f7290000 x6 : 0000000000000018
x5 : 0000000000000000 x4 : ffff8003c1c03cb0
x3 : ffff8003c1c03ca0 x2 : 00000017ffe80000
x1 : ffff8003cc6c3af8 x0 : ffff8003d3e9e000
Process ps (pid: 11696, stack limit = 0xffff8003cc6c0000)
Stack: (0xffff8003cc6c3ab0 to 0xffff8003cc6c4000)
[...]
[<ffff00000808892c>] unwind_frame+0x12c/0x180
[<ffff000008305008>] do_task_stat+0x864/0x870
[<ffff000008305c44>] proc_tgid_stat+0x3c/0x48
[<ffff0000082fde0c>] proc_single_show+0x5c/0xb8
[<ffff0000082b27e0>] seq_read+0x160/0x414
[<ffff000008289e6c>] __vfs_read+0x58/0x164
[<ffff00000828b164>] vfs_read+0x88/0x144
[<ffff00000828c2e8>] SyS_read+0x60/0xc0
[<ffff0000080834a0>] __sys_trace_return+0x0/0x4
Fixes: 20380bb390 (arm64: ftrace: fix a stack tracer's output under function graph tracer)
Signed-off-by: Pratyush Anand <panand@redhat.com>
Signed-off-by: Jerome Marchand <jmarchan@redhat.com>
[catalin.marinas@arm.com: replace WARN_ON with WARN_ON_ONCE]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
David S. Miller