The default behavior for source MACVLAN is to duplicate packets to
appropriate type source devices, and then do the normal destination MACVLAN
flow. This patch adds an option to skip destination MACVLAN processing if
any matching source MACVLAN device has the option set.
This allows setting up a "catch all" device for source MACVLAN: create one
or more devices with type source nodst, and one device with e.g. type vepa,
and incoming traffic will be received on exactly one device.
v2: netdev wants non-standard line length
Signed-off-by: Jethro Beekman <kernel@jbeekman.nl> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 26 Apr 2021 01:31:35 +0000 (18:31 -0700)]
Merge tag 'mlx5-updates-2021-04-21' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux
Saeed Mahameed says:
====================
mlx5-updates-2021-04-21
devlink external port attribute for SF (Sub-Function) port flavour
This adds the support to instantiate Sub-Functions on external hosts
E.g when Eswitch manager is enabled on the ARM SmarNic SoC CPU, users
are now able to spawn new Sub-Functions on the Host server CPU.
Parav Pandit Says:
==================
This series introduces and uses external attribute for the SF port to
indicate that a SF port belongs to an external controller.
This is needed to generate unique phys_port_name when PF and SF numbers
are overlapping between local and external controllers.
For example two controllers 0 and 1, both of these controller have a SF.
having PF number 0, SF number 77. Here, phys_port_name has duplicate
entry which doesn't have controller number in it.
Hence, add controller number optionally when a SF port is for an
external controller. This extension is similar to existing PF and VF
eswitch ports of the external controller.
When a SF is for external controller an example view of external SF
port and config sequence:
On eswitch system:
$ devlink dev eswitch set pci/0033:01:00.0 mode switchdev
$ devlink port show
pci/0033:01:00.0/196607: type eth netdev enP51p1s0f0np0 flavour physical port 0 splittable false
pci/0033:01:00.0/131072: type eth netdev eth0 flavour pcipf controller 1 pfnum 0 external true splittable false
function:
hw_addr 00:00:00:00:00:00
$ devlink port add pci/0033:01:00.0 flavour pcisf pfnum 0 sfnum 77 controller 1
pci/0033:01:00.0/163840: type eth netdev eth1 flavour pcisf controller 1 pfnum 0 sfnum 77 splittable false
function:
hw_addr 00:00:00:00:00:00 state inactive opstate detached
Patch summary:
First 3 patches prepares the eswitch to handle vports in more generic
way using xarray to lookup vport from its unique vport number.
Patch-1 returns maximum eswitch ports only when eswitch is enabled
Patch-2 prepares eswitch to return eswitch max ports from a struct
Patch-3 uses xarray for vport and representor lookup
Patch-4 considers SF for an additioanl range of SF vports
Patch-5 relies on SF hw table to check SF support
Patch-6 extends SF devlink port attribute for external flag
Patch-7 stores the per controller SF allocation attributes
Patch-8 uses SF function id for filtering events
Patch-9 uses helper for allocation and free
Patch-10 splits hw table into per controller table and generic one
Patch-11 extends sf table for additional range
==================
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
net: ethernet: ixp4xx: Retire ancient phy retrieveal
This driver was using a really dated way of obtaining the
phy by printing a string and using it with phy_connect().
Switch to using more reasonable modern interfaces.
Suggested-by: Andrew Lunn <andrew@lunn.ch> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Linus Walleij <linus.walleij@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Hayes Wang [Sat, 24 Apr 2021 06:09:03 +0000 (14:09 +0800)]
r8152: remove some bit operations
Remove DELL_TB_RX_AGG_BUG and LENOVO_MACPASSTHRU flags of rtl8152_flags.
They are only set when initializing and wouldn't be change. It is enough
to record them with variables.
Signed-off-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Dexuan Cui [Sat, 24 Apr 2021 01:12:35 +0000 (18:12 -0700)]
hv_netvsc: Make netvsc/VF binding check both MAC and serial number
Currently the netvsc/VF binding logic only checks the PCI serial number.
The Microsoft Azure Network Adapter (MANA) supports multiple net_device
interfaces (each such interface is called a "vPort", and has its unique
MAC address) which are backed by the same VF PCI device, so the binding
logic should check both the MAC address and the PCI serial number.
The change should not break any other existing VF drivers, because
Hyper-V NIC SR-IOV implementation requires the netvsc network
interface and the VF network interface have the same MAC address.
Co-developed-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com> Co-developed-by: Shachar Raindel <shacharr@microsoft.com> Signed-off-by: Shachar Raindel <shacharr@microsoft.com> Acked-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Variable result is being assigned a value from a calculation
however the variable is never read, so this redundant variable
can be removed.
Cleans up the following clang-analyzer warning:
drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c:1488:2:
warning: Value stored to 'pos' is never read
[clang-analyzer-deadcode.DeadStores].
drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c:876:3:
warning: Value stored to 'pos' is never read
[clang-analyzer-deadcode.DeadStores].
drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c:36:3:
warning: Value stored to 'start' is never read
[clang-analyzer-deadcode.DeadStores].
Reported-by: Abaci Robot <abaci@linux.alibaba.com> Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Parav Pandit [Mon, 8 Mar 2021 11:19:53 +0000 (13:19 +0200)]
net/mlx5: SF, Extend SF table for additional SF id range
Extended the SF table to cover additioanl SF id range of external
controller.
A user optionallly provides the external controller number when user
wants to create SF on the external controller.
An example on eswitch system:
$ devlink dev eswitch set pci/0033:01:00.0 mode switchdev
$ devlink port show
pci/0033:01:00.0/196607: type eth netdev enP51p1s0f0np0 flavour physical port 0 splittable false
pci/0033:01:00.0/131072: type eth netdev eth0 flavour pcipf controller 1 pfnum 0 external true splittable false
function:
hw_addr 00:00:00:00:00:00
$ devlink port add pci/0033:01:00.0 flavour pcisf pfnum 0 sfnum 77 controller 1
pci/0033:01:00.0/163840: type eth netdev eth1 flavour pcisf controller 1 pfnum 0 sfnum 77 external true splittable false
function:
hw_addr 00:00:00:00:00:00 state inactive opstate detached
Parav Pandit [Fri, 5 Mar 2021 08:06:06 +0000 (10:06 +0200)]
net/mlx5: SF, Split mlx5_sf_hw_table into two parts
Device has SF ids in two different contiguous ranges. One for the local
controller and second for the external controller's PF.
Each such range has its own maximum number of functions and base id.
To allocate SF from either of the range, prepare code to split into
range specific fields into its own structure.
Parav Pandit [Wed, 10 Mar 2021 13:35:03 +0000 (15:35 +0200)]
devlink: Extend SF port attributes to have external attribute
Extended SF port attributes to have optional external flag similar to
PCI PF and VF port attributes.
External atttibute is required to generate unique phys_port_name when PF number
and SF number are overlapping between two controllers similar to SR-IOV
VFs.
When a SF is for external controller an example view of external SF
port and config sequence.
On eswitch system:
$ devlink dev eswitch set pci/0033:01:00.0 mode switchdev
$ devlink port show
pci/0033:01:00.0/196607: type eth netdev enP51p1s0f0np0 flavour physical port 0 splittable false
pci/0033:01:00.0/131072: type eth netdev eth0 flavour pcipf controller 1 pfnum 0 external true splittable false
function:
hw_addr 00:00:00:00:00:00
$ devlink port add pci/0033:01:00.0 flavour pcisf pfnum 0 sfnum 77 controller 1
pci/0033:01:00.0/163840: type eth netdev eth1 flavour pcisf controller 1 pfnum 0 sfnum 77 splittable false
function:
hw_addr 00:00:00:00:00:00 state inactive opstate detached
Parav Pandit [Fri, 19 Mar 2021 03:21:31 +0000 (05:21 +0200)]
net/mlx5: E-Switch, Use xarray for vport number to vport and rep mapping
Currently vport number to vport and its representor are mapped using an
array and an index.
Vport numbers of different types of functions are not contiguous. Adding
new such discontiguous range using index and number mapping is increasingly
complex and hard to maintain.
Hence, maintain an xarray of vport and rep whose lookup is done based on
the vport number.
Each VF and SF entry is marked with a xarray mark to identify the function
type. Additionally PF and VF needs special handling for legacy inline
mode. They are additionally marked as host function using additional
HOST_FN mark.
Signed-off-by: Parav Pandit <parav@nvidia.com> Reviewed-by: Roi Dayan <roid@nvidia.com> Reviewed-by: Vu Pham <vuhuong@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Parav Pandit [Tue, 2 Mar 2021 12:10:49 +0000 (14:10 +0200)]
net/mlx5: E-Switch, Prepare to return total vports from eswitch struct
Total vports are already stored during eswitch initialization. Instead
of calculating everytime, read directly from eswitch.
Additionally, host PF's SF vport information is available using
QUERY_HCA_CAP command. It is not available through HCA_CAP of the
eswitch manager PF.
Hence, this patch prepares the return total eswitch vport count from the
existing eswitch struct.
This further helps to keep eswitch port counting macros and logic within
eswitch.
When MLX5_ESWITCH is disabled, FS layer continues to initialize eswitch
specific ACL namespaces.
Instead, start honoring MLX5_ESWITCH flag and perform vport specific
initialization only when vport count is non zero.
Signed-off-by: Parav Pandit <parav@nvidia.com> Reviewed-by: Roi Dayan <roid@nvidia.com> Reviewed-by: Vu Pham <vuhuong@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Tiezhu Yang [Fri, 23 Apr 2021 01:23:30 +0000 (09:23 +0800)]
bpf: Document the pahole release info related to libbpf in bpf_devel_QA.rst
pahole starts to use libbpf definitions and APIs since v1.13 after the
commit 21507cd3e97b ("pahole: add libbpf as submodule under lib/bpf").
It works well with the git repository because the libbpf submodule will
use "git submodule update --init --recursive" to update.
Unfortunately, the default github release source code does not contain
libbpf submodule source code and this will cause build issues, the tarball
from https://git.kernel.org/pub/scm/devel/pahole/pahole.git/ is same with
github, you can get the source tarball with corresponding libbpf submodule
codes from
https://fedorapeople.org/~acme/dwarves
This change documents the above issues to give more information so that
we can get the tarball from the right place, early discussion is here:
Added .config_intr and .handle_interrupt callbacks.
Link event interrupt will trigger an interrupt every time when the link
goes up or down.
Signed-off-by: Radu Pirea (NXP OSS) <radu-nicolae.pirea@oss.nxp.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Po-Hsu Lin [Fri, 23 Apr 2021 11:15:38 +0000 (19:15 +0800)]
selftests/net: bump timeout to 5 minutes
We found that with the latest mainline kernel (5.12.0-051200rc8) on
some KVM instances / bare-metal systems, the following tests will take
longer than the kselftest framework default timeout (45 seconds) to
run and thus got terminated with TIMEOUT error:
* xfrm_policy.sh - took about 2m20s
* pmtu.sh - took about 3m5s
* udpgso_bench.sh - took about 60s
Bump the timeout setting to 5 minutes to allow them have a chance to
finish.
https://bugs.launchpad.net/bugs/1856010 Signed-off-by: Po-Hsu Lin <po-hsu.lin@canonical.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 23 Apr 2021 21:06:32 +0000 (14:06 -0700)]
Merge branch 'mptcp-msg-flags'
Mat Martineau says:
====================
mptcp: Compatibility with common msg flags
These patches from the MPTCP tree handle some of the msg flags that are
typically used with TCP, to make it easier to adapt userspace programs
for use with MPTCP.
Patches 1, 2, and 4 add support for MSG_ERRQUEUE (no-op for now),
MSG_TRUNC, and MSG_PEEK on the receive side.
Patch 3 ignores unsupported msg flags for send and receive.
Patch 5 adds a selftest for MSG_PEEK.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Yonglong Li [Fri, 23 Apr 2021 18:17:09 +0000 (11:17 -0700)]
selftests: mptcp: add a test case for MSG_PEEK
Extend mptcp_connect tool with MSG_PEEK support and add a test case in
mptcp_connect.sh that checks the data received from/after recv() with
MSG_PEEK.
Acked-by: Paolo Abeni <pabeni@redhat.com> Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net> Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net> Signed-off-by: Yonglong Li <liyonglong@chinatelecom.cn> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yonglong Li [Fri, 23 Apr 2021 18:17:08 +0000 (11:17 -0700)]
mptcp: add MSG_PEEK support
This patch adds support for MSG_PEEK flag. Packets are not removed
from the receive_queue if MSG_PEEK set in recv() system call.
Acked-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Yonglong Li <liyonglong@chinatelecom.cn> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Fri, 23 Apr 2021 18:17:07 +0000 (11:17 -0700)]
mptcp: ignore unsupported msg flags
Currently mptcp_sendmsg() fails with EOPNOTSUPP if the
user-space provides some unsupported flag. That is unexpected
and may foul existing applications migrated to MPTCP, which
expect a different behavior.
Change the mentioned function to silently ignore the unsupported
flags except MSG_FASTOPEN. This is the only flags currently not
supported by MPTCP with user-space visible side-effects.
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/162 Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Fri, 23 Apr 2021 18:17:06 +0000 (11:17 -0700)]
mptcp: implement MSG_TRUNC support
The mentioned flag is currently silenlty ignored. This
change implements the TCP-like behaviour, dropping the
pending data up to the specified length.
Signed-off-by: Paolo Abeni <pabeni@redhat.com> Sigend-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Fri, 23 Apr 2021 18:17:05 +0000 (11:17 -0700)]
mptcp: implement dummy MSG_ERRQUEUE support
mptcp_recvmsg() currently silently ignores MSG_ERRQUEUE, returning
input data instead of error cmsg.
This change provides a dummy implementation for MSG_ERRQUEUE - always
returns no data. That is consistent with the current lack of a suitable
IP_RECVERR setsockopt() support.
Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Add BPF static linker support for extern resolution of global variables,
functions, and BTF-defined maps.
This patch set consists of 4 parts:
- few patches are extending bpftool to simplify working with BTF dump;
- libbpf object loading logic is extended to support __hidden functions and
overriden (unused) __weak functions; also BTF-defined map parsing logic is
refactored to be re-used by linker;
- the crux of the patch set is BPF static linker logic extension to perform
extern resolution for three categories: global variables, BPF
sub-programs, BTF-defined maps;
- a set of selftests that validate that all the combinations of
extern/weak/__hidden are working as expected.
See respective patches for more details.
One aspect hasn't been addressed yet and is going to be resolved in the next
patch set, but is worth mentioning. With BPF static linking of multiple .o
files, dealing with static everything becomes more problematic for BPF
skeleton and in general for any by name look up APIs. This is due to static
entities are allowed to have non-unique name. Historically this was never
a problem due to BPF programs were always confined to a single C file. That
changes now and needs to be addressed. The thinking so far is for BPF static
linker to prepend filename to each static variable and static map (which is
currently not supported by libbpf, btw), so that they can be unambiguously
resolved by (mostly) unique name. Mostly, because even filenames can be
duplicated, but that should be rare and easy to address by wiser choice of
filenames by users. Fortunately, static BPF subprograms don't suffer from this
issues, as they are not independent entities and are neither exposed in BPF
skeleton, nor is lookup-able by any of libbpf APIs (and there is little reason
to do that anyways).
This and few other things will be the topic of the next set of patches.
Some tests rely on Clang fix ([0]), so need latest Clang built from main.
v2->v3:
- allow only STV_DEFAULT and STV_HIDDEN ELF symbol visibility (Yonghong);
- update selftests' README for required Clang 13 fix dependency (Alexei);
- comments, typos, slight code changes (Yonghong, Alexei);
v1->v2:
- make map externs support full attribute list, adjust linked_maps selftest
to demonstrate that typedef works now (though no shared header file was
added to simplicity sake) (Alexei);
- remove commented out parts from selftests and fix few minor code style
issues;
- special __weak map definition semantics not yet implemented and will be
addressed in a follow up.
====================
Add selftest validating various aspects of statically linking BTF-defined map
definitions. Legacy map definitions do not support extern resolution between
object files. Some of the aspects validated:
- correct resolution of extern maps against concrete map definitions;
- extern maps can currently only specify map type and key/value size and/or
type information;
- weak concrete map definitions are resolved properly.
Static map definitions are not yet supported by libbpf, so they are not
explicitly tested, though manual testing showes that BPF linker handles them
properly.
selftests/bpf: Add global variables linking selftest
Add selftest validating various aspects of statically linking global
variables:
- correct resolution of extern variables across .bss, .data, and .rodata
sections;
- correct handling of weak definitions;
- correct de-duplication of repeating special externs (.kconfig, .ksyms).
Add selftest validating various aspects of statically linking functions:
- no conflicts and correct resolution for name-conflicting static funcs;
- correct resolution of extern functions;
- correct handling of weak functions, both resolution itself and libbpf's
handling of unused weak function that "lost" (it leaves gaps in code with
no ELF symbols);
- correct handling of hidden visibility to turn global function into
"static" for the purpose of BPF verification.
selftests/bpf: Omit skeleton generation for multi-linked BPF object files
Skip generating individual BPF skeletons for files that are supposed to be
linked together to form the final BPF object file. Very often such files are
"incomplete" BPF object files, which will fail libbpf bpf_object__open() step,
if used individually, thus failing BPF skeleton generation. This is by design,
so skip individual BPF skeletons and only validate them as part of their
linked final BPF object file and skeleton.
selftests/bpf: Use -O0 instead of -Og in selftests builds
While -Og is designed to work well with debugger, it's still inferior to -O0
in terms of debuggability experience. It will cause some variables to still be
inlined, it will also prevent single-stepping some statements and otherwise
interfere with debugging experience. So switch to -O0 which turns off any
optimization and provides the best debugging experience.
libbpf: Support extern resolution for BTF-defined maps in .maps section
Add extra logic to handle map externs (only BTF-defined maps are supported for
linking). Re-use the map parsing logic used during bpf_object__open(). Map
externs are currently restricted to always match complete map definition. So
all the specified attributes will be compared (down to pining, map_flags,
numa_node, etc). In the future this restriction might be relaxed with no
backwards compatibility issues. If any attribute is mismatched between extern
and actual map definition, linker will report an error, pointing out which one
mismatches.
The original intent was to allow for extern to specify attributes that matters
(to user) to enforce. E.g., if you specify just key information and omit
value, then any value fits. Similarly, it should have been possible to enforce
map_flags, pinning, and any other possible map attribute. Unfortunately, that
means that multiple externs can be only partially overlapping with each other,
which means linker would need to combine their type definitions to end up with
the most restrictive and fullest map definition. This requires an extra amount
of BTF manipulation which at this time was deemed unnecessary and would
require further extending generic BTF writer APIs. So that is left for future
follow ups, if there will be demand for that. But the idea seems intresting
and useful, so I want to document it here.
Weak definitions are also supported, but are pretty strict as well, just
like externs: all weak map definitions have to match exactly. In the follow up
patches this most probably will be relaxed, with __weak map definitions being
able to differ between each other (with non-weak definition always winning, of
course).
libbpf: Add linker extern resolution support for functions and global variables
Add BPF static linker logic to resolve extern variables and functions across
multiple linked together BPF object files.
For that, linker maintains a separate list of struct glob_sym structures,
which keeps track of few pieces of metadata (is it extern or resolved global,
is it a weak symbol, which ELF section it belongs to, etc) and ties together
BTF type info and ELF symbol information and keeps them in sync.
With adding support for extern variables/funcs, it's now possible for some
sections to contain both extern and non-extern definitions. This means that
some sections may start out as ephemeral (if only externs are present and thus
there is not corresponding ELF section), but will be "upgraded" to actual ELF
section as symbols are resolved or new non-extern definitions are appended.
Additional care is taken to not duplicate extern entries in sections like
.kconfig and .ksyms.
Given libbpf requires BTF type to always be present for .kconfig/.ksym
externs, linker extends this requirement to all the externs, even those that
are supposed to be resolved during static linking and which won't be visible
to libbpf. With BTF information always present, static linker will check not
just ELF symbol matches, but entire BTF type signature match as well. That
logic is stricter that BPF CO-RE checks. It probably should be re-used by
.ksym resolution logic in libbpf as well, but that's left for follow up
patches.
To make it unnecessary to rewrite ELF symbols and minimize BTF type
rewriting/removal, ELF symbols that correspond to externs initially will be
updated in place once they are resolved. Similarly for BTF type info, VAR/FUNC
and var_secinfo's (sec_vars in struct bpf_linker) are staying stable, but
types they point to might get replaced when extern is resolved. This might
leave some left-over types (even though we try to minimize this for common
cases of having extern funcs with not argument names vs concrete function with
names properly specified). That can be addresses later with a generic BTF
garbage collection. That's left for a follow up as well.
Given BTF type appending phase is separate from ELF symbol
appending/resolution, special struct glob_sym->underlying_btf_id variable is
used to communicate resolution and rewrite decisions. 0 means
underlying_btf_id needs to be appended (it's not yet in final linker->btf), <0
values are used for temporary storage of source BTF type ID (not yet
rewritten), so -glob_sym->underlying_btf_id is BTF type id in obj-btf. But by
the end of linker_append_btf() phase, that underlying_btf_id will be remapped
and will always be > 0. This is the uglies part of the whole process, but
keeps the other parts much simpler due to stability of sec_var and VAR/FUNC
types, as well as ELF symbol, so please keep that in mind while reviewing.
BTF-defined maps require some extra custom logic and is addressed separate in
the next patch, so that to keep this one smaller and easier to review.
libbpf: Extend sanity checking ELF symbols with externs validation
Add logic to validate extern symbols, plus some other minor extra checks, like
ELF symbol #0 validation, general symbol visibility and binding validations.
Factor out logic for sanity checking SHT_SYMTAB and SHT_REL sections into
separate sections. They are already quite extensive and are suffering from too
deep indentation. Subsequent changes will extend SYMTAB sanity checking
further, so it's better to factor each into a separate function.
Refactor BTF-defined maps parsing logic to allow it to be nicely reused by BPF
static linker. Further, at least for BPF static linker, it's important to know
which attributes of a BPF map were defined explicitly, so provide a bit set
for each known portion of BTF map definition. This allows BPF static linker to
do a simple check when dealing with extern map declarations.
The same capabilities allow to distinguish attributes explicitly set to zero
(e.g., __uint(max_entries, 0)) vs the case of not specifying it at all (no
max_entries attribute at all). Libbpf is currently not utilizing that, but it
could be useful for backwards compatibility reasons later.
libbpf: Allow gaps in BPF program sections to support overriden weak functions
Currently libbpf is very strict about parsing BPF program instruction
sections. No gaps are allowed between sequential BPF programs within a given
ELF section. Libbpf enforced that by keeping track of the next section offset
that should start a new BPF (sub)program and cross-checks that by searching
for a corresponding STT_FUNC ELF symbol.
But this is too restrictive once we allow to have weak BPF programs and link
together two or more BPF object files. In such case, some weak BPF programs
might be "overridden" by either non-weak BPF program with the same name and
signature, or even by another weak BPF program that just happened to be linked
first. That, in turn, leaves BPF instructions of the "lost" BPF (sub)program
intact, but there is no corresponding ELF symbol, because no one is going to
be referencing it.
Libbpf already correctly handles such cases in the sense that it won't append
such dead code to actual BPF programs loaded into kernel. So the only change
that needs to be done is to relax the logic of parsing BPF instruction
sections. Instead of assuming next BPF (sub)program section offset, iterate
available STT_FUNC ELF symbols to discover all available BPF subprograms and
programs.
libbpf: Mark BPF subprogs with hidden visibility as static for BPF verifier
Define __hidden helper macro in bpf_helpers.h, which is a short-hand for
__attribute__((visibility("hidden"))). Add libbpf support to mark BPF
subprograms marked with __hidden as static in BTF information to enforce BPF
verifier's static function validation algorithm, which takes more information
(caller's context) into account during a subprogram validation.
libbpf: Suppress compiler warning when using SEC() macro with externs
When used on externs SEC() macro will trigger compilation warning about
inapplicable `__attribute__((used))`. That's expected for extern declarations,
so suppress it with the corresponding _Pragma.
Dump succinct information for each member of DATASEC: its kinds and name. This
is extremely helpful to see at a quick glance what is inside each DATASEC of
a given BTF. Without this, one has to jump around BTF data to just find out
the name of a VAR or FUNC. DATASEC's var_secinfo member is special in that
regard because it doesn't itself contain the name of the member, delegating
that to the referenced VAR and FUNC kinds. Other kinds, like
STRUCT/UNION/FUNC/ENUM, encode member names directly and thus are clearly
identifiable in BTF dump.
David S. Miller [Fri, 23 Apr 2021 21:04:43 +0000 (14:04 -0700)]
Merge branch '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue
Tony Nguyen says:
====================
40GbE Intel Wired LAN Driver Updates 2021-04-23
This series contains updates to i40e and iavf drivers.
Aleksandr adds support for VIRTCHNL_VF_CAP_ADV_LINK_SPEED in i40e which
allows for reporting link speed to VF as a value instead of using an
enum; helper functions are created to remove repeated code.
Coiby Xu reduces memory use of i40e when using kdump by reducing Tx, Rx,
and admin queue to minimum values. Current use causes failure of kdump.
Stefan Assmann removes duplicated free calls in iavf.
Haiyue cleans up a loop to return directly when if the value is found
and changes some magic numbers to defines for better maintainability
in iavf.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 23 Apr 2021 21:01:28 +0000 (14:01 -0700)]
Merge branch 'mlxsw-selftest-fixes'
Petr Machata says:
====================
selftests: mlxsw: Fixes
This patch set carries fixes to selftest issues that we have hit in our
nightly regression run. Almost all are in mlxsw selftests, though one is in
a generic forwarding selftest.
- In patch #1, in an ERSPAN test, install an FDB entry as static instead of
(implicitly) as local.
- In the mlxsw resource-scale test, an if statement overrides the value of
$?, which is supposed to contain the result of the test. As a result, the
resource scale test can spuriously pass.
In patches #2 and #3, remove the if statements to fix the issue in,
respectively, port_scale test and tc_flower_scale tests.
- Again in the mlxsw resource-scale test, when more then one sub-test is
run, a successful sub-test overrides any previous failures. This causes a
spurious pass of the overall test. This is fixed in patch #4.
- In patch #5, increase a tolerance in a mlxsw-specific RED backlog test.
This test is very noisy, due to rounding errors and the unpredictability
of software traffic generation. By bumping the tolerance from 5 % to 10,
get the failure rate to zero. This shouldn't impact the accuracy,
mistakes in backlog configuration (e.g. due to wrong cell size) are
likely to cause a much larger discrepancy.
- In patch #6, fix mausezahn invocation in the mlxsw ERSPAN scale
test. The test failed because of the wrong invocation.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Petr Machata [Fri, 23 Apr 2021 12:19:48 +0000 (14:19 +0200)]
selftests: mlxsw: Fix mausezahn invocation in ERSPAN scale test
The mirror_gre_scale test creates as many ERSPAN sessions as the underlying
chip supports, and tests that they all work. In order to determine that it
issues a stream of ICMP packets and checks if they are mirrored as
expected.
However, the mausezahn invocation missed the -6 flag to identify the use of
IPv6 protocol, and was sending ICMP messages over IPv6, as opposed to
ICMP6. It also didn't pass an explicit source IP address, which apparently
worked at some point in the past, but does not anymore.
To fix these issues, extend the function mirror_test() in mirror_lib by
detecting the IPv6 protocol addresses, and using a different ICMP scheme.
Fix __mirror_gre_test() in the selftest itself to pass a source IP address.
Signed-off-by: Petr Machata <petrm@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Petr Machata [Fri, 23 Apr 2021 12:19:47 +0000 (14:19 +0200)]
selftests: mlxsw: Increase the tolerance of backlog buildup
The intention behind this test is to make sure that qdisc limit is
correctly projected to the HW. However, first, due to rounding in the
qdisc, and then in the driver, the number cannot actually be accurate. And
second, the approach to testing this is to oversubscribe the port with
traffic generated on the same switch. The actual backlog size therefore
fluctuates.
In practice, this test proved to be noisier than the rest, and spuriously
fails every now and then. Increase the tolerance to 10 % to avoid these
issues.
Signed-off-by: Petr Machata <petrm@nvidia.com> Acked-by: Jiri Pirko <jiri@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
selftests: mlxsw: Return correct error code in resource scale tests
Currently, the resource scale test checks a few cases, when the error code
resets between the cases. So for example, if one case fails and the
consecutive case passes, the error code eventually will fit the last test
and will be 0.
Save a new return code that will hold the 'or' return codes of all the
cases, so the final return code will consider all the cases.
Signed-off-by: Danielle Ratson <danieller@nvidia.com> Reviewed-by: Petr Machata <petrm@nvidia.com> Signed-off-by: Petr Machata <petrm@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
selftests: mlxsw: Remove a redundant if statement in tc_flower_scale test
Currently, the error return code of the failure condition is lost after
using an if statement, so the test doesn't fail when it should.
Remove the if statement that separates the condition and the error code
check, so the test won't always pass.
Fixes: abfce9e062021 ("selftests: mlxsw: Reduce running time using offload indication") Reported-by: Ido Schimmel <idosch@nvidia.com> Signed-off-by: Danielle Ratson <danieller@nvidia.com> Reviewed-by: Petr Machata <petrm@nvidia.com> Signed-off-by: Petr Machata <petrm@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
selftests: mlxsw: Remove a redundant if statement in port_scale test
Currently, the error return code of the failure condition is lost after
using an if statement, so the test doesn't fail when it should.
Remove the if statement that separates the condition and the error code
check, so the test won't always pass.
Fixes: 5154b1b826d9b ("selftests: mlxsw: Add a scale test for physical ports") Reported-by: Ido Schimmel <idosch@nvidia.com> Signed-off-by: Danielle Ratson <danieller@nvidia.com> Reviewed-by: Petr Machata <petrm@nvidia.com> Signed-off-by: Petr Machata <petrm@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Petr Machata [Fri, 23 Apr 2021 12:19:43 +0000 (14:19 +0200)]
selftests: net: mirror_gre_vlan_bridge_1q: Make an FDB entry static
The FDB roaming test installs a destination MAC address on the wrong
interface of an FDB database and tests whether the mirroring fails, because
packets are sent to the wrong port. The test by mistake installs the FDB
entry as local. This worked previously, because drivers were notified of
local FDB entries in the same way as of static entries. However that has
been fixed in the commit 6ab4c3117aec ("net: bridge: don't notify switchdev
for local FDB addresses"), and local entries are not notified anymore. As a
result, the HW is not reconfigured for the FDB roam, and mirroring keeps
working, failing the test.
To fix the issue, mark the FDB entry as static.
Fixes: 9c7c8a82442c ("selftests: forwarding: mirror_gre_vlan_bridge_1q: Add more tests") Signed-off-by: Petr Machata <petrm@nvidia.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 23 Apr 2021 20:58:25 +0000 (13:58 -0700)]
Merge tag 'wireless-drivers-next-2021-04-23' of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers-next
Kalle Valo says:
====================
wireless-drivers-next patches for v5.13
Third, and final, set of patches for v5.13. We got one more week
before the merge window and this includes from that extra week.
Smaller features to rtw88 and mt76, but mostly this contains fixes.
rtw88
* 8822c: Add gap-k calibration to improve long range performance
mt76
* parse rate power limits from DT
* debugfs file to test firmware crash
* debugfs to disable NAPI threaded mode
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Hayes Wang [Fri, 23 Apr 2021 09:44:54 +0000 (17:44 +0800)]
r8152: remove NCM mode from REALTEK_USB_DEVICE macro
The RTL8156 support CDC NCM mode. And users could set the configuration
of the USB device between vendor and NCM mode dynamically by themselves.
That is, the driver doesn't need to set vendor mode from NCM mode.
Fixes: 195aae321c82 ("r8152: support new chips") Signed-off-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yangbo Lu [Fri, 23 Apr 2021 09:33:55 +0000 (17:33 +0800)]
enetc: fix locking for one-step timestamping packet transfer
The previous patch to support PTP Sync packet one-step timestamping
described one-step timestamping packet handling logic as below in
commit message:
- Trasmit packet immediately if no other one in transfer, or queue to
skb queue if there is already one in transfer.
The test_and_set_bit_lock() is used here to lock and check state.
- Start a work when complete transfer on hardware, to release the bit
lock and to send one skb in skb queue if has.
There was not problem of the description, but there was a mistake in
implementation. The locking/test_and_set_bit_lock() should be put in
enetc_start_xmit() which may be called by worker, rather than in
enetc_xmit(). Otherwise, the worker calling enetc_start_xmit() after
bit lock released is not able to lock again for transfer.
Fixes: 7294380c5211 ("enetc: support PTP Sync packet one-step timestamping") Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com> Reviewed-by: Claudiu Manoil <claudiu.manoil@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 23 Apr 2021 20:31:58 +0000 (13:31 -0700)]
Merge branch 'mk_eth_soc_fixes-perf-improvements'
Ilya Lipnitskiy says:
====================
mtk_eth_soc: fixes and performance improvements
Most of these changes come from OpenWrt where they have been present and
tested for months.
First three patches are bug fixes. The rest are performance
improvements. The last patch is a cleanup to use the iopoll.h macro for
busy-waiting instead of a custom loop.
v2:
- Reverse christmas tree in "use iopoll.h macro for DMA init"
- Use cond_resched() instead of iopoll.h macro in "reduce MDIO bus
access latency"
- Use napi_complete_done and rework NAPI callbacks in a new patch
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
net: ethernet: mtk_eth_soc: use iopoll.h macro for DMA init
Replace a tight busy-wait loop without a pause with a standard
readx_poll_timeout_atomic routine with a 5 us poll period.
Tested by booting a MT7621 device to ensure the driver initializes
properly.
Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Felix Fietkau [Fri, 23 Apr 2021 05:21:07 +0000 (22:21 -0700)]
net: ethernet: mtk_eth_soc: set PPE flow hash as skb hash if present
This improves GRO performance
Signed-off-by: Felix Fietkau <nbd@nbd.name>
[Ilya: Use MTK_RXD4_FOE_ENTRY instead of GENMASK(13, 0)] Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Use napi_complete_done to communicate total TX and RX work done to NAPI.
Count total RX work up instead of remaining work down for clarity.
Remove unneeded local variables for clarity. Use do {} while instead of
goto for clarity.
Suggested-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Avoid rearming interrupt if napi_complete returns false
Signed-off-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Felix Fietkau [Fri, 23 Apr 2021 05:21:04 +0000 (22:21 -0700)]
net: ethernet: mtk_eth_soc: only read the full RX descriptor if DMA is done
Uncached memory access is expensive, and there is no need to access all
descriptor words if we can't process them anyway
Signed-off-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Felix Fietkau [Fri, 23 Apr 2021 05:21:03 +0000 (22:21 -0700)]
net: ethernet: mtk_eth_soc: cache HW pointer of last freed TX descriptor
The value is only updated by the CPU, so it is cheaper to access from the
ring data structure than from a hardware register.
Signed-off-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
[Ilya: add documentation for new struct fields] Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Felix Fietkau [Fri, 23 Apr 2021 05:21:01 +0000 (22:21 -0700)]
net: ethernet: mtk_eth_soc: increase DMA ring sizes
256 descriptors is not enough for multi-gigabit traffic under load on
MT7622. Bump it to 512 to improve performance.
Signed-off-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Felix Fietkau [Fri, 23 Apr 2021 05:21:00 +0000 (22:21 -0700)]
net: ethernet: mtk_eth_soc: use larger burst size for QDMA TX
Improves tx performance
Signed-off-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
When running short on descriptors, only stop the queue for the netdev that
tx was attempted for. By the time something tries to send on the other
netdev, the ring might have some more room already.
Signed-off-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Felix Fietkau [Fri, 23 Apr 2021 05:20:58 +0000 (22:20 -0700)]
net: ethernet: mtk_eth_soc: reduce MDIO bus access latency
usleep_range often ends up sleeping much longer than the 10-20us provided
as a range here. This causes significant latency in mdio bus acceses,
which easily adds multiple seconds to the boot time on MT7621 when polling
DSA slave ports.
Use cond_resched instead of usleep_range, since the MDIO access does not
take much time
Signed-off-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Felix Fietkau [Fri, 23 Apr 2021 05:20:57 +0000 (22:20 -0700)]
net: ethernet: mtk_eth_soc: use napi_consume_skb
Should improve performance, since it can use bulk free
Signed-off-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
In case build_skb fails, call skb_free_frag on the correct pointer. Also
update the DMA structures with the new mapping before exiting, because
the mapping was successful
Suggested-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Felix Fietkau [Fri, 23 Apr 2021 05:20:55 +0000 (22:20 -0700)]
net: ethernet: mtk_eth_soc: unmap RX data before calling build_skb
Since build_skb accesses the data area (for initializing shinfo), dma unmap
needs to happen before that call
Signed-off-by: Felix Fietkau <nbd@nbd.name>
[Ilya: split build_skb cleanup fix into a separate commit] Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Felix Fietkau [Fri, 23 Apr 2021 05:20:54 +0000 (22:20 -0700)]
net: ethernet: mtk_eth_soc: fix RX VLAN offload
The VLAN ID in the rx descriptor is only valid if the RX_DMA_VTAG bit is
set. Fixes frames wrongly marked with VLAN tags.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
[Ilya: fix commit message] Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Dexuan Cui [Thu, 22 Apr 2021 20:08:16 +0000 (13:08 -0700)]
net: mana: Use int to check the return value of mana_gd_poll_cq()
mana_gd_poll_cq() may return -1 if an overflow error is detected (this
should never happen unless there is a bug in the driver or the hardware).
Fix the type of the variable "comp_read" by using int rather than u32.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Fixes: ca9c54d2d6a5 ("net: mana: Add a driver for Microsoft Azure Network Adapter (MANA)") Signed-off-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes: fb32856b16ad ("virtio-net: page_to_skb() use build_skb when there's sufficient tailroom") Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com> Reported-by: Ido Schimmel <idosch@nvidia.com> Tested-by: Ido Schimmel <idosch@nvidia.com> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: sock: remove the unnecessary check in proto_register
tw_prot_cleanup will check the twsk_prot.
Fixes: 0f5907af3913 ("net: Fix potential memory leak in proto_register()") Cc: Miaohe Lin <linmiaohe@huawei.com> Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
bpf: Remove unnecessary map checks for ARG_PTR_TO_CONST_STR
reg->type is enforced by check_reg_type() and map should never be NULL
(it would already have been dereferenced anyway) so these checks are
unnecessary.
bpf: Notify user if we ever hit a bpf_snprintf verifier bug
In check_bpf_snprintf_call(), a map_direct_value_addr() of the fmt map
should never fail because it has already been checked by
ARG_PTR_TO_CONST_STR. But if it ever fails, it's better to error out
with an explicit debug message rather than silently fail.
Haiyue Wang [Wed, 31 Mar 2021 02:08:37 +0000 (10:08 +0800)]
iavf: redefine the magic number for FDIR GTP-U header fields
The flex-byte for GTP-U protocol header fields uses the magic number,
which is hard to maintain and understand, define the interested fields
with meaningful macro name, based on the GTP-U protocol stack:
Coiby Xu [Thu, 4 Mar 2021 02:55:43 +0000 (10:55 +0800)]
i40e: use minimal admin queue for kdump
The minimum size of admin send/receive queue is 1 and 2 respectively.
The admin send queue can't be set to 1 because in that case, the
firmware would fail to init.
Signed-off-by: Coiby Xu <coxu@redhat.com> Tested-by: Dave Switzer <david.switzer@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Coiby Xu [Thu, 4 Mar 2021 02:55:41 +0000 (10:55 +0800)]
i40e: use minimal Tx and Rx pairs for kdump
Set the number of the MSI-X vectors to 1. When MSI-X is enabled,
it's not allowed to use more TC queue pairs than MSI-X vectors
(pf->num_lan_msix) exist. Thus the number of Tx and Rx pairs
(vsi->num_queue_pairs) will be equal to the number of MSI-X vectors,
i.e., 1.
Signed-off-by: Coiby Xu <coxu@redhat.com> Tested-by: Dave Switzer <david.switzer@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Refactor repeated link state reporting code into a separate helper
functions: i40e_set_vf_link_state() i40e_vc_link_speed2mbps().
Add support of VIRTCHNL_VF_CAP_ADV_LINK_SPEED;
Signed-off-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com> Signed-off-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com> Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
This patch series to add setting for HW descriptor prefetch for DWMAC
version 5.20 onwards. For Intel platform, enable the capability by
default.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
stmmac: intel: Enable HW descriptor prefetch by default
Enable HW descriptor prefetch by default by setting plat->dma_cfg->dche =
true in intel_mgbe_common_data(). Need to be noted that this capability
only be supported in DWMAC core version 5.20 onwards. In stmmac, there is
a checking to check the core version. If the core version is below 5.20,
this capability wouldn`t be configured.
Below is the iperf result comparison between HW descriptor prefetch
disabled(DCHE=0b) and enabled(DCHE=1b). Tested on Intel Elkhartlake
platform with DWMAC Core 5.20. Observed line rate performance
improvement with HW descriptor prefetch enabled.
DWMAC Core 5.20 onwards supports HW descriptor prefetching.
Additionally, it also depends on platform specific RTL configuration.
This capability could be enabled by setting DMA_Mode bit-19 (DCHE).
So, to enable this cability, platform must set plat->dma_cfg->dche = true
and the DWMAC core version must be 5.20 onwards. Else, this capability
wouldn`t be configured
Signed-off-by: Mohammad Athari Bin Ismail <mohammad.athari.ismail@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net/mlx4: Treat VFs fair when handling comm_channel_events
Handling comm_channel_event in mlx4_master_comm_channel uses a double
loop to determine which slaves have requested work. The search is
always started at lowest slave. This leads to unfairness; lower VFs
tends to be prioritized over higher VFs.
The patch uses find_next_bit to determine which slaves to handle.
Fairness is implemented by always starting at the next to the last
start.
An MPI program has been used to measure improvements. It runs 500
ibv_reg_mr, synchronizes with all other instances and then runs 500
ibv_dereg_mr.
The results running 500 processes, time reported is for running 500
calls:
Suggested-by: Håkon Bugge <haakon.bugge@oracle.com> Signed-off-by: Hans Westgaard Ry <hans.westgaard.ry@oracle.com> Reviewed-by: Tariq Toukan <tariqt@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Tiezhu Yang [Thu, 22 Apr 2021 03:36:00 +0000 (11:36 +0800)]
bpf, doc: Fix some invalid links in bpf_devel_QA.rst
There exist some errors "404 Not Found" when I click the link
of "MAINTAINERS" [1], "samples/bpf/" [2] and "selftests" [3]
in the documentation "HOWTO interact with BPF subsystem" [4].
As Alexei Starovoitov suggested, just remove "MAINTAINERS" and
"samples/bpf/" links and use correct link of "selftests".
Martin Willi [Mon, 19 Apr 2021 14:15:59 +0000 (16:15 +0200)]
net, xdp: Update pkt_type if generic XDP changes unicast MAC
If a generic XDP program changes the destination MAC address from/to
multicast/broadcast, the skb->pkt_type is updated to properly handle
the packet when passed up the stack. When changing the MAC from/to
the NICs MAC, PACKET_HOST/OTHERHOST is not updated, though, making
the behavior different from that of native XDP.
Remember the PACKET_HOST/OTHERHOST state before calling the program
in generic XDP, and update pkt_type accordingly if the destination
MAC address has changed. As eth_type_trans() assumes a default
pkt_type of PACKET_HOST, restore that before calling it.
The use case for this is when a XDP program wants to push received
packets up the stack by rewriting the MAC to the NICs MAC, for
example by cluster nodes sharing MAC addresses.
Fixes: 297249569932 ("net: fix generic XDP to handle if eth header was mangled") Signed-off-by: Martin Willi <martin@strongswan.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210419141559.8611-1-martin@strongswan.org
projects / mirror_ubuntu-kernels.git / log