Stanislav Fomichev says:

====================
C flow dissector supports input flags that tell it to customize parsing
by either stopping early or trying to parse as deep as possible.
BPF flow dissector always parses as deep as possible which is sub-optimal.
Pass input flags to the BPF flow dissector as well so it can make the same
decisions.

Series outline:
* remove unused FLOW_DISSECTOR_F_STOP_AT_L3 flag
* export FLOW_DISSECTOR_F_XXX flags as uapi and pass them to BPF
  flow dissector
* add documentation for the export flags
* support input flags in BPF_PROG_TEST_RUN via ctx_{in,out}
* sync uapi to tools
* support FLOW_DISSECTOR_F_PARSE_1ST_FRAG in selftest
* support FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL in kernel and selftest
* support FLOW_DISSECTOR_F_STOP_AT_ENCAP in selftest

Pros:
* makes BPF flow dissector faster by avoiding burning extra cycles
* existing BPF progs continue to work by ignoring the flags and always
  parsing as deep as possible

Cons:
* new UAPI which we need to support (OTOH, if we need to deprecate some
  flags, we can just stop setting them upon calling BPF programs)

Some numbers (with .repeat = 4000000 in test_flow_dissector):
        test_flow_dissector:PASS:ipv4-frag 35 nsec
        test_flow_dissector:PASS:ipv4-frag 35 nsec
        test_flow_dissector:PASS:ipv4-no-frag 32 nsec
        test_flow_dissector:PASS:ipv4-no-frag 32 nsec

        test_flow_dissector:PASS:ipv6-frag 39 nsec
        test_flow_dissector:PASS:ipv6-frag 39 nsec
        test_flow_dissector:PASS:ipv6-no-frag 36 nsec
        test_flow_dissector:PASS:ipv6-no-frag 36 nsec

        test_flow_dissector:PASS:ipv6-flow-label 36 nsec
        test_flow_dissector:PASS:ipv6-flow-label 36 nsec
        test_flow_dissector:PASS:ipv6-no-flow-label 33 nsec
        test_flow_dissector:PASS:ipv6-no-flow-label 33 nsec

        test_flow_dissector:PASS:ipip-encap 38 nsec
        test_flow_dissector:PASS:ipip-encap 38 nsec
        test_flow_dissector:PASS:ipip-no-encap 32 nsec
        test_flow_dissector:PASS:ipip-no-encap 32 nsec

The improvement is around 10%, but it's in a tight cache-hot
BPF_PROG_TEST_RUN loop.
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Exit as soon as we found that packet is encapped when
BPF_FLOW_DISSECTOR_F_STOP_AT_ENCAP is passed.
Add appropriate selftest cases.

v2:
* Subtract sizeof(struct iphdr) from .iph_inner.tot_len (Willem de Bruijn)
Acked-by: Petar Penkov <ppenkov@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Willem de Bruijn <willemb@google.com>
Cc: Petar Penkov <ppenkov@google.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Add support for exporting ipv6 flow label via bpf_flow_keys.
Export flow label from bpf_flow.c and also return early when
BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL is passed.
Acked-by: Petar Penkov <ppenkov@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Willem de Bruijn <willemb@google.com>
Cc: Petar Penkov <ppenkov@google.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

bpf_flow.c: exit early unless BPF_FLOW_DISSECTOR_F_PARSE_1ST_FRAG is
passed in flags. Also, set ip_proto earlier, this makes sure we have
correct value with fragmented packets.

Add selftest cases to test ipv4/ipv6 fragments and skip eth_get_headlen
tests that don't have BPF_FLOW_DISSECTOR_F_PARSE_1ST_FRAG flag.

eth_get_headlen calls flow dissector with
BPF_FLOW_DISSECTOR_F_PARSE_1ST_FRAG flag so we can't run tests that
have different set of input flags against it.

v2:
 * sefltests -> selftests (Willem de Bruijn)
 * Reword a comment about eth_get_headlen flags (Song Liu)
Acked-by: Petar Penkov <ppenkov@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Willem de Bruijn <willemb@google.com>
Cc: Petar Penkov <ppenkov@google.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Export bpf_flow_keys flags to tools/libbpf/selftests.
Acked-by: Petar Penkov <ppenkov@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Willem de Bruijn <willemb@google.com>
Cc: Petar Penkov <ppenkov@google.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

This will allow us to write tests for those flags.

v2:
* Swap kfree(data) and kfree(user_ctx) (Song Liu)
Acked-by: Petar Penkov <ppenkov@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Willem de Bruijn <willemb@google.com>
Cc: Petar Penkov <ppenkov@google.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Describe what each input flag does and who uses it.
Acked-by: Petar Penkov <ppenkov@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Willem de Bruijn <willemb@google.com>
Cc: Petar Penkov <ppenkov@google.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

C flow dissector supports input flags that tell it to customize parsing
by either stopping early or trying to parse as deep as possible. Pass
those flags to the BPF flow dissector so it can make the same
decisions. In the next commits I'll add support for those flags to
our reference bpf_flow.c

v3:
* Export copy of flow dissector flags instead of moving (Alexei Starovoitov)
Acked-by: Petar Penkov <ppenkov@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Willem de Bruijn <willemb@google.com>
Cc: Petar Penkov <ppenkov@google.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Software event output is only enabled by a few prog types.
This test is to ensure that all supported types are enabled for
bpf_perf_event_output successfully.
Signed-off-by: Allan Zhang <allanzhang@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Software event output is only enabled by a few prog types right now (TC,
LWT out, XDP, sockops). Many other skb based prog types need
bpf_skb_event_output to produce software event.

Added socket_filter, cg_skb, sk_skb prog types to generate sw event.

Test bpf code is generated from code snippet:

struct TMP {
    uint64_t tmp;
} tt;
tt.tmp = 5;
bpf_perf_event_output(skb, &connection_tracking_event_map, 0,
                      &tt, sizeof(tt));
return 1;

the bpf assembly from llvm is:
       0:       b7 02 00 00 05 00 00 00         r2 = 5
       1:       7b 2a f8 ff 00 00 00 00         *(u64 *)(r10 - 8) = r2
       2:       bf a4 00 00 00 00 00 00         r4 = r10
       3:       07 04 00 00 f8 ff ff ff         r4 += -8
       4:       18 02 00 00 00 00 00 00 00 00 00 00 00 00 00 00    r2 = 0ll
       6:       b7 03 00 00 00 00 00 00         r3 = 0
       7:       b7 05 00 00 08 00 00 00         r5 = 8
       8:       85 00 00 00 19 00 00 00         call 25
       9:       b7 00 00 00 01 00 00 00         r0 = 1
      10:       95 00 00 00 00 00 00 00         exit
Signed-off-by: Allan Zhang <allanzhang@google.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Andrii Nakryiko says:

====================
There were few more tests and samples that were using custom perf buffer setup
code from trace_helpers.h. This patch set gets rid of all the usages of those
and removes helpers themselves. Libbpf provides nicer, but equally powerful
set of APIs to work with perf ring buffers, so let's have all the samples use

v1->v2:
- make logging message one long line instead of two (Song).
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

libbpf's perf_buffer API supersedes trace_helper.h's helpers.
Remove those helpers after all existing users were already moved to
perf_buffer API.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Convert trace_output sample to libbpf's perf_buffer API.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Convert xdp_sample_pkts_user to libbpf's perf_buffer API.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Switch test_tcpnotify test to use libbpf's perf_buffer API instead of
re-implementing portion of it.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Convert test_get_stack_raw_tp test to new perf_buffer API.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

When BPF program defines uninitialized global variable, it's put into
a special COMMON section. Libbpf will reject such programs, but will
provide very unhelpful message with garbage-looking section index.

This patch detects special section cases and gives more explicit error
message.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Signed-off-by: Roman Mashak <mrv@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The rest of Hyper-V code is using new types for UUID handling.
Convert hv_sock as well.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

John Hurley says:

====================
nfp: Offload MPLS actions

The module act_mpls has recently been added to the kernel. This allows the
manipulation of MPLS headers on packets including push, pop and modify.
Add these new actions and parameters to the intermediate representation
API for hardware offload. Follow this by implementing the offload of these
MPLS actions in the NFP driver.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Recent additions to the kernel include a TC action module to manipulate
MPLS headers on packets. Such actions are available to offload via the
flow_offload intermediate representation API.

Modify the NFP driver to allow the offload of MPLS set actions to
firmware. Set actions update the outermost MPLS header. The offload
includes a mask to specify which fields should be set.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Recent additions to the kernel include a TC action module to manipulate
MPLS headers on packets. Such actions are available to offload via the
flow_offload intermediate representation API.

Modify the NFP driver to allow the offload of MPLS pop actions to
firmware. The act_mpls TC module enforces that the next protocol is
supplied along with the pop action. Passing this to firmware allows it
to properly rebuild the underlying packet after the pop.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Recent additions to the kernel include a TC action module to manipulate
MPLS headers on packets. Such actions are available to offload via the
flow_offload intermediate representation API.

Modify the NFP driver to allow the offload of MPLS push actions to
firmware.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

A recent addition to TC actions is the ability to manipulate the MPLS
headers on packets.

In preparation to offload such actions to hardware, update the IR code to
accept and prepare the new actions.

Note that no driver currently impliments the MPLS dec_ttl action so this
is not included.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The structure is too large to put on the stack, resulting in a
warning on 32-bit ARM:

drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c:59:5: error: stack frame size of 1344 bytes in function
      'mlx5e_open_xsk' [-Werror,-Wframe-larger-than=]

Use kvzalloc() instead.

Fixes: a038e9794541 ("net/mlx5e: Add XSK zero-copy support")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of using to_pci_dev + pci_get_drvdata,
use dev_get_drvdata to make code simpler.
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of using to_pci_dev + pci_get_drvdata,
use dev_get_drvdata to make code simpler.
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of using to_pci_dev + pci_get_drvdata,
use dev_get_drvdata to make code simpler.
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of using to_pci_dev + pci_get_drvdata,
use dev_get_drvdata to make code simpler.
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of using to_pci_dev + pci_get_drvdata,
use dev_get_drvdata to make code simpler.
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of using to_pci_dev + pci_get_drvdata,
use dev_get_drvdata to make code simpler.
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of using to_pci_dev + pci_get_drvdata,
use dev_get_drvdata to make code simpler.
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of using to_pci_dev + pci_get_drvdata,
use dev_get_drvdata to make code simpler.
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of using to_pci_dev + pci_get_drvdata,
use dev_get_drvdata to make code simpler.
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/faraday/ftgmac100.c:777:13: error: 'skb_frag_t {aka struct bio_vec}' has no member named 'size'

Fallout from the skb_frag_t conversion to bio_vec, simply
use skb_frag_size().

Fixes: b8b576a1 ("net: Rename skb_frag_t size to bv_len")
Reported-by: René van Dorst <opensource@vdorst.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The hardware has been declared EOL by the vendor more than 5 years ago.
What's more relevant to the Linux kernel is that the quality of this driver
is not on par with many other mainline drivers.

Cc: Manish Chopra <manishc@marvell.com>
Message-id: <20190617074858.32467-1-bpoirier@suse.com>
Signed-off-by: Benjamin Poirier <bpoirier@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Matthew Wilcox says:

====================
Convert skb_frag_t to bio_vec

The skb_frag_t and bio_vec are fundamentally the same (page, offset,
length) tuple.  This patch series unifies the two, leaving the
skb_frag_t typedef in place.  This has the immediate advantage that
we already have iov_iter support for bvecs and don't need to add
support for iterating skbuffs.  It enables a long-term plan to use
bvecs more broadly within the kernel and should make network-storage
drivers able to do less work converting between skbuffs and biovecs.

It will consume more memory on 32-bit kernels.  If that proves
problematic, we can look at ways of addressing it.

v3: Rebase on latest Linus with net-next merged.
  - Reorder the uncontroversial 'Use skb accessors' patches first so you
    can apply just those two if you want to hold off on the full
    conversion.
  - Convert all the users of 'struct skb_frag_struct' to skb_frag_t.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

There are a lot of users of frag->page_offset, so use a union
to avoid converting those users today.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Improved compatibility with bvec
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

One step closer to turning the skb_frag_t into a bio_vec.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>