It's follow-up for discussion [1]

CHECK and CHECK_FAIL macros in test_progs.h can affect errno in some
circumstances, e.g. if some code accidentally closes stdout. It makes
checking errno in patterns like this unreliable:

	if (CHECK(!bpf_prog_attach_xattr(...), "tag", "msg"))
		goto err;
	CHECK_FAIL(errno != ENOENT);

, since by CHECK_FAIL time errno could be affected not only by
bpf_prog_attach_xattr but by CHECK as well.

Fix it by saving and restoring errno in the macros. There is no "Fixes"
tag since no problems were discovered yet and it's rather precaution.

test_progs was run with this change and no difference was identified.

[1] https://lore.kernel.org/bpf/20191219210907.GD16266@rdna-mbp.dhcp.thefacebook.com/Signed-off-by: NAndrey Ignatov <rdna@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NAndrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20191220000511.1684853-1-rdna@fb.com

Magnus Karlsson says:

====================
This patch set cleans up the ring access functions of AF_XDP in hope
that it will now be easier to understand and maintain. I used to get a
headache every time I looked at this code in order to really understand it,
but now I do think it is a lot less painful.

The code has been simplified a lot and as a bonus we get better
performance in nearly all cases. On my new 2.1 GHz Cascade Lake
machine with a standard default config plus AF_XDP support and
CONFIG_PREEMPT on I get the following results in percent performance
increases with this patch set compared to without it:

Zero-copy (-N):
          rxdrop        txpush        l2fwd
1 core:    -2%            0%            3%
2 cores:    4%            0%            3%

Zero-copy with poll() (-N -p):
          rxdrop        txpush        l2fwd
1 core:     3%            0%            1%
2 cores:   21%            0%            9%

Skb mode (-S):
Shows a 0% to 5% performance improvement over the same benchmarks as
above.

Here 1 core means that we are running the driver processing and the
application on the same core, while 2 cores means that they execute on
separate cores. The applications are from the xdpsock sample app.

On my older 2.0 Ghz Broadwell machine that I used for the v1, I get
the following results:

Zero-copy (-N):
          rxdrop        txpush        l2fwd
1 core:     4%            5%            4%
2 cores:    1%            0%            2%

Zero-copy with poll() (-N -p):
          rxdrop        txpush        l2fwd
1 core:     1%            3%            3%
2 cores:   22%            0%            5%

Skb mode (-S):
Shows a 0% to 1% performance improvement over the same benchmarks as
above.

When a results says 21 or 22% better, as in the case of poll mode with
2 cores and rxdrop, my first reaction is that it must be a
bug. Everything else shows between 0% and 5% performance
improvement. What is giving rise to 22%? A quick bisect indicates that
it is patches 2, 3, 4, 5, and 6 that are giving rise to most of this
improvement. So not one patch in particular, but something around 4%
improvement from each one of them. Note that exactly this benchmark
has previously had an extraordinary slow down compared to when running
without poll syscalls. For all the other poll tests above, the
slowdown has always been around 4% for using poll syscalls. But with
the bad performing test in question, it was above 25%. Interestingly,
after this clean up, the slow down is 4%, just like all the other poll
tests. Please take an extra peek at this so I have not messed up
something.

The 0% for several txpush results are due to the test bottlenecking on
a non-CPU HW resource. If I eliminated that bottleneck on my system, I
would expect to see an increase there too.

Changes v1 -> v2:
* Corrected textual errors in the commit logs (Sergei and Martin)
* Fixed the functions that detect empty and full rings so that they
  now operate on the global ring state (Maxim)

This patch has been applied against commit a352a824 ("Merge branch 'libbpf-extern-followups'")

Structure of the patch set:

Patch 1: Eliminate the lazy update threshold used when preallocating
         entries in the completion ring
Patch 2: Simplify the detection of empty and full rings
Patch 3: Consolidate the two local producer pointers into one
Patch 4: Standardize the naming of the producer ring access functions
Patch 5: Eliminate the Rx batch size used for the fill ring
Patch 6: Simplify the functions xskq_nb_avail and xskq_nb_free
Patch 7: Simplify and standardize the naming of the consumer ring
         access functions
Patch 8: Change the names of the validation functions to improve
         readability and also the return value of these functions
Patch 9: Change the name of xsk_umem_discard_addr() to
         xsk_umem_release_addr() to better reflect the new
         names. Requires a name change in the drivers that support AF_XDP
         zero-copy.
Patch 10: Remove unnecessary READ_ONCE of data in the ring
Patch 11: Add overall function naming comment and reorder the functions
          for easier reference
Patch 12: Use the struct_size helper function when allocating rings
====================
Reviewed-by: NBjörn Töpel <bjorn.topel@intel.com>
Tested-by: NBjörn Töpel <bjorn.topel@intel.com>
Acked-by: NBjörn Töpel <bjorn.topel@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>

Improve readability and maintainability by using the struct_size()
helper when allocating the AF_XDP rings.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-13-git-send-email-magnus.karlsson@intel.com

Add comments on how the ring access functions are named and how they
are supposed to be used for producers and consumers. The functions are
also reordered so that the consumer functions are in the beginning and
the producer functions in the end, for easier reference. Put this in a
separate patch as the diff might look a little odd, but no
functionality has changed in this patch.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-12-git-send-email-magnus.karlsson@intel.com

There are two unnecessary READ_ONCE of descriptor data. These are not
needed since the data is written by the producer before it signals
that the data is available by incrementing the producer pointer. As the
access to this producer pointer is serialized and the consumer always
reads the descriptor after it has read and synchronized with the
producer counter, the write of the descriptor will have fully
completed and it does not matter if the consumer has any read tearing.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-11-git-send-email-magnus.karlsson@intel.com

Change the name of xsk_umem_discard_addr to xsk_umem_release_addr to
better reflect the new naming of the AF_XDP queue manipulation
functions. As this functions is used by drivers implementing support
for AF_XDP zero-copy, it requires a name change to these drivers. The
function xsk_umem_release_addr_rq has also changed name in the same
fashion.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-10-git-send-email-magnus.karlsson@intel.com

Change the names of the validation functions to better reflect what
they are doing. The uppermost ones are reading entries from the rings
and only the bottom ones validate entries. So xskq_cons_read_ is a
better prefix name.

Also change the xskq_cons_read_ functions to return a bool
as the the descriptor or address is already returned by reference
in the parameters. Everyone is using the return value as a bool
anyway.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-9-git-send-email-magnus.karlsson@intel.com

Simplify and refactor consumer ring functions. The consumer first
"peeks" to find descriptors or addresses that are available to
read from the ring, then reads them and finally "releases" these
descriptors once it is done. The two local variables cons_tail
and cons_head are turned into one single variable called
cached_cons. cached_tail referred to the cached value of the
global consumer pointer and will be stored in cached_cons. For
cached_head, we just use cached_prod instead as it was not used
for a consumer queue before. It also better reflects what it
really is now: a cached copy of the producer pointer.

The names of the functions are also renamed in the same manner as
the producer functions. The new functions are called xskq_cons_
followed by what it does.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-8-git-send-email-magnus.karlsson@intel.com

At this point, there are no users of the functions xskq_nb_avail and
xskq_nb_free that take any other number of entries argument than 1, so
let us get rid of the second argument that takes the number of
entries.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-7-git-send-email-magnus.karlsson@intel.com

In the xsk consumer ring code there is a variable called RX_BATCH_SIZE
that dictates the minimum number of entries that we try to grab from
the fill and Tx rings. In fact, the code always try to grab the
maximum amount of entries from these rings. The only thing this
variable does is to throw an error if there is less than 16 (as it is
defined) entries on the ring. There is no reason to do this and it
will just lead to weird behavior from user space's point of view. So
eliminate this variable.

With this change, we will be able to simplify the xskq_nb_free and
xskq_nb_avail code in the next commit.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-6-git-send-email-magnus.karlsson@intel.com

Adopt the naming of the producer ring access functions to have a
similar naming convention as the functions in libbpf, but adapted to
the kernel. You first reserve a number of entries that you later
submit to the global state of the ring. This is much clearer, IMO,
than the one that was in the kernel part. Once renamed, we also
discover that two functions are actually the same, so remove one of
them. Some of the primitive ring submission operations are also the
same so break these out into __xskq_prod_submit that the upper level
ring access functions can use.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-5-git-send-email-magnus.karlsson@intel.com

Currently, the xsk ring code has two cached producer pointers:
prod_head and prod_tail. This patch consolidates these two into a
single one called cached_prod to make the code simpler and easier to
maintain. This will be in line with the user space part of the the
code found in libbpf, that only uses a single cached pointer.

The Rx path only uses the two top level functions
xskq_produce_batch_desc and xskq_produce_flush_desc and they both use
prod_head and never prod_tail. So just move them over to
cached_prod.

The Tx XDP_DRV path uses xskq_produce_addr_lazy and
xskq_produce_flush_addr_n and unnecessarily operates on both prod_tail
and prod_head, so move them over to just use cached_prod by skipping
the intermediate step of updating prod_tail.

The Tx path in XDP_SKB mode uses xskq_reserve_addr and
xskq_produce_addr. They currently use both cached pointers, but we can
operate on the global producer pointer in xskq_produce_addr since it
has to be updated anyway, thus eliminating the use of both cached
pointers. We can also remove the xskq_nb_free in xskq_produce_addr
since it is already called in xskq_reserve_addr. No need to do it
twice.

When there is only one cached producer pointer, we can also simplify
xskq_nb_free by removing one argument.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-4-git-send-email-magnus.karlsson@intel.com

In order to set the correct return flags for poll, the xsk code has to
check if the Rx queue is empty and if the Tx queue is full. This code
was unnecessarily large and complex as it used the functions that are
used to update the local state from the global state (xskq_nb_free and
xskq_nb_avail). Since we are not doing this nor updating any data
dependent on this state, we can simplify the functions. Another
benefit from this is that we can also simplify the xskq_nb_free and
xskq_nb_avail functions in a later commit.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-3-git-send-email-magnus.karlsson@intel.com

The lazy update threshold was introduced to keep the producer and
consumer some distance apart in the completion ring. This was
important in the beginning of the development of AF_XDP as the ring
format as that point in time was very sensitive to the producer and
consumer being on the same cache line. This is not the case
anymore as the current ring format does not degrade in any noticeable
way when this happens. Moreover, this threshold makes it impossible
to run the system with rings that have less than 128 entries.

So let us remove this threshold and just get one entry from the ring
as in all other functions. This will enable us to remove this function
in a later commit. Note that xskq_produce_addr_lazy followed by
xskq_produce_flush_addr_n are still not the same function as
xskq_produce_addr() as it operates on another cached pointer.
Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-2-git-send-email-magnus.karlsson@intel.com

Andrey Ignatov says:

====================
v3->v4:
- use OPTS_VALID and OPTS_GET to handle bpf_prog_attach_opts.

v2->v3:
- rely on DECLARE_LIBBPF_OPTS from libbpf_common.h;
- separate "required" and "optional" arguments in bpf_prog_attach_xattr;
- convert test_cgroup_attach to prog_tests;
- move the new selftest to prog_tests/cgroup_attach_multi.

v1->v2:
- move DECLARE_LIBBPF_OPTS from libbpf.h to bpf.h (patch 4);
- switch new libbpf API to OPTS framework;
- switch selftest to libbpf OPTS framework.

This patch set adds support for replacing cgroup-bpf programs attached with
BPF_F_ALLOW_MULTI flag so that any program in a list can be updated to a new
version without service interruption and order of programs can be preserved.

Please see patch 3 for details on the use-case and API changes.

Other patches:
Patch 1 is preliminary refactoring of __cgroup_bpf_attach to simplify it.
Patch 2 is minor cleanup of hierarchy_allows_attach.
Patch 4 extends libbpf API to support new set of attach attributes.
Patch 5 converts test_cgroup_attach to prog_tests.
Patch 6 adds selftest coverage for the new API.
====================
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>

Test replacing a cgroup-bpf program attached with BPF_F_ALLOW_MULTI and
possible failure modes: invalid combination of flags, invalid
replace_bpf_fd, replacing a non-attachd to specified cgroup program.

Example of program replacing:

  # gdb -q --args ./test_progs --name=cgroup_attach_multi
  ...
  Breakpoint 1, test_cgroup_attach_multi () at cgroup_attach_multi.c:227
  (gdb)
  [1]+  Stopped                 gdb -q --args ./test_progs --name=cgroup_attach_multi
  # bpftool c s /mnt/cgroup2/cgroup-test-work-dir/cg1
  ID       AttachType      AttachFlags     Name
  2133     egress          multi
  2134     egress          multi
  # fg
  gdb -q --args ./test_progs --name=cgroup_attach_multi
  (gdb) c
  Continuing.

  Breakpoint 2, test_cgroup_attach_multi () at cgroup_attach_multi.c:233
  (gdb)
  [1]+  Stopped                 gdb -q --args ./test_progs --name=cgroup_attach_multi
  # bpftool c s /mnt/cgroup2/cgroup-test-work-dir/cg1
  ID       AttachType      AttachFlags     Name
  2139     egress          multi
  2134     egress          multi
Signed-off-by: NAndrey Ignatov <rdna@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NAndrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/7b9b83e8d5fb82e15b034341bd40b6fb2431eeba.1576741281.git.rdna@fb.com

Convert test_cgroup_attach to prog_tests.

This change does a lot of things but in many cases it's pretty expensive
to separate them, so they go in one commit. Nevertheless the logic is
ketp as is and changes made are just moving things around, simplifying
them (w/o changing the meaning of the tests) and making prog_tests
compatible:

* split the 3 tests in the file into 3 separate files in prog_tests/;

* rename the test functions to test_<file_base_name>;

* remove unused includes, constants, variables and functions from every
  test;

* replace `if`-s with or `if (CHECK())` where additional context should
  be logged and with `if (CHECK_FAIL())` where line number is enough;

* switch from `log_err()` to logging via `CHECK()`;

* replace `assert`-s with `CHECK_FAIL()` to avoid crashing the whole
  test_progs if one assertion fails;

* replace cgroup_helpers with test__join_cgroup() in
  cgroup_attach_override only, other tests need more fine-grained
  control for cgroup creation/deletion so cgroup_helpers are still used
  there;

* simplify cgroup_attach_autodetach by switching to easiest possible
  program since this test doesn't really need such a complicated program
  as cgroup_attach_multi does;

* remove test_cgroup_attach.c itself.
Signed-off-by: NAndrey Ignatov <rdna@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NAndrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/0ff19cc64d2dc5cf404349f07131119480e10e32.1576741281.git.rdna@fb.com

Introduce a new bpf_prog_attach_xattr function that, in addition to
program fd, target fd and attach type, accepts an extendable struct
bpf_prog_attach_opts.

bpf_prog_attach_opts relies on DECLARE_LIBBPF_OPTS macro to maintain
backward and forward compatibility and has the following "optional"
attach attributes:

* existing attach_flags, since it's not required when attaching in NONE
  mode. Even though it's quite often used in MULTI and OVERRIDE mode it
  seems to be a good idea to reduce number of arguments to
  bpf_prog_attach_xattr;

* newly introduced attribute of BPF_PROG_ATTACH command: replace_prog_fd
  that is fd of previously attached cgroup-bpf program to replace if
  BPF_F_REPLACE flag is used.

The new function is named to be consistent with other xattr-functions
(bpf_prog_test_run_xattr, bpf_create_map_xattr, bpf_load_program_xattr).

The struct bpf_prog_attach_opts is supposed to be used with
DECLARE_LIBBPF_OPTS macro.
Signed-off-by: NAndrey Ignatov <rdna@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NAndrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/bd6e0732303eb14e4b79cb128268d9e9ad6db208.1576741281.git.rdna@fb.com

The common use-case in production is to have multiple cgroup-bpf
programs per attach type that cover multiple use-cases. Such programs
are attached with BPF_F_ALLOW_MULTI and can be maintained by different
people.

Order of programs usually matters, for example imagine two egress
programs: the first one drops packets and the second one counts packets.
If they're swapped the result of counting program will be different.

It brings operational challenges with updating cgroup-bpf program(s)
attached with BPF_F_ALLOW_MULTI since there is no way to replace a
program:

* One way to update is to detach all programs first and then attach the
  new version(s) again in the right order. This introduces an
  interruption in the work a program is doing and may not be acceptable
  (e.g. if it's egress firewall);

* Another way is attach the new version of a program first and only then
  detach the old version. This introduces the time interval when two
  versions of same program are working, what may not be acceptable if a
  program is not idempotent. It also imposes additional burden on
  program developers to make sure that two versions of their program can
  co-exist.

Solve the problem by introducing a "replace" mode in BPF_PROG_ATTACH
command for cgroup-bpf programs being attached with BPF_F_ALLOW_MULTI
flag. This mode is enabled by newly introduced BPF_F_REPLACE attach flag
and bpf_attr.replace_bpf_fd attribute to pass fd of the old program to
replace

That way user can replace any program among those attached with
BPF_F_ALLOW_MULTI flag without the problems described above.

Details of the new API:

* If BPF_F_REPLACE is set but replace_bpf_fd doesn't have valid
  descriptor of BPF program, BPF_PROG_ATTACH will return corresponding
  error (EINVAL or EBADF).

* If replace_bpf_fd has valid descriptor of BPF program but such a
  program is not attached to specified cgroup, BPF_PROG_ATTACH will
  return ENOENT.

BPF_F_REPLACE is introduced to make the user intent clear, since
replace_bpf_fd alone can't be used for this (its default value, 0, is a
valid fd). BPF_F_REPLACE also makes it possible to extend the API in the
future (e.g. add BPF_F_BEFORE and BPF_F_AFTER if needed).
Signed-off-by: NAndrey Ignatov <rdna@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NMartin KaFai Lau <kafai@fb.com>
Acked-by: NAndrii Narkyiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/30cd850044a0057bdfcaaf154b7d2f39850ba813.1576741281.git.rdna@fb.com

new_flags is unused, remove it.
Signed-off-by: NAndrey Ignatov <rdna@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NMartin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/2c49b30ab750f93cfef04a1e40b097d70c3a39a1.1576741281.git.rdna@fb.com

__cgroup_bpf_attach has a lot of identical code to handle two scenarios:
BPF_F_ALLOW_MULTI is set and unset.

Simplify it by splitting the two main steps:

* First, the decision is made whether a new bpf_prog_list entry should
  be allocated or existing entry should be reused for the new program.
  This decision is saved in replace_pl pointer;

* Next, replace_pl pointer is used to handle both possible states of
  BPF_F_ALLOW_MULTI flag (set / unset) instead of doing similar work for
  them separately.

This splitting, in turn, allows to make further simplifications:

* The check for attaching same program twice in BPF_F_ALLOW_MULTI mode
  can be done before allocating cgroup storage, so that if user tries to
  attach same program twice no alloc/free happens as it was before;

* pl_was_allocated becomes redundant so it's removed.
Signed-off-by: NAndrey Ignatov <rdna@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NMartin KaFai Lau <kafai@fb.com>
Acked-by: NAndrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/c6193db6fe630797110b0d3ff06c125d093b834c.1576741281.git.rdna@fb.com

Björn Töpel says:

====================
This series aims to simplify the XDP maps and
xdp_do_redirect_map()/xdp_do_flush_map(), and to crank out some more
performance from XDP_REDIRECT scenarios.

The first part of the series simplifies all XDP_REDIRECT capable maps,
so that __XXX_flush_map() does not require the map parameter, by
moving the flush list from the map to global scope.

This results in that the map_to_flush member can be removed from
struct bpf_redirect_info, and its corresponding logic.

Simpler code, and more performance due to that checks/code per-packet
is moved to flush.

Pre-series performance:
  $ sudo taskset -c 22 ./xdpsock -i enp134s0f0 -q 20 -n 1 -r -z

   sock0@enp134s0f0:20 rxdrop xdp-drv
                  pps         pkts        1.00
  rx              20,797,350  230,942,399
  tx              0           0

  $ sudo ./xdp_redirect_cpu --dev enp134s0f0 --cpu 22 xdp_cpu_map0

  Running XDP/eBPF prog_name:xdp_cpu_map5_lb_hash_ip_pairs
  XDP-cpumap      CPU:to  pps            drop-pps    extra-info
  XDP-RX          20      7723038        0           0
  XDP-RX          total   7723038        0
  cpumap_kthread  total   0              0           0
  redirect_err    total   0              0
  xdp_exception   total   0              0

Post-series performance:
  $ sudo taskset -c 22 ./xdpsock -i enp134s0f0 -q 20 -n 1 -r -z

   sock0@enp134s0f0:20 rxdrop xdp-drv
                  pps         pkts        1.00
  rx              21,524,979  86,835,327
  tx              0           0

  $ sudo ./xdp_redirect_cpu --dev enp134s0f0 --cpu 22 xdp_cpu_map0

  Running XDP/eBPF prog_name:xdp_cpu_map5_lb_hash_ip_pairs
  XDP-cpumap      CPU:to  pps            drop-pps    extra-info
  XDP-RX          20      7840124        0           0
  XDP-RX          total   7840124        0
  cpumap_kthread  total   0              0           0
  redirect_err    total   0              0
  xdp_exception   total   0              0

Results: +3.5% and +1.5% for the ubenchmarks.

v1->v2 [1]:
  * Removed 'unused-variable' compiler warning (Jakub)

[1] https://lore.kernel.org/bpf/20191218105400.2895-1-bjorn.topel@gmail.com/
====================
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>

The explicit error checking is not needed. Simply return the error
instead.
Signed-off-by: NBjörn Töpel <bjorn.topel@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NToke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-9-bjorn.topel@gmail.com

Now that all XDP maps that can be used with bpf_redirect_map() tracks
entries to be flushed in a global fashion, there is not need to track
that the map has changed and flush from xdp_do_generic_map()
anymore. All entries will be flushed in xdp_do_flush_map().

This means that the map_to_flush can be removed, and the corresponding
checks. Moving the flush logic to one place, xdp_do_flush_map(), give
a bulking behavior and performance boost.
Signed-off-by: NBjörn Töpel <bjorn.topel@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NToke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-8-bjorn.topel@gmail.com

The cpumap flush list is used to track entries that need to flushed
from via the xdp_do_flush_map() function. This list used to be
per-map, but there is really no reason for that. Instead make the
flush list global for all devmaps, which simplifies __cpu_map_flush()
and cpu_map_alloc().
Signed-off-by: NBjörn Töpel <bjorn.topel@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NToke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-7-bjorn.topel@gmail.com

The devmap flush list is used to track entries that need to flushed
from via the xdp_do_flush_map() function. This list used to be
per-map, but there is really no reason for that. Instead make the
flush list global for all devmaps, which simplifies __dev_map_flush()
and dev_map_init_map().
Signed-off-by: NBjörn Töpel <bjorn.topel@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NToke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-6-bjorn.topel@gmail.com

The xskmap flush list is used to track entries that need to flushed
from via the xdp_do_flush_map() function. This list used to be
per-map, but there is really no reason for that. Instead make the
flush list global for all xskmaps, which simplifies __xsk_map_flush()
and xsk_map_alloc().
Signed-off-by: NBjörn Töpel <bjorn.topel@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NToke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-5-bjorn.topel@gmail.com

Simple spelling fix.
Signed-off-by: NBjörn Töpel <bjorn.topel@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NToke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-4-bjorn.topel@gmail.com

After the RCU flavor consolidation [1], call_rcu() and
synchronize_rcu() waits for preempt-disable regions (NAPI) in addition
to the read-side critical sections. As a result of this, the cleanup
code in cpumap can be simplified

* There is no longer a need to flush in __cpu_map_entry_free, since we
  know that this has been done when the call_rcu() callback is
  triggered.

* When freeing the map, there is no need to explicitly wait for a
  flush. It's guaranteed to be done after the synchronize_rcu() call
  in cpu_map_free().

[1] https://lwn.net/Articles/777036/Signed-off-by: NBjörn Töpel <bjorn.topel@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NToke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-3-bjorn.topel@gmail.com

After the RCU flavor consolidation [1], call_rcu() and
synchronize_rcu() waits for preempt-disable regions (NAPI) in addition
to the read-side critical sections. As a result of this, the cleanup
code in devmap can be simplified

* There is no longer a need to flush in __dev_map_entry_free, since we
  know that this has been done when the call_rcu() callback is
  triggered.

* When freeing the map, there is no need to explicitly wait for a
  flush. It's guaranteed to be done after the synchronize_rcu() call
  in dev_map_free(). The rcu_barrier() is still needed, so that the
  map is not freed prior the elements.

[1] https://lwn.net/Articles/777036/Signed-off-by: NBjörn Töpel <bjorn.topel@intel.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NToke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-2-bjorn.topel@gmail.com

The two callers of bpf_prog_realloc - bpf_patch_insn_single and
bpf_migrate_filter dereference the struct fp_old, before passing
it to the function. Thus assertion to check fp_old is unnecessary
and can be removed.
Signed-off-by: NAditya Pakki <pakki001@umn.edu>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191219175735.19231-1-pakki001@umn.edu

Fix yet another printf warning for %llu specifier on ppc64le. This time size_t
casting won't work, so cast to verbose `unsigned long long`.

Fixes: 166750bc ("libbpf: Support libbpf-provided extern variables")
Signed-off-by: NAndrii Nakryiko <andriin@fb.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191219052103.3515-1-andriin@fb.com

Naresh pointed out that libbpf builds fail on 32-bit architectures because
rlimit.rlim_cur is defined as 'unsigned long long' on those architectures.
Fix this by using %zu in printf and casting to size_t.

Fixes: dc3a2d25 ("libbpf: Print hint about ulimit when getting permission denied error")
Reported-by: NNaresh Kamboju <naresh.kamboju@linaro.org>
Signed-off-by: NToke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191219090236.905059-1-toke@redhat.com

Fix two issues in test_attach_probe:

1. it was not able to parse /proc/self/maps beyond the first line,
   since %s means parse string until white space.
2. offset has to be accounted for otherwise uprobed address is incorrect.

Fixes: 1e8611bb ("selftests/bpf: add kprobe/uprobe selftests")
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NYonghong Song <yhs@fb.com>
Acked-by: NAndrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20191219020442.1922617-1-ast@kernel.org

The error log output in the opts validation macro was missing a newline.

Fixes: 2ce8450e ("libbpf: add bpf_object__open_{file, mem} w/ extensible opts")
Signed-off-by: NToke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191219120714.928380-1-toke@redhat.com

Björn Töpel says:

====================
This series contain one non-critical fix, support for far jumps, and
some optimizations for the BPF JIT.

Previously, the JIT only supported 12b branch targets for conditional
branches, and 21b for unconditional branches. Starting with this
series, 32b branching is supported.

As part of supporting far jumps, branch relaxation was introduced. The
idea is to start with a pessimistic jump (e.g. auipc/jalr) and for
each pass the JIT will have an opportunity to pick a better
instruction (e.g. jal) and shrink the image. Instead of two passes,
the JIT requires more passes. It typically converges after 3 passes.

The optimizations mentioned in the subject are for calls and tail
calls. In the tail call generation we can save one instruction by
using the offset in jalr. Calls are optimized by doing (auipc)/jal(r)
relative jumps instead of loading the entire absolute address and
doing jalr. This required that the JIT image allocator was made RISC-V
specific, so we can ensure that the JIT image and the kernel text are
in range (32b).

The last two patches of the series is not critical to the series, but
are two UAPI build issues for BPF events. A closer look from the
RV-folks would be much appreciated.

The test_bpf.ko module, selftests/bpf/test_verifier and
selftests/seccomp/seccomp_bpf pass all tests.

RISC-V is still missing proper kprobe and tracepoint support, so a lot
of BPF selftests cannot be run.

v1->v2: [1]
 * Removed unused function parameter from emit_branch()
 * Added patch to support far branch in tail call emit

[1] https://lore.kernel.org/bpf/20191209173136.29615-1-bjorn.topel@gmail.com/
====================
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

RISC-V was missing a proper perf_arch_bpf_user_pt_regs macro for
CONFIG_PERF_EVENT builds.
Signed-off-by: NBjörn Töpel <bjorn.topel@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216091343.23260-10-bjorn.topel@gmail.com

Add missing uapi header the BPF_PROG_TYPE_PERF_EVENT programs by
exporting struct user_regs_struct instead of struct pt_regs which is
in-kernel only.
Signed-off-by: NBjörn Töpel <bjorn.topel@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216091343.23260-9-bjorn.topel@gmail.com

Instead of using emit_imm() and emit_jalr() which can expand to six
instructions, start using jal or auipc+jalr.
Signed-off-by: NBjörn Töpel <bjorn.topel@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216091343.23260-8-bjorn.topel@gmail.com

This commit makes sure that the JIT images is kept close to the kernel
text, so BPF calls can use relative calling with auipc/jalr or jal
instead of loading the full 64-bit address and jalr.

The BPF JIT image region is 128 MB before the kernel text.
Signed-off-by: NBjörn Töpel <bjorn.topel@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191216091343.23260-7-bjorn.topel@gmail.com