Roman Gushchin [Thu, 30 May 2019 01:03:58 +0000 (18:03 -0700)]
bpf: rework memlock-based memory accounting for maps
In order to unify the existing memlock charging code with the
memcg-based memory accounting, which will be added later, let's
rework the current scheme.
Currently the following design is used:
1) .alloc() callback optionally checks if the allocation will likely
succeed using bpf_map_precharge_memlock()
2) .alloc() performs actual allocations
3) .alloc() callback calculates map cost and sets map.memory.pages
4) map_create() calls bpf_map_init_memlock() which sets map.memory.user
and performs actual charging; in case of failure the map is
destroyed
<map is in use>
1) bpf_map_free_deferred() calls bpf_map_release_memlock(), which
performs uncharge and releases the user
2) .map_free() callback releases the memory
The scheme can be simplified and made more robust:
1) .alloc() calculates map cost and calls bpf_map_charge_init()
2) bpf_map_charge_init() sets map.memory.user and performs actual
charge
3) .alloc() performs actual allocations
<map is in use>
1) .map_free() callback releases the memory
2) bpf_map_charge_finish() performs uncharge and releases the user
The new scheme also allows to reuse bpf_map_charge_init()/finish()
functions for memcg-based accounting. Because charges are performed
before actual allocations and uncharges after freeing the memory,
no bogus memory pressure can be created.
In cases when the map structure is not available (e.g. it's not
created yet, or is already destroyed), on-stack bpf_map_memory
structure is used. The charge can be transferred with the
bpf_map_charge_move() function.
Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Roman Gushchin [Thu, 30 May 2019 01:03:57 +0000 (18:03 -0700)]
bpf: group memory related fields in struct bpf_map_memory
Group "user" and "pages" fields of bpf_map into the bpf_map_memory
structure. Later it can be extended with "memcg" and other related
information.
The main reason for a such change (beside cosmetics) is to pass
bpf_map_memory structure to charging functions before the actual
allocation of bpf_map.
Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
====================
This patchset adds support for propagating congestion notifications (cn)
to TCP from cgroup inet skb egress BPF programs.
Current cgroup skb BPF programs cannot trigger TCP congestion window
reductions, even when they drop a packet. This patch-set adds support
for cgroup skb BPF programs to send congestion notifications in the
return value when the packets are TCP packets. Rather than the
current 1 for keeping the packet and 0 for dropping it, they can
now return:
NET_XMIT_SUCCESS (0) - continue with packet output
NET_XMIT_DROP (1) - drop packet and do cn
NET_XMIT_CN (2) - continue with packet output and do cn
-EPERM - drop packet
Finally, HBM programs are modified to collect and return more
statistics.
There has been some discussion regarding the best place to manage
bandwidths. Some believe this should be done in the qdisc where it can
also be managed with a BPF program. We believe there are advantages
for doing it with a BPF program in the cgroup/skb callback. For example,
it reduces overheads in the cases where there is on primary workload and
one or more secondary workloads, where each workload is running on its
own cgroupv2. In this scenario, we only need to throttle the secondary
workloads and there is no overhead for the primary workload since there
will be no BPF program attached to its cgroup.
Regardless, we agree that this mechanism should not penalize those that
are not using it. We tested this by doing 1 byte req/reply RPCs over
loopback. Each test consists of 30 sec of back-to-back 1 byte RPCs.
Each test was repeated 50 times with a 1 minute delay between each set
of 10. We then calculated the average RPCs/sec over the 50 tests. We
compare upstream with upstream + patchset and no BPF program as well
as upstream + patchset and a BPF program that just returns ALLOW_PKT.
Here are the results:
upstream 80937 RPCs/sec
upstream + patches, no BPF program 80894 RPCs/sec
upstream + patches, BPF program 80634 RPCs/sec
These numbers indicate that there is no penalty for these patches
The use of congestion notifications improves the performance of HBM when
using Cubic. Without congestion notifications, Cubic will not decrease its
cwnd and HBM will need to drop a large percentage of the packets.
The following results are obtained for rate limits of 1Gbps,
between two servers using netperf, and only one flow. We also show how
reducing the max delayed ACK timer can improve the performance when
using Cubic.
Command used was:
./do_hbm_test.sh -l -D --stats -N -r=<rate> [--no_cn] [dctcp] \
-s=<server running netserver>
where:
<rate> is 1000
--no_cn specifies no cwr notifications
dctcp uses dctcp
Notes:
--no_cn has no effect with DCTCP
Lim = rate limit
DA = maximum delay ack timer
cred = credit in packets
drops = % packets dropped
v1->v2: Insures that only BPF_CGROUP_INET_EGRESS can return values 2 and 3
New egress values apply to all protocols, not just TCP
Cleaned up patch 4, Update BPF_CGROUP_RUN_PROG_INET_EGRESS callers
Removed changes to __tcp_transmit_skb (patch 5), no longer needed
Removed sample use of EDT
v2->v3: Removed the probe timer related changes
v3->v4: Replaced preempt_enable_no_resched() by preempt_enable()
in BPF_PROG_CGROUP_INET_EGRESS_RUN_ARRAY() macro
====================
Acked-by: Martin KaFai Lau <kafai@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
brakmo [Tue, 28 May 2019 23:59:37 +0000 (16:59 -0700)]
bpf: Update __cgroup_bpf_run_filter_skb with cn
For egress packets, __cgroup_bpf_fun_filter_skb() will now call
BPF_PROG_CGROUP_INET_EGRESS_RUN_ARRAY() instead of PROG_CGROUP_RUN_ARRAY()
in order to propagate congestion notifications (cn) requests to TCP
callers.
For egress packets, this function can return:
NET_XMIT_SUCCESS (0) - continue with packet output
NET_XMIT_DROP (1) - drop packet and notify TCP to call cwr
NET_XMIT_CN (2) - continue with packet output and notify TCP
to call cwr
-EPERM - drop packet
For ingress packets, this function will return -EPERM if any attached
program was found and if it returned != 1 during execution. Otherwise 0
is returned.
Signed-off-by: Lawrence Brakmo <brakmo@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
brakmo [Tue, 28 May 2019 23:59:36 +0000 (16:59 -0700)]
bpf: cgroup inet skb programs can return 0 to 3
Allows cgroup inet skb programs to return values in the range [0, 3].
The second bit is used to deterine if congestion occurred and higher
level protocol should decrease rate. E.g. TCP would call tcp_enter_cwr()
The bpf_prog must set expected_attach_type to BPF_CGROUP_INET_EGRESS
at load time if it uses the new return values (i.e. 2 or 3).
The expected_attach_type is currently not enforced for
BPF_PROG_TYPE_CGROUP_SKB. e.g Meaning the current bpf_prog with
expected_attach_type setting to BPF_CGROUP_INET_EGRESS can attach to
BPF_CGROUP_INET_INGRESS. Blindly enforcing expected_attach_type will
break backward compatibility.
This patch adds a enforce_expected_attach_type bit to only
enforce the expected_attach_type when it uses the new
return value.
Signed-off-by: Lawrence Brakmo <brakmo@fb.com> Signed-off-by: Martin KaFai Lau <kafai@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
brakmo [Tue, 28 May 2019 23:59:35 +0000 (16:59 -0700)]
bpf: Create BPF_PROG_CGROUP_INET_EGRESS_RUN_ARRAY
Create new macro BPF_PROG_CGROUP_INET_EGRESS_RUN_ARRAY() to be used by
__cgroup_bpf_run_filter_skb for EGRESS BPF progs so BPF programs can
request cwr for TCP packets.
Current cgroup skb programs can only return 0 or 1 (0 to drop the
packet. This macro changes the behavior so the low order bit
indicates whether the packet should be dropped (0) or not (1)
and the next bit is used for congestion notification (cn).
Hence, new allowed return values of CGROUP EGRESS BPF programs are:
0: drop packet
1: keep packet
2: drop packet and call cwr
3: keep packet and call cwr
This macro then converts it to one of NET_XMIT values or -EPERM
that has the effect of dropping the packet with no cn.
0: NET_XMIT_SUCCESS skb should be transmitted (no cn)
1: NET_XMIT_DROP skb should be dropped and cwr called
2: NET_XMIT_CN skb should be transmitted and cwr called
3: -EPERM skb should be dropped (no cn)
Note that when more than one BPF program is called, the packet is
dropped if at least one of programs requests it be dropped, and
there is cn if at least one program returns cn.
Signed-off-by: Lawrence Brakmo <brakmo@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Andrii Nakryiko [Wed, 29 May 2019 17:36:08 +0000 (10:36 -0700)]
libbpf: use negative fd to specify missing BTF
0 is a valid FD, so it's better to initialize it to -1, as is done in
other places. Also, technically, BTF type ID 0 is valid (it's a VOID
type), so it's more reliable to check btf_fd, instead of
btf_key_type_id, to determine if there is any BTF associated with a map.
Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Andrii Nakryiko [Wed, 29 May 2019 17:36:05 +0000 (10:36 -0700)]
libbpf: simplify endianness check
Rewrite endianness check to use "more canonical" way, using
compiler-defined macros, similar to few other places in libbpf. It also
is more obvious and shorter.
Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Quentin Monnet [Wed, 29 May 2019 14:26:41 +0000 (15:26 +0100)]
libbpf: prevent overwriting of log_level in bpf_object__load_progs()
There are two functions in libbpf that support passing a log_level
parameter for the verifier for loading programs:
bpf_object__load_xattr() and bpf_prog_load_xattr(). Both accept an
attribute object containing the log_level, and apply it to the programs
to load.
It turns out that to effectively load the programs, the latter function
eventually relies on the former. This was not taken into account when
adding support for log_level in bpf_object__load_xattr(), and the
log_level passed to bpf_prog_load_xattr() later gets overwritten with a
zero value, thus disabling verifier logs for the program in all cases:
Now that we don't have __rcu markers on the bpf_prog_array helpers,
let's use proper rcu_dereference_protected to obtain array pointer
under mutex.
We also don't need __rcu annotations on cgroup_bpf.inactive since
it's not read/updated concurrently.
v4:
* drop cgroup_rcu_xyz wrappers and use rcu APIs directly; presumably
should be more clear to understand which mutex/refcount protects
each particular place
v3:
* amend cgroup_rcu_dereference to include percpu_ref_is_dying;
cgroup_bpf is now reference counted and we don't hold cgroup_mutex
anymore in cgroup_bpf_release
v2:
* replace xchg with rcu_swap_protected
Cc: Roman Gushchin <guro@fb.com> Signed-off-by: Stanislav Fomichev <sdf@google.com> Acked-by: Roman Gushchin <guro@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Now that we don't have __rcu markers on the bpf_prog_array helpers,
let's use proper rcu_dereference_protected to obtain array pointer
under mutex.
Cc: linux-media@vger.kernel.org Cc: Mauro Carvalho Chehab <mchehab@kernel.org> Cc: Sean Young <sean@mess.org> Signed-off-by: Stanislav Fomichev <sdf@google.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Drop __rcu annotations and rcu read sections from bpf_prog_array
helper functions. They are not needed since all existing callers
call those helpers from the rcu update side while holding a mutex.
This guarantees that use-after-free could not happen.
In the next patches I'll fix the callers with missing
rcu_dereference_protected to make sparse/lockdep happy, the proper
way to use these helpers is:
No functional changes! rcu_dereference_protected with lockdep_is_held
should catch any cases where we update prog array without a mutex
(I've looked at existing call sites and I think we hold a mutex
everywhere).
Motivation is to fix sparse warnings:
kernel/bpf/core.c:1803:9: warning: incorrect type in argument 1 (different address spaces)
kernel/bpf/core.c:1803:9: expected struct callback_head *head
kernel/bpf/core.c:1803:9: got struct callback_head [noderef] <asn:4> *
kernel/bpf/core.c:1877:44: warning: incorrect type in initializer (different address spaces)
kernel/bpf/core.c:1877:44: expected struct bpf_prog_array_item *item
kernel/bpf/core.c:1877:44: got struct bpf_prog_array_item [noderef] <asn:4> *
kernel/bpf/core.c:1901:26: warning: incorrect type in assignment (different address spaces)
kernel/bpf/core.c:1901:26: expected struct bpf_prog_array_item *existing
kernel/bpf/core.c:1901:26: got struct bpf_prog_array_item [noderef] <asn:4> *
kernel/bpf/core.c:1935:26: warning: incorrect type in assignment (different address spaces)
kernel/bpf/core.c:1935:26: expected struct bpf_prog_array_item *[assigned] existing
kernel/bpf/core.c:1935:26: got struct bpf_prog_array_item [noderef] <asn:4> *
v2:
* remove comment about potential race; that can't happen
because all callers are in rcu-update section
Cc: Roman Gushchin <guro@fb.com> Acked-by: Roman Gushchin <guro@fb.com> Signed-off-by: Stanislav Fomichev <sdf@google.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Alan Maguire [Wed, 29 May 2019 09:48:14 +0000 (10:48 +0100)]
selftests/bpf: fix compilation error for flow_dissector.c
When building the tools/testing/selftest/bpf subdirectory,
(running both a local directory "make" and a
"make -C tools/testing/selftests/bpf") I keep hitting the
following compilation error:
prog_tests/flow_dissector.c: In function ‘create_tap’:
prog_tests/flow_dissector.c:150:38: error: ‘IFF_NAPI’ undeclared (first
use in this function)
.ifr_flags = IFF_TAP | IFF_NO_PI | IFF_NAPI | IFF_NAPI_FRAGS,
^
prog_tests/flow_dissector.c:150:38: note: each undeclared identifier is
reported only once for each function it appears in
prog_tests/flow_dissector.c:150:49: error: ‘IFF_NAPI_FRAGS’ undeclared
Adding include/uapi/linux/if_tun.h to tools/include/uapi/linux
resolves the problem and ensures the compilation of the file
does not depend on having up-to-date kernel headers locally.
Signed-off-by: Alan Maguire <alan.maguire@oracle.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
====================
This patchset implements a cgroup bpf auto-detachment functionality:
bpf programs are detached as soon as possible after removal of the
cgroup, without waiting for the release of all associated resources.
Patches 2 and 3 are required to implement a corresponding kselftest
in patch 4.
v5:
1) rebase
v4:
1) release cgroup bpf data using a workqueue
2) add test_cgroup_attach to .gitignore
v3:
1) some minor changes and typo fixes
v2:
1) removed a bogus check in patch 4
2) moved buf[len] = 0 in patch 2
====================
Roman Gushchin [Sat, 25 May 2019 16:37:42 +0000 (09:37 -0700)]
selftests/bpf: add auto-detach test
Add a kselftest to cover bpf auto-detachment functionality.
The test creates a cgroup, associates some resources with it,
attaches a couple of bpf programs and deletes the cgroup.
Then it checks that bpf programs are going away in 5 seconds.
Roman Gushchin [Sat, 25 May 2019 16:37:41 +0000 (09:37 -0700)]
selftests/bpf: enable all available cgroup v2 controllers
Enable all available cgroup v2 controllers when setting up
the environment for the bpf kselftests. It's required to properly test
the bpf prog auto-detach feature. Also it will generally increase
the code coverage.
Signed-off-by: Roman Gushchin <guro@fb.com> Acked-by: Yonghong Song <yhs@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Roman Gushchin [Sat, 25 May 2019 16:37:40 +0000 (09:37 -0700)]
selftests/bpf: convert test_cgrp2_attach2 example into kselftest
Convert test_cgrp2_attach2 example into a proper test_cgroup_attach
kselftest. It's better because we do run kselftest on a constant
basis, so there are better chances to spot a potential regression.
Also make it slightly less verbose to conform kselftests output style.
Roman Gushchin [Sat, 25 May 2019 16:37:39 +0000 (09:37 -0700)]
bpf: decouple the lifetime of cgroup_bpf from cgroup itself
Currently the lifetime of bpf programs attached to a cgroup is bound
to the lifetime of the cgroup itself. It means that if a user
forgets (or intentionally avoids) to detach a bpf program before
removing the cgroup, it will stay attached up to the release of the
cgroup. Since the cgroup can stay in the dying state (the state
between being rmdir()'ed and being released) for a very long time, it
leads to a waste of memory. Also, it blocks a possibility to implement
the memcg-based memory accounting for bpf objects, because a circular
reference dependency will occur. Charged memory pages are pinning the
corresponding memory cgroup, and if the memory cgroup is pinning
the attached bpf program, nothing will be ever released.
A dying cgroup can not contain any processes, so the only chance for
an attached bpf program to be executed is a live socket associated
with the cgroup. So in order to release all bpf data early, let's
count associated sockets using a new percpu refcounter. On cgroup
removal the counter is transitioned to the atomic mode, and as soon
as it reaches 0, all bpf programs are detached.
Because cgroup_bpf_release() can block, it can't be called from
the percpu ref counter callback directly, so instead an asynchronous
work is scheduled.
The reference counter is not socket specific, and can be used for any
other types of programs, which can be executed from a cgroup-bpf hook
outside of the process context, had such a need arise in the future.
Signed-off-by: Roman Gushchin <guro@fb.com> Cc: jolsa@redhat.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
selftests/bpf: fail test_tunnel.sh if subtests fail
Right now test_tunnel.sh always exits with success even if some
of the subtests fail. Since the output is very verbose, it's
hard to spot the issues with subtests. Let's fail the script
if any subtest fails.
Signed-off-by: Stanislav Fomichev <sdf@google.com> Acked-by: Yonghong Song <yhs@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Daniel Borkmann [Tue, 28 May 2019 09:05:43 +0000 (11:05 +0200)]
Merge branch 'bpf-bpftool-dbg-output'
Quentin Monnet says:
====================
This series adds an option to bpftool to make it print additional
information via libbpf and the kernel verifier when attempting to load
programs.
A new API function is added to libbpf in order to pass the log_level
from bpftool with the bpf_object__* part of the API.
Options for a finer control over the log levels to use for libbpf and
the verifier could be added in the future, if desired.
v3:
- Fix and clarify commit logs.
v2:
- Do not add distinct options for libbpf and verifier logs, just keep the
one that sets all log levels to their maximum. Rename the option.
- Do not offer a way to pick desired log levels. The choice is "use the
option to print all logs" or "stick with the defaults".
- Do not export BPF_LOG_* flags to user header.
- Update all man pages (most bpftool operations use libbpf and may print
libbpf logs). Verifier logs are only used when attempting to load
programs for now, so bpftool-prog.rst and bpftool.rst remain the only
pages updated in that regard.
Quentin Monnet [Fri, 24 May 2019 10:36:48 +0000 (11:36 +0100)]
tools: bpftool: make -d option print debug output from verifier
The "-d" option is used to require all logs available for bpftool. So
far it meant telling libbpf to print even debug-level information. But
there is another source of info that can be made more verbose: when we
attemt to load programs with bpftool, we can pass a log_level parameter
to the verifier in order to control the amount of information that is
printed to the console.
Reuse the "-d" option to print all information the verifier can tell. At
this time, this means logs related to BPF_LOG_LEVEL1, BPF_LOG_LEVEL2 and
BPF_LOG_STATS. As mentioned in the discussion on the first version of
this set, these macros are internal to the kernel
(include/linux/bpf_verifier.h) and are not meant to be part of the
stable user API, therefore we simply use the related constants to print
whatever we can at this time, without trying to tell users what is
log_level1 or what is statistics.
Verifier logs are only used when loading programs for now (In the
future: for loading BTF objects with bpftool? Although libbpf does not
currently offer to print verifier info at debug level if no error
occurred when loading BTF objects), so bpftool.rst and bpftool-prog.rst
are the only man pages to get the update.
v3:
- Add details on log level and BTF loading at the end of commit log.
v2:
- Remove the possibility to select the log levels to use (v1 offered a
combination of "log_level1", "log_level2" and "stats").
- The macros from kernel header bpf_verifier.h are not used (and
therefore not moved to UAPI header).
- In v1 this was a distinct option, but is now merged in the only "-d"
switch to activate libbpf and verifier debug-level logs all at the
same time.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Quentin Monnet [Fri, 24 May 2019 10:36:47 +0000 (11:36 +0100)]
libbpf: add bpf_object__load_xattr() API function to pass log_level
libbpf was recently made aware of the log_level attribute for programs,
used to specify the level of information expected to be dumped by the
verifier. Function bpf_prog_load_xattr() got support for this log_level
parameter.
But some applications using libbpf rely on another function to load
programs, bpf_object__load(), which does accept any parameter for log
level. Create an API function based on bpf_object__load(), but accepting
an "attr" object as a parameter. Then add a log_level field to that
object, so that applications calling the new bpf_object__load_xattr()
can pick the desired log level.
v3:
- Rewrite commit log.
v2:
- We are in a new cycle, bump libbpf extraversion number.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Quentin Monnet [Fri, 24 May 2019 10:36:46 +0000 (11:36 +0100)]
tools: bpftool: add -d option to get debug output from libbpf
libbpf has three levels of priority for output messages: warn, info,
debug. By default, debug output is not printed to the console.
Add a new "--debug" (short name: "-d") option to bpftool to print libbpf
logs for all three levels.
Internally, we simply use the function provided by libbpf to replace the
default printing function by one that prints logs regardless of their
level.
v2:
- Remove the possibility to select the log-levels to use (v1 offered a
combination of "warn", "info" and "debug").
- Rename option and offer a short name: -d|--debug.
- Add option description to all bpftool manual pages (instead of
bpftool-prog.rst only), as all commands use libbpf.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Yonghong Song [Sat, 25 May 2019 18:57:53 +0000 (11:57 -0700)]
bpf: check signal validity in nmi for bpf_send_signal() helper
Commit 8b401f9ed244 ("bpf: implement bpf_send_signal() helper")
introduced bpf_send_signal() helper. If the context is nmi,
the sending signal work needs to be deferred to irq_work.
If the signal is invalid, the error will appear in irq_work
and it won't be propagated to user.
This patch did an early check in the helper itself to notify
user invalid signal, as suggested by Daniel.
Suggested-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Yonghong Song <yhs@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Andrii Nakryiko [Sun, 26 May 2019 00:01:01 +0000 (17:01 -0700)]
bpftool: auto-complete BTF IDs for btf dump
Auto-complete BTF IDs for `btf dump id` sub-command. List of possible BTF
IDs is scavenged from loaded BPF programs that have associated BTFs, as
there is currently no API in libbpf to fetch list of all BTFs in the
system.
====================
v9:
- Split patch 5 in v8.
make bpf uapi header file sync a separate patch. (Alexei)
v8:
- For stack slot read, mark them as REG_LIVE_READ64. (Alexei)
- Change DEF_NOT_SUBREG from -1 to 0. (Alexei)
- Rebased on top of latest bpf-next.
v7:
- Drop the first patch in v6, the one adding 32-bit return value and
argument type. (Alexei)
- Rename bpf_jit_hardware_zext to bpf_jit_needs_zext. (Alexei)
- Use mov32 with imm == 1 to indicate it is zext. (Alexei)
- JIT back-ends peephole next insn to optimize out unnecessary zext
inserted by verifier. (Alexei)
- Testing:
+ patch set tested (bpf selftest) on x64 host with llvm 9.0
no regression observed no both JIT and interpreter modes.
+ patch set tested (bpf selftest) on x32 host.
By Yanqing Wang, thanks!
no regression observed on both JIT and interpreter modes.
+ patch set tested (bpf selftest) on RV64 host with llvm 9.0,
By Björn Töpel, thanks!
no regression observed before and after this set with JIT_ALWAYS_ON.
test_progs_32 also enabled as LLVM 9.0 is used by Björn.
+ cross compiled the other affected targets, arm, PowerPC, SPARC, S390.
v6:
- Fixed s390 kbuild test robot error. (kbuild)
- Make comment style in backends patches more consistent.
v5:
- Adjusted several test_verifier helpers to make them works on hosts
w and w/o hardware zext. (Naveen)
- Make sure zext flag not set when verifier by-passed, for example,
libtest_bpf.ko. (Naveen)
- Conservatively mark bpf main return value as 64-bit. (Alexei)
- Make sure read flag is either READ64 or READ32, not the mix of both.
(Alexei)
- Merged patch 1 and 2 in v4. (Alexei)
- Fixed kbuild test robot warning on NFP. (kbuild)
- Proposed new BPF_ZEXT insn to have optimal code-gen for various JIT
back-ends.
- Conservately set zext flags for patched-insn.
- Fixed return value zext for helper function calls.
- Also adjusted test_verifier scalability unit test to avoid triggerring
too many insn patch which will hang computer.
- re-tested on x86 host with llvm 9.0, no regression on test_verifier,
test_progs, test_progs_32.
- re-tested offload target (nfp), no regression on local testsuite.
v4:
- added the two missing fixes which addresses two Jakub's reviewes in v3.
- rebase on top of bpf-next.
v3:
- remove redundant check in "propagate_liveness_reg". (Jakub)
- add extra check in "mark_reg_read" to prune more search. (Jakub)
- re-implemented "prog_flags" passing mechanism, removed use of
global switch inside libbpf.
- enabled high 32-bit randomization beyond "test_verifier" and
"test_progs". Now it should have been enabled for all possible
tests. Re-run all tests, haven't noticed regression.
- remove RFC tag.
v2:
- rebased on top of bpf-next master.
- added comments for what is sub-register def index. (Edward, Alexei)
- removed patch 1 which turns bit mask from enum to macro. (Alexei)
- removed sysctl/bpf_jit_32bit_opt. (Alexei)
- merged sub-register def insn index into reg state. (Alexei)
- change test methodology (Alexei):
+ instead of simple unit tests on x86_64 for which this optimization
doesn't enabled due to there is hardware support, poison high
32-bit for whose def identified as safe to do so. this could let
the correctness of this patch set checked when daily bpf selftest
ran which delivers very stressful test on host machine like x86_64.
+ hi32 poisoning is gated by a new BPF_F_TEST_RND_HI32 prog flags.
+ BPF_F_TEST_RND_HI32 is enabled for all tests of "test_progs" and
"test_verifier", the latter needs minor tweak on two unit tests,
please see the patch for the change.
+ introduced a new global variable "libbpf_test_mode" into libbpf.
once it is set to true, it will set BPF_F_TEST_RND_HI32 for all the
later PROG_LOAD syscall, the goal is to easy the enable of hi32
poison on exsiting testsuite.
we could also introduce new APIs, for example "bpf_prog_test_load",
then use -Dbpf_prog_load=bpf_prog_test_load to migrate tests under
test_progs, but there are several load APIs, and such new API need
some change on struture like "struct bpf_prog_load_attr".
+ removed old unit tests. it is based on insn scan and requires quite
a few test_verifier generic code change. given hi32 randomization
could offer good test coverage, the unit tests doesn't add much
extra test value.
- enhanced register width check ("is_reg64") when record sub-register
write, now, it returns more accurate width.
- Re-run all tests under "test_progs" and "test_verifier" on x86_64, no
regression. Fixed a couple of bugs exposed:
1. ctx field size transformation was not taken into account.
2. insn patch could cause lost of original aux data which is
important for ctx field conversion.
3. return value for propagate_liveness was wrong and caused
regression on processed insn number.
4. helper call arg wasn't handled properly that path prune may cause
64-bit read info in pruned path lost.
- Re-run Cilium bpf prog for processed-insn-number benchmarking, no
regression.
v1:
- Fixed the missing handling on callee-saved for bpf-to-bpf call,
sub-register defs therefore moved to frame state. (Jakub Kicinski)
- Removed redundant "cross_reg". (Jakub Kicinski)
- Various coding styles & grammar fixes. (Jakub Kicinski, Quentin Monnet)
eBPF ISA specification requires high 32-bit cleared when low 32-bit
sub-register is written. This applies to destination register of ALU32 etc.
JIT back-ends must guarantee this semantic when doing code-gen. x86_64 and
AArch64 ISA has the same semantics, so the corresponding JIT back-end
doesn't need to do extra work.
However, 32-bit arches (arm, x86, nfp etc.) and some other 64-bit arches
(PowerPC, SPARC etc) need to do explicit zero extension to meet this
requirement, otherwise code like the following will fail.
u64_value = (u64) u32_value
... other uses of u64_value
This is because compiler could exploit the semantic described above and
save those zero extensions for extending u32_value to u64_value, these JIT
back-ends are expected to guarantee this through inserting extra zero
extensions which however could be a significant increase on the code size.
Some benchmarks show there could be ~40% sub-register writes out of total
insns, meaning at least ~40% extra code-gen.
One observation is these extra zero extensions are not always necessary.
Take above code snippet for example, it is possible u32_value will never be
casted into a u64, the value of high 32-bit of u32_value then could be
ignored and extra zero extension could be eliminated.
This patch implements this idea, insns defining sub-registers will be
marked when the high 32-bit of the defined sub-register matters. For
those unmarked insns, it is safe to eliminate high 32-bit clearnace for
them.
Algo
====
We could use insn scan based static analysis to tell whether one
sub-register def doesn't need zero extension. However, using such static
analysis, we must do conservative assumption at branching point where
multiple uses could be introduced. So, for any sub-register def that is
active at branching point, we need to mark it as needing zero extension.
This could introducing quite a few false alarms, for example ~25% on
Cilium bpf_lxc.
It will be far better to use dynamic data-flow tracing which verifier
fortunately already has and could be easily extend to serve the purpose of
this patch set.
- Split read flags into READ32 and READ64.
- Record index of insn that does sub-register write. Keep the index inside
reg state and update it during verifier insn walking.
- A full register read on a sub-register marks its definition insn as
needing zero extension on dst register.
A new sub-register write overrides the old one.
- When propagating read64 during path pruning, also mark any insn defining
a sub-register that is read in the pruned path as full-register.
Benchmark
=========
- I estimate the JITed image could be 10% ~ 30% smaller on these affected
arches (nfp, arm, x32, risv, ppc, sparc, s390), depending on the prog.
- For Cilium bpf_lxc, there is ~11500 insns in the compiled binary (use
latest LLVM snapshot, and with -mcpu=v3 -mattr=+alu32 enabled), 4460 of
them has sub-register writes (~40%). Calculated by:
After this patch set enabled, > 25% of those 4460 could be identified as
doesn't needing zero extension on the destination, and the percentage
could go further up to more than 50% with some follow up optimizations
based on the infrastructure offered by this set. This leads to
significant save on JITed image.
====================
Jiong Wang [Fri, 24 May 2019 22:25:28 +0000 (23:25 +0100)]
nfp: bpf: eliminate zero extension code-gen
This patch eliminate zero extension code-gen for instructions including
both alu and load/store. The only exception is for ctx load, because
offload target doesn't go through host ctx convert logic so we do
customized load and ignores zext flag set by verifier.
Cc: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Jiong Wang [Fri, 24 May 2019 22:25:27 +0000 (23:25 +0100)]
riscv: bpf: eliminate zero extension code-gen
Cc: Björn Töpel <bjorn.topel@gmail.com> Acked-by: Björn Töpel <bjorn.topel@gmail.com> Tested-by: Björn Töpel <bjorn.topel@gmail.com> Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Jiong Wang [Fri, 24 May 2019 22:25:21 +0000 (23:25 +0100)]
selftests: bpf: enable hi32 randomization for all tests
The previous libbpf patch allows user to specify "prog_flags" to bpf
program load APIs. To enable high 32-bit randomization for a test, we need
to set BPF_F_TEST_RND_HI32 in "prog_flags".
To enable such randomization for all tests, we need to make sure all places
are passing BPF_F_TEST_RND_HI32. Changing them one by one is not
convenient, also, it would be better if a test could be switched to
"normal" running mode without code change.
Given the program load APIs used across bpf selftests are mostly:
bpf_prog_load: load from file
bpf_load_program: load from raw insns
A test_stub.c is implemented for bpf seltests, it offers two functions for
testing purpose:
bpf_prog_test_load
bpf_test_load_program
The are the same as "bpf_prog_load" and "bpf_load_program", except they
also set BPF_F_TEST_RND_HI32. Given *_xattr functions are the APIs to
customize any "prog_flags", it makes little sense to put these two
functions into libbpf.
Then, the following CFLAGS are passed to compilations for host programs:
-Dbpf_prog_load=bpf_prog_test_load
-Dbpf_load_program=bpf_test_load_program
They migrate the used load APIs to the test version, hence enable high
32-bit randomization for these tests without changing source code.
Besides all these, there are several testcases are using
"bpf_prog_load_attr" directly, their call sites are updated to pass
BPF_F_TEST_RND_HI32.
Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Jiong Wang [Fri, 24 May 2019 22:25:20 +0000 (23:25 +0100)]
selftests: bpf: adjust several test_verifier helpers for insn insertion
- bpf_fill_ld_abs_vlan_push_pop:
Prevent zext happens inside PUSH_CNT loop. This could happen because
of BPF_LD_ABS (32-bit def) + BPF_JMP (64-bit use), or BPF_LD_ABS +
EXIT (64-bit use of R0). So, change BPF_JMP to BPF_JMP32 and redefine
R0 at exit path to cut off the data-flow from inside the loop.
- bpf_fill_jump_around_ld_abs:
Jump range is limited to 16 bit. every ld_abs is replaced by 6 insns,
but on arches like arm, ppc etc, there will be one BPF_ZEXT inserted
to extend the error value of the inlined ld_abs sequence which then
contains 7 insns. so, set the dividend to 7 so the testcase could
work on all arches.
- bpf_fill_scale1/bpf_fill_scale2:
Both contains ~1M BPF_ALU32_IMM which will trigger ~1M insn patcher
call because of hi32 randomization later when BPF_F_TEST_RND_HI32 is
set for bpf selftests. Insn patcher is not efficient that 1M call to
it will hang computer. So , change to BPF_ALU64_IMM to avoid hi32
randomization.
Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Jiong Wang [Fri, 24 May 2019 22:25:19 +0000 (23:25 +0100)]
libbpf: add "prog_flags" to bpf_program/bpf_prog_load_attr/bpf_load_program_attr
libbpf doesn't allow passing "prog_flags" during bpf program load in a
couple of load related APIs, "bpf_load_program_xattr", "load_program" and
"bpf_prog_load_xattr".
It makes sense to allow passing "prog_flags" which is useful for
customizing program loading.
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Jiong Wang [Fri, 24 May 2019 22:25:16 +0000 (23:25 +0100)]
bpf: introduce new bpf prog load flags "BPF_F_TEST_RND_HI32"
x86_64 and AArch64 perhaps are two arches that running bpf testsuite
frequently, however the zero extension insertion pass is not enabled for
them because of their hardware support.
It is critical to guarantee the pass correction as it is supposed to be
enabled at default for a couple of other arches, for example PowerPC,
SPARC, arm, NFP etc. Therefore, it would be very useful if there is a way
to test this pass on for example x86_64.
The test methodology employed by this set is "poisoning" useless bits. High
32-bit of a definition is randomized if it is identified as not used by any
later insn. Such randomization is only enabled under testing mode which is
gated by the new bpf prog load flags "BPF_F_TEST_RND_HI32".
Jiong Wang [Fri, 24 May 2019 22:25:15 +0000 (23:25 +0100)]
bpf: verifier: insert zero extension according to analysis result
After previous patches, verifier will mark a insn if it really needs zero
extension on dst_reg.
It is then for back-ends to decide how to use such information to eliminate
unnecessary zero extension code-gen during JIT compilation.
One approach is verifier insert explicit zero extension for those insns
that need zero extension in a generic way, JIT back-ends then do not
generate zero extension for sub-register write at default.
However, only those back-ends which do not have hardware zero extension
want this optimization. Back-ends like x86_64 and AArch64 have hardware
zero extension support that the insertion should be disabled.
This patch introduces new target hook "bpf_jit_needs_zext" which returns
false at default, meaning verifier zero extension insertion is disabled at
default. A back-end could override this hook to return true if it doesn't
have hardware support and want verifier insert zero extension explicitly.
Offload targets do not use this native target hook, instead, they could
get the optimization results using bpf_prog_offload_ops.finalize.
NOTE: arches could have diversified features, it is possible for one arch
to have hardware zero extension support for some sub-register write insns
but not for all. For example, PowerPC, SPARC have zero extended loads, but
not for alu32. So when verifier zero extension insertion enabled, these JIT
back-ends need to peephole insns to remove those zero extension inserted
for insn that actually has hardware zero extension support. The peephole
could be as simple as looking the next insn, if it is a special zero
extension insn then it is safe to eliminate it if the current insn has
hardware zero extension support.
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
We use this new form for doing zero extension for which verifier will
guarantee SRC == DST.
Implications on JIT back-ends when doing code-gen for
BPF_ALU | BPF_MOV | BPF_X:
1. No change if hardware already does zero extension unconditionally for
sub-register write.
2. Otherwise, when seeing imm == 1, just generate insns to clear high
32-bit. No need to generate insns for the move because when imm == 1,
dst_reg is the same as src_reg at the moment.
Interpreter doesn't need change as well. It is doing unconditionally zero
extension for mov32 already.
One helper macro BPF_ZEXT_REG is added to help creating zero extension
insn using this new mov32 variant.
One helper function insn_is_zext is added for checking one insn is an
zero extension on dst. This will be widely used by a few JIT back-ends in
later patches in this set.
Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Jiong Wang [Fri, 24 May 2019 22:25:12 +0000 (23:25 +0100)]
bpf: verifier: mark verified-insn with sub-register zext flag
eBPF ISA specification requires high 32-bit cleared when low 32-bit
sub-register is written. This applies to destination register of ALU32 etc.
JIT back-ends must guarantee this semantic when doing code-gen. x86_64 and
AArch64 ISA has the same semantics, so the corresponding JIT back-end
doesn't need to do extra work.
However, 32-bit arches (arm, x86, nfp etc.) and some other 64-bit arches
(PowerPC, SPARC etc) need to do explicit zero extension to meet this
requirement, otherwise code like the following will fail.
u64_value = (u64) u32_value
... other uses of u64_value
This is because compiler could exploit the semantic described above and
save those zero extensions for extending u32_value to u64_value, these JIT
back-ends are expected to guarantee this through inserting extra zero
extensions which however could be a significant increase on the code size.
Some benchmarks show there could be ~40% sub-register writes out of total
insns, meaning at least ~40% extra code-gen.
One observation is these extra zero extensions are not always necessary.
Take above code snippet for example, it is possible u32_value will never be
casted into a u64, the value of high 32-bit of u32_value then could be
ignored and extra zero extension could be eliminated.
This patch implements this idea, insns defining sub-registers will be
marked when the high 32-bit of the defined sub-register matters. For
those unmarked insns, it is safe to eliminate high 32-bit clearnace for
them.
Algo:
- Split read flags into READ32 and READ64.
- Record index of insn that does sub-register write. Keep the index inside
reg state and update it during verifier insn walking.
- A full register read on a sub-register marks its definition insn as
needing zero extension on dst register.
A new sub-register write overrides the old one.
- When propagating read64 during path pruning, also mark any insn defining
a sub-register that is read in the pruned path as full-register.
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Daniel Borkmann [Fri, 24 May 2019 21:26:49 +0000 (23:26 +0200)]
Merge branch 'bpf-send-sig'
Yonghong Song says:
====================
This patch tries to solve the following specific use case.
Currently, bpf program can already collect stack traces
through kernel function get_perf_callchain()
when certain events happens (e.g., cache miss counter or
cpu clock counter overflows). But such stack traces are
not enough for jitted programs, e.g., hhvm (jited php).
To get real stack trace, jit engine internal data structures
need to be traversed in order to get the real user functions.
bpf program itself may not be the best place to traverse
the jit engine as the traversing logic could be complex and
it is not a stable interface either.
Instead, hhvm implements a signal handler,
e.g. for SIGALARM, and a set of program locations which
it can dump stack traces. When it receives a signal, it will
dump the stack in next such program location.
This patch implements bpf_send_signal() helper to send
a signal to hhvm in real time, resulting in intended stack traces.
Patch #1 implemented the bpf_send_helper() in the kernel.
Patch #2 synced uapi header bpf.h to tools directory.
Patch #3 added a self test which covers tracepoint
and perf_event bpf programs.
Changelogs:
v4 => v5:
. pass the "current" task struct to irq_work as well
since the current task struct may change between
nmi and subsequent irq_work_interrupt.
Discovered by Daniel.
v3 => v4:
. fix one typo and declare "const char *id_path = ..."
to avoid directly use the long string in the func body
in Patch #3.
v2 => v3:
. change the standalone test to be part of prog_tests.
RFC v1 => v2:
. previous version allows to send signal to an arbitrary
pid. This version just sends the signal to current
task to avoid unstable pid and potential races between
sending signals and task state changes for the pid.
====================
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Yonghong Song [Thu, 23 May 2019 21:47:45 +0000 (14:47 -0700)]
bpf: implement bpf_send_signal() helper
This patch tries to solve the following specific use case.
Currently, bpf program can already collect stack traces
through kernel function get_perf_callchain()
when certain events happens (e.g., cache miss counter or
cpu clock counter overflows). But such stack traces are
not enough for jitted programs, e.g., hhvm (jited php).
To get real stack trace, jit engine internal data structures
need to be traversed in order to get the real user functions.
bpf program itself may not be the best place to traverse
the jit engine as the traversing logic could be complex and
it is not a stable interface either.
Instead, hhvm implements a signal handler,
e.g. for SIGALARM, and a set of program locations which
it can dump stack traces. When it receives a signal, it will
dump the stack in next such program location.
Such a mechanism can be implemented in the following way:
. a perf ring buffer is created between bpf program
and tracing app.
. once a particular event happens, bpf program writes
to the ring buffer and the tracing app gets notified.
. the tracing app sends a signal SIGALARM to the hhvm.
But this method could have large delays and causing profiling
results skewed.
This patch implements bpf_send_signal() helper to send
a signal to hhvm in real time, resulting in intended stack traces.
Acked-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Yonghong Song <yhs@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
====================
This patch set adds BTF-to-C dumping APIs to libbpf, allowing to output
a subset of BTF types as a compilable C type definitions. This is useful by
itself, as raw BTF output is not easy to inspect and comprehend. But it's also
a big part of BPF CO-RE (compile once - run everywhere) initiative aimed at
allowing to write relocatable BPF programs, that won't require on-the-host
kernel headers (and would be able to inspect internal kernel structures, not
exposed through kernel headers).
This patch set consists of three groups of patches and one pre-patch, with the
BTF-to-C dumper API depending on the first two groups.
Pre-patch #1 fixes issue with libbpf_internal.h.
btf__parse_elf() API patches:
- patch #2 adds btf__parse_elf() API to libbpf, allowing to load BTF and/or
BTF.ext from ELF file;
- patch #3 utilizies btf__parse_elf() from bpftool for `btf dump file` command;
- patch #4 switches test_btf.c to use btf__parse_elf() to check for presence
of BTF data in object file.
libbpf's internal hashmap patches:
- patch #5 adds resizeable non-thread safe generic hashmap to libbpf;
- patch #6 adds tests for that hashmap;
- patch #7 migrates btf_dedup()'s dedup_table to use hashmap w/ APPEND.
BTF-to-C dumper API patches:
- patch #8 adds btf_dump APIs with all the logic for laying out type
definitions in correct order and emitting C syntax for them;
- patch #9 adds lots of tests for common and quirky parts of C type system;
- patch #10 adds support for C-syntax btf dumping to bpftool;
- patch #11 updates bpftool documentation to mention C-syntax dump option;
- patch #12 update bash-completion for btf dump sub-command.
v1->v2:
- removed unuseful file header (Jakub);
- removed inlines in .c (Jakub);
- added 'format {c|raw}' keyword/option (Jakub);
- re-use i var for iteration in btf_dump_c() (Jakub);
- bumped libbpf version to 0.0.4;
v0->v1:
- fix bug in hashmap__for_each_bucket_entry() not handling empty hashmap;
- removed `btf dump`-specific libbpf logging hook up (Quentin has more generic
patchset);
- change btf__parse_elf() to always load .BTF and return it as a result, with
.BTF.ext being optional and returned through struct btf_ext** arg (Alexei);
- endianness check to use __BYTE_ORDER__ (Alexei);
- bool:1 to __u8:1 in type_aux_state (Alexei);
- added HASHMAP_APPEND strategy to hashmap, changed
hashmap__for_each_key_entry() to also check for key equality during
iteration (multimap iteration for key);
- added new tests for empty hashmap and hashmap as a multimap;
- tried to clarify weak/strong dependency ordering comments (Alexei)
- btf dump test's expected output - support better commenting aproach (Alexei);
- added bash-completion for a new "c" option (Alexei).
====================
Andrii Nakryiko [Fri, 24 May 2019 18:59:03 +0000 (11:59 -0700)]
libbpf: add btf_dump API for BTF-to-C conversion
BTF contains enough type information to allow generating valid
compilable C header w/ correct layout of structs/unions and all the
typedef/enum definitions. This patch adds a new "object" - btf_dump to
facilitate dumping BTF as valid C. btf_dump__dump_type() is the main API
which takes care of dumping out (through user-provided printf-like
callback function) C definitions for given type ID and it's required
dependencies. This allows for not just dumping out entirety of BTF types,
but also selective filtering based on user-provided criterias w/ minimal
set of dependent types.
There is a need for fast point lookups inside libbpf for multiple use
cases (e.g., name resolution for BTF-to-C conversion, by-name lookups in
BTF for upcoming BPF CO-RE relocation support, etc). This patch
implements simple resizable non-thread safe hashmap using single linked
list chains.
Four different insert strategies are supported:
- HASHMAP_ADD - only add key/value if key doesn't exist yet;
- HASHMAP_SET - add key/value pair if key doesn't exist yet; otherwise,
update value;
- HASHMAP_UPDATE - update value, if key already exists; otherwise, do
nothing and return -ENOENT;
- HASHMAP_APPEND - always add key/value pair, even if key already exists.
This turns hashmap into a multimap by allowing multiple values to be
associated with the same key. Most useful read API for such hashmap is
hashmap__for_each_key_entry() iteration. If hashmap__find() is still
used, it will return last inserted key/value entry (first in a bucket
chain).
For HASHMAP_SET and HASHMAP_UPDATE, old key/value pair is returned, so
that calling code can handle proper memory management, if necessary.
Andrii Nakryiko [Fri, 24 May 2019 18:58:57 +0000 (11:58 -0700)]
libbpf: add btf__parse_elf API to load .BTF and .BTF.ext
Loading BTF and BTF.ext from ELF file is a common need. Instead of
requiring every user to re-implement it, let's provide this API from
libbpf itself. It's mostly copy/paste from `bpftool btf dump`
implementation, which will be switched to libbpf's version in next
patch. btf__parse_elf allows to load BTF and optionally BTF.ext.
This is also useful for tests that need to load/work with BTF, loaded
from test ELF files.
Michal Rostecki [Thu, 23 May 2019 12:53:54 +0000 (14:53 +0200)]
selftests: bpf: Move bpf_printk to bpf_helpers.h
bpf_printk is a macro which is commonly used to print out debug messages
in BPF programs and it was copied in many selftests and samples. Since
all of them include bpf_helpers.h, this change moves the macro there.
Signed-off-by: Michal Rostecki <mrostecki@opensuse.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Daniel Borkmann [Thu, 23 May 2019 23:46:26 +0000 (01:46 +0200)]
Merge branch 'bpf-explored-states'
Alexei Starovoitov says:
====================
Convert explored_states array into hash table and use simple hash
to reduce verifier peak memory consumption for programs with bpf2bpf
calls. More details in patch 3.
v1->v2: fixed Jakub's small nit in patch 1
====================
Acked-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
All prune points inside a callee bpf function most likely will have
different callsites. For example, if function foo() is called from
two callsites the half of explored states in all prune points in foo()
will be useless for subsequent walking of one of those callsites.
Fortunately explored_states pruning heuristics keeps the number of states
per prune point small, but walking these states is still a waste of cpu
time when the callsite of the current state is different from the callsite
of the explored state.
To improve pruning logic convert explored_states into hash table and
use simple insn_idx ^ callsite hash to select hash bucket.
This optimization has no effect on programs without bpf2bpf calls
and drastically improves programs with calls.
In the later case it reduces total memory consumption in 1M scale tests
by almost 3 times (peak_states drops from 5752 to 2016).
Care should be taken when comparing the states for equivalency.
Since the same hash bucket can now contain states with different indices
the insn_idx has to be part of verifier_state and compared.
Different hash table sizes and different hash functions were explored,
but the results were not significantly better vs this patch.
They can be improved in the future.
Hit/miss heuristic is not counting index miscompare as a miss.
Otherwise verifier stats become unstable when experimenting
with different hash functions.
Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
split explored_states into prune_point boolean mark
and link list of explored states.
This removes STATE_LIST_MARK hack and allows marks to be separate from states.
Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Daniel Borkmann [Thu, 23 May 2019 14:20:58 +0000 (16:20 +0200)]
Merge branch 'bpf-jmp-seq-limit'
Alexei Starovoitov says:
====================
Patch 1 - jmp sequence limit
Patch 2 - improve existing tests
Patch 3 - add pyperf-based realistic bpf program that takes
advantage of higher limit and use it as a stress test
v1->v2: fixed nit in patch 3. added Andrii's acks
====================
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Adjust scale tests to check for new jmp sequence limit.
BPF_JGT had to be changed to BPF_JEQ because the verifier was
too smart. It tracked the known safe range of R0 values
and pruned the search earlier before hitting exact 8192 limit.
bpf_semi_rand_get() was too (un)?lucky.
k = 0; was missing in bpf_fill_scale2.
It was testing a bit shorter sequence of jumps than intended.
Andrii Nakryiko [Wed, 22 May 2019 17:51:28 +0000 (10:51 -0700)]
libbpf: emit diff of mismatched public API, if any
It's easy to have a mismatch of "intended to be public" vs really
exposed API functions. While Makefile does check for this mismatch, if
it actually occurs it's not trivial to determine which functions are
accidentally exposed. This patch dumps out a diff showing what's not
supposed to be exposed facilitating easier fixing.
Sunil Muthuswamy [Wed, 22 May 2019 23:10:44 +0000 (23:10 +0000)]
hv_sock: perf: loop in send() to maximize bandwidth
Currently, the hv_sock send() iterates once over the buffer, puts data into
the VMBUS channel and returns. It doesn't maximize on the case when there
is a simultaneous reader draining data from the channel. In such a case,
the send() can maximize the bandwidth (and consequently minimize the cpu
cycles) by iterating until the channel is found to be full.
Perf data:
Total Data Transfer: 10GB/iteration
Single threaded reader/writer, Linux hvsocket writer with Windows hvsocket
reader
Packet size: 64KB
CPU sys time was captured using the 'time' command for the writer to send
10GB of data.
'Send Buffer Loop' is with the patch applied.
The values below are over 10 iterations.
|--------------------------------------------------------|
| | Current | Send Buffer Loop |
|--------------------------------------------------------|
| | Throughput | CPU sys | Throughput | CPU sys |
| | (MB/s) | time (s) | (MB/s) | time (s) |
|--------------------------------------------------------|
| Min | 407 | 7.048 | 401 | 5.958 |
|--------------------------------------------------------|
| Max | 455 | 7.563 | 542 | 6.993 |
|--------------------------------------------------------|
| Avg | 440 | 7.411 | 451 | 6.639 |
|--------------------------------------------------------|
| Median | 446 | 7.417 | 447 | 6.761 |
|--------------------------------------------------------|
Observation:
1. The avg throughput doesn't really change much with this change for this
scenario. This is most probably because the bottleneck on throughput is
somewhere else.
2. The average system (or kernel) cpu time goes down by 10%+ with this
change, for the same amount of data transfer.
Signed-off-by: Sunil Muthuswamy <sunilmut@microsoft.com> Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sunil Muthuswamy [Wed, 22 May 2019 22:56:07 +0000 (22:56 +0000)]
hv_sock: perf: Allow the socket buffer size options to influence the actual socket buffers
Currently, the hv_sock buffer size is static and can't scale to the
bandwidth requirements of the application. This change allows the
applications to influence the socket buffer sizes using the SO_SNDBUF and
the SO_RCVBUF socket options.
Few interesting points to note:
1. Since the VMBUS does not allow a resize operation of the ring size, the
socket buffer size option should be set prior to establishing the
connection for it to take effect.
2. Setting the socket option comes with the cost of that much memory being
reserved/allocated by the kernel, for the lifetime of the connection.
Perf data:
Total Data Transfer: 1GB
Single threaded reader/writer
Results below are summarized over 10 iterations.
Signed-off-by: Sunil Muthuswamy <sunilmut@microsoft.com> Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Wed, 22 May 2019 19:11:06 +0000 (12:11 -0700)]
selftests: pmtu: Simplify cleanup and namespace names
The point of the pause-on-fail argument is to leave the setup as is after
a test fails to allow a user to debug why it failed. Move the cleanup
after posting the result to the user to make it so.
Random names for the namespaces are not user friendly when trying to
debug a failure. Make them simpler and more direct for the tests. Run
cleanup at the beginning to ensure they are cleaned up if they already
exist.
Remove cleanup_done. There is no harm in doing cleanup twice; just
ignore any errors related to not existing - which is already done.
Signed-off-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Wed, 22 May 2019 19:09:16 +0000 (12:09 -0700)]
selftests: fib-onlink: Make quiet by default
Add VERBOSE argument to fib-onlink-tests.sh and make output quiet by
default. Add getopt parsing of inputs and support for -v (verbose) and
-p (pause on fail).
Signed-off-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Wed, 22 May 2019 19:07:43 +0000 (12:07 -0700)]
net: Set strict_start_type for routes and rules
New userspace on an older kernel can send unknown and unsupported
attributes resulting in an incompelete config which is almost
always wrong for routing (few exceptions are passthrough settings
like the protocol that installed the route).
Set strict_start_type in the policies for IPv4 and IPv6 routes and
rules to detect new, unsupported attributes and fail the route add.
Signed-off-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
net: Export functions for nexthop code
This set exports ipv4 and ipv6 fib functions for use by the nexthop
code. It also adds new ones to send route notifications if a nexthop
configuration changes.
v2
- repost of patches dropped at the end of the last dev window
added patch 8 which exports nh_update_mtu since it is inline with
the other patches
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
projects / kernel.git / log