| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Stanislav Fomichev says:
====================
This series implements new lsm flavor for attaching per-cgroup programs to
existing lsm hooks. The cgroup is taken out of 'current', unless
the first argument of the hook is 'struct socket'. In this case,
the cgroup association is taken out of socket. The attachment
looks like a regular per-cgroup attachment: we add new BPF_LSM_CGROUP
attach type which, together with attach_btf_id, signals per-cgroup lsm.
Behind the scenes, we allocate trampoline shim program and
attach to lsm. This program looks up cgroup from current/socket
and runs cgroup's effective prog array. The rest of the per-cgroup BPF
stays the same: hierarchy, local storage, retval conventions
(return 1 == success).
Current limitations:
* haven't considered sleepable bpf; can be extended later on
* not sure the verifier does the right thing with null checks;
see latest selftest for details
* total of 10 (global) per-cgroup LSM attach points
v11:
- Martin: address selftest memory & fd leaks
- Martin: address moving into root (instead have another temp leaf cgroup)
- Martin: move tools/include/uapi/linux/bpf.h change from libbpf patch
into 'sync tools' patch
v10:
- Martin: reword commit message, drop outdated items
- Martin: remove rcu_real_lock from __cgroup_bpf_run_lsm_current
- Martin: remove CONFIG_BPF_LSM from cgroup_bpf_release
- Martin: fix leaking shim reference in bpf_cgroup_link_release
- Martin: WARN_ON_ONCE for bpf_trampoline_lookup in bpf_trampoline_unlink_cgroup_shim
- Martin: sync tools/include/linux/btf_ids.h
- Martin: move progs/flags closer to the places where they are used in __cgroup_bpf_query
- Martin: remove sk_clone_security & sctp_bind_connect from bpf_lsm_locked_sockopt_hooks
- Martin: try to determine vmlinux btf_id in bpftool
- Martin: update tools header in a separate commit
- Quentin: do libbpf_find_kernel_btf from the ops that need it
- lkp@intel.com: another build failure
v9:
Major change since last version is the switch to bpf_setsockopt to
change the socket state instead of letting the progs poke socket directly.
This, in turn, highlights the challenge that we need to care about whether
the socket is locked or not when we call bpf_setsockopt. (with my original
example selftest, the hooks are running early in the init phase for this
not to matter).
For now, I've added two btf id lists:
* hooks where we know the socket is locked and it's safe to call bpf_setsockopt
* hooks where we know the socket is _not_ locked, but the hook works on
the socket that's not yet exposed to userspace so it should be safe
(for this mode, special new set of bpf_{s,g}etsockopt helpers
is added; they don't have sock_owned_by_me check)
Going forward, for the rest of the hooks, this might be a good motivation
to expand lsm cgroup to support sleeping bpf and allow the callers to
lock/unlock sockets or have a new bpf_setsockopt variant that does the
locking.
- ifdef around cleanup in cgroup_bpf_release
- Andrii: a few nits in libbpf patches
- Martin: remove unused btf_id_set_index
- Martin: bring back refcnt for cgroup_atype
- Martin: make __cgroup_bpf_query a bit more readable
- Martin: expose dst_prog->aux->attach_btf as attach_btf_obj_id as well
- Martin: reorg check_return_code path for BPF_LSM_CGROUP
- Martin: return directly from check_helper_call (instead of goto err)
- Martin: add note to new warning in check_return_code, print only for void hooks
- Martin: remove confusing shim reuse
- Martin: use bpf_{s,g}etsockopt instead of poking into socket data
- Martin: use CONFIG_CGROUP_BPF in bpf_prog_alloc_no_stats/bpf_prog_free_deferred
v8:
- CI: fix compile issue
- CI: fix broken bpf_cookie
- Yonghong: remove __bpf_trampoline_unlink_prog comment
- Yonghong: move cgroup_atype around to fill the gap
- Yonghong: make bpf_lsm_find_cgroup_shim void
- Yonghong: rename regs to args
- Yonghong: remove if(current) check
- Martin: move refcnt into bpf_link
- Martin: move shim management to bpf_link ops
- Martin: use cgroup_atype for shim only
- Martin: go back to arrays for managing cgroup_atype(s)
- Martin: export bpf_obj_id(aux->attach_btf)
- Andrii: reorder SEC_DEF("lsm_cgroup+")
- Andrii: OPTS_SET instead of OPTS_HAS
- Andrii: rename attach_btf_func_id
- Andrii: move into 1.0 map
v7:
- there were a lot of comments last time, hope I didn't forget anything,
some of the bigger ones:
- Martin: use/extend BTF_SOCK_TYPE_SOCKET
- Martin: expose bpf_set_retval
- Martin: reject 'return 0' at the verifier for 'void' hooks
- Martin: prog_query returns all BPF_LSM_CGROUP, prog_info
returns attach_btf_func_id
- Andrii: split libbpf changes
- Andrii: add field access test to test_progs, not test_verifier (still
using asm though)
- things that I haven't addressed, stating them here explicitly, let
me know if some of these are still problematic:
1. Andrii: exposing only link-based api: seems like the changes
to support non-link-based ones are minimal, couple of lines,
so seems like it worth having it?
2. Alexei: applying cgroup_atype for all cgroup hooks, not only
cgroup lsm: looks a bit harder to apply everywhere that I
originally thought; with lsm cgroup, we have a shim_prog pointer where
we store cgroup_atype; for non-lsm programs, we don't have a
trace program where to store it, so we still need some kind
of global table to map from "static" hook to "dynamic" slot.
So I'm dropping this "can be easily extended" clause from the
description for now. I have converted this whole machinery
to an RCU-managed list to remove synchronize_rcu().
- also note that I had to introduce new bpf_shim_tramp_link and
moved refcnt there; we need something to manage new bpf_tramp_link
v6:
- remove active count & stats for shim program (Martin KaFai Lau)
- remove NULL/error check for btf_vmlinux (Martin)
- don't check cgroup_atype in bpf_cgroup_lsm_shim_release (Martin)
- use old_prog (instead of passed one) in __cgroup_bpf_detach (Martin)
- make sure attach_btf_id is the same in __cgroup_bpf_replace (Martin)
- enable cgroup local storage and test it (Martin)
- properly implement prog query and add bpftool & tests (Martin)
- prohibit non-shared cgroup storage mode for BPF_LSM_CGROUP (Martin)
v5:
- __cgroup_bpf_run_lsm_socket remove NULL sock/sk checks (Martin KaFai Lau)
- __cgroup_bpf_run_lsm_{socket,current} s/prog/shim_prog/ (Martin)
- make sure bpf_lsm_find_cgroup_shim works for hooks without args (Martin)
- __cgroup_bpf_attach make sure attach_btf_id is the same when replacing (Martin)
- call bpf_cgroup_lsm_shim_release only for LSM_CGROUP (Martin)
- drop BPF_LSM_CGROUP from bpf_attach_type_to_tramp (Martin)
- drop jited check from cgroup_shim_find (Martin)
- new patch to convert cgroup_bpf to hlist_node (Jakub Sitnicki)
- new shim flavor for 'struct sock' + list of exceptions (Martin)
v4:
- fix build when jit is on but syscall is off
v3:
- add BPF_LSM_CGROUP to bpftool
- use simple int instead of refcnt_t (to avoid use-after-free
false positive)
v2:
- addressed build bot failures
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Functional test that exercises the following:
1. apply default sk_priority policy
2. permit TX-only AF_PACKET socket
3. cgroup attach/detach/replace
4. reusing trampoline shim
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-12-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
$ bpftool --nomount prog loadall $KDIR/tools/testing/selftests/bpf/lsm_cgroup.o /sys/fs/bpf/x
$ bpftool cgroup attach /sys/fs/cgroup lsm_cgroup pinned /sys/fs/bpf/x/socket_alloc
$ bpftool cgroup attach /sys/fs/cgroup lsm_cgroup pinned /sys/fs/bpf/x/socket_bind
$ bpftool cgroup attach /sys/fs/cgroup lsm_cgroup pinned /sys/fs/bpf/x/socket_clone
$ bpftool cgroup attach /sys/fs/cgroup lsm_cgroup pinned /sys/fs/bpf/x/socket_post_create
$ bpftool cgroup tree
CgroupPath
ID AttachType AttachFlags Name
/sys/fs/cgroup
6 lsm_cgroup socket_post_create bpf_lsm_socket_post_create
8 lsm_cgroup socket_bind bpf_lsm_socket_bind
10 lsm_cgroup socket_alloc bpf_lsm_sk_alloc_security
11 lsm_cgroup socket_clone bpf_lsm_inet_csk_clone
$ bpftool cgroup detach /sys/fs/cgroup lsm_cgroup pinned /sys/fs/bpf/x/socket_post_create
$ bpftool cgroup tree
CgroupPath
ID AttachType AttachFlags Name
/sys/fs/cgroup
8 lsm_cgroup socket_bind bpf_lsm_socket_bind
10 lsm_cgroup socket_alloc bpf_lsm_sk_alloc_security
11 lsm_cgroup socket_clone bpf_lsm_inet_csk_clone
Reviewed-by: Quentin Monnet <quentin@isovalent.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-11-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Implement bpf_prog_query_opts as a more expendable version of
bpf_prog_query. Expose new prog_attach_flags and attach_btf_func_id as
well:
* prog_attach_flags is a per-program attach_type; relevant only for
lsm cgroup program which might have different attach_flags
per attach_btf_id
* attach_btf_func_id is a new field expose for prog_query which
specifies real btf function id for lsm cgroup attachments
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-10-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
lsm_cgroup/ is the prefix for BPF_LSM_CGROUP.
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-9-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Has been slowly getting out of sync, let's update it.
resolve_btfids usage has been updated to match the header changes.
Also bring new parts of tools/include/uapi/linux/bpf.h.
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-8-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
I don't see how to make it nice without introducing btf id lists
for the hooks where these helpers are allowed. Some LSM hooks
work on the locked sockets, some are triggering early and
don't grab any locks, so have two lists for now:
1. LSM hooks which trigger under socket lock - minority of the hooks,
but ideal case for us, we can expose existing BTF-based helpers
2. LSM hooks which trigger without socket lock, but they trigger
early in the socket creation path where it should be safe to
do setsockopt without any locks
3. The rest are prohibited. I'm thinking that this use-case might
be a good gateway to sleeping lsm cgroup hooks in the future.
We can either expose lock/unlock operations (and add tracking
to the verifier) or have another set of bpf_setsockopt
wrapper that grab the locks and might sleep.
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-7-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
We have two options:
1. Treat all BPF_LSM_CGROUP the same, regardless of attach_btf_id
2. Treat BPF_LSM_CGROUP+attach_btf_id as a separate hook point
I was doing (2) in the original patch, but switching to (1) here:
* bpf_prog_query returns all attached BPF_LSM_CGROUP programs
regardless of attach_btf_id
* attach_btf_id is exported via bpf_prog_info
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-6-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Previous patch adds 1:1 mapping between all 211 LSM hooks
and bpf_cgroup program array. Instead of reserving a slot per
possible hook, reserve 10 slots per cgroup for lsm programs.
Those slots are dynamically allocated on demand and reclaimed.
struct cgroup_bpf {
struct bpf_prog_array * effective[33]; /* 0 264 */
/* --- cacheline 4 boundary (256 bytes) was 8 bytes ago --- */
struct hlist_head progs[33]; /* 264 264 */
/* --- cacheline 8 boundary (512 bytes) was 16 bytes ago --- */
u8 flags[33]; /* 528 33 */
/* XXX 7 bytes hole, try to pack */
struct list_head storages; /* 568 16 */
/* --- cacheline 9 boundary (576 bytes) was 8 bytes ago --- */
struct bpf_prog_array * inactive; /* 584 8 */
struct percpu_ref refcnt; /* 592 16 */
struct work_struct release_work; /* 608 72 */
/* size: 680, cachelines: 11, members: 7 */
/* sum members: 673, holes: 1, sum holes: 7 */
/* last cacheline: 40 bytes */
};
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-5-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Allow attaching to lsm hooks in the cgroup context.
Attaching to per-cgroup LSM works exactly like attaching
to other per-cgroup hooks. New BPF_LSM_CGROUP is added
to trigger new mode; the actual lsm hook we attach to is
signaled via existing attach_btf_id.
For the hooks that have 'struct socket' or 'struct sock' as its first
argument, we use the cgroup associated with that socket. For the rest,
we use 'current' cgroup (this is all on default hierarchy == v2 only).
Note that for some hooks that work on 'struct sock' we still
take the cgroup from 'current' because some of them work on the socket
that hasn't been properly initialized yet.
Behind the scenes, we allocate a shim program that is attached
to the trampoline and runs cgroup effective BPF programs array.
This shim has some rudimentary ref counting and can be shared
between several programs attaching to the same lsm hook from
different cgroups.
Note that this patch bloats cgroup size because we add 211
cgroup_bpf_attach_type(s) for simplicity sake. This will be
addressed in the subsequent patch.
Also note that we only add non-sleepable flavor for now. To enable
sleepable use-cases, bpf_prog_run_array_cg has to grab trace rcu,
shim programs have to be freed via trace rcu, cgroup_bpf.effective
should be also trace-rcu-managed + maybe some other changes that
I'm not aware of.
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-4-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
This lets us reclaim some space to be used by new cgroup lsm slots.
Before:
struct cgroup_bpf {
struct bpf_prog_array * effective[23]; /* 0 184 */
/* --- cacheline 2 boundary (128 bytes) was 56 bytes ago --- */
struct list_head progs[23]; /* 184 368 */
/* --- cacheline 8 boundary (512 bytes) was 40 bytes ago --- */
u32 flags[23]; /* 552 92 */
/* XXX 4 bytes hole, try to pack */
/* --- cacheline 10 boundary (640 bytes) was 8 bytes ago --- */
struct list_head storages; /* 648 16 */
struct bpf_prog_array * inactive; /* 664 8 */
struct percpu_ref refcnt; /* 672 16 */
struct work_struct release_work; /* 688 32 */
/* size: 720, cachelines: 12, members: 7 */
/* sum members: 716, holes: 1, sum holes: 4 */
/* last cacheline: 16 bytes */
};
After:
struct cgroup_bpf {
struct bpf_prog_array * effective[23]; /* 0 184 */
/* --- cacheline 2 boundary (128 bytes) was 56 bytes ago --- */
struct hlist_head progs[23]; /* 184 184 */
/* --- cacheline 5 boundary (320 bytes) was 48 bytes ago --- */
u8 flags[23]; /* 368 23 */
/* XXX 1 byte hole, try to pack */
/* --- cacheline 6 boundary (384 bytes) was 8 bytes ago --- */
struct list_head storages; /* 392 16 */
struct bpf_prog_array * inactive; /* 408 8 */
struct percpu_ref refcnt; /* 416 16 */
struct work_struct release_work; /* 432 72 */
/* size: 504, cachelines: 8, members: 7 */
/* sum members: 503, holes: 1, sum holes: 1 */
/* last cacheline: 56 bytes */
};
Suggested-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-3-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
I'll be adding lsm cgroup specific helpers that grab
trampoline mutex.
No functional changes.
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-2-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Andrii Nakryiko says:
====================
This patch set removes all the deprecated APIs in preparation for 1.0 release.
It also makes libbpf_set_strict_mode() a no-op (but keeps it to let per-1.0
applications buildable and dynamically linkable against libbpf 1.0 if they
were already libbpf-1.0 ready) and starts enforcing all the
behaviors that were previously opt-in through libbpf_set_strict_mode().
xsk.{c,h} parts that are now properly provided by libxdp ([0]) are still used
by xdpxceiver.c in selftest/bpf, so I've moved xsk.{c,h} with barely any
changes to under selftests/bpf.
Other than that, apart from removing all the LIBBPF_DEPRECATED-marked APIs,
there is a bunch of internal clean ups allowed by that. I've also "restored"
libbpf.map inheritance chain while removing all the deprecated APIs. I think
that's the right way to do this, as applications using libbpf as shared
library but not relying on any deprecated APIs (i.e., "good citizens" that
prepared for libbpf 1.0 release ahead of time to minimize disruption) should
be able to link both against 0.x and 1.x versions. Either way, it doesn't seem
to break anything and preserve a history on when each "surviving" API was
added.
[0] https://github.com/xdp-project/xdp-tools/tree/master/lib/libxdp
v1->v2:
- rebase on latest bpf-next now that Jiri's perf patches landed;
- fix xsk.o dependency in Makefile to ensure libbpf headers are installed
reliably.
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Seems like we missed to add 2 APIs to libbpf.map and another API was
misspelled. Fix it in libbpf.map.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-16-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Remove support for legacy features and behaviors that previously had to
be disabled by calling libbpf_set_strict_mode():
- legacy BPF map definitions are not supported now;
- RLIMIT_MEMLOCK auto-setting, if necessary, is always on (but see
libbpf_set_memlock_rlim());
- program name is used for program pinning (instead of section name);
- cleaned up error returning logic;
- entry BPF programs should have SEC() always.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-15-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Libbpf 1.0 stops support legacy-style BPF map definitions. Selftests has
been migrated away from using legacy BPF map definitions except for two
selftests, to make sure that legacy functionality still worked in
pre-1.0 libbpf. Now it's time to let those tests go as libbpf 1.0 is
imminent.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-14-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Get rid of sloppy prefix logic and remove deprecated xdp_{devmap,cpumap}
sections.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-13-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Clean up internals that had to deal with the possibility of
multi-instance bpf_programs. Libbpf 1.0 doesn't support this, so all
this is not necessary now and can be simplified.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-12-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Keep the LIBBPF_DEPRECATED_SINCE macro "framework" for future
deprecations, but clean up 0.x related helper macros.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-11-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Remove all the public APIs that are related to creating multi-instance
bpf_programs through custom preprocessing callback and generally working
with them.
Also remove all the bpf_{object,map,program}__[set_]priv() APIs.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-10-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Remove a bunch of high-level bpf_object/bpf_map/bpf_program related
APIs. All the APIs related to private per-object/map/prog state,
program preprocessing callback, and generally everything multi-instance
related is removed in a separate patch.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-9-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Remove prog_info_linear-related APIs previously used by perf.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-8-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Remove deprecated perfbuf APIs and clean up opts structs.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-7-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Get rid of deprecated BTF-related APIs.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-6-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Get rid of deprecated feature-probing APIs.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-5-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Get rid of deprecated bpf_set_link*() and bpf_get_link*() APIs.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-4-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Drop low-level APIs as well as high-level (and very confusingly named)
BPF object loading bpf_prog_load_xattr() and bpf_prog_load_deprecated()
APIs.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Remove deprecated xsk APIs from libbpf. But given we have selftests
relying on this, move those files (with minimal adjustments to make them
compilable) under selftests/bpf.
We also remove all the removed APIs from libbpf.map, while overall
keeping version inheritance chain, as most APIs are backwards
compatible so there is no need to reassign them as LIBBPF_1.0.0 versions.
Cc: Magnus Karlsson <magnus.karlsson@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220627211527.2245459-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
syzbot reproduced the bug ...
BUG: sleeping function called from invalid context at kernel/workqueue.c:3010
... with the following stack trace fragment ...
start_flush_work kernel/workqueue.c:3010 [inline]
__flush_work+0x109/0xb10 kernel/workqueue.c:3074
__cancel_work_timer+0x3f9/0x570 kernel/workqueue.c:3162
sk_psock_stop+0x4cb/0x630 net/core/skmsg.c:802
sock_map_destroy+0x333/0x760 net/core/sock_map.c:1581
inet_csk_destroy_sock+0x196/0x440 net/ipv4/inet_connection_sock.c:1130
__tcp_close+0xd5b/0x12b0 net/ipv4/tcp.c:2897
tcp_close+0x29/0xc0 net/ipv4/tcp.c:2909
... introduced by d8616ee2affc. Do a quick trace of the code path and the
bug is obvious:
inet_csk_destroy_sock(sk)
sk_prot->destroy(sk); <--- sock_map_destroy
sk_psock_stop(, true); <--- true so cancel workqueue
cancel_work_sync() <--- splat, because *_bh_disable()
We can not call cancel_work_sync() from inside destroy path. So mark
the sk_psock_stop call to skip this cancel_work_sync(). This will avoid
the BUG, but means we may run sk_psock_backlog after or during the
destroy op. We zapped the ingress_skb queue in sk_psock_stop (safe to
do with local_bh_disable) so its empty and the sk_psock_backlog work
item will not find any pkts to process here. However, because we are
not going to wait for it or clear its ->state its possible it kicks off
or is already running. This should be 'safe' up until psock drops its
refcnt to psock->sk. The sock_put() that drops this reference is only
done at psock destroy time from sk_psock_destroy(). This is done through
workqueue when sk_psock_drop() is called on psock refnt reaches 0.
And importantly sk_psock_destroy() does a cancel_work_sync(). So trivial
fix works.
I've had hit or miss luck reproducing this caught it once or twice with
the provided reproducer when running with many runners. However, syzkaller
is very good at reproducing so relying on syzkaller to verify fix.
Fixes: d8616ee2affc ("bpf, sockmap: Fix sk->sk_forward_alloc warn_on in sk_stream_kill_queues")
Reported-by: syzbot+140186ceba0c496183bc@syzkaller.appspotmail.com
Suggested-by: Hillf Danton <hdanton@sina.com>
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Wang Yufen <wangyufen@huawei.com>
Link: https://lore.kernel.org/bpf/20220628035803.317876-1-john.fastabend@gmail.com
|
|
BPF type compatibility checks (bpf_core_types_are_compat()) are
currently duplicated between kernel and user space. That's a historical
artifact more than intentional doing and can lead to subtle bugs where
one implementation is adjusted but another is forgotten.
That happened with the enum64 work, for example, where the libbpf side
was changed (commit 23b2a3a8f63a ("libbpf: Add enum64 relocation
support")) to use the btf_kind_core_compat() helper function but the
kernel side was not (commit 6089fb325cf7 ("bpf: Add btf enum64
support")).
This patch addresses both the duplication issue, by merging both
implementations and moving them into relo_core.c, and fixes the alluded
to kind check (by giving preference to libbpf's already adjusted logic).
For discussion of the topic, please refer to:
https://lore.kernel.org/bpf/CAADnVQKbWR7oarBdewgOBZUPzryhRYvEbkhyPJQHHuxq=0K1gw@mail.gmail.com/T/#mcc99f4a33ad9a322afaf1b9276fb1f0b7add9665
Changelog:
v1 -> v2:
- limited libbpf recursion limit to 32
- changed name to __bpf_core_types_are_compat
- included warning previously present in libbpf version
- merged kernel and user space changes into a single patch
Signed-off-by: Daniel Müller <deso@posteo.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20220623182934.2582827-1-deso@posteo.net
|
|
A minor typo fix to include "| BPF_LD" into its previous
code phrase:
``BPF_IND`` | BPF_LD --> ``BPF_IND | BPF_LD``
Signed-off-by: Shahab Vahedi <shahab@synopsys.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/b6120b31-3d1d-bf2d-2f2a-aa768d91257b@synopsys.com
|
|
Jiri Olsa says:
====================
hi,
sending change we discussed some time ago [1] to get rid of
some deprecated functions we use in perf prologue code.
Despite the gloomy discussion I think the final code does
not look that bad ;-)
This patchset removes following libbpf functions from perf:
bpf_program__set_prep
bpf_program__nth_fd
struct bpf_prog_prep_result
v5 changes:
- squashed patches together so we don't break bisection [Arnaldo]
v4 changes:
- fix typo [Andrii]
v3 changes:
- removed R0/R1 zero init in libbpf_prog_prepare_load_fn,
because it's not needed [Andrii]
- rebased/post on top of bpf-next/master which now has
all the needed perf/core changes
v2 changes:
- use fallback section prog handler, so we don't need to
use section prefix [Andrii]
- realloc prog->insns array in bpf_program__set_insns [Andrii]
- squash patch 1 from previous version with
bpf_program__set_insns change [Daniel]
- patch 3 already merged [Arnaldo]
- added more comments
thanks,
jirka
[1] https://lore.kernel.org/bpf/CAEf4BzaiBO3_617kkXZdYJ8hS8YF--ZLgapNbgeeEJ-pY0H88g@mail.gmail.com/
====================
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
|
|
Some functions we use for bpf prologue generation are going to be
deprecated. This change reworks current code not to use them.
We need to replace following functions/struct:
bpf_program__set_prep
bpf_program__nth_fd
struct bpf_prog_prep_result
Currently we use bpf_program__set_prep to hook perf callback before
program is loaded and provide new instructions with the prologue.
We replace this function/ality by taking instructions for specific
program, attaching prologue to them and load such new ebpf programs
with prologue using separate bpf_prog_load calls (outside libbpf
load machinery).
Before we can take and use program instructions, we need libbpf to
actually load it. This way we get the final shape of its instructions
with all relocations and verifier adjustments).
There's one glitch though.. perf kprobe program already assumes
generated prologue code with proper values in argument registers,
so loading such program directly will fail in the verifier.
That's where the fallback pre-load handler fits in and prepends
the initialization code to the program. Once such program is loaded
we take its instructions, cut off the initialization code and prepend
the prologue.
I know.. sorry ;-)
To have access to the program when loading this patch adds support to
register 'fallback' section handler to take care of perf kprobe programs.
The fallback means that it handles any section definition besides the
ones that libbpf handles.
The handler serves two purposes:
- allows perf programs to have special arguments in section name
- allows perf to use pre-load callback where we can attach init
code (zeroing all argument registers) to each perf program
Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Link: https://lore.kernel.org/bpf/20220616202214.70359-2-jolsa@kernel.org
|
|
This test verifies that bpf_loop() inlining works as expected when
address of `env->prog` is updated. This address is updated upon BPF
program reallocation.
Reallocation is handled by bpf_prog_realloc(), which reuses old memory
if page boundary is not crossed. The value of `len` in the test is
chosen to cross this boundary on bpf_loop() patching.
Verify that the use-after-free bug in inline_bpf_loop() reported by
Dan Carpenter is fixed.
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20220624020613.548108-3-eddyz87@gmail.com
|
|
As reported by Dan Carpenter, the following statements in inline_bpf_loop()
might cause a use-after-free bug:
struct bpf_prog *new_prog;
// ...
new_prog = bpf_patch_insn_data(env, position, insn_buf, *cnt);
// ...
env->prog->insnsi[call_insn_offset].imm = callback_offset;
The bpf_patch_insn_data() might free the memory used by env->prog.
Fixes: 1ade23711971 ("bpf: Inline calls to bpf_loop when callback is known")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20220624020613.548108-2-eddyz87@gmail.com
|
|
Enhance readability a bit.
Signed-off-by: Simon Wang <wangchuanguo@inspur.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20220622031923.65692-1-wangchuanguo@inspur.com
|
|
test_sock_fields__detach() got called with a null pointer here when one
of the CHECKs or ASSERTs up to the test_sock_fields__open_and_load()
call resulted in a jump to the "done" label.
A skeletons *__detach() is not safe to call with a null pointer, though.
This led to a segfault.
Go the easy route and only call test_sock_fields__destroy() which is
null-pointer safe and includes detaching.
Came across this while looking[1] to introduce the usage of
bpf_tcp_helpers.h (included in progs/test_sock_fields.c) together with
vmlinux.h.
[1] https://lore.kernel.org/bpf/629bc069dd807d7ac646f836e9dca28bbc1108e2.camel@mailbox.tu-berlin.de/
Fixes: 8f50f16ff39d ("selftests/bpf: Extend verifier and bpf_sock tests for dst_port loads")
Signed-off-by: Jörn-Thorben Hinz <jthinz@mailbox.tu-berlin.de>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20220621070116.307221-1-jthinz@mailbox.tu-berlin.de
|
|
Jörn-Thorben Hinz says:
====================
This series corrects some inconveniences for a BPF TCP CC that
implements and uses tcp_congestion_ops.cong_control(). Until now, such a
CC did not have all necessary write access to struct sock and
unnecessarily needed to implement cong_avoid().
v4:
- Remove braces around single statements after if
- Don’t check pointer passed to bpf_link__destroy()
v3:
- Add a selftest writing sk_pacing_*
- Add a selftest with incomplete tcp_congestion_ops
- Add a selftest with unsupported get_info()
- Remove an unused variable
- Reword a comment about reg() in bpf_struct_ops_map_update_elem()
v2:
- Drop redundant check for required functions and just rely on
tcp_register_congestion_control() (Martin KaFai Lau)
====================
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Test whether a TCP CC implemented in BPF providing get_info() is
rejected correctly. get_info() is unsupported in a BPF CC. The check for
required functions in a BPF CC has moved, this test ensures unsupported
functions are still rejected correctly.
Signed-off-by: Jörn-Thorben Hinz <jthinz@mailbox.tu-berlin.de>
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/r/20220622191227.898118-6-jthinz@mailbox.tu-berlin.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Test whether a TCP CC implemented in BPF providing neither cong_avoid()
nor cong_control() is correctly rejected. This check solely depends on
tcp_register_congestion_control() now, which is invoked during
bpf_map__attach_struct_ops().
Signed-off-by: Jörn-Thorben Hinz <jthinz@mailbox.tu-berlin.de>
Link: https://lore.kernel.org/r/20220622191227.898118-5-jthinz@mailbox.tu-berlin.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Test whether a TCP CC implemented in BPF is allowed to write
sk_pacing_rate and sk_pacing_status in struct sock. This is needed when
cong_control() is implemented and used.
Signed-off-by: Jörn-Thorben Hinz <jthinz@mailbox.tu-berlin.de>
Link: https://lore.kernel.org/r/20220622191227.898118-4-jthinz@mailbox.tu-berlin.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Remove the check for required and optional functions in a struct
tcp_congestion_ops from bpf_tcp_ca.c. Rely on
tcp_register_congestion_control() to reject a BPF CC that does not
implement all required functions, as it will do for a non-BPF CC.
When a CC implements tcp_congestion_ops.cong_control(), the alternate
cong_avoid() is not in use in the TCP stack. Previously, a BPF CC was
still forced to implement cong_avoid() as a no-op since it was
non-optional in bpf_tcp_ca.c.
Signed-off-by: Jörn-Thorben Hinz <jthinz@mailbox.tu-berlin.de>
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/r/20220622191227.898118-3-jthinz@mailbox.tu-berlin.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
A CC that implements tcp_congestion_ops.cong_control() should be able to
control sk_pacing_rate and set sk_pacing_status, since
tcp_update_pacing_rate() is never called in this case. A built-in CC or
one from a kernel module is already able to write to both struct sock
members. For a BPF program, write access has not been allowed, yet.
Signed-off-by: Jörn-Thorben Hinz <jthinz@mailbox.tu-berlin.de>
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/r/20220622191227.898118-2-jthinz@mailbox.tu-berlin.de
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
The prototype of .features is netdev_features_t, it should use
NETIF_F_LLTX and NETIF_F_HW_VLAN_STAG_TX, not NETIF_F_LLTX_BIT
and NETIF_F_HW_VLAN_STAG_TX_BIT.
Fixes: cf204a718357 ("bpf, testing: Introduce 'gso_linear_no_head_frag' skb_segment test")
Signed-off-by: Jian Shen <shenjian15@huawei.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/r/20220622135002.8263-1-shenjian15@huawei.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Add a benchmarks to demonstrate the performance cliff for local_storage
get as the number of local_storage maps increases beyond current
local_storage implementation's cache size.
"sequential get" and "interleaved get" benchmarks are added, both of
which do many bpf_task_storage_get calls on sets of task local_storage
maps of various counts, while considering a single specific map to be
'important' and counting task_storage_gets to the important map
separately in addition to normal 'hits' count of all gets. Goal here is
to mimic scenario where a particular program using one map - the
important one - is running on a system where many other local_storage
maps exist and are accessed often.
While "sequential get" benchmark does bpf_task_storage_get for map 0, 1,
..., {9, 99, 999} in order, "interleaved" benchmark interleaves 4
bpf_task_storage_gets for the important map for every 10 map gets. This
is meant to highlight performance differences when important map is
accessed far more frequently than non-important maps.
A "hashmap control" benchmark is also included for easy comparison of
standard bpf hashmap lookup vs local_storage get. The benchmark is
similar to "sequential get", but creates and uses BPF_MAP_TYPE_HASH
instead of local storage. Only one inner map is created - a hashmap
meant to hold tid -> data mapping for all tasks. Size of the hashmap is
hardcoded to my system's PID_MAX_LIMIT (4,194,304). The number of these
keys which are actually fetched as part of the benchmark is
configurable.
Addition of this benchmark is inspired by conversation with Alexei in a
previous patchset's thread [0], which highlighted the need for such a
benchmark to motivate and validate improvements to local_storage
implementation. My approach in that series focused on improving
performance for explicitly-marked 'important' maps and was rejected
with feedback to make more generally-applicable improvements while
avoiding explicitly marking maps as important. Thus the benchmark
reports both general and important-map-focused metrics, so effect of
future work on both is clear.
Regarding the benchmark results. On a powerful system (Skylake, 20
cores, 256gb ram):
Hashmap Control
===============
num keys: 10
hashmap (control) sequential get: hits throughput: 20.900 ± 0.334 M ops/s, hits latency: 47.847 ns/op, important_hits throughput: 20.900 ± 0.334 M ops/s
num keys: 1000
hashmap (control) sequential get: hits throughput: 13.758 ± 0.219 M ops/s, hits latency: 72.683 ns/op, important_hits throughput: 13.758 ± 0.219 M ops/s
num keys: 10000
hashmap (control) sequential get: hits throughput: 6.995 ± 0.034 M ops/s, hits latency: 142.959 ns/op, important_hits throughput: 6.995 ± 0.034 M ops/s
num keys: 100000
hashmap (control) sequential get: hits throughput: 4.452 ± 0.371 M ops/s, hits latency: 224.635 ns/op, important_hits throughput: 4.452 ± 0.371 M ops/s
num keys: 4194304
hashmap (control) sequential get: hits throughput: 3.043 ± 0.033 M ops/s, hits latency: 328.587 ns/op, important_hits throughput: 3.043 ± 0.033 M ops/s
Local Storage
=============
num_maps: 1
local_storage cache sequential get: hits throughput: 47.298 ± 0.180 M ops/s, hits latency: 21.142 ns/op, important_hits throughput: 47.298 ± 0.180 M ops/s
local_storage cache interleaved get: hits throughput: 55.277 ± 0.888 M ops/s, hits latency: 18.091 ns/op, important_hits throughput: 55.277 ± 0.888 M ops/s
num_maps: 10
local_storage cache sequential get: hits throughput: 40.240 ± 0.802 M ops/s, hits latency: 24.851 ns/op, important_hits throughput: 4.024 ± 0.080 M ops/s
local_storage cache interleaved get: hits throughput: 48.701 ± 0.722 M ops/s, hits latency: 20.533 ns/op, important_hits throughput: 17.393 ± 0.258 M ops/s
num_maps: 16
local_storage cache sequential get: hits throughput: 44.515 ± 0.708 M ops/s, hits latency: 22.464 ns/op, important_hits throughput: 2.782 ± 0.044 M ops/s
local_storage cache interleaved get: hits throughput: 49.553 ± 2.260 M ops/s, hits latency: 20.181 ns/op, important_hits throughput: 15.767 ± 0.719 M ops/s
num_maps: 17
local_storage cache sequential get: hits throughput: 38.778 ± 0.302 M ops/s, hits latency: 25.788 ns/op, important_hits throughput: 2.284 ± 0.018 M ops/s
local_storage cache interleaved get: hits throughput: 43.848 ± 1.023 M ops/s, hits latency: 22.806 ns/op, important_hits throughput: 13.349 ± 0.311 M ops/s
num_maps: 24
local_storage cache sequential get: hits throughput: 19.317 ± 0.568 M ops/s, hits latency: 51.769 ns/op, important_hits throughput: 0.806 ± 0.024 M ops/s
local_storage cache interleaved get: hits throughput: 24.397 ± 0.272 M ops/s, hits latency: 40.989 ns/op, important_hits throughput: 6.863 ± 0.077 M ops/s
num_maps: 32
local_storage cache sequential get: hits throughput: 13.333 ± 0.135 M ops/s, hits latency: 75.000 ns/op, important_hits throughput: 0.417 ± 0.004 M ops/s
local_storage cache interleaved get: hits throughput: 16.898 ± 0.383 M ops/s, hits latency: 59.178 ns/op, important_hits throughput: 4.717 ± 0.107 M ops/s
num_maps: 100
local_storage cache sequential get: hits throughput: 6.360 ± 0.107 M ops/s, hits latency: 157.233 ns/op, important_hits throughput: 0.064 ± 0.001 M ops/s
local_storage cache interleaved get: hits throughput: 7.303 ± 0.362 M ops/s, hits latency: 136.930 ns/op, important_hits throughput: 1.907 ± 0.094 M ops/s
num_maps: 1000
local_storage cache sequential get: hits throughput: 0.452 ± 0.010 M ops/s, hits latency: 2214.022 ns/op, important_hits throughput: 0.000 ± 0.000 M ops/s
local_storage cache interleaved get: hits throughput: 0.542 ± 0.007 M ops/s, hits latency: 1843.341 ns/op, important_hits throughput: 0.136 ± 0.002 M ops/s
Looking at the "sequential get" results, it's clear that as the
number of task local_storage maps grows beyond the current cache size
(16), there's a significant reduction in hits throughput. Note that
current local_storage implementation assigns a cache_idx to maps as they
are created. Since "sequential get" is creating maps 0..n in order and
then doing bpf_task_storage_get calls in the same order, the benchmark
is effectively ensuring that a map will not be in cache when the program
tries to access it.
For "interleaved get" results, important-map hits throughput is greatly
increased as the important map is more likely to be in cache by virtue
of being accessed far more frequently. Throughput still reduces as #
maps increases, though.
To get a sense of the overhead of the benchmark program, I
commented out bpf_task_storage_get/bpf_map_lookup_elem in
local_storage_bench.c and ran the benchmark on the same host as the
'real' run. Results:
Hashmap Control
===============
num keys: 10
hashmap (control) sequential get: hits throughput: 54.288 ± 0.655 M ops/s, hits latency: 18.420 ns/op, important_hits throughput: 54.288 ± 0.655 M ops/s
num keys: 1000
hashmap (control) sequential get: hits throughput: 52.913 ± 0.519 M ops/s, hits latency: 18.899 ns/op, important_hits throughput: 52.913 ± 0.519 M ops/s
num keys: 10000
hashmap (control) sequential get: hits throughput: 53.480 ± 1.235 M ops/s, hits latency: 18.699 ns/op, important_hits throughput: 53.480 ± 1.235 M ops/s
num keys: 100000
hashmap (control) sequential get: hits throughput: 54.982 ± 1.902 M ops/s, hits latency: 18.188 ns/op, important_hits throughput: 54.982 ± 1.902 M ops/s
num keys: 4194304
hashmap (control) sequential get: hits throughput: 50.858 ± 0.707 M ops/s, hits latency: 19.662 ns/op, important_hits throughput: 50.858 ± 0.707 M ops/s
Local Storage
=============
num_maps: 1
local_storage cache sequential get: hits throughput: 110.990 ± 4.828 M ops/s, hits latency: 9.010 ns/op, important_hits throughput: 110.990 ± 4.828 M ops/s
local_storage cache interleaved get: hits throughput: 161.057 ± 4.090 M ops/s, hits latency: 6.209 ns/op, important_hits throughput: 161.057 ± 4.090 M ops/s
num_maps: 10
local_storage cache sequential get: hits throughput: 112.930 ± 1.079 M ops/s, hits latency: 8.855 ns/op, important_hits throughput: 11.293 ± 0.108 M ops/s
local_storage cache interleaved get: hits throughput: 115.841 ± 2.088 M ops/s, hits latency: 8.633 ns/op, important_hits throughput: 41.372 ± 0.746 M ops/s
num_maps: 16
local_storage cache sequential get: hits throughput: 115.653 ± 0.416 M ops/s, hits latency: 8.647 ns/op, important_hits throughput: 7.228 ± 0.026 M ops/s
local_storage cache interleaved get: hits throughput: 138.717 ± 1.649 M ops/s, hits latency: 7.209 ns/op, important_hits throughput: 44.137 ± 0.525 M ops/s
num_maps: 17
local_storage cache sequential get: hits throughput: 112.020 ± 1.649 M ops/s, hits latency: 8.927 ns/op, important_hits throughput: 6.598 ± 0.097 M ops/s
local_storage cache interleaved get: hits throughput: 128.089 ± 1.960 M ops/s, hits latency: 7.807 ns/op, important_hits throughput: 38.995 ± 0.597 M ops/s
num_maps: 24
local_storage cache sequential get: hits throughput: 92.447 ± 5.170 M ops/s, hits latency: 10.817 ns/op, important_hits throughput: 3.855 ± 0.216 M ops/s
local_storage cache interleaved get: hits throughput: 128.844 ± 2.808 M ops/s, hits latency: 7.761 ns/op, important_hits throughput: 36.245 ± 0.790 M ops/s
num_maps: 32
local_storage cache sequential get: hits throughput: 102.042 ± 1.462 M ops/s, hits latency: 9.800 ns/op, important_hits throughput: 3.194 ± 0.046 M ops/s
local_storage cache interleaved get: hits throughput: 126.577 ± 1.818 M ops/s, hits latency: 7.900 ns/op, important_hits throughput: 35.332 ± 0.507 M ops/s
num_maps: 100
local_storage cache sequential get: hits throughput: 111.327 ± 1.401 M ops/s, hits latency: 8.983 ns/op, important_hits throughput: 1.113 ± 0.014 M ops/s
local_storage cache interleaved get: hits throughput: 131.327 ± 1.339 M ops/s, hits latency: 7.615 ns/op, important_hits throughput: 34.302 ± 0.350 M ops/s
num_maps: 1000
local_storage cache sequential get: hits throughput: 101.978 ± 0.563 M ops/s, hits latency: 9.806 ns/op, important_hits throughput: 0.102 ± 0.001 M ops/s
local_storage cache interleaved get: hits throughput: 141.084 ± 1.098 M ops/s, hits latency: 7.088 ns/op, important_hits throughput: 35.430 ± 0.276 M ops/s
Adjusting for overhead, latency numbers for "hashmap control" and
"sequential get" are:
hashmap_control_1k: ~53.8ns
hashmap_control_10k: ~124.2ns
hashmap_control_100k: ~206.5ns
sequential_get_1: ~12.1ns
sequential_get_10: ~16.0ns
sequential_get_16: ~13.8ns
sequential_get_17: ~16.8ns
sequential_get_24: ~40.9ns
sequential_get_32: ~65.2ns
sequential_get_100: ~148.2ns
sequential_get_1000: ~2204ns
Clearly demonstrating a cliff.
In the discussion for v1 of this patch, Alexei noted that local_storage
was 2.5x faster than a large hashmap when initially implemented [1]. The
benchmark results show that local_storage is 5-10x faster: a
long-running BPF application putting some pid-specific info into a
hashmap for each pid it sees will probably see on the order of 10-100k
pids. Bench numbers for hashmaps of this size are ~10x slower than
sequential_get_16, but as the number of local_storage maps grows far
past local_storage cache size the performance advantage shrinks and
eventually reverses.
When running the benchmarks it may be necessary to bump 'open files'
ulimit for a successful run.
[0]: https://lore.kernel.org/all/20220420002143.1096548-1-davemarchevsky@fb.com
[1]: https://lore.kernel.org/bpf/20220511173305.ftldpn23m4ski3d3@MBP-98dd607d3435.dhcp.thefacebook.com/
Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/r/20220620222554.270578-1-davemarchevsky@fb.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
This changes the section name for the bpf program embedded in these
files to "xdp.frags" to allow the programs to be loaded on drivers that
are using an MTU greater than PAGE_SIZE. Rather than directly accessing
the buffers, the packet data is now accessed via xdp helper functions to
provide an example for those who may need to write more complex
programs.
v2: remove new unnecessary variable
Signed-off-by: Andy Gospodarek <gospo@broadcom.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Lorenzo Bianconi <lorenzo@kernel.org>
Link: https://lore.kernel.org/r/20220621175402.35327-1-gospo@broadcom.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Today doing a BPF tail call after a BPF to BPF call, that is from a
subprogram, is allowed only by the x86-64 BPF JIT. Mixing these features
requires support from JIT. Tail call count has to be tracked through BPF to
BPF calls, as well as through BPF tail calls to prevent unbounded chains of
tail calls.
arm64 BPF JIT stores the tail call count (TCC) in a dedicated
register (X26). This makes it easier to support bpf2bpf calls mixed with
tail calls than on x86 platform.
In order to keep the tail call count in tact throughout bpf2bpf calls, all
we need to do is tweak the program prologue generator. When emitting
prologue for a subprogram, we skip the block that initializes the tail call
count and emits a jump pad for the tail call.
With this change, a sample execution flow where a bpf2bpf call is followed
by a tail call would look like so:
int entry(struct __sk_buff *skb):
0xffffffc0090151d4: paciasp
0xffffffc0090151d8: stp x29, x30, [sp, #-16]!
0xffffffc0090151dc: mov x29, sp
0xffffffc0090151e0: stp x19, x20, [sp, #-16]!
0xffffffc0090151e4: stp x21, x22, [sp, #-16]!
0xffffffc0090151e8: stp x25, x26, [sp, #-16]!
0xffffffc0090151ec: stp x27, x28, [sp, #-16]!
0xffffffc0090151f0: mov x25, sp
0xffffffc0090151f4: mov x26, #0x0 // <- init TCC only
0xffffffc0090151f8: bti j // in main prog
0xffffffc0090151fc: sub x27, x25, #0x0
0xffffffc009015200: sub sp, sp, #0x10
0xffffffc009015204: mov w1, #0x0
0xffffffc009015208: mov x10, #0xffffffffffffffff
0xffffffc00901520c: strb w1, [x25, x10]
0xffffffc009015210: mov x10, #0xffffffffffffd25c
0xffffffc009015214: movk x10, #0x902, lsl #16
0xffffffc009015218: movk x10, #0xffc0, lsl #32
0xffffffc00901521c: blr x10 -------------------. // bpf2bpf call
0xffffffc009015220: add x7, x0, #0x0 <-------------.
0xffffffc009015224: add sp, sp, #0x10 | |
0xffffffc009015228: ldp x27, x28, [sp], #16 | |
0xffffffc00901522c: ldp x25, x26, [sp], #16 | |
0xffffffc009015230: ldp x21, x22, [sp], #16 | |
0xffffffc009015234: ldp x19, x20, [sp], #16 | |
0xffffffc009015238: ldp x29, x30, [sp], #16 | |
0xffffffc00901523c: add x0, x7, #0x0 | |
0xffffffc009015240: autiasp | |
0xffffffc009015244: ret | |
| |
int subprog_tail(struct __sk_buff *skb): | |
0xffffffc00902d25c: paciasp <----------------------' |
0xffffffc00902d260: stp x29, x30, [sp, #-16]! |
0xffffffc00902d264: mov x29, sp |
0xffffffc00902d268: stp x19, x20, [sp, #-16]! |
0xffffffc00902d26c: stp x21, x22, [sp, #-16]! |
0xffffffc00902d270: stp x25, x26, [sp, #-16]! |
0xffffffc00902d274: stp x27, x28, [sp, #-16]! |
0xffffffc00902d278: mov x25, sp |
0xffffffc00902d27c: sub x27, x25, #0x0 |
0xffffffc00902d280: sub sp, sp, #0x10 | // <- end of prologue, notice:
0xffffffc00902d284: add x19, x0, #0x0 | // 1) TCC not touched, and
0xffffffc00902d288: mov w0, #0x1 | // 2) no tail call jump pad
0xffffffc00902d28c: mov x10, #0xfffffffffffffffc |
0xffffffc00902d290: str w0, [x25, x10] |
0xffffffc00902d294: mov x20, #0xffffff80ffffffff |
0xffffffc00902d298: movk x20, #0xc033, lsl #16 |
0xffffffc00902d29c: movk x20, #0x4e00 |
0xffffffc00902d2a0: add x0, x19, #0x0 |
0xffffffc00902d2a4: add x1, x20, #0x0 |
0xffffffc00902d2a8: mov x2, #0x0 |
0xffffffc00902d2ac: mov w10, #0x24 |
0xffffffc00902d2b0: ldr w10, [x1, x10] |
0xffffffc00902d2b4: add w2, w2, #0x0 |
0xffffffc00902d2b8: cmp w2, w10 |
0xffffffc00902d2bc: b.cs 0xffffffc00902d2f8 |
0xffffffc00902d2c0: mov w10, #0x21 |
0xffffffc00902d2c4: cmp x26, x10 | // TCC >= MAX_TAIL_CALL_CNT?
0xffffffc00902d2c8: b.cs 0xffffffc00902d2f8 |
0xffffffc00902d2cc: add x26, x26, #0x1 | // TCC++
0xffffffc00902d2d0: mov w10, #0x110 |
0xffffffc00902d2d4: add x10, x1, x10 |
0xffffffc00902d2d8: lsl x11, x2, #3 |
0xffffffc00902d2dc: ldr x11, [x10, x11] |
0xffffffc00902d2e0: cbz x11, 0xffffffc00902d2f8 |
0xffffffc00902d2e4: mov w10, #0x30 |
0xffffffc00902d2e8: ldr x10, [x11, x10] |
0xffffffc00902d2ec: add x10, x10, #0x24 |
0xffffffc00902d2f0: add sp, sp, #0x10 | // <- destroy just current
0xffffffc00902d2f4: br x10 ---------------------. | // BPF stack frame
0xffffffc00902d2f8: mov x10, #0xfffffffffffffffc | | // before the tail call
0xffffffc00902d2fc: ldr w7, [x25, x10] | |
0xffffffc00902d300: add sp, sp, #0x10 | |
0xffffffc00902d304: ldp x27, x28, [sp], #16 | |
0xffffffc00902d308: ldp x25, x26, [sp], #16 | |
0xffffffc00902d30c: ldp x21, x22, [sp], #16 | |
0xffffffc00902d310: ldp x19, x20, [sp], #16 | |
0xffffffc00902d314: ldp x29, x30, [sp], #16 | |
0xffffffc00902d318: add x0, x7, #0x0 | |
0xffffffc00902d31c: autiasp | |
0xffffffc00902d320: ret | |
| |
int classifier_0(struct __sk_buff *skb): | |
0xffffffc008ff5874: paciasp | |
0xffffffc008ff5878: stp x29, x30, [sp, #-16]! | |
0xffffffc008ff587c: mov x29, sp | |
0xffffffc008ff5880: stp x19, x20, [sp, #-16]! | |
0xffffffc008ff5884: stp x21, x22, [sp, #-16]! | |
0xffffffc008ff5888: stp x25, x26, [sp, #-16]! | |
0xffffffc008ff588c: stp x27, x28, [sp, #-16]! | |
0xffffffc008ff5890: mov x25, sp | |
0xffffffc008ff5894: mov x26, #0x0 | |
0xffffffc008ff5898: bti j <----------------------' |
0xffffffc008ff589c: sub x27, x25, #0x0 |
0xffffffc008ff58a0: sub sp, sp, #0x0 |
0xffffffc008ff58a4: mov x0, #0xffffffc0ffffffff |
0xffffffc008ff58a8: movk x0, #0x8fc, lsl #16 |
0xffffffc008ff58ac: movk x0, #0x6000 |
0xffffffc008ff58b0: mov w1, #0x1 |
0xffffffc008ff58b4: str w1, [x0] |
0xffffffc008ff58b8: mov w7, #0x0 |
0xffffffc008ff58bc: mov sp, sp |
0xffffffc008ff58c0: ldp x27, x28, [sp], #16 |
0xffffffc008ff58c4: ldp x25, x26, [sp], #16 |
0xffffffc008ff58c8: ldp x21, x22, [sp], #16 |
0xffffffc008ff58cc: ldp x19, x20, [sp], #16 |
0xffffffc008ff58d0: ldp x29, x30, [sp], #16 |
0xffffffc008ff58d4: add x0, x7, #0x0 |
0xffffffc008ff58d8: autiasp |
0xffffffc008ff58dc: ret -------------------------------'
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20220617105735.733938-3-jakub@cloudflare.com
|
|
The BPF core/verifier is hard-coded to permit mixing bpf2bpf and tail
calls for only x86-64. Change the logic to instead rely on a new weak
function 'bool bpf_jit_supports_subprog_tailcalls(void)', which a capable
JIT backend can override.
Update the x86-64 eBPF JIT to reflect this.
Signed-off-by: Tony Ambardar <Tony.Ambardar@gmail.com>
[jakub: drop MIPS bits and tweak patch subject]
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20220617105735.733938-2-jakub@cloudflare.com
|
|
Eduard Zingerman says:
====================
Hi Everyone,
This is the next iteration of the patch. It includes changes suggested
by Song, Joanne and Alexei. Please find updated intro message and
change log below.
This patch implements inlining of calls to bpf_loop helper function
when bpf_loop's callback is statically known. E.g. the rewrite does
the following transformation during BPF program processing:
bpf_loop(10, foo, NULL, 0);
->
for (int i = 0; i < 10; ++i)
foo(i, NULL);
The transformation leads to measurable latency change for simple
loops. Measurements using `benchs/run_bench_bpf_loop.sh` inside QEMU /
KVM on i7-4710HQ CPU show a drop in latency from 14 ns/op to 2 ns/op.
The change is split in five parts:
* Update to test_verifier.c to specify expected and unexpected
instruction sequences. This allows to check BPF program rewrites
applied by e.g. do_mix_fixups function.
* Update to test_verifier.c to specify BTF function infos and types
per test case. This is necessary for tests that load sub-program
addresses to a variable because of the checks applied by
check_ld_imm function.
* The update to verifier.c that tracks state of the parameters for
each bpf_loop call in a program and decides whether it could be
replaced by a loop.
* A set of test cases for `test_verifier` that use capabilities added
by the first two patches to verify instructions produced by inlining
logic.
* Two test cases for `test_prog` to check that possible corner cases
behave as expected.
Additional details are available in commit messages for each patch.
Changes since v7:
- Call to `mark_chain_precision` is added in `loop_flag_is_zero` to
avoid potential issues with state pruning and precision tracking.
- `flags non-zero` test_verifier test case is updated to have two
execution paths reaching `bpf_loop` call, one with flags = 0,
another with flags = 1. Potentially this test case should be able
to show that call to `mark_chain_precision` is necessary in
`loop_flag_is_zero` but not at the moment. Please refer to
discussion for [PATCH bpf-next v7 3/5] for additional details.
- `stack_depth_extra` computation is updated to guarantee that R6, R7
and R8 offsets are always aligned on 8 byte boundary.
- `stack locations for loop vars` test_verifier test case updated to
show that R6, R7, R8 offsets are indeed aligned when function stack
depth is not a multiple of 8.
- I removed Song Liu's ACK from commit message for [PATCH bpf-next v8
4/5] because I updated the patch. (Please let me know if I had to
keep the ACK tag).
Changes since v6:
- Return value of the `optimize_bpf_loop` function is no longer
ignored. This is necessary to properly propagate -ENOMEM error.
Changes since v5:
- Added function `loop_flag_is_zero` to skip a few checks in
`update_loop_inline_state` when loop instruction is not fit for
inline.
Changes since v4:
- Added missing `static` modifier for `update_loop_inline_state` and
`inline_bpf_loop` functions.
- `update_loop_inline_state` updated for better readability.
- Fields `initialized` and `fit_for_inline` of `struct
bpf_loop_inline_state` are changed back from `bool` to bitfields.
- Acks from Song Liu added to comments for patches 1/5, 2/5, 4/5,
5/5.
Changes since v3:
- Function `adjust_stack_depth_for_loop_inlining` is replaced by
function `optimize_bpf_loop`. Function `optimize_bpf_loop` is
responsible for both stack depth adjustment and call instruction
replacement.
- Changes in `do_misc_fixups` are reverted.
- Changes in `adjust_subprog_starts_after_remove` are reverted and
function `adjust_loop_inline_subprogno` is removed. This is
possible because call to `optimize_bpf_loop` is placed before the
dead code removal in `opt_remove_dead_code` (in contrast to the
position of `do_misc_fixups` where inlining was done in v3).
- Field `bpf_insn_aux_data.loop_inline_state` is now a part of
anonymous union at the start of the `bpf_insn_aux_data`.
- Data structure `bpf_loop_inline_state` is simplified to use single
flag field `fit_for_inline` instead of separate fields
`flags_is_zero` & `callback_is_constant`.
- Macro definition `BPF_MAX_LOOPS` is moved from
`include/linux/bpf_verifier.h` to `include/linux/bpf.h` to avoid
include of `include/linux/bpf_verifier.h` in `bpf_iter.c`.
- `inline_bpf_loop` changed back to use array initialization and hard
coded offsets as in v2.
- Style / formatting updates.
Changes since v2:
- fix for `stack_check` test case in `test_progs-no_alu32`, all tests
are passing now;
- v2 3/3 patch is split in three parts:
- kernel changes
- test_verifier changes
- test_prog changes
- updated `inline_bpf_loop` in `verifier.c` to calculate each offset
used in instructions to avoid "magic" numbers;
- removed newline handling logic in `fail_log` branch of
`do_single_test` in `test_verifier.c` to simplify the patch set;
- styling fixes suggested in review for v2 of this patch set.
Changes since v1:
- allow to use SKIP_INSNS in instruction pattern specification in
test_verifier tests;
- fix for a bug in spill offset assignement for loop vars when
bpf_loop is located in a non-main function.
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Two new test BPF programs for test_prog selftests checking bpf_loop
behavior. Both are corner cases for bpf_loop inlinig transformation:
- check that bpf_loop behaves correctly when callback function is not
a compile time constant
- check that local function variables are not affected by allocating
additional stack storage for registers spilled by loop inlining
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/r/20220620235344.569325-6-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
