linux.git/kernel/bpf/offload.c, branch v4.18.12

bpf: offload: allow offloaded programs to use perf event arrays

2018-05-04T21:41:03+00:00

BPF_MAP_TYPE_PERF_EVENT_ARRAY is special as far as offload goes.
The map only holds glue to perf ring, not actual data.  Allow
non-offloaded perf event arrays to be used in offloaded programs.
Offload driver can extract the events from HW and put them in
the map for user space to retrieve.

Signed-off-by: Jakub Kicinski 
Reviewed-by: Quentin Monnet 
Reviewed-by: Jiong Wang 
Signed-off-by: Daniel Borkmann

bpf: offload: report device information about offloaded maps

2018-01-18T21:54:25+00:00

Tell user space about device on which the map was created.
Unfortunate reality of user ABI makes sharing this code
with program offload difficult but the information is the
same.

Signed-off-by: Jakub Kicinski 
Acked-by: Alexei Starovoitov 
Signed-off-by: Daniel Borkmann

bpf: offload: allow array map offload

2018-01-18T21:54:25+00:00

The special handling of different map types is left to the driver.
Allow offload of array maps by simply adding it to accepted types.
For nfp we have to make sure array elements are not deleted.

Signed-off-by: Jakub Kicinski 
Signed-off-by: Daniel Borkmann

bpf: add new jited info fields in bpf_dev_offload and bpf_prog_info

2018-01-18T00:26:15+00:00

For host JIT, there are "jited_len"/"bpf_func" fields in struct bpf_prog
used by all host JIT targets to get jited image and it's length. While for
offload, targets are likely to have different offload mechanisms that these
info are kept in device private data fields.

Therefore, BPF_OBJ_GET_INFO_BY_FD syscall needs an unified way to get JIT
length and contents info for offload targets.

One way is to introduce new callback to parse device private data then fill
those fields in bpf_prog_info. This might be a little heavy, the other way
is to add generic fields which will be initialized by all offload targets.

This patch follow the second approach to introduce two new fields in
struct bpf_dev_offload and teach bpf_prog_get_info_by_fd about them to fill
correct jited_prog_len and jited_prog_insns in bpf_prog_info.

Reviewed-by: Jakub Kicinski 
Signed-off-by: Jiong Wang 
Acked-by: Alexei Starovoitov 
Signed-off-by: Daniel Borkmann

bpf: offload: make bpf_offload_dev_match() reject host+host case

2018-01-17T00:15:05+00:00

Daniel suggests it would be more logical for bpf_offload_dev_match()
to return false is either the program or the map are not offloaded,
rather than treating the both not offloaded case as a "matching
CPU/host device".

This makes no functional difference today, since verifier only calls
bpf_offload_dev_match() when one of the objects is offloaded.

Signed-off-by: Jakub Kicinski 
Signed-off-by: Daniel Borkmann

bpf: offload: add map offload infrastructure

2018-01-14T22:36:30+00:00

BPF map offload follow similar path to program offload.  At creation
time users may specify ifindex of the device on which they want to
create the map.  Map will be validated by the kernel's
.map_alloc_check callback and device driver will be called for the
actual allocation.  Map will have an empty set of operations
associated with it (save for alloc and free callbacks).  The real
device callbacks are kept in map->offload->dev_ops because they
have slightly different signatures.  Map operations are called in
process context so the driver may communicate with HW freely,
msleep(), wait() etc.

Map alloc and free callbacks are muxed via existing .ndo_bpf, and
are always called with rtnl lock held.  Maps and programs are
guaranteed to be destroyed before .ndo_uninit (i.e. before
unregister_netdev() returns).  Map callbacks are invoked with
bpf_devs_lock *read* locked, drivers must take care of exclusive
locking if necessary.

All offload-specific branches are marked with unlikely() (through
bpf_map_is_dev_bound()), given that branch penalty will be
negligible compared to IO anyway, and we don't want to penalize
SW path unnecessarily.

Signed-off-by: Jakub Kicinski 
Reviewed-by: Quentin Monnet 
Signed-off-by: Daniel Borkmann

bpf: offload: factor out netdev checking at allocation time

2018-01-14T22:36:29+00:00

Add a helper to check if netdev could be found and whether it
has .ndo_bpf callback.  There is no need to check the callback
every time it's invoked, ndos can't reasonably be swapped for
a set without .ndp_bpf while program is loaded.

bpf_dev_offload_check() will also be used by map offload.

Signed-off-by: Jakub Kicinski 
Reviewed-by: Quentin Monnet 
Signed-off-by: Daniel Borkmann

bpf: rename bpf_dev_offload -> bpf_prog_offload

2018-01-14T22:36:29+00:00

With map offload coming, we need to call program offload structure
something less ambiguous.  Pure rename, no functional changes.

Signed-off-by: Jakub Kicinski 
Reviewed-by: Quentin Monnet 
Signed-off-by: Daniel Borkmann

bpf: offload: report device information for offloaded programs

2017-12-31T15:12:23+00:00

Report to the user ifindex and namespace information of offloaded
programs.  If device has disappeared return -ENODEV.  Specify the
namespace using dev/inode combination.

CC: Eric W. Biederman 
Signed-off-by: Jakub Kicinski 
Signed-off-by: Daniel Borkmann

bpf: offload: free program id when device disappears

2017-12-31T15:12:23+00:00

Bound programs are quite useless after their device disappears.
They are simply waiting for reference count to go to zero,
don't list them in BPF_PROG_GET_NEXT_ID by freeing their ID
early.

Note that orphaned offload programs will return -ENODEV on
BPF_OBJ_GET_INFO_BY_FD so user will never see ID 0.

Signed-off-by: Jakub Kicinski 
Reviewed-by: Quentin Monnet 
Acked-by: Alexei Starovoitov 
Signed-off-by: Daniel Borkmann