bpf, arm64: support more atomic ops #2577
Conversation
|
Master branch: 9b6eb04 |
|
Master branch: 1b8c924 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
c06cb47
to
04bb3a4
Compare
|
Master branch: 9e98ace |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
04bb3a4
to
1a3e904
Compare
kernel-patches-bot
force-pushed
the
bpf-next
branch
from
9e98ace
to
1b8c924
Compare
|
Master branch: 1b8c924 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
1a3e904
to
f1a2ba0
Compare
|
Master branch: b38101c |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
f1a2ba0
to
bd7afae
Compare
|
Master branch: b75daca |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
bd7afae
to
6055e91
Compare
|
Master branch: d24d2a2 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
6055e91
to
4fa61d4
Compare
|
Master branch: 086d490 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
4fa61d4
to
e9161eb
Compare
|
Master branch: 9087c6f |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
e9161eb
to
5eda530
Compare
|
Master branch: a33c0c7 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
5eda530
to
b886ec2
Compare
|
Master branch: d0b3822 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
b886ec2
to
3994681
Compare
|
Master branch: b03e194 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
3994681
to
a31d028
Compare
|
Master branch: 13c6a37 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
a31d028
to
ebbdb16
Compare
|
Master branch: 6966d4c |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
ebbdb16
to
d94bbb1
Compare
|
Master branch: b4f7278 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
d94bbb1
to
e381dc5
Compare
|
Master branch: a19df71 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
e381dc5
to
6d3ea7e
Compare
|
Master branch: 08894d9 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
6d3ea7e
to
ce28a09
Compare
|
Master branch: c561d11 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
ce28a09
to
1892705
Compare
|
Master branch: 08d4dba |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
1892705
to
8db16b6
Compare
|
Master branch: 80bebeb |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
8db16b6
to
5a15c84
Compare
|
Master branch: dda7596 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
5a15c84
to
4a24cc6
Compare
|
Master branch: 5e214f2 |
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
4a24cc6
to
581d044
Compare
If CONFIG_ARM64_LSE_ATOMICS is off, encoders for LSE-related instructions can return AARCH64_BREAK_FAULT directly in insn.h. In order to access AARCH64_BREAK_FAULT in insn.h, we can not include debug-monitors.h in insn.h, because debug-monitors.h has already depends on insn.h, so just move AARCH64_BREAK_FAULT into insn-def.h. It will be used by the following patch to eliminate unnecessary LSE-related encoders when CONFIG_ARM64_LSE_ATOMICS is off. Signed-off-by: Hou Tao <houtao1@huawei.com>
It is a preparation patch for eBPF atomic supports under arm64. eBPF
needs support atomic[64]_fetch_add, atomic[64]_[fetch_]{and,or,xor} and
atomic[64]_{xchg|cmpxchg}. The ordering semantics of eBPF atomics are
the same with the implementations in linux kernel.
Add three helpers to support LDCLR/LDEOR/LDSET/SWP, CAS and DMB
instructions. STADD/STCLR/STEOR/STSET are simply encoded as aliases for
LDADD/LDCLR/LDEOR/LDSET with XZR as the destination register, so no extra
helper is added. atomic_fetch_add() and other atomic ops needs support for
STLXR instruction, so extend enum aarch64_insn_ldst_type to do that.
LDADD/LDEOR/LDSET/SWP and CAS instructions are only available when LSE
atomics is enabled, so just return AARCH64_BREAK_FAULT directly in
these newly-added helpers if CONFIG_ARM64_LSE_ATOMICS is disabled.
Signed-off-by: Hou Tao <houtao1@huawei.com>
Atomics for eBPF patch series adds support for atomic[64]_fetch_add,
atomic[64]_[fetch_]{and,or,xor} and atomic[64]_{xchg|cmpxchg}, but
it only adds support for x86-64, so support these atomic operations
for arm64 as well.
Basically the implementation procedure is almost mechanical translation
of code snippets in atomic_ll_sc.h & atomic_lse.h & cmpxchg.h located
under arch/arm64/include/asm.
When LSE atomic is unavailable, an extra temporary register is needed for
(BPF_ADD | BPF_FETCH) to save the value of src register, instead of adding
TMP_REG_4 just use BPF_REG_AX instead. Also make emit_lse_atomic() as an
empty inline function when CONFIG_ARM64_LSE_ATOMICS is disabled.
For cpus_have_cap(ARM64_HAS_LSE_ATOMICS) case and no-LSE-ATOMICS case, the
following three tests: "./test_verifier", "./test_progs -t atomic" and
"insmod ./test_bpf.ko" are exercised and passed.
Signed-off-by: Hou Tao <houtao1@huawei.com>
|
Master branch: c62dd8a |
Now atomic tests will attach fentry program and run it through bpf_prog_test_run_opts(), but attaching fentry program depends on bpf trampoline which is only available under x86-64. Considering many archs have atomic support, using raw_tp program instead. Signed-off-by: Hou Tao <houtao1@huawei.com> Acked-by: Andrii Nakryiko <andrii@kernel.org>
kernel-patches-bot
force-pushed
the
series/611843=>bpf-next
branch
from
581d044
to
d6993df
Compare
|
At least one diff in series https://patchwork.kernel.org/project/netdevbpf/list/?series=615259 irrelevant now. Closing PR. |
Pull request for series with
subject: bpf, arm64: support more atomic ops
version: 4
url: https://patchwork.kernel.org/project/netdevbpf/list/?series=615259