pull-request: bpf 2019-01-31
Subsystems (2)
Hi David,
The following pull-request contains BPF updates for your *net* tree.
The main changes are:
1) disable preemption in sender side of socket filters, from Alexei.
2) fix two potential deadlocks in syscall bpf lookup and prog_register, from Martin and Alexei.
3) fix BTF to allow typedef on func_proto, from Yonghong.
4) two bpftool fixes, from Jiri and Paolo.
Please consider pulling these changes from:
git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf.git
Thanks a lot!
----------------------------------------------------------------
The following changes since commit ab06418b7fb8cc1bf82cbce3ebd74e1bbf446d06:
net: phy: Fixup GPLv2+ SPDX tags based on license text (2019-01-22 20:57:03 -0800)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf.git
for you to fetch changes up to f01c2803873e83ea5f1b160c3169ed6018704be8:
Merge branch 'bpf-lockdep-fixes' (2019-01-31 23:18:23 +0100)
----------------------------------------------------------------
Alexei Starovoitov (4):
Merge branch 'typedef-func_proto'
bpf: run bpf programs with preemption disabled
bpf: fix lockdep false positive in percpu_freelist
bpf: fix potential deadlock in bpf_prog_register
Daniel Borkmann (2):
bpf, doc: add reviewers to maintainers entry Merge branch 'bpf-lockdep-fixes'
Jakub Kicinski (1):
tools: bpftool: fix crash with un-owned prog arrays
Jakub Sitnicki (1):
sk_msg: Always cancel strp work before freeing the psock
Jiri Olsa (1):
bpftool: Fix prog dump by tag
Martin KaFai Lau (1):
bpf: Fix syscall's stackmap lookup potential deadlock
Martynas Pumputis (1):
bpf, selftests: fix handling of sparse CPU allocations
Paolo Abeni (1):
bpftool: fix percpu maps updating
Yafang Shao (1):
bpf: sock recvbuff must be limited by rmem_max in bpf_setsockopt()
Yonghong Song (2):
bpf: btf: allow typedef func_proto tools/bpf: fix test_btf for typedef func_proto case
MAINTAINERS | 5 +++++ include/linux/filter.h | 21 ++++++++++++++--- kernel/bpf/btf.c | 3 ++- kernel/bpf/cgroup.c | 2 +- kernel/bpf/hashtab.c | 4 ++-- kernel/bpf/percpu_freelist.c | 41 ++++++++++++++++++++++++---------- kernel/bpf/percpu_freelist.h | 4 ++++ kernel/bpf/syscall.c | 12 ++++++++-- kernel/trace/bpf_trace.c | 14 ++---------- net/core/filter.c | 2 ++ net/core/skmsg.c | 3 +-- tools/bpf/bpftool/common.c | 6 +---- tools/bpf/bpftool/map.c | 33 +++++++++++++++++++-------- tools/bpf/bpftool/prog.c | 5 +++-- tools/testing/selftests/bpf/bpf_util.h | 30 ++++++++++++++++--------- tools/testing/selftests/bpf/test_btf.c | 9 +++----- 16 files changed, 127 insertions(+), 67 deletions(-)
David Miller
maintainerFrom: Alexei Starovoitov <[email protected]>
Date: Thu, 31 Jan 2019 14:50:40 -0800
Quoted 9 lines
The following pull-request contains BPF updates for your *net* tree.
The main changes are:
1) disable preemption in sender side of socket filters, from Alexei.
2) fix two potential deadlocks in syscall bpf lookup and
prog_register, from Martin and Alexei.
Quoted 7 lines
3) fix BTF to allow typedef on
func_proto, from Yonghong.4) two bpftool fixes, from Jiri and Paolo.
Please consider pulling these changes from:
git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf.git
Pulled, thanks Alexei.