| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Adjust the mode eBPF programs are opened in so 0400 pinned bpf programs
work without requiring CAP_DAC_OVERRIDE.
This matches Linux 5.2's:
commit e547ff3f803e779a3898f1f48447b29f43c54085
Author: Chenbo Feng <fengc@google.com>
Date: Tue May 14 19:42:57 2019 -0700
bpf: relax inode permission check for retrieving bpf program
For iptable module to load a bpf program from a pinned location, it
only retrieve a loaded program and cannot change the program content so
requiring a write permission for it might not be necessary.
Also when adding or removing an unrelated iptable rule, it might need to
flush and reload the xt_bpf related rules as well and triggers the inode
permission check. It might be better to remove the write premission
check for the inode so we won't need to grant write access to all the
processes that flush and restore iptables rules.
kernel/bpf/inode.c:
- int ret = inode_permission(inode, MAY_READ | MAY_WRITE);
+ int ret = inode_permission(inode, MAY_READ);
In practice, AFAICT, the xt_bpf match .fd field isn't even used by new
kernels, but I believe it might be needed for compatibility with old ones
(though I'm pretty sure table modifications on them will outright fail).
Test: builds, passes Android test suite (albeit on an older iptables base),
git grep bpf_obj_get - finds no other users
Cc: Chenbo Feng <fengc@google.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: Maciej Żenczykowski <maze@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In kernel 3.18 the union bpf_attr does not have a pathname attribute and
BPF_OBJ_GET is also not defined in these versions.
This was added in Linux commit b2197755b263 ("bpf: add support for
persistent maps/progs"). Check for the BPF_FS_MAGIC define which was
also added in this Linux commit and only activate this code in case we
find that define.
This fixes a build problem with Linux 3.18.
Netfilter bug: #1231
Fixes: f17f9ace8a8 ("extensions: libxt_bpf: support ebpf pinned objects")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
|
|
|
|
|
| |
This include is needed to compile the bpf_obj_get function properly,
as it brings in the __NR_bpf declaration.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Exercise the new kernel feature introduced in commit 2c16d6033264
("netfilter: xt_bpf: support ebpf") to load pinned eBPF programs.
The new interface allows instantiating a bpf match using
-m bpf --object-pinned ${PATH}
where ${PATH} points to a node in a bpf virtual filesystem. See
also the revised man page.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Add user-space code to support the new BPF iptables extension.
Pablo has mangled the original patch to:
* include a copy of include/linux/netfilter/xt_bpf.h in the tree.
* I have also remove the --bytecode-file option. The original
proposal was to accept BPF code in a file in human readable
format. Now, with the nfbpf_compile utility, it's very easy
to generate the filter using tcpdump-like syntax.
* I have remove the trailing comma in the backtick format, the
parser works just fine for me here.
* Fix error message if --bytecode is missing.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|