KF_CHANGES_PKT indicates that the kfunc call may change packet data.
This is analogous to bpf_helper_changes_pkt_data().
Signed-off-by: Daniel Xu <[email protected]>
---
Documentation/bpf/kfuncs.rst | 7 +++++++
include/linux/btf.h | 1 +
kernel/bpf/verifier.c | 8 ++++++++
3 files changed, 16 insertions(+)
diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
index 9fd7fb539f85..061ab392a02f 100644
--- a/Documentation/bpf/kfuncs.rst
+++ b/Documentation/bpf/kfuncs.rst
@@ -200,6 +200,13 @@ single argument which must be a trusted argument or a MEM_RCU pointer.
The argument may have reference count of 0 and the kfunc must take this
into consideration.
+2.4.9 KF_CHANGES_PKT flag
+-----------------
+
+The KF_CHANGES_PKT is used for kfuncs that may change packet data.
+After calls to such kfuncs, existing packet pointers will be invalidated
+and must be revalidated before the prog can access packet data.
+
2.5 Registering the kfuncs
--------------------------
diff --git a/include/linux/btf.h b/include/linux/btf.h
index 5f628f323442..0575f530e40b 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -71,6 +71,7 @@
#define KF_SLEEPABLE (1 << 5) /* kfunc may sleep */
#define KF_DESTRUCTIVE (1 << 6) /* kfunc performs destructive actions */
#define KF_RCU (1 << 7) /* kfunc only takes rcu pointer arguments */
+#define KF_CHANGES_PKT (1 << 8) /* kfunc may change packet data */
/*
* Return the name of the passed struct, if exists, or halt the build if for
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index a5255a0dcbb6..0ac505cbd6ba 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -8213,6 +8213,11 @@ static bool is_kfunc_rcu(struct bpf_kfunc_call_arg_meta *meta)
return meta->kfunc_flags & KF_RCU;
}
+static bool is_kfunc_changes_pkt(struct bpf_kfunc_call_arg_meta *meta)
+{
+ return meta->kfunc_flags & KF_CHANGES_PKT;
+}
+
static bool is_kfunc_arg_kptr_get(struct bpf_kfunc_call_arg_meta *meta, int arg)
{
return arg == 0 && (meta->kfunc_flags & KF_KPTR_GET);
@@ -9313,6 +9318,9 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
mark_btf_func_reg_size(env, regno, t->size);
}
+ if (is_kfunc_changes_pkt(&meta))
+ clear_all_pkt_pointers(env);
+
return 0;
}
--
2.39.0
Hi Daniel,
Thank you for the patch! Perhaps something to improve:
[auto build test WARNING on bpf-next/master]
url: https://github.com/intel-lab-lkp/linux/commits/Daniel-Xu/Support-defragmenting-IPv4-packets-in-BPF/20221215-072858
base: https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master
patch link: https://lore.kernel.org/r/fc37530d6cf954c3ebd1173d2bdc6b731c733134.1671049840.git.dxu%40dxuuu.xyz
patch subject: [PATCH bpf-next 2/6] bpf: verifier: Support KF_CHANGES_PKT flag
reproduce:
# https://github.com/intel-lab-lkp/linux/commit/ca42b5d5e5d93debd023634aca7e206550fee180
git remote add linux-review https://github.com/intel-lab-lkp/linux
git fetch --no-tags linux-review Daniel-Xu/Support-defragmenting-IPv4-packets-in-BPF/20221215-072858
git checkout ca42b5d5e5d93debd023634aca7e206550fee180
make menuconfig
# enable CONFIG_COMPILE_TEST, CONFIG_WARN_MISSING_DOCUMENTS, CONFIG_WARN_ABI_ERRORS
make htmldocs
If you fix the issue, kindly add following tag where applicable
| Reported-by: kernel test robot <[email protected]>
All warnings (new ones prefixed by >>):
>> Documentation/bpf/kfuncs.rst:204: WARNING: Title underline too short.
vim +204 Documentation/bpf/kfuncs.rst
202
203 2.4.9 KF_CHANGES_PKT flag
> 204 -----------------
205
--
0-DAY CI Kernel Test Service
https://01.org/lkp