* [PATCH bpf-next v3 4/9] bpf: Introduce mem, size argument pair support for kfunc
From: Kumar Kartikeya Dwivedi @ 2021-12-10 13:02 UTC (permalink / raw)
To: bpf
Cc: Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
Maxim Mikityanskiy, Florian Westphal, Jesper Dangaard Brouer,
Toke Høiland-Jørgensen, netdev, netfilter-devel
In-Reply-To: <20211210130230.4128676-1-memxor@gmail.com>
BPF helpers can associate two adjacent arguments together to pass memory
of certain size, using ARG_PTR_TO_MEM and ARG_CONST_SIZE arguments.
Since we don't use bpf_func_proto for kfunc, we need to leverage BTF to
implement similar support.
The ARG_CONST_SIZE processing for helpers is refactored into a common
check_mem_size_reg helper that is shared with kfunc as well. kfunc
ptr_to_mem support follows logic similar to global functions, where
verification is done as if pointer is not null, even when it may be
null.
This leads to a simple to follow rule for writing kfunc: always check
the argument pointer for NULL, except when it is PTR_TO_CTX.
Currently, we require the size argument to be prefixed with "len__" in
the parameter name. This information is then recorded in kernel BTF and
verified during function argument checking. In the future we can use BTF
tagging instead, and modify the kernel function definitions. This will
be a purely kernel-side change.
This allows us to have some form of backwards compatibility for
structures that are passed in to the kernel function with their size,
and allow variable length structures to be passed in if they are
accompanied by a size parameter.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
---
include/linux/bpf_verifier.h | 2 +
kernel/bpf/btf.c | 41 +++++++++++-
kernel/bpf/verifier.c | 124 ++++++++++++++++++++++-------------
3 files changed, 119 insertions(+), 48 deletions(-)
diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index 182b16a91084..b80fe5bf2a02 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -507,6 +507,8 @@ bpf_prog_offload_remove_insns(struct bpf_verifier_env *env, u32 off, u32 cnt);
int check_ctx_reg(struct bpf_verifier_env *env,
const struct bpf_reg_state *reg, int regno);
+int check_kfunc_mem_size_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
+ u32 regno);
int check_mem_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
u32 regno, u32 mem_size);
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 63b22ff73550..df9a3f77fc4a 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -5618,6 +5618,25 @@ static bool __btf_type_is_scalar_struct(struct bpf_verifier_log *log,
return true;
}
+static bool is_kfunc_arg_mem_size(const struct btf *btf,
+ const struct btf_param *arg,
+ const struct bpf_reg_state *reg)
+{
+ const struct btf_type *t;
+ const char *param_name;
+
+ t = btf_type_skip_modifiers(btf, arg->type, NULL);
+ if (!btf_type_is_scalar(t) || reg->type != SCALAR_VALUE)
+ return false;
+
+ /* In the future, this can be ported to use BTF tagging */
+ param_name = btf_name_by_offset(btf, arg->name_off);
+ if (strncmp(param_name, "len__", sizeof("len__") - 1))
+ return false;
+
+ return true;
+}
+
static int btf_check_func_arg_match(struct bpf_verifier_env *env,
const struct btf *btf, u32 func_id,
struct bpf_reg_state *regs,
@@ -5729,16 +5748,32 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env,
u32 type_size;
if (is_kfunc) {
+ bool arg_mem_size = i + 1 < nargs && is_kfunc_arg_mem_size(btf, &args[i + 1], ®s[regno + 1]);
+
/* Permit pointer to mem, but only when argument
* type is pointer to scalar, or struct composed
* (recursively) of scalars.
+ * When arg_mem_size is true, the pointer can be
+ * void *.
*/
- if (!btf_type_is_scalar(ref_t) && !__btf_type_is_scalar_struct(log, btf, ref_t, 0)) {
+ if (!btf_type_is_scalar(ref_t) && !__btf_type_is_scalar_struct(log, btf, ref_t, 0) &&
+ (arg_mem_size ? !btf_type_is_void(ref_t) : 1)) {
bpf_log(log,
- "arg#%d pointer type %s %s must point to scalar or struct with scalar\n",
- i, btf_type_str(ref_t), ref_tname);
+ "arg#%d pointer type %s %s must point to %sscalar, or struct with scalar\n",
+ i, btf_type_str(ref_t), ref_tname, arg_mem_size ? "void, " : "");
return -EINVAL;
}
+
+ /* Check for mem, len pair */
+ if (arg_mem_size) {
+ if (check_kfunc_mem_size_reg(env, ®s[regno + 1], regno + 1)) {
+ bpf_log(log, "arg#%d arg#%d memory, len pair leads to invalid memory access\n",
+ i, i + 1);
+ return -EINVAL;
+ }
+ i++;
+ continue;
+ }
}
resolve_ret = btf_resolve_size(btf, ref_t, &type_size);
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 1126b75fe650..074a78a0efa4 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -4780,6 +4780,62 @@ static int check_helper_mem_access(struct bpf_verifier_env *env, int regno,
}
}
+static int check_mem_size_reg(struct bpf_verifier_env *env,
+ struct bpf_reg_state *reg, u32 regno,
+ bool zero_size_allowed,
+ struct bpf_call_arg_meta *meta)
+{
+ int err;
+
+ /* This is used to refine r0 return value bounds for helpers
+ * that enforce this value as an upper bound on return values.
+ * See do_refine_retval_range() for helpers that can refine
+ * the return value. C type of helper is u32 so we pull register
+ * bound from umax_value however, if negative verifier errors
+ * out. Only upper bounds can be learned because retval is an
+ * int type and negative retvals are allowed.
+ */
+ if (meta)
+ meta->msize_max_value = reg->umax_value;
+
+ /* The register is SCALAR_VALUE; the access check
+ * happens using its boundaries.
+ */
+ if (!tnum_is_const(reg->var_off))
+ /* For unprivileged variable accesses, disable raw
+ * mode so that the program is required to
+ * initialize all the memory that the helper could
+ * just partially fill up.
+ */
+ meta = NULL;
+
+ if (reg->smin_value < 0) {
+ verbose(env, "R%d min value is negative, either use unsigned or 'var &= const'\n",
+ regno);
+ return -EACCES;
+ }
+
+ if (reg->umin_value == 0) {
+ err = check_helper_mem_access(env, regno - 1, 0,
+ zero_size_allowed,
+ meta);
+ if (err)
+ return err;
+ }
+
+ if (reg->umax_value >= BPF_MAX_VAR_SIZ) {
+ verbose(env, "R%d unbounded memory access, use 'var &= const' or 'if (var < const)'\n",
+ regno);
+ return -EACCES;
+ }
+ err = check_helper_mem_access(env, regno - 1,
+ reg->umax_value,
+ zero_size_allowed, meta);
+ if (!err)
+ err = mark_chain_precision(env, regno);
+ return err;
+}
+
int check_mem_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
u32 regno, u32 mem_size)
{
@@ -4803,6 +4859,28 @@ int check_mem_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
return check_helper_mem_access(env, regno, mem_size, true, NULL);
}
+int check_kfunc_mem_size_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
+ u32 regno)
+{
+ struct bpf_reg_state *mem_reg = &cur_regs(env)[regno - 1];
+ bool may_be_null = reg_type_may_be_null(mem_reg->type);
+ struct bpf_reg_state saved_reg;
+ int err;
+
+ WARN_ON_ONCE(regno < BPF_REG_2 || regno > BPF_REG_5);
+
+ if (may_be_null) {
+ saved_reg = *mem_reg;
+ mark_ptr_not_null_reg(mem_reg);
+ }
+
+ err = check_mem_size_reg(env, reg, regno, true, NULL);
+
+ if (may_be_null)
+ *mem_reg = saved_reg;
+ return err;
+}
+
/* Implementation details:
* bpf_map_lookup returns PTR_TO_MAP_VALUE_OR_NULL
* Two bpf_map_lookups (even with the same key) will have different reg->id.
@@ -5316,51 +5394,7 @@ static int check_func_arg(struct bpf_verifier_env *env, u32 arg,
} else if (arg_type_is_mem_size(arg_type)) {
bool zero_size_allowed = (arg_type == ARG_CONST_SIZE_OR_ZERO);
- /* This is used to refine r0 return value bounds for helpers
- * that enforce this value as an upper bound on return values.
- * See do_refine_retval_range() for helpers that can refine
- * the return value. C type of helper is u32 so we pull register
- * bound from umax_value however, if negative verifier errors
- * out. Only upper bounds can be learned because retval is an
- * int type and negative retvals are allowed.
- */
- meta->msize_max_value = reg->umax_value;
-
- /* The register is SCALAR_VALUE; the access check
- * happens using its boundaries.
- */
- if (!tnum_is_const(reg->var_off))
- /* For unprivileged variable accesses, disable raw
- * mode so that the program is required to
- * initialize all the memory that the helper could
- * just partially fill up.
- */
- meta = NULL;
-
- if (reg->smin_value < 0) {
- verbose(env, "R%d min value is negative, either use unsigned or 'var &= const'\n",
- regno);
- return -EACCES;
- }
-
- if (reg->umin_value == 0) {
- err = check_helper_mem_access(env, regno - 1, 0,
- zero_size_allowed,
- meta);
- if (err)
- return err;
- }
-
- if (reg->umax_value >= BPF_MAX_VAR_SIZ) {
- verbose(env, "R%d unbounded memory access, use 'var &= const' or 'if (var < const)'\n",
- regno);
- return -EACCES;
- }
- err = check_helper_mem_access(env, regno - 1,
- reg->umax_value,
- zero_size_allowed, meta);
- if (!err)
- err = mark_chain_precision(env, regno);
+ err = check_mem_size_reg(env, reg, regno, zero_size_allowed, meta);
} else if (arg_type_is_alloc_size(arg_type)) {
if (!tnum_is_const(reg->var_off)) {
verbose(env, "R%d is not a known constant'\n",
--
2.34.1
^ permalink raw reply related
* [PATCH bpf-next v3 5/9] bpf: Add reference tracking support to kfunc
From: Kumar Kartikeya Dwivedi @ 2021-12-10 13:02 UTC (permalink / raw)
To: bpf
Cc: Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
Maxim Mikityanskiy, Florian Westphal, Jesper Dangaard Brouer,
Toke Høiland-Jørgensen, netdev, netfilter-devel
In-Reply-To: <20211210130230.4128676-1-memxor@gmail.com>
This patch adds verifier support for PTR_TO_BTF_ID return type of kfunc
to be a reference, by reusing acquire_reference_state/release_reference
support for existing in-kernel bpf helpers.
Verifier ops struct is extended with three callbacks:
- is_acquire_kfunc
Return true if kfunc_btf_id, module pair is an acquire kfunc. This
will acquire_reference_state for the returned PTR_TO_BTF_ID (this is
the only allow return value). Note that acquire kfunc must always
return a PTR_TO_BTF_ID{_OR_NULL}, otherwise the program is rejected.
- is_release_kfunc
Return true if kfunc_btf_id, module pair is a release kfunc. This
will release the reference to the passed in PTR_TO_BTF_ID which has a
reference state (from earlier acquire kfunc).
The btf_check_func_arg_match returns the regno (of argument register,
hence > 0) if the kfunc is a release kfunc, and a proper referenced
PTR_TO_BTF_ID is being passed to it.
This is similar to how helper call check uses bpf_call_arg_meta to
store the ref_obj_id that is later used to release the reference.
Similar to in-kernel helper, we only allow passing one referenced
PTR_TO_BTF_ID as an argument. It can also be passed in to normal
kfunc, but in case of release kfunc there must always be one
PTR_TO_BTF_ID argument that is referenced.
- is_kfunc_ret_type_null
For kfunc returning PTR_TO_BTF_ID, tells if it can be NULL, hence
force caller to mark the pointer not null (using check) before
accessing it. Note that taking into account the case fixed by commit
93c230e3f5bd6, we assign a non-zero id for mark_ptr_or_null_reg logic.
Later, if more return types are supported by kfunc, which have a
_OR_NULL variant, it might be better to move this id generation under
a common reg_type_may_be_null check, similar to the case in the
commit.
Later patches will implement these callbacks.
Referenced PTR_TO_BTF_ID is currently only limited to kfunc, but can be
extended in the future to other BPF helpers as well. For now, we can
rely on the btf_struct_ids_match check to ensure we get the pointer to
the expected struct type. In the future, care needs to be taken to avoid
ambiguity for reference PTR_TO_BTF_ID passed to release function, in
case multiple candidates can release same BTF ID.
e.g. there might be two release kfuncs (or kfunc and helper):
foo(struct abc *p);
bar(struct abc *p);
... such that both release a PTR_TO_BTF_ID with btf_id of struct abc. In
this case we would need to track the acquire function corresponding to
the release function to avoid type confusion, and store this information
in the register state so that an incorrect program can be rejected. This
is not a problem right now, hence it is left as an exercise for the
future patch introducing such a case in the kernel.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
---
include/linux/bpf.h | 6 ++++-
kernel/bpf/btf.c | 42 ++++++++++++++++++++++++++++-----
kernel/bpf/verifier.c | 54 +++++++++++++++++++++++++++++++++++++------
3 files changed, 88 insertions(+), 14 deletions(-)
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 8bbf08fbab66..6a14072c72a0 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -521,6 +521,9 @@ struct bpf_verifier_ops {
enum bpf_access_type atype,
u32 *next_btf_id);
bool (*check_kfunc_call)(u32 kfunc_btf_id, struct module *owner);
+ bool (*is_acquire_kfunc)(u32 kfunc_btf_id, struct module *owner);
+ bool (*is_release_kfunc)(u32 kfunc_btf_id, struct module *owner);
+ bool (*is_kfunc_ret_type_null)(u32 kfunc_btf_id, struct module *owner);
};
struct bpf_prog_offload_ops {
@@ -1717,7 +1720,8 @@ int btf_check_subprog_arg_match(struct bpf_verifier_env *env, int subprog,
struct bpf_reg_state *regs);
int btf_check_kfunc_arg_match(struct bpf_verifier_env *env,
const struct btf *btf, u32 func_id,
- struct bpf_reg_state *regs);
+ struct bpf_reg_state *regs,
+ struct module *btf_mod);
int btf_prepare_func_args(struct bpf_verifier_env *env, int subprog,
struct bpf_reg_state *reg);
int btf_check_type_match(struct bpf_verifier_log *log, const struct bpf_prog *prog,
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index df9a3f77fc4a..a790ba6d93d8 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -5640,14 +5640,17 @@ static bool is_kfunc_arg_mem_size(const struct btf *btf,
static int btf_check_func_arg_match(struct bpf_verifier_env *env,
const struct btf *btf, u32 func_id,
struct bpf_reg_state *regs,
- bool ptr_to_mem_ok)
+ bool ptr_to_mem_ok,
+ struct module *btf_mod)
{
struct bpf_verifier_log *log = &env->log;
+ u32 i, nargs, ref_id, ref_obj_id = 0;
bool is_kfunc = btf_is_kernel(btf);
const char *func_name, *ref_tname;
const struct btf_type *t, *ref_t;
const struct btf_param *args;
- u32 i, nargs, ref_id;
+ int ref_regno = 0;
+ bool rel = false;
t = btf_type_by_id(btf, func_id);
if (!t || !btf_type_is_func(t)) {
@@ -5725,6 +5728,16 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env,
if (reg->type == PTR_TO_BTF_ID) {
reg_btf = reg->btf;
reg_ref_id = reg->btf_id;
+ /* Ensure only one argument is referenced PTR_TO_BTF_ID */
+ if (reg->ref_obj_id) {
+ if (ref_obj_id) {
+ bpf_log(log, "verifier internal error: more than one arg with ref_obj_id R%d %u %u\n",
+ regno, reg->ref_obj_id, ref_obj_id);
+ return -EFAULT;
+ }
+ ref_regno = regno;
+ ref_obj_id = reg->ref_obj_id;
+ }
} else {
reg_btf = btf_vmlinux;
reg_ref_id = *reg2btf_ids[reg->type];
@@ -5794,7 +5807,23 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env,
}
}
- return 0;
+ /* Either both are set, or neither */
+ WARN_ON_ONCE((ref_obj_id && !ref_regno) || (!ref_obj_id && ref_regno));
+ if (is_kfunc) {
+ rel = env->ops->is_release_kfunc &&
+ env->ops->is_release_kfunc(func_id, btf_mod);
+ /* We already made sure ref_obj_id is set only for one argument */
+ if (rel && !ref_obj_id) {
+ bpf_log(log, "release kernel function %s expects refcounted PTR_TO_BTF_ID\n",
+ func_name);
+ return -EINVAL;
+ }
+ /* Allow (!rel && ref_obj_id), so that passing such referenced PTR_TO_BTF_ID to
+ * other kfuncs works
+ */
+ }
+ /* returns argument register number > 0 in case of reference release kfunc */
+ return rel ? ref_regno : 0;
}
/* Compare BTF of a function with given bpf_reg_state.
@@ -5824,7 +5853,7 @@ int btf_check_subprog_arg_match(struct bpf_verifier_env *env, int subprog,
return -EINVAL;
is_global = prog->aux->func_info_aux[subprog].linkage == BTF_FUNC_GLOBAL;
- err = btf_check_func_arg_match(env, btf, btf_id, regs, is_global);
+ err = btf_check_func_arg_match(env, btf, btf_id, regs, is_global, NULL);
/* Compiler optimizations can remove arguments from static functions
* or mismatched type can be passed into a global function.
@@ -5837,9 +5866,10 @@ int btf_check_subprog_arg_match(struct bpf_verifier_env *env, int subprog,
int btf_check_kfunc_arg_match(struct bpf_verifier_env *env,
const struct btf *btf, u32 func_id,
- struct bpf_reg_state *regs)
+ struct bpf_reg_state *regs,
+ struct module *btf_mod)
{
- return btf_check_func_arg_match(env, btf, func_id, regs, true);
+ return btf_check_func_arg_match(env, btf, func_id, regs, true, btf_mod);
}
/* Convert BTF of a function into bpf_reg_state if possible
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 074a78a0efa4..b8685fb7ff15 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -466,7 +466,9 @@ static bool reg_type_may_be_refcounted_or_null(enum bpf_reg_type type)
type == PTR_TO_TCP_SOCK ||
type == PTR_TO_TCP_SOCK_OR_NULL ||
type == PTR_TO_MEM ||
- type == PTR_TO_MEM_OR_NULL;
+ type == PTR_TO_MEM_OR_NULL ||
+ type == PTR_TO_BTF_ID ||
+ type == PTR_TO_BTF_ID_OR_NULL;
}
static bool arg_type_may_be_refcounted(enum bpf_arg_type type)
@@ -6753,16 +6755,18 @@ static void mark_btf_func_reg_size(struct bpf_verifier_env *env, u32 regno,
}
}
-static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn)
+static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
+ int *insn_idx_p)
{
const struct btf_type *t, *func, *func_proto, *ptr_type;
struct bpf_reg_state *regs = cur_regs(env);
const char *func_name, *ptr_type_name;
u32 i, nargs, func_id, ptr_type_id;
+ int err, insn_idx = *insn_idx_p;
struct module *btf_mod = NULL;
const struct btf_param *args;
struct btf *desc_btf;
- int err;
+ bool acq;
/* skip for now, but return error when we find this in fixup_kfunc_call */
if (!insn->imm)
@@ -6784,16 +6788,37 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn)
return -EACCES;
}
+ /* Is this an acquire kfunc? */
+ acq = env->ops->is_acquire_kfunc &&
+ env->ops->is_acquire_kfunc(func_id, btf_mod);
+
/* Check the arguments */
- err = btf_check_kfunc_arg_match(env, desc_btf, func_id, regs);
- if (err)
+ err = btf_check_kfunc_arg_match(env, desc_btf, func_id, regs, btf_mod);
+ if (err < 0)
return err;
+ /* In case of release function, we get register number of refcounted
+ * PTR_TO_BTF_ID back from btf_check_kfunc_arg_match, do the release now
+ */
+ if (err) {
+ err = release_reference(env, regs[err].ref_obj_id);
+ if (err) {
+ verbose(env, "kfunc %s#%d reference has not been acquired before\n",
+ func_name, func_id);
+ return err;
+ }
+ }
for (i = 0; i < CALLER_SAVED_REGS; i++)
mark_reg_not_init(env, regs, caller_saved[i]);
/* Check return type */
t = btf_type_skip_modifiers(desc_btf, func_proto->type, NULL);
+
+ if (acq && !btf_type_is_ptr(t)) {
+ verbose(env, "acquire kernel function does not return PTR_TO_BTF_ID\n");
+ return -EINVAL;
+ }
+
if (btf_type_is_scalar(t)) {
mark_reg_unknown(env, regs, BPF_REG_0);
mark_btf_func_reg_size(env, BPF_REG_0, t->size);
@@ -6808,11 +6833,26 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn)
ptr_type_name);
return -EINVAL;
}
+ if (env->ops->is_kfunc_ret_type_null &&
+ env->ops->is_kfunc_ret_type_null(func_id, btf_mod)) {
+ regs[BPF_REG_0].type = PTR_TO_BTF_ID_OR_NULL;
+ /* For mark_ptr_or_null_reg, see 93c230e3f5bd6 */
+ regs[BPF_REG_0].id = ++env->id_gen;
+ } else {
+ regs[BPF_REG_0].type = PTR_TO_BTF_ID;
+ }
mark_reg_known_zero(env, regs, BPF_REG_0);
regs[BPF_REG_0].btf = desc_btf;
- regs[BPF_REG_0].type = PTR_TO_BTF_ID;
regs[BPF_REG_0].btf_id = ptr_type_id;
mark_btf_func_reg_size(env, BPF_REG_0, sizeof(void *));
+ if (acq) {
+ int id = acquire_reference_state(env, insn_idx);
+
+ if (id < 0)
+ return id;
+ regs[BPF_REG_0].id = id;
+ regs[BPF_REG_0].ref_obj_id = id;
+ }
} /* else { add_kfunc_call() ensures it is btf_type_is_void(t) } */
nargs = btf_type_vlen(func_proto);
@@ -11457,7 +11497,7 @@ static int do_check(struct bpf_verifier_env *env)
if (insn->src_reg == BPF_PSEUDO_CALL)
err = check_func_call(env, insn, &env->insn_idx);
else if (insn->src_reg == BPF_PSEUDO_KFUNC_CALL)
- err = check_kfunc_call(env, insn);
+ err = check_kfunc_call(env, insn, &env->insn_idx);
else
err = check_helper_call(env, insn, &env->insn_idx);
if (err)
--
2.34.1
^ permalink raw reply related
* [PATCH bpf-next v3 6/9] bpf: Track provenance for pointers formed from referenced PTR_TO_BTF_ID
From: Kumar Kartikeya Dwivedi @ 2021-12-10 13:02 UTC (permalink / raw)
To: bpf
Cc: Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
Maxim Mikityanskiy, Florian Westphal, Jesper Dangaard Brouer,
Toke Høiland-Jørgensen, netdev, netfilter-devel
In-Reply-To: <20211210130230.4128676-1-memxor@gmail.com>
In the previous commit, we implemented support in the verifier for
working with referenced PTR_TO_BTF_ID registers. These are invalidated
when their corresponding release function is called.
However, PTR_TO_BTF_ID is a bit special, in that distinct PTR_TO_BTF_ID
can be formed by walking pointers in the struct represented by the BTF
ID. mark_btf_ld_reg will copy the relevant register state to the
destination register.
However, we cannot simply copy ref_obj_id (such that
release_reg_references will match and invalidate all pointers formed by
pointer walking), as we obtain the same BTF ID in the destination
register, leading to confusion during release. An example is show below:
For a type like so:
struct foo { struct foo *next; };
r1 = acquire(...); // BTF ID of struct foo
if (r1) {
r2 = r1->next; // BTF ID of struct foo, and we copied ref_obj_id in
// mark_btf_ld_reg.
release(r2);
}
With this logic, the above snippet succeeds. Hence we need to
distinguish the canonical reference and pointers formed from it.
We introduce a 'parent_ref_obj_id' member in bpf_reg_state, for a
referenced register, only one of ref_obj_id or parent_ref_obj_id may be
set, i.e. either a register holds a canonical reference, or it is
related to a canonical reference for invalidation purposes (contains an
edge pointing to it by way of having the same ref_obj_id in
parent_ref_obj_id, in the graph of objects).
When releasing reference, we ensure that both are not set at once, and
then release if either of them match the requested ref_obj_id to be
released. This ensures that the example given above will not succeed.
A test to this end has been added in later patches.
Typically, kernel objects have a nested object lifetime (where the
parent object 'owns' the objects it holds references to). However, this
is not always true. For now, we don't need support to hold on to
references to objects obtained from a refcounted PTR_TO_BTF_ID after its
release, but this can be relaxed on a case by case basis (i.e. based on
the BTF ID and program type/attach type) in the future.
The safest assumption for the verifier to make in absence of any other
hints, is that all such pointers formed from refcounted PTR_TO_BTF_ID
shall be invalidated.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
---
include/linux/bpf_verifier.h | 10 +++++++
kernel/bpf/verifier.c | 54 ++++++++++++++++++++++++++++--------
2 files changed, 52 insertions(+), 12 deletions(-)
diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index b80fe5bf2a02..a6ef11db6823 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -128,6 +128,16 @@ struct bpf_reg_state {
* allowed and has the same effect as bpf_sk_release(sk).
*/
u32 ref_obj_id;
+ /* This is set for pointers which are derived from referenced
+ * pointer (e.g. PTR_TO_BTF_ID pointer walking), so that the
+ * pointers obtained by walking referenced PTR_TO_BTF_ID
+ * are appropriately invalidated when the lifetime of their
+ * parent object ends.
+ *
+ * Only one of ref_obj_id and parent_ref_obj_id can be set,
+ * never both at once.
+ */
+ u32 parent_ref_obj_id;
/* For scalar types (SCALAR_VALUE), this represents our knowledge of
* the actual value.
* For pointer types, this represents the variable part of the offset
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index b8685fb7ff15..16e30d7f631b 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -654,7 +654,8 @@ static void print_verifier_state(struct bpf_verifier_env *env,
verbose(env, "%s", kernel_type_name(reg->btf, reg->btf_id));
verbose(env, "(id=%d", reg->id);
if (reg_type_may_be_refcounted_or_null(t))
- verbose(env, ",ref_obj_id=%d", reg->ref_obj_id);
+ verbose(env, ",%sref_obj_id=%d", reg->ref_obj_id ? "" : "parent_",
+ reg->ref_obj_id ?: reg->parent_ref_obj_id);
if (t != SCALAR_VALUE)
verbose(env, ",off=%d", reg->off);
if (type_is_pkt_pointer(t))
@@ -1500,7 +1501,8 @@ static void mark_reg_not_init(struct bpf_verifier_env *env,
static void mark_btf_ld_reg(struct bpf_verifier_env *env,
struct bpf_reg_state *regs, u32 regno,
enum bpf_reg_type reg_type,
- struct btf *btf, u32 btf_id)
+ struct btf *btf, u32 btf_id,
+ u32 parent_ref_obj_id)
{
if (reg_type == SCALAR_VALUE) {
mark_reg_unknown(env, regs, regno);
@@ -1509,6 +1511,7 @@ static void mark_btf_ld_reg(struct bpf_verifier_env *env,
mark_reg_known_zero(env, regs, regno);
regs[regno].type = PTR_TO_BTF_ID;
regs[regno].btf = btf;
+ regs[regno].parent_ref_obj_id = parent_ref_obj_id;
regs[regno].btf_id = btf_id;
}
@@ -4136,8 +4139,14 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
if (ret < 0)
return ret;
- if (atype == BPF_READ && value_regno >= 0)
- mark_btf_ld_reg(env, regs, value_regno, ret, reg->btf, btf_id);
+ if (atype == BPF_READ && value_regno >= 0) {
+ if (WARN_ON_ONCE(reg->ref_obj_id && reg->parent_ref_obj_id)) {
+ verbose(env, "verifier internal error: both ref and parent ref set\n");
+ return -EACCES;
+ }
+ mark_btf_ld_reg(env, regs, value_regno, ret, reg->btf, btf_id,
+ reg->ref_obj_id ?: reg->parent_ref_obj_id);
+ }
return 0;
}
@@ -4191,8 +4200,14 @@ static int check_ptr_to_map_access(struct bpf_verifier_env *env,
if (ret < 0)
return ret;
- if (value_regno >= 0)
- mark_btf_ld_reg(env, regs, value_regno, ret, btf_vmlinux, btf_id);
+ if (value_regno >= 0) {
+ if (WARN_ON_ONCE(reg->ref_obj_id && reg->parent_ref_obj_id)) {
+ verbose(env, "verifier internal error: both ref and parent ref set\n");
+ return -EACCES;
+ }
+ mark_btf_ld_reg(env, regs, value_regno, ret, btf_vmlinux, btf_id,
+ reg->ref_obj_id ?: reg->parent_ref_obj_id);
+ }
return 0;
}
@@ -5865,23 +5880,35 @@ static void mark_pkt_end(struct bpf_verifier_state *vstate, int regn, bool range
reg->range = AT_PKT_END;
}
-static void release_reg_references(struct bpf_verifier_env *env,
+static int release_reg_references(struct bpf_verifier_env *env,
struct bpf_func_state *state,
int ref_obj_id)
{
struct bpf_reg_state *regs = state->regs, *reg;
int i;
- for (i = 0; i < MAX_BPF_REG; i++)
- if (regs[i].ref_obj_id == ref_obj_id)
+ for (i = 0; i < MAX_BPF_REG; i++) {
+ if (WARN_ON_ONCE(regs[i].ref_obj_id && regs[i].parent_ref_obj_id)) {
+ verbose(env, "verifier internal error: both ref and parent ref set\n");
+ return -EACCES;
+ }
+ if (regs[i].ref_obj_id == ref_obj_id ||
+ regs[i].parent_ref_obj_id == ref_obj_id)
mark_reg_unknown(env, regs, i);
+ }
bpf_for_each_spilled_reg(i, state, reg) {
if (!reg)
continue;
- if (reg->ref_obj_id == ref_obj_id)
+ if (WARN_ON_ONCE(reg->ref_obj_id && reg->parent_ref_obj_id)) {
+ verbose(env, "verifier internal error: both ref and parent ref set\n");
+ return -EACCES;
+ }
+ if (reg->ref_obj_id == ref_obj_id ||
+ reg->parent_ref_obj_id == ref_obj_id)
__mark_reg_unknown(env, reg);
}
+ return 0;
}
/* The pointer with the specified id has released its reference to kernel
@@ -5898,8 +5925,11 @@ static int release_reference(struct bpf_verifier_env *env,
if (err)
return err;
- for (i = 0; i <= vstate->curframe; i++)
- release_reg_references(env, vstate->frame[i], ref_obj_id);
+ for (i = 0; i <= vstate->curframe; i++) {
+ err = release_reg_references(env, vstate->frame[i], ref_obj_id);
+ if (err)
+ return err;
+ }
return 0;
}
--
2.34.1
^ permalink raw reply related
* [PATCH bpf-next v3 7/9] net/netfilter: Add unstable CT lookup helpers for XDP and TC-BPF
From: Kumar Kartikeya Dwivedi @ 2021-12-10 13:02 UTC (permalink / raw)
To: bpf
Cc: Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
Maxim Mikityanskiy, Florian Westphal, Jesper Dangaard Brouer,
Toke Høiland-Jørgensen, netdev, netfilter-devel
In-Reply-To: <20211210130230.4128676-1-memxor@gmail.com>
This change adds conntrack lookup helpers using the unstable kfunc call
interface for the XDP and TC-BPF hooks. The primary usecase is
implementing a synproxy in XDP, see Maxim's patchset at [0].
Also add acquire/release functions (randomly returning NULL), and also
exercise the PTR_TO_BTF_ID_OR_NULL path so that BPF program caller has
to check for NULL before dereferencing the pointer, for the TC hook.
Introduce kfunc that take various argument types (for PTR_TO_MEM) that
will pass and fail the verifier checks. These will be used in selftests.
Export get_net_ns_by_id as nf_conntrack needs to call it.
Note that we search for acquire, release, and null returning kfuncs in
the intersection of those sets and main set.
This implies that the kfunc_btf_id_list acq_set, rel_set, null_set may
contain BTF ID not in main set, this is explicitly allowed and
recommended (to save on definining more and more sets), since
check_kfunc_call verifier operation would filter out the invalid BTF ID
fairly early, so later checks for acquire, release, and ret_type_null
kfunc will only consider allowed BTF IDs for that program that are
allowed in main set. This is why the nf_conntrack_acq_ids set has BTF
IDs for both xdp and tc hook kfuncs.
[0]: https://lore.kernel.org/bpf/20211019144655.3483197-1-maximmi@nvidia.com
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
---
include/linux/bpf.h | 21 +++
include/linux/btf.h | 31 +++-
kernel/bpf/btf.c | 19 +++
net/bpf/test_run.c | 147 +++++++++++++++++
net/core/filter.c | 27 ++++
net/core/net_namespace.c | 1 +
net/netfilter/nf_conntrack_core.c | 252 ++++++++++++++++++++++++++++++
7 files changed, 497 insertions(+), 1 deletion(-)
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 6a14072c72a0..7b370ed06b1d 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1675,6 +1675,9 @@ int bpf_prog_test_run_sk_lookup(struct bpf_prog *prog,
const union bpf_attr *kattr,
union bpf_attr __user *uattr);
bool bpf_prog_test_check_kfunc_call(u32 kfunc_id, struct module *owner);
+bool bpf_prog_test_is_acquire_kfunc(u32 kfunc_id, struct module *owner);
+bool bpf_prog_test_is_release_kfunc(u32 kfunc_id, struct module *owner);
+bool bpf_prog_test_is_kfunc_ret_type_null(u32 kfunc_id, struct module *owner);
bool btf_ctx_access(int off, int size, enum bpf_access_type type,
const struct bpf_prog *prog,
struct bpf_insn_access_aux *info);
@@ -1933,6 +1936,24 @@ static inline bool bpf_prog_test_check_kfunc_call(u32 kfunc_id,
return false;
}
+static inline bool bpf_prog_test_is_acquire_kfunc(u32 kfunc_id,
+ struct module *owner)
+{
+ return false;
+}
+
+static inline bool bpf_prog_test_is_release_kfunc(u32 kfunc_id,
+ struct module *owner)
+{
+ return false;
+}
+
+static inline bool bpf_prog_test_is_kfunc_ret_type_null(u32 kfunc_id,
+ struct module *owner)
+{
+ return false;
+}
+
static inline void bpf_map_put(struct bpf_map *map)
{
}
diff --git a/include/linux/btf.h b/include/linux/btf.h
index 3730e845d266..bc92825d6e3f 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -321,7 +321,10 @@ static inline const char *btf_name_by_offset(const struct btf *btf,
#endif
enum kfunc_btf_id_set_types {
- BTF_SET_CHECK,
+ BTF_SET_CHECK, /* Allowed kfunc set */
+ BTF_SET_ACQUIRE, /* Acquire kfunc set */
+ BTF_SET_RELEASE, /* Release kfunc set */
+ BTF_SET_RET_NULL, /* kfunc with 'return type PTR_TO_BTF_ID_OR_NULL' set */
__BTF_SET_MAX,
};
@@ -331,6 +334,9 @@ struct kfunc_btf_id_set {
struct btf_id_set *sets[__BTF_SET_MAX];
struct {
struct btf_id_set *set;
+ struct btf_id_set *acq_set;
+ struct btf_id_set *rel_set;
+ struct btf_id_set *null_set;
};
};
struct module *owner;
@@ -345,6 +351,12 @@ void unregister_kfunc_btf_id_set(struct kfunc_btf_id_list *l,
struct kfunc_btf_id_set *s);
bool bpf_check_mod_kfunc_call(struct kfunc_btf_id_list *klist, u32 kfunc_id,
struct module *owner);
+bool bpf_is_mod_acquire_kfunc(struct kfunc_btf_id_list *klist, u32 kfunc_id,
+ struct module *owner);
+bool bpf_is_mod_release_kfunc(struct kfunc_btf_id_list *klist, u32 kfunc_id,
+ struct module *owner);
+bool bpf_is_mod_kfunc_ret_type_null(struct kfunc_btf_id_list *klist,
+ u32 kfunc_id, struct module *owner);
#else
static inline void register_kfunc_btf_id_set(struct kfunc_btf_id_list *l,
struct kfunc_btf_id_set *s)
@@ -359,9 +371,26 @@ static inline bool bpf_check_mod_kfunc_call(struct kfunc_btf_id_list *klist,
{
return false;
}
+static inline bool bpf_is_mod_acquire_kfunc(struct kfunc_btf_id_list *klist,
+ u32 kfunc_id, struct module *owner)
+{
+ return false;
+}
+static inline bool bpf_is_mod_release_kfunc(struct kfunc_btf_id_list *klist,
+ u32 kfunc_id, struct module *owner)
+{
+ return false;
+}
+static inline bool
+bpf_is_mod_kfunc_ret_type_null(struct kfunc_btf_id_list *klist, u32 kfunc_id,
+ struct module *owner)
+{
+ return false;
+}
#endif
extern struct kfunc_btf_id_list bpf_tcp_ca_kfunc_list;
extern struct kfunc_btf_id_list prog_test_kfunc_list;
+extern struct kfunc_btf_id_list xdp_kfunc_list;
#endif
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index a790ba6d93d8..4b955f47b800 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -6539,6 +6539,24 @@ bool bpf_check_mod_kfunc_call(struct kfunc_btf_id_list *klist, u32 kfunc_id,
return kfunc_btf_id_set_contains(klist, kfunc_id, owner, BTF_SET_CHECK);
}
+bool bpf_is_mod_acquire_kfunc(struct kfunc_btf_id_list *klist, u32 kfunc_id,
+ struct module *owner)
+{
+ return kfunc_btf_id_set_contains(klist, kfunc_id, owner, BTF_SET_ACQUIRE);
+}
+
+bool bpf_is_mod_release_kfunc(struct kfunc_btf_id_list *klist, u32 kfunc_id,
+ struct module *owner)
+{
+ return kfunc_btf_id_set_contains(klist, kfunc_id, owner, BTF_SET_RELEASE);
+}
+
+bool bpf_is_mod_kfunc_ret_type_null(struct kfunc_btf_id_list *klist,
+ u32 kfunc_id, struct module *owner)
+{
+ return kfunc_btf_id_set_contains(klist, kfunc_id, owner, BTF_SET_RET_NULL);
+}
+
#endif
#define DEFINE_KFUNC_BTF_ID_LIST(name) \
@@ -6548,6 +6566,7 @@ bool bpf_check_mod_kfunc_call(struct kfunc_btf_id_list *klist, u32 kfunc_id,
DEFINE_KFUNC_BTF_ID_LIST(bpf_tcp_ca_kfunc_list);
DEFINE_KFUNC_BTF_ID_LIST(prog_test_kfunc_list);
+DEFINE_KFUNC_BTF_ID_LIST(xdp_kfunc_list);
int bpf_core_types_are_compat(const struct btf *local_btf, __u32 local_id,
const struct btf *targ_btf, __u32 targ_id)
diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 46dd95755967..779377f4af25 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -232,6 +232,117 @@ struct sock * noinline bpf_kfunc_call_test3(struct sock *sk)
return sk;
}
+struct prog_test_ref_kfunc {
+ int a;
+ int b;
+ struct prog_test_ref_kfunc *next;
+};
+
+static struct prog_test_ref_kfunc prog_test_struct = {
+ .a = 42,
+ .b = 108,
+ .next = &prog_test_struct,
+};
+
+noinline struct prog_test_ref_kfunc *
+bpf_kfunc_call_test_acquire(unsigned long *scalar_ptr)
+{
+ /* randomly return NULL */
+ if (get_jiffies_64() % 2)
+ return NULL;
+ return &prog_test_struct;
+}
+
+noinline void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p)
+{
+}
+
+struct prog_test_pass1 {
+ int x0;
+ struct {
+ int x1;
+ struct {
+ int x2;
+ struct {
+ int x3;
+ struct {
+ int x4;
+ struct {
+ int x5;
+ struct {
+ int x6;
+ struct {
+ int x7;
+ };
+ };
+ };
+ };
+ };
+ };
+ };
+};
+
+struct prog_test_pass2 {
+ int len;
+ short arr1[4];
+ struct {
+ char arr2[4];
+ unsigned long arr3[8];
+ } x;
+};
+
+struct prog_test_fail1 {
+ void *p;
+ int x;
+};
+
+struct prog_test_fail2 {
+ int x8;
+ struct prog_test_pass1 x;
+};
+
+struct prog_test_fail3 {
+ int len;
+ char arr1[2];
+ char arr2[0];
+};
+
+noinline void bpf_kfunc_call_test_pass_ctx(struct __sk_buff *skb)
+{
+}
+
+noinline void bpf_kfunc_call_test_pass1(struct prog_test_pass1 *p)
+{
+}
+
+noinline void bpf_kfunc_call_test_pass2(struct prog_test_pass2 *p)
+{
+}
+
+noinline void bpf_kfunc_call_test_fail1(struct prog_test_fail1 *p)
+{
+}
+
+noinline void bpf_kfunc_call_test_fail2(struct prog_test_fail2 *p)
+{
+}
+
+noinline void bpf_kfunc_call_test_fail3(struct prog_test_fail3 *p)
+{
+}
+
+noinline void bpf_kfunc_call_test_mem_len_pass1(void *mem, int len__mem)
+{
+}
+
+noinline void bpf_kfunc_call_test_mem_len_fail1(void *mem, int len)
+{
+}
+
+noinline void bpf_kfunc_call_test_mem_len_fail2(u64 *mem, int len)
+{
+}
+
__diag_pop();
ALLOW_ERROR_INJECTION(bpf_modify_return_test, ERRNO);
@@ -240,8 +351,23 @@ BTF_SET_START(test_sk_kfunc_ids)
BTF_ID(func, bpf_kfunc_call_test1)
BTF_ID(func, bpf_kfunc_call_test2)
BTF_ID(func, bpf_kfunc_call_test3)
+BTF_ID(func, bpf_kfunc_call_test_acquire)
+BTF_ID(func, bpf_kfunc_call_test_release)
+BTF_ID(func, bpf_kfunc_call_test_pass_ctx)
+BTF_ID(func, bpf_kfunc_call_test_pass1)
+BTF_ID(func, bpf_kfunc_call_test_pass2)
+BTF_ID(func, bpf_kfunc_call_test_fail1)
+BTF_ID(func, bpf_kfunc_call_test_fail2)
+BTF_ID(func, bpf_kfunc_call_test_fail3)
+BTF_ID(func, bpf_kfunc_call_test_mem_len_pass1)
+BTF_ID(func, bpf_kfunc_call_test_mem_len_fail1)
+BTF_ID(func, bpf_kfunc_call_test_mem_len_fail2)
BTF_SET_END(test_sk_kfunc_ids)
+BTF_ID_LIST(test_sk_acq_rel)
+BTF_ID(func, bpf_kfunc_call_test_acquire)
+BTF_ID(func, bpf_kfunc_call_test_release)
+
bool bpf_prog_test_check_kfunc_call(u32 kfunc_id, struct module *owner)
{
if (btf_id_set_contains(&test_sk_kfunc_ids, kfunc_id))
@@ -249,6 +375,27 @@ bool bpf_prog_test_check_kfunc_call(u32 kfunc_id, struct module *owner)
return bpf_check_mod_kfunc_call(&prog_test_kfunc_list, kfunc_id, owner);
}
+bool bpf_prog_test_is_acquire_kfunc(u32 kfunc_id, struct module *owner)
+{
+ if (kfunc_id == test_sk_acq_rel[0])
+ return true;
+ return bpf_is_mod_acquire_kfunc(&prog_test_kfunc_list, kfunc_id, owner);
+}
+
+bool bpf_prog_test_is_release_kfunc(u32 kfunc_id, struct module *owner)
+{
+ if (kfunc_id == test_sk_acq_rel[1])
+ return true;
+ return bpf_is_mod_release_kfunc(&prog_test_kfunc_list, kfunc_id, owner);
+}
+
+bool bpf_prog_test_is_kfunc_ret_type_null(u32 kfunc_id, struct module *owner)
+{
+ if (kfunc_id == test_sk_acq_rel[0])
+ return true;
+ return bpf_is_mod_kfunc_ret_type_null(&prog_test_kfunc_list, kfunc_id, owner);
+}
+
static void *bpf_test_init(const union bpf_attr *kattr, u32 size,
u32 headroom, u32 tailroom)
{
diff --git a/net/core/filter.c b/net/core/filter.c
index fe27c91e3758..ad84ec6e0daa 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -10000,17 +10000,44 @@ const struct bpf_verifier_ops tc_cls_act_verifier_ops = {
.gen_prologue = tc_cls_act_prologue,
.gen_ld_abs = bpf_gen_ld_abs,
.check_kfunc_call = bpf_prog_test_check_kfunc_call,
+ .is_acquire_kfunc = bpf_prog_test_is_acquire_kfunc,
+ .is_release_kfunc = bpf_prog_test_is_release_kfunc,
+ .is_kfunc_ret_type_null = bpf_prog_test_is_kfunc_ret_type_null,
};
const struct bpf_prog_ops tc_cls_act_prog_ops = {
.test_run = bpf_prog_test_run_skb,
};
+static bool xdp_check_kfunc_call(u32 kfunc_id, struct module *owner)
+{
+ return bpf_check_mod_kfunc_call(&xdp_kfunc_list, kfunc_id, owner);
+}
+
+static bool xdp_is_acquire_kfunc(u32 kfunc_id, struct module *owner)
+{
+ return bpf_is_mod_acquire_kfunc(&xdp_kfunc_list, kfunc_id, owner);
+}
+
+static bool xdp_is_release_kfunc(u32 kfunc_id, struct module *owner)
+{
+ return bpf_is_mod_release_kfunc(&xdp_kfunc_list, kfunc_id, owner);
+}
+
+static bool xdp_is_kfunc_ret_type_null(u32 kfunc_id, struct module *owner)
+{
+ return bpf_is_mod_kfunc_ret_type_null(&xdp_kfunc_list, kfunc_id, owner);
+}
+
const struct bpf_verifier_ops xdp_verifier_ops = {
.get_func_proto = xdp_func_proto,
.is_valid_access = xdp_is_valid_access,
.convert_ctx_access = xdp_convert_ctx_access,
.gen_prologue = bpf_noop_prologue,
+ .check_kfunc_call = xdp_check_kfunc_call,
+ .is_acquire_kfunc = xdp_is_acquire_kfunc,
+ .is_release_kfunc = xdp_is_release_kfunc,
+ .is_kfunc_ret_type_null = xdp_is_kfunc_ret_type_null,
};
const struct bpf_prog_ops xdp_prog_ops = {
diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index 202fa5eacd0f..7b4bfe793002 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -299,6 +299,7 @@ struct net *get_net_ns_by_id(const struct net *net, int id)
return peer;
}
+EXPORT_SYMBOL_GPL(get_net_ns_by_id);
/*
* setup_net runs the initializers for the network namespace object.
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 770a63103c7a..85042cb6f82e 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -11,6 +11,9 @@
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#include <linux/bpf.h>
+#include <linux/btf.h>
+#include <linux/btf_ids.h>
#include <linux/types.h>
#include <linux/netfilter.h>
#include <linux/module.h>
@@ -2451,6 +2454,249 @@ static int kill_all(struct nf_conn *i, void *data)
return net_eq(nf_ct_net(i), data);
}
+/* Unstable Kernel Helpers for XDP and TC-BPF hook
+ *
+ * These are called from the XDP and SCHED_CLS BPF programs. Note that it is
+ * allowed to break compatibility for these functions since the interface they
+ * are exposed through to BPF programs is explicitly unstable.
+ */
+static struct nf_conn *__bpf_nf_ct_lookup(struct net *net,
+ struct bpf_sock_tuple *bpf_tuple,
+ u32 tuple_len, u8 protonum,
+ s32 netns_id)
+{
+ struct nf_conntrack_tuple_hash *hash;
+ struct nf_conntrack_tuple tuple;
+
+ if (unlikely(protonum != IPPROTO_TCP && protonum != IPPROTO_UDP))
+ return ERR_PTR(-EPROTO);
+ if (unlikely(netns_id < BPF_F_CURRENT_NETNS))
+ return ERR_PTR(-EINVAL);
+
+ memset(&tuple, 0, sizeof(tuple));
+ switch (tuple_len) {
+ case sizeof(bpf_tuple->ipv4):
+ tuple.src.l3num = AF_INET;
+ tuple.src.u3.ip = bpf_tuple->ipv4.saddr;
+ tuple.src.u.tcp.port = bpf_tuple->ipv4.sport;
+ tuple.dst.u3.ip = bpf_tuple->ipv4.daddr;
+ tuple.dst.u.tcp.port = bpf_tuple->ipv4.dport;
+ break;
+ case sizeof(bpf_tuple->ipv6):
+ tuple.src.l3num = AF_INET6;
+ memcpy(tuple.src.u3.ip6, bpf_tuple->ipv6.saddr, sizeof(bpf_tuple->ipv6.saddr));
+ tuple.src.u.tcp.port = bpf_tuple->ipv6.sport;
+ memcpy(tuple.dst.u3.ip6, bpf_tuple->ipv6.daddr, sizeof(bpf_tuple->ipv6.daddr));
+ tuple.dst.u.tcp.port = bpf_tuple->ipv6.dport;
+ break;
+ default:
+ return ERR_PTR(-EAFNOSUPPORT);
+ }
+
+ tuple.dst.protonum = protonum;
+
+ if (netns_id >= 0) {
+ net = get_net_ns_by_id(net, netns_id);
+ if (unlikely(!net))
+ return ERR_PTR(-ENONET);
+ }
+
+ hash = nf_conntrack_find_get(net, &nf_ct_zone_dflt, &tuple);
+ if (netns_id >= 0)
+ put_net(net);
+ if (!hash)
+ return ERR_PTR(-ENOENT);
+ return nf_ct_tuplehash_to_ctrack(hash);
+}
+
+/* bpf_ct_opts - Options for CT lookup helpers
+ *
+ * Members:
+ * @error - Out parameter, set for any errors encountered
+ * Values:
+ * -EINVAL - Passed NULL for bpf_tuple pointer
+ * -EINVAL - opts->reserved is not 0
+ * -EINVAL - netns_id is less than -1
+ * -EINVAL - len__opts isn't NF_BPF_CT_OPTS_SZ (12)
+ * -EPROTO - l4proto isn't one of IPPROTO_TCP or IPPROTO_UDP
+ * -ENONET - No network namespace found for netns_id
+ * -ENOENT - Conntrack lookup could not find entry for tuple
+ * -EAFNOSUPPORT - len__tuple isn't one of sizeof(tuple->ipv4)
+ * or sizeof(tuple->ipv6)
+ * @l4proto - Layer 4 protocol
+ * Values:
+ * IPPROTO_TCP, IPPROTO_UDP
+ * @reserved - Reserved member, will be reused for more options in future
+ * Values:
+ * 0
+ * @netns_id - Specify the network namespace for lookup
+ * Values:
+ * BPF_F_CURRENT_NETNS (-1)
+ * Use namespace associated with ctx (xdp_md, __sk_buff)
+ * [0, S32_MAX]
+ * Network Namespace ID
+ */
+struct bpf_ct_opts {
+ s32 netns_id;
+ s32 error;
+ u8 l4proto;
+ u8 reserved[3];
+};
+
+enum {
+ NF_BPF_CT_OPTS_SZ = 12,
+};
+
+/* bpf_xdp_ct_lookup - Lookup CT entry for the given tuple, and acquire a
+ * reference to it
+ *
+ * Parameters:
+ * @xdp_ctx - Pointer to ctx (xdp_md) in XDP program
+ * Cannot be NULL
+ * @bpf_tuple - Pointer to memory representing the tuple to look up
+ * Cannot be NULL
+ * @len__tuple - Length of the tuple structure
+ * Must be one of sizeof(bpf_tuple->ipv4) or
+ * sizeof(bpf_tuple->ipv6)
+ * @opts - Additional options for lookup (documented above)
+ * Cannot be NULL
+ * @len__opts - Length of the bpf_ct_opts structure
+ * Must be NF_BPF_CT_OPTS_SZ (12)
+ */
+struct nf_conn *bpf_xdp_ct_lookup(struct xdp_md *xdp_ctx,
+ struct bpf_sock_tuple *bpf_tuple,
+ u32 len__tuple, struct bpf_ct_opts *opts,
+ u32 len__opts)
+{
+ struct xdp_buff *ctx = (struct xdp_buff *)xdp_ctx;
+ struct net *caller_net;
+ struct nf_conn *nfct;
+
+ BUILD_BUG_ON(sizeof(struct bpf_ct_opts) != NF_BPF_CT_OPTS_SZ);
+
+ if (!opts)
+ return NULL;
+ if (!bpf_tuple || opts->reserved[0] || opts->reserved[1] ||
+ opts->reserved[2] || len__opts != NF_BPF_CT_OPTS_SZ) {
+ opts->error = -EINVAL;
+ return NULL;
+ }
+ caller_net = dev_net(ctx->rxq->dev);
+ nfct = __bpf_nf_ct_lookup(caller_net, bpf_tuple, len__tuple, opts->l4proto,
+ opts->netns_id);
+ if (IS_ERR(nfct)) {
+ opts->error = PTR_ERR(nfct);
+ return NULL;
+ }
+ return nfct;
+}
+
+/* bpf_skb_ct_lookup - Lookup CT entry for the given tuple, and acquire a
+ * reference to it
+ *
+ * Parameters:
+ * @skb_ctx - Pointer to ctx (__sk_buff) in TC program
+ * Cannot be NULL
+ * @bpf_tuple - Pointer to memory representing the tuple to look up
+ * Cannot be NULL
+ * @len__tuple - Length of the tuple structure
+ * Must be one of sizeof(bpf_tuple->ipv4) or
+ * sizeof(bpf_tuple->ipv6)
+ * @opts - Additional options for lookup (documented above)
+ * Cannot be NULL
+ * @len__opts - Length of the bpf_ct_opts structure
+ * Must be NF_BPF_CT_OPTS_SZ (12)
+ */
+struct nf_conn *bpf_skb_ct_lookup(struct __sk_buff *skb_ctx,
+ struct bpf_sock_tuple *bpf_tuple,
+ u32 len__tuple, struct bpf_ct_opts *opts,
+ u32 len__opts)
+{
+ struct sk_buff *skb = (struct sk_buff *)skb_ctx;
+ struct net *caller_net;
+ struct nf_conn *nfct;
+
+ BUILD_BUG_ON(sizeof(struct bpf_ct_opts) != NF_BPF_CT_OPTS_SZ);
+
+ if (!opts)
+ return NULL;
+ if (!bpf_tuple || opts->reserved[0] || opts->reserved[1] ||
+ opts->reserved[2] || len__opts != NF_BPF_CT_OPTS_SZ) {
+ opts->error = -EINVAL;
+ return NULL;
+ }
+ caller_net = skb->dev ? dev_net(skb->dev) : sock_net(skb->sk);
+ nfct = __bpf_nf_ct_lookup(caller_net, bpf_tuple, len__tuple, opts->l4proto,
+ opts->netns_id);
+ if (IS_ERR(nfct)) {
+ opts->error = PTR_ERR(nfct);
+ return NULL;
+ }
+ return nfct;
+}
+
+/* bpf_ct_release - Release acquired nf_conn object
+ *
+ * This must be invoked for referenced PTR_TO_BTF_ID, and the verifier rejects
+ * the program if any references remain in the program in all of the explored
+ * states.
+ *
+ * Parameters:
+ * @nf_conn - Pointer to referenced nf_conn object, obtained using
+ * bpf_xdp_ct_lookup or bpf_skb_ct_lookup.
+ */
+void bpf_ct_release(struct nf_conn *nfct)
+{
+ if (!nfct)
+ return;
+ nf_ct_put(nfct);
+}
+
+/* kfuncs that may return NULL PTR_TO_BTF_ID */
+BTF_SET_START(nf_conntrack_null_ids)
+BTF_ID(func, bpf_xdp_ct_lookup)
+BTF_ID(func, bpf_skb_ct_lookup)
+BTF_SET_END(nf_conntrack_null_ids)
+
+/* XDP hook allowed kfuncs */
+BTF_SET_START(nf_conntrack_xdp_ids)
+BTF_ID(func, bpf_xdp_ct_lookup)
+BTF_ID(func, bpf_ct_release)
+BTF_SET_END(nf_conntrack_xdp_ids)
+
+/* TC-BPF hook allowed kfuncs */
+BTF_SET_START(nf_conntrack_skb_ids)
+BTF_ID(func, bpf_skb_ct_lookup)
+BTF_ID(func, bpf_ct_release)
+BTF_SET_END(nf_conntrack_skb_ids)
+
+/* XDP and TC-BPF hook acquire kfuncs */
+BTF_SET_START(nf_conntrack_acq_ids)
+BTF_ID(func, bpf_xdp_ct_lookup)
+BTF_ID(func, bpf_skb_ct_lookup)
+BTF_SET_END(nf_conntrack_acq_ids)
+
+/* XDP and TC-BPF hook release kfuncs */
+BTF_SET_START(nf_conntrack_rel_ids)
+BTF_ID(func, bpf_ct_release)
+BTF_SET_END(nf_conntrack_rel_ids)
+
+static struct kfunc_btf_id_set nf_ct_xdp_kfunc_set = {
+ .owner = THIS_MODULE,
+ .set = &nf_conntrack_xdp_ids,
+ .acq_set = &nf_conntrack_acq_ids,
+ .rel_set = &nf_conntrack_rel_ids,
+ .null_set = &nf_conntrack_null_ids,
+};
+
+static struct kfunc_btf_id_set nf_ct_skb_kfunc_set = {
+ .owner = THIS_MODULE,
+ .set = &nf_conntrack_skb_ids,
+ .acq_set = &nf_conntrack_acq_ids,
+ .rel_set = &nf_conntrack_rel_ids,
+ .null_set = &nf_conntrack_null_ids,
+};
+
void nf_conntrack_cleanup_start(void)
{
conntrack_gc_work.exiting = true;
@@ -2459,6 +2705,9 @@ void nf_conntrack_cleanup_start(void)
void nf_conntrack_cleanup_end(void)
{
+ unregister_kfunc_btf_id_set(&xdp_kfunc_list, &nf_ct_xdp_kfunc_set);
+ unregister_kfunc_btf_id_set(&prog_test_kfunc_list, &nf_ct_skb_kfunc_set);
+
RCU_INIT_POINTER(nf_ct_hook, NULL);
cancel_delayed_work_sync(&conntrack_gc_work.dwork);
kvfree(nf_conntrack_hash);
@@ -2745,6 +2994,9 @@ int nf_conntrack_init_start(void)
conntrack_gc_work_init(&conntrack_gc_work);
queue_delayed_work(system_power_efficient_wq, &conntrack_gc_work.dwork, HZ);
+ register_kfunc_btf_id_set(&prog_test_kfunc_list, &nf_ct_skb_kfunc_set);
+ register_kfunc_btf_id_set(&xdp_kfunc_list, &nf_ct_xdp_kfunc_set);
+
return 0;
err_proto:
--
2.34.1
^ permalink raw reply related
* [PATCH bpf-next v3 8/9] selftests/bpf: Extend kfunc selftests
From: Kumar Kartikeya Dwivedi @ 2021-12-10 13:02 UTC (permalink / raw)
To: bpf
Cc: Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
Maxim Mikityanskiy, Florian Westphal, Jesper Dangaard Brouer,
Toke Høiland-Jørgensen, netdev, netfilter-devel
In-Reply-To: <20211210130230.4128676-1-memxor@gmail.com>
Use the prog_test kfuncs to test the referenced PTR_TO_BTF_ID kfunc
support, and PTR_TO_CTX, PTR_TO_MEM argument passing support. Also
testing the various failure cases.
The failure selftests will test the following cases for kfunc:
kfunc_call_test_fail1 - Argument struct type has non-scalar member
kfunc_call_test_fail2 - Nesting depth of type > 8
kfunc_call_test_fail3 - Struct type has trailing zero-sized FAM
kfunc_call_test_fail4 - Trying to pass reg->type != PTR_TO_CTX when
argument struct type is a ctx type
kfunc_call_test_fail5 - void * not part of mem, len pair
kfunc_call_test_fail6 - u64 * not part of mem, len pair
kfunc_call_test_fail7 - mark_btf_ld_reg copies ref_obj_id
kfunc_call_test_fail8 - Same type btf_struct_walk reference copy handled
correctly during release (i.e. only parent
object can be released)
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
---
.../selftests/bpf/prog_tests/kfunc_call.c | 28 ++++++++++
.../selftests/bpf/progs/kfunc_call_test.c | 52 ++++++++++++++++++-
.../bpf/progs/kfunc_call_test_fail1.c | 16 ++++++
.../bpf/progs/kfunc_call_test_fail2.c | 16 ++++++
.../bpf/progs/kfunc_call_test_fail3.c | 16 ++++++
.../bpf/progs/kfunc_call_test_fail4.c | 16 ++++++
.../bpf/progs/kfunc_call_test_fail5.c | 16 ++++++
.../bpf/progs/kfunc_call_test_fail6.c | 16 ++++++
.../bpf/progs/kfunc_call_test_fail7.c | 24 +++++++++
.../bpf/progs/kfunc_call_test_fail8.c | 22 ++++++++
10 files changed, 220 insertions(+), 2 deletions(-)
create mode 100644 tools/testing/selftests/bpf/progs/kfunc_call_test_fail1.c
create mode 100644 tools/testing/selftests/bpf/progs/kfunc_call_test_fail2.c
create mode 100644 tools/testing/selftests/bpf/progs/kfunc_call_test_fail3.c
create mode 100644 tools/testing/selftests/bpf/progs/kfunc_call_test_fail4.c
create mode 100644 tools/testing/selftests/bpf/progs/kfunc_call_test_fail5.c
create mode 100644 tools/testing/selftests/bpf/progs/kfunc_call_test_fail6.c
create mode 100644 tools/testing/selftests/bpf/progs/kfunc_call_test_fail7.c
create mode 100644 tools/testing/selftests/bpf/progs/kfunc_call_test_fail8.c
diff --git a/tools/testing/selftests/bpf/prog_tests/kfunc_call.c b/tools/testing/selftests/bpf/prog_tests/kfunc_call.c
index 7d7445ccc141..b6630b9427d0 100644
--- a/tools/testing/selftests/bpf/prog_tests/kfunc_call.c
+++ b/tools/testing/selftests/bpf/prog_tests/kfunc_call.c
@@ -5,11 +5,33 @@
#include "kfunc_call_test.lskel.h"
#include "kfunc_call_test_subprog.skel.h"
#include "kfunc_call_test_subprog.lskel.h"
+#include "kfunc_call_test_fail1.skel.h"
+#include "kfunc_call_test_fail2.skel.h"
+#include "kfunc_call_test_fail3.skel.h"
+#include "kfunc_call_test_fail4.skel.h"
+#include "kfunc_call_test_fail5.skel.h"
+#include "kfunc_call_test_fail6.skel.h"
+#include "kfunc_call_test_fail7.skel.h"
+#include "kfunc_call_test_fail8.skel.h"
static void test_main(void)
{
struct kfunc_call_test_lskel *skel;
int prog_fd, retval, err;
+ void *fskel;
+
+#define FAIL(nr) \
+ ({ \
+ fskel = kfunc_call_test_fail##nr##__open_and_load(); \
+ if (!ASSERT_EQ(fskel, NULL, \
+ "kfunc_call_test_fail" #nr \
+ "__open_and_load")) { \
+ kfunc_call_test_fail##nr##__destroy(fskel); \
+ return; \
+ } \
+ })
+
+ FAIL(1); FAIL(2); FAIL(3); FAIL(4); FAIL(5); FAIL(6); FAIL(7); FAIL(8);
skel = kfunc_call_test_lskel__open_and_load();
if (!ASSERT_OK_PTR(skel, "skel"))
@@ -27,6 +49,12 @@ static void test_main(void)
ASSERT_OK(err, "bpf_prog_test_run(test2)");
ASSERT_EQ(retval, 3, "test2-retval");
+ prog_fd = skel->progs.kfunc_call_test_ref_btf_id.prog_fd;
+ err = bpf_prog_test_run(prog_fd, 1, &pkt_v4, sizeof(pkt_v4),
+ NULL, NULL, (__u32 *)&retval, NULL);
+ ASSERT_OK(err, "bpf_prog_test_run(test_ref_btf_id)");
+ ASSERT_EQ(retval, 0, "test_ref_btf_id-retval");
+
kfunc_call_test_lskel__destroy(skel);
}
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_test.c b/tools/testing/selftests/bpf/progs/kfunc_call_test.c
index 8a8cf59017aa..5aecbb9fdc68 100644
--- a/tools/testing/selftests/bpf/progs/kfunc_call_test.c
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_test.c
@@ -1,13 +1,20 @@
// SPDX-License-Identifier: GPL-2.0
/* Copyright (c) 2021 Facebook */
-#include <linux/bpf.h>
+#include <vmlinux.h>
#include <bpf/bpf_helpers.h>
-#include "bpf_tcp_helpers.h"
extern int bpf_kfunc_call_test2(struct sock *sk, __u32 a, __u32 b) __ksym;
extern __u64 bpf_kfunc_call_test1(struct sock *sk, __u32 a, __u64 b,
__u32 c, __u64 d) __ksym;
+extern struct prog_test_ref_kfunc *bpf_kfunc_call_test_acquire(unsigned long *sp) __ksym;
+extern void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym;
+extern void bpf_kfunc_call_test_pass_ctx(struct __sk_buff *skb) __ksym;
+extern void bpf_kfunc_call_test_pass1(struct prog_test_pass1 *p) __ksym;
+extern void bpf_kfunc_call_test_pass2(struct prog_test_pass2 *p) __ksym;
+extern void bpf_kfunc_call_test_mem_len_pass1(void *mem, int len) __ksym;
+extern void bpf_kfunc_call_test_mem_len_fail2(__u64 *mem, int len) __ksym;
+
SEC("tc")
int kfunc_call_test2(struct __sk_buff *skb)
{
@@ -44,4 +51,45 @@ int kfunc_call_test1(struct __sk_buff *skb)
return ret;
}
+SEC("tc")
+int kfunc_call_test_ref_btf_id(struct __sk_buff *skb)
+{
+ struct prog_test_ref_kfunc *pt;
+ unsigned long s = 0;
+ int ret = 0;
+
+ pt = bpf_kfunc_call_test_acquire(&s);
+ if (pt) {
+ if (pt->a != 42 || pt->b != 108)
+ ret = -1;
+ bpf_kfunc_call_test_release(pt);
+ }
+ return ret;
+}
+
+SEC("tc")
+int kfunc_call_test_pass(struct __sk_buff *skb)
+{
+ struct prog_test_pass1 p1 = {};
+ struct prog_test_pass2 p2 = {};
+ short a = 0;
+ __u64 b = 0;
+ long c = 0;
+ char d = 0;
+ int e = 0;
+
+ bpf_kfunc_call_test_pass_ctx(skb);
+ bpf_kfunc_call_test_pass1(&p1);
+ bpf_kfunc_call_test_pass2(&p2);
+
+ bpf_kfunc_call_test_mem_len_pass1(&a, sizeof(a));
+ bpf_kfunc_call_test_mem_len_pass1(&b, sizeof(b));
+ bpf_kfunc_call_test_mem_len_pass1(&c, sizeof(c));
+ bpf_kfunc_call_test_mem_len_pass1(&d, sizeof(d));
+ bpf_kfunc_call_test_mem_len_pass1(&e, sizeof(e));
+ bpf_kfunc_call_test_mem_len_fail2(&b, -1);
+
+ return 0;
+}
+
char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_test_fail1.c b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail1.c
new file mode 100644
index 000000000000..4088000dcfc0
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail1.c
@@ -0,0 +1,16 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+extern void bpf_kfunc_call_test_fail1(struct prog_test_fail1 *p) __ksym;
+
+SEC("tc")
+int kfunc_call_test_fail1(struct __sk_buff *skb)
+{
+ struct prog_test_fail1 s = {};
+
+ bpf_kfunc_call_test_fail1(&s);
+ return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_test_fail2.c b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail2.c
new file mode 100644
index 000000000000..0c9779693576
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail2.c
@@ -0,0 +1,16 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+extern void bpf_kfunc_call_test_fail2(struct prog_test_fail2 *p) __ksym;
+
+SEC("tc")
+int kfunc_call_test_fail2(struct __sk_buff *skb)
+{
+ struct prog_test_fail2 s = {};
+
+ bpf_kfunc_call_test_fail2(&s);
+ return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_test_fail3.c b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail3.c
new file mode 100644
index 000000000000..4e5a7493cdf7
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail3.c
@@ -0,0 +1,16 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+extern void bpf_kfunc_call_test_fail3(struct prog_test_fail3 *p) __ksym;
+
+SEC("tc")
+int kfunc_call_test_fail3(struct __sk_buff *skb)
+{
+ struct prog_test_fail3 s = {};
+
+ bpf_kfunc_call_test_fail3(&s);
+ return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_test_fail4.c b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail4.c
new file mode 100644
index 000000000000..01c3523c7c50
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail4.c
@@ -0,0 +1,16 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+extern void bpf_kfunc_call_test_pass_ctx(struct __sk_buff *skb) __ksym;
+
+SEC("tc")
+int kfunc_call_test_fail4(struct __sk_buff *skb)
+{
+ struct __sk_buff local_skb = {};
+
+ bpf_kfunc_call_test_pass_ctx(&local_skb);
+ return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_test_fail5.c b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail5.c
new file mode 100644
index 000000000000..e32f13709357
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail5.c
@@ -0,0 +1,16 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+extern void bpf_kfunc_call_test_mem_len_fail1(void *mem, int len) __ksym;
+
+SEC("tc")
+int kfunc_call_test_fail5(struct __sk_buff *skb)
+{
+ int a = 0;
+
+ bpf_kfunc_call_test_mem_len_fail1(&a, sizeof(a));
+ return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_test_fail6.c b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail6.c
new file mode 100644
index 000000000000..998626aaca35
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail6.c
@@ -0,0 +1,16 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+extern void bpf_kfunc_call_test_mem_len_fail2(__u64 *mem, int len) __ksym;
+
+SEC("tc")
+int kfunc_call_test_fail6(struct __sk_buff *skb)
+{
+ int a = 0;
+
+ bpf_kfunc_call_test_mem_len_fail2((void *)&a, sizeof(a));
+ return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_test_fail7.c b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail7.c
new file mode 100644
index 000000000000..05d4914b0533
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail7.c
@@ -0,0 +1,24 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+extern struct prog_test_ref_kfunc *bpf_kfunc_call_test_acquire(unsigned long *sp) __ksym;
+extern void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym;
+
+SEC("tc")
+int kfunc_call_test_fail7(struct __sk_buff *skb)
+{
+ struct prog_test_ref_kfunc *p, *p2;
+ unsigned long sp = 0;
+
+ p = bpf_kfunc_call_test_acquire(&sp);
+ if (p) {
+ p2 = p->next->next;
+ bpf_kfunc_call_test_release(p);
+ if (p2->a == 42)
+ return 1;
+ }
+ return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_test_fail8.c b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail8.c
new file mode 100644
index 000000000000..eac8637ce841
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_test_fail8.c
@@ -0,0 +1,22 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+extern struct prog_test_ref_kfunc *bpf_kfunc_call_test_acquire(unsigned long *sp) __ksym;
+extern void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym;
+
+SEC("tc")
+int kfunc_call_test_fail8(struct __sk_buff *skb)
+{
+ struct prog_test_ref_kfunc *p, *p2;
+ unsigned long sp = 0;
+
+ p = bpf_kfunc_call_test_acquire(&sp);
+ if (p) {
+ p2 = p->next->next;
+ bpf_kfunc_call_test_release(p2);
+ }
+ return 0;
+}
+
+char _license[] SEC("license") = "GPL";
--
2.34.1
^ permalink raw reply related
* [PATCH bpf-next v3 9/9] selftests/bpf: Add test for unstable CT lookup API
From: Kumar Kartikeya Dwivedi @ 2021-12-10 13:02 UTC (permalink / raw)
To: bpf
Cc: Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
Maxim Mikityanskiy, Florian Westphal, Jesper Dangaard Brouer,
Toke Høiland-Jørgensen, netdev, netfilter-devel
In-Reply-To: <20211210130230.4128676-1-memxor@gmail.com>
This tests that we return errors as documented, and also that the kfunc
calls work from both XDP and TC hooks.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
---
tools/testing/selftests/bpf/config | 4 +
.../testing/selftests/bpf/prog_tests/bpf_nf.c | 48 ++++++++
.../testing/selftests/bpf/progs/test_bpf_nf.c | 113 ++++++++++++++++++
3 files changed, 165 insertions(+)
create mode 100644 tools/testing/selftests/bpf/prog_tests/bpf_nf.c
create mode 100644 tools/testing/selftests/bpf/progs/test_bpf_nf.c
diff --git a/tools/testing/selftests/bpf/config b/tools/testing/selftests/bpf/config
index 5192305159ec..4a2a47fcd6ef 100644
--- a/tools/testing/selftests/bpf/config
+++ b/tools/testing/selftests/bpf/config
@@ -46,3 +46,7 @@ CONFIG_IMA_READ_POLICY=y
CONFIG_BLK_DEV_LOOP=y
CONFIG_FUNCTION_TRACER=y
CONFIG_DYNAMIC_FTRACE=y
+CONFIG_NETFILTER=y
+CONFIG_NF_DEFRAG_IPV4=y
+CONFIG_NF_DEFRAG_IPV6=y
+CONFIG_NF_CONNTRACK=y
diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_nf.c b/tools/testing/selftests/bpf/prog_tests/bpf_nf.c
new file mode 100644
index 000000000000..56e8d745b8c8
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_nf.c
@@ -0,0 +1,48 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <test_progs.h>
+#include <network_helpers.h>
+#include "test_bpf_nf.skel.h"
+
+enum {
+ TEST_XDP,
+ TEST_TC_BPF,
+};
+
+void test_bpf_nf_ct(int mode)
+{
+ struct test_bpf_nf *skel;
+ int prog_fd, err, retval;
+
+ skel = test_bpf_nf__open_and_load();
+ if (!ASSERT_OK_PTR(skel, "test_bpf_nf__open_and_load"))
+ return;
+
+ if (mode == TEST_XDP)
+ prog_fd = bpf_program__fd(skel->progs.nf_xdp_ct_test);
+ else
+ prog_fd = bpf_program__fd(skel->progs.nf_skb_ct_test);
+
+ err = bpf_prog_test_run(prog_fd, 1, &pkt_v4, sizeof(pkt_v4), NULL, NULL,
+ (__u32 *)&retval, NULL);
+ if (!ASSERT_OK(err, "bpf_prog_test_run"))
+ goto end;
+
+ ASSERT_EQ(skel->bss->test_einval_bpf_tuple, -EINVAL, "Test EINVAL for NULL bpf_tuple");
+ ASSERT_EQ(skel->bss->test_einval_reserved, -EINVAL, "Test EINVAL for reserved not set to 0");
+ ASSERT_EQ(skel->bss->test_einval_netns_id, -EINVAL, "Test EINVAL for netns_id < -1");
+ ASSERT_EQ(skel->bss->test_einval_len_opts, -EINVAL, "Test EINVAL for len__opts != NF_BPF_CT_OPTS_SZ");
+ ASSERT_EQ(skel->bss->test_eproto_l4proto, -EPROTO, "Test EPROTO for l4proto != TCP or UDP");
+ ASSERT_EQ(skel->bss->test_enonet_netns_id, -ENONET, "Test ENONET for bad but valid netns_id");
+ ASSERT_EQ(skel->bss->test_enoent_lookup, -ENOENT, "Test ENOENT for failed lookup");
+ ASSERT_EQ(skel->bss->test_eafnosupport, -EAFNOSUPPORT,"Test EAFNOSUPPORT for invalid len__tuple");
+end:
+ test_bpf_nf__destroy(skel);
+}
+
+void test_bpf_nf(void)
+{
+ if (test__start_subtest("xdp-ct"))
+ test_bpf_nf_ct(TEST_XDP);
+ if (test__start_subtest("tc-bpf-ct"))
+ test_bpf_nf_ct(TEST_TC_BPF);
+}
diff --git a/tools/testing/selftests/bpf/progs/test_bpf_nf.c b/tools/testing/selftests/bpf/progs/test_bpf_nf.c
new file mode 100644
index 000000000000..7cfff245b24f
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_bpf_nf.c
@@ -0,0 +1,113 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+#define EAFNOSUPPORT 97
+#define EPROTO 71
+#define ENONET 64
+#define EINVAL 22
+#define ENOENT 2
+
+int test_einval_bpf_tuple = 0;
+int test_einval_reserved = 0;
+int test_einval_netns_id = 0;
+int test_einval_len_opts = 0;
+int test_eproto_l4proto = 0;
+int test_enonet_netns_id = 0;
+int test_enoent_lookup = 0;
+int test_eafnosupport = 0;
+
+struct nf_conn *bpf_xdp_ct_lookup(struct xdp_md *, struct bpf_sock_tuple *, u32,
+ struct bpf_ct_opts *, u32) __weak __ksym;
+struct nf_conn *bpf_skb_ct_lookup(struct __sk_buff *, struct bpf_sock_tuple *, u32,
+ struct bpf_ct_opts *, u32) __weak __ksym;
+void bpf_ct_release(struct nf_conn *) __weak __ksym;
+
+#define nf_ct_test(func, ctx) \
+ ({ \
+ struct bpf_ct_opts opts_def = { .l4proto = IPPROTO_TCP, \
+ .netns_id = -1 }; \
+ struct bpf_sock_tuple bpf_tuple; \
+ struct nf_conn *ct; \
+ \
+ __builtin_memset(&bpf_tuple, 0, sizeof(bpf_tuple.ipv4)); \
+ ct = func(ctx, NULL, 0, &opts_def, sizeof(opts_def)); \
+ if (ct) \
+ bpf_ct_release(ct); \
+ else \
+ test_einval_bpf_tuple = opts_def.error; \
+ \
+ opts_def.reserved[0] = 1; \
+ ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, \
+ sizeof(opts_def)); \
+ opts_def.reserved[0] = 0; \
+ opts_def.l4proto = IPPROTO_TCP; \
+ if (ct) \
+ bpf_ct_release(ct); \
+ else \
+ test_einval_reserved = opts_def.error; \
+ \
+ opts_def.netns_id = -2; \
+ ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, \
+ sizeof(opts_def)); \
+ opts_def.netns_id = -1; \
+ if (ct) \
+ bpf_ct_release(ct); \
+ else \
+ test_einval_netns_id = opts_def.error; \
+ \
+ ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, \
+ sizeof(opts_def) - 1); \
+ if (ct) \
+ bpf_ct_release(ct); \
+ else \
+ test_einval_len_opts = opts_def.error; \
+ \
+ opts_def.l4proto = IPPROTO_ICMP; \
+ ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, \
+ sizeof(opts_def)); \
+ opts_def.l4proto = IPPROTO_TCP; \
+ if (ct) \
+ bpf_ct_release(ct); \
+ else \
+ test_eproto_l4proto = opts_def.error; \
+ \
+ opts_def.netns_id = 0xf00f; \
+ ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, \
+ sizeof(opts_def)); \
+ opts_def.netns_id = -1; \
+ if (ct) \
+ bpf_ct_release(ct); \
+ else \
+ test_enonet_netns_id = opts_def.error; \
+ \
+ ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, \
+ sizeof(opts_def)); \
+ if (ct) \
+ bpf_ct_release(ct); \
+ else \
+ test_enoent_lookup = opts_def.error; \
+ \
+ ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4) - 1, \
+ &opts_def, sizeof(opts_def)); \
+ if (ct) \
+ bpf_ct_release(ct); \
+ else \
+ test_eafnosupport = opts_def.error; \
+ })
+
+SEC("xdp")
+int nf_xdp_ct_test(struct xdp_md *ctx)
+{
+ nf_ct_test(bpf_xdp_ct_lookup, ctx);
+ return 0;
+}
+
+SEC("tc")
+int nf_skb_ct_test(struct __sk_buff *ctx)
+{
+ nf_ct_test(bpf_skb_ct_lookup, ctx);
+ return 0;
+}
+
+char _license[] SEC("license") = "GPL";
--
2.34.1
^ permalink raw reply related
* Re: [PATCH v1 1/1] can: mcp251x: Get rid of duplicate of_node assignment
From: Marc Kleine-Budde @ 2021-12-10 13:06 UTC (permalink / raw)
To: Andy Shevchenko
Cc: linux-can, netdev, linux-kernel, Wolfgang Grandegger,
David S. Miller, Jakub Kicinski
In-Reply-To: <YbHvcDhtZFTyfThT@smile.fi.intel.com>
[-- Attachment #1: Type: text/plain, Size: 755 bytes --]
On 09.12.2021 13:58:40, Andy Shevchenko wrote:
> On Thu, Dec 02, 2021 at 10:58:55PM +0200, Andy Shevchenko wrote:
> > GPIO library does copy the of_node from the parent device of
> > the GPIO chip, there is no need to repeat this in the individual
> > drivers. Remove assignment here.
> >
> > For the details one may look into the of_gpio_dev_init() implementation.
>
> Marc, what do you think about this change?
LGTM, added to linux-can-next/testing.
regards,
Marc
--
Pengutronix e.K. | Marc Kleine-Budde |
Embedded Linux | https://www.pengutronix.de |
Vertretung West/Dortmund | Phone: +49-231-2826-924 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
^ permalink raw reply
* [PATCH V2] net: bonding: Add support for IPV6 ns/na
From: Sun Shouxin @ 2021-12-10 13:08 UTC (permalink / raw)
To: j.vosburgh, vfalico, andy, davem, kuba; +Cc: netdev, linux-kernel, huyd12
Since ipv6 neighbor solicitation and advertisement messages
isn't handled gracefully in bonding6 driver, we can see packet
drop due to inconsistency bewteen mac address in the option
message and source MAC .
Another examples is ipv6 neighbor solicitation and advertisement
messages from VM via tap attached to host brighe, the src mac
mighe be changed through balance-alb mode, but it is not synced
with Link-layer address in the option message.
The patch implements bond6's tx handle for ipv6 neighbor
solicitation and advertisement messages.
Border-Leaf
/ \
/ \
Tunnel1 Tunnel2
/ \
/ \
Leaf-1--Tunnel3--Leaf-2
\ /
\ /
\ /
\ /
NIC1 NIC2
\ /
server
We can see in our lab the Border-Leaf receives occasionally
a NA packet which is assigned to NIC1 mac in ND/NS option
message, but actaully send out via NIC2 mac due to tx-alb,
as a result, it will cause inconsistency between MAC table
and ND Table in Border-Leaf, i.e, NIC1 = Tunnel2 in ND table
and NIC1 = Tunnel1 in mac table.
And then, Border-Leaf starts to forward packet destinated
to the Server, it will only check the ND table entry in some
switch to encapsulate the destination MAC of the message as
NIC1 MAC, and then send it out from Tunnel2 by ND table.
Then, Leaf-2 receives the packet, it notices the destination
MAC of message is NIC1 MAC and should forword it to Tunne1
by Tunnel3.
However, this traffic forward will be failure due to split
horizon of VxLAN tunnels.
Suggested-by: Hu Yadi <huyd12@chinatelecom.cn>
Signed-off-by: Sun Shouxin <sunshouxin@chinatelecom.cn>
---
drivers/net/bonding/bond_alb.c | 131 +++++++++++++++++++++++++++++++++++++++++
1 file changed, 131 insertions(+)
diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index 533e476..afa386b 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -22,6 +22,7 @@
#include <asm/byteorder.h>
#include <net/bonding.h>
#include <net/bond_alb.h>
+#include <net/ndisc.h>
static const u8 mac_v6_allmcast[ETH_ALEN + 2] __long_aligned = {
0x33, 0x33, 0x00, 0x00, 0x00, 0x01
@@ -1269,6 +1270,119 @@ static int alb_set_mac_address(struct bonding *bond, void *addr)
return res;
}
+/*determine if the packet is NA or NS*/
+static bool alb_determine_nd(struct icmp6hdr *hdr)
+{
+ if (hdr->icmp6_type == NDISC_NEIGHBOUR_ADVERTISEMENT ||
+ hdr->icmp6_type == NDISC_NEIGHBOUR_SOLICITATION) {
+ return true;
+ }
+
+ return false;
+}
+
+static void alb_change_nd_option(struct sk_buff *skb, void *data)
+{
+ struct nd_msg *msg = (struct nd_msg *)skb_transport_header(skb);
+ struct nd_opt_hdr *nd_opt = (struct nd_opt_hdr *)msg->opt;
+ struct net_device *dev = skb->dev;
+ struct icmp6hdr *icmp6h = icmp6_hdr(skb);
+ struct ipv6hdr *ip6hdr = ipv6_hdr(skb);
+ u8 *lladdr = NULL;
+ u32 ndoptlen = skb_tail_pointer(skb) - (skb_transport_header(skb) +
+ offsetof(struct nd_msg, opt));
+
+ while (ndoptlen) {
+ int l;
+
+ switch (nd_opt->nd_opt_type) {
+ case ND_OPT_SOURCE_LL_ADDR:
+ case ND_OPT_TARGET_LL_ADDR:
+ lladdr = ndisc_opt_addr_data(nd_opt, dev);
+ break;
+
+ default:
+ lladdr = NULL;
+ break;
+ }
+
+ l = nd_opt->nd_opt_len << 3;
+
+ if (ndoptlen < l || l == 0)
+ return;
+
+ if (lladdr) {
+ memcpy(lladdr, data, dev->addr_len);
+ icmp6h->icmp6_cksum = 0;
+
+ icmp6h->icmp6_cksum = csum_ipv6_magic(&ip6hdr->saddr,
+ &ip6hdr->daddr,
+ ntohs(ip6hdr->payload_len),
+ IPPROTO_ICMPV6,
+ csum_partial(icmp6h,
+ ntohs(ip6hdr->payload_len), 0));
+ }
+ ndoptlen -= l;
+ nd_opt = ((void *)nd_opt) + l;
+ }
+}
+
+static u8 *alb_get_lladdr(struct sk_buff *skb)
+{
+ struct nd_msg *msg = (struct nd_msg *)skb_transport_header(skb);
+ struct nd_opt_hdr *nd_opt = (struct nd_opt_hdr *)msg->opt;
+ struct net_device *dev = skb->dev;
+ u8 *lladdr = NULL;
+ u32 ndoptlen = skb_tail_pointer(skb) - (skb_transport_header(skb) +
+ offsetof(struct nd_msg, opt));
+
+ while (ndoptlen) {
+ int l;
+
+ switch (nd_opt->nd_opt_type) {
+ case ND_OPT_SOURCE_LL_ADDR:
+ case ND_OPT_TARGET_LL_ADDR:
+ lladdr = ndisc_opt_addr_data(nd_opt, dev);
+ break;
+
+ default:
+ break;
+ }
+
+ l = nd_opt->nd_opt_len << 3;
+
+ if (ndoptlen < l || l == 0)
+ return lladdr;
+
+ if (lladdr)
+ return lladdr;
+
+ ndoptlen -= l;
+ nd_opt = ((void *)nd_opt) + l;
+ }
+
+ return lladdr;
+}
+
+static void alb_set_nd_option(struct sk_buff *skb, struct bonding *bond,
+ struct slave *tx_slave)
+{
+ struct ipv6hdr *ip6hdr;
+ struct icmp6hdr *hdr = NULL;
+
+ if (skb->protocol == htons(ETH_P_IPV6)) {
+ if (tx_slave && tx_slave !=
+ rcu_access_pointer(bond->curr_active_slave)) {
+ ip6hdr = ipv6_hdr(skb);
+ if (ip6hdr->nexthdr == IPPROTO_ICMPV6) {
+ hdr = icmp6_hdr(skb);
+ if (alb_determine_nd(hdr))
+ alb_change_nd_option(skb, tx_slave->dev->dev_addr);
+ }
+ }
+ }
+}
+
/************************ exported alb functions ************************/
int bond_alb_initialize(struct bonding *bond, int rlb_enabled)
@@ -1415,6 +1529,7 @@ struct slave *bond_xmit_alb_slave_get(struct bonding *bond,
}
case ETH_P_IPV6: {
const struct ipv6hdr *ip6hdr;
+ struct icmp6hdr *hdr = NULL;
/* IPv6 doesn't really use broadcast mac address, but leave
* that here just in case.
@@ -1446,6 +1561,21 @@ struct slave *bond_xmit_alb_slave_get(struct bonding *bond,
break;
}
+ if (ip6hdr->nexthdr == IPPROTO_ICMPV6) {
+ hdr = icmp6_hdr(skb);
+ if (alb_determine_nd(hdr)) {
+ u8 *lladdr = NULL;
+
+ lladdr = alb_get_lladdr(skb);
+ if (lladdr) {
+ if (!bond_slave_has_mac_rx(bond, lladdr)) {
+ do_tx_balance = false;
+ break;
+ }
+ }
+ }
+ }
+
hash_start = (char *)&ip6hdr->daddr;
hash_size = sizeof(ip6hdr->daddr);
break;
@@ -1489,6 +1619,7 @@ netdev_tx_t bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
struct slave *tx_slave = NULL;
tx_slave = bond_xmit_alb_slave_get(bond, skb);
+ alb_set_nd_option(skb, bond, tx_slave);
return bond_do_alb_xmit(skb, bond, tx_slave);
}
--
1.8.3.1
^ permalink raw reply related
* [PATCH] netfilter: fix regression in looped (broad|multi)cast's MAC handing
From: Ignacy Gawędzki @ 2021-12-10 12:26 UTC (permalink / raw)
To: netdev
In 5648b5e1169f, the test for non-empty MAC header introduced in
2c38de4c1f8da7 has been replaced with a test for a set MAC header,
which breaks the case when the MAC header has been reset (using
skb_reset_mac_header), as is the case with looped-back multicast
packets.
This patch adds a test for a non-empty MAC header in addition to the
test for a set MAC header. The same two tests are also implemented in
nfnetlink_log.c, where the initial code of 2c38de4c1f8da7 has not been
touched, but where supposedly the same situation may happen.
Signed-off-by: Ignacy Gawędzki <ignacy.gawedzki@green-communications.fr>
---
net/netfilter/nfnetlink_log.c | 3 ++-
net/netfilter/nfnetlink_queue.c | 3 ++-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c
index 691ef4cffdd9..7f83f9697fc1 100644
--- a/net/netfilter/nfnetlink_log.c
+++ b/net/netfilter/nfnetlink_log.c
@@ -556,7 +556,8 @@ __build_packet_message(struct nfnl_log_net *log,
goto nla_put_failure;
if (indev && skb->dev &&
- skb->mac_header != skb->network_header) {
+ skb_mac_header_was_set(skb) &&
+ skb_mac_header_len(skb) != 0) {
struct nfulnl_msg_packet_hw phw;
int len;
diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c
index 4acc4b8e9fe5..959527708e38 100644
--- a/net/netfilter/nfnetlink_queue.c
+++ b/net/netfilter/nfnetlink_queue.c
@@ -560,7 +560,8 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
goto nla_put_failure;
if (indev && entskb->dev &&
- skb_mac_header_was_set(entskb)) {
+ skb_mac_header_was_set(entskb) &&
+ skb_mac_header_len(entskb) != 0) {
struct nfqnl_msg_packet_hw phw;
int len;
--
2.32.0
^ permalink raw reply related
* [syzbot] KMSAN: uninit-value in _ieee802_11_parse_elems_crc
From: syzbot @ 2021-12-10 13:09 UTC (permalink / raw)
To: davem, glider, johannes, kuba, linux-kernel, linux-wireless,
netdev, syzkaller-bugs
Hello,
syzbot found the following issue on:
HEAD commit: 093998ececa3 [PATCH net] tcp: fix another uninit-value (sk..
git tree: https://github.com/google/kmsan.git master
console output: https://syzkaller.appspot.com/x/log.txt?x=110b5395b00000
kernel config: https://syzkaller.appspot.com/x/.config?x=e00a8959fdd3f3e8
dashboard link: https://syzkaller.appspot.com/bug?extid=59bdff68edce82e393b6
compiler: clang version 14.0.0 (git@github.com:llvm/llvm-project.git 0996585c8e3b3d409494eb5f1cad714b9e1f7fb5), GNU ld (GNU Binutils for Debian) 2.35.2
Unfortunately, I don't have any reproducer for this issue yet.
IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+59bdff68edce82e393b6@syzkaller.appspotmail.com
=====================================================
BUG: KMSAN: uninit-value in ieee80211_parse_extension_element net/mac80211/util.c:948 [inline]
BUG: KMSAN: uninit-value in _ieee802_11_parse_elems_crc+0x3448/0x4310 net/mac80211/util.c:1348
ieee80211_parse_extension_element net/mac80211/util.c:948 [inline]
_ieee802_11_parse_elems_crc+0x3448/0x4310 net/mac80211/util.c:1348
ieee802_11_parse_elems_crc+0x17e3/0x1e30 net/mac80211/util.c:1490
ieee802_11_parse_elems net/mac80211/ieee80211_i.h:2208 [inline]
ieee80211_rx_mgmt_probe_beacon net/mac80211/ibss.c:1605 [inline]
ieee80211_ibss_rx_queued_mgmt+0x7e5/0x4350 net/mac80211/ibss.c:1639
ieee80211_iface_process_skb net/mac80211/iface.c:1468 [inline]
ieee80211_iface_work+0xeda/0x1990 net/mac80211/iface.c:1522
process_one_work+0xdc2/0x1820 kernel/workqueue.c:2298
worker_thread+0x10f1/0x2290 kernel/workqueue.c:2445
kthread+0x721/0x850 kernel/kthread.c:327
ret_from_fork+0x1f/0x30
Uninit was created at:
slab_post_alloc_hook mm/slab.h:524 [inline]
slab_alloc_node mm/slub.c:3251 [inline]
__kmalloc_node_track_caller+0xe0c/0x1510 mm/slub.c:4974
kmalloc_reserve net/core/skbuff.c:354 [inline]
__alloc_skb+0x545/0xf90 net/core/skbuff.c:426
skb_copy+0x191/0xb90 net/core/skbuff.c:1586
mac80211_hwsim_tx_frame_no_nl+0x1fcf/0x2c00 drivers/net/wireless/mac80211_hwsim.c:1565
mac80211_hwsim_tx+0x10d3/0x1760 drivers/net/wireless/mac80211_hwsim.c:1669
drv_tx net/mac80211/driver-ops.h:35 [inline]
ieee80211_tx_frags+0x7bf/0x1250 net/mac80211/tx.c:1714
__ieee80211_tx+0x5a8/0x7d0 net/mac80211/tx.c:1768
ieee80211_tx+0x776/0x790 net/mac80211/tx.c:1948
ieee80211_xmit+0x849/0x890 net/mac80211/tx.c:2040
__ieee80211_tx_skb_tid_band+0x297/0x3a0 net/mac80211/tx.c:5701
ieee80211_tx_skb_tid net/mac80211/ieee80211_i.h:2186 [inline]
ieee80211_tx_skb net/mac80211/ieee80211_i.h:2195 [inline]
ieee80211_mgmt_tx+0x1721/0x1d00 net/mac80211/offchannel.c:927
rdev_mgmt_tx+0x117/0x4e0 net/wireless/rdev-ops.h:742
cfg80211_mlme_mgmt_tx+0x910/0x1330 net/wireless/mlme.c:759
nl80211_tx_mgmt+0x112a/0x1870 net/wireless/nl80211.c:11708
genl_family_rcv_msg_doit net/netlink/genetlink.c:731 [inline]
genl_family_rcv_msg net/netlink/genetlink.c:775 [inline]
genl_rcv_msg+0x157f/0x1660 net/netlink/genetlink.c:792
netlink_rcv_skb+0x447/0x800 net/netlink/af_netlink.c:2491
genl_rcv+0x63/0x80 net/netlink/genetlink.c:803
netlink_unicast_kernel net/netlink/af_netlink.c:1319 [inline]
netlink_unicast+0x1095/0x1360 net/netlink/af_netlink.c:1345
netlink_sendmsg+0x16f3/0x1870 net/netlink/af_netlink.c:1916
sock_sendmsg_nosec net/socket.c:704 [inline]
sock_sendmsg net/socket.c:724 [inline]
____sys_sendmsg+0xe11/0x12c0 net/socket.c:2409
___sys_sendmsg net/socket.c:2463 [inline]
__sys_sendmsg+0x4a5/0x640 net/socket.c:2492
__do_sys_sendmsg net/socket.c:2501 [inline]
__se_sys_sendmsg net/socket.c:2499 [inline]
__x64_sys_sendmsg+0xe2/0x120 net/socket.c:2499
do_syscall_x64 arch/x86/entry/common.c:51 [inline]
do_syscall_64+0x54/0xd0 arch/x86/entry/common.c:82
entry_SYSCALL_64_after_hwframe+0x44/0xae
=====================================================
---
This report is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzkaller@googlegroups.com.
syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
^ permalink raw reply
* [PATCH net 0/2] net: hns3: add some fixes for -net
From: Guangbin Huang @ 2021-12-10 13:09 UTC (permalink / raw)
To: davem, kuba, wangjie125
Cc: netdev, linux-kernel, lipeng321, huangguangbin2, chenhao288
This series adds some fixes for the HNS3 ethernet driver.
Jie Wang (1):
net: hns3: fix use-after-free bug in hclgevf_send_mbx_msg
Yufeng Mo (1):
net: hns3: fix race condition in debugfs
drivers/net/ethernet/hisilicon/hns3/hnae3.h | 2 ++
.../ethernet/hisilicon/hns3/hns3_debugfs.c | 20 +++++++++++++------
.../hisilicon/hns3/hns3vf/hclgevf_mbx.c | 3 ++-
3 files changed, 18 insertions(+), 7 deletions(-)
--
2.33.0
^ permalink raw reply
* [PATCH net 1/2] net: hns3: fix use-after-free bug in hclgevf_send_mbx_msg
From: Guangbin Huang @ 2021-12-10 13:09 UTC (permalink / raw)
To: davem, kuba, wangjie125
Cc: netdev, linux-kernel, lipeng321, huangguangbin2, chenhao288
In-Reply-To: <20211210130934.36278-1-huangguangbin2@huawei.com>
From: Jie Wang <wangjie125@huawei.com>
Currently, the hns3_remove function firstly uninstall client instance,
and then uninstall acceletion engine device. The netdevice is freed in
client instance uninstall process, but acceletion engine device uninstall
process still use it to trace runtime information. This causes a use after
free problem.
So fixes it by check the instance register state to avoid use after free.
Fixes: d8355240cf8f ("net: hns3: add trace event support for PF/VF mailbox")
Signed-off-by: Jie Wang <wangjie125@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
---
drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c
index fdc66fae0960..c5ac6ecf36e1 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c
@@ -114,7 +114,8 @@ int hclgevf_send_mbx_msg(struct hclgevf_dev *hdev,
memcpy(&req->msg, send_msg, sizeof(struct hclge_vf_to_pf_msg));
- trace_hclge_vf_mbx_send(hdev, req);
+ if (test_bit(HCLGEVF_STATE_NIC_REGISTERED, &hdev->state))
+ trace_hclge_vf_mbx_send(hdev, req);
/* synchronous send */
if (need_resp) {
--
2.33.0
^ permalink raw reply related
* [PATCH net 2/2] net: hns3: fix race condition in debugfs
From: Guangbin Huang @ 2021-12-10 13:09 UTC (permalink / raw)
To: davem, kuba, wangjie125
Cc: netdev, linux-kernel, lipeng321, huangguangbin2, chenhao288
In-Reply-To: <20211210130934.36278-1-huangguangbin2@huawei.com>
From: Yufeng Mo <moyufeng@huawei.com>
When multiple threads concurrently access the debugfs content, data
and pointer exceptions may occur. Therefore, mutex lock protection is
added for debugfs.
Fixes: 5e69ea7ee2a6 ("net: hns3: refactor the debugfs process")
Signed-off-by: Yufeng Mo <moyufeng@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
---
drivers/net/ethernet/hisilicon/hns3/hnae3.h | 2 ++
.../ethernet/hisilicon/hns3/hns3_debugfs.c | 20 +++++++++++++------
2 files changed, 16 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
index 3f7a9a4c59d5..63f5abcc6bf4 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
@@ -839,6 +839,8 @@ struct hnae3_handle {
u8 netdev_flags;
struct dentry *hnae3_dbgfs;
+ /* protects concurrent contention between debugfs commands */
+ struct mutex dbgfs_lock;
/* Network interface message level enabled bits */
u32 msg_enable;
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
index 081295bff765..c381f8af67f0 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
@@ -1226,6 +1226,7 @@ static ssize_t hns3_dbg_read(struct file *filp, char __user *buffer,
if (ret)
return ret;
+ mutex_lock(&handle->dbgfs_lock);
save_buf = &hns3_dbg_cmd[index].buf;
if (!test_bit(HNS3_NIC_STATE_INITED, &priv->state) ||
@@ -1238,15 +1239,15 @@ static ssize_t hns3_dbg_read(struct file *filp, char __user *buffer,
read_buf = *save_buf;
} else {
read_buf = kvzalloc(hns3_dbg_cmd[index].buf_len, GFP_KERNEL);
- if (!read_buf)
- return -ENOMEM;
+ if (!read_buf) {
+ ret = -ENOMEM;
+ goto out;
+ }
/* save the buffer addr until the last read operation */
*save_buf = read_buf;
- }
- /* get data ready for the first time to read */
- if (!*ppos) {
+ /* get data ready for the first time to read */
ret = hns3_dbg_read_cmd(dbg_data, hns3_dbg_cmd[index].cmd,
read_buf, hns3_dbg_cmd[index].buf_len);
if (ret)
@@ -1255,8 +1256,10 @@ static ssize_t hns3_dbg_read(struct file *filp, char __user *buffer,
size = simple_read_from_buffer(buffer, count, ppos, read_buf,
strlen(read_buf));
- if (size > 0)
+ if (size > 0) {
+ mutex_unlock(&handle->dbgfs_lock);
return size;
+ }
out:
/* free the buffer for the last read operation */
@@ -1265,6 +1268,7 @@ static ssize_t hns3_dbg_read(struct file *filp, char __user *buffer,
*save_buf = NULL;
}
+ mutex_unlock(&handle->dbgfs_lock);
return ret;
}
@@ -1337,6 +1341,8 @@ int hns3_dbg_init(struct hnae3_handle *handle)
debugfs_create_dir(hns3_dbg_dentry[i].name,
handle->hnae3_dbgfs);
+ mutex_init(&handle->dbgfs_lock);
+
for (i = 0; i < ARRAY_SIZE(hns3_dbg_cmd); i++) {
if ((hns3_dbg_cmd[i].cmd == HNAE3_DBG_CMD_TM_NODES &&
ae_dev->dev_version <= HNAE3_DEVICE_VERSION_V2) ||
@@ -1363,6 +1369,7 @@ int hns3_dbg_init(struct hnae3_handle *handle)
return 0;
out:
+ mutex_destroy(&handle->dbgfs_lock);
debugfs_remove_recursive(handle->hnae3_dbgfs);
handle->hnae3_dbgfs = NULL;
return ret;
@@ -1378,6 +1385,7 @@ void hns3_dbg_uninit(struct hnae3_handle *handle)
hns3_dbg_cmd[i].buf = NULL;
}
+ mutex_destroy(&handle->dbgfs_lock);
debugfs_remove_recursive(handle->hnae3_dbgfs);
handle->hnae3_dbgfs = NULL;
}
--
2.33.0
^ permalink raw reply related
* Re: [PATCH v1 1/1] can: mcp251x: Get rid of duplicate of_node assignment
From: Andy Shevchenko @ 2021-12-10 13:19 UTC (permalink / raw)
To: Marc Kleine-Budde
Cc: linux-can, netdev, linux-kernel, Wolfgang Grandegger,
David S. Miller, Jakub Kicinski
In-Reply-To: <20211210130607.rajkkzr7lf6l4tok@pengutronix.de>
On Fri, Dec 10, 2021 at 02:06:07PM +0100, Marc Kleine-Budde wrote:
> On 09.12.2021 13:58:40, Andy Shevchenko wrote:
> > On Thu, Dec 02, 2021 at 10:58:55PM +0200, Andy Shevchenko wrote:
...
> > Marc, what do you think about this change?
>
> LGTM, added to linux-can-next/testing.
Thanks for applying this and hi311x patches!
--
With Best Regards,
Andy Shevchenko
^ permalink raw reply
* Re: [PATCH V6 3/5] hyper-v: Enable swiotlb bounce buffer for Isolation VM
From: Tianyu Lan @ 2021-12-10 13:25 UTC (permalink / raw)
To: Michael Kelley (LINUX), KY Srinivasan, Haiyang Zhang,
Stephen Hemminger, wei.liu@kernel.org, Dexuan Cui,
tglx@linutronix.de, mingo@redhat.com, bp@alien8.de,
dave.hansen@linux.intel.com, x86@kernel.org, hpa@zytor.com,
davem@davemloft.net, kuba@kernel.org, jejb@linux.ibm.com,
martin.petersen@oracle.com, arnd@arndb.de, hch@infradead.org,
m.szyprowski@samsung.com, robin.murphy@arm.com, Tianyu Lan,
thomas.lendacky@amd.com
Cc: iommu@lists.linux-foundation.org, linux-arch@vger.kernel.org,
linux-hyperv@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-scsi@vger.kernel.org, netdev@vger.kernel.org, vkuznets,
brijesh.singh@amd.com, konrad.wilk@oracle.com, hch@lst.de,
joro@8bytes.org, parri.andrea@gmail.com, dave.hansen@intel.com
In-Reply-To: <MWHPR21MB159359667085776793988EACD7709@MWHPR21MB1593.namprd21.prod.outlook.com>
On 12/10/2021 4:09 AM, Michael Kelley (LINUX) wrote:
>> @@ -319,8 +320,16 @@ static void __init ms_hyperv_init_platform(void)
>> pr_info("Hyper-V: Isolation Config: Group A 0x%x, Group B 0x%x\n",
>> ms_hyperv.isolation_config_a, ms_hyperv.isolation_config_b);
>>
>> - if (hv_get_isolation_type() == HV_ISOLATION_TYPE_SNP)
>> + if (hv_get_isolation_type() == HV_ISOLATION_TYPE_SNP) {
>> static_branch_enable(&isolation_type_snp);
>> + swiotlb_unencrypted_base = ms_hyperv.shared_gpa_boundary;
>> + }
>> +
>> + /*
>> + * Enable swiotlb force mode in Isolation VM to
>> + * use swiotlb bounce buffer for dma transaction.
>> + */
>> + swiotlb_force = SWIOTLB_FORCE;
> I'm good with this approach that directly updates the swiotlb settings here
>
> rather than in IOMMU initialization code. It's a lot more straightforward.
>
> However, there's an issue if building for X86_32 without PAE, in that the
> swiotlb module may not be built, resulting in compile and link errors. The
> swiotlb.h file needs to be updated to provide a stub function for
> swiotlb_update_mem_attributes(). swiotlb_unencrypted_base probably
> needs wrapper functions to get/set it, which can be stubs when
> CONFIG_SWIOTLB is not set. swiotlb_force is a bit of a mess in that it already
> has a stub definition that assumes it will only be read, and not set. A bit of
> thinking will be needed to sort that out.
It's ok to fix the issue via selecting swiotlb when CONFIG_HYPERV is
set?
>
>> }
>>
>> if (hv_max_functions_eax >= HYPERV_CPUID_NESTED_FEATURES) {
>> diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h
>> index b823311eac79..1f037e114dc8 100644
>> --- a/include/linux/hyperv.h
>> +++ b/include/linux/hyperv.h
>> @@ -1726,6 +1726,14 @@ int hyperv_write_cfg_blk(struct pci_dev *dev, void *buf, unsigned int len,
>> int hyperv_reg_block_invalidate(struct pci_dev *dev, void *context,
>> void (*block_invalidate)(void *context,
>> u64 block_mask));
>> +#if IS_ENABLED(CONFIG_HYPERV)
>> +int __init hyperv_swiotlb_detect(void);
>> +#else
>> +static inline int __init hyperv_swiotlb_detect(void)
>> +{
>> + return 0;
>> +}
>> +#endif
> I don't think hyperv_swiotlb_detect() is used any longer, so this change
> should be dropped.
Yes, will update.
^ permalink raw reply
* Re: [PATCH v4 net-next 1/9] i40e: don't reserve excessive XDP_PACKET_HEADROOM on XSK Rx to skb
From: Jesper Dangaard Brouer @ 2021-12-10 13:31 UTC (permalink / raw)
To: Alexander Lobakin, Jesper Dangaard Brouer
Cc: brouer, intel-wired-lan, Jesse Brandeburg, Tony Nguyen,
David S. Miller, Jakub Kicinski, Alexei Starovoitov,
Daniel Borkmann, Jesper Dangaard Brouer, John Fastabend,
Björn Töpel, Maciej Fijalkowski, Michal Swiatkowski,
Jithu Joseph, Martin KaFai Lau, Song Liu, KP Singh, Yonghong Song,
Andrii Nakryiko, netdev, linux-kernel, bpf
In-Reply-To: <20211209173307.5003-1-alexandr.lobakin@intel.com>
On 09/12/2021 18.33, Alexander Lobakin wrote:
> From: Jesper Dangaard Brouer <jbrouer@redhat.com>
> Date: Thu, 9 Dec 2021 09:19:46 +0100
>
>> On 08/12/2021 15.06, Alexander Lobakin wrote:
>>> {__,}napi_alloc_skb() allocates and reserves additional NET_SKB_PAD
>>> + NET_IP_ALIGN for any skb.
>>> OTOH, i40e_construct_skb_zc() currently allocates and reserves
>>> additional `xdp->data - xdp->data_hard_start`, which is
>>> XDP_PACKET_HEADROOM for XSK frames.
>>> There's no need for that at all as the frame is post-XDP and will
>>> go only to the networking stack core.
>>
>> I disagree with this assumption, that headroom is not needed by netstack.
>> Why "no need for that at all" for netstack?
>
> napi_alloc_skb() in our particular case will reserve 64 bytes, it is
> sufficient for {TCP,UDP,SCTP,...}/IPv{4,6} etc.
My bad, I misunderstood you. I now see (looking at code) that (as you
say) 64 bytes of headroom *is* reserved (in bottom of __napi_alloc_skb).
Thus, the SKB *do* have headroom, so this patch should be fine.
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Do watch out that 64 bytes is not always enough. Notice the define
LL_MAX_HEADER and MAX_HEADER in include/linux/netdevice.h (that tries to
determine worst-case header length) which is above 64 bytes. It is also
affected by HyperV and WiFi configs.
^ permalink raw reply
* Re: [PATCH] igc: Avoid possible deadlock during suspend/resume
From: Stefan Dietrich @ 2021-12-10 13:45 UTC (permalink / raw)
To: Thorsten Leemhuis, Vinicius Costa Gomes
Cc: kuba, greg, netdev, intel-wired-lan, regressions
In-Reply-To: <063995d8-acf3-9f33-5667-f284233c94b4@leemhuis.info>
Hi Thorsten,
thanks for keeping an eye on the issue. I've sent the files in private
because I did not want to spam the mailing lists with them. Please let
me know if this is the correct procedure.
Cheers,
Stefan
On Fri, 2021-12-10 at 10:40 +0100, Thorsten Leemhuis wrote:
> Hi, this is your Linux kernel regression tracker speaking.
>
> On 02.12.21 23:34, Vinicius Costa Gomes wrote:
> > Hi Stefan,
> >
> > Stefan Dietrich <roots@gmx.de> writes:
> >
> > > Hi Vinicius,
> > >
> > > thanks for the patch - unfortunately it did not solve the issue
> > > and I
> > > am still getting reboots/lockups.
> > >
> >
> > Thanks for the test. We learned something, not a lot, but
> > something: the
> > problem you are facing is PTM related and it's not the same bug as
> > that
> > PM deadlock.
> >
> > I am still trying to understand what's going on.
> >
> > Are you able to send me the 'dmesg' output for the two kernel
> > configs
> > (CONFIG_PCIE_PTM enabled and disabled)? (no need to bring the
> > network
> > interface up or down). Your kernel .config would be useful as well.
>
> Stefan, could you provide the data Vinicius asked for? Or did you do
> that in private already? Or was progress made somewhere else and I
> simply missed this?
>
> Ciao, Thorsten, your Linux kernel regression tracker.
>
> P.S.: As a Linux kernel regression tracker I'm getting a lot of
> reports
> on my table. I can only look briefly into most of them. Unfortunately
> therefore I sometimes will get things wrong or miss something
> important.
> I hope that's not the case here; if you think it is, don't hesitate
> to
> tell me about it in a public reply. That's in everyone's interest, as
> what I wrote above might be misleading to everyone reading this; any
> suggestion I gave they thus might sent someone reading this down the
> wrong rabbit hole, which none of us wants.
>
> BTW, I have no personal interest in this issue, which is tracked
> using
> regzbot, my Linux kernel regression tracking bot
> (https://linux-regtracking.leemhuis.info/regzbot/). I'm only posting
> this mail to get things rolling again and hence don't need to be CC
> on
> all further activities wrt to this regression.
>
> #regzbot poke
>
> > > On Wed, 2021-12-01 at 10:57 -0800, Vinicius Costa Gomes wrote:
> > > > Inspired by:
> > > > https://bugzilla.kernel.org/show_bug.cgi?id=215129
> > > >
> > > > Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
> > > > ---
> > > > Just to see if it's indeed the same problem as the bug report
> > > > above.
> > > >
> > > > drivers/net/ethernet/intel/igc/igc_main.c | 19 +++++++++++++
> > > > ------
> > > > 1 file changed, 13 insertions(+), 6 deletions(-)
> > > >
> > > > diff --git a/drivers/net/ethernet/intel/igc/igc_main.c
> > > > b/drivers/net/ethernet/intel/igc/igc_main.c
> > > > index 0e19b4d02e62..c58bf557a2a1 100644
> > > > --- a/drivers/net/ethernet/intel/igc/igc_main.c
> > > > +++ b/drivers/net/ethernet/intel/igc/igc_main.c
> > > > @@ -6619,7 +6619,7 @@ static void
> > > > igc_deliver_wake_packet(struct
> > > > net_device *netdev)
> > > > netif_rx(skb);
> > > > }
> > > >
> > > > -static int __maybe_unused igc_resume(struct device *dev)
> > > > +static int __maybe_unused __igc_resume(struct device *dev,
> > > > bool rpm)
> > > > {
> > > > struct pci_dev *pdev = to_pci_dev(dev);
> > > > struct net_device *netdev = pci_get_drvdata(pdev);
> > > > @@ -6661,20 +6661,27 @@ static int __maybe_unused
> > > > igc_resume(struct
> > > > device *dev)
> > > >
> > > > wr32(IGC_WUS, ~0);
> > > >
> > > > - rtnl_lock();
> > > > + if (!rpm)
> > > > + rtnl_lock();
> > > > if (!err && netif_running(netdev))
> > > > err = __igc_open(netdev, true);
> > > >
> > > > if (!err)
> > > > netif_device_attach(netdev);
> > > > - rtnl_unlock();
> > > > + if (!rpm)
> > > > + rtnl_unlock();
> > > >
> > > > return err;
> > > > }
> > > >
> > > > static int __maybe_unused igc_runtime_resume(struct device
> > > > *dev)
> > > > {
> > > > - return igc_resume(dev);
> > > > + return __igc_resume(dev, true);
> > > > +}
> > > > +
> > > > +static int __maybe_unused igc_resume(struct device *dev)
> > > > +{
> > > > + return __igc_resume(dev, false);
> > > > }
> > > >
> > > > static int __maybe_unused igc_suspend(struct device *dev)
> > > > @@ -6738,7 +6745,7 @@ static pci_ers_result_t
> > > > igc_io_error_detected(struct pci_dev *pdev,
> > > > * @pdev: Pointer to PCI device
> > > > *
> > > > * Restart the card from scratch, as if from a cold-boot.
> > > > Implementation
> > > > - * resembles the first-half of the igc_resume routine.
> > > > + * resembles the first-half of the __igc_resume routine.
> > > > **/
> > > > static pci_ers_result_t igc_io_slot_reset(struct pci_dev
> > > > *pdev)
> > > > {
> > > > @@ -6777,7 +6784,7 @@ static pci_ers_result_t
> > > > igc_io_slot_reset(struct pci_dev *pdev)
> > > > *
> > > > * This callback is called when the error recovery driver
> > > > tells us
> > > > that
> > > > * its OK to resume normal operation. Implementation
> > > > resembles the
> > > > - * second-half of the igc_resume routine.
> > > > + * second-half of the __igc_resume routine.
> > > > */
> > > > static void igc_io_resume(struct pci_dev *pdev)
> > > > {
> >
> > Cheers,
> >
^ permalink raw reply
* Re: [RFC PATCH 0/2] Reset PHY in fec_resume if it got powered down
From: Philippe Schenker @ 2021-12-10 13:51 UTC (permalink / raw)
To: festevam@gmail.com, linux@armlinux.org.uk, kuba@kernel.org,
netdev@vger.kernel.org, davem@davemloft.net,
qiangqing.zhang@nxp.com, andrew@lunn.ch
Cc: linux-kernel@vger.kernel.org
In-Reply-To: <DB8PR04MB679536F1DD900564B9A957C4E66E9@DB8PR04MB6795.eurprd04.prod.outlook.com>
On Tue, 2021-12-07 at 01:58 +0000, Joakim Zhang wrote:
>
> Hi Philippe,
>
> > -----Original Message-----
> > From: Philippe Schenker <philippe.schenker@toradex.com>
> > Sent: 2021年12月6日 18:13
> > To: netdev@vger.kernel.org; Joakim Zhang <qiangqing.zhang@nxp.com>;
> > Fabio Estevam <festevam@gmail.com>; Fugang Duan
> > <fugang.duan@nxp.com>; David S . Miller <davem@davemloft.net>;
> > Russell
> > King <linux@armlinux.org.uk>; Andrew Lunn <andrew@lunn.ch>; Jakub
> > Kicinski <kuba@kernel.org>
> > Cc: Philippe Schenker <philippe.schenker@toradex.com>;
> > linux-kernel@vger.kernel.org
> > Subject: [RFC PATCH 0/2] Reset PHY in fec_resume if it got powered
> > down
> >
> >
> > If a hardware-design is able to control power to the Ethernet PHY
> > and relying
> > on software to do a reset, the PHY does no longer work after
> > resuming from
> > suspend, given the PHY does need a hardware-reset.
> > The Freescale fec driver does currently control the reset-signal of
> > a phy but
> > does not issue a reset on resume.
> >
> > On Toradex Apalis iMX8 board we do have such a design where we also
> > don't
> > place the RC circuit to delay the reset-line by hardware. Hence we
> > fully rely
> > on software to do so.
> > Since I didn't manage to get the needed parts of Apalis iMX8 working
> > with
> > mainline this patchset was only tested on the downstream kernel
> > toradex_5.4-2.3.x-imx. [1] This kernel is based on NXP's release
> > imx_5.4.70_2.3.0. [2] The affected code is still the same on
> > mainline kernel,
> > which would actually make me comfortable merging this patch, but due
> > to
> > this fact I'm sending this as RFC maybe someone else is able to test
> > this code.
> >
> > This patchset aims to change the behavior by resetting the ethernet
> > PHY in
> > fec_resume. A short description of the patches can be found below,
> > please
> > find a detailed description in the commit-messages of the respective
> > patches.
> >
> > [PATCH 2/2] net: fec: reset phy in resume if it was powered down
> >
> > This patch calls fec_reset_phy just after regulator enable in
> > fec_resume,
> > when the phy is resumed
> >
> > [PATCH 1/2] net: fec: make fec_reset_phy not only usable once
> >
> > This patch prepares the function fec_reset_phy to be called multiple
> > times. It
> > stores the data around the reset-gpio in fec_enet_private.
> > This patch aims to do no functional changes.
> >
> > [1]
> > https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fgit.tor
> > adex.com%2Fcgit%2Flinux-toradex.git%2Flog%2F%3Fh%3Dtoradex_5.4-2.3.x
> > -imx&data=04%7C01%7Cqiangqing.zhang%40nxp.com%7Cf3c138ed9232
> > 4a8d75e708d9b8a11b9a%7C686ea1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%
> > 7C637743824364193423%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAw
> > MDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sda
> > ta=Bw%2BZdqhAjPXqKJFZCXp0mtId1x9mkX6f6MW2ky6U1ww%3D&res
> > erved=0
> > [2]
> > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsourc
> > e.codeaurora.org%2Fexternal%2Fimx%2Flinux-imx%2Flog%2F%3Fh%3Dimx_
> > 5.4.70_2.3.0&data=04%7C01%7Cqiangqing.zhang%40nxp.com%7Cf3c13
> > 8ed92324a8d75e708d9b8a11b9a%7C686ea1d3bc2b4c6fa92cd99c5c301635%7C
> > 0%7C0%7C637743824364193423%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiM
> > C4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000
> > &sdata=of9z9hfVhHakVScLxCdEo%2BXmd2B9Ad9X8Rry6GjEZE4%3D&a
> > mp;reserved=0
> >
>
> In fec driver, it has supported hardware reset for PHY when MAC resume
> back,
>
> fec_resume() -> phy_init_hw() -> phy_device_reset() de-assert the
> reset signal, you only need implement
> the properties which PHY core provided.
>
> I think you should not use deprecated reset properties provided by fec
> driver, instead the common
> reset properties provided by PHY core.
>
> Please check the dt-bindings for more details:
> Documentation/devicetree/bindings/net/fsl,fec.yaml
> Documentation/devicetree/bindings/net/ethernet-phy.yaml
>
> Best Regards,
> Joakim Zhang
Hi Joakim and many thanks for that hint! I tried that out but
unfortunately it still does not work due to phy_init_hw() only
deasserting the reset. For that to work, for example a call of
phy_device_reset(ndev->phydev, 1); in fec_suspend or in fec_resume
before enabling the supply would work, in order to assert that reset.
I see now two ways to go to fix our issue:
1. Assert the phy-reset gpio in fec_suspend() or fec_resume()
2. Add support for regulators in drivers/net/phy/phy-core.c and handle
the phy-reset properly in there with assert-us and deassert-us delays.
As you probably have a much better overview: Do you see another
possibility to handle phy-reset after resuming? Or which way shall I
choose to go forward?
Thanks in advance for any advice
Philippe
> > Philippe Schenker (2):
> > net: fec: make fec_reset_phy not only usable once
> > net: fec: reset phy in resume if it was powered down
> >
> > drivers/net/ethernet/freescale/fec.h | 6 ++
> > drivers/net/ethernet/freescale/fec_main.c | 98 ++++++++++++++++----
> > ---
> > 2 files changed, 73 insertions(+), 31 deletions(-)
> >
> > --
> > 2.34.0
>
^ permalink raw reply
* Re: [PATCH v6 00/14] genirq: Cleanup the abuse of irq_set_affinity_hint()
From: Nitesh Lal @ 2021-12-10 13:51 UTC (permalink / raw)
To: Martin K. Petersen
Cc: Juri Lelli, Dick Kennedy, Thomas Gleixner, Marcelo Tosatti,
linux-kernel, linux-pci, linux-scsi, netdev, davem, ajit.khaparde,
sriharsha.basavapatna, somnath.kotur, huangguangbin2, huangdaode,
Frederic Weisbecker, Alex Belits, Bjorn Helgaas, rostedt,
Peter Zijlstra, Jesse Brandeburg, Robin Murphy, Ingo Molnar,
jbrandeb, akpm, sfr, stephen, rppt, chris.friesen, Marc Zyngier,
Neil Horman, pjwaskiewicz, Stefan Assmann, Tomas Henzl,
james.smart, Ken Cox, faisal.latif, shiraz.saleem, tariqt,
Alaa Hleihel, Kamal Heib, borisp, saeedm, Nikolova, Tatyana E,
Ismail, Mustafa, Al Stone, Leon Romanovsky, Chandrakanth Patil,
bjorn.andersson, chunkuang.hu, yongqiang.niu, baolin.wang7,
Petr Oros, Ming Lei, Ewan Milne, jejb, kabel, Viresh Kumar,
Jakub Kicinski, kashyap.desai, Sumit Saxena,
shivasharan.srikanteshwara, sathya.prakash, Sreekanth Reddy,
suganath-prabu.subramani, ley.foon.tan, jbrunet, johannes,
snelson, lewis.hanly, benve, _govind, jassisinghbrar
In-Reply-To: <CAFki+Lmrv-UjZpuTQWr9c-Rymfm-tuCw9WpwmHgyfjVhJgp--g@mail.gmail.com>
Hi Martin,
On Wed, Nov 24, 2021 at 5:16 PM Nitesh Lal <nilal@redhat.com> wrote:
>
> On Wed, Nov 24, 2021 at 2:30 PM Thomas Gleixner <tglx@linutronix.de> wrote:
> >
> > Nitesh,
> >
> > On Mon, Sep 13 2021 at 10:34, Nitesh Lal wrote:
> > > On Fri, Sep 3, 2021 at 11:25 AM Nitesh Narayan Lal <nitesh@redhat.com> wrote:
> > >>
> > >> The drivers currently rely on irq_set_affinity_hint() to either set the
> > >> affinity_hint that is consumed by the userspace and/or to enforce a custom
> > >> affinity.
> > >>
> > >> irq_set_affinity_hint() as the name suggests is originally introduced to
> > >> only set the affinity_hint to help the userspace in guiding the interrupts
> > >> and not the affinity itself. However, since the commit
> > >>
> > >> e2e64a932556 "genirq: Set initial affinity in irq_set_affinity_hint()"
> >
> > sorry for ignoring this. It fell through the cracks.
>
>
> No worries, thank you for reviewing.
>
> >
> >
> > >> Thomas Gleixner (1):
> > >> genirq: Provide new interfaces for affinity hints
> >
> > Did I actually write this?
>
>
> Yeap, the idea and the initial patch came from you. :)
>
> >
> >
> > > Any suggestions on what should be the next steps here? Unfortunately, I haven't
> > > been able to get any reviews on the following two patches:
> > > be2net: Use irq_update_affinity_hint
> > > hinic: Use irq_set_affinity_and_hint
> > >
> > > One option would be to proceed with the remaining patches and I can try
> > > posting these two again when I post patches for the remaining drivers?
> >
> > The more general question is whether I should queue all the others or
> > whether some subsystem would prefer to pull in a tagged commit on top of
> > rc1. I'm happy to carry them all of course.
> >
>
> I am fine either way.
> In the past, while I was asking for more testing help I was asked if the
> SCSI changes are part of Martins's scsi-fixes tree as that's something
> Broadcom folks test to check for regression.
> So, maybe Martin can pull this up?
>
Gentle ping.
Any thoughts on the above query?
--
Thanks
Nitesh
^ permalink raw reply
* [PATCH v5 net-next 0/4] Add ethtool interface for RClocks
From: Maciej Machnikowski @ 2021-12-10 13:45 UTC (permalink / raw)
To: maciej.machnikowski, netdev, intel-wired-lan,
arkadiusz.kubalewski
Cc: richardcochran, abyagowi, anthony.l.nguyen, davem, kuba,
linux-kselftest, idosch, mkubecek, saeed, michael.chan, petrm
Synchronous Ethernet networks use a physical layer clock to syntonize
the frequency across different network elements.
Basic SyncE node defined in the ITU-T G.8264 consist of an Ethernet
Equipment Clock (EEC) and have the ability to synchronize to reference
frequency sources.
This patch series is a prerequisite for EEC object and adds ability
to enable recovered clocks in the physical layer of the netdev object.
Recovered clocks can be used as one of the reference signal by the EEC.
Further work is required to add the DPLL subsystem, link it to the
netdev object and create API to read the EEC DPLL state.
v5:
- rewritten the documentation
- fixed doxygen headers
v4:
- Dropped EEC_STATE reporting (TBD: DPLL subsystem)
- moved recovered clock configuration to ethtool netlink
v3:
- remove RTM_GETRCLKRANGE
- return state of all possible pins in the RTM_GETRCLKSTATE
- clarify documentation
v2:
- improved documentation
- fixed kdoc warning
RFC history:
v2:
- removed whitespace changes
- fix issues reported by test robot
v3:
- Changed naming from SyncE to EEC
- Clarify cover letter and commit message for patch 1
v4:
- Removed sync_source and pin_idx info
- Changed one structure to attributes
- Added EEC_SRC_PORT flag to indicate that the EEC is synchronized
to the recovered clock of a port that returns the state
v5:
- add EEC source as an optiona attribute
- implement support for recovered clocks
- align states returned by EEC to ITU-T G.781
v6:
- fix EEC clock state reporting
- add documentation
- fix descriptions in code comments
Maciej Machnikowski (4):
ice: add support detecting features based on netlist
ethtool: Add ability to configure recovered clock for SyncE feature
ice: add support for monitoring SyncE DPLL state
ice: add support for recovered clocks
Documentation/networking/ethtool-netlink.rst | 62 ++++
drivers/net/ethernet/intel/ice/ice.h | 7 +
.../net/ethernet/intel/ice/ice_adminq_cmd.h | 70 ++++-
drivers/net/ethernet/intel/ice/ice_common.c | 224 +++++++++++++++
drivers/net/ethernet/intel/ice/ice_common.h | 20 +-
drivers/net/ethernet/intel/ice/ice_devids.h | 3 +
drivers/net/ethernet/intel/ice/ice_ethtool.c | 96 +++++++
drivers/net/ethernet/intel/ice/ice_lib.c | 6 +-
drivers/net/ethernet/intel/ice/ice_ptp.c | 35 +++
drivers/net/ethernet/intel/ice/ice_ptp_hw.c | 49 ++++
drivers/net/ethernet/intel/ice/ice_ptp_hw.h | 36 +++
drivers/net/ethernet/intel/ice/ice_type.h | 1 +
include/linux/ethtool.h | 9 +
include/uapi/linux/ethtool_netlink.h | 21 ++
net/ethtool/Makefile | 3 +-
net/ethtool/netlink.c | 20 ++
net/ethtool/netlink.h | 4 +
net/ethtool/synce.c | 267 ++++++++++++++++++
18 files changed, 929 insertions(+), 4 deletions(-)
create mode 100644 net/ethtool/synce.c
--
2.26.3
^ permalink raw reply
* [PATCH v5 net-next 1/4] ice: add support detecting features based on netlist
From: Maciej Machnikowski @ 2021-12-10 13:45 UTC (permalink / raw)
To: maciej.machnikowski, netdev, intel-wired-lan,
arkadiusz.kubalewski
Cc: richardcochran, abyagowi, anthony.l.nguyen, davem, kuba,
linux-kselftest, idosch, mkubecek, saeed, michael.chan, petrm
In-Reply-To: <20211210134550.1195182-1-maciej.machnikowski@intel.com>
Add new functions to check netlist of a given board for:
- Recovered Clock device,
- Clock Generation Unit,
- Clock Multiplexer,
Initialize feature bits depending on detected components.
Signed-off-by: Maciej Machnikowski <maciej.machnikowski@intel.com>
---
drivers/net/ethernet/intel/ice/ice.h | 2 +
.../net/ethernet/intel/ice/ice_adminq_cmd.h | 7 +-
drivers/net/ethernet/intel/ice/ice_common.c | 123 ++++++++++++++++++
drivers/net/ethernet/intel/ice/ice_common.h | 9 ++
drivers/net/ethernet/intel/ice/ice_lib.c | 6 +-
drivers/net/ethernet/intel/ice/ice_ptp_hw.c | 1 +
drivers/net/ethernet/intel/ice/ice_type.h | 1 +
7 files changed, 147 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index b67ad51cbcc9..cb6b4c53584b 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -183,6 +183,8 @@
enum ice_feature {
ICE_F_DSCP,
+ ICE_F_CGU,
+ ICE_F_PHY_RCLK,
ICE_F_SMA_CTRL,
ICE_F_MAX
};
diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index 4eef3488d86f..339c2a86f680 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -1297,6 +1297,8 @@ struct ice_aqc_link_topo_params {
#define ICE_AQC_LINK_TOPO_NODE_TYPE_CAGE 6
#define ICE_AQC_LINK_TOPO_NODE_TYPE_MEZZ 7
#define ICE_AQC_LINK_TOPO_NODE_TYPE_ID_EEPROM 8
+#define ICE_AQC_LINK_TOPO_NODE_TYPE_CLK_CTRL 9
+#define ICE_AQC_LINK_TOPO_NODE_TYPE_CLK_MUX 10
#define ICE_AQC_LINK_TOPO_NODE_CTX_S 4
#define ICE_AQC_LINK_TOPO_NODE_CTX_M \
(0xF << ICE_AQC_LINK_TOPO_NODE_CTX_S)
@@ -1333,7 +1335,10 @@ struct ice_aqc_link_topo_addr {
struct ice_aqc_get_link_topo {
struct ice_aqc_link_topo_addr addr;
u8 node_part_num;
-#define ICE_AQC_GET_LINK_TOPO_NODE_NR_PCA9575 0x21
+#define ICE_AQC_GET_LINK_TOPO_NODE_NR_PCA9575 0x21
+#define ICE_ACQ_GET_LINK_TOPO_NODE_NR_ZL30632_80032 0x24
+#define ICE_ACQ_GET_LINK_TOPO_NODE_NR_PKVL 0x31
+#define ICE_ACQ_GET_LINK_TOPO_NODE_NR_GEN_CLK_MUX 0x47
u8 rsvd[9];
};
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index b3066d0fea8b..35903b282885 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -274,6 +274,79 @@ ice_aq_get_link_topo_handle(struct ice_port_info *pi, u8 node_type,
return ice_aq_send_cmd(pi->hw, &desc, NULL, 0, cd);
}
+/**
+ * ice_aq_get_netlist_node
+ * @hw: pointer to the hw struct
+ * @cmd: get_link_topo AQ structure
+ * @node_part_number: output node part number if node found
+ * @node_handle: output node handle parameter if node found
+ */
+enum ice_status
+ice_aq_get_netlist_node(struct ice_hw *hw, struct ice_aqc_get_link_topo *cmd,
+ u8 *node_part_number, u16 *node_handle)
+{
+ struct ice_aq_desc desc;
+
+ ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_get_link_topo);
+ desc.params.get_link_topo = *cmd;
+
+ if (ice_aq_send_cmd(hw, &desc, NULL, 0, NULL))
+ return ICE_ERR_NOT_SUPPORTED;
+
+ if (node_handle)
+ *node_handle =
+ le16_to_cpu(desc.params.get_link_topo.addr.handle);
+ if (node_part_number)
+ *node_part_number = desc.params.get_link_topo.node_part_num;
+
+ return ICE_SUCCESS;
+}
+
+#define MAX_NETLIST_SIZE 10
+/**
+ * ice_find_netlist_node
+ * @hw: pointer to the hw struct
+ * @node_type_ctx: type of netlist node to look for
+ * @node_part_number: node part number to look for
+ * @node_handle: output parameter if node found - optional
+ *
+ * Find and return the node handle for a given node type and part number in the
+ * netlist. When found ICE_SUCCESS is returned, ICE_ERR_DOES_NOT_EXIST
+ * otherwise. If @node_handle provided, it would be set to found node handle.
+ */
+enum ice_status
+ice_find_netlist_node(struct ice_hw *hw, u8 node_type_ctx, u8 node_part_number,
+ u16 *node_handle)
+{
+ struct ice_aqc_get_link_topo cmd;
+ u8 rec_node_part_number;
+ enum ice_status status;
+ u16 rec_node_handle;
+ u8 idx;
+
+ for (idx = 0; idx < MAX_NETLIST_SIZE; idx++) {
+ memset(&cmd, 0, sizeof(cmd));
+
+ cmd.addr.topo_params.node_type_ctx =
+ (node_type_ctx << ICE_AQC_LINK_TOPO_NODE_TYPE_S);
+ cmd.addr.topo_params.index = idx;
+
+ status = ice_aq_get_netlist_node(hw, &cmd,
+ &rec_node_part_number,
+ &rec_node_handle);
+ if (status)
+ return status;
+
+ if (rec_node_part_number == node_part_number) {
+ if (node_handle)
+ *node_handle = rec_node_handle;
+ return ICE_SUCCESS;
+ }
+ }
+
+ return ICE_ERR_DOES_NOT_EXIST;
+}
+
/**
* ice_is_media_cage_present
* @pi: port information structure
@@ -5083,3 +5156,53 @@ bool ice_fw_supports_report_dflt_cfg(struct ice_hw *hw)
}
return false;
}
+
+/**
+ * ice_is_phy_rclk_present_e810t
+ * @hw: pointer to the hw struct
+ *
+ * Check if the PHY Recovered Clock device is present in the netlist
+ */
+bool ice_is_phy_rclk_present_e810t(struct ice_hw *hw)
+{
+ if (ice_find_netlist_node(hw, ICE_AQC_LINK_TOPO_NODE_TYPE_CLK_CTRL,
+ ICE_ACQ_GET_LINK_TOPO_NODE_NR_PKVL, NULL))
+ return false;
+
+ return true;
+}
+
+/**
+ * ice_is_cgu_present_e810t
+ * @hw: pointer to the hw struct
+ *
+ * Check if the Clock Generation Unit (CGU) device is present in the netlist
+ */
+bool ice_is_cgu_present_e810t(struct ice_hw *hw)
+{
+ if (!ice_find_netlist_node(hw, ICE_AQC_LINK_TOPO_NODE_TYPE_CLK_CTRL,
+ ICE_ACQ_GET_LINK_TOPO_NODE_NR_ZL30632_80032,
+ NULL)) {
+ hw->cgu_part_number =
+ ICE_ACQ_GET_LINK_TOPO_NODE_NR_ZL30632_80032;
+ return true;
+ }
+ return false;
+}
+
+/**
+ * ice_is_clock_mux_present_e810t
+ * @hw: pointer to the hw struct
+ *
+ * Check if the Clock Multiplexer device is present in the netlist
+ */
+bool ice_is_clock_mux_present_e810t(struct ice_hw *hw)
+{
+ if (ice_find_netlist_node(hw, ICE_AQC_LINK_TOPO_NODE_TYPE_CLK_MUX,
+ ICE_ACQ_GET_LINK_TOPO_NODE_NR_GEN_CLK_MUX,
+ NULL))
+ return false;
+
+ return true;
+}
+
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index 65c1b3244264..b20a5c085246 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -89,6 +89,12 @@ ice_aq_get_phy_caps(struct ice_port_info *pi, bool qual_mods, u8 report_mode,
struct ice_aqc_get_phy_caps_data *caps,
struct ice_sq_cd *cd);
enum ice_status
+ice_aq_get_netlist_node(struct ice_hw *hw, struct ice_aqc_get_link_topo *cmd,
+ u8 *node_part_number, u16 *node_handle);
+enum ice_status
+ice_find_netlist_node(struct ice_hw *hw, u8 node_type_ctx, u8 node_part_number,
+ u16 *node_handle);
+enum ice_status
ice_aq_list_caps(struct ice_hw *hw, void *buf, u16 buf_size, u32 *cap_count,
enum ice_adminq_opc opc, struct ice_sq_cd *cd);
enum ice_status
@@ -206,4 +212,7 @@ bool ice_fw_supports_lldp_fltr_ctrl(struct ice_hw *hw);
enum ice_status
ice_lldp_fltr_add_remove(struct ice_hw *hw, u16 vsi_num, bool add);
bool ice_fw_supports_report_dflt_cfg(struct ice_hw *hw);
+bool ice_is_phy_rclk_present_e810t(struct ice_hw *hw);
+bool ice_is_cgu_present_e810t(struct ice_hw *hw);
+bool ice_is_clock_mux_present_e810t(struct ice_hw *hw);
#endif /* _ICE_COMMON_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index 09a3297cd63c..18c30b2912e3 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -4188,8 +4188,12 @@ void ice_init_feature_support(struct ice_pf *pf)
case ICE_DEV_ID_E810C_QSFP:
case ICE_DEV_ID_E810C_SFP:
ice_set_feature_support(pf, ICE_F_DSCP);
- if (ice_is_e810t(&pf->hw))
+ if (ice_is_clock_mux_present_e810t(&pf->hw))
ice_set_feature_support(pf, ICE_F_SMA_CTRL);
+ if (ice_is_phy_rclk_present_e810t(&pf->hw))
+ ice_set_feature_support(pf, ICE_F_PHY_RCLK);
+ if (ice_is_cgu_present_e810t(&pf->hw))
+ ice_set_feature_support(pf, ICE_F_CGU);
break;
default:
break;
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index 29f947c0cd2e..aa257db36765 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -800,3 +800,4 @@ bool ice_is_pca9575_present(struct ice_hw *hw)
return !status && handle;
}
+
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index 9e0c2923c62e..a9dc16641bd4 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -920,6 +920,7 @@ struct ice_hw {
struct list_head rss_list_head;
struct ice_mbx_snapshot mbx_snapshot;
u16 io_expander_handle;
+ u8 cgu_part_number;
};
/* Statistics collected by each port, VSI, VEB, and S-channel */
--
2.26.3
^ permalink raw reply related
* [PATCH v5 net-next 2/4] ethtool: Add ability to configure recovered clock for SyncE feature
From: Maciej Machnikowski @ 2021-12-10 13:45 UTC (permalink / raw)
To: maciej.machnikowski, netdev, intel-wired-lan,
arkadiusz.kubalewski
Cc: richardcochran, abyagowi, anthony.l.nguyen, davem, kuba,
linux-kselftest, idosch, mkubecek, saeed, michael.chan, petrm
In-Reply-To: <20211210134550.1195182-1-maciej.machnikowski@intel.com>
Add netlink ethtool messages:
- ETHTOOL_MSG_RCLK_GET
- ETHTOOL_MSG_RCLK_SET
Required for controling basic SyncE functionality - configuration of
recovered reference clocks from PHY port on netdevice supporting SyncE.
- ETHTOOL_MSG_RCLK_GET - will read the current state if pin index is
given or will return allowed range if it's not
- ETHTOOL_MSG_RCLK_SET - will configure given recovered clock pin to
output recovered clock
Co-developed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Maciej Machnikowski <maciej.machnikowski@intel.com>
---
Documentation/networking/ethtool-netlink.rst | 62 +++++
include/linux/ethtool.h | 9 +
include/uapi/linux/ethtool_netlink.h | 21 ++
net/ethtool/Makefile | 3 +-
net/ethtool/netlink.c | 20 ++
net/ethtool/netlink.h | 4 +
net/ethtool/synce.c | 267 +++++++++++++++++++
7 files changed, 385 insertions(+), 1 deletion(-)
create mode 100644 net/ethtool/synce.c
diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 9d98e0511249..6399cdb3fdca 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -220,6 +220,8 @@ Userspace to kernel:
``ETHTOOL_MSG_PHC_VCLOCKS_GET`` get PHC virtual clocks info
``ETHTOOL_MSG_MODULE_SET`` set transceiver module parameters
``ETHTOOL_MSG_MODULE_GET`` get transceiver module parameters
+ ``ETHTOOL_RCLK_GET`` get recovered clock parameters
+ ``ETHTOOL_RCLK_SET`` set recovered clock parameters
===================================== =================================
Kernel to userspace:
@@ -260,6 +262,8 @@ Kernel to userspace:
``ETHTOOL_MSG_STATS_GET_REPLY`` standard statistics
``ETHTOOL_MSG_PHC_VCLOCKS_GET_REPLY`` PHC virtual clocks info
``ETHTOOL_MSG_MODULE_GET_REPLY`` transceiver module parameters
+ ``ETHTOOL_MSG_RCLK_GET_REPLY`` reference recovered clock config
+ ``ETHTOOL_MSG_RCLK_NTF`` reference recovered clock config
======================================== =================================
``GET`` requests are sent by userspace applications to retrieve device
@@ -1598,6 +1602,62 @@ For SFF-8636 modules, low power mode is forced by the host according to table
For CMIS modules, low power mode is forced by the host according to table 6-12
in revision 5.0 of the specification.
+RCLK_GET
+========
+
+Get status of an output pin for PHY recovered frequency clock signal.
+
+Request contents:
+
+ ====================================== ====== ==========================
+ ``ETHTOOL_A_RCLK_HEADER`` nested request header
+ ``ETHTOOL_A_RCLK_OUT_PIN_IDX`` u32 index of a pin
+ ====================================== ====== ==========================
+
+Kernel response contents:
+
+ ====================================== ====== ==========================
+ ``ETHTOOL_A_RCLK_OUT_PIN_IDX`` u32 index of a pin
+ ``ETHTOOL_A_RCLK_PIN_FLAGS`` u32 state of a pin
+ ``ETHTOOL_A_RCLK_RANGE_MIN_PIN`` u32 min index of RCLK pins
+ ``ETHTOOL_A_RCLK_RANGE_MAX_PIN`` u32 max index of RCLK pins
+ ====================================== ====== ==========================
+
+A device can support multiple recovered clock output pins. They typically are
+connected to the EEC and are used as a frequency reference.
+When a given pin on a given port is enabled, the PHY recovered frequency is
+fed onto that pin.
+Pins can be freely indexed and don't have to start with index 0.
+
+The ``ETHTOOL_A_RCLK_OUT_PIN_IDX`` is optional parameter. If present in
+the RCLK_GET request, the ``ETHTOOL_A_RCLK_PIN_FLAGS`` will be returned in a
+response, containing the state of the pin pointed by the index.
+
+If ``ETHTOOL_A_RCLK_OUT_PIN_IDX`` is not present in the RCLK_GET request,
+the range of available pins is returned:
+``ETHTOOL_A_RCLK_RANGE_MIN_PIN`` is the lowest recovered clock pin index.
+``ETHTOOL_A_RCLK_RANGE_MAX_PIN`` is the highest recovered clock pin index.
+
+RCLK_SET
+==========
+
+Set status of an output pin for PHY recovered frequency clock.
+
+Request contents:
+
+ ====================================== ====== ========================
+ ``ETHTOOL_A_RCLK_HEADER`` nested request header
+ ``ETHTOOL_A_RCLK_OUT_PIN_IDX`` u32 index of a pin
+ ``ETHTOOL_A_RCLK_PIN_FLAGS`` u32 requested state
+ ====================================== ====== ========================
+
+``ETHTOOL_A_RCLK_OUT_PIN_IDX`` is the index of the pin for which a change of
+state is requested.
+``ETHTOOL_A_RCLK_PIN_FLAGS`` if the ``ETHTOOL_RCLK_PIN_FLAGS_ENA`` flag
+is set, the given output pin will be enabled, if not - the pin will be disabled
+
+
+
Request translation
===================
@@ -1699,4 +1759,6 @@ are netlink only.
n/a ``ETHTOOL_MSG_PHC_VCLOCKS_GET``
n/a ``ETHTOOL_MSG_MODULE_GET``
n/a ``ETHTOOL_MSG_MODULE_SET``
+ n/a ``ETHTOOL_MSG_RCLK_GET``
+ n/a ``ETHTOOL_MSG_RCLK_SET``
=================================== =====================================
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index a26f37a27167..e344e5153f9b 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -614,6 +614,9 @@ struct ethtool_module_power_mode_params {
* plugged-in.
* @set_module_power_mode: Set the power mode policy for the plug-in module
* used by the network device.
+ * @get_rclk_range: Get range of valid Reference Clock input pins for the port.
+ * @get_rclk_state: Get state of Reference Clock input signal pin.
+ * @set_rclk_out: Enable/disable Reference Clock input signal pin.
*
* All operations are optional (i.e. the function pointer may be set
* to %NULL) and callers must take this into account. Callers must
@@ -750,6 +753,12 @@ struct ethtool_ops {
int (*set_module_power_mode)(struct net_device *dev,
const struct ethtool_module_power_mode_params *params,
struct netlink_ext_ack *extack);
+ int (*get_rclk_range)(struct net_device *dev, u32 *min_idx,
+ u32 *max_idx, struct netlink_ext_ack *extack);
+ int (*get_rclk_state)(struct net_device *dev, u32 out_idx,
+ bool *ena, struct netlink_ext_ack *extack);
+ int (*set_rclk_out)(struct net_device *dev, u32 out_idx, bool ena,
+ struct netlink_ext_ack *extack);
};
int ethtool_check_ops(const struct ethtool_ops *ops);
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index cca6e474a085..9a860f052a36 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -49,6 +49,8 @@ enum {
ETHTOOL_MSG_PHC_VCLOCKS_GET,
ETHTOOL_MSG_MODULE_GET,
ETHTOOL_MSG_MODULE_SET,
+ ETHTOOL_MSG_RCLK_GET,
+ ETHTOOL_MSG_RCLK_SET,
/* add new constants above here */
__ETHTOOL_MSG_USER_CNT,
@@ -94,6 +96,8 @@ enum {
ETHTOOL_MSG_PHC_VCLOCKS_GET_REPLY,
ETHTOOL_MSG_MODULE_GET_REPLY,
ETHTOOL_MSG_MODULE_NTF,
+ ETHTOOL_MSG_RCLK_GET_REPLY,
+ ETHTOOL_MSG_RCLK_NTF,
/* add new constants above here */
__ETHTOOL_MSG_KERNEL_CNT,
@@ -853,6 +857,23 @@ enum {
ETHTOOL_A_MODULE_MAX = (__ETHTOOL_A_MODULE_CNT - 1)
};
+/* REF CLK */
+
+enum {
+ ETHTOOL_A_RCLK_UNSPEC,
+ ETHTOOL_A_RCLK_HEADER, /* nest - _A_HEADER_* */
+ ETHTOOL_A_RCLK_OUT_PIN_IDX, /* u32 */
+ ETHTOOL_A_RCLK_PIN_FLAGS, /* u32 */
+ ETHTOOL_A_RCLK_PIN_MIN, /* u32 */
+ ETHTOOL_A_RCLK_PIN_MAX, /* u32 */
+
+ /* add new constants above here */
+ __ETHTOOL_A_RCLK_CNT,
+ ETHTOOL_A_RCLK_MAX = (__ETHTOOL_A_RCLK_CNT - 1)
+};
+
+#define ETHTOOL_RCLK_PIN_FLAGS_ENA (1 << 0)
+
/* generic netlink info */
#define ETHTOOL_GENL_NAME "ethtool"
#define ETHTOOL_GENL_VERSION 1
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index b76432e70e6b..dd6de311a9c2 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -7,4 +7,5 @@ obj-$(CONFIG_ETHTOOL_NETLINK) += ethtool_nl.o
ethtool_nl-y := netlink.o bitset.o strset.o linkinfo.o linkmodes.o \
linkstate.o debug.o wol.o features.o privflags.o rings.o \
channels.o coalesce.o pause.o eee.o tsinfo.o cabletest.o \
- tunnels.o fec.o eeprom.o stats.o phc_vclocks.o module.o
+ tunnels.o fec.o eeprom.o stats.o phc_vclocks.o module.o \
+ synce.o
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 23f32a995099..a88e4e91d015 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -285,6 +285,7 @@ ethnl_default_requests[__ETHTOOL_MSG_USER_CNT] = {
[ETHTOOL_MSG_STATS_GET] = ðnl_stats_request_ops,
[ETHTOOL_MSG_PHC_VCLOCKS_GET] = ðnl_phc_vclocks_request_ops,
[ETHTOOL_MSG_MODULE_GET] = ðnl_module_request_ops,
+ [ETHTOOL_MSG_RCLK_GET] = ðnl_rclk_request_ops,
};
static struct ethnl_dump_ctx *ethnl_dump_context(struct netlink_callback *cb)
@@ -597,6 +598,7 @@ ethnl_default_notify_ops[ETHTOOL_MSG_KERNEL_MAX + 1] = {
[ETHTOOL_MSG_EEE_NTF] = ðnl_eee_request_ops,
[ETHTOOL_MSG_FEC_NTF] = ðnl_fec_request_ops,
[ETHTOOL_MSG_MODULE_NTF] = ðnl_module_request_ops,
+ [ETHTOOL_MSG_RCLK_NTF] = ðnl_rclk_request_ops,
};
/* default notification handler */
@@ -692,6 +694,7 @@ static const ethnl_notify_handler_t ethnl_notify_handlers[] = {
[ETHTOOL_MSG_EEE_NTF] = ethnl_default_notify,
[ETHTOOL_MSG_FEC_NTF] = ethnl_default_notify,
[ETHTOOL_MSG_MODULE_NTF] = ethnl_default_notify,
+ [ETHTOOL_MSG_RCLK_NTF] = ethnl_default_notify,
};
void ethtool_notify(struct net_device *dev, unsigned int cmd, const void *data)
@@ -1021,6 +1024,23 @@ static const struct genl_ops ethtool_genl_ops[] = {
.policy = ethnl_module_set_policy,
.maxattr = ARRAY_SIZE(ethnl_module_set_policy) - 1,
},
+ {
+ .cmd = ETHTOOL_MSG_RCLK_GET,
+ .flags = GENL_UNS_ADMIN_PERM,
+ .doit = ethnl_default_doit,
+ .start = ethnl_default_start,
+ .dumpit = ethnl_default_dumpit,
+ .done = ethnl_default_done,
+ .policy = ethnl_rclk_get_policy,
+ .maxattr = ARRAY_SIZE(ethnl_rclk_get_policy) - 1,
+ },
+ {
+ .cmd = ETHTOOL_MSG_RCLK_SET,
+ .flags = GENL_UNS_ADMIN_PERM,
+ .doit = ethnl_set_rclk,
+ .policy = ethnl_rclk_set_policy,
+ .maxattr = ARRAY_SIZE(ethnl_rclk_set_policy) - 1,
+ },
};
static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index a779bbb0c524..cfc7c45513b1 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -340,6 +340,7 @@ extern const struct ethnl_request_ops ethnl_module_eeprom_request_ops;
extern const struct ethnl_request_ops ethnl_stats_request_ops;
extern const struct ethnl_request_ops ethnl_phc_vclocks_request_ops;
extern const struct ethnl_request_ops ethnl_module_request_ops;
+extern const struct ethnl_request_ops ethnl_rclk_request_ops;
extern const struct nla_policy ethnl_header_policy[ETHTOOL_A_HEADER_FLAGS + 1];
extern const struct nla_policy ethnl_header_policy_stats[ETHTOOL_A_HEADER_FLAGS + 1];
@@ -378,6 +379,8 @@ extern const struct nla_policy ethnl_stats_get_policy[ETHTOOL_A_STATS_GROUPS + 1
extern const struct nla_policy ethnl_phc_vclocks_get_policy[ETHTOOL_A_PHC_VCLOCKS_HEADER + 1];
extern const struct nla_policy ethnl_module_get_policy[ETHTOOL_A_MODULE_HEADER + 1];
extern const struct nla_policy ethnl_module_set_policy[ETHTOOL_A_MODULE_POWER_MODE_POLICY + 1];
+extern const struct nla_policy ethnl_rclk_get_policy[ETHTOOL_A_RCLK_OUT_PIN_IDX + 1];
+extern const struct nla_policy ethnl_rclk_set_policy[ETHTOOL_A_RCLK_PIN_FLAGS + 1];
int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info);
int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info);
@@ -397,6 +400,7 @@ int ethnl_tunnel_info_start(struct netlink_callback *cb);
int ethnl_tunnel_info_dumpit(struct sk_buff *skb, struct netlink_callback *cb);
int ethnl_set_fec(struct sk_buff *skb, struct genl_info *info);
int ethnl_set_module(struct sk_buff *skb, struct genl_info *info);
+int ethnl_set_rclk(struct sk_buff *skb, struct genl_info *info);
extern const char stats_std_names[__ETHTOOL_STATS_CNT][ETH_GSTRING_LEN];
extern const char stats_eth_phy_names[__ETHTOOL_A_STATS_ETH_PHY_CNT][ETH_GSTRING_LEN];
diff --git a/net/ethtool/synce.c b/net/ethtool/synce.c
new file mode 100644
index 000000000000..f4ebb4c57d4d
--- /dev/null
+++ b/net/ethtool/synce.c
@@ -0,0 +1,267 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/ethtool.h>
+#include "netlink.h"
+
+struct rclk_out_pin_info {
+ u32 idx;
+ bool valid;
+};
+
+struct rclk_request_data {
+ struct ethnl_req_info base;
+ struct rclk_out_pin_info out_pin;
+};
+
+struct rclk_pin_state_info {
+ u32 range_min;
+ u32 range_max;
+ u32 flags;
+ u32 idx;
+};
+
+struct rclk_reply_data {
+ struct ethnl_reply_data base;
+ struct rclk_pin_state_info pin_state;
+};
+
+#define RCLK_REPDATA(__reply_base) \
+ container_of(__reply_base, struct rclk_reply_data, base)
+
+#define RCLK_REQDATA(__req_base) \
+ container_of(__req_base, struct rclk_request_data, base)
+
+/* RCLK_GET */
+
+const struct nla_policy
+ethnl_rclk_get_policy[ETHTOOL_A_RCLK_OUT_PIN_IDX + 1] = {
+ [ETHTOOL_A_RCLK_HEADER] = NLA_POLICY_NESTED(ethnl_header_policy),
+ [ETHTOOL_A_RCLK_OUT_PIN_IDX] = { .type = NLA_U32 },
+};
+
+static int rclk_parse_request(struct ethnl_req_info *req_base,
+ struct nlattr **tb,
+ struct netlink_ext_ack *extack)
+{
+ struct rclk_request_data *req = RCLK_REQDATA(req_base);
+
+ if (tb[ETHTOOL_A_RCLK_OUT_PIN_IDX]) {
+ req->out_pin.idx = nla_get_u32(tb[ETHTOOL_A_RCLK_OUT_PIN_IDX]);
+ req->out_pin.valid = true;
+ }
+
+ return 0;
+}
+
+static int rclk_state_get(struct net_device *dev,
+ struct rclk_reply_data *data,
+ struct netlink_ext_ack *extack,
+ u32 out_idx)
+{
+ const struct ethtool_ops *ops = dev->ethtool_ops;
+ bool pin_state;
+ int ret;
+
+ if (!ops->get_rclk_state)
+ return -EOPNOTSUPP;
+
+ ret = ops->get_rclk_state(dev, out_idx, &pin_state, extack);
+ if (ret)
+ return ret;
+
+ data->pin_state.flags = pin_state ? ETHTOOL_RCLK_PIN_FLAGS_ENA : 0;
+ data->pin_state.idx = out_idx;
+
+ return ret;
+}
+
+static int rclk_range_get(struct net_device *dev,
+ struct rclk_reply_data *data,
+ struct netlink_ext_ack *extack)
+{
+ const struct ethtool_ops *ops = dev->ethtool_ops;
+ u32 min_idx, max_idx;
+ int ret;
+
+ if (!ops->get_rclk_range)
+ return -EOPNOTSUPP;
+
+ ret = ops->get_rclk_range(dev, &min_idx, &max_idx, extack);
+ if (ret)
+ return ret;
+
+ data->pin_state.range_min = min_idx;
+ data->pin_state.range_max = max_idx;
+
+ return ret;
+}
+
+static int rclk_prepare_data(const struct ethnl_req_info *req_base,
+ struct ethnl_reply_data *reply_base,
+ struct genl_info *info)
+{
+ struct rclk_reply_data *reply = RCLK_REPDATA(reply_base);
+ struct rclk_request_data *request = RCLK_REQDATA(req_base);
+ struct netlink_ext_ack *extack = info ? info->extack : NULL;
+ struct net_device *dev = reply_base->dev;
+ int ret;
+
+ memset(&reply->pin_state, 0, sizeof(reply->pin_state));
+ ret = ethnl_ops_begin(dev);
+ if (ret < 0)
+ return ret;
+
+ if (request->out_pin.valid)
+ ret = rclk_state_get(dev, reply, extack,
+ request->out_pin.idx);
+ else
+ ret = rclk_range_get(dev, reply, extack);
+
+ ethnl_ops_complete(dev);
+
+ return ret;
+}
+
+static int rclk_fill_reply(struct sk_buff *skb,
+ const struct ethnl_req_info *req_base,
+ const struct ethnl_reply_data *reply_base)
+{
+ const struct rclk_reply_data *reply = RCLK_REPDATA(reply_base);
+ const struct rclk_request_data *request = RCLK_REQDATA(req_base);
+
+ if (request->out_pin.valid) {
+ if (nla_put_u32(skb, ETHTOOL_A_RCLK_PIN_FLAGS,
+ reply->pin_state.flags))
+ return -EMSGSIZE;
+ if (nla_put_u32(skb, ETHTOOL_A_RCLK_OUT_PIN_IDX,
+ reply->pin_state.idx))
+ return -EMSGSIZE;
+ } else {
+ if (nla_put_u32(skb, ETHTOOL_A_RCLK_PIN_MIN,
+ reply->pin_state.range_min))
+ return -EMSGSIZE;
+ if (nla_put_u32(skb, ETHTOOL_A_RCLK_PIN_MAX,
+ reply->pin_state.range_max))
+ return -EMSGSIZE;
+ }
+
+ return 0;
+}
+
+static int rclk_reply_size(const struct ethnl_req_info *req_base,
+ const struct ethnl_reply_data *reply_base)
+{
+ const struct rclk_request_data *request = RCLK_REQDATA(req_base);
+
+ if (request->out_pin.valid)
+ return nla_total_size(sizeof(u32)) + /* ETHTOOL_A_RCLK_PIN_FLAGS */
+ nla_total_size(sizeof(u32)); /* ETHTOOL_A_RCLK_OUT_PIN_IDX */
+ else
+ return nla_total_size(sizeof(u32)) + /* ETHTOOL_A_RCLK_PIN_MIN */
+ nla_total_size(sizeof(u32)); /* ETHTOOL_A_RCLK_PIN_MAX */
+}
+
+const struct ethnl_request_ops ethnl_rclk_request_ops = {
+ .request_cmd = ETHTOOL_MSG_RCLK_GET,
+ .reply_cmd = ETHTOOL_MSG_RCLK_GET_REPLY,
+ .hdr_attr = ETHTOOL_A_RCLK_HEADER,
+ .req_info_size = sizeof(struct rclk_request_data),
+ .reply_data_size = sizeof(struct rclk_reply_data),
+
+ .parse_request = rclk_parse_request,
+ .prepare_data = rclk_prepare_data,
+ .reply_size = rclk_reply_size,
+ .fill_reply = rclk_fill_reply,
+};
+
+/* RCLK SET */
+
+const struct nla_policy
+ethnl_rclk_set_policy[ETHTOOL_A_RCLK_PIN_FLAGS + 1] = {
+ [ETHTOOL_A_RCLK_HEADER] = NLA_POLICY_NESTED(ethnl_header_policy),
+ [ETHTOOL_A_RCLK_OUT_PIN_IDX] = { .type = NLA_U32 },
+ [ETHTOOL_A_RCLK_PIN_FLAGS] = { .type = NLA_U32 },
+};
+
+static int rclk_set_state(struct net_device *dev, struct nlattr **tb,
+ bool *p_mod, struct netlink_ext_ack *extack)
+{
+ const struct ethtool_ops *ops = dev->ethtool_ops;
+ bool old_state, new_state;
+ u32 min_idx, max_idx;
+ u32 out_idx;
+ int ret;
+
+ if (!tb[ETHTOOL_A_RCLK_PIN_FLAGS] &&
+ !tb[ETHTOOL_A_RCLK_OUT_PIN_IDX])
+ return 0;
+
+ if (!ops->set_rclk_out || !ops->get_rclk_range) {
+ NL_SET_ERR_MSG_ATTR(extack,
+ tb[ETHTOOL_A_RCLK_PIN_FLAGS],
+ "Setting recovered clock state is not supported by this device");
+ return -EOPNOTSUPP;
+ }
+
+ ret = ops->get_rclk_range(dev, &min_idx, &max_idx, extack);
+ if (ret)
+ return ret;
+
+ out_idx = nla_get_u32(tb[ETHTOOL_A_RCLK_OUT_PIN_IDX]);
+ if (out_idx < min_idx || out_idx > max_idx) {
+ NL_SET_ERR_MSG_ATTR(extack,
+ tb[ETHTOOL_A_RCLK_OUT_PIN_IDX],
+ "Requested recovered clock pin index is out of range");
+ return -EINVAL;
+ }
+
+ ret = ops->get_rclk_state(dev, out_idx, &old_state, extack);
+ if (ret < 0)
+ return ret;
+
+ new_state = !!(nla_get_u32(tb[ETHTOOL_A_RCLK_PIN_FLAGS]) &
+ ETHTOOL_RCLK_PIN_FLAGS_ENA);
+
+ /* If state changed - flag need for sending the notification */
+ *p_mod = old_state != new_state;
+
+ return ops->set_rclk_out(dev, out_idx, new_state, extack);
+}
+
+int ethnl_set_rclk(struct sk_buff *skb, struct genl_info *info)
+{
+ struct ethnl_req_info req_info = {};
+ struct nlattr **tb = info->attrs;
+ struct net_device *dev;
+ bool mod = false;
+ int ret;
+
+ ret = ethnl_parse_header_dev_get(&req_info, tb[ETHTOOL_A_RCLK_HEADER],
+ genl_info_net(info), info->extack,
+ true);
+ if (ret < 0)
+ return ret;
+ dev = req_info.dev;
+
+ rtnl_lock();
+ ret = ethnl_ops_begin(dev);
+ if (ret < 0)
+ goto out_rtnl;
+
+ ret = rclk_set_state(dev, tb, &mod, info->extack);
+ if (ret < 0)
+ goto out_ops;
+
+ if (!mod)
+ goto out_ops;
+
+ ethtool_notify(dev, ETHTOOL_MSG_RCLK_NTF, NULL);
+
+out_ops:
+ ethnl_ops_complete(dev);
+out_rtnl:
+ rtnl_unlock();
+ dev_put(dev);
+ return ret;
+}
+
--
2.26.3
^ permalink raw reply related
* [PATCH v5 net-next 3/4] ice: add support for monitoring SyncE DPLL state
From: Maciej Machnikowski @ 2021-12-10 13:45 UTC (permalink / raw)
To: maciej.machnikowski, netdev, intel-wired-lan,
arkadiusz.kubalewski
Cc: richardcochran, abyagowi, anthony.l.nguyen, davem, kuba,
linux-kselftest, idosch, mkubecek, saeed, michael.chan, petrm
In-Reply-To: <20211210134550.1195182-1-maciej.machnikowski@intel.com>
Implement SyncE DPLL monitoring for E810-T devices.
Poll loop will periodically check the state of the DPLL and cache it
in the pf structure. State changes will be logged in the system log.
Signed-off-by: Maciej Machnikowski <maciej.machnikowski@intel.com>
---
drivers/net/ethernet/intel/ice/ice.h | 5 ++
.../net/ethernet/intel/ice/ice_adminq_cmd.h | 34 +++++++++++++
drivers/net/ethernet/intel/ice/ice_common.c | 36 ++++++++++++++
drivers/net/ethernet/intel/ice/ice_common.h | 5 +-
drivers/net/ethernet/intel/ice/ice_devids.h | 3 ++
drivers/net/ethernet/intel/ice/ice_ptp.c | 35 ++++++++++++++
drivers/net/ethernet/intel/ice/ice_ptp_hw.c | 48 +++++++++++++++++++
drivers/net/ethernet/intel/ice/ice_ptp_hw.h | 34 +++++++++++++
8 files changed, 199 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index cb6b4c53584b..2dcc8fd6dff5 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -607,6 +607,11 @@ struct ice_pf {
#define ICE_VF_AGG_NODE_ID_START 65
#define ICE_MAX_VF_AGG_NODES 32
struct ice_agg_node vf_agg_node[ICE_MAX_VF_AGG_NODES];
+
+ enum ice_eec_state synce_dpll_state;
+ u8 synce_dpll_pin;
+ enum ice_eec_state ptp_dpll_state;
+ u8 ptp_dpll_pin;
};
struct ice_netdev_priv {
diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index 339c2a86f680..11226af7a9a4 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -1808,6 +1808,36 @@ struct ice_aqc_add_rdma_qset_data {
struct ice_aqc_add_tx_rdma_qset_entry rdma_qsets[];
};
+/* Get CGU DPLL status (direct 0x0C66) */
+struct ice_aqc_get_cgu_dpll_status {
+ u8 dpll_num;
+ u8 ref_state;
+#define ICE_AQC_GET_CGU_DPLL_STATUS_REF_SW_LOS BIT(0)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_REF_SW_SCM BIT(1)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_REF_SW_CFM BIT(2)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_REF_SW_GST BIT(3)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_REF_SW_PFM BIT(4)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_REF_SW_ESYNC BIT(6)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_FAST_LOCK_EN BIT(7)
+ __le16 dpll_state;
+#define ICE_AQC_GET_CGU_DPLL_STATUS_STATE_LOCK BIT(0)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_STATE_HO BIT(1)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_STATE_HO_READY BIT(2)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_STATE_FLHIT BIT(5)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_STATE_PSLHIT BIT(7)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_STATE_CLK_REF_SHIFT 8
+#define ICE_AQC_GET_CGU_DPLL_STATUS_STATE_CLK_REF_SEL \
+ ICE_M(0x1F, ICE_AQC_GET_CGU_DPLL_STATUS_STATE_CLK_REF_SHIFT)
+#define ICE_AQC_GET_CGU_DPLL_STATUS_STATE_MODE_SHIFT 13
+#define ICE_AQC_GET_CGU_DPLL_STATUS_STATE_MODE \
+ ICE_M(0x7, ICE_AQC_GET_CGU_DPLL_STATUS_STATE_MODE_SHIFT)
+ __le32 phase_offset_h;
+ __le32 phase_offset_l;
+ u8 eec_mode;
+ u8 rsvd[1];
+ __le16 node_handle;
+};
+
/* Configure Firmware Logging Command (indirect 0xFF09)
* Logging Information Read Response (indirect 0xFF10)
* Note: The 0xFF10 command has no input parameters.
@@ -2039,6 +2069,7 @@ struct ice_aq_desc {
struct ice_aqc_fw_logging fw_logging;
struct ice_aqc_get_clear_fw_log get_clear_fw_log;
struct ice_aqc_download_pkg download_pkg;
+ struct ice_aqc_get_cgu_dpll_status get_cgu_dpll_status;
struct ice_aqc_driver_shared_params drv_shared_params;
struct ice_aqc_set_mac_lb set_mac_lb;
struct ice_aqc_alloc_free_res_cmd sw_res_ctrl;
@@ -2205,6 +2236,9 @@ enum ice_adminq_opc {
ice_aqc_opc_update_pkg = 0x0C42,
ice_aqc_opc_get_pkg_info_list = 0x0C43,
+ /* 1588/SyncE commands/events */
+ ice_aqc_opc_get_cgu_dpll_status = 0x0C66,
+
ice_aqc_opc_driver_shared_params = 0x0C90,
/* Standalone Commands/Events */
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 35903b282885..8069141ac105 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -4644,6 +4644,42 @@ ice_dis_vsi_rdma_qset(struct ice_port_info *pi, u16 count, u32 *qset_teid,
return ice_status_to_errno(status);
}
+/**
+ * ice_aq_get_cgu_dpll_status
+ * @hw: pointer to the HW struct
+ * @dpll_num: DPLL index
+ * @ref_state: Reference clock state
+ * @dpll_state: DPLL state
+ * @phase_offset: Phase offset in ps
+ * @eec_mode: EEC_mode
+ *
+ * Get CGU DPLL status (0x0C66)
+ */
+enum ice_status
+ice_aq_get_cgu_dpll_status(struct ice_hw *hw, u8 dpll_num, u8 *ref_state,
+ u16 *dpll_state, u64 *phase_offset, u8 *eec_mode)
+{
+ struct ice_aqc_get_cgu_dpll_status *cmd;
+ struct ice_aq_desc desc;
+ enum ice_status status;
+
+ ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_get_cgu_dpll_status);
+ cmd = &desc.params.get_cgu_dpll_status;
+ cmd->dpll_num = dpll_num;
+
+ status = ice_aq_send_cmd(hw, &desc, NULL, 0, NULL);
+ if (!status) {
+ *ref_state = cmd->ref_state;
+ *dpll_state = le16_to_cpu(cmd->dpll_state);
+ *phase_offset = le32_to_cpu(cmd->phase_offset_h);
+ *phase_offset <<= 32;
+ *phase_offset += le32_to_cpu(cmd->phase_offset_l);
+ *eec_mode = cmd->eec_mode;
+ }
+
+ return status;
+}
+
/**
* ice_replay_pre_init - replay pre initialization
* @hw: pointer to the HW struct
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index b20a5c085246..aaed388a40a8 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -106,6 +106,7 @@ enum ice_status
ice_aq_manage_mac_write(struct ice_hw *hw, const u8 *mac_addr, u8 flags,
struct ice_sq_cd *cd);
bool ice_is_e810(struct ice_hw *hw);
+bool ice_is_e810t(struct ice_hw *hw);
enum ice_status ice_clear_pf_cfg(struct ice_hw *hw);
enum ice_status
ice_aq_set_phy_cfg(struct ice_hw *hw, struct ice_port_info *pi,
@@ -162,6 +163,9 @@ ice_cfg_vsi_rdma(struct ice_port_info *pi, u16 vsi_handle, u16 tc_bitmap,
int
ice_ena_vsi_rdma_qset(struct ice_port_info *pi, u16 vsi_handle, u8 tc,
u16 *rdma_qset, u16 num_qsets, u32 *qset_teid);
+enum ice_status
+ice_aq_get_cgu_dpll_status(struct ice_hw *hw, u8 dpll_num, u8 *ref_state,
+ u16 *dpll_state, u64 *phase_offset, u8 *eec_mode);
int
ice_dis_vsi_rdma_qset(struct ice_port_info *pi, u16 count, u32 *qset_teid,
u16 *q_id);
@@ -189,7 +193,6 @@ ice_stat_update40(struct ice_hw *hw, u32 reg, bool prev_stat_loaded,
void
ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded,
u64 *prev_stat, u64 *cur_stat);
-bool ice_is_e810t(struct ice_hw *hw);
enum ice_status
ice_sched_query_elem(struct ice_hw *hw, u32 node_teid,
struct ice_aqc_txsched_elem_data *buf);
diff --git a/drivers/net/ethernet/intel/ice/ice_devids.h b/drivers/net/ethernet/intel/ice/ice_devids.h
index 61dd2f18dee8..0b654d417d29 100644
--- a/drivers/net/ethernet/intel/ice/ice_devids.h
+++ b/drivers/net/ethernet/intel/ice/ice_devids.h
@@ -58,4 +58,7 @@
/* Intel(R) Ethernet Connection E822-L 1GbE */
#define ICE_DEV_ID_E822L_SGMII 0x189A
+#define ICE_SUBDEV_ID_E810T 0x000E
+#define ICE_SUBDEV_ID_E810T2 0x000F
+
#endif /* _ICE_DEVIDS_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index bf7247c6f58e..bb502c19d53a 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -1766,6 +1766,36 @@ static void ice_ptp_tx_tstamp_cleanup(struct ice_ptp_tx *tx)
}
}
+static void ice_handle_cgu_state(struct ice_pf *pf)
+{
+ enum ice_eec_state cgu_state;
+ u8 pin;
+
+ cgu_state = ice_get_zl_dpll_state(&pf->hw, ICE_CGU_DPLL_SYNCE, &pin);
+ if (pf->synce_dpll_state != cgu_state) {
+ pf->synce_dpll_state = cgu_state;
+ pf->synce_dpll_pin = pin;
+
+ dev_warn(ice_pf_to_dev(pf),
+ "<DPLL%i> state changed to: %d, pin %d",
+ ICE_CGU_DPLL_SYNCE,
+ pf->synce_dpll_state,
+ pin);
+ }
+
+ cgu_state = ice_get_zl_dpll_state(&pf->hw, ICE_CGU_DPLL_PTP, &pin);
+ if (pf->ptp_dpll_state != cgu_state) {
+ pf->ptp_dpll_state = cgu_state;
+ pf->ptp_dpll_pin = pin;
+
+ dev_warn(ice_pf_to_dev(pf),
+ "<DPLL%i> state changed to: %d, pin %d",
+ ICE_CGU_DPLL_PTP,
+ pf->ptp_dpll_state,
+ pin);
+ }
+}
+
static void ice_ptp_periodic_work(struct kthread_work *work)
{
struct ice_ptp *ptp = container_of(work, struct ice_ptp, work.work);
@@ -1774,6 +1804,10 @@ static void ice_ptp_periodic_work(struct kthread_work *work)
if (!test_bit(ICE_FLAG_PTP, pf->flags))
return;
+ if (ice_is_feature_supported(pf, ICE_F_CGU) &&
+ pf->hw.func_caps.ts_func_info.src_tmr_owned)
+ ice_handle_cgu_state(pf);
+
ice_ptp_update_cached_phctime(pf);
ice_ptp_tx_tstamp_cleanup(&pf->ptp.port.tx);
@@ -1958,3 +1992,4 @@ void ice_ptp_release(struct ice_pf *pf)
dev_info(ice_pf_to_dev(pf), "Removed PTP clock\n");
}
+
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index aa257db36765..b4300bf3e4ce 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -375,6 +375,54 @@ static int ice_ptp_port_cmd_e810(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd)
return 0;
}
+/**
+ * ice_get_zl_dpll_state - get the state of the DPLL
+ * @hw: pointer to the hw struct
+ * @dpll_idx: Index of internal DPLL unit
+ * @pin: pointer to a buffer for returning currently active pin
+ *
+ * This function will read the state of the DPLL(dpll_idx). If optional
+ * parameter pin is given it'll be used to retrieve currently active pin.
+ *
+ * Return: state of the DPLL
+ */
+enum ice_eec_state
+ice_get_zl_dpll_state(struct ice_hw *hw, u8 dpll_idx, u8 *pin)
+{
+ enum ice_status status;
+ u64 phase_offset;
+ u16 dpll_state;
+ u8 ref_state;
+ u8 eec_mode;
+
+ if (dpll_idx >= ICE_CGU_DPLL_MAX)
+ return ICE_EEC_STATE_INVALID;
+
+ status = ice_aq_get_cgu_dpll_status(hw, dpll_idx, &ref_state,
+ &dpll_state, &phase_offset,
+ &eec_mode);
+ if (status)
+ return ICE_EEC_STATE_INVALID;
+
+ if (pin) {
+ /* current ref pin in dpll_state_refsel_status_X register */
+ *pin = (dpll_state &
+ ICE_AQC_GET_CGU_DPLL_STATUS_STATE_CLK_REF_SEL) >>
+ ICE_AQC_GET_CGU_DPLL_STATUS_STATE_CLK_REF_SHIFT;
+ }
+
+ if (dpll_state & ICE_AQC_GET_CGU_DPLL_STATUS_STATE_LOCK) {
+ if (dpll_state & ICE_AQC_GET_CGU_DPLL_STATUS_STATE_HO_READY)
+ return ICE_EEC_STATE_LOCKED_HO_ACQ;
+ else
+ return ICE_EEC_STATE_LOCKED;
+ } else if ((dpll_state & ICE_AQC_GET_CGU_DPLL_STATUS_STATE_HO) &&
+ (dpll_state & ICE_AQC_GET_CGU_DPLL_STATUS_STATE_HO_READY)) {
+ return ICE_EEC_STATE_HOLDOVER;
+ }
+ return ICE_EEC_STATE_FREERUN;
+}
+
/* Device agnostic functions
*
* The following functions implement useful behavior to hide the differences
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
index b2984b5c22c1..28b04ec40bae 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
@@ -12,6 +12,18 @@ enum ice_ptp_tmr_cmd {
READ_TIME
};
+enum ice_eec_state {
+ ICE_EEC_STATE_INVALID = 0, /* state is not valid */
+ ICE_EEC_STATE_FREERUN, /* clock is free-running */
+ ICE_EEC_STATE_LOCKED, /* clock is locked to the reference,
+ * but the holdover memory is not valid
+ */
+ ICE_EEC_STATE_LOCKED_HO_ACQ, /* clock is locked to the reference
+ * and holdover memory is valid
+ */
+ ICE_EEC_STATE_HOLDOVER, /* clock is in holdover mode */
+};
+
/* Increment value to generate nanoseconds in the GLTSYN_TIME_L register for
* the E810 devices. Based off of a PLL with an 812.5 MHz frequency.
*/
@@ -33,6 +45,8 @@ int ice_ptp_init_phy_e810(struct ice_hw *hw);
int ice_read_sma_ctrl_e810t(struct ice_hw *hw, u8 *data);
int ice_write_sma_ctrl_e810t(struct ice_hw *hw, u8 data);
bool ice_is_pca9575_present(struct ice_hw *hw);
+enum ice_eec_state
+ice_get_zl_dpll_state(struct ice_hw *hw, u8 dpll_idx, u8 *pin);
#define PFTSYN_SEM_BYTES 4
@@ -98,4 +112,24 @@ bool ice_is_pca9575_present(struct ice_hw *hw);
#define ICE_SMA_MAX_BIT_E810T 7
#define ICE_PCA9575_P1_OFFSET 8
+enum ice_e810t_cgu_dpll {
+ ICE_CGU_DPLL_SYNCE,
+ ICE_CGU_DPLL_PTP,
+ ICE_CGU_DPLL_MAX
+};
+
+enum ice_e810t_cgu_pins {
+ REF0P,
+ REF0N,
+ REF1P,
+ REF1N,
+ REF2P,
+ REF2N,
+ REF3P,
+ REF3N,
+ REF4P,
+ REF4N,
+ NUM_E810T_CGU_PINS
+};
+
#endif /* _ICE_PTP_HW_H_ */
--
2.26.3
^ permalink raw reply related
* [PATCH v5 net-next 4/4] ice: add support for recovered clocks
From: Maciej Machnikowski @ 2021-12-10 13:45 UTC (permalink / raw)
To: maciej.machnikowski, netdev, intel-wired-lan,
arkadiusz.kubalewski
Cc: richardcochran, abyagowi, anthony.l.nguyen, davem, kuba,
linux-kselftest, idosch, mkubecek, saeed, michael.chan, petrm
In-Reply-To: <20211210134550.1195182-1-maciej.machnikowski@intel.com>
Implement ethtool netlink functions for handling recovered clocks
configuration on ice driver:
- ETHTOOL_MSG_RCLK_SET
- ETHTOOL_MSG_RCLK_GET
Co-developed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Maciej Machnikowski <maciej.machnikowski@intel.com>
---
.../net/ethernet/intel/ice/ice_adminq_cmd.h | 29 ++++++
drivers/net/ethernet/intel/ice/ice_common.c | 65 +++++++++++++
drivers/net/ethernet/intel/ice/ice_common.h | 6 ++
drivers/net/ethernet/intel/ice/ice_ethtool.c | 96 +++++++++++++++++++
drivers/net/ethernet/intel/ice/ice_ptp_hw.h | 2 +
5 files changed, 198 insertions(+)
diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index 11226af7a9a4..aed03200bb99 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -1281,6 +1281,31 @@ struct ice_aqc_set_mac_lb {
u8 reserved[15];
};
+/* Set PHY recovered clock output (direct 0x0630) */
+struct ice_aqc_set_phy_rec_clk_out {
+ u8 phy_output;
+ u8 port_num;
+ u8 flags;
+#define ICE_AQC_SET_PHY_REC_CLK_OUT_OUT_EN BIT(0)
+#define ICE_AQC_SET_PHY_REC_CLK_OUT_CURR_PORT 0xFF
+ u8 rsvd;
+ __le32 freq;
+ u8 rsvd2[6];
+ __le16 node_handle;
+};
+
+/* Get PHY recovered clock output (direct 0x0631) */
+struct ice_aqc_get_phy_rec_clk_out {
+ u8 phy_output;
+ u8 port_num;
+ u8 flags;
+#define ICE_AQC_GET_PHY_REC_CLK_OUT_OUT_EN BIT(0)
+ u8 rsvd;
+ __le32 freq;
+ u8 rsvd2[6];
+ __le16 node_handle;
+};
+
struct ice_aqc_link_topo_params {
u8 lport_num;
u8 lport_num_valid;
@@ -2033,6 +2058,8 @@ struct ice_aq_desc {
struct ice_aqc_get_phy_caps get_phy;
struct ice_aqc_set_phy_cfg set_phy;
struct ice_aqc_restart_an restart_an;
+ struct ice_aqc_set_phy_rec_clk_out set_phy_rec_clk_out;
+ struct ice_aqc_get_phy_rec_clk_out get_phy_rec_clk_out;
struct ice_aqc_gpio read_write_gpio;
struct ice_aqc_sff_eeprom read_write_sff_param;
struct ice_aqc_set_port_id_led set_port_id_led;
@@ -2188,6 +2215,8 @@ enum ice_adminq_opc {
ice_aqc_opc_get_link_status = 0x0607,
ice_aqc_opc_set_event_mask = 0x0613,
ice_aqc_opc_set_mac_lb = 0x0620,
+ ice_aqc_opc_set_phy_rec_clk_out = 0x0630,
+ ice_aqc_opc_get_phy_rec_clk_out = 0x0631,
ice_aqc_opc_get_link_topo = 0x06E0,
ice_aqc_opc_set_port_id_led = 0x06E9,
ice_aqc_opc_set_gpio = 0x06EC,
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 8069141ac105..29d302ea1e56 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -5242,3 +5242,68 @@ bool ice_is_clock_mux_present_e810t(struct ice_hw *hw)
return true;
}
+/**
+ * ice_aq_set_phy_rec_clk_out - set RCLK phy out
+ * @hw: pointer to the HW struct
+ * @phy_output: PHY reference clock output pin
+ * @enable: GPIO state to be applied
+ * @freq: PHY output frequency
+ *
+ * Set CGU reference priority (0x0630)
+ * Return 0 on success or negative value on failure.
+ */
+enum ice_status
+ice_aq_set_phy_rec_clk_out(struct ice_hw *hw, u8 phy_output, bool enable,
+ u32 *freq)
+{
+ struct ice_aqc_set_phy_rec_clk_out *cmd;
+ struct ice_aq_desc desc;
+ enum ice_status status;
+
+ ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_set_phy_rec_clk_out);
+ cmd = &desc.params.set_phy_rec_clk_out;
+ cmd->phy_output = phy_output;
+ cmd->port_num = ICE_AQC_SET_PHY_REC_CLK_OUT_CURR_PORT;
+ cmd->flags = enable & ICE_AQC_SET_PHY_REC_CLK_OUT_OUT_EN;
+ cmd->freq = cpu_to_le32(*freq);
+
+ status = ice_aq_send_cmd(hw, &desc, NULL, 0, NULL);
+ if (!status)
+ *freq = le32_to_cpu(cmd->freq);
+
+ return status;
+}
+
+/**
+ * ice_aq_get_phy_rec_clk_out
+ * @hw: pointer to the HW struct
+ * @phy_output: PHY reference clock output pin
+ * @port_num: Port number
+ * @flags: PHY flags
+ * @freq: PHY output frequency
+ *
+ * Get PHY recovered clock output (0x0631)
+ */
+enum ice_status
+ice_aq_get_phy_rec_clk_out(struct ice_hw *hw, u8 phy_output, u8 *port_num,
+ u8 *flags, u32 *freq)
+{
+ struct ice_aqc_get_phy_rec_clk_out *cmd;
+ struct ice_aq_desc desc;
+ enum ice_status status;
+
+ ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_get_phy_rec_clk_out);
+ cmd = &desc.params.get_phy_rec_clk_out;
+ cmd->phy_output = phy_output;
+ cmd->port_num = *port_num;
+
+ status = ice_aq_send_cmd(hw, &desc, NULL, 0, NULL);
+ if (!status) {
+ *port_num = cmd->port_num;
+ *flags = cmd->flags;
+ *freq = le32_to_cpu(cmd->freq);
+ }
+
+ return status;
+}
+
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index aaed388a40a8..8a99c8364173 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -166,6 +166,12 @@ ice_ena_vsi_rdma_qset(struct ice_port_info *pi, u16 vsi_handle, u8 tc,
enum ice_status
ice_aq_get_cgu_dpll_status(struct ice_hw *hw, u8 dpll_num, u8 *ref_state,
u16 *dpll_state, u64 *phase_offset, u8 *eec_mode);
+enum ice_status
+ice_aq_set_phy_rec_clk_out(struct ice_hw *hw, u8 phy_output, bool enable,
+ u32 *freq);
+enum ice_status
+ice_aq_get_phy_rec_clk_out(struct ice_hw *hw, u8 phy_output, u8 *port_num,
+ u8 *flags, u32 *freq);
int
ice_dis_vsi_rdma_qset(struct ice_port_info *pi, u16 count, u32 *qset_teid,
u16 *q_id);
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 5af2faaa21e1..2ae58f23c845 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -4076,6 +4076,99 @@ ice_get_module_eeprom(struct net_device *netdev,
return 0;
}
+/**
+ * ice_get_rclk_range - get range of recovered clock indices
+ * @netdev: network interface device structure
+ * @min_idx: min rclk index
+ * @max_idx: max rclk index
+ * @extack: netlink extended ack
+ */
+static int
+ice_get_rclk_range(struct net_device *netdev, u32 *min_idx, u32 *max_idx,
+ struct netlink_ext_ack *extack)
+{
+ struct ice_netdev_priv *np = netdev_priv(netdev);
+ struct ice_vsi *vsi = np->vsi;
+ struct ice_pf *pf = vsi->back;
+
+ if (!ice_is_feature_supported(pf, ICE_F_CGU))
+ return -EOPNOTSUPP;
+
+ *min_idx = 0;
+ *max_idx = ICE_RCLK_PIN_MAX;
+
+ return 0;
+}
+
+/**
+ * ice_get_rclk_state - get state of a recovered frequency output pin
+ * @netdev: network interface device structure
+ * @out_idx: index of a questioned pin
+ * @ena: returned state of a pin
+ * @extack: netlink extended ack
+ */
+static int
+ice_get_rclk_state(struct net_device *netdev, u32 out_idx,
+ bool *ena, struct netlink_ext_ack *extack)
+{
+ u8 port_num = ICE_AQC_SET_PHY_REC_CLK_OUT_CURR_PORT, flags;
+ struct ice_netdev_priv *np = netdev_priv(netdev);
+ struct ice_vsi *vsi = np->vsi;
+ struct ice_pf *pf = vsi->back;
+ u32 freq;
+ int ret;
+
+ if (!ice_is_feature_supported(pf, ICE_F_CGU))
+ return -EOPNOTSUPP;
+
+ if (out_idx > ICE_RCLK_PIN_MAX)
+ return -EINVAL;
+
+ ret = ice_aq_get_phy_rec_clk_out(&pf->hw, out_idx,
+ &port_num, &flags, &freq);
+ if (ret)
+ return ret;
+
+ if (flags & ICE_AQC_GET_PHY_REC_CLK_OUT_OUT_EN)
+ *ena = true;
+ else
+ *ena = false;
+
+ return ret;
+}
+
+/**
+ * ice_set_rclk_out - enable/disable recovered clock redirection to the
+ * output pin
+ * @netdev: network interface device structure
+ * @out_idx: index of pin being configured
+ * @ena: requested state of a pin
+ * @extack: netlink extended ack
+ */
+static int
+ice_set_rclk_out(struct net_device *netdev, u32 out_idx, bool ena,
+ struct netlink_ext_ack *extack)
+{
+ struct ice_netdev_priv *np = netdev_priv(netdev);
+ struct ice_vsi *vsi = np->vsi;
+ struct ice_pf *pf = vsi->back;
+ enum ice_status ret;
+ u32 freq;
+
+ if (!ice_is_feature_supported(pf, ICE_F_CGU))
+ return -EOPNOTSUPP;
+
+ if (out_idx > ICE_RCLK_PIN_MAX)
+ return -EINVAL;
+
+ ret = ice_aq_set_phy_rec_clk_out(&pf->hw, out_idx,
+ ena, &freq);
+ if (ret)
+ return ret;
+
+ return ret;
+}
+
static const struct ethtool_ops ice_ethtool_ops = {
.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
ETHTOOL_COALESCE_USE_ADAPTIVE |
@@ -4121,6 +4214,9 @@ static const struct ethtool_ops ice_ethtool_ops = {
.set_fecparam = ice_set_fecparam,
.get_module_info = ice_get_module_info,
.get_module_eeprom = ice_get_module_eeprom,
+ .get_rclk_range = ice_get_rclk_range,
+ .get_rclk_state = ice_get_rclk_state,
+ .set_rclk_out = ice_set_rclk_out,
};
static const struct ethtool_ops ice_ethtool_safe_mode_ops = {
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
index 28b04ec40bae..865ca680b62e 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
@@ -132,4 +132,6 @@ enum ice_e810t_cgu_pins {
NUM_E810T_CGU_PINS
};
+#define ICE_RCLK_PIN_MAX (REF1N - REF1P)
+
#endif /* _ICE_PTP_HW_H_ */
--
2.26.3
^ permalink raw reply related
* Re: [PATCH V6 3/5] hyper-v: Enable swiotlb bounce buffer for Isolation VM
From: Tianyu Lan @ 2021-12-10 14:01 UTC (permalink / raw)
To: Michael Kelley (LINUX), KY Srinivasan, Haiyang Zhang,
Stephen Hemminger, wei.liu@kernel.org, Dexuan Cui,
tglx@linutronix.de, mingo@redhat.com, bp@alien8.de,
dave.hansen@linux.intel.com, x86@kernel.org, hpa@zytor.com,
davem@davemloft.net, kuba@kernel.org, jejb@linux.ibm.com,
martin.petersen@oracle.com, arnd@arndb.de, hch@infradead.org,
m.szyprowski@samsung.com, robin.murphy@arm.com, Tianyu Lan,
thomas.lendacky@amd.com
Cc: iommu@lists.linux-foundation.org, linux-arch@vger.kernel.org,
linux-hyperv@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-scsi@vger.kernel.org, netdev@vger.kernel.org, vkuznets,
brijesh.singh@amd.com, konrad.wilk@oracle.com, hch@lst.de,
joro@8bytes.org, parri.andrea@gmail.com, dave.hansen@intel.com
In-Reply-To: <4d60fcd1-97df-f4a1-1b79-643e65f66b3e@gmail.com>
On 12/10/2021 9:25 PM, Tianyu Lan wrote:
>>> @@ -319,8 +320,16 @@ static void __init ms_hyperv_init_platform(void)
>>> pr_info("Hyper-V: Isolation Config: Group A 0x%x, Group B
>>> 0x%x\n",
>>> ms_hyperv.isolation_config_a,
>>> ms_hyperv.isolation_config_b);
>>>
>>> - if (hv_get_isolation_type() == HV_ISOLATION_TYPE_SNP)
>>> + if (hv_get_isolation_type() == HV_ISOLATION_TYPE_SNP) {
>>> static_branch_enable(&isolation_type_snp);
>>> + swiotlb_unencrypted_base = ms_hyperv.shared_gpa_boundary;
>>> + }
>>> +
>>> + /*
>>> + * Enable swiotlb force mode in Isolation VM to
>>> + * use swiotlb bounce buffer for dma transaction.
>>> + */
>>> + swiotlb_force = SWIOTLB_FORCE;
>> I'm good with this approach that directly updates the swiotlb settings
>> here
>>
>> rather than in IOMMU initialization code. It's a lot more
>> straightforward.
>>
>> However, there's an issue if building for X86_32 without PAE, in that the
>> swiotlb module may not be built, resulting in compile and link
>> errors. The
>> swiotlb.h file needs to be updated to provide a stub function for
>> swiotlb_update_mem_attributes(). swiotlb_unencrypted_base probably
>> needs wrapper functions to get/set it, which can be stubs when
>> CONFIG_SWIOTLB is not set. swiotlb_force is a bit of a mess in that
>> it already
>> has a stub definition that assumes it will only be read, and not set.
>> A bit of
>> thinking will be needed to sort that out.
>
> It's ok to fix the issue via selecting swiotlb when CONFIG_HYPERV is
> set?
>
Sorry. ignore the previous statement. These codes doesn't depend on
CONFIG_HYPERV.
How about making these code under #ifdef CONFIG_X86_64 or CONFIG_SWIOTLB?
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox