* [PATCH v4 bpf-next 02/14] libbpf: convert libbpf code to use new btf helpers
From: Andrii Nakryiko @ 2019-08-07 5:37 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Simplify code by relying on newly added BTF helper functions.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
---
tools/lib/bpf/btf.c | 181 ++++++++++++++++++---------------------
tools/lib/bpf/btf_dump.c | 136 ++++++++++-------------------
tools/lib/bpf/libbpf.c | 60 +++++++------
3 files changed, 157 insertions(+), 220 deletions(-)
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 467224feb43b..f49a54c2575f 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -19,13 +19,6 @@
#define BTF_MAX_NR_TYPES 0x7fffffff
#define BTF_MAX_STR_OFFSET 0x7fffffff
-#define IS_MODIFIER(k) (((k) == BTF_KIND_TYPEDEF) || \
- ((k) == BTF_KIND_VOLATILE) || \
- ((k) == BTF_KIND_CONST) || \
- ((k) == BTF_KIND_RESTRICT))
-
-#define IS_VAR(k) ((k) == BTF_KIND_VAR)
-
static struct btf_type btf_void;
struct btf {
@@ -192,9 +185,9 @@ static int btf_parse_str_sec(struct btf *btf)
static int btf_type_size(struct btf_type *t)
{
int base_size = sizeof(struct btf_type);
- __u16 vlen = BTF_INFO_VLEN(t->info);
+ __u16 vlen = btf_vlen(t);
- switch (BTF_INFO_KIND(t->info)) {
+ switch (btf_kind(t)) {
case BTF_KIND_FWD:
case BTF_KIND_CONST:
case BTF_KIND_VOLATILE:
@@ -219,7 +212,7 @@ static int btf_type_size(struct btf_type *t)
case BTF_KIND_DATASEC:
return base_size + vlen * sizeof(struct btf_var_secinfo);
default:
- pr_debug("Unsupported BTF_KIND:%u\n", BTF_INFO_KIND(t->info));
+ pr_debug("Unsupported BTF_KIND:%u\n", btf_kind(t));
return -EINVAL;
}
}
@@ -263,7 +256,7 @@ const struct btf_type *btf__type_by_id(const struct btf *btf, __u32 type_id)
static bool btf_type_is_void(const struct btf_type *t)
{
- return t == &btf_void || BTF_INFO_KIND(t->info) == BTF_KIND_FWD;
+ return t == &btf_void || btf_is_fwd(t);
}
static bool btf_type_is_void_or_null(const struct btf_type *t)
@@ -284,7 +277,7 @@ __s64 btf__resolve_size(const struct btf *btf, __u32 type_id)
t = btf__type_by_id(btf, type_id);
for (i = 0; i < MAX_RESOLVE_DEPTH && !btf_type_is_void_or_null(t);
i++) {
- switch (BTF_INFO_KIND(t->info)) {
+ switch (btf_kind(t)) {
case BTF_KIND_INT:
case BTF_KIND_STRUCT:
case BTF_KIND_UNION:
@@ -303,7 +296,7 @@ __s64 btf__resolve_size(const struct btf *btf, __u32 type_id)
type_id = t->type;
break;
case BTF_KIND_ARRAY:
- array = (const struct btf_array *)(t + 1);
+ array = btf_array(t);
if (nelems && array->nelems > UINT32_MAX / nelems)
return -E2BIG;
nelems *= array->nelems;
@@ -334,8 +327,7 @@ int btf__resolve_type(const struct btf *btf, __u32 type_id)
t = btf__type_by_id(btf, type_id);
while (depth < MAX_RESOLVE_DEPTH &&
!btf_type_is_void_or_null(t) &&
- (IS_MODIFIER(BTF_INFO_KIND(t->info)) ||
- IS_VAR(BTF_INFO_KIND(t->info)))) {
+ (btf_is_mod(t) || btf_is_typedef(t) || btf_is_var(t))) {
type_id = t->type;
t = btf__type_by_id(btf, type_id);
depth++;
@@ -554,11 +546,11 @@ static int compare_vsi_off(const void *_a, const void *_b)
static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
struct btf_type *t)
{
- __u32 size = 0, off = 0, i, vars = BTF_INFO_VLEN(t->info);
+ __u32 size = 0, off = 0, i, vars = btf_vlen(t);
const char *name = btf__name_by_offset(btf, t->name_off);
const struct btf_type *t_var;
struct btf_var_secinfo *vsi;
- struct btf_var *var;
+ const struct btf_var *var;
int ret;
if (!name) {
@@ -574,12 +566,11 @@ static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
t->size = size;
- for (i = 0, vsi = (struct btf_var_secinfo *)(t + 1);
- i < vars; i++, vsi++) {
+ for (i = 0, vsi = (void *)btf_var_secinfos(t); i < vars; i++, vsi++) {
t_var = btf__type_by_id(btf, vsi->type);
- var = (struct btf_var *)(t_var + 1);
+ var = btf_var(t_var);
- if (BTF_INFO_KIND(t_var->info) != BTF_KIND_VAR) {
+ if (!btf_is_var(t_var)) {
pr_debug("Non-VAR type seen in section %s\n", name);
return -EINVAL;
}
@@ -595,7 +586,8 @@ static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
ret = bpf_object__variable_offset(obj, name, &off);
if (ret) {
- pr_debug("No offset found in symbol table for VAR %s\n", name);
+ pr_debug("No offset found in symbol table for VAR %s\n",
+ name);
return -ENOENT;
}
@@ -619,7 +611,7 @@ int btf__finalize_data(struct bpf_object *obj, struct btf *btf)
* is section size and global variable offset. We use
* the info from the ELF itself for this purpose.
*/
- if (BTF_INFO_KIND(t->info) == BTF_KIND_DATASEC) {
+ if (btf_is_datasec(t)) {
err = btf_fixup_datasec(obj, btf, t);
if (err)
break;
@@ -774,14 +766,13 @@ int btf__get_map_kv_tids(const struct btf *btf, const char *map_name,
return -EINVAL;
}
- if (BTF_INFO_KIND(container_type->info) != BTF_KIND_STRUCT ||
- BTF_INFO_VLEN(container_type->info) < 2) {
+ if (!btf_is_struct(container_type) || btf_vlen(container_type) < 2) {
pr_warning("map:%s container_name:%s is an invalid container struct\n",
map_name, container_name);
return -EINVAL;
}
- key = (struct btf_member *)(container_type + 1);
+ key = btf_members(container_type);
value = key + 1;
key_size = btf__resolve_size(btf, key->type);
@@ -1440,10 +1431,9 @@ static struct btf_dedup *btf_dedup_new(struct btf *btf, struct btf_ext *btf_ext,
d->map[0] = 0;
for (i = 1; i <= btf->nr_types; i++) {
struct btf_type *t = d->btf->types[i];
- __u16 kind = BTF_INFO_KIND(t->info);
/* VAR and DATASEC are never deduped and are self-canonical */
- if (kind == BTF_KIND_VAR || kind == BTF_KIND_DATASEC)
+ if (btf_is_var(t) || btf_is_datasec(t))
d->map[i] = i;
else
d->map[i] = BTF_UNPROCESSED_ID;
@@ -1484,11 +1474,11 @@ static int btf_for_each_str_off(struct btf_dedup *d, str_off_fn_t fn, void *ctx)
if (r)
return r;
- switch (BTF_INFO_KIND(t->info)) {
+ switch (btf_kind(t)) {
case BTF_KIND_STRUCT:
case BTF_KIND_UNION: {
- struct btf_member *m = (struct btf_member *)(t + 1);
- __u16 vlen = BTF_INFO_VLEN(t->info);
+ struct btf_member *m = (void *)btf_members(t);
+ __u16 vlen = btf_vlen(t);
for (j = 0; j < vlen; j++) {
r = fn(&m->name_off, ctx);
@@ -1499,8 +1489,8 @@ static int btf_for_each_str_off(struct btf_dedup *d, str_off_fn_t fn, void *ctx)
break;
}
case BTF_KIND_ENUM: {
- struct btf_enum *m = (struct btf_enum *)(t + 1);
- __u16 vlen = BTF_INFO_VLEN(t->info);
+ struct btf_enum *m = (void *)btf_enum(t);
+ __u16 vlen = btf_vlen(t);
for (j = 0; j < vlen; j++) {
r = fn(&m->name_off, ctx);
@@ -1511,8 +1501,8 @@ static int btf_for_each_str_off(struct btf_dedup *d, str_off_fn_t fn, void *ctx)
break;
}
case BTF_KIND_FUNC_PROTO: {
- struct btf_param *m = (struct btf_param *)(t + 1);
- __u16 vlen = BTF_INFO_VLEN(t->info);
+ struct btf_param *m = (void *)btf_params(t);
+ __u16 vlen = btf_vlen(t);
for (j = 0; j < vlen; j++) {
r = fn(&m->name_off, ctx);
@@ -1801,16 +1791,16 @@ static long btf_hash_enum(struct btf_type *t)
/* Check structural equality of two ENUMs. */
static bool btf_equal_enum(struct btf_type *t1, struct btf_type *t2)
{
- struct btf_enum *m1, *m2;
+ const struct btf_enum *m1, *m2;
__u16 vlen;
int i;
if (!btf_equal_common(t1, t2))
return false;
- vlen = BTF_INFO_VLEN(t1->info);
- m1 = (struct btf_enum *)(t1 + 1);
- m2 = (struct btf_enum *)(t2 + 1);
+ vlen = btf_vlen(t1);
+ m1 = btf_enum(t1);
+ m2 = btf_enum(t2);
for (i = 0; i < vlen; i++) {
if (m1->name_off != m2->name_off || m1->val != m2->val)
return false;
@@ -1822,8 +1812,7 @@ static bool btf_equal_enum(struct btf_type *t1, struct btf_type *t2)
static inline bool btf_is_enum_fwd(struct btf_type *t)
{
- return BTF_INFO_KIND(t->info) == BTF_KIND_ENUM &&
- BTF_INFO_VLEN(t->info) == 0;
+ return btf_is_enum(t) && btf_vlen(t) == 0;
}
static bool btf_compat_enum(struct btf_type *t1, struct btf_type *t2)
@@ -1843,8 +1832,8 @@ static bool btf_compat_enum(struct btf_type *t1, struct btf_type *t2)
*/
static long btf_hash_struct(struct btf_type *t)
{
- struct btf_member *member = (struct btf_member *)(t + 1);
- __u32 vlen = BTF_INFO_VLEN(t->info);
+ const struct btf_member *member = btf_members(t);
+ __u32 vlen = btf_vlen(t);
long h = btf_hash_common(t);
int i;
@@ -1864,16 +1853,16 @@ static long btf_hash_struct(struct btf_type *t)
*/
static bool btf_shallow_equal_struct(struct btf_type *t1, struct btf_type *t2)
{
- struct btf_member *m1, *m2;
+ const struct btf_member *m1, *m2;
__u16 vlen;
int i;
if (!btf_equal_common(t1, t2))
return false;
- vlen = BTF_INFO_VLEN(t1->info);
- m1 = (struct btf_member *)(t1 + 1);
- m2 = (struct btf_member *)(t2 + 1);
+ vlen = btf_vlen(t1);
+ m1 = btf_members(t1);
+ m2 = btf_members(t2);
for (i = 0; i < vlen; i++) {
if (m1->name_off != m2->name_off || m1->offset != m2->offset)
return false;
@@ -1890,7 +1879,7 @@ static bool btf_shallow_equal_struct(struct btf_type *t1, struct btf_type *t2)
*/
static long btf_hash_array(struct btf_type *t)
{
- struct btf_array *info = (struct btf_array *)(t + 1);
+ const struct btf_array *info = btf_array(t);
long h = btf_hash_common(t);
h = hash_combine(h, info->type);
@@ -1908,13 +1897,13 @@ static long btf_hash_array(struct btf_type *t)
*/
static bool btf_equal_array(struct btf_type *t1, struct btf_type *t2)
{
- struct btf_array *info1, *info2;
+ const struct btf_array *info1, *info2;
if (!btf_equal_common(t1, t2))
return false;
- info1 = (struct btf_array *)(t1 + 1);
- info2 = (struct btf_array *)(t2 + 1);
+ info1 = btf_array(t1);
+ info2 = btf_array(t2);
return info1->type == info2->type &&
info1->index_type == info2->index_type &&
info1->nelems == info2->nelems;
@@ -1927,14 +1916,10 @@ static bool btf_equal_array(struct btf_type *t1, struct btf_type *t2)
*/
static bool btf_compat_array(struct btf_type *t1, struct btf_type *t2)
{
- struct btf_array *info1, *info2;
-
if (!btf_equal_common(t1, t2))
return false;
- info1 = (struct btf_array *)(t1 + 1);
- info2 = (struct btf_array *)(t2 + 1);
- return info1->nelems == info2->nelems;
+ return btf_array(t1)->nelems == btf_array(t2)->nelems;
}
/*
@@ -1944,8 +1929,8 @@ static bool btf_compat_array(struct btf_type *t1, struct btf_type *t2)
*/
static long btf_hash_fnproto(struct btf_type *t)
{
- struct btf_param *member = (struct btf_param *)(t + 1);
- __u16 vlen = BTF_INFO_VLEN(t->info);
+ const struct btf_param *member = btf_params(t);
+ __u16 vlen = btf_vlen(t);
long h = btf_hash_common(t);
int i;
@@ -1966,16 +1951,16 @@ static long btf_hash_fnproto(struct btf_type *t)
*/
static bool btf_equal_fnproto(struct btf_type *t1, struct btf_type *t2)
{
- struct btf_param *m1, *m2;
+ const struct btf_param *m1, *m2;
__u16 vlen;
int i;
if (!btf_equal_common(t1, t2))
return false;
- vlen = BTF_INFO_VLEN(t1->info);
- m1 = (struct btf_param *)(t1 + 1);
- m2 = (struct btf_param *)(t2 + 1);
+ vlen = btf_vlen(t1);
+ m1 = btf_params(t1);
+ m2 = btf_params(t2);
for (i = 0; i < vlen; i++) {
if (m1->name_off != m2->name_off || m1->type != m2->type)
return false;
@@ -1992,7 +1977,7 @@ static bool btf_equal_fnproto(struct btf_type *t1, struct btf_type *t2)
*/
static bool btf_compat_fnproto(struct btf_type *t1, struct btf_type *t2)
{
- struct btf_param *m1, *m2;
+ const struct btf_param *m1, *m2;
__u16 vlen;
int i;
@@ -2000,9 +1985,9 @@ static bool btf_compat_fnproto(struct btf_type *t1, struct btf_type *t2)
if (t1->name_off != t2->name_off || t1->info != t2->info)
return false;
- vlen = BTF_INFO_VLEN(t1->info);
- m1 = (struct btf_param *)(t1 + 1);
- m2 = (struct btf_param *)(t2 + 1);
+ vlen = btf_vlen(t1);
+ m1 = btf_params(t1);
+ m2 = btf_params(t2);
for (i = 0; i < vlen; i++) {
if (m1->name_off != m2->name_off)
return false;
@@ -2028,7 +2013,7 @@ static int btf_dedup_prim_type(struct btf_dedup *d, __u32 type_id)
__u32 cand_id;
long h;
- switch (BTF_INFO_KIND(t->info)) {
+ switch (btf_kind(t)) {
case BTF_KIND_CONST:
case BTF_KIND_VOLATILE:
case BTF_KIND_RESTRICT:
@@ -2141,13 +2126,13 @@ static uint32_t resolve_fwd_id(struct btf_dedup *d, uint32_t type_id)
{
__u32 orig_type_id = type_id;
- if (BTF_INFO_KIND(d->btf->types[type_id]->info) != BTF_KIND_FWD)
+ if (!btf_is_fwd(d->btf->types[type_id]))
return type_id;
while (is_type_mapped(d, type_id) && d->map[type_id] != type_id)
type_id = d->map[type_id];
- if (BTF_INFO_KIND(d->btf->types[type_id]->info) != BTF_KIND_FWD)
+ if (!btf_is_fwd(d->btf->types[type_id]))
return type_id;
return orig_type_id;
@@ -2156,7 +2141,7 @@ static uint32_t resolve_fwd_id(struct btf_dedup *d, uint32_t type_id)
static inline __u16 btf_fwd_kind(struct btf_type *t)
{
- return BTF_INFO_KFLAG(t->info) ? BTF_KIND_UNION : BTF_KIND_STRUCT;
+ return btf_kflag(t) ? BTF_KIND_UNION : BTF_KIND_STRUCT;
}
/*
@@ -2277,8 +2262,8 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
cand_type = d->btf->types[cand_id];
canon_type = d->btf->types[canon_id];
- cand_kind = BTF_INFO_KIND(cand_type->info);
- canon_kind = BTF_INFO_KIND(canon_type->info);
+ cand_kind = btf_kind(cand_type);
+ canon_kind = btf_kind(canon_type);
if (cand_type->name_off != canon_type->name_off)
return 0;
@@ -2327,12 +2312,12 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
return btf_dedup_is_equiv(d, cand_type->type, canon_type->type);
case BTF_KIND_ARRAY: {
- struct btf_array *cand_arr, *canon_arr;
+ const struct btf_array *cand_arr, *canon_arr;
if (!btf_compat_array(cand_type, canon_type))
return 0;
- cand_arr = (struct btf_array *)(cand_type + 1);
- canon_arr = (struct btf_array *)(canon_type + 1);
+ cand_arr = btf_array(cand_type);
+ canon_arr = btf_array(canon_type);
eq = btf_dedup_is_equiv(d,
cand_arr->index_type, canon_arr->index_type);
if (eq <= 0)
@@ -2342,14 +2327,14 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
case BTF_KIND_STRUCT:
case BTF_KIND_UNION: {
- struct btf_member *cand_m, *canon_m;
+ const struct btf_member *cand_m, *canon_m;
__u16 vlen;
if (!btf_shallow_equal_struct(cand_type, canon_type))
return 0;
- vlen = BTF_INFO_VLEN(cand_type->info);
- cand_m = (struct btf_member *)(cand_type + 1);
- canon_m = (struct btf_member *)(canon_type + 1);
+ vlen = btf_vlen(cand_type);
+ cand_m = btf_members(cand_type);
+ canon_m = btf_members(canon_type);
for (i = 0; i < vlen; i++) {
eq = btf_dedup_is_equiv(d, cand_m->type, canon_m->type);
if (eq <= 0)
@@ -2362,7 +2347,7 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
}
case BTF_KIND_FUNC_PROTO: {
- struct btf_param *cand_p, *canon_p;
+ const struct btf_param *cand_p, *canon_p;
__u16 vlen;
if (!btf_compat_fnproto(cand_type, canon_type))
@@ -2370,9 +2355,9 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
eq = btf_dedup_is_equiv(d, cand_type->type, canon_type->type);
if (eq <= 0)
return eq;
- vlen = BTF_INFO_VLEN(cand_type->info);
- cand_p = (struct btf_param *)(cand_type + 1);
- canon_p = (struct btf_param *)(canon_type + 1);
+ vlen = btf_vlen(cand_type);
+ cand_p = btf_params(cand_type);
+ canon_p = btf_params(canon_type);
for (i = 0; i < vlen; i++) {
eq = btf_dedup_is_equiv(d, cand_p->type, canon_p->type);
if (eq <= 0)
@@ -2427,8 +2412,8 @@ static void btf_dedup_merge_hypot_map(struct btf_dedup *d)
targ_type_id = d->hypot_map[cand_type_id];
t_id = resolve_type_id(d, targ_type_id);
c_id = resolve_type_id(d, cand_type_id);
- t_kind = BTF_INFO_KIND(d->btf->types[t_id]->info);
- c_kind = BTF_INFO_KIND(d->btf->types[c_id]->info);
+ t_kind = btf_kind(d->btf->types[t_id]);
+ c_kind = btf_kind(d->btf->types[c_id]);
/*
* Resolve FWD into STRUCT/UNION.
* It's ok to resolve FWD into STRUCT/UNION that's not yet
@@ -2497,7 +2482,7 @@ static int btf_dedup_struct_type(struct btf_dedup *d, __u32 type_id)
return 0;
t = d->btf->types[type_id];
- kind = BTF_INFO_KIND(t->info);
+ kind = btf_kind(t);
if (kind != BTF_KIND_STRUCT && kind != BTF_KIND_UNION)
return 0;
@@ -2592,7 +2577,7 @@ static int btf_dedup_ref_type(struct btf_dedup *d, __u32 type_id)
t = d->btf->types[type_id];
d->map[type_id] = BTF_IN_PROGRESS_ID;
- switch (BTF_INFO_KIND(t->info)) {
+ switch (btf_kind(t)) {
case BTF_KIND_CONST:
case BTF_KIND_VOLATILE:
case BTF_KIND_RESTRICT:
@@ -2616,7 +2601,7 @@ static int btf_dedup_ref_type(struct btf_dedup *d, __u32 type_id)
break;
case BTF_KIND_ARRAY: {
- struct btf_array *info = (struct btf_array *)(t + 1);
+ struct btf_array *info = (void *)btf_array(t);
ref_type_id = btf_dedup_ref_type(d, info->type);
if (ref_type_id < 0)
@@ -2650,8 +2635,8 @@ static int btf_dedup_ref_type(struct btf_dedup *d, __u32 type_id)
return ref_type_id;
t->type = ref_type_id;
- vlen = BTF_INFO_VLEN(t->info);
- param = (struct btf_param *)(t + 1);
+ vlen = btf_vlen(t);
+ param = (void *)btf_params(t);
for (i = 0; i < vlen; i++) {
ref_type_id = btf_dedup_ref_type(d, param->type);
if (ref_type_id < 0)
@@ -2791,7 +2776,7 @@ static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
struct btf_type *t = d->btf->types[type_id];
int i, r;
- switch (BTF_INFO_KIND(t->info)) {
+ switch (btf_kind(t)) {
case BTF_KIND_INT:
case BTF_KIND_ENUM:
break;
@@ -2811,7 +2796,7 @@ static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
break;
case BTF_KIND_ARRAY: {
- struct btf_array *arr_info = (struct btf_array *)(t + 1);
+ struct btf_array *arr_info = (void *)btf_array(t);
r = btf_dedup_remap_type_id(d, arr_info->type);
if (r < 0)
@@ -2826,8 +2811,8 @@ static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
case BTF_KIND_STRUCT:
case BTF_KIND_UNION: {
- struct btf_member *member = (struct btf_member *)(t + 1);
- __u16 vlen = BTF_INFO_VLEN(t->info);
+ struct btf_member *member = (void *)btf_members(t);
+ __u16 vlen = btf_vlen(t);
for (i = 0; i < vlen; i++) {
r = btf_dedup_remap_type_id(d, member->type);
@@ -2840,8 +2825,8 @@ static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
}
case BTF_KIND_FUNC_PROTO: {
- struct btf_param *param = (struct btf_param *)(t + 1);
- __u16 vlen = BTF_INFO_VLEN(t->info);
+ struct btf_param *param = (void *)btf_params(t);
+ __u16 vlen = btf_vlen(t);
r = btf_dedup_remap_type_id(d, t->type);
if (r < 0)
@@ -2859,8 +2844,8 @@ static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
}
case BTF_KIND_DATASEC: {
- struct btf_var_secinfo *var = (struct btf_var_secinfo *)(t + 1);
- __u16 vlen = BTF_INFO_VLEN(t->info);
+ struct btf_var_secinfo *var = (void *)btf_var_secinfos(t);
+ __u16 vlen = btf_vlen(t);
for (i = 0; i < vlen; i++) {
r = btf_dedup_remap_type_id(d, var->type);
diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c
index 7065bb5b2752..e8987bfa4d0e 100644
--- a/tools/lib/bpf/btf_dump.c
+++ b/tools/lib/bpf/btf_dump.c
@@ -100,21 +100,6 @@ static bool str_equal_fn(const void *a, const void *b, void *ctx)
return strcmp(a, b) == 0;
}
-static __u16 btf_kind_of(const struct btf_type *t)
-{
- return BTF_INFO_KIND(t->info);
-}
-
-static __u16 btf_vlen_of(const struct btf_type *t)
-{
- return BTF_INFO_VLEN(t->info);
-}
-
-static bool btf_kflag_of(const struct btf_type *t)
-{
- return BTF_INFO_KFLAG(t->info);
-}
-
static const char *btf_name_of(const struct btf_dump *d, __u32 name_off)
{
return btf__name_by_offset(d->btf, name_off);
@@ -349,7 +334,7 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
*/
struct btf_dump_type_aux_state *tstate = &d->type_states[id];
const struct btf_type *t;
- __u16 kind, vlen;
+ __u16 vlen;
int err, i;
/* return true, letting typedefs know that it's ok to be emitted */
@@ -357,18 +342,16 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
return 1;
t = btf__type_by_id(d->btf, id);
- kind = btf_kind_of(t);
if (tstate->order_state == ORDERING) {
/* type loop, but resolvable through fwd declaration */
- if ((kind == BTF_KIND_STRUCT || kind == BTF_KIND_UNION) &&
- through_ptr && t->name_off != 0)
+ if (btf_is_composite(t) && through_ptr && t->name_off != 0)
return 0;
pr_warning("unsatisfiable type cycle, id:[%u]\n", id);
return -ELOOP;
}
- switch (kind) {
+ switch (btf_kind(t)) {
case BTF_KIND_INT:
tstate->order_state = ORDERED;
return 0;
@@ -378,14 +361,12 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
tstate->order_state = ORDERED;
return err;
- case BTF_KIND_ARRAY: {
- const struct btf_array *a = (void *)(t + 1);
+ case BTF_KIND_ARRAY:
+ return btf_dump_order_type(d, btf_array(t)->type, through_ptr);
- return btf_dump_order_type(d, a->type, through_ptr);
- }
case BTF_KIND_STRUCT:
case BTF_KIND_UNION: {
- const struct btf_member *m = (void *)(t + 1);
+ const struct btf_member *m = btf_members(t);
/*
* struct/union is part of strong link, only if it's embedded
* (so no ptr in a path) or it's anonymous (so has to be
@@ -396,7 +377,7 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
tstate->order_state = ORDERING;
- vlen = btf_vlen_of(t);
+ vlen = btf_vlen(t);
for (i = 0; i < vlen; i++, m++) {
err = btf_dump_order_type(d, m->type, false);
if (err < 0)
@@ -447,7 +428,7 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
return btf_dump_order_type(d, t->type, through_ptr);
case BTF_KIND_FUNC_PROTO: {
- const struct btf_param *p = (void *)(t + 1);
+ const struct btf_param *p = btf_params(t);
bool is_strong;
err = btf_dump_order_type(d, t->type, through_ptr);
@@ -455,7 +436,7 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
return err;
is_strong = err > 0;
- vlen = btf_vlen_of(t);
+ vlen = btf_vlen(t);
for (i = 0; i < vlen; i++, p++) {
err = btf_dump_order_type(d, p->type, through_ptr);
if (err < 0)
@@ -553,7 +534,7 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
return;
t = btf__type_by_id(d->btf, id);
- kind = btf_kind_of(t);
+ kind = btf_kind(t);
if (top_level_def && t->name_off == 0) {
pr_warning("unexpected nameless definition, id:[%u]\n", id);
@@ -618,12 +599,9 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
case BTF_KIND_RESTRICT:
btf_dump_emit_type(d, t->type, cont_id);
break;
- case BTF_KIND_ARRAY: {
- const struct btf_array *a = (void *)(t + 1);
-
- btf_dump_emit_type(d, a->type, cont_id);
+ case BTF_KIND_ARRAY:
+ btf_dump_emit_type(d, btf_array(t)->type, cont_id);
break;
- }
case BTF_KIND_FWD:
btf_dump_emit_fwd_def(d, id, t);
btf_dump_printf(d, ";\n\n");
@@ -656,8 +634,8 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
* applicable
*/
if (top_level_def || t->name_off == 0) {
- const struct btf_member *m = (void *)(t + 1);
- __u16 vlen = btf_vlen_of(t);
+ const struct btf_member *m = btf_members(t);
+ __u16 vlen = btf_vlen(t);
int i, new_cont_id;
new_cont_id = t->name_off == 0 ? cont_id : id;
@@ -678,8 +656,8 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
}
break;
case BTF_KIND_FUNC_PROTO: {
- const struct btf_param *p = (void *)(t + 1);
- __u16 vlen = btf_vlen_of(t);
+ const struct btf_param *p = btf_params(t);
+ __u16 vlen = btf_vlen(t);
int i;
btf_dump_emit_type(d, t->type, cont_id);
@@ -696,7 +674,7 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
static int btf_align_of(const struct btf *btf, __u32 id)
{
const struct btf_type *t = btf__type_by_id(btf, id);
- __u16 kind = btf_kind_of(t);
+ __u16 kind = btf_kind(t);
switch (kind) {
case BTF_KIND_INT:
@@ -709,15 +687,12 @@ static int btf_align_of(const struct btf *btf, __u32 id)
case BTF_KIND_CONST:
case BTF_KIND_RESTRICT:
return btf_align_of(btf, t->type);
- case BTF_KIND_ARRAY: {
- const struct btf_array *a = (void *)(t + 1);
-
- return btf_align_of(btf, a->type);
- }
+ case BTF_KIND_ARRAY:
+ return btf_align_of(btf, btf_array(t)->type);
case BTF_KIND_STRUCT:
case BTF_KIND_UNION: {
- const struct btf_member *m = (void *)(t + 1);
- __u16 vlen = btf_vlen_of(t);
+ const struct btf_member *m = btf_members(t);
+ __u16 vlen = btf_vlen(t);
int i, align = 1;
for (i = 0; i < vlen; i++, m++)
@@ -726,7 +701,7 @@ static int btf_align_of(const struct btf *btf, __u32 id)
return align;
}
default:
- pr_warning("unsupported BTF_KIND:%u\n", btf_kind_of(t));
+ pr_warning("unsupported BTF_KIND:%u\n", btf_kind(t));
return 1;
}
}
@@ -737,20 +712,18 @@ static bool btf_is_struct_packed(const struct btf *btf, __u32 id,
const struct btf_member *m;
int align, i, bit_sz;
__u16 vlen;
- bool kflag;
align = btf_align_of(btf, id);
/* size of a non-packed struct has to be a multiple of its alignment*/
if (t->size % align)
return true;
- m = (void *)(t + 1);
- kflag = btf_kflag_of(t);
- vlen = btf_vlen_of(t);
+ m = btf_members(t);
+ vlen = btf_vlen(t);
/* all non-bitfield fields have to be naturally aligned */
for (i = 0; i < vlen; i++, m++) {
align = btf_align_of(btf, m->type);
- bit_sz = kflag ? BTF_MEMBER_BITFIELD_SIZE(m->offset) : 0;
+ bit_sz = btf_member_bitfield_size(t, i);
if (bit_sz == 0 && m->offset % (8 * align) != 0)
return true;
}
@@ -807,7 +780,7 @@ static void btf_dump_emit_struct_fwd(struct btf_dump *d, __u32 id,
const struct btf_type *t)
{
btf_dump_printf(d, "%s %s",
- btf_kind_of(t) == BTF_KIND_STRUCT ? "struct" : "union",
+ btf_is_struct(t) ? "struct" : "union",
btf_dump_type_name(d, id));
}
@@ -816,12 +789,11 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
const struct btf_type *t,
int lvl)
{
- const struct btf_member *m = (void *)(t + 1);
- bool kflag = btf_kflag_of(t), is_struct;
+ const struct btf_member *m = btf_members(t);
+ bool is_struct = btf_is_struct(t);
int align, i, packed, off = 0;
- __u16 vlen = btf_vlen_of(t);
+ __u16 vlen = btf_vlen(t);
- is_struct = btf_kind_of(t) == BTF_KIND_STRUCT;
packed = is_struct ? btf_is_struct_packed(d->btf, id, t) : 0;
align = packed ? 1 : btf_align_of(d->btf, id);
@@ -835,8 +807,8 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
int m_off, m_sz;
fname = btf_name_of(d, m->name_off);
- m_sz = kflag ? BTF_MEMBER_BITFIELD_SIZE(m->offset) : 0;
- m_off = kflag ? BTF_MEMBER_BIT_OFFSET(m->offset) : m->offset;
+ m_sz = btf_member_bitfield_size(t, i);
+ m_off = btf_member_bit_offset(t, i);
align = packed ? 1 : btf_align_of(d->btf, m->type);
btf_dump_emit_bit_padding(d, off, m_off, m_sz, align, lvl + 1);
@@ -871,7 +843,7 @@ static void btf_dump_emit_enum_def(struct btf_dump *d, __u32 id,
int lvl)
{
const struct btf_enum *v = (void *)(t+1);
- __u16 vlen = btf_vlen_of(t);
+ __u16 vlen = btf_vlen(t);
const char *name;
size_t dup_cnt;
int i;
@@ -905,7 +877,7 @@ static void btf_dump_emit_fwd_def(struct btf_dump *d, __u32 id,
{
const char *name = btf_dump_type_name(d, id);
- if (btf_kflag_of(t))
+ if (btf_kflag(t))
btf_dump_printf(d, "union %s", name);
else
btf_dump_printf(d, "struct %s", name);
@@ -987,7 +959,6 @@ static void btf_dump_emit_type_decl(struct btf_dump *d, __u32 id,
struct id_stack decl_stack;
const struct btf_type *t;
int err, stack_start;
- __u16 kind;
stack_start = d->decl_stack_cnt;
for (;;) {
@@ -1008,8 +979,7 @@ static void btf_dump_emit_type_decl(struct btf_dump *d, __u32 id,
break;
t = btf__type_by_id(d->btf, id);
- kind = btf_kind_of(t);
- switch (kind) {
+ switch (btf_kind(t)) {
case BTF_KIND_PTR:
case BTF_KIND_VOLATILE:
case BTF_KIND_CONST:
@@ -1017,12 +987,9 @@ static void btf_dump_emit_type_decl(struct btf_dump *d, __u32 id,
case BTF_KIND_FUNC_PROTO:
id = t->type;
break;
- case BTF_KIND_ARRAY: {
- const struct btf_array *a = (void *)(t + 1);
-
- id = a->type;
+ case BTF_KIND_ARRAY:
+ id = btf_array(t)->type;
break;
- }
case BTF_KIND_INT:
case BTF_KIND_ENUM:
case BTF_KIND_FWD:
@@ -1032,7 +999,7 @@ static void btf_dump_emit_type_decl(struct btf_dump *d, __u32 id,
goto done;
default:
pr_warning("unexpected type in decl chain, kind:%u, id:[%u]\n",
- kind, id);
+ btf_kind(t), id);
goto done;
}
}
@@ -1070,7 +1037,7 @@ static void btf_dump_emit_mods(struct btf_dump *d, struct id_stack *decl_stack)
id = decl_stack->ids[decl_stack->cnt - 1];
t = btf__type_by_id(d->btf, id);
- switch (btf_kind_of(t)) {
+ switch (btf_kind(t)) {
case BTF_KIND_VOLATILE:
btf_dump_printf(d, "volatile ");
break;
@@ -1087,20 +1054,6 @@ static void btf_dump_emit_mods(struct btf_dump *d, struct id_stack *decl_stack)
}
}
-static bool btf_is_mod_kind(const struct btf *btf, __u32 id)
-{
- const struct btf_type *t = btf__type_by_id(btf, id);
-
- switch (btf_kind_of(t)) {
- case BTF_KIND_VOLATILE:
- case BTF_KIND_CONST:
- case BTF_KIND_RESTRICT:
- return true;
- default:
- return false;
- }
-}
-
static void btf_dump_emit_name(const struct btf_dump *d,
const char *name, bool last_was_ptr)
{
@@ -1139,7 +1092,7 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
}
t = btf__type_by_id(d->btf, id);
- kind = btf_kind_of(t);
+ kind = btf_kind(t);
switch (kind) {
case BTF_KIND_INT:
@@ -1185,7 +1138,7 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
btf_dump_printf(d, " restrict");
break;
case BTF_KIND_ARRAY: {
- const struct btf_array *a = (void *)(t + 1);
+ const struct btf_array *a = btf_array(t);
const struct btf_type *next_t;
__u32 next_id;
bool multidim;
@@ -1201,7 +1154,8 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
*/
while (decls->cnt) {
next_id = decls->ids[decls->cnt - 1];
- if (btf_is_mod_kind(d->btf, next_id))
+ next_t = btf__type_by_id(d->btf, next_id);
+ if (btf_is_mod(next_t))
decls->cnt--;
else
break;
@@ -1214,7 +1168,7 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
}
next_t = btf__type_by_id(d->btf, next_id);
- multidim = btf_kind_of(next_t) == BTF_KIND_ARRAY;
+ multidim = btf_is_array(next_t);
/* we need space if we have named non-pointer */
if (fname[0] && !last_was_ptr)
btf_dump_printf(d, " ");
@@ -1228,8 +1182,8 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
return;
}
case BTF_KIND_FUNC_PROTO: {
- const struct btf_param *p = (void *)(t + 1);
- __u16 vlen = btf_vlen_of(t);
+ const struct btf_param *p = btf_params(t);
+ __u16 vlen = btf_vlen(t);
int i;
btf_dump_emit_mods(d, decls);
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index ead915aec349..fafcc75b33b1 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -1049,9 +1049,9 @@ static bool get_map_field_int(const char *map_name, const struct btf *btf,
const struct btf_array *arr_info;
const struct btf_type *arr_t;
- if (BTF_INFO_KIND(t->info) != BTF_KIND_PTR) {
+ if (!btf_is_ptr(t)) {
pr_warning("map '%s': attr '%s': expected PTR, got %u.\n",
- map_name, name, BTF_INFO_KIND(t->info));
+ map_name, name, btf_kind(t));
return false;
}
@@ -1061,12 +1061,12 @@ static bool get_map_field_int(const char *map_name, const struct btf *btf,
map_name, name, t->type);
return false;
}
- if (BTF_INFO_KIND(arr_t->info) != BTF_KIND_ARRAY) {
+ if (!btf_is_array(arr_t)) {
pr_warning("map '%s': attr '%s': expected ARRAY, got %u.\n",
- map_name, name, BTF_INFO_KIND(arr_t->info));
+ map_name, name, btf_kind(arr_t));
return false;
}
- arr_info = (const void *)(arr_t + 1);
+ arr_info = btf_array(arr_t);
*res = arr_info->nelems;
return true;
}
@@ -1084,11 +1084,11 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
struct bpf_map *map;
int vlen, i;
- vi = (const struct btf_var_secinfo *)(const void *)(sec + 1) + var_idx;
+ vi = btf_var_secinfos(sec) + var_idx;
var = btf__type_by_id(obj->btf, vi->type);
- var_extra = (const void *)(var + 1);
+ var_extra = btf_var(var);
map_name = btf__name_by_offset(obj->btf, var->name_off);
- vlen = BTF_INFO_VLEN(var->info);
+ vlen = btf_vlen(var);
if (map_name == NULL || map_name[0] == '\0') {
pr_warning("map #%d: empty name.\n", var_idx);
@@ -1098,9 +1098,9 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
pr_warning("map '%s' BTF data is corrupted.\n", map_name);
return -EINVAL;
}
- if (BTF_INFO_KIND(var->info) != BTF_KIND_VAR) {
+ if (!btf_is_var(var)) {
pr_warning("map '%s': unexpected var kind %u.\n",
- map_name, BTF_INFO_KIND(var->info));
+ map_name, btf_kind(var));
return -EINVAL;
}
if (var_extra->linkage != BTF_VAR_GLOBAL_ALLOCATED &&
@@ -1111,9 +1111,9 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
}
def = skip_mods_and_typedefs(obj->btf, var->type);
- if (BTF_INFO_KIND(def->info) != BTF_KIND_STRUCT) {
+ if (!btf_is_struct(def)) {
pr_warning("map '%s': unexpected def kind %u.\n",
- map_name, BTF_INFO_KIND(var->info));
+ map_name, btf_kind(var));
return -EINVAL;
}
if (def->size > vi->size) {
@@ -1136,8 +1136,8 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
pr_debug("map '%s': at sec_idx %d, offset %zu.\n",
map_name, map->sec_idx, map->sec_offset);
- vlen = BTF_INFO_VLEN(def->info);
- m = (const void *)(def + 1);
+ vlen = btf_vlen(def);
+ m = btf_members(def);
for (i = 0; i < vlen; i++, m++) {
const char *name = btf__name_by_offset(obj->btf, m->name_off);
@@ -1187,9 +1187,9 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
map_name, m->type);
return -EINVAL;
}
- if (BTF_INFO_KIND(t->info) != BTF_KIND_PTR) {
+ if (!btf_is_ptr(t)) {
pr_warning("map '%s': key spec is not PTR: %u.\n",
- map_name, BTF_INFO_KIND(t->info));
+ map_name, btf_kind(t));
return -EINVAL;
}
sz = btf__resolve_size(obj->btf, t->type);
@@ -1230,9 +1230,9 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
map_name, m->type);
return -EINVAL;
}
- if (BTF_INFO_KIND(t->info) != BTF_KIND_PTR) {
+ if (!btf_is_ptr(t)) {
pr_warning("map '%s': value spec is not PTR: %u.\n",
- map_name, BTF_INFO_KIND(t->info));
+ map_name, btf_kind(t));
return -EINVAL;
}
sz = btf__resolve_size(obj->btf, t->type);
@@ -1293,7 +1293,7 @@ static int bpf_object__init_user_btf_maps(struct bpf_object *obj, bool strict)
nr_types = btf__get_nr_types(obj->btf);
for (i = 1; i <= nr_types; i++) {
t = btf__type_by_id(obj->btf, i);
- if (BTF_INFO_KIND(t->info) != BTF_KIND_DATASEC)
+ if (!btf_is_datasec(t))
continue;
name = btf__name_by_offset(obj->btf, t->name_off);
if (strcmp(name, MAPS_ELF_SEC) == 0) {
@@ -1307,7 +1307,7 @@ static int bpf_object__init_user_btf_maps(struct bpf_object *obj, bool strict)
return -ENOENT;
}
- vlen = BTF_INFO_VLEN(sec->info);
+ vlen = btf_vlen(sec);
for (i = 0; i < vlen; i++) {
err = bpf_object__init_user_btf_map(obj, sec, i,
obj->efile.btf_maps_shndx,
@@ -1368,24 +1368,22 @@ static void bpf_object__sanitize_btf(struct bpf_object *obj)
struct btf *btf = obj->btf;
struct btf_type *t;
int i, j, vlen;
- __u16 kind;
if (!obj->btf || (has_func && has_datasec))
return;
for (i = 1; i <= btf__get_nr_types(btf); i++) {
t = (struct btf_type *)btf__type_by_id(btf, i);
- kind = BTF_INFO_KIND(t->info);
- if (!has_datasec && kind == BTF_KIND_VAR) {
+ if (!has_datasec && btf_is_var(t)) {
/* replace VAR with INT */
t->info = BTF_INFO_ENC(BTF_KIND_INT, 0, 0);
t->size = sizeof(int);
- *(int *)(t+1) = BTF_INT_ENC(0, 0, 32);
- } else if (!has_datasec && kind == BTF_KIND_DATASEC) {
+ *(int *)(t + 1) = BTF_INT_ENC(0, 0, 32);
+ } else if (!has_datasec && btf_is_datasec(t)) {
/* replace DATASEC with STRUCT */
- struct btf_var_secinfo *v = (void *)(t + 1);
- struct btf_member *m = (void *)(t + 1);
+ const struct btf_var_secinfo *v = btf_var_secinfos(t);
+ struct btf_member *m = (void *)btf_members(t);
struct btf_type *vt;
char *name;
@@ -1396,7 +1394,7 @@ static void bpf_object__sanitize_btf(struct bpf_object *obj)
name++;
}
- vlen = BTF_INFO_VLEN(t->info);
+ vlen = btf_vlen(t);
t->info = BTF_INFO_ENC(BTF_KIND_STRUCT, 0, vlen);
for (j = 0; j < vlen; j++, v++, m++) {
/* order of field assignments is important */
@@ -1406,12 +1404,12 @@ static void bpf_object__sanitize_btf(struct bpf_object *obj)
vt = (void *)btf__type_by_id(btf, v->type);
m->name_off = vt->name_off;
}
- } else if (!has_func && kind == BTF_KIND_FUNC_PROTO) {
+ } else if (!has_func && btf_is_func_proto(t)) {
/* replace FUNC_PROTO with ENUM */
- vlen = BTF_INFO_VLEN(t->info);
+ vlen = btf_vlen(t);
t->info = BTF_INFO_ENC(BTF_KIND_ENUM, 0, vlen);
t->size = sizeof(__u32); /* kernel enforced */
- } else if (!has_func && kind == BTF_KIND_FUNC) {
+ } else if (!has_func && btf_is_func(t)) {
/* replace FUNC with TYPEDEF */
t->info = BTF_INFO_ENC(BTF_KIND_TYPEDEF, 0, 0);
}
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 14/14] selftests/bpf: add CO-RE relocs misc tests
From: Andrii Nakryiko @ 2019-08-07 5:38 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Add tests validating few edge-cases of capturing offset relocations.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
---
.../selftests/bpf/prog_tests/core_reloc.c | 19 +++++++
.../bpf/progs/btf__core_reloc_misc.c | 5 ++
.../selftests/bpf/progs/core_reloc_types.h | 25 ++++++++
.../bpf/progs/test_core_reloc_misc.c | 57 +++++++++++++++++++
4 files changed, 106 insertions(+)
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_misc.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_misc.c
diff --git a/tools/testing/selftests/bpf/prog_tests/core_reloc.c b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
index 11daf0dbf2a3..bcc21db3c974 100644
--- a/tools/testing/selftests/bpf/prog_tests/core_reloc.c
+++ b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
@@ -260,6 +260,25 @@ static struct core_reloc_test_case test_cases[] = {
INTS_ERR_CASE(ints___err_wrong_sz_16),
INTS_ERR_CASE(ints___err_wrong_sz_32),
INTS_ERR_CASE(ints___err_wrong_sz_64),
+
+ /* validate edge cases of capturing relocations */
+ {
+ .case_name = "misc",
+ .bpf_obj_file = "test_core_reloc_misc.o",
+ .btf_src_file = "btf__core_reloc_misc.o",
+ .input = (const char *)&(struct core_reloc_misc_extensible[]){
+ { .a = 1 },
+ { .a = 2 }, /* not read */
+ { .a = 3 },
+ },
+ .input_len = 4 * sizeof(int),
+ .output = STRUCT_TO_CHAR_PTR(core_reloc_misc_output) {
+ .a = 1,
+ .b = 1,
+ .c = 0, /* BUG in clang, should be 3 */
+ },
+ .output_len = sizeof(struct core_reloc_misc_output),
+ },
};
struct data {
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_misc.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_misc.c
new file mode 100644
index 000000000000..ed9ad8b5b4f8
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_misc.c
@@ -0,0 +1,5 @@
+#include "core_reloc_types.h"
+
+void f1(struct core_reloc_misc___a x) {}
+void f2(struct core_reloc_misc___b x) {}
+void f3(struct core_reloc_misc_extensible x) {}
diff --git a/tools/testing/selftests/bpf/progs/core_reloc_types.h b/tools/testing/selftests/bpf/progs/core_reloc_types.h
index 5f3ebd4f6dc3..10a252b6da55 100644
--- a/tools/testing/selftests/bpf/progs/core_reloc_types.h
+++ b/tools/testing/selftests/bpf/progs/core_reloc_types.h
@@ -640,3 +640,28 @@ struct core_reloc_ints___err_wrong_sz_64 {
uint32_t u64_field; /* not 64-bit anymore */
int32_t s64_field; /* not 64-bit anymore */
};
+
+/*
+ * MISC
+ */
+struct core_reloc_misc_output {
+ int a, b, c;
+};
+
+struct core_reloc_misc___a {
+ int a1;
+ int a2;
+};
+
+struct core_reloc_misc___b {
+ int b1;
+ int b2;
+};
+
+/* this one extends core_reloc_misc_extensible struct from BPF prog */
+struct core_reloc_misc_extensible {
+ int a;
+ int b;
+ int c;
+ int d;
+};
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_misc.c b/tools/testing/selftests/bpf/progs/test_core_reloc_misc.c
new file mode 100644
index 000000000000..c59984bd3e23
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_misc.c
@@ -0,0 +1,57 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2019 Facebook
+
+#include <linux/bpf.h>
+#include <stdint.h>
+#include "bpf_helpers.h"
+
+char _license[] SEC("license") = "GPL";
+
+static volatile struct data {
+ char in[256];
+ char out[256];
+} data;
+
+struct core_reloc_misc_output {
+ int a, b, c;
+};
+
+struct core_reloc_misc___a {
+ int a1;
+ int a2;
+};
+
+struct core_reloc_misc___b {
+ int b1;
+ int b2;
+};
+
+/* fixed two first members, can be extended with new fields */
+struct core_reloc_misc_extensible {
+ int a;
+ int b;
+};
+
+SEC("raw_tracepoint/sys_enter")
+int test_core_misc(void *ctx)
+{
+ struct core_reloc_misc___a *in_a = (void *)&data.in;
+ struct core_reloc_misc___b *in_b = (void *)&data.in;
+ struct core_reloc_misc_extensible *in_ext = (void *)&data.in;
+ struct core_reloc_misc_output *out = (void *)&data.out;
+
+ /* record two different relocations with the same accessor string */
+ if (BPF_CORE_READ(&out->a, &in_a->a1) || /* accessor: 0:0 */
+ BPF_CORE_READ(&out->b, &in_b->b1)) /* accessor: 0:0 */
+ return 1;
+
+ /* Validate relocations capture array-only accesses for structs with
+ * fixed header, but with potentially extendable tail. This will read
+ * first 4 bytes of 2nd element of in_ext array of potentially
+ * variably sized struct core_reloc_misc_extensible. */
+ if (BPF_CORE_READ(&out->c, &in_ext[2])) /* accessor: 2 */
+ return 1;
+
+ return 0;
+}
+
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 13/14] selftests/bpf: add CO-RE relocs ints tests
From: Andrii Nakryiko @ 2019-08-07 5:38 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Add various tests validating handling compatible/incompatible integer
types.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
---
.../selftests/bpf/prog_tests/core_reloc.c | 40 +++++++
.../bpf/progs/btf__core_reloc_ints.c | 3 +
.../bpf/progs/btf__core_reloc_ints___bool.c | 3 +
.../btf__core_reloc_ints___err_bitfield.c | 3 +
.../btf__core_reloc_ints___err_wrong_sz_16.c | 3 +
.../btf__core_reloc_ints___err_wrong_sz_32.c | 3 +
.../btf__core_reloc_ints___err_wrong_sz_64.c | 3 +
.../btf__core_reloc_ints___err_wrong_sz_8.c | 3 +
.../btf__core_reloc_ints___reverse_sign.c | 3 +
.../selftests/bpf/progs/core_reloc_types.h | 101 ++++++++++++++++++
.../bpf/progs/test_core_reloc_ints.c | 44 ++++++++
11 files changed, 209 insertions(+)
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___bool.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_bitfield.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_16.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_32.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_64.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_8.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___reverse_sign.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_ints.c
diff --git a/tools/testing/selftests/bpf/prog_tests/core_reloc.c b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
index 77478b51699b..11daf0dbf2a3 100644
--- a/tools/testing/selftests/bpf/prog_tests/core_reloc.c
+++ b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
@@ -145,6 +145,35 @@
.output_len = sizeof(struct core_reloc_ptr_as_arr), \
}
+#define INTS_DATA(struct_name) STRUCT_TO_CHAR_PTR(struct_name) { \
+ .u8_field = 1, \
+ .s8_field = 2, \
+ .u16_field = 3, \
+ .s16_field = 4, \
+ .u32_field = 5, \
+ .s32_field = 6, \
+ .u64_field = 7, \
+ .s64_field = 8, \
+}
+
+#define INTS_CASE_COMMON(name) \
+ .case_name = #name, \
+ .bpf_obj_file = "test_core_reloc_ints.o", \
+ .btf_src_file = "btf__core_reloc_" #name ".o"
+
+#define INTS_CASE(name) { \
+ INTS_CASE_COMMON(name), \
+ .input = INTS_DATA(core_reloc_##name), \
+ .input_len = sizeof(struct core_reloc_##name), \
+ .output = INTS_DATA(core_reloc_ints), \
+ .output_len = sizeof(struct core_reloc_ints), \
+}
+
+#define INTS_ERR_CASE(name) { \
+ INTS_CASE_COMMON(name), \
+ .fails = true, \
+}
+
struct core_reloc_test_case {
const char *case_name;
const char *bpf_obj_file;
@@ -220,6 +249,17 @@ static struct core_reloc_test_case test_cases[] = {
/* handling "ptr is an array" semantics */
PTR_AS_ARR_CASE(ptr_as_arr),
PTR_AS_ARR_CASE(ptr_as_arr___diff_sz),
+
+ /* int signedness/sizing/bitfield handling */
+ INTS_CASE(ints),
+ INTS_CASE(ints___bool),
+ INTS_CASE(ints___reverse_sign),
+
+ INTS_ERR_CASE(ints___err_bitfield),
+ INTS_ERR_CASE(ints___err_wrong_sz_8),
+ INTS_ERR_CASE(ints___err_wrong_sz_16),
+ INTS_ERR_CASE(ints___err_wrong_sz_32),
+ INTS_ERR_CASE(ints___err_wrong_sz_64),
};
struct data {
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_ints.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints.c
new file mode 100644
index 000000000000..7d0f041042c5
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_ints x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___bool.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___bool.c
new file mode 100644
index 000000000000..f9359450186e
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___bool.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_ints___bool x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_bitfield.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_bitfield.c
new file mode 100644
index 000000000000..50369e8320a0
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_bitfield.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_ints___err_bitfield x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_16.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_16.c
new file mode 100644
index 000000000000..823bac13d641
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_16.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_ints___err_wrong_sz_16 x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_32.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_32.c
new file mode 100644
index 000000000000..b44f3be18535
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_32.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_ints___err_wrong_sz_32 x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_64.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_64.c
new file mode 100644
index 000000000000..9a3dd2099c0f
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_64.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_ints___err_wrong_sz_64 x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_8.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_8.c
new file mode 100644
index 000000000000..9f11ef5f6e88
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_8.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_ints___err_wrong_sz_8 x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___reverse_sign.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___reverse_sign.c
new file mode 100644
index 000000000000..aafb1c5819d7
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_ints___reverse_sign.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_ints___reverse_sign x) {}
diff --git a/tools/testing/selftests/bpf/progs/core_reloc_types.h b/tools/testing/selftests/bpf/progs/core_reloc_types.h
index c17c9279deae..5f3ebd4f6dc3 100644
--- a/tools/testing/selftests/bpf/progs/core_reloc_types.h
+++ b/tools/testing/selftests/bpf/progs/core_reloc_types.h
@@ -1,3 +1,6 @@
+#include <stdint.h>
+#include <stdbool.h>
+
/*
* FLAVORS
*/
@@ -539,3 +542,101 @@ struct core_reloc_ptr_as_arr___diff_sz {
char __some_more_padding;
int a;
};
+
+/*
+ * INTS
+ */
+struct core_reloc_ints {
+ uint8_t u8_field;
+ int8_t s8_field;
+ uint16_t u16_field;
+ int16_t s16_field;
+ uint32_t u32_field;
+ int32_t s32_field;
+ uint64_t u64_field;
+ int64_t s64_field;
+};
+
+/* signed/unsigned types swap */
+struct core_reloc_ints___reverse_sign {
+ int8_t u8_field;
+ uint8_t s8_field;
+ int16_t u16_field;
+ uint16_t s16_field;
+ int32_t u32_field;
+ uint32_t s32_field;
+ int64_t u64_field;
+ uint64_t s64_field;
+};
+
+struct core_reloc_ints___bool {
+ bool u8_field; /* bool instead of uint8 */
+ int8_t s8_field;
+ uint16_t u16_field;
+ int16_t s16_field;
+ uint32_t u32_field;
+ int32_t s32_field;
+ uint64_t u64_field;
+ int64_t s64_field;
+};
+
+struct core_reloc_ints___err_bitfield {
+ uint8_t u8_field;
+ int8_t s8_field;
+ uint16_t u16_field;
+ int16_t s16_field;
+ uint32_t u32_field: 32; /* bitfields are not supported */
+ int32_t s32_field;
+ uint64_t u64_field;
+ int64_t s64_field;
+};
+
+struct core_reloc_ints___err_wrong_sz_8 {
+ uint16_t u8_field; /* not 8-bit anymore */
+ int16_t s8_field; /* not 8-bit anymore */
+
+ uint16_t u16_field;
+ int16_t s16_field;
+ uint32_t u32_field;
+ int32_t s32_field;
+ uint64_t u64_field;
+ int64_t s64_field;
+};
+
+struct core_reloc_ints___err_wrong_sz_16 {
+ uint8_t u8_field;
+ int8_t s8_field;
+
+ uint32_t u16_field; /* not 16-bit anymore */
+ int32_t s16_field; /* not 16-bit anymore */
+
+ uint32_t u32_field;
+ int32_t s32_field;
+ uint64_t u64_field;
+ int64_t s64_field;
+};
+
+struct core_reloc_ints___err_wrong_sz_32 {
+ uint8_t u8_field;
+ int8_t s8_field;
+ uint16_t u16_field;
+ int16_t s16_field;
+
+ uint64_t u32_field; /* not 32-bit anymore */
+ int64_t s32_field; /* not 32-bit anymore */
+
+ uint64_t u64_field;
+ int64_t s64_field;
+};
+
+struct core_reloc_ints___err_wrong_sz_64 {
+ uint8_t u8_field;
+ int8_t s8_field;
+ uint16_t u16_field;
+ int16_t s16_field;
+ uint32_t u32_field;
+ int32_t s32_field;
+
+ uint32_t u64_field; /* not 64-bit anymore */
+ int32_t s64_field; /* not 64-bit anymore */
+};
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_ints.c b/tools/testing/selftests/bpf/progs/test_core_reloc_ints.c
new file mode 100644
index 000000000000..d99233c8008a
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_ints.c
@@ -0,0 +1,44 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2019 Facebook
+
+#include <linux/bpf.h>
+#include <stdint.h>
+#include "bpf_helpers.h"
+
+char _license[] SEC("license") = "GPL";
+
+static volatile struct data {
+ char in[256];
+ char out[256];
+} data;
+
+struct core_reloc_ints {
+ uint8_t u8_field;
+ int8_t s8_field;
+ uint16_t u16_field;
+ int16_t s16_field;
+ uint32_t u32_field;
+ int32_t s32_field;
+ uint64_t u64_field;
+ int64_t s64_field;
+};
+
+SEC("raw_tracepoint/sys_enter")
+int test_core_ints(void *ctx)
+{
+ struct core_reloc_ints *in = (void *)&data.in;
+ struct core_reloc_ints *out = (void *)&data.out;
+
+ if (BPF_CORE_READ(&out->u8_field, &in->u8_field) ||
+ BPF_CORE_READ(&out->s8_field, &in->s8_field) ||
+ BPF_CORE_READ(&out->u16_field, &in->u16_field) ||
+ BPF_CORE_READ(&out->s16_field, &in->s16_field) ||
+ BPF_CORE_READ(&out->u32_field, &in->u32_field) ||
+ BPF_CORE_READ(&out->s32_field, &in->s32_field) ||
+ BPF_CORE_READ(&out->u64_field, &in->u64_field) ||
+ BPF_CORE_READ(&out->s64_field, &in->s64_field))
+ return 1;
+
+ return 0;
+}
+
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 12/14] selftests/bpf: add CO-RE relocs ptr-as-array tests
From: Andrii Nakryiko @ 2019-08-07 5:38 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Add test validating correct relocation handling for cases where pointer
to something is used as an array. E.g.:
int *ptr = ...;
int x = ptr[42];
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
---
.../selftests/bpf/prog_tests/core_reloc.c | 20 +++++++++++++
.../bpf/progs/btf__core_reloc_ptr_as_arr.c | 3 ++
.../btf__core_reloc_ptr_as_arr___diff_sz.c | 3 ++
.../selftests/bpf/progs/core_reloc_types.h | 13 ++++++++
.../bpf/progs/test_core_reloc_ptr_as_arr.c | 30 +++++++++++++++++++
5 files changed, 69 insertions(+)
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ptr_as_arr.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ptr_as_arr___diff_sz.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c
diff --git a/tools/testing/selftests/bpf/prog_tests/core_reloc.c b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
index cf38c787537a..77478b51699b 100644
--- a/tools/testing/selftests/bpf/prog_tests/core_reloc.c
+++ b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
@@ -129,6 +129,22 @@
.output_len = sizeof(struct core_reloc_mods_output), \
}
+#define PTR_AS_ARR_CASE(name) { \
+ .case_name = #name, \
+ .bpf_obj_file = "test_core_reloc_ptr_as_arr.o", \
+ .btf_src_file = "btf__core_reloc_" #name ".o", \
+ .input = (const char *)&(struct core_reloc_##name []){ \
+ { .a = 1 }, \
+ { .a = 2 }, \
+ { .a = 3 }, \
+ }, \
+ .input_len = 3 * sizeof(struct core_reloc_##name), \
+ .output = STRUCT_TO_CHAR_PTR(core_reloc_ptr_as_arr) { \
+ .a = 3, \
+ }, \
+ .output_len = sizeof(struct core_reloc_ptr_as_arr), \
+}
+
struct core_reloc_test_case {
const char *case_name;
const char *bpf_obj_file;
@@ -200,6 +216,10 @@ static struct core_reloc_test_case test_cases[] = {
MODS_CASE(mods),
MODS_CASE(mods___mod_swap),
MODS_CASE(mods___typedefs),
+
+ /* handling "ptr is an array" semantics */
+ PTR_AS_ARR_CASE(ptr_as_arr),
+ PTR_AS_ARR_CASE(ptr_as_arr___diff_sz),
};
struct data {
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_ptr_as_arr.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_ptr_as_arr.c
new file mode 100644
index 000000000000..8da52432ba17
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_ptr_as_arr.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_ptr_as_arr x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_ptr_as_arr___diff_sz.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_ptr_as_arr___diff_sz.c
new file mode 100644
index 000000000000..003acfc9a3e7
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_ptr_as_arr___diff_sz.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_ptr_as_arr___diff_sz x) {}
diff --git a/tools/testing/selftests/bpf/progs/core_reloc_types.h b/tools/testing/selftests/bpf/progs/core_reloc_types.h
index 3401e8342e57..c17c9279deae 100644
--- a/tools/testing/selftests/bpf/progs/core_reloc_types.h
+++ b/tools/testing/selftests/bpf/progs/core_reloc_types.h
@@ -526,3 +526,16 @@ struct core_reloc_mods___typedefs {
int3_t b;
int3_t a;
};
+
+/*
+ * PTR_AS_ARR
+ */
+struct core_reloc_ptr_as_arr {
+ int a;
+};
+
+struct core_reloc_ptr_as_arr___diff_sz {
+ int :32; /* padding */
+ char __some_more_padding;
+ int a;
+};
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c b/tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c
new file mode 100644
index 000000000000..526b7ddc7ea1
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c
@@ -0,0 +1,30 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2019 Facebook
+
+#include <linux/bpf.h>
+#include <stdint.h>
+#include "bpf_helpers.h"
+
+char _license[] SEC("license") = "GPL";
+
+static volatile struct data {
+ char in[256];
+ char out[256];
+} data;
+
+struct core_reloc_ptr_as_arr {
+ int a;
+};
+
+SEC("raw_tracepoint/sys_enter")
+int test_core_ptr_as_arr(void *ctx)
+{
+ struct core_reloc_ptr_as_arr *in = (void *)&data.in;
+ struct core_reloc_ptr_as_arr *out = (void *)&data.out;
+
+ if (BPF_CORE_READ(&out->a, &in[2].a))
+ return 1;
+
+ return 0;
+}
+
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 11/14] selftests/bpf: add CO-RE relocs modifiers/typedef tests
From: Andrii Nakryiko @ 2019-08-07 5:38 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Add tests validating correct handling of various combinations of
typedefs and const/volatile/restrict modifiers.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
---
.../selftests/bpf/prog_tests/core_reloc.c | 27 +++++++
.../bpf/progs/btf__core_reloc_mods.c | 3 +
.../progs/btf__core_reloc_mods___mod_swap.c | 3 +
.../progs/btf__core_reloc_mods___typedefs.c | 3 +
.../selftests/bpf/progs/core_reloc_types.h | 72 +++++++++++++++++++
.../bpf/progs/test_core_reloc_mods.c | 62 ++++++++++++++++
6 files changed, 170 insertions(+)
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_mods.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_mods___mod_swap.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_mods___typedefs.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_mods.c
diff --git a/tools/testing/selftests/bpf/prog_tests/core_reloc.c b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
index c924c8ae67af..cf38c787537a 100644
--- a/tools/testing/selftests/bpf/prog_tests/core_reloc.c
+++ b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
@@ -107,6 +107,28 @@
.fails = true, \
}
+#define MODS_CASE(name) { \
+ .case_name = #name, \
+ .bpf_obj_file = "test_core_reloc_mods.o", \
+ .btf_src_file = "btf__core_reloc_" #name ".o", \
+ .input = STRUCT_TO_CHAR_PTR(core_reloc_##name) { \
+ .a = 1, \
+ .b = 2, \
+ .c = (void *)3, \
+ .d = (void *)4, \
+ .e = { [2] = 5 }, \
+ .f = { [1] = 6 }, \
+ .g = { .x = 7 }, \
+ .h = { .y = 8 }, \
+ }, \
+ .input_len = sizeof(struct core_reloc_##name), \
+ .output = STRUCT_TO_CHAR_PTR(core_reloc_mods_output) { \
+ .a = 1, .b = 2, .c = 3, .d = 4, \
+ .e = 5, .f = 6, .g = 7, .h = 8, \
+ }, \
+ .output_len = sizeof(struct core_reloc_mods_output), \
+}
+
struct core_reloc_test_case {
const char *case_name;
const char *bpf_obj_file;
@@ -173,6 +195,11 @@ static struct core_reloc_test_case test_cases[] = {
PRIMITIVES_ERR_CASE(primitives___err_non_enum),
PRIMITIVES_ERR_CASE(primitives___err_non_int),
PRIMITIVES_ERR_CASE(primitives___err_non_ptr),
+
+ /* const/volatile/restrict and typedefs scenarios */
+ MODS_CASE(mods),
+ MODS_CASE(mods___mod_swap),
+ MODS_CASE(mods___typedefs),
};
struct data {
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_mods.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_mods.c
new file mode 100644
index 000000000000..124197a2e813
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_mods.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_mods x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_mods___mod_swap.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_mods___mod_swap.c
new file mode 100644
index 000000000000..f8a6592ca75f
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_mods___mod_swap.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_mods___mod_swap x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_mods___typedefs.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_mods___typedefs.c
new file mode 100644
index 000000000000..5c0d73687247
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_mods___typedefs.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_mods___typedefs x) {}
diff --git a/tools/testing/selftests/bpf/progs/core_reloc_types.h b/tools/testing/selftests/bpf/progs/core_reloc_types.h
index 7526a5f5755b..3401e8342e57 100644
--- a/tools/testing/selftests/bpf/progs/core_reloc_types.h
+++ b/tools/testing/selftests/bpf/progs/core_reloc_types.h
@@ -454,3 +454,75 @@ struct core_reloc_primitives___err_non_ptr {
int d; /* int instead of ptr */
int (*f)(const char *);
};
+
+/*
+ * MODS
+ */
+struct core_reloc_mods_output {
+ int a, b, c, d, e, f, g, h;
+};
+
+typedef const int int_t;
+typedef const char *char_ptr_t;
+typedef const int arr_t[7];
+
+struct core_reloc_mods_substruct {
+ int x;
+ int y;
+};
+
+typedef struct {
+ int x;
+ int y;
+} core_reloc_mods_substruct_t;
+
+struct core_reloc_mods {
+ int a;
+ int_t b;
+ char *c;
+ char_ptr_t d;
+ int e[3];
+ arr_t f;
+ struct core_reloc_mods_substruct g;
+ core_reloc_mods_substruct_t h;
+};
+
+/* a/b, c/d, e/f, and g/h pairs are swapped */
+struct core_reloc_mods___mod_swap {
+ int b;
+ int_t a;
+ char *d;
+ char_ptr_t c;
+ int f[3];
+ arr_t e;
+ struct {
+ int y;
+ int x;
+ } h;
+ core_reloc_mods_substruct_t g;
+};
+
+typedef int int1_t;
+typedef int1_t int2_t;
+typedef int2_t int3_t;
+
+typedef int arr1_t[5];
+typedef arr1_t arr2_t;
+typedef arr2_t arr3_t;
+typedef arr3_t arr4_t;
+
+typedef const char * const volatile restrict fancy_char_ptr_t;
+
+typedef core_reloc_mods_substruct_t core_reloc_mods_substruct_tt;
+
+/* we need more typedefs */
+struct core_reloc_mods___typedefs {
+ core_reloc_mods_substruct_tt g;
+ core_reloc_mods_substruct_tt h;
+ arr4_t f;
+ arr4_t e;
+ fancy_char_ptr_t d;
+ fancy_char_ptr_t c;
+ int3_t b;
+ int3_t a;
+};
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_mods.c b/tools/testing/selftests/bpf/progs/test_core_reloc_mods.c
new file mode 100644
index 000000000000..f98b942c062b
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_mods.c
@@ -0,0 +1,62 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2019 Facebook
+
+#include <linux/bpf.h>
+#include <stdint.h>
+#include "bpf_helpers.h"
+
+char _license[] SEC("license") = "GPL";
+
+static volatile struct data {
+ char in[256];
+ char out[256];
+} data;
+
+struct core_reloc_mods_output {
+ int a, b, c, d, e, f, g, h;
+};
+
+typedef const int int_t;
+typedef const char *char_ptr_t;
+typedef const int arr_t[7];
+
+struct core_reloc_mods_substruct {
+ int x;
+ int y;
+};
+
+typedef struct {
+ int x;
+ int y;
+} core_reloc_mods_substruct_t;
+
+struct core_reloc_mods {
+ int a;
+ int_t b;
+ char *c;
+ char_ptr_t d;
+ int e[3];
+ arr_t f;
+ struct core_reloc_mods_substruct g;
+ core_reloc_mods_substruct_t h;
+};
+
+SEC("raw_tracepoint/sys_enter")
+int test_core_mods(void *ctx)
+{
+ struct core_reloc_mods *in = (void *)&data.in;
+ struct core_reloc_mods_output *out = (void *)&data.out;
+
+ if (BPF_CORE_READ(&out->a, &in->a) ||
+ BPF_CORE_READ(&out->b, &in->b) ||
+ BPF_CORE_READ(&out->c, &in->c) ||
+ BPF_CORE_READ(&out->d, &in->d) ||
+ BPF_CORE_READ(&out->e, &in->e[2]) ||
+ BPF_CORE_READ(&out->f, &in->f[1]) ||
+ BPF_CORE_READ(&out->g, &in->g.x) ||
+ BPF_CORE_READ(&out->h, &in->h.y))
+ return 1;
+
+ return 0;
+}
+
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 09/14] selftests/bpf: add CO-RE relocs array tests
From: Andrii Nakryiko @ 2019-08-07 5:38 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Add tests for various array handling/relocation scenarios.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
---
.../selftests/bpf/prog_tests/core_reloc.c | 41 ++++++++++
.../bpf/progs/btf__core_reloc_arrays.c | 3 +
.../btf__core_reloc_arrays___diff_arr_dim.c | 3 +
...btf__core_reloc_arrays___diff_arr_val_sz.c | 3 +
.../btf__core_reloc_arrays___err_non_array.c | 3 +
...btf__core_reloc_arrays___err_too_shallow.c | 3 +
.../btf__core_reloc_arrays___err_too_small.c | 3 +
..._core_reloc_arrays___err_wrong_val_type1.c | 3 +
..._core_reloc_arrays___err_wrong_val_type2.c | 3 +
.../selftests/bpf/progs/core_reloc_types.h | 81 +++++++++++++++++++
.../bpf/progs/test_core_reloc_arrays.c | 55 +++++++++++++
11 files changed, 201 insertions(+)
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___diff_arr_dim.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___diff_arr_val_sz.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_non_array.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_too_shallow.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_too_small.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_wrong_val_type1.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_wrong_val_type2.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
diff --git a/tools/testing/selftests/bpf/prog_tests/core_reloc.c b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
index 2c0e695f958f..30e2f5c5d3b5 100644
--- a/tools/testing/selftests/bpf/prog_tests/core_reloc.c
+++ b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
@@ -51,6 +51,36 @@
.fails = true, \
}
+#define ARRAYS_DATA(struct_name) STRUCT_TO_CHAR_PTR(struct_name) { \
+ .a = { [2] = 1 }, \
+ .b = { [1] = { [2] = { [3] = 2 } } }, \
+ .c = { [1] = { .c = 3 } }, \
+ .d = { [0] = { [0] = { .d = 4 } } }, \
+}
+
+#define ARRAYS_CASE_COMMON(name) \
+ .case_name = #name, \
+ .bpf_obj_file = "test_core_reloc_arrays.o", \
+ .btf_src_file = "btf__core_reloc_" #name ".o"
+
+#define ARRAYS_CASE(name) { \
+ ARRAYS_CASE_COMMON(name), \
+ .input = ARRAYS_DATA(core_reloc_##name), \
+ .input_len = sizeof(struct core_reloc_##name), \
+ .output = STRUCT_TO_CHAR_PTR(core_reloc_arrays_output) { \
+ .a2 = 1, \
+ .b123 = 2, \
+ .c1c = 3, \
+ .d00d = 4, \
+ }, \
+ .output_len = sizeof(struct core_reloc_arrays_output) \
+}
+
+#define ARRAYS_ERR_CASE(name) { \
+ ARRAYS_CASE_COMMON(name), \
+ .fails = true, \
+}
+
struct core_reloc_test_case {
const char *case_name;
const char *bpf_obj_file;
@@ -96,6 +126,17 @@ static struct core_reloc_test_case test_cases[] = {
NESTING_ERR_CASE(nesting___err_dup_incompat_types),
NESTING_ERR_CASE(nesting___err_partial_match_dups),
NESTING_ERR_CASE(nesting___err_too_deep),
+
+ /* various array access relocation scenarios */
+ ARRAYS_CASE(arrays),
+ ARRAYS_CASE(arrays___diff_arr_dim),
+ ARRAYS_CASE(arrays___diff_arr_val_sz),
+
+ ARRAYS_ERR_CASE(arrays___err_too_small),
+ ARRAYS_ERR_CASE(arrays___err_too_shallow),
+ ARRAYS_ERR_CASE(arrays___err_non_array),
+ ARRAYS_ERR_CASE(arrays___err_wrong_val_type1),
+ ARRAYS_ERR_CASE(arrays___err_wrong_val_type2),
};
struct data {
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays.c
new file mode 100644
index 000000000000..018ed7fbba3a
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___diff_arr_dim.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___diff_arr_dim.c
new file mode 100644
index 000000000000..13d662c57014
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___diff_arr_dim.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___diff_arr_dim x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___diff_arr_val_sz.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___diff_arr_val_sz.c
new file mode 100644
index 000000000000..a351f418c85d
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___diff_arr_val_sz.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___diff_arr_val_sz x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_non_array.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_non_array.c
new file mode 100644
index 000000000000..a8735009becc
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_non_array.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___err_non_array x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_too_shallow.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_too_shallow.c
new file mode 100644
index 000000000000..2a67c28b1e75
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_too_shallow.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___err_too_shallow x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_too_small.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_too_small.c
new file mode 100644
index 000000000000..1142c08c925f
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_too_small.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___err_too_small x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_wrong_val_type1.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_wrong_val_type1.c
new file mode 100644
index 000000000000..795a5b729176
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_wrong_val_type1.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___err_wrong_val_type1 x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_wrong_val_type2.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_wrong_val_type2.c
new file mode 100644
index 000000000000..3af74b837c4d
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_wrong_val_type2.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___err_wrong_val_type2 x) {}
diff --git a/tools/testing/selftests/bpf/progs/core_reloc_types.h b/tools/testing/selftests/bpf/progs/core_reloc_types.h
index 340ee2bcd463..45de7986ea2e 100644
--- a/tools/testing/selftests/bpf/progs/core_reloc_types.h
+++ b/tools/testing/selftests/bpf/progs/core_reloc_types.h
@@ -306,3 +306,84 @@ struct core_reloc_nesting___err_too_deep {
} b;
} b;
};
+
+/*
+ * ARRAYS
+ */
+struct core_reloc_arrays_output {
+ int a2;
+ char b123;
+ int c1c;
+ int d00d;
+};
+
+struct core_reloc_arrays_substruct {
+ int c;
+ int d;
+};
+
+struct core_reloc_arrays {
+ int a[5];
+ char b[2][3][4];
+ struct core_reloc_arrays_substruct c[3];
+ struct core_reloc_arrays_substruct d[1][2];
+};
+
+/* bigger array dimensions */
+struct core_reloc_arrays___diff_arr_dim {
+ int a[7];
+ char b[3][4][5];
+ struct core_reloc_arrays_substruct c[4];
+ struct core_reloc_arrays_substruct d[2][3];
+};
+
+/* different size of array's value (struct) */
+struct core_reloc_arrays___diff_arr_val_sz {
+ int a[5];
+ char b[2][3][4];
+ struct {
+ int __padding1;
+ int c;
+ int __padding2;
+ } c[3];
+ struct {
+ int __padding1;
+ int d;
+ int __padding2;
+ } d[1][2];
+};
+
+struct core_reloc_arrays___err_too_small {
+ int a[2]; /* this one is too small */
+ char b[2][3][4];
+ struct core_reloc_arrays_substruct c[3];
+ struct core_reloc_arrays_substruct d[1][2];
+};
+
+struct core_reloc_arrays___err_too_shallow {
+ int a[5];
+ char b[2][3]; /* this one lacks one dimension */
+ struct core_reloc_arrays_substruct c[3];
+ struct core_reloc_arrays_substruct d[1][2];
+};
+
+struct core_reloc_arrays___err_non_array {
+ int a; /* not an array */
+ char b[2][3][4];
+ struct core_reloc_arrays_substruct c[3];
+ struct core_reloc_arrays_substruct d[1][2];
+};
+
+struct core_reloc_arrays___err_wrong_val_type1 {
+ char a[5]; /* char instead of int */
+ char b[2][3][4];
+ struct core_reloc_arrays_substruct c[3];
+ struct core_reloc_arrays_substruct d[1][2];
+};
+
+struct core_reloc_arrays___err_wrong_val_type2 {
+ int a[5];
+ char b[2][3][4];
+ int c[3]; /* value is not a struct */
+ struct core_reloc_arrays_substruct d[1][2];
+};
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c b/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
new file mode 100644
index 000000000000..bf67f0fdf743
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
@@ -0,0 +1,55 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2019 Facebook
+
+#include <linux/bpf.h>
+#include <stdint.h>
+#include "bpf_helpers.h"
+
+char _license[] SEC("license") = "GPL";
+
+static volatile struct data {
+ char in[256];
+ char out[256];
+} data;
+
+struct core_reloc_arrays_output {
+ int a2;
+ char b123;
+ int c1c;
+ int d00d;
+};
+
+struct core_reloc_arrays_substruct {
+ int c;
+ int d;
+};
+
+struct core_reloc_arrays {
+ int a[5];
+ char b[2][3][4];
+ struct core_reloc_arrays_substruct c[3];
+ struct core_reloc_arrays_substruct d[1][2];
+};
+
+SEC("raw_tracepoint/sys_enter")
+int test_core_arrays(void *ctx)
+{
+ struct core_reloc_arrays *in = (void *)&data.in;
+ struct core_reloc_arrays_output *out = (void *)&data.out;
+
+ /* in->a[2] */
+ if (BPF_CORE_READ(&out->a2, &in->a[2]))
+ return 1;
+ /* in->b[1][2][3] */
+ if (BPF_CORE_READ(&out->b123, &in->b[1][2][3]))
+ return 1;
+ /* in->c[1].c */
+ if (BPF_CORE_READ(&out->c1c, &in->c[1].c))
+ return 1;
+ /* in->d[0][0].d */
+ if (BPF_CORE_READ(&out->d00d, &in->d[0][0].d))
+ return 1;
+
+ return 0;
+}
+
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 07/14] selftests/bpf: add CO-RE relocs struct flavors tests
From: Andrii Nakryiko @ 2019-08-07 5:37 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Add tests verifying that BPF program can use various struct/union
"flavors" to extract data from the same target struct/union.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
---
.../selftests/bpf/prog_tests/core_reloc.c | 34 ++++++++++
.../bpf/progs/btf__core_reloc_flavors.c | 3 +
.../btf__core_reloc_flavors__err_wrong_name.c | 3 +
.../selftests/bpf/progs/core_reloc_types.h | 15 +++++
.../bpf/progs/test_core_reloc_flavors.c | 62 +++++++++++++++++++
5 files changed, 117 insertions(+)
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_flavors.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_flavors__err_wrong_name.c
create mode 100644 tools/testing/selftests/bpf/progs/core_reloc_types.h
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c
diff --git a/tools/testing/selftests/bpf/prog_tests/core_reloc.c b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
index f31cede82112..995d5e9fd4c3 100644
--- a/tools/testing/selftests/bpf/prog_tests/core_reloc.c
+++ b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
@@ -1,5 +1,32 @@
// SPDX-License-Identifier: GPL-2.0
#include <test_progs.h>
+#include "progs/core_reloc_types.h"
+
+#define STRUCT_TO_CHAR_PTR(struct_name) (const char *)&(struct struct_name)
+
+#define FLAVORS_DATA(struct_name) STRUCT_TO_CHAR_PTR(struct_name) { \
+ .a = 42, \
+ .b = 0xc001, \
+ .c = 0xbeef, \
+}
+
+#define FLAVORS_CASE_COMMON(name) \
+ .case_name = #name, \
+ .bpf_obj_file = "test_core_reloc_flavors.o", \
+ .btf_src_file = "btf__core_reloc_" #name ".o" \
+
+#define FLAVORS_CASE(name) { \
+ FLAVORS_CASE_COMMON(name), \
+ .input = FLAVORS_DATA(core_reloc_##name), \
+ .input_len = sizeof(struct core_reloc_##name), \
+ .output = FLAVORS_DATA(core_reloc_flavors), \
+ .output_len = sizeof(struct core_reloc_flavors), \
+}
+
+#define FLAVORS_ERR_CASE(name) { \
+ FLAVORS_CASE_COMMON(name), \
+ .fails = true, \
+}
struct core_reloc_test_case {
const char *case_name;
@@ -23,6 +50,13 @@ static struct core_reloc_test_case test_cases[] = {
.output = "\1", /* true */
.output_len = 1,
},
+
+ /* validate BPF program can use multiple flavors to match against
+ * single target BTF type
+ */
+ FLAVORS_CASE(flavors),
+
+ FLAVORS_ERR_CASE(flavors__err_wrong_name),
};
struct data {
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_flavors.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_flavors.c
new file mode 100644
index 000000000000..b74455b91227
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_flavors.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_flavors x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_flavors__err_wrong_name.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_flavors__err_wrong_name.c
new file mode 100644
index 000000000000..7b6035f86ee6
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_flavors__err_wrong_name.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_flavors__err_wrong_name x) {}
diff --git a/tools/testing/selftests/bpf/progs/core_reloc_types.h b/tools/testing/selftests/bpf/progs/core_reloc_types.h
new file mode 100644
index 000000000000..33b0c6a61912
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/core_reloc_types.h
@@ -0,0 +1,15 @@
+/*
+ * FLAVORS
+ */
+struct core_reloc_flavors {
+ int a;
+ int b;
+ int c;
+};
+
+/* this is not a flavor, as it doesn't have triple underscore */
+struct core_reloc_flavors__err_wrong_name {
+ int a;
+ int b;
+ int c;
+};
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c b/tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c
new file mode 100644
index 000000000000..9fda73e87972
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c
@@ -0,0 +1,62 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2019 Facebook
+
+#include <linux/bpf.h>
+#include <stdint.h>
+#include "bpf_helpers.h"
+
+char _license[] SEC("license") = "GPL";
+
+static volatile struct data {
+ char in[256];
+ char out[256];
+} data;
+
+struct core_reloc_flavors {
+ int a;
+ int b;
+ int c;
+};
+
+/* local flavor with reversed layout */
+struct core_reloc_flavors___reversed {
+ int c;
+ int b;
+ int a;
+};
+
+/* local flavor with nested/overlapping layout */
+struct core_reloc_flavors___weird {
+ struct {
+ int b;
+ };
+ /* a and c overlap in local flavor, but this should still work
+ * correctly with target original flavor
+ */
+ union {
+ int a;
+ int c;
+ };
+};
+
+SEC("raw_tracepoint/sys_enter")
+int test_core_flavors(void *ctx)
+{
+ struct core_reloc_flavors *in_orig = (void *)&data.in;
+ struct core_reloc_flavors___reversed *in_rev = (void *)&data.in;
+ struct core_reloc_flavors___weird *in_weird = (void *)&data.in;
+ struct core_reloc_flavors *out = (void *)&data.out;
+
+ /* read a using weird layout */
+ if (BPF_CORE_READ(&out->a, &in_weird->a))
+ return 1;
+ /* read b using reversed layout */
+ if (BPF_CORE_READ(&out->b, &in_rev->b))
+ return 1;
+ /* read c using original layout */
+ if (BPF_CORE_READ(&out->c, &in_orig->c))
+ return 1;
+
+ return 0;
+}
+
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 08/14] selftests/bpf: add CO-RE relocs nesting tests
From: Andrii Nakryiko @ 2019-08-07 5:38 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Add a bunch of test validating correct handling of nested
structs/unions.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
---
.../selftests/bpf/prog_tests/core_reloc.c | 39 +++
.../bpf/progs/btf__core_reloc_nesting.c | 3 +
.../btf__core_reloc_nesting___anon_embed.c | 3 +
...f__core_reloc_nesting___dup_compat_types.c | 5 +
...core_reloc_nesting___err_array_container.c | 3 +
...tf__core_reloc_nesting___err_array_field.c | 3 +
...e_reloc_nesting___err_dup_incompat_types.c | 4 +
...re_reloc_nesting___err_missing_container.c | 3 +
...__core_reloc_nesting___err_missing_field.c | 3 +
..._reloc_nesting___err_nonstruct_container.c | 3 +
...e_reloc_nesting___err_partial_match_dups.c | 4 +
.../btf__core_reloc_nesting___err_too_deep.c | 3 +
.../btf__core_reloc_nesting___extra_nesting.c | 3 +
..._core_reloc_nesting___struct_union_mixup.c | 3 +
.../selftests/bpf/progs/core_reloc_types.h | 293 ++++++++++++++++++
.../bpf/progs/test_core_reloc_nesting.c | 46 +++
16 files changed, 421 insertions(+)
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___anon_embed.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___dup_compat_types.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_array_container.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_array_field.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_dup_incompat_types.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_missing_container.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_missing_field.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_nonstruct_container.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_partial_match_dups.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_too_deep.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___extra_nesting.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___struct_union_mixup.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c
diff --git a/tools/testing/selftests/bpf/prog_tests/core_reloc.c b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
index 995d5e9fd4c3..2c0e695f958f 100644
--- a/tools/testing/selftests/bpf/prog_tests/core_reloc.c
+++ b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
@@ -28,6 +28,29 @@
.fails = true, \
}
+#define NESTING_DATA(struct_name) STRUCT_TO_CHAR_PTR(struct_name) { \
+ .a = { .a = { .a = 42 } }, \
+ .b = { .b = { .b = 0xc001 } }, \
+}
+
+#define NESTING_CASE_COMMON(name) \
+ .case_name = #name, \
+ .bpf_obj_file = "test_core_reloc_nesting.o", \
+ .btf_src_file = "btf__core_reloc_" #name ".o"
+
+#define NESTING_CASE(name) { \
+ NESTING_CASE_COMMON(name), \
+ .input = NESTING_DATA(core_reloc_##name), \
+ .input_len = sizeof(struct core_reloc_##name), \
+ .output = NESTING_DATA(core_reloc_nesting), \
+ .output_len = sizeof(struct core_reloc_nesting) \
+}
+
+#define NESTING_ERR_CASE(name) { \
+ NESTING_CASE_COMMON(name), \
+ .fails = true, \
+}
+
struct core_reloc_test_case {
const char *case_name;
const char *bpf_obj_file;
@@ -57,6 +80,22 @@ static struct core_reloc_test_case test_cases[] = {
FLAVORS_CASE(flavors),
FLAVORS_ERR_CASE(flavors__err_wrong_name),
+
+ /* various struct/enum nesting and resolution scenarios */
+ NESTING_CASE(nesting),
+ NESTING_CASE(nesting___anon_embed),
+ NESTING_CASE(nesting___struct_union_mixup),
+ NESTING_CASE(nesting___extra_nesting),
+ NESTING_CASE(nesting___dup_compat_types),
+
+ NESTING_ERR_CASE(nesting___err_missing_field),
+ NESTING_ERR_CASE(nesting___err_array_field),
+ NESTING_ERR_CASE(nesting___err_missing_container),
+ NESTING_ERR_CASE(nesting___err_nonstruct_container),
+ NESTING_ERR_CASE(nesting___err_array_container),
+ NESTING_ERR_CASE(nesting___err_dup_incompat_types),
+ NESTING_ERR_CASE(nesting___err_partial_match_dups),
+ NESTING_ERR_CASE(nesting___err_too_deep),
};
struct data {
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting.c
new file mode 100644
index 000000000000..4480fcc0f183
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_nesting x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___anon_embed.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___anon_embed.c
new file mode 100644
index 000000000000..13e108f76ece
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___anon_embed.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_nesting___anon_embed x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___dup_compat_types.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___dup_compat_types.c
new file mode 100644
index 000000000000..76b54fda5fbb
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___dup_compat_types.c
@@ -0,0 +1,5 @@
+#include "core_reloc_types.h"
+
+void f1(struct core_reloc_nesting___dup_compat_types x) {}
+void f2(struct core_reloc_nesting___dup_compat_types__2 x) {}
+void f3(struct core_reloc_nesting___dup_compat_types__3 x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_array_container.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_array_container.c
new file mode 100644
index 000000000000..975fb95db810
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_array_container.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_nesting___err_array_container x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_array_field.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_array_field.c
new file mode 100644
index 000000000000..ad66c67e7980
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_array_field.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_nesting___err_array_field x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_dup_incompat_types.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_dup_incompat_types.c
new file mode 100644
index 000000000000..35c5f8da6812
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_dup_incompat_types.c
@@ -0,0 +1,4 @@
+#include "core_reloc_types.h"
+
+void f1(struct core_reloc_nesting___err_dup_incompat_types__1 x) {}
+void f2(struct core_reloc_nesting___err_dup_incompat_types__2 x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_missing_container.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_missing_container.c
new file mode 100644
index 000000000000..142e332041db
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_missing_container.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_nesting___err_missing_container x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_missing_field.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_missing_field.c
new file mode 100644
index 000000000000..efcae167fab9
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_missing_field.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_nesting___err_missing_field x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_nonstruct_container.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_nonstruct_container.c
new file mode 100644
index 000000000000..97aaaedd8ada
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_nonstruct_container.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_nesting___err_nonstruct_container x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_partial_match_dups.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_partial_match_dups.c
new file mode 100644
index 000000000000..ffde35086e90
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_partial_match_dups.c
@@ -0,0 +1,4 @@
+#include "core_reloc_types.h"
+
+void f1(struct core_reloc_nesting___err_partial_match_dups__a x) {}
+void f2(struct core_reloc_nesting___err_partial_match_dups__b x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_too_deep.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_too_deep.c
new file mode 100644
index 000000000000..39a2fadd8e95
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_too_deep.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_nesting___err_too_deep x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___extra_nesting.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___extra_nesting.c
new file mode 100644
index 000000000000..a09d9dfb20df
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___extra_nesting.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_nesting___extra_nesting x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___struct_union_mixup.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___struct_union_mixup.c
new file mode 100644
index 000000000000..3d8a1a74012f
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___struct_union_mixup.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_nesting___struct_union_mixup x) {}
diff --git a/tools/testing/selftests/bpf/progs/core_reloc_types.h b/tools/testing/selftests/bpf/progs/core_reloc_types.h
index 33b0c6a61912..340ee2bcd463 100644
--- a/tools/testing/selftests/bpf/progs/core_reloc_types.h
+++ b/tools/testing/selftests/bpf/progs/core_reloc_types.h
@@ -13,3 +13,296 @@ struct core_reloc_flavors__err_wrong_name {
int b;
int c;
};
+
+/*
+ * NESTING
+ */
+/* original set up, used to record relocations in BPF program */
+struct core_reloc_nesting_substruct {
+ int a;
+};
+
+union core_reloc_nesting_subunion {
+ int b;
+};
+
+struct core_reloc_nesting {
+ union {
+ struct core_reloc_nesting_substruct a;
+ } a;
+ struct {
+ union core_reloc_nesting_subunion b;
+ } b;
+};
+
+/* inlined anonymous struct/union instead of named structs in original */
+struct core_reloc_nesting___anon_embed {
+ int __just_for_padding;
+ union {
+ struct {
+ int a;
+ } a;
+ } a;
+ struct {
+ union {
+ int b;
+ } b;
+ } b;
+};
+
+/* different mix of nested structs/unions than in original */
+struct core_reloc_nesting___struct_union_mixup {
+ int __a;
+ struct {
+ int __a;
+ union {
+ char __a;
+ int a;
+ } a;
+ } a;
+ int __b;
+ union {
+ int __b;
+ union {
+ char __b;
+ int b;
+ } b;
+ } b;
+};
+
+/* extra anon structs/unions, but still valid a.a.a and b.b.b accessors */
+struct core_reloc_nesting___extra_nesting {
+ int __padding;
+ struct {
+ struct {
+ struct {
+ struct {
+ union {
+ int a;
+ } a;
+ };
+ };
+ } a;
+ int __some_more;
+ struct {
+ union {
+ union {
+ union {
+ struct {
+ int b;
+ };
+ } b;
+ };
+ } b;
+ };
+ };
+};
+
+/* three flavors of same struct with different structure but same layout for
+ * a.a.a and b.b.b, thus successfully resolved and relocatable */
+struct core_reloc_nesting___dup_compat_types {
+ char __just_for_padding;
+ /* 3 more bytes of padding */
+ struct {
+ struct {
+ int a; /* offset 4 */
+ } a;
+ } a;
+ long long __more_padding;
+ struct {
+ struct {
+ int b; /* offset 16 */
+ } b;
+ } b;
+};
+
+struct core_reloc_nesting___dup_compat_types__2 {
+ int __aligned_padding;
+ struct {
+ int __trickier_noop[0];
+ struct {
+ char __some_more_noops[0];
+ int a; /* offset 4 */
+ } a;
+ } a;
+ int __more_padding;
+ struct {
+ struct {
+ struct {
+ int __critical_padding;
+ int b; /* offset 16 */
+ } b;
+ int __does_not_matter;
+ };
+ } b;
+ int __more_irrelevant_stuff;
+};
+
+struct core_reloc_nesting___dup_compat_types__3 {
+ char __correct_padding[4];
+ struct {
+ struct {
+ int a; /* offset 4 */
+ } a;
+ } a;
+ /* 8 byte padding due to next struct's alignment */
+ struct {
+ struct {
+ int b;
+ } b;
+ } b __attribute__((aligned(16)));
+};
+
+/* b.b.b field is missing */
+struct core_reloc_nesting___err_missing_field {
+ struct {
+ struct {
+ int a;
+ } a;
+ } a;
+ struct {
+ struct {
+ int x;
+ } b;
+ } b;
+};
+
+/* b.b.b field is an array of integers instead of plain int */
+struct core_reloc_nesting___err_array_field {
+ struct {
+ struct {
+ int a;
+ } a;
+ } a;
+ struct {
+ struct {
+ int b[1];
+ } b;
+ } b;
+};
+
+/* middle b container is missing */
+struct core_reloc_nesting___err_missing_container {
+ struct {
+ struct {
+ int a;
+ } a;
+ } a;
+ struct {
+ int x;
+ } b;
+};
+
+/* middle b container is referenced through pointer instead of being embedded */
+struct core_reloc_nesting___err_nonstruct_container {
+ struct {
+ struct {
+ int a;
+ } a;
+ } a;
+ struct {
+ struct {
+ int b;
+ } *b;
+ } b;
+};
+
+/* middle b container is an array of structs instead of plain struct */
+struct core_reloc_nesting___err_array_container {
+ struct {
+ struct {
+ int a;
+ } a;
+ } a;
+ struct {
+ struct {
+ int b;
+ } b[1];
+ } b;
+};
+
+/* two flavors of same struct with incompatible layout for b.b.b */
+struct core_reloc_nesting___err_dup_incompat_types__1 {
+ struct {
+ struct {
+ int a; /* offset 0 */
+ } a;
+ } a;
+ struct {
+ struct {
+ int b; /* offset 4 */
+ } b;
+ } b;
+};
+
+struct core_reloc_nesting___err_dup_incompat_types__2 {
+ struct {
+ struct {
+ int a; /* offset 0 */
+ } a;
+ } a;
+ int __extra_padding;
+ struct {
+ struct {
+ int b; /* offset 8 (!) */
+ } b;
+ } b;
+};
+
+/* two flavors of same struct having one of a.a.a and b.b.b, but not both */
+struct core_reloc_nesting___err_partial_match_dups__a {
+ struct {
+ struct {
+ int a;
+ } a;
+ } a;
+};
+
+struct core_reloc_nesting___err_partial_match_dups__b {
+ struct {
+ struct {
+ int b;
+ } b;
+ } b;
+};
+
+struct core_reloc_nesting___err_too_deep {
+ struct {
+ struct {
+ int a;
+ } a;
+ } a;
+ /* 65 levels of nestedness for b.b.b */
+ struct {
+ struct {
+ struct { struct { struct { struct { struct {
+ struct { struct { struct { struct { struct {
+ struct { struct { struct { struct { struct {
+ struct { struct { struct { struct { struct {
+ struct { struct { struct { struct { struct {
+ struct { struct { struct { struct { struct {
+ struct { struct { struct { struct { struct {
+ struct { struct { struct { struct { struct {
+ struct { struct { struct { struct { struct {
+ struct { struct { struct { struct { struct {
+ struct { struct { struct { struct { struct {
+ struct { struct { struct { struct { struct {
+ /* this one is one too much */
+ struct {
+ int b;
+ };
+ }; }; }; }; };
+ }; }; }; }; };
+ }; }; }; }; };
+ }; }; }; }; };
+ }; }; }; }; };
+ }; }; }; }; };
+ }; }; }; }; };
+ }; }; }; }; };
+ }; }; }; }; };
+ }; }; }; }; };
+ }; }; }; }; };
+ }; }; }; }; };
+ } b;
+ } b;
+};
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c b/tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c
new file mode 100644
index 000000000000..3ca30cec2b39
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c
@@ -0,0 +1,46 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2019 Facebook
+
+#include <linux/bpf.h>
+#include <stdint.h>
+#include "bpf_helpers.h"
+
+char _license[] SEC("license") = "GPL";
+
+static volatile struct data {
+ char in[256];
+ char out[256];
+} data;
+
+struct core_reloc_nesting_substruct {
+ int a;
+};
+
+union core_reloc_nesting_subunion {
+ int b;
+};
+
+/* int a.a.a and b.b.b accesses */
+struct core_reloc_nesting {
+ union {
+ struct core_reloc_nesting_substruct a;
+ } a;
+ struct {
+ union core_reloc_nesting_subunion b;
+ } b;
+};
+
+SEC("raw_tracepoint/sys_enter")
+int test_core_nesting(void *ctx)
+{
+ struct core_reloc_nesting *in = (void *)&data.in;
+ struct core_reloc_nesting *out = (void *)&data.out;
+
+ if (BPF_CORE_READ(&out->a.a.a, &in->a.a.a))
+ return 1;
+ if (BPF_CORE_READ(&out->b.b.b, &in->b.b.b))
+ return 1;
+
+ return 0;
+}
+
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 05/14] selftests/bpf: add BPF_CORE_READ relocatable read macro
From: Andrii Nakryiko @ 2019-08-07 5:37 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Add BPF_CORE_READ macro used in tests to do bpf_core_read(), which
automatically captures offset relocation.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
---
tools/testing/selftests/bpf/bpf_helpers.h | 20 ++++++++++++++++++++
1 file changed, 20 insertions(+)
diff --git a/tools/testing/selftests/bpf/bpf_helpers.h b/tools/testing/selftests/bpf/bpf_helpers.h
index 120aa86c58d3..8b503ea142f0 100644
--- a/tools/testing/selftests/bpf/bpf_helpers.h
+++ b/tools/testing/selftests/bpf/bpf_helpers.h
@@ -504,4 +504,24 @@ struct pt_regs;
(void *)(PT_REGS_FP(ctx) + sizeof(ip))); })
#endif
+/*
+ * BPF_CORE_READ abstracts away bpf_probe_read() call and captures offset
+ * relocation for source address using __builtin_preserve_access_index()
+ * built-in, provided by Clang.
+ *
+ * __builtin_preserve_access_index() takes as an argument an expression of
+ * taking an address of a field within struct/union. It makes compiler emit
+ * a relocation, which records BTF type ID describing root struct/union and an
+ * accessor string which describes exact embedded field that was used to take
+ * an address. See detailed description of this relocation format and
+ * semantics in comments to struct bpf_offset_reloc in libbpf_internal.h.
+ *
+ * This relocation allows libbpf to adjust BPF instruction to use correct
+ * actual field offset, based on target kernel BTF type that matches original
+ * (local) BTF, used to record relocation.
+ */
+#define BPF_CORE_READ(dst, src) \
+ bpf_probe_read((dst), sizeof(*(src)), \
+ __builtin_preserve_access_index(src))
+
#endif
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 04/14] libbpf: implement BPF CO-RE offset relocation algorithm
From: Andrii Nakryiko @ 2019-08-07 5:37 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
This patch implements the core logic for BPF CO-RE offsets relocations.
Every instruction that needs to be relocated has corresponding
bpf_offset_reloc as part of BTF.ext. Relocations are performed by trying
to match recorded "local" relocation spec against potentially many
compatible "target" types, creating corresponding spec. Details of the
algorithm are noted in corresponding comments in the code.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
---
tools/lib/bpf/libbpf.c | 881 ++++++++++++++++++++++++++++++++++++++++-
tools/lib/bpf/libbpf.h | 1 +
2 files changed, 864 insertions(+), 18 deletions(-)
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index fafcc75b33b1..3528392c0f7f 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -38,6 +38,7 @@
#include <sys/stat.h>
#include <sys/types.h>
#include <sys/vfs.h>
+#include <sys/utsname.h>
#include <tools/libc_compat.h>
#include <libelf.h>
#include <gelf.h>
@@ -47,6 +48,7 @@
#include "btf.h"
#include "str_error.h"
#include "libbpf_internal.h"
+#include "hashmap.h"
#ifndef EM_BPF
#define EM_BPF 247
@@ -1015,23 +1017,21 @@ static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict)
return 0;
}
-static const struct btf_type *skip_mods_and_typedefs(const struct btf *btf,
- __u32 id)
+static const struct btf_type *
+skip_mods_and_typedefs(const struct btf *btf, __u32 id, __u32 *res_id)
{
const struct btf_type *t = btf__type_by_id(btf, id);
- while (true) {
- switch (BTF_INFO_KIND(t->info)) {
- case BTF_KIND_VOLATILE:
- case BTF_KIND_CONST:
- case BTF_KIND_RESTRICT:
- case BTF_KIND_TYPEDEF:
- t = btf__type_by_id(btf, t->type);
- break;
- default:
- return t;
- }
+ if (res_id)
+ *res_id = id;
+
+ while (btf_is_mod(t) || btf_is_typedef(t)) {
+ if (res_id)
+ *res_id = t->type;
+ t = btf__type_by_id(btf, t->type);
}
+
+ return t;
}
/*
@@ -1044,7 +1044,7 @@ static const struct btf_type *skip_mods_and_typedefs(const struct btf *btf,
static bool get_map_field_int(const char *map_name, const struct btf *btf,
const struct btf_type *def,
const struct btf_member *m, __u32 *res) {
- const struct btf_type *t = skip_mods_and_typedefs(btf, m->type);
+ const struct btf_type *t = skip_mods_and_typedefs(btf, m->type, NULL);
const char *name = btf__name_by_offset(btf, m->name_off);
const struct btf_array *arr_info;
const struct btf_type *arr_t;
@@ -1110,7 +1110,7 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
return -EOPNOTSUPP;
}
- def = skip_mods_and_typedefs(obj->btf, var->type);
+ def = skip_mods_and_typedefs(obj->btf, var->type, NULL);
if (!btf_is_struct(def)) {
pr_warning("map '%s': unexpected def kind %u.\n",
map_name, btf_kind(var));
@@ -2290,6 +2290,844 @@ bpf_program_reloc_btf_ext(struct bpf_program *prog, struct bpf_object *obj,
return 0;
}
+#define BPF_CORE_SPEC_MAX_LEN 64
+
+/* represents BPF CO-RE field or array element accessor */
+struct bpf_core_accessor {
+ __u32 type_id; /* struct/union type or array element type */
+ __u32 idx; /* field index or array index */
+ const char *name; /* field name or NULL for array accessor */
+};
+
+struct bpf_core_spec {
+ const struct btf *btf;
+ /* high-level spec: named fields and array indices only */
+ struct bpf_core_accessor spec[BPF_CORE_SPEC_MAX_LEN];
+ /* high-level spec length */
+ int len;
+ /* raw, low-level spec: 1-to-1 with accessor spec string */
+ int raw_spec[BPF_CORE_SPEC_MAX_LEN];
+ /* raw spec length */
+ int raw_len;
+ /* field byte offset represented by spec */
+ __u32 offset;
+};
+
+static bool str_is_empty(const char *s)
+{
+ return !s || !s[0];
+}
+
+/*
+ * Turn bpf_offset_reloc into a low- and high-level spec representation,
+ * validating correctness along the way, as well as calculating resulting
+ * field offset (in bytes), specified by accessor string. Low-level spec
+ * captures every single level of nestedness, including traversing anonymous
+ * struct/union members. High-level one only captures semantically meaningful
+ * "turning points": named fields and array indicies.
+ * E.g., for this case:
+ *
+ * struct sample {
+ * int __unimportant;
+ * struct {
+ * int __1;
+ * int __2;
+ * int a[7];
+ * };
+ * };
+ *
+ * struct sample *s = ...;
+ *
+ * int x = &s->a[3]; // access string = '0:1:2:3'
+ *
+ * Low-level spec has 1:1 mapping with each element of access string (it's
+ * just a parsed access string representation): [0, 1, 2, 3].
+ *
+ * High-level spec will capture only 3 points:
+ * - intial zero-index access by pointer (&s->... is the same as &s[0]...);
+ * - field 'a' access (corresponds to '2' in low-level spec);
+ * - array element #3 access (corresponds to '3' in low-level spec).
+ *
+ */
+static int bpf_core_spec_parse(const struct btf *btf,
+ __u32 type_id,
+ const char *spec_str,
+ struct bpf_core_spec *spec)
+{
+ int access_idx, parsed_len, i;
+ const struct btf_type *t;
+ const char *name;
+ __u32 id;
+ __s64 sz;
+
+ if (str_is_empty(spec_str) || *spec_str == ':')
+ return -EINVAL;
+
+ memset(spec, 0, sizeof(*spec));
+ spec->btf = btf;
+
+ /* parse spec_str="0:1:2:3:4" into array raw_spec=[0, 1, 2, 3, 4] */
+ while (*spec_str) {
+ if (*spec_str == ':')
+ ++spec_str;
+ if (sscanf(spec_str, "%d%n", &access_idx, &parsed_len) != 1)
+ return -EINVAL;
+ if (spec->raw_len == BPF_CORE_SPEC_MAX_LEN)
+ return -E2BIG;
+ spec_str += parsed_len;
+ spec->raw_spec[spec->raw_len++] = access_idx;
+ }
+
+ if (spec->raw_len == 0)
+ return -EINVAL;
+
+ /* first spec value is always reloc type array index */
+ t = skip_mods_and_typedefs(btf, type_id, &id);
+ if (!t)
+ return -EINVAL;
+
+ access_idx = spec->raw_spec[0];
+ spec->spec[0].type_id = id;
+ spec->spec[0].idx = access_idx;
+ spec->len++;
+
+ sz = btf__resolve_size(btf, id);
+ if (sz < 0)
+ return sz;
+ spec->offset = access_idx * sz;
+
+ for (i = 1; i < spec->raw_len; i++) {
+ t = skip_mods_and_typedefs(btf, id, &id);
+ if (!t)
+ return -EINVAL;
+
+ access_idx = spec->raw_spec[i];
+
+ if (btf_is_composite(t)) {
+ const struct btf_member *m;
+ __u32 offset;
+
+ if (access_idx >= btf_vlen(t))
+ return -EINVAL;
+ if (btf_member_bitfield_size(t, access_idx))
+ return -EINVAL;
+
+ offset = btf_member_bit_offset(t, access_idx);
+ if (offset % 8)
+ return -EINVAL;
+ spec->offset += offset / 8;
+
+ m = btf_members(t) + access_idx;
+ if (m->name_off) {
+ name = btf__name_by_offset(btf, m->name_off);
+ if (str_is_empty(name))
+ return -EINVAL;
+
+ spec->spec[spec->len].type_id = id;
+ spec->spec[spec->len].idx = access_idx;
+ spec->spec[spec->len].name = name;
+ spec->len++;
+ }
+
+ id = m->type;
+ } else if (btf_is_array(t)) {
+ const struct btf_array *a = btf_array(t);
+
+ t = skip_mods_and_typedefs(btf, a->type, &id);
+ if (!t || access_idx >= a->nelems)
+ return -EINVAL;
+
+ spec->spec[spec->len].type_id = id;
+ spec->spec[spec->len].idx = access_idx;
+ spec->len++;
+
+ sz = btf__resolve_size(btf, id);
+ if (sz < 0)
+ return sz;
+ spec->offset += access_idx * sz;
+ } else {
+ pr_warning("relo for [%u] %s (at idx %d) captures type [%d] of unexpected kind %d\n",
+ type_id, spec_str, i, id, btf_kind(t));
+ return -EINVAL;
+ }
+ }
+
+ return 0;
+}
+
+static bool bpf_core_is_flavor_sep(const char *s)
+{
+ /* check X___Y name pattern, where X and Y are not underscores */
+ return s[0] != '_' && /* X */
+ s[1] == '_' && s[2] == '_' && s[3] == '_' && /* ___ */
+ s[4] != '_'; /* Y */
+}
+
+/* Given 'some_struct_name___with_flavor' return the length of a name prefix
+ * before last triple underscore. Struct name part after last triple
+ * underscore is ignored by BPF CO-RE relocation during relocation matching.
+ */
+static size_t bpf_core_essential_name_len(const char *name)
+{
+ size_t n = strlen(name);
+ int i;
+
+ for (i = n - 5; i >= 0; i--) {
+ if (bpf_core_is_flavor_sep(name + i))
+ return i + 1;
+ }
+ return n;
+}
+
+/* dynamically sized list of type IDs */
+struct ids_vec {
+ __u32 *data;
+ int len;
+};
+
+static void bpf_core_free_cands(struct ids_vec *cand_ids)
+{
+ free(cand_ids->data);
+ free(cand_ids);
+}
+
+static struct ids_vec *bpf_core_find_cands(const struct btf *local_btf,
+ __u32 local_type_id,
+ const struct btf *targ_btf)
+{
+ size_t local_essent_len, targ_essent_len;
+ const char *local_name, *targ_name;
+ const struct btf_type *t;
+ struct ids_vec *cand_ids;
+ __u32 *new_ids;
+ int i, err, n;
+
+ t = btf__type_by_id(local_btf, local_type_id);
+ if (!t)
+ return ERR_PTR(-EINVAL);
+
+ local_name = btf__name_by_offset(local_btf, t->name_off);
+ if (str_is_empty(local_name))
+ return ERR_PTR(-EINVAL);
+ local_essent_len = bpf_core_essential_name_len(local_name);
+
+ cand_ids = calloc(1, sizeof(*cand_ids));
+ if (!cand_ids)
+ return ERR_PTR(-ENOMEM);
+
+ n = btf__get_nr_types(targ_btf);
+ for (i = 1; i <= n; i++) {
+ t = btf__type_by_id(targ_btf, i);
+ targ_name = btf__name_by_offset(targ_btf, t->name_off);
+ if (str_is_empty(targ_name))
+ continue;
+
+ targ_essent_len = bpf_core_essential_name_len(targ_name);
+ if (targ_essent_len != local_essent_len)
+ continue;
+
+ if (strncmp(local_name, targ_name, local_essent_len) == 0) {
+ pr_debug("[%d] %s: found candidate [%d] %s\n",
+ local_type_id, local_name, i, targ_name);
+ new_ids = realloc(cand_ids->data, cand_ids->len + 1);
+ if (!new_ids) {
+ err = -ENOMEM;
+ goto err_out;
+ }
+ cand_ids->data = new_ids;
+ cand_ids->data[cand_ids->len++] = i;
+ }
+ }
+ return cand_ids;
+err_out:
+ bpf_core_free_cands(cand_ids);
+ return ERR_PTR(err);
+}
+
+/* Check two types for compatibility, skipping const/volatile/restrict and
+ * typedefs, to ensure we are relocating offset to the compatible entities:
+ * - any two STRUCTs/UNIONs are compatible and can be mixed;
+ * - any two FWDs are compatible;
+ * - any two PTRs are always compatible;
+ * - for ENUMs, check sizes, names are ignored;
+ * - for INT, size and bitness should match, signedness is ignored;
+ * - for ARRAY, dimensionality is ignored, element types are checked for
+ * compatibility recursively;
+ * - everything else shouldn't be ever a target of relocation.
+ * These rules are not set in stone and probably will be adjusted as we get
+ * more experience with using BPF CO-RE relocations.
+ */
+static int bpf_core_fields_are_compat(const struct btf *local_btf,
+ __u32 local_id,
+ const struct btf *targ_btf,
+ __u32 targ_id)
+{
+ const struct btf_type *local_type, *targ_type;
+
+recur:
+ local_type = skip_mods_and_typedefs(local_btf, local_id, &local_id);
+ targ_type = skip_mods_and_typedefs(targ_btf, targ_id, &targ_id);
+ if (!local_type || !targ_type)
+ return -EINVAL;
+
+ if (btf_is_composite(local_type) && btf_is_composite(targ_type))
+ return 1;
+ if (btf_kind(local_type) != btf_kind(targ_type))
+ return 0;
+
+ switch (btf_kind(local_type)) {
+ case BTF_KIND_FWD:
+ case BTF_KIND_PTR:
+ return 1;
+ case BTF_KIND_ENUM:
+ return local_type->size == targ_type->size;
+ case BTF_KIND_INT:
+ return btf_int_offset(local_type) == 0 &&
+ btf_int_offset(targ_type) == 0 &&
+ local_type->size == targ_type->size &&
+ btf_int_bits(local_type) == btf_int_bits(targ_type);
+ case BTF_KIND_ARRAY:
+ local_id = btf_array(local_type)->type;
+ targ_id = btf_array(targ_type)->type;
+ goto recur;
+ default:
+ pr_warning("unexpected kind %d relocated, local [%d], target [%d]\n",
+ btf_kind(local_type), local_id, targ_id);
+ return 0;
+ }
+}
+
+/*
+ * Given single high-level named field accessor in local type, find
+ * corresponding high-level accessor for a target type. Along the way,
+ * maintain low-level spec for target as well. Also keep updating target
+ * offset.
+ *
+ * Searching is performed through recursive exhaustive enumeration of all
+ * fields of a struct/union. If there are any anonymous (embedded)
+ * structs/unions, they are recursively searched as well. If field with
+ * desired name is found, check compatibility between local and target types,
+ * before returning result.
+ *
+ * 1 is returned, if field is found.
+ * 0 is returned if no compatible field is found.
+ * <0 is returned on error.
+ */
+static int bpf_core_match_member(const struct btf *local_btf,
+ const struct bpf_core_accessor *local_acc,
+ const struct btf *targ_btf,
+ __u32 targ_id,
+ struct bpf_core_spec *spec,
+ __u32 *next_targ_id)
+{
+ const struct btf_type *local_type, *targ_type;
+ const struct btf_member *local_member, *m;
+ const char *local_name, *targ_name;
+ __u32 local_id;
+ int i, n, found;
+
+ targ_type = skip_mods_and_typedefs(targ_btf, targ_id, &targ_id);
+ if (!targ_type)
+ return -EINVAL;
+ if (!btf_is_composite(targ_type))
+ return 0;
+
+ local_id = local_acc->type_id;
+ local_type = btf__type_by_id(local_btf, local_id);
+ local_member = btf_members(local_type) + local_acc->idx;
+ local_name = btf__name_by_offset(local_btf, local_member->name_off);
+
+ n = btf_vlen(targ_type);
+ m = btf_members(targ_type);
+ for (i = 0; i < n; i++, m++) {
+ __u32 offset;
+
+ /* bitfield relocations not supported */
+ if (btf_member_bitfield_size(targ_type, i))
+ continue;
+ offset = btf_member_bit_offset(targ_type, i);
+ if (offset % 8)
+ continue;
+
+ /* too deep struct/union/array nesting */
+ if (spec->raw_len == BPF_CORE_SPEC_MAX_LEN)
+ return -E2BIG;
+
+ /* speculate this member will be the good one */
+ spec->offset += offset / 8;
+ spec->raw_spec[spec->raw_len++] = i;
+
+ targ_name = btf__name_by_offset(targ_btf, m->name_off);
+ if (str_is_empty(targ_name)) {
+ /* embedded struct/union, we need to go deeper */
+ found = bpf_core_match_member(local_btf, local_acc,
+ targ_btf, m->type,
+ spec, next_targ_id);
+ if (found) /* either found or error */
+ return found;
+ } else if (strcmp(local_name, targ_name) == 0) {
+ /* matching named field */
+ struct bpf_core_accessor *targ_acc;
+
+ targ_acc = &spec->spec[spec->len++];
+ targ_acc->type_id = targ_id;
+ targ_acc->idx = i;
+ targ_acc->name = targ_name;
+
+ *next_targ_id = m->type;
+ found = bpf_core_fields_are_compat(local_btf,
+ local_member->type,
+ targ_btf, m->type);
+ if (!found)
+ spec->len--; /* pop accessor */
+ return found;
+ }
+ /* member turned out not to be what we looked for */
+ spec->offset -= offset / 8;
+ spec->raw_len--;
+ }
+
+ return 0;
+}
+
+/*
+ * Try to match local spec to a target type and, if successful, produce full
+ * target spec (high-level, low-level + offset).
+ */
+static int bpf_core_spec_match(struct bpf_core_spec *local_spec,
+ const struct btf *targ_btf, __u32 targ_id,
+ struct bpf_core_spec *targ_spec)
+{
+ const struct btf_type *targ_type;
+ const struct bpf_core_accessor *local_acc;
+ struct bpf_core_accessor *targ_acc;
+ int i, sz, matched;
+
+ memset(targ_spec, 0, sizeof(*targ_spec));
+ targ_spec->btf = targ_btf;
+
+ local_acc = &local_spec->spec[0];
+ targ_acc = &targ_spec->spec[0];
+
+ for (i = 0; i < local_spec->len; i++, local_acc++, targ_acc++) {
+ targ_type = skip_mods_and_typedefs(targ_spec->btf, targ_id,
+ &targ_id);
+ if (!targ_type)
+ return -EINVAL;
+
+ if (local_acc->name) {
+ matched = bpf_core_match_member(local_spec->btf,
+ local_acc,
+ targ_btf, targ_id,
+ targ_spec, &targ_id);
+ if (matched <= 0)
+ return matched;
+ } else {
+ /* for i=0, targ_id is already treated as array element
+ * type (because it's the original struct), for others
+ * we should find array element type first
+ */
+ if (i > 0) {
+ const struct btf_array *a;
+
+ if (!btf_is_array(targ_type))
+ return 0;
+
+ a = btf_array(targ_type);
+ if (local_acc->idx >= a->nelems)
+ return 0;
+ if (!skip_mods_and_typedefs(targ_btf, a->type,
+ &targ_id))
+ return -EINVAL;
+ }
+
+ /* too deep struct/union/array nesting */
+ if (targ_spec->raw_len == BPF_CORE_SPEC_MAX_LEN)
+ return -E2BIG;
+
+ targ_acc->type_id = targ_id;
+ targ_acc->idx = local_acc->idx;
+ targ_acc->name = NULL;
+ targ_spec->len++;
+ targ_spec->raw_spec[targ_spec->raw_len] = targ_acc->idx;
+ targ_spec->raw_len++;
+
+ sz = btf__resolve_size(targ_btf, targ_id);
+ if (sz < 0)
+ return sz;
+ targ_spec->offset += local_acc->idx * sz;
+ }
+ }
+
+ return 1;
+}
+
+/*
+ * Patch relocatable BPF instruction.
+ * Expected insn->imm value is provided for validation, as well as the new
+ * relocated value.
+ *
+ * Currently three kinds of BPF instructions are supported:
+ * 1. rX = <imm> (assignment with immediate operand);
+ * 2. rX += <imm> (arithmetic operations with immediate operand);
+ * 3. *(rX) = <imm> (indirect memory assignment with immediate operand).
+ *
+ * If actual insn->imm value is wrong, bail out.
+ */
+static int bpf_core_reloc_insn(struct bpf_program *prog, int insn_off,
+ __u32 orig_off, __u32 new_off)
+{
+ struct bpf_insn *insn;
+ int insn_idx;
+ __u8 class;
+
+ if (insn_off % sizeof(struct bpf_insn))
+ return -EINVAL;
+ insn_idx = insn_off / sizeof(struct bpf_insn);
+
+ insn = &prog->insns[insn_idx];
+ class = BPF_CLASS(insn->code);
+
+ if (class == BPF_ALU || class == BPF_ALU64) {
+ if (BPF_SRC(insn->code) != BPF_K)
+ return -EINVAL;
+ if (insn->imm != orig_off)
+ return -EINVAL;
+ insn->imm = new_off;
+ pr_debug("prog '%s': patched insn #%d (ALU/ALU64) imm %d -> %d\n",
+ bpf_program__title(prog, false),
+ insn_idx, orig_off, new_off);
+ } else {
+ pr_warning("prog '%s': trying to relocate unrecognized insn #%d, code:%x, src:%x, dst:%x, off:%x, imm:%x\n",
+ bpf_program__title(prog, false),
+ insn_idx, insn->code, insn->src_reg, insn->dst_reg,
+ insn->off, insn->imm);
+ return -EINVAL;
+ }
+ return 0;
+}
+
+/*
+ * Probe few well-known locations for vmlinux kernel image and try to load BTF
+ * data out of it to use for target BTF.
+ */
+static struct btf *bpf_core_find_kernel_btf(void)
+{
+ const char *locations[] = {
+ "/lib/modules/%1$s/vmlinux-%1$s",
+ "/usr/lib/modules/%1$s/kernel/vmlinux",
+ };
+ char path[PATH_MAX + 1];
+ struct utsname buf;
+ struct btf *btf;
+ int i;
+
+ uname(&buf);
+
+ for (i = 0; i < ARRAY_SIZE(locations); i++) {
+ snprintf(path, PATH_MAX, locations[i], buf.release);
+
+ if (access(path, R_OK))
+ continue;
+
+ btf = btf__parse_elf(path, NULL);
+ pr_debug("kernel BTF load from '%s': %ld\n",
+ path, PTR_ERR(btf));
+ if (IS_ERR(btf))
+ continue;
+
+ return btf;
+ }
+
+ pr_warning("failed to find valid kernel BTF\n");
+ return ERR_PTR(-ESRCH);
+}
+
+/* Output spec definition in the format:
+ * [<type-id>] (<type-name>) + <raw-spec> => <offset>@<spec>,
+ * where <spec> is a C-syntax view of recorded field access, e.g.: x.a[3].b
+ */
+static void bpf_core_dump_spec(int level, const struct bpf_core_spec *spec)
+{
+ const struct btf_type *t;
+ const char *s;
+ __u32 type_id;
+ int i;
+
+ type_id = spec->spec[0].type_id;
+ t = btf__type_by_id(spec->btf, type_id);
+ s = btf__name_by_offset(spec->btf, t->name_off);
+ libbpf_print(level, "[%u] %s + ", type_id, s);
+
+ for (i = 0; i < spec->raw_len; i++)
+ libbpf_print(level, "%d%s", spec->raw_spec[i],
+ i == spec->raw_len - 1 ? " => " : ":");
+
+ libbpf_print(level, "%u @ &x", spec->offset);
+
+ for (i = 0; i < spec->len; i++) {
+ if (spec->spec[i].name)
+ libbpf_print(level, ".%s", spec->spec[i].name);
+ else
+ libbpf_print(level, "[%u]", spec->spec[i].idx);
+ }
+
+}
+
+static size_t bpf_core_hash_fn(const void *key, void *ctx)
+{
+ return (size_t)key;
+}
+
+static bool bpf_core_equal_fn(const void *k1, const void *k2, void *ctx)
+{
+ return k1 == k2;
+}
+
+static void *u32_as_hash_key(__u32 x)
+{
+ return (void *)(uintptr_t)x;
+}
+
+/*
+ * CO-RE relocate single instruction.
+ *
+ * The outline and important points of the algorithm:
+ * 1. For given local type, find corresponding candidate target types.
+ * Candidate type is a type with the same "essential" name, ignoring
+ * everything after last triple underscore (___). E.g., `sample`,
+ * `sample___flavor_one`, `sample___flavor_another_one`, are all candidates
+ * for each other. Names with triple underscore are referred to as
+ * "flavors" and are useful, among other things, to allow to
+ * specify/support incompatible variations of the same kernel struct, which
+ * might differ between different kernel versions and/or build
+ * configurations.
+ *
+ * N.B. Struct "flavors" could be generated by bpftool's BTF-to-C
+ * converter, when deduplicated BTF of a kernel still contains more than
+ * one different types with the same name. In that case, ___2, ___3, etc
+ * are appended starting from second name conflict. But start flavors are
+ * also useful to be defined "locally", in BPF program, to extract same
+ * data from incompatible changes between different kernel
+ * versions/configurations. For instance, to handle field renames between
+ * kernel versions, one can use two flavors of the struct name with the
+ * same common name and use conditional relocations to extract that field,
+ * depending on target kernel version.
+ * 2. For each candidate type, try to match local specification to this
+ * candidate target type. Matching involves finding corresponding
+ * high-level spec accessors, meaning that all named fields should match,
+ * as well as all array accesses should be within the actual bounds. Also,
+ * types should be compatible (see bpf_core_fields_are_compat for details).
+ * 3. It is supported and expected that there might be multiple flavors
+ * matching the spec. As long as all the specs resolve to the same set of
+ * offsets across all candidates, there is not error. If there is any
+ * ambiguity, CO-RE relocation will fail. This is necessary to accomodate
+ * imprefection of BTF deduplication, which can cause slight duplication of
+ * the same BTF type, if some directly or indirectly referenced (by
+ * pointer) type gets resolved to different actual types in different
+ * object files. If such situation occurs, deduplicated BTF will end up
+ * with two (or more) structurally identical types, which differ only in
+ * types they refer to through pointer. This should be OK in most cases and
+ * is not an error.
+ * 4. Candidate types search is performed by linearly scanning through all
+ * types in target BTF. It is anticipated that this is overall more
+ * efficient memory-wise and not significantly worse (if not better)
+ * CPU-wise compared to prebuilding a map from all local type names to
+ * a list of candidate type names. It's also sped up by caching resolved
+ * list of matching candidates per each local "root" type ID, that has at
+ * least one bpf_offset_reloc associated with it. This list is shared
+ * between multiple relocations for the same type ID and is updated as some
+ * of the candidates are pruned due to structural incompatibility.
+ */
+static int bpf_core_reloc_offset(struct bpf_program *prog,
+ const struct bpf_offset_reloc *relo,
+ int relo_idx,
+ const struct btf *local_btf,
+ const struct btf *targ_btf,
+ struct hashmap *cand_cache)
+{
+ const char *prog_name = bpf_program__title(prog, false);
+ struct bpf_core_spec local_spec, cand_spec, targ_spec;
+ const void *type_key = u32_as_hash_key(relo->type_id);
+ const struct btf_type *local_type, *cand_type;
+ const char *local_name, *cand_name;
+ struct ids_vec *cand_ids;
+ __u32 local_id, cand_id;
+ const char *spec_str;
+ int i, j, err;
+
+ local_id = relo->type_id;
+ local_type = btf__type_by_id(local_btf, local_id);
+ if (!local_type)
+ return -EINVAL;
+
+ local_name = btf__name_by_offset(local_btf, local_type->name_off);
+ if (str_is_empty(local_name))
+ return -EINVAL;
+
+ spec_str = btf__name_by_offset(local_btf, relo->access_str_off);
+ if (str_is_empty(spec_str))
+ return -EINVAL;
+
+ err = bpf_core_spec_parse(local_btf, local_id, spec_str, &local_spec);
+ if (err) {
+ pr_warning("prog '%s': relo #%d: parsing [%d] %s + %s failed: %d\n",
+ prog_name, relo_idx, local_id, local_name, spec_str,
+ err);
+ return -EINVAL;
+ }
+
+ pr_debug("prog '%s': relo #%d: spec is ", prog_name, relo_idx);
+ bpf_core_dump_spec(LIBBPF_DEBUG, &local_spec);
+ libbpf_print(LIBBPF_DEBUG, "\n");
+
+ if (!hashmap__find(cand_cache, type_key, (void **)&cand_ids)) {
+ cand_ids = bpf_core_find_cands(local_btf, local_id, targ_btf);
+ if (IS_ERR(cand_ids)) {
+ pr_warning("prog '%s': relo #%d: target candidate search failed for [%d] %s: %ld",
+ prog_name, relo_idx, local_id, local_name,
+ PTR_ERR(cand_ids));
+ return PTR_ERR(cand_ids);
+ }
+ err = hashmap__set(cand_cache, type_key, cand_ids, NULL, NULL);
+ if (err) {
+ bpf_core_free_cands(cand_ids);
+ return err;
+ }
+ }
+
+ for (i = 0, j = 0; i < cand_ids->len; i++) {
+ cand_id = cand_ids->data[i];
+ cand_type = btf__type_by_id(targ_btf, cand_id);
+ cand_name = btf__name_by_offset(targ_btf, cand_type->name_off);
+
+ err = bpf_core_spec_match(&local_spec, targ_btf,
+ cand_id, &cand_spec);
+ pr_debug("prog '%s': relo #%d: matching candidate #%d %s against spec ",
+ prog_name, relo_idx, i, cand_name);
+ bpf_core_dump_spec(LIBBPF_DEBUG, &cand_spec);
+ libbpf_print(LIBBPF_DEBUG, ": %d\n", err);
+ if (err < 0) {
+ pr_warning("prog '%s': relo #%d: matching error: %d\n",
+ prog_name, relo_idx, err);
+ return err;
+ }
+ if (err == 0)
+ continue;
+
+ if (j == 0) {
+ targ_spec = cand_spec;
+ } else if (cand_spec.offset != targ_spec.offset) {
+ /* if there are many candidates, they should all
+ * resolve to the same offset
+ */
+ pr_warning("prog '%s': relo #%d: offset ambiguity: %u != %u\n",
+ prog_name, relo_idx, cand_spec.offset,
+ targ_spec.offset);
+ return -EINVAL;
+ }
+
+ cand_ids->data[j++] = cand_spec.spec[0].type_id;
+ }
+
+ cand_ids->len = j;
+ if (cand_ids->len == 0) {
+ pr_warning("prog '%s': relo #%d: no matching targets found for [%d] %s + %s\n",
+ prog_name, relo_idx, local_id, local_name, spec_str);
+ return -ESRCH;
+ }
+
+ err = bpf_core_reloc_insn(prog, relo->insn_off,
+ local_spec.offset, targ_spec.offset);
+ if (err) {
+ pr_warning("prog '%s': relo #%d: failed to patch insn at offset %d: %d\n",
+ prog_name, relo_idx, relo->insn_off, err);
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int
+bpf_core_reloc_offsets(struct bpf_object *obj, const char *targ_btf_path)
+{
+ const struct btf_ext_info_sec *sec;
+ const struct bpf_offset_reloc *rec;
+ const struct btf_ext_info *seg;
+ struct hashmap_entry *entry;
+ struct hashmap *cand_cache = NULL;
+ struct bpf_program *prog;
+ struct btf *targ_btf;
+ const char *sec_name;
+ int i, err = 0;
+
+ if (targ_btf_path)
+ targ_btf = btf__parse_elf(targ_btf_path, NULL);
+ else
+ targ_btf = bpf_core_find_kernel_btf();
+ if (IS_ERR(targ_btf)) {
+ pr_warning("failed to get target BTF: %ld\n",
+ PTR_ERR(targ_btf));
+ return PTR_ERR(targ_btf);
+ }
+
+ cand_cache = hashmap__new(bpf_core_hash_fn, bpf_core_equal_fn, NULL);
+ if (IS_ERR(cand_cache)) {
+ err = PTR_ERR(cand_cache);
+ goto out;
+ }
+
+ seg = &obj->btf_ext->offset_reloc_info;
+ for_each_btf_ext_sec(seg, sec) {
+ sec_name = btf__name_by_offset(obj->btf, sec->sec_name_off);
+ if (str_is_empty(sec_name)) {
+ err = -EINVAL;
+ goto out;
+ }
+ prog = bpf_object__find_program_by_title(obj, sec_name);
+ if (!prog) {
+ pr_warning("failed to find program '%s' for CO-RE offset relocation\n",
+ sec_name);
+ err = -EINVAL;
+ goto out;
+ }
+
+ pr_debug("prog '%s': performing %d CO-RE offset relocs\n",
+ sec_name, sec->num_info);
+
+ for_each_btf_ext_rec(seg, sec, i, rec) {
+ err = bpf_core_reloc_offset(prog, rec, i, obj->btf,
+ targ_btf, cand_cache);
+ if (err) {
+ pr_warning("prog '%s': relo #%d: failed to relocate: %d\n",
+ sec_name, i, err);
+ goto out;
+ }
+ }
+ }
+
+out:
+ btf__free(targ_btf);
+ if (!IS_ERR_OR_NULL(cand_cache)) {
+ hashmap__for_each_entry(cand_cache, entry, i) {
+ bpf_core_free_cands(entry->value);
+ }
+ hashmap__free(cand_cache);
+ }
+ return err;
+}
+
+static int
+bpf_object__relocate_core(struct bpf_object *obj, const char *targ_btf_path)
+{
+ int err = 0;
+
+ if (obj->btf_ext->offset_reloc_info.len)
+ err = bpf_core_reloc_offsets(obj, targ_btf_path);
+
+ return err;
+}
+
static int
bpf_program__reloc_text(struct bpf_program *prog, struct bpf_object *obj,
struct reloc_desc *relo)
@@ -2397,14 +3235,21 @@ bpf_program__relocate(struct bpf_program *prog, struct bpf_object *obj)
return 0;
}
-
static int
-bpf_object__relocate(struct bpf_object *obj)
+bpf_object__relocate(struct bpf_object *obj, const char *targ_btf_path)
{
struct bpf_program *prog;
size_t i;
int err;
+ if (obj->btf_ext) {
+ err = bpf_object__relocate_core(obj, targ_btf_path);
+ if (err) {
+ pr_warning("failed to perform CO-RE relocations: %d\n",
+ err);
+ return err;
+ }
+ }
for (i = 0; i < obj->nr_programs; i++) {
prog = &obj->programs[i];
@@ -2805,7 +3650,7 @@ int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
obj->loaded = true;
CHECK_ERR(bpf_object__create_maps(obj), err, out);
- CHECK_ERR(bpf_object__relocate(obj), err, out);
+ CHECK_ERR(bpf_object__relocate(obj, attr->target_btf_path), err, out);
CHECK_ERR(bpf_object__load_progs(obj, attr->log_level), err, out);
return 0;
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 8a9d462a6f6d..e8f70977d137 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -92,6 +92,7 @@ LIBBPF_API void bpf_object__close(struct bpf_object *object);
struct bpf_object_load_attr {
struct bpf_object *obj;
int log_level;
+ const char *target_btf_path;
};
/* Load/unload object into/from kernel */
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 06/14] selftests/bpf: add CO-RE relocs testing setup
From: Andrii Nakryiko @ 2019-08-07 5:37 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Add CO-RE relocation test runner. Add one simple test validating that
libbpf's logic for searching for kernel image and loading BTF out of it
works.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
---
.../selftests/bpf/prog_tests/core_reloc.c | 130 ++++++++++++++++++
.../bpf/progs/test_core_reloc_kernel.c | 36 +++++
2 files changed, 166 insertions(+)
create mode 100644 tools/testing/selftests/bpf/prog_tests/core_reloc.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c
diff --git a/tools/testing/selftests/bpf/prog_tests/core_reloc.c b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
new file mode 100644
index 000000000000..f31cede82112
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
@@ -0,0 +1,130 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <test_progs.h>
+
+struct core_reloc_test_case {
+ const char *case_name;
+ const char *bpf_obj_file;
+ const char *btf_src_file;
+ const char *input;
+ int input_len;
+ const char *output;
+ int output_len;
+ bool fails;
+};
+
+static struct core_reloc_test_case test_cases[] = {
+ /* validate we can find kernel image and use its BTF for relocs */
+ {
+ .case_name = "kernel",
+ .bpf_obj_file = "test_core_reloc_kernel.o",
+ .btf_src_file = NULL, /* load from /lib/modules/$(uname -r) */
+ .input = "",
+ .input_len = 0,
+ .output = "\1", /* true */
+ .output_len = 1,
+ },
+};
+
+struct data {
+ char in[256];
+ char out[256];
+};
+
+void test_core_reloc(void)
+{
+ const char *probe_name = "raw_tracepoint/sys_enter";
+ struct bpf_object_load_attr load_attr = {};
+ struct core_reloc_test_case *test_case;
+ int err, duration = 0, i, equal;
+ struct bpf_link *link = NULL;
+ struct bpf_map *data_map;
+ struct bpf_program *prog;
+ struct bpf_object *obj;
+ const int zero = 0;
+ struct data data;
+
+ for (i = 0; i < ARRAY_SIZE(test_cases); i++) {
+ test_case = &test_cases[i];
+
+ if (!test__start_subtest(test_case->case_name))
+ continue;
+
+ obj = bpf_object__open(test_case->bpf_obj_file);
+ if (CHECK(IS_ERR_OR_NULL(obj), "obj_open",
+ "failed to open '%s': %ld\n",
+ test_case->bpf_obj_file, PTR_ERR(obj)))
+ continue;
+
+ prog = bpf_object__find_program_by_title(obj, probe_name);
+ if (CHECK(!prog, "find_probe",
+ "prog '%s' not found\n", probe_name))
+ goto cleanup;
+ bpf_program__set_type(prog, BPF_PROG_TYPE_RAW_TRACEPOINT);
+
+ load_attr.obj = obj;
+ load_attr.log_level = 0;
+ load_attr.target_btf_path = test_case->btf_src_file;
+ err = bpf_object__load_xattr(&load_attr);
+ if (test_case->fails) {
+ CHECK(!err, "obj_load_fail",
+ "should fail to load prog '%s'\n", probe_name);
+ goto cleanup;
+ } else {
+ if (CHECK(err, "obj_load",
+ "failed to load prog '%s': %d\n",
+ probe_name, err))
+ goto cleanup;
+ }
+
+ link = bpf_program__attach_raw_tracepoint(prog, "sys_enter");
+ if (CHECK(IS_ERR(link), "attach_raw_tp", "err %ld\n",
+ PTR_ERR(link)))
+ goto cleanup;
+
+ data_map = bpf_object__find_map_by_name(obj, "test_cor.bss");
+ if (CHECK(!data_map, "find_data_map", "data map not found\n"))
+ goto cleanup;
+
+ memset(&data, 0, sizeof(data));
+ memcpy(data.in, test_case->input, test_case->input_len);
+
+ err = bpf_map_update_elem(bpf_map__fd(data_map),
+ &zero, &data, 0);
+ if (CHECK(err, "update_data_map",
+ "failed to update .data map: %d\n", err))
+ goto cleanup;
+
+ /* trigger test run */
+ usleep(1);
+
+ err = bpf_map_lookup_elem(bpf_map__fd(data_map), &zero, &data);
+ if (CHECK(err, "get_result",
+ "failed to get output data: %d\n", err))
+ goto cleanup;
+
+ equal = memcmp(data.out, test_case->output,
+ test_case->output_len) == 0;
+ if (CHECK(!equal, "check_result",
+ "input/output data don't match\n")) {
+ int j;
+
+ for (j = 0; j < test_case->input_len; j++) {
+ test__printf("input byte #%d: 0x%02hhx\n",
+ j, test_case->input[j]);
+ }
+ for (j = 0; j < test_case->output_len; j++) {
+ test__printf("output byte #%d: EXP 0x%02hhx GOT 0x%02hhx\n",
+ j, test_case->output[j],
+ data.out[j]);
+ }
+ goto cleanup;
+ }
+
+cleanup:
+ if (!IS_ERR_OR_NULL(link)) {
+ bpf_link__destroy(link);
+ link = NULL;
+ }
+ bpf_object__close(obj);
+ }
+}
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c b/tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c
new file mode 100644
index 000000000000..37e02aa3f0c8
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c
@@ -0,0 +1,36 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2019 Facebook
+
+#include <linux/bpf.h>
+#include <stdint.h>
+#include "bpf_helpers.h"
+
+char _license[] SEC("license") = "GPL";
+
+static volatile struct data {
+ char in[256];
+ char out[256];
+} data;
+
+struct task_struct {
+ int pid;
+ int tgid;
+};
+
+SEC("raw_tracepoint/sys_enter")
+int test_core_kernel(void *ctx)
+{
+ struct task_struct *task = (void *)bpf_get_current_task();
+ uint64_t pid_tgid = bpf_get_current_pid_tgid();
+ int pid, tgid;
+
+ if (BPF_CORE_READ(&pid, &task->pid) ||
+ BPF_CORE_READ(&tgid, &task->tgid))
+ return 1;
+
+ /* validate pid + tgid matches */
+ data.out[0] = (((uint64_t)pid << 32) | tgid) == pid_tgid;
+
+ return 0;
+}
+
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 03/14] libbpf: add .BTF.ext offset relocation section loading
From: Andrii Nakryiko @ 2019-08-07 5:37 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Add support for BPF CO-RE offset relocations. Add section/record
iteration macros for .BTF.ext. These macro are useful for iterating over
each .BTF.ext record, either for dumping out contents or later for BPF
CO-RE relocation handling.
To enable other parts of libbpf to work with .BTF.ext contents, moved
a bunch of type definitions into libbpf_internal.h.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
---
tools/lib/bpf/btf.c | 69 ++++++++-------------
tools/lib/bpf/btf.h | 4 ++
tools/lib/bpf/libbpf_internal.h | 105 ++++++++++++++++++++++++++++++++
3 files changed, 136 insertions(+), 42 deletions(-)
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index f49a54c2575f..e3a448bb768a 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -35,47 +35,6 @@ struct btf {
int fd;
};
-struct btf_ext_info {
- /*
- * info points to the individual info section (e.g. func_info and
- * line_info) from the .BTF.ext. It does not include the __u32 rec_size.
- */
- void *info;
- __u32 rec_size;
- __u32 len;
-};
-
-struct btf_ext {
- union {
- struct btf_ext_header *hdr;
- void *data;
- };
- struct btf_ext_info func_info;
- struct btf_ext_info line_info;
- __u32 data_size;
-};
-
-struct btf_ext_info_sec {
- __u32 sec_name_off;
- __u32 num_info;
- /* Followed by num_info * record_size number of bytes */
- __u8 data[0];
-};
-
-/* The minimum bpf_func_info checked by the loader */
-struct bpf_func_info_min {
- __u32 insn_off;
- __u32 type_id;
-};
-
-/* The minimum bpf_line_info checked by the loader */
-struct bpf_line_info_min {
- __u32 insn_off;
- __u32 file_name_off;
- __u32 line_off;
- __u32 line_col;
-};
-
static inline __u64 ptr_to_u64(const void *ptr)
{
return (__u64) (unsigned long) ptr;
@@ -822,6 +781,9 @@ static int btf_ext_setup_info(struct btf_ext *btf_ext,
/* The start of the info sec (including the __u32 record_size). */
void *info;
+ if (ext_sec->len == 0)
+ return 0;
+
if (ext_sec->off & 0x03) {
pr_debug(".BTF.ext %s section is not aligned to 4 bytes\n",
ext_sec->desc);
@@ -925,11 +887,24 @@ static int btf_ext_setup_line_info(struct btf_ext *btf_ext)
return btf_ext_setup_info(btf_ext, ¶m);
}
+static int btf_ext_setup_offset_reloc(struct btf_ext *btf_ext)
+{
+ struct btf_ext_sec_setup_param param = {
+ .off = btf_ext->hdr->offset_reloc_off,
+ .len = btf_ext->hdr->offset_reloc_len,
+ .min_rec_size = sizeof(struct bpf_offset_reloc),
+ .ext_info = &btf_ext->offset_reloc_info,
+ .desc = "offset_reloc",
+ };
+
+ return btf_ext_setup_info(btf_ext, ¶m);
+}
+
static int btf_ext_parse_hdr(__u8 *data, __u32 data_size)
{
const struct btf_ext_header *hdr = (struct btf_ext_header *)data;
- if (data_size < offsetof(struct btf_ext_header, func_info_off) ||
+ if (data_size < offsetofend(struct btf_ext_header, hdr_len) ||
data_size < hdr->hdr_len) {
pr_debug("BTF.ext header not found");
return -EINVAL;
@@ -987,6 +962,9 @@ struct btf_ext *btf_ext__new(__u8 *data, __u32 size)
}
memcpy(btf_ext->data, data, size);
+ if (btf_ext->hdr->hdr_len <
+ offsetofend(struct btf_ext_header, line_info_len))
+ goto done;
err = btf_ext_setup_func_info(btf_ext);
if (err)
goto done;
@@ -995,6 +973,13 @@ struct btf_ext *btf_ext__new(__u8 *data, __u32 size)
if (err)
goto done;
+ if (btf_ext->hdr->hdr_len <
+ offsetofend(struct btf_ext_header, offset_reloc_len))
+ goto done;
+ err = btf_ext_setup_offset_reloc(btf_ext);
+ if (err)
+ goto done;
+
done:
if (err) {
btf_ext__free(btf_ext);
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index 63c939b71efe..c566026ee62b 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -58,6 +58,10 @@ struct btf_ext_header {
__u32 func_info_len;
__u32 line_info_off;
__u32 line_info_len;
+
+ /* optional part of .BTF.ext header */
+ __u32 offset_reloc_off;
+ __u32 offset_reloc_len;
};
LIBBPF_API void btf__free(struct btf *btf);
diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index 2ac29bd36226..2e83a34f8c79 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -29,6 +29,10 @@
#ifndef max
# define max(x, y) ((x) < (y) ? (y) : (x))
#endif
+#ifndef offsetofend
+# define offsetofend(TYPE, FIELD) \
+ (offsetof(TYPE, FIELD) + sizeof(((TYPE *)0)->FIELD))
+#endif
extern void libbpf_print(enum libbpf_print_level level,
const char *format, ...)
@@ -46,4 +50,105 @@ do { \
int libbpf__load_raw_btf(const char *raw_types, size_t types_len,
const char *str_sec, size_t str_len);
+struct btf_ext_info {
+ /*
+ * info points to the individual info section (e.g. func_info and
+ * line_info) from the .BTF.ext. It does not include the __u32 rec_size.
+ */
+ void *info;
+ __u32 rec_size;
+ __u32 len;
+};
+
+#define for_each_btf_ext_sec(seg, sec) \
+ for (sec = (seg)->info; \
+ (void *)sec < (seg)->info + (seg)->len; \
+ sec = (void *)sec + sizeof(struct btf_ext_info_sec) + \
+ (seg)->rec_size * sec->num_info)
+
+#define for_each_btf_ext_rec(seg, sec, i, rec) \
+ for (i = 0, rec = (void *)&(sec)->data; \
+ i < (sec)->num_info; \
+ i++, rec = (void *)rec + (seg)->rec_size)
+
+struct btf_ext {
+ union {
+ struct btf_ext_header *hdr;
+ void *data;
+ };
+ struct btf_ext_info func_info;
+ struct btf_ext_info line_info;
+ struct btf_ext_info offset_reloc_info;
+ __u32 data_size;
+};
+
+struct btf_ext_info_sec {
+ __u32 sec_name_off;
+ __u32 num_info;
+ /* Followed by num_info * record_size number of bytes */
+ __u8 data[0];
+};
+
+/* The minimum bpf_func_info checked by the loader */
+struct bpf_func_info_min {
+ __u32 insn_off;
+ __u32 type_id;
+};
+
+/* The minimum bpf_line_info checked by the loader */
+struct bpf_line_info_min {
+ __u32 insn_off;
+ __u32 file_name_off;
+ __u32 line_off;
+ __u32 line_col;
+};
+
+/* The minimum bpf_offset_reloc checked by the loader
+ *
+ * Offset relocation captures the following data:
+ * - insn_off - instruction offset (in bytes) within a BPF program that needs
+ * its insn->imm field to be relocated with actual offset;
+ * - type_id - BTF type ID of the "root" (containing) entity of a relocatable
+ * offset;
+ * - access_str_off - offset into corresponding .BTF string section. String
+ * itself encodes an accessed field using a sequence of field and array
+ * indicies, separated by colon (:). It's conceptually very close to LLVM's
+ * getelementptr ([0]) instruction's arguments for identifying offset to
+ * a field.
+ *
+ * Example to provide a better feel.
+ *
+ * struct sample {
+ * int a;
+ * struct {
+ * int b[10];
+ * };
+ * };
+ *
+ * struct sample *s = ...;
+ * int x = &s->a; // encoded as "0:0" (a is field #0)
+ * int y = &s->b[5]; // encoded as "0:1:0:5" (anon struct is field #1,
+ * // b is field #0 inside anon struct, accessing elem #5)
+ * int z = &s[10]->b; // encoded as "10:1" (ptr is used as an array)
+ *
+ * type_id for all relocs in this example will capture BTF type id of
+ * `struct sample`.
+ *
+ * Such relocation is emitted when using __builtin_preserve_access_index()
+ * Clang built-in, passing expression that captures field address, e.g.:
+ *
+ * bpf_probe_read(&dst, sizeof(dst),
+ * __builtin_preserve_access_index(&src->a.b.c));
+ *
+ * In this case Clang will emit offset relocation recording necessary data to
+ * be able to find offset of embedded `a.b.c` field within `src` struct.
+ *
+ * [0] https://llvm.org/docs/LangRef.html#getelementptr-instruction
+ */
+struct bpf_offset_reloc {
+ __u32 insn_off;
+ __u32 type_id;
+ __u32 access_str_off;
+};
+
#endif /* __LIBBPF_LIBBPF_INTERNAL_H */
--
2.17.1
^ permalink raw reply related
* [PATCH v4 bpf-next 00/14] CO-RE offset relocations
From: Andrii Nakryiko @ 2019-08-07 5:37 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
This patch set implements central part of CO-RE (Compile Once - Run
Everywhere, see [0] and [1] for slides and video): relocating fields offsets.
Most of the details are written down as comments to corresponding parts of the
code.
Patch #1 adds a bunch of commonly useful btf_xxx helpers to simplify working
with BTF types.
Patch #2 converts existing libbpf code to these new helpers and removes some
of pre-existing ones.
Patch #3 adds loading of .BTF.ext offset relocations section and macros to
work with its contents.
Patch #4 implements CO-RE relocations algorithm in libbpf.
Patch #5 introduced BPF_CORE_READ macro, hiding usage of Clang's
__builtin_preserve_access_index intrinsic that records offset relocation.
Patches #6-#14 adds selftests validating various parts of relocation handling,
type compatibility, etc.
For all tests to work, you'll need latest Clang/LLVM supporting
__builtin_preserve_access_index intrinsic, used for recording offset
relocations. Kernel on which selftests run should have BTF information built
in (CONFIG_DEBUG_INFO_BTF=y).
[0] http://vger.kernel.org/bpfconf2019.html#session-2
[1] http://vger.kernel.org/lpc-bpf2018.html#session-2
v3->v4:
- added btf_xxx helpers (Alexei);
- switched libbpf code to new helpers;
- reduced amount of logging and simplified format in few places (Alexei);
- made flavor name parsing logic more strict (exactly three underscores);
- no uname() error checking (Alexei);
- updated misc tests to reflect latest Clang fixes (Yonghong);
v2->v3:
- enclose BPF_CORE_READ args in parens (Song);
v1->v2:
- add offsetofend(), fix btf_ext optional fields checks (Song);
- add bpf_core_dump_spec() for logging spec representation;
- move special first element processing out of the loop (Song);
- typo fixes (Song);
- drop BPF_ST | BPF_MEM insn relocation (Alexei);
- extracted BPF_CORE_READ into bpf_helpers (Alexei);
- added extra tests validating Clang capturing relocs correctly (Yonghong);
- switch core_relocs.c to use sub-tests;
- updated mods tests after Clang bug was fixed (Yonghong);
- fix bug enumerating candidate types;
Andrii Nakryiko (14):
libbpf: add helpers for working with BTF types
libbpf: convert libbpf code to use new btf helpers
libbpf: add .BTF.ext offset relocation section loading
libbpf: implement BPF CO-RE offset relocation algorithm
selftests/bpf: add BPF_CORE_READ relocatable read macro
selftests/bpf: add CO-RE relocs testing setup
selftests/bpf: add CO-RE relocs struct flavors tests
selftests/bpf: add CO-RE relocs nesting tests
selftests/bpf: add CO-RE relocs array tests
selftests/bpf: add CO-RE relocs enum/ptr/func_proto tests
selftests/bpf: add CO-RE relocs modifiers/typedef tests
selftests/bpf: add CO-RE relocs ptr-as-array tests
selftests/bpf: add CO-RE relocs ints tests
selftests/bpf: add CO-RE relocs misc tests
tools/lib/bpf/btf.c | 250 ++---
tools/lib/bpf/btf.h | 180 ++++
tools/lib/bpf/btf_dump.c | 136 +--
tools/lib/bpf/libbpf.c | 941 +++++++++++++++++-
tools/lib/bpf/libbpf.h | 1 +
tools/lib/bpf/libbpf_internal.h | 105 ++
tools/testing/selftests/bpf/bpf_helpers.h | 20 +
.../selftests/bpf/prog_tests/core_reloc.c | 386 +++++++
.../bpf/progs/btf__core_reloc_arrays.c | 3 +
.../btf__core_reloc_arrays___diff_arr_dim.c | 3 +
...btf__core_reloc_arrays___diff_arr_val_sz.c | 3 +
.../btf__core_reloc_arrays___err_non_array.c | 3 +
...btf__core_reloc_arrays___err_too_shallow.c | 3 +
.../btf__core_reloc_arrays___err_too_small.c | 3 +
..._core_reloc_arrays___err_wrong_val_type1.c | 3 +
..._core_reloc_arrays___err_wrong_val_type2.c | 3 +
.../bpf/progs/btf__core_reloc_flavors.c | 3 +
.../btf__core_reloc_flavors__err_wrong_name.c | 3 +
.../bpf/progs/btf__core_reloc_ints.c | 3 +
.../bpf/progs/btf__core_reloc_ints___bool.c | 3 +
.../btf__core_reloc_ints___err_bitfield.c | 3 +
.../btf__core_reloc_ints___err_wrong_sz_16.c | 3 +
.../btf__core_reloc_ints___err_wrong_sz_32.c | 3 +
.../btf__core_reloc_ints___err_wrong_sz_64.c | 3 +
.../btf__core_reloc_ints___err_wrong_sz_8.c | 3 +
.../btf__core_reloc_ints___reverse_sign.c | 3 +
.../bpf/progs/btf__core_reloc_misc.c | 5 +
.../bpf/progs/btf__core_reloc_mods.c | 3 +
.../progs/btf__core_reloc_mods___mod_swap.c | 3 +
.../progs/btf__core_reloc_mods___typedefs.c | 3 +
.../bpf/progs/btf__core_reloc_nesting.c | 3 +
.../btf__core_reloc_nesting___anon_embed.c | 3 +
...f__core_reloc_nesting___dup_compat_types.c | 5 +
...core_reloc_nesting___err_array_container.c | 3 +
...tf__core_reloc_nesting___err_array_field.c | 3 +
...e_reloc_nesting___err_dup_incompat_types.c | 4 +
...re_reloc_nesting___err_missing_container.c | 3 +
...__core_reloc_nesting___err_missing_field.c | 3 +
..._reloc_nesting___err_nonstruct_container.c | 3 +
...e_reloc_nesting___err_partial_match_dups.c | 4 +
.../btf__core_reloc_nesting___err_too_deep.c | 3 +
.../btf__core_reloc_nesting___extra_nesting.c | 3 +
..._core_reloc_nesting___struct_union_mixup.c | 3 +
.../bpf/progs/btf__core_reloc_primitives.c | 3 +
...f__core_reloc_primitives___diff_enum_def.c | 3 +
..._core_reloc_primitives___diff_func_proto.c | 3 +
...f__core_reloc_primitives___diff_ptr_type.c | 3 +
...tf__core_reloc_primitives___err_non_enum.c | 3 +
...btf__core_reloc_primitives___err_non_int.c | 3 +
...btf__core_reloc_primitives___err_non_ptr.c | 3 +
.../bpf/progs/btf__core_reloc_ptr_as_arr.c | 3 +
.../btf__core_reloc_ptr_as_arr___diff_sz.c | 3 +
.../selftests/bpf/progs/core_reloc_types.h | 667 +++++++++++++
.../bpf/progs/test_core_reloc_arrays.c | 55 +
.../bpf/progs/test_core_reloc_flavors.c | 62 ++
.../bpf/progs/test_core_reloc_ints.c | 44 +
.../bpf/progs/test_core_reloc_kernel.c | 36 +
.../bpf/progs/test_core_reloc_misc.c | 57 ++
.../bpf/progs/test_core_reloc_mods.c | 62 ++
.../bpf/progs/test_core_reloc_nesting.c | 46 +
.../bpf/progs/test_core_reloc_primitives.c | 43 +
.../bpf/progs/test_core_reloc_ptr_as_arr.c | 30 +
62 files changed, 2979 insertions(+), 280 deletions(-)
create mode 100644 tools/testing/selftests/bpf/prog_tests/core_reloc.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___diff_arr_dim.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___diff_arr_val_sz.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_non_array.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_too_shallow.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_too_small.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_wrong_val_type1.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_wrong_val_type2.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_flavors.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_flavors__err_wrong_name.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___bool.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_bitfield.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_16.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_32.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_64.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___err_wrong_sz_8.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ints___reverse_sign.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_misc.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_mods.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_mods___mod_swap.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_mods___typedefs.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___anon_embed.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___dup_compat_types.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_array_container.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_array_field.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_dup_incompat_types.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_missing_container.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_missing_field.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_nonstruct_container.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_partial_match_dups.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___err_too_deep.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___extra_nesting.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_nesting___struct_union_mixup.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_primitives.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_primitives___diff_enum_def.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_primitives___diff_func_proto.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_primitives___diff_ptr_type.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_primitives___err_non_enum.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_primitives___err_non_int.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_primitives___err_non_ptr.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ptr_as_arr.c
create mode 100644 tools/testing/selftests/bpf/progs/btf__core_reloc_ptr_as_arr___diff_sz.c
create mode 100644 tools/testing/selftests/bpf/progs/core_reloc_types.h
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_ints.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_misc.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_mods.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_primitives.c
create mode 100644 tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c
--
2.17.1
^ permalink raw reply
* [PATCH v4 bpf-next 01/14] libbpf: add helpers for working with BTF types
From: Andrii Nakryiko @ 2019-08-07 5:37 UTC (permalink / raw)
To: bpf, netdev, ast, daniel, yhs
Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
In-Reply-To: <20190807053806.1534571-1-andriin@fb.com>
Add lots of frequently used helpers that simplify working with BTF
types.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
---
tools/lib/bpf/btf.h | 176 ++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 176 insertions(+)
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index 88a52ae56fc6..63c939b71efe 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -5,6 +5,7 @@
#define __LIBBPF_BTF_H
#include <stdarg.h>
+#include <linux/btf.h>
#include <linux/types.h>
#ifdef __cplusplus
@@ -120,6 +121,181 @@ LIBBPF_API void btf_dump__free(struct btf_dump *d);
LIBBPF_API int btf_dump__dump_type(struct btf_dump *d, __u32 id);
+/*
+ * A set of helpers for easier BTF types handling
+ */
+static inline __u16 btf_kind(const struct btf_type *t)
+{
+ return BTF_INFO_KIND(t->info);
+}
+
+static inline __u16 btf_vlen(const struct btf_type *t)
+{
+ return BTF_INFO_VLEN(t->info);
+}
+
+static inline bool btf_kflag(const struct btf_type *t)
+{
+ return BTF_INFO_KFLAG(t->info);
+}
+
+static inline bool btf_is_int(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_INT;
+}
+
+static inline bool btf_is_ptr(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_PTR;
+}
+
+static inline bool btf_is_array(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_ARRAY;
+}
+
+static inline bool btf_is_struct(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_STRUCT;
+}
+
+static inline bool btf_is_union(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_UNION;
+}
+
+static inline bool btf_is_composite(const struct btf_type *t)
+{
+ __u16 kind = btf_kind(t);
+
+ return kind == BTF_KIND_STRUCT || kind == BTF_KIND_UNION;
+}
+
+static inline bool btf_is_enum(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_ENUM;
+}
+
+static inline bool btf_is_fwd(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_FWD;
+}
+
+static inline bool btf_is_typedef(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_TYPEDEF;
+}
+
+static inline bool btf_is_volatile(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_VOLATILE;
+}
+
+static inline bool btf_is_const(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_CONST;
+}
+
+static inline bool btf_is_restrict(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_RESTRICT;
+}
+
+static inline bool btf_is_mod(const struct btf_type *t)
+{
+ __u16 kind = btf_kind(t);
+
+ return kind == BTF_KIND_VOLATILE ||
+ kind == BTF_KIND_CONST ||
+ kind == BTF_KIND_RESTRICT;
+}
+
+static inline bool btf_is_func(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_FUNC;
+}
+
+static inline bool btf_is_func_proto(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_FUNC_PROTO;
+}
+
+static inline bool btf_is_var(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_VAR;
+}
+
+static inline bool btf_is_datasec(const struct btf_type *t)
+{
+ return btf_kind(t) == BTF_KIND_DATASEC;
+}
+
+static inline __u8 btf_int_encoding(const struct btf_type *t)
+{
+ return BTF_INT_ENCODING(*(__u32 *)(t + 1));
+}
+
+static inline __u8 btf_int_offset(const struct btf_type *t)
+{
+ return BTF_INT_OFFSET(*(__u32 *)(t + 1));
+}
+
+static inline __u8 btf_int_bits(const struct btf_type *t)
+{
+ return BTF_INT_BITS(*(__u32 *)(t + 1));
+}
+
+static inline const struct btf_array *btf_array(const struct btf_type *t)
+{
+ return (const struct btf_array *)(t + 1);
+}
+
+static inline const struct btf_enum *btf_enum(const struct btf_type *t)
+{
+ return (const struct btf_enum *)(t + 1);
+}
+
+static inline const struct btf_member *btf_members(const struct btf_type *t)
+{
+ return (const struct btf_member *)(t + 1);
+}
+
+/* get bit offset of a member with specified index */
+static inline const __u32 btf_member_bit_offset(const struct btf_type *t,
+ __u32 member_idx)
+{
+ const struct btf_member *m = btf_members(t) + member_idx;
+ bool kflag = btf_kflag(t);
+
+ return kflag ? BTF_MEMBER_BIT_OFFSET(m->offset) : m->offset;
+}
+/* get bitfield size of a member, assuming t is BTF_KIND_STRUCT or
+ * BTF_KIND_UNION. If member is not a bitfield, zero is returned. */
+static inline const __u32 btf_member_bitfield_size(const struct btf_type *t,
+ __u32 member_idx)
+{
+ const struct btf_member *m = btf_members(t) + member_idx;
+ bool kflag = btf_kflag(t);
+
+ return kflag ? BTF_MEMBER_BITFIELD_SIZE(m->offset) : 0;
+}
+
+static inline const struct btf_param *btf_params(const struct btf_type *t)
+{
+ return (const struct btf_param *)(t + 1);
+}
+
+static inline const struct btf_var *btf_var(const struct btf_type *t)
+{
+ return (const struct btf_var *)(t + 1);
+}
+
+static inline const struct btf_var_secinfo *
+btf_var_secinfos(const struct btf_type *t)
+{
+ return (const struct btf_var_secinfo *)(t + 1);
+}
+
#ifdef __cplusplus
} /* extern "C" */
#endif
--
2.17.1
^ permalink raw reply related
* Re: [PATCH v2 bpf-next 1/4] bpf: unprivileged BPF access via /dev/bpf
From: Andy Lutomirski @ 2019-08-07 5:24 UTC (permalink / raw)
To: Alexei Starovoitov
Cc: Andy Lutomirski, Song Liu, Kees Cook, Networking, bpf,
Alexei Starovoitov, Daniel Borkmann, Kernel Team, Lorenz Bauer,
Jann Horn, Greg KH, Linux API, LSM List
In-Reply-To: <20190806011134.p5baub5l3t5fkmou@ast-mbp>
On Mon, Aug 5, 2019 at 6:11 PM Alexei Starovoitov
<alexei.starovoitov@gmail.com> wrote:
>
> On Mon, Aug 05, 2019 at 02:25:35PM -0700, Andy Lutomirski wrote:
> > It tries to make the kernel respect the access modes for fds. Without
> > this patch, there seem to be some holes: nothing looked at program fds
> > and, unless I missed something, you could take a readonly fd for a
> > program, pin the program, and reopen it RW.
>
> I think it's by design. iirc Daniel had a use case for something like this.
That seems odd. Daniel, can you elaborate?
> Hence unprivileged bpf is actually something that can be deprecated.
I hope not. There are a couple setsockopt uses right now, and and
seccomp will surely want it someday. And the bpf-inside-container use
case really is unprivileged bpf -- containers are, in many (most?)
cases, explicitly not trusted by the host. But regardless, see below.
> > The ability to obtain an fd to a cgroup does *not* imply any right to
> > modify that cgroup. The ability to write to a cgroup directory
> > already means something else -- it's the ability to create cgroups
> > under the group in question. I'm suggesting that a new API be added
> > that allows attaching a bpf program to a cgroup without capabilities
> > and that instead requires write access to a new file in the cgroup
> > directory. (It could be a single file for all bpf types or one file
> > per type. I prefer the latter -- it gives the admin finer-grained
> > control.)
>
> This is something to discuss. I don't mind something like this,
> but in general bpf is not for untrusted users.
> Hence I don't want to overdesign.
I think the code would end up being extremely simple. It would be an
ABI change, but I think that it could easily be arranged so that new
and old libbpf would be fully functional on new and old kernels except
that only new libbpf with a new kernel would be able to attach cgroup
programs without privilege. I see no reason to remove the current
cgroup attach API.
> See https://github.com/systemd/systemd/blob/01234e1fe777602265ffd25ab6e73823c62b0efe/src/core/bpf-firewall.c#L671-L674
> bpf based IP sandboxing doesn't work in 'systemd --user'.
> That is just one of the problems that people complained about.
This isn't privilege *dropping* -- this is not having privilege in the
first place. Giving systemd --user unrestricted use of bpf() is
tantamount to making the user root. But again, see below.
>
> Inside containers and inside nested containers we need to start processes
> that will use bpf. All of the processes are trusted.
Trusted by whom? In a non-nested container, the container manager
*might* be trusted by the outside world. In a *nested* container,
unless the inner container management is controlled from outside the
outer container, it's not trusted. I don't know much about how
Facebook's containers work, but the LXC/LXD/Podman world is moving
very strongly toward user namespaces and maximally-untrusted
containers, and I think bpf() should work in that context.
> To solve your concern of bypassing all capable checks...
> How about we do /dev/bpf/full_verifier first?
> It will replace capable() checks in the verifier only.
I'm not convinced that "in the verifier" is the right distinction.
Telling administrators that some setting lets certain users bypass
bpf() verifier checks doesn't have a clear enough meaning. I propose,
instead, that the current capable() checks be divided into three
categories:
a) Those that, by design, control privileged operations. This
includes most attach calls, but it also includes allow_ptr_leaks,
bpf_probe_read(), and quite a few other things. It also includes all
of the by_id calls, I think, unless some clever modification to the
way they worked would isolate different users' objects. I think that
persistent objects can do pretty much everything that by_id users
would need, so this isn't a big deal.
b) Those that protect code that is considered to be at higher risk of
exposing security bugs. In other words, these are things that would
have no need for privilege if we could prove that the implementation
was perfect. This includes bpf2bpf, LPM, etc.
c) Those that serve to prevent excessive resource usage. This
includes the bounded loop code (I think -- this might also be category
(b)) and several explicit checks for program size.
I suppose that the speculative execution mitigation stuff could be
split out from (b) into its own category.
Based on this, how about a different division: /dev/bpf_dangerous for
(b) and /dev/bpf_resources for (c)? And (a) is dealt with by
gradually reducing the privilege needed for the important operations.
/dev/bpf_dangerous isn't ideal in the long run though, since a lot of
the operations in that category will probably become less dangerous as
the code is better vetted, and people granting "dangerous" permission
might want to restrict it to only the specific parts that they need.
/dev/bpf_lpm, /dev/bpf2bpf, etc could work better.
This type of thing actually fits quite nicely into an idea I've been
thinking about for a while called "implicit rights". In very brief
summary, there would be objects called /dev/rights/xyz, where xyz is
the same of a "right". If there is a readable object of the right
type at the literal path "/dev/rights/xyz", then you have right xyz.
There's a bit more flexibility on top of this. BPF could use
/dev/rights/bpf/maptypes/lpm and
/dev/rights/bpf/verifier/bounded_loops, for example. Other non-BPF
use cases include a biggie:
/dev/rights/namespace/create_unprivileged_userns.
/dev/rights/bind_port/80 would be nice, too.
I may have a bit of time over the next few days to actually prototype
this. Would you be interested in using this for BPF?
^ permalink raw reply
* Re: [PATCH -next] iwlwifi: dbg: work around clang bug by marking debug strings static
From: Luciano Coelho @ 2019-08-07 5:24 UTC (permalink / raw)
To: Nathan Chancellor, Nick Desaulniers
Cc: Johannes Berg, Michael Ellerman, Kalle Valo, Arnd Bergmann,
Emmanuel Grumbach, Intel Linux Wireless, David S. Miller,
Shahar S Matityahu, Sara Sharon, linux-wireless, netdev, LKML,
clang-built-linux
In-Reply-To: <20190807051516.GA117639@archlinux-threadripper>
On Tue, 2019-08-06 at 22:15 -0700, Nathan Chancellor wrote:
> On Tue, Aug 06, 2019 at 03:37:42PM -0700, Nick Desaulniers wrote:
> > On Thu, Aug 1, 2019 at 12:11 AM Johannes Berg <johannes@sipsolutions.net> wrote:
> > >
> > > > Luca, you said this was already fixed in your internal tree, and the fix
> > > > would appear soon in next, but I don't see anything in linux-next?
> > >
> > > Luca is still on vacation, but I just sent out a version of the patch we
> > > had applied internally.
> > >
> > > Also turns out it wasn't actually _fixed_, just _moved_, so those
> > > internal patches wouldn't have helped anyway.
> >
> > Thanks for the report. Do you have a link?
> > I'll rebase my patch then.
> > --
> > Thanks,
> > ~Nick Desaulniers
>
> Just for everyone else (since I commented on our issue tracker), this is
> now fixed in Linus's tree as of commit 1f6607250331 ("iwlwifi: dbg_ini:
> fix compile time assert build errors").
Yes, thanks Nathan! I was just digging for this patch to reply to you,
I'm still catching up with what happened during my vacations.
--
Cheers,
Luca.
^ permalink raw reply
* Re: [PATCH -next] iwlwifi: dbg: work around clang bug by marking debug strings static
From: Nathan Chancellor @ 2019-08-07 5:15 UTC (permalink / raw)
To: Nick Desaulniers
Cc: Johannes Berg, Michael Ellerman, Kalle Valo, Luca Coelho,
Arnd Bergmann, Emmanuel Grumbach, Intel Linux Wireless,
David S. Miller, Shahar S Matityahu, Sara Sharon, linux-wireless,
netdev, LKML, clang-built-linux
In-Reply-To: <CAKwvOdmBeB1BezsGh=cK=U9m8goKzZnngDRzNM7B1voZfh8yWg@mail.gmail.com>
On Tue, Aug 06, 2019 at 03:37:42PM -0700, Nick Desaulniers wrote:
> On Thu, Aug 1, 2019 at 12:11 AM Johannes Berg <johannes@sipsolutions.net> wrote:
> >
> >
> > > Luca, you said this was already fixed in your internal tree, and the fix
> > > would appear soon in next, but I don't see anything in linux-next?
> >
> > Luca is still on vacation, but I just sent out a version of the patch we
> > had applied internally.
> >
> > Also turns out it wasn't actually _fixed_, just _moved_, so those
> > internal patches wouldn't have helped anyway.
>
> Thanks for the report. Do you have a link?
> I'll rebase my patch then.
> --
> Thanks,
> ~Nick Desaulniers
Just for everyone else (since I commented on our issue tracker), this is
now fixed in Linus's tree as of commit 1f6607250331 ("iwlwifi: dbg_ini:
fix compile time assert build errors").
Cheers,
Nathan
^ permalink raw reply
* Re: [PATCH v2] net: mdio-octeon: Fix Kconfig warnings and build errors
From: Nathan Chancellor @ 2019-08-07 5:10 UTC (permalink / raw)
To: David Miller
Cc: andrew, broonie, devel, f.fainelli, gregkh, hkallweit1,
kernel-build-reports, linux-arm-kernel, linux-next, lkp, netdev,
rdunlap, willy
In-Reply-To: <20190806.141133.1365654857955536268.davem@davemloft.net>
On Tue, Aug 06, 2019 at 02:11:33PM -0700, David Miller wrote:
> From: Nathan Chancellor <natechancellor@gmail.com>
> Date: Fri, 2 Aug 2019 23:01:56 -0700
>
> > After commit 171a9bae68c7 ("staging/octeon: Allow test build on
> > !MIPS"), the following combination of configs cause a few Kconfig
> > warnings and build errors (distilled from arm allyesconfig and Randy's
> > randconfig builds):
> >
> > CONFIG_NETDEVICES=y
> > CONFIG_STAGING=y
> > CONFIG_COMPILE_TEST=y
> >
> > and CONFIG_OCTEON_ETHERNET as either a module or built-in.
> >
> > WARNING: unmet direct dependencies detected for MDIO_OCTEON
> > Depends on [n]: NETDEVICES [=y] && MDIO_DEVICE [=y] && MDIO_BUS [=y]
> > && 64BIT [=n] && HAS_IOMEM [=y] && OF_MDIO [=n]
> > Selected by [y]:
> > - OCTEON_ETHERNET [=y] && STAGING [=y] && (CAVIUM_OCTEON_SOC ||
> > COMPILE_TEST [=y]) && NETDEVICES [=y]
> >
> > In file included from ../drivers/net/phy/mdio-octeon.c:14:
> > ../drivers/net/phy/mdio-cavium.h:111:36: error: implicit declaration of
> > function ‘writeq’; did you mean ‘writel’?
> > [-Werror=implicit-function-declaration]
> > 111 | #define oct_mdio_writeq(val, addr) writeq(val, (void *)addr)
> > | ^~~~~~
> >
> > CONFIG_64BIT is not strictly necessary if the proper readq/writeq
> > definitions are included from io-64-nonatomic-lo-hi.h.
> >
> > CONFIG_OF_MDIO is not needed when CONFIG_COMPILE_TEST is enabled because
> > of commit f9dc9ac51610 ("of/mdio: Add dummy functions in of_mdio.h.").
> >
> > Fixes: 171a9bae68c7 ("staging/octeon: Allow test build on !MIPS")
> > Reported-by: kbuild test robot <lkp@intel.com>
> > Reported-by: Mark Brown <broonie@kernel.org>
> > Reported-by: Randy Dunlap <rdunlap@infradead.org>
> > Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
>
> Applied to net-next.
>
> Please make it clear what tree your changes are targetting in the future,
> thank you.
Sorry for the confusion, I'll do my best to add a patch suffix in the
future.
Thank you for picking this up!
Nathan
^ permalink raw reply
* [PATCH] tipc: set addr_trail_end when using explicit node addresses
From: Chris Packham @ 2019-08-07 4:55 UTC (permalink / raw)
To: jon.maloy, ying.xue, davem
Cc: netdev, tipc-discussion, linux-kernel, Chris Packham
When tipc uses auto-generated node addresses it goes through a duplicate
address detection phase to ensure the address is unique.
When using explicitly configured node names the DAD phase is skipped.
However addr_trail_end was being left set to 0 which causes parts of the
tipc state machine to assume that the address is not yet valid and
unnecessarily delays the discovery phase. By setting addr_trail_end to
jiffies when using explicit addresses we ensure that we move straight to
discovery.
Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
---
net/tipc/discover.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/net/tipc/discover.c b/net/tipc/discover.c
index c138d68e8a69..f83bfe8c9443 100644
--- a/net/tipc/discover.c
+++ b/net/tipc/discover.c
@@ -361,6 +361,8 @@ int tipc_disc_create(struct net *net, struct tipc_bearer *b,
if (!tipc_own_addr(net)) {
tn->addr_trial_end = jiffies + msecs_to_jiffies(1000);
msg_set_type(buf_msg(d->skb), DSC_TRIAL_MSG);
+ } else {
+ tn->addr_trial_end = jiffies;
}
memcpy(&d->dest, dest, sizeof(*dest));
d->net = net;
--
2.22.0
^ permalink raw reply related
* RE: [PATCH net-next 4/5] r8152: support skb_add_rx_frag
From: Hayes Wang @ 2019-08-07 4:34 UTC (permalink / raw)
To: Jakub Kicinski
Cc: netdev@vger.kernel.org, nic_swsd, linux-kernel@vger.kernel.org,
linux-usb@vger.kernel.org
In-Reply-To: <20190806150802.72e0ef02@cakuba.netronome.com>
Jakub Kicinski [mailto:jakub.kicinski@netronome.com]
> Sent: Wednesday, August 07, 2019 6:08 AM
[...]
> > #define RTL8152_REQT_READ 0xc0
> > @@ -720,7 +723,7 @@ struct r8152 {
> > struct net_device *netdev;
> > struct urb *intr_urb;
> > struct tx_agg tx_info[RTL8152_MAX_TX];
> > - struct list_head rx_info;
> > + struct list_head rx_info, rx_used;
>
> I don't see where entries on the rx_used list get freed when driver is
> unloaded, could you explain how that's taken care of?
When the driver is unloaded, all rx_agg would be freed from
info_list list.
The info_list includes all rx_agg buffers which may be idle
or be busy. The rx_done and rx_use are used to determine
the status of rx_agg buffer included in info_list.
info_list: the rx_agg buffer would be inserted in this list
when it is allocated.
rx_done: the rx_agg buffer is ready (contains rx data). Or
it needs to be resubmitted.
rx_use: the rx_agg buffer is busy and couldn't be submitted
yet.
Best Regards,
Hayes
^ permalink raw reply
* RE: [PATCH net-next 2/5] r8152: replace array with linking list for rx information
From: Hayes Wang @ 2019-08-07 4:34 UTC (permalink / raw)
To: Jakub Kicinski
Cc: netdev@vger.kernel.org, nic_swsd, linux-kernel@vger.kernel.org,
linux-usb@vger.kernel.org
In-Reply-To: <20190806125342.4620a94f@cakuba.netronome.com>
Jakub Kicinski [mailto:jakub.kicinski@netronome.com]
[...]
> > static int rtl_stop_rx(struct r8152 *tp)
> > {
> > - int i;
> > + struct list_head *cursor, *next, tmp_list;
> > + unsigned long flags;
> > +
> > + INIT_LIST_HEAD(&tmp_list);
> >
> > - for (i = 0; i < RTL8152_MAX_RX; i++)
> > - usb_kill_urb(tp->rx_info[i].urb);
> > + /* The usb_kill_urb() couldn't be used in atomic.
> > + * Therefore, move the list of rx_info to a tmp one.
> > + * Then, list_for_each_safe could be used without
> > + * spin lock.
> > + */
>
> Would you mind explaining in a little more detail why taking the
> entries from the list for a brief period of time is safe?
Usually, it needs the spin lock before accessing the entry
of the list "tp->rx_info". However, for some reasons,
if we want to access the entry without spin lock, we
cloud move all entries to a local list temporally. Then,
we could make sure no other one could access the entries
included in the temporal local list.
For this case, when I move all entries to a temporal
local list, no other one could access them. Therefore,
I could access the entries included in the temporal local
list without spin lock.
Best Regards,
Hayes
^ permalink raw reply
* Re: [PATCH net v2] net: dsa: Check existence of .port_mdb_add callback before calling it
From: Vivien Didelot @ 2019-08-07 4:33 UTC (permalink / raw)
To: Chen-Yu Tsai
Cc: Chen-Yu Tsai, Andrew Lunn, Florian Fainelli, David S. Miller,
netdev, linux-kernel
In-Reply-To: <CAGb2v666sNV41rejf=wTJB2gYeDGAHNHn1NyxyRh3E6y9C=11A@mail.gmail.com>
Hi Chen-Yu,
On Wed, 7 Aug 2019 11:18:28 +0800, Chen-Yu Tsai <wens@kernel.org> wrote:
> On Wed, Aug 7, 2019 at 4:34 AM Vivien Didelot <vivien.didelot@gmail.com> wrote:
> >
> > Hi Chen-Yu,
> >
> > On Wed, 7 Aug 2019 01:49:37 +0800, Chen-Yu Tsai <wens@kernel.org> wrote:
> > > On Wed, Aug 7, 2019 at 1:15 AM Vivien Didelot <vivien.didelot@gmail.com> wrote:
> > > >
> > > > Hi Chen-Yu,
> > > >
> > > > On Tue, 6 Aug 2019 15:53:25 +0800, Chen-Yu Tsai <wens@kernel.org> wrote:
> > > > > From: Chen-Yu Tsai <wens@csie.org>
> > > > >
> > > > > With the recent addition of commit 75dad2520fc3 ("net: dsa: b53: Disable
> > > > > all ports on setup"), users of b53 (BCM53125 on Lamobo R1 in my case)
> > > > > are forced to use the dsa subsystem to enable the switch, instead of
> > > > > having it in the default transparent "forward-to-all" mode.
> > > > >
> > > > > The b53 driver does not support mdb bitmap functions. However the dsa
> > > > > layer does not check for the existence of the .port_mdb_add callback
> > > > > before actually using it. This results in a NULL pointer dereference,
> > > > > as shown in the kernel oops below.
> > > > >
> > > > > The other functions seem to be properly guarded. Do the same for
> > > > > .port_mdb_add in dsa_switch_mdb_add_bitmap() as well.
> > > > >
> > > > > b53 is not the only driver that doesn't support mdb bitmap functions.
> > > > > Others include bcm_sf2, dsa_loop, lantiq_gswip, mt7530, mv88e6060,
> > > > > qca8k, realtek-smi, and vitesse-vsc73xx.
> > > >
> > > > I don't know what you mean by that, there's no "mdb bitmap function"
> > > > support for drivers, only the port_mdb_{prepare,add,del} callbacks...
> > >
> > > The term was coined from commit e6db98db8a95 ("net: dsa: add switch mdb
> > > bitmap functions"). But yeah, .port_mdb_* ops/callbacks would be more
> > > appropriate.
> > >
> > > > > 8<--- cut here ---
> > > > > Unable to handle kernel NULL pointer dereference at virtual address 00000000
> > > > > pgd = (ptrval)
> > > > > [00000000] *pgd=00000000
> > > > > Internal error: Oops: 80000005 [#1] SMP ARM
> > > > > Modules linked in: rtl8xxxu rtl8192cu rtl_usb rtl8192c_common rtlwifi mac80211 cfg80211
> > > > > CPU: 1 PID: 134 Comm: kworker/1:2 Not tainted 5.3.0-rc1-00247-gd3519030752a #1
> > > > > Hardware name: Allwinner sun7i (A20) Family
> > > > > Workqueue: events switchdev_deferred_process_work
> > > > > PC is at 0x0
> > > > > LR is at dsa_switch_event+0x570/0x620
> > > > > pc : [<00000000>] lr : [<c08533ec>] psr: 80070013
> > > > > sp : ee871db8 ip : 00000000 fp : ee98d0a4
> > > > > r10: 0000000c r9 : 00000008 r8 : ee89f710
> > > > > r7 : ee98d040 r6 : ee98d088 r5 : c0f04c48 r4 : ee98d04c
> > > > > r3 : 00000000 r2 : ee89f710 r1 : 00000008 r0 : ee98d040
> > > > > Flags: Nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none
> > > > > Control: 10c5387d Table: 6deb406a DAC: 00000051
> > > > > Process kworker/1:2 (pid: 134, stack limit = 0x(ptrval))
> > > > > Stack: (0xee871db8 to 0xee872000)
> > > > > 1da0: ee871e14 103ace2d
> > > > > 1dc0: 00000000 ffffffff 00000000 ee871e14 00000005 00000000 c08524a0 00000000
> > > > > 1de0: ffffe000 c014bdfc c0f04c48 ee871e98 c0f04c48 ee9e5000 c0851120 c014bef0
> > > > > 1e00: 00000000 b643aea2 ee9b4068 c08509a8 ee2bf940 ee89f710 ee871ecb 00000000
> > > > > 1e20: 00000008 103ace2d 00000000 c087e248 ee29c868 103ace2d 00000001 ffffffff
> > > > > 1e40: 00000000 ee871e98 00000006 00000000 c0fb2a50 c087e2d0 ffffffff c08523c4
> > > > > 1e60: ffffffff c014bdfc 00000006 c0fad2d0 ee871e98 ee89f710 00000000 c014c500
> > > > > 1e80: 00000000 ee89f3c0 c0f04c48 00000000 ee9e5000 c087dfb4 ee9e5000 00000000
> > > > > 1ea0: ee89f710 ee871ecb 00000001 103ace2d 00000000 c0f04c48 00000000 c087e0a8
> > > > > 1ec0: 00000000 efd9a3e0 0089f3c0 103ace2d ee89f700 ee89f710 ee9e5000 00000122
> > > > > 1ee0: 00000100 c087e130 ee89f700 c0fad2c8 c1003ef0 c087de4c 2e928000 c0fad2ec
> > > > > 1f00: c0fad2ec ee839580 ef7a62c0 ef7a9400 00000000 c087def8 c0fad2ec c01447dc
> > > > > 1f20: ef315640 ef7a62c0 00000008 ee839580 ee839594 ef7a62c0 00000008 c0f03d00
> > > > > 1f40: ef7a62d8 ef7a62c0 ffffe000 c0145b84 ffffe000 c0fb2420 c0bfaa8c 00000000
> > > > > 1f60: ffffe000 ee84b600 ee84b5c0 00000000 ee870000 ee839580 c0145b40 ef0e5ea4
> > > > > 1f80: ee84b61c c014a6f8 00000001 ee84b5c0 c014a5b0 00000000 00000000 00000000
> > > > > 1fa0: 00000000 00000000 00000000 c01010e8 00000000 00000000 00000000 00000000
> > > > > 1fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
> > > > > 1fe0: 00000000 00000000 00000000 00000000 00000013 00000000 00000000 00000000
> > > > > [<c08533ec>] (dsa_switch_event) from [<c014bdfc>] (notifier_call_chain+0x48/0x84)
> > > > > [<c014bdfc>] (notifier_call_chain) from [<c014bef0>] (raw_notifier_call_chain+0x18/0x20)
> > > > > [<c014bef0>] (raw_notifier_call_chain) from [<c08509a8>] (dsa_port_mdb_add+0x48/0x74)
> > > > > [<c08509a8>] (dsa_port_mdb_add) from [<c087e248>] (__switchdev_handle_port_obj_add+0x54/0xd4)
> > > > > [<c087e248>] (__switchdev_handle_port_obj_add) from [<c087e2d0>] (switchdev_handle_port_obj_add+0x8/0x14)
> > > > > [<c087e2d0>] (switchdev_handle_port_obj_add) from [<c08523c4>] (dsa_slave_switchdev_blocking_event+0x94/0xa4)
> > > > > [<c08523c4>] (dsa_slave_switchdev_blocking_event) from [<c014bdfc>] (notifier_call_chain+0x48/0x84)
> > > > > [<c014bdfc>] (notifier_call_chain) from [<c014c500>] (blocking_notifier_call_chain+0x50/0x68)
> > > > > [<c014c500>] (blocking_notifier_call_chain) from [<c087dfb4>] (switchdev_port_obj_notify+0x44/0xa8)
> > > > > [<c087dfb4>] (switchdev_port_obj_notify) from [<c087e0a8>] (switchdev_port_obj_add_now+0x90/0x104)
> > > > > [<c087e0a8>] (switchdev_port_obj_add_now) from [<c087e130>] (switchdev_port_obj_add_deferred+0x14/0x5c)
> > > > > [<c087e130>] (switchdev_port_obj_add_deferred) from [<c087de4c>] (switchdev_deferred_process+0x64/0x104)
> > > > > [<c087de4c>] (switchdev_deferred_process) from [<c087def8>] (switchdev_deferred_process_work+0xc/0x14)
> > > > > [<c087def8>] (switchdev_deferred_process_work) from [<c01447dc>] (process_one_work+0x218/0x50c)
> > > > > [<c01447dc>] (process_one_work) from [<c0145b84>] (worker_thread+0x44/0x5bc)
> > > > > [<c0145b84>] (worker_thread) from [<c014a6f8>] (kthread+0x148/0x150)
> > > > > [<c014a6f8>] (kthread) from [<c01010e8>] (ret_from_fork+0x14/0x2c)
> > > > > Exception stack(0xee871fb0 to 0xee871ff8)
> > > > > 1fa0: 00000000 00000000 00000000 00000000
> > > > > 1fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
> > > > > 1fe0: 00000000 00000000 00000000 00000000 00000013 00000000
> > > > > Code: bad PC value
> > > > > ---[ end trace 1292c61abd17b130 ]---
> > > > >
> > > > > [<c08533ec>] (dsa_switch_event) from [<c014bdfc>] (notifier_call_chain+0x48/0x84)
> > > > > corresponds to
> > > > >
> > > > > $ arm-linux-gnueabihf-addr2line -C -i -e vmlinux c08533ec
> > > > >
> > > > > linux/net/dsa/switch.c:156
> > > > > linux/net/dsa/switch.c:178
> > > > > linux/net/dsa/switch.c:328
> > > > >
> > > > > Fixes: e6db98db8a95 ("net: dsa: add switch mdb bitmap functions")
> > > > > Signed-off-by: Chen-Yu Tsai <wens@csie.org>
> > > > > ---
> > > > > Changes since v1:
> > > > >
> > > > > - Moved the check to the beginning of dsa_switch_mdb_add()
> > > > >
> > > > > Looks like we could also move the ops check out of
> > > > > dsa_switch_mdb_prepare_bitmap(), though I suppose keeping the code the
> > > > > way it is now is clearer.
> > > > >
> > > > > ---
> > > > > net/dsa/switch.c | 3 +++
> > > > > 1 file changed, 3 insertions(+)
> > > > >
> > > > > diff --git a/net/dsa/switch.c b/net/dsa/switch.c
> > > > > index 4ec5b7f85d51..231af5268656 100644
> > > > > --- a/net/dsa/switch.c
> > > > > +++ b/net/dsa/switch.c
> > > > > @@ -164,6 +164,9 @@ static int dsa_switch_mdb_add(struct dsa_switch *ds,
> > > > > struct switchdev_trans *trans = info->trans;
> > > > > int port;
> > > > >
> > > > > + if (!ds->ops->port_mdb_add)
> > > > > + return -EOPNOTSUPP;
> > > > > +
> > > > > /* Build a mask of Multicast group members */
> > > > > bitmap_zero(ds->bitmap, ds->num_ports);
> > > > > if (ds->index == info->sw_index)
> > > > > --
> > > > > 2.20.1
> > > > >
> > > >
> > > > I don't understand the crash here, nor the fix. dsa_switch_mdb_add()
> > > > is supposed to be called through switchdev with a prepare phase,
> > > > which checks for ds->ops->port_mdb_add. Do you mean that a switchdev
> > > > MDB object is added somewhere without a prepare phase? If that's
> > > > the case, this is what the commit message must say. Then the
> > >
> > > I had pretty much zero understanding of how switchdev and dsa work.
> > > The symptom is a NULL pointer reference, resulting from an unsupported
> > > callback that was not checked before being called, as described above.
> > > And that is what I mean. A NULL pointer reference happened when it
> > > should not have.
> > >
> > > Based on what you just mentioned, yes it does look like an object was
> > > added without a prepare phase. Randomly looking through the net/dsa
> > > code, it seems only dsa_port_vid_add() does a prepare phase, judging
> > > by .ph_prepare being set. dsa_port_{vlan,mbd,fdb}_add directly call
> > > the add phase, without the prepare phase. So I'm guessing "supposed
> > > to be called with a prepare phase" is not quite accurate. This also
> > > exceeds the scope of the simple fix I had in mind.
> > >
> > > > ds->ops->port_mdb_add check must go where it is used, that is to say
> > > > at the beginning of dsa_switch_mdb_add_bitmap() (similarly to what
> > > > dsa_switch_mdb_prepare_bitmap() does), not in dsa_switch_mdb_add.
> > >
> > > Andrew asked me to move it to where it is now. Please take a look at
> > > v1 [2] if it's what you would like.
> > >
> > > I'm ok either way.
> >
> > I still cannot find in the code where a SWITCHDEV_OBJ_ID_PORT_MDB object
> > gets added without a prepare phase or a trans object, but it wouldn't hurt to
> > double check the presence of ds->ops->port_mdb_add before calling it anyway,
> > since a patch may actually bypass this prepare phase.
>
> I dug a bit more and I couldn't find it either. AFAICS the only place it
> gets called is through the notification chain invoked in
> switchdev_port_obj_add_now(),
> which specifically has prepare and commit phases.
>
> > Your v1 patch was a bit confusing and changed the signature of the function
> > at the same time. Please check the callback where it is used, like this:
> >
> > diff --git a/net/dsa/switch.c b/net/dsa/switch.c
> > index 4ec5b7f85d51..09d9286b27cc 100644
> > --- a/net/dsa/switch.c
> > +++ b/net/dsa/switch.c
> > @@ -153,6 +153,9 @@ static void dsa_switch_mdb_add_bitmap(struct dsa_switch *ds,
> > {
> > int port;
> >
> > + if (!ds->ops->port_mdb_add)
> > + return;
> > +
> > for_each_set_bit(port, bitmap, ds->num_ports)
> > ds->ops->port_mdb_add(ds, port, mdb);
> > }
> >
> >
> > This will be easier to maintain. Please provide a simpler commit message,
>
> OK. How about:
>
> The dsa framework has optional .port_mdb_{prepare,add,delete} callback fields
s/delete/del/
> for drivers to handle multicast database entries. When adding an entry, the
> framework goes through a prepare phase, then a commit phase. Drivers not
> providing these callbacks should be detected in the prepare phase.
>
> For some unknown reason, the guard in the prepare phase is insufficient,
I'd change the line above with something like:
DSA core may still bypass the bridge layer and call the dsa_port_mdb_add
function directly with no prepare phase or no switchdev trans object,
> and the framework ends up calling an undefined .port_mdb_add callback.
> This results in a NULL pointer dereference, as shown in the log below.
But yes you get the idea right ;-)
> Add a check for .port_mdb_add before calling it in dsa_switch_mdb_add_bitmap().
>
> <attach log>
>
> > this one is not relevant. Are you actually able to reproduce this stack
> > trace? If not, that is not necessary to add it to the commit message...
>
> Yes I can reproduce it. Just did on a vanilla v5.3-rc3 kernel.
>
> [ 71.827837] br0: port 1(eth0.1) entered blocking state
> [ 71.844072] br0: port 1(eth0.1) entered disabled state
> [ 71.849669] device eth0.1 entered promiscuous mode
> [ 71.865848] device eth0 entered promiscuous mode
> [ 71.876535] br0: port 2(wan) entered blocking state
> [ 71.881475] br0: port 2(wan) entered disabled state
> [ 71.888631] device wan entered promiscuous mode
> [ 71.914341] bcm53xx stmmac-0:1e wan: configuring for phy/gmii link mode
> [ 71.921505] 8021q: adding VLAN 0 to HW filter on device wan
> [ 71.930146] bcm53xx stmmac-0:1e wan: Link is Up - 1Gbps/Full - flow
> control rx/tx
> [ 71.932360] br0: port 3(lan1) entered blocking state
> [ 71.942751] br0: port 3(lan1) entered disabled state
> [ 71.950128] device lan1 entered promiscuous mode
> [ 71.970024] bcm53xx stmmac-0:1e lan1: configuring for phy/gmii link mode
> [ 71.977157] 8021q: adding VLAN 0 to HW filter on device lan1
> [ 71.988159] br0: port 4(lan2) entered blocking state
> [ 71.993158] br0: port 4(lan2) entered disabled state
> [ 72.001235] device lan2 entered promiscuous mode
> [ 72.020994] bcm53xx stmmac-0:1e lan2: configuring for phy/gmii link mode
> [ 72.028081] 8021q: adding VLAN 0 to HW filter on device lan2
> [ 72.035797] bcm53xx stmmac-0:1e lan2: Link is Up - 1Gbps/Full -
> flow control rx/tx
> [ 72.038767] br0: port 5(lan3) entered blocking state
> [ 72.048450] br0: port 5(lan3) entered disabled state
> [ 72.057129] device lan3 entered promiscuous mode
> [ 72.076956] bcm53xx stmmac-0:1e lan3: configuring for phy/gmii link mode
> [ 72.084004] 8021q: adding VLAN 0 to HW filter on device lan3
> [ 72.091352] bcm53xx stmmac-0:1e lan3: Link is Up - 1Gbps/Full -
> flow control rx/tx
> [ 72.095168] br0: port 6(lan4) entered blocking state
> [ 72.103926] br0: port 6(lan4) entered disabled state
> [ 72.113313] device lan4 entered promiscuous mode
> [ 72.133293] bcm53xx stmmac-0:1e lan4: configuring for phy/gmii link mode
> [ 72.140410] 8021q: adding VLAN 0 to HW filter on device lan4
> [ 72.147682] bcm53xx stmmac-0:1e lan4: Link is Up - 100Mbps/Full -
> flow control rx/tx
> [ 72.155915] br0: port 6(lan4) entered blocking state
> [ 72.160905] br0: port 6(lan4) entered forwarding state
> [ 72.166135] br0: port 5(lan3) entered blocking state
> [ 72.171111] br0: port 5(lan3) entered forwarding state
> [ 72.176300] br0: port 4(lan2) entered blocking state
> [ 72.181277] br0: port 4(lan2) entered forwarding state
> [ 72.186472] br0: port 2(wan) entered blocking state
> [ 72.191361] br0: port 2(wan) entered forwarding state
> [ 72.196453] br0: port 1(eth0.1) entered blocking state
> [ 72.201601] br0: port 1(eth0.1) entered forwarding state
>
> Waiting for br0 to get ready (MAXWAIT is 32 seconds).
> [ 72.244956] 8<--- cut here ---
> [ 72.248072] Unable to handle kernel NULL pointer dereference at
> virtual address 00000000
> [ 72.256331] pgd = (ptrval)
> [ 72.259048] [00000000] *pgd=00000000
> [ 72.262642] Internal error: Oops: 80000005 [#1] SMP ARM
> [ 72.267869] Modules linked in:
> [ 72.270934] CPU: 1 PID: 134 Comm: kworker/1:2 Not tainted 5.3.0-rc3 #1
> [ 72.277458] Hardware name: Allwinner sun7i (A20) Family
> [ 72.282699] Workqueue: events switchdev_deferred_process_work
> [ 72.288447] PC is at 0x0
> [ 72.290988] LR is at dsa_switch_event+0x570/0x620
> [ 72.295691] pc : [<00000000>] lr : [<c0853890>] psr: 80070113
> [ 72.301955] sp : ef3cfdb8 ip : 00000000 fp : ef0ad0a4
> [ 72.307180] r10: 0000000c r9 : 00000008 r8 : eeb3b450
> [ 72.312405] r7 : ef0ad040 r6 : ef0ad088 r5 : c0f04c48 r4 : ef0ad04c
> [ 72.318931] r3 : 00000000 r2 : eeb3b450 r1 : 00000008 r0 : ef0ad040
> [ 72.325459] Flags: Nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none
> [ 72.332593] Control: 10c5387d Table: 6de6406a DAC: 00000051
> [ 72.338340] Process kworker/1:2 (pid: 134, stack limit = 0x(ptrval))
> [ 72.344692] Stack: (0xef3cfdb8 to 0xef3d0000)
> [ 72.349085] fda0:
> ef3cfe14 46893f30
> [ 72.357329] fdc0: 00000000 ffffffff 00000000 ef3cfe14 00000005
> 00000000 c0852944 00000000
> [ 72.365550] fde0: ffffe000 c014be24 c0f04c48 ef3cfe98 c0f04c48
> ee9ea800 c08515d8 c014bf18
> [ 72.373739] fe00: 00000000 63a7c876 ee9b0068 c0850e60 ef26e800
> eeb3b450 ef3cfecb 00000000
> [ 72.381921] fe20: 00000008 46893f30 00000000 c087e718 eebe2068
> 46893f30 c0e72400 ffffffff
> [ 72.390102] fe40: 00000000 ef3cfe98 00000006 00000000 c0fb2a50
> c087e7a0 ffffffff c0852868
> [ 72.398283] fe60: ffffffff c014be24 00000006 c0fad2d0 ef3cfe98
> eeb3b450 00000000 c014c528
> [ 72.406463] fe80: 00000000 c015dcf8 c0f04c48 00000000 ee9ea800
> c087e484 ee9ea800 00000000
> [ 72.414644] fea0: eeb3b450 ef3cfecb 00000001 46893f30 00000000
> c0f04c48 00000000 c087e578
> [ 72.422825] fec0: 00000000 ef26e780 00024400 46893f30 eeb3b440
> eeb3b450 ee9ea800 00000122
> [ 72.430998] fee0: 00000100 c087e600 eeb3b440 c0fad2c8 c1003ef0
> c087e31c 2e928000 c0fad2ec
> [ 72.439179] ff00: c0fad2ec ee858080 ef7a62c0 ef7a9400 00000000
> c087e3c8 c0fad2ec c0144804
> [ 72.447360] ff20: ef312280 ef7a62c0 00000008 ee858080 ee858094
> ef7a62c0 00000008 c0f03d00
> [ 72.455541] ff40: ef7a62d8 ef7a62c0 ffffe000 c0145bac ffffe000
> c0fb2420 c0bfa1ec 00000000
> [ 72.463721] ff60: ffffe000 ee855c40 ee855c00 00000000 ef3ce000
> ee858080 c0145b68 ef0e5ea4
> [ 72.471902] ff80: ee855c5c c014a720 0000000b ee855c00 c014a5d8
> 00000000 00000000 00000000
> [ 72.480082] ffa0: 00000000 00000000 00000000 c01010e8 00000000
> 00000000 00000000 00000000
> [ 72.488262] ffc0: 00000000 00000000 00000000 00000000 00000000
> 00000000 00000000 00000000
> [ 72.496442] ffe0: 00000000 00000000 00000000 00000000 00000013
> 00000000 00000000 00000000
> [ 72.504652] [<c0853890>] (dsa_switch_event) from [<c014be24>]
> (notifier_call_chain+0x48/0x84)
> [ 72.513184] [<c014be24>] (notifier_call_chain) from [<c014bf18>]
> (raw_notifier_call_chain+0x18/0x20)
> [ 72.522321] [<c014bf18>] (raw_notifier_call_chain) from
> [<c0850e60>] (dsa_port_mdb_add+0x48/0x74)
> [ 72.531200] [<c0850e60>] (dsa_port_mdb_add) from [<c087e718>]
> (__switchdev_handle_port_obj_add+0x54/0xd4)
> [ 72.540773] [<c087e718>] (__switchdev_handle_port_obj_add) from
> [<c087e7a0>] (switchdev_handle_port_obj_add+0x8/0x14)
> [ 72.551385] [<c087e7a0>] (switchdev_handle_port_obj_add) from
> [<c0852868>] (dsa_slave_switchdev_blocking_event+0x94/0xa4)
> [ 72.562343] [<c0852868>] (dsa_slave_switchdev_blocking_event) from
> [<c014be24>] (notifier_call_chain+0x48/0x84)
> [ 72.572434] [<c014be24>] (notifier_call_chain) from [<c014c528>]
> (blocking_notifier_call_chain+0x50/0x68)
> [ 72.582004] [<c014c528>] (blocking_notifier_call_chain) from
> [<c087e484>] (switchdev_port_obj_notify+0x44/0xa8)
> [ 72.592094] [<c087e484>] (switchdev_port_obj_notify) from
> [<c087e578>] (switchdev_port_obj_add_now+0x90/0x104)
> [ 72.602098] [<c087e578>] (switchdev_port_obj_add_now) from
> [<c087e600>] (switchdev_port_obj_add_deferred+0x14/0x5c)
> [ 72.612537] [<c087e600>] (switchdev_port_obj_add_deferred) from
> [<c087e31c>] (switchdev_deferred_process+0x64/0x104)
> [ 72.623060] [<c087e31c>] (switchdev_deferred_process) from
> [<c087e3c8>] (switchdev_deferred_process_work+0xc/0x14)
> [ 72.633412] [<c087e3c8>] (switchdev_deferred_process_work) from
> [<c0144804>] (process_one_work+0x218/0x50c)
> [ 72.643156] [<c0144804>] (process_one_work) from [<c0145bac>]
> (worker_thread+0x44/0x5bc)
> [ 72.651253] [<c0145bac>] (worker_thread) from [<c014a720>]
> (kthread+0x148/0x150)
> [ 72.658657] [<c014a720>] (kthread) from [<c01010e8>]
> (ret_from_fork+0x14/0x2c)
> [ 72.665879] Exception stack(0xef3cffb0 to 0xef3cfff8)
> [ 72.670932] ffa0: 00000000
> 00000000 00000000 00000000
> [ 72.679112] ffc0: 00000000 00000000 00000000 00000000 00000000
> 00000000 00000000 00000000
> [ 72.687291] ffe0: 00000000 00000000 00000000 00000000 00000013 00000000
> [ 72.693911] Code: bad PC value
> [ 72.697022] ---[ end trace c7626868564873c8 ]---
Good to know!
Thank you,
Vivien
^ permalink raw reply
* Re: Slowness forming TIPC cluster with explicit node addresses
From: Chris Packham @ 2019-08-07 3:45 UTC (permalink / raw)
To: jon.maloy@ericsson.com, tipc-discussion@lists.sourceforge.net
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org
In-Reply-To: <CH2PR15MB35759E27F2A01FAE59AB66809AD40@CH2PR15MB3575.namprd15.prod.outlook.com>
Hi Jon,
On Wed, 2019-08-07 at 02:55 +0000, Jon Maloy wrote:
>
> >
> > -----Original Message-----
> > From: Chris Packham <Chris.Packham@alliedtelesis.co.nz>
> > Sent: 4-Aug-19 19:05
> > To: Jon Maloy <jon.maloy@ericsson.com>; tipc-
> > discussion@lists.sourceforge.net
> > Cc: netdev@vger.kernel.org; linux-kernel@vger.kernel.org
> > Subject: Re: Slowness forming TIPC cluster with explicit node
> > addresses
> >
> > On Sun, 2019-08-04 at 21:53 +0000, Jon Maloy wrote:
> > >
> > >
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: netdev-owner@vger.kernel.org <netdev-owner@vger.kernel.or
> > > > g>
> > On
> > >
> > > >
> > > > Behalf Of Chris Packham
> > > > Sent: 2-Aug-19 01:11
> > > > To: Jon Maloy <jon.maloy@ericsson.com>; tipc-
> > > > discussion@lists.sourceforge.net
> > > > Cc: netdev@vger.kernel.org; linux-kernel@vger.kernel.org
> > > > Subject: Re: Slowness forming TIPC cluster with explicit node
> > > > addresses
> > > >
> > > > On Mon, 2019-07-29 at 09:04 +1200, Chris Packham wrote:
> > > > >
> > > > >
> > > > > On Fri, 2019-07-26 at 13:31 +0000, Jon Maloy wrote:
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > -----Original Message-----
> > > > > > > From: netdev-owner@vger.kernel.org <netdev-
> > > > owner@vger.kernel.org>
> > > > >
> > > > >
> > > > > >
> > > > > >
> > > > > > >
> > > > > > >
> > > > > > > On Behalf Of Chris Packham
> > > > > > > Sent: 25-Jul-19 19:37
> > > > > > > To: tipc-discussion@lists.sourceforge.net
> > > > > > > Cc: netdev@vger.kernel.org; linux-kernel@vger.kernel.org
> > > > > > > Subject: Slowness forming TIPC cluster with explicit node
> > > > > > > addresses
> > > > > > >
> > > > > > > Hi,
> > > > > > >
> > > > > > > I'm having problems forming a TIPC cluster between 2
> > > > > > > nodes.
> > > > > > >
> > > > > > > This is the basic steps I'm going through on each node.
> > > > > > >
> > > > > > > modprobe tipc
> > > > > > > ip link set eth2 up
> > > > > > > tipc node set addr 1.1.5 # or 1.1.6 tipc bearer enable
> > > > > > > media
> > > > > > > eth dev eth0
> > > > > > eth2, I assume...
> > > > > >
> > > > > Yes sorry I keep switching between between Ethernet ports for
> > > > > testing
> > > > > so I hand edited the email.
> > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > Then to confirm if the cluster is formed I use tipc link
> > > > > > > list
> > > > > > >
> > > > > > > [root@node-5 ~]# tipc link list
> > > > > > > broadcast-link: up
> > > > > > > ...
> > > > > > >
> > > > > > > Looking at tcpdump the two nodes are sending packets
> > > > > > >
> > > > > > > 22:30:05.782320 TIPC v2.0 1.1.5 > 0.0.0, headerlength 60
> > > > > > > bytes,
> > > > > > > MessageSize
> > > > > > > 76 bytes, Neighbor Detection Protocol internal,
> > > > > > > messageType
> > > > > > > Link
> > > > > > > request
> > > > > > > 22:30:05.863555 TIPC v2.0 1.1.6 > 0.0.0, headerlength 60
> > > > > > > bytes,
> > > > > > > MessageSize
> > > > > > > 76 bytes, Neighbor Detection Protocol internal,
> > > > > > > messageType
> > > > > > > Link
> > > > > > > request
> > > > > > >
> > > > > > > Eventually (after a few minutes) the link does come up
> > > > > > >
> > > > > > > [root@node-6 ~]# tipc link list
> > > > > > > broadcast-link: up
> > > > > > > 1001006:eth2-1001005:eth2: up
> > > > > > >
> > > > > > > [root@node-5 ~]# tipc link list
> > > > > > > broadcast-link: up
> > > > > > > 1001005:eth2-1001006:eth2: up
> > > > > > >
> > > > > > > When I remove the "tipc node set addr" things seem to
> > > > > > > kick
> > > > > > > into
> > > > > > > life straight away
> > > > > > >
> > > > > > > [root@node-5 ~]# tipc link list
> > > > > > > broadcast-link: up
> > > > > > > 0050b61bd2aa:eth2-0050b61e6dfa:eth2: up
> > > > > > >
> > > > > > > So there appears to be some difference in behaviour
> > > > > > > between
> > > > > > > having
> > > > > > > an explicit node address and using the default.
> > > > > > > Unfortunately
> > > > > > > our
> > > > > > > application relies on setting the node addresses.
> > > > > > I do this many times a day, without any problems. If there
> > > > > > would be
> > > > > > any time difference, I would expect the 'auto configurable'
> > > > > > version
> > > > > > to be slower, because it involves a DAD step.
> > > > > > Are you sure you don't have any other nodes running in your
> > > > > > system?
> > > > > >
> > > > > > ///jon
> > > > > >
> > > > > Nope the two nodes are connected back to back. Does the
> > > > > number of
> > > > > Ethernet interfaces make a difference? As you can see I've
> > > > > got 3
> > > > > on
> > > > > each node. One is completely disconnected, one is for booting
> > > > > over
> > > > > TFTP
> > > > > (only used by U-boot) and the other is the USB Ethernet I'm
> > > > > using for
> > > > > testing.
> > > > >
> > > > So I can still reproduce this on nodes that only have one
> > > > network
> > > > interface and
> > > > are the only things connected.
> > > >
> > > > I did find one thing that helps
> > > >
> > > > diff --git a/net/tipc/discover.c b/net/tipc/discover.c index
> > > > c138d68e8a69..49921dad404a 100644
> > > > --- a/net/tipc/discover.c
> > > > +++ b/net/tipc/discover.c
> > > > @@ -358,10 +358,10 @@ int tipc_disc_create(struct net *net,
> > > > struct
> > > > tipc_bearer *b,
> > > > tipc_disc_init_msg(net, d->skb, DSC_REQ_MSG, b);
> > > >
> > > > /* Do we need an address trial period first ? */
> > > > - if (!tipc_own_addr(net)) {
> > > > +// if (!tipc_own_addr(net)) {
> > > > tn->addr_trial_end = jiffies +
> > > > msecs_to_jiffies(1000);
> > > > msg_set_type(buf_msg(d->skb), DSC_TRIAL_MSG);
> > > > - }
> > > > +// }
> > > > memcpy(&d->dest, dest, sizeof(*dest));
> > > > d->net = net;
> > > > d->bearer_id = b->identity;
> > > >
> > > > I think because with pre-configured addresses the duplicate
> > > > address
> > > > detection
> > > > is skipped the shorter init phase is skipped. Would is make
> > > > sense
> > > > to
> > > > unconditionally do the trial step? Or is there some better way
> > > > to
> > > > get things to
> > > > transition with pre-assigned addresses.
> > > I am on vacation until the end of next-week, so I can't give you
> > > any
> > > good analysis right now.
> > Thanks for taking the time to respond.
> >
> > >
> > > To do the trial step doesn’t make much sense to me, -it would
> > > only
> > > delay the setup unnecessarily (but with only 1 second).
> > > Can you check the initial value of addr_trial_end when there a
> > > pre-
> > > configured address?
> > I had the same thought. For both my devices 'addr_trial_end = 0' so
> > I
> > think tipc_disc_addr_trial_msg should end up with trial == false
> I suggest you try initializing it to jiffies and see what happens.
>
Setting addr_trial_end to jiffies seems to do the trick. I'll prepare a
patch and send it through.
^ permalink raw reply
* Re: [PATCH net v2] net: dsa: Check existence of .port_mdb_add callback before calling it
From: Chen-Yu Tsai @ 2019-08-07 3:18 UTC (permalink / raw)
To: Vivien Didelot
Cc: Chen-Yu Tsai, Andrew Lunn, Florian Fainelli, David S. Miller,
netdev, linux-kernel
In-Reply-To: <20190806163402.GB16656@t480s.localdomain>
On Wed, Aug 7, 2019 at 4:34 AM Vivien Didelot <vivien.didelot@gmail.com> wrote:
>
> Hi Chen-Yu,
>
> On Wed, 7 Aug 2019 01:49:37 +0800, Chen-Yu Tsai <wens@kernel.org> wrote:
> > On Wed, Aug 7, 2019 at 1:15 AM Vivien Didelot <vivien.didelot@gmail.com> wrote:
> > >
> > > Hi Chen-Yu,
> > >
> > > On Tue, 6 Aug 2019 15:53:25 +0800, Chen-Yu Tsai <wens@kernel.org> wrote:
> > > > From: Chen-Yu Tsai <wens@csie.org>
> > > >
> > > > With the recent addition of commit 75dad2520fc3 ("net: dsa: b53: Disable
> > > > all ports on setup"), users of b53 (BCM53125 on Lamobo R1 in my case)
> > > > are forced to use the dsa subsystem to enable the switch, instead of
> > > > having it in the default transparent "forward-to-all" mode.
> > > >
> > > > The b53 driver does not support mdb bitmap functions. However the dsa
> > > > layer does not check for the existence of the .port_mdb_add callback
> > > > before actually using it. This results in a NULL pointer dereference,
> > > > as shown in the kernel oops below.
> > > >
> > > > The other functions seem to be properly guarded. Do the same for
> > > > .port_mdb_add in dsa_switch_mdb_add_bitmap() as well.
> > > >
> > > > b53 is not the only driver that doesn't support mdb bitmap functions.
> > > > Others include bcm_sf2, dsa_loop, lantiq_gswip, mt7530, mv88e6060,
> > > > qca8k, realtek-smi, and vitesse-vsc73xx.
> > >
> > > I don't know what you mean by that, there's no "mdb bitmap function"
> > > support for drivers, only the port_mdb_{prepare,add,del} callbacks...
> >
> > The term was coined from commit e6db98db8a95 ("net: dsa: add switch mdb
> > bitmap functions"). But yeah, .port_mdb_* ops/callbacks would be more
> > appropriate.
> >
> > > > 8<--- cut here ---
> > > > Unable to handle kernel NULL pointer dereference at virtual address 00000000
> > > > pgd = (ptrval)
> > > > [00000000] *pgd=00000000
> > > > Internal error: Oops: 80000005 [#1] SMP ARM
> > > > Modules linked in: rtl8xxxu rtl8192cu rtl_usb rtl8192c_common rtlwifi mac80211 cfg80211
> > > > CPU: 1 PID: 134 Comm: kworker/1:2 Not tainted 5.3.0-rc1-00247-gd3519030752a #1
> > > > Hardware name: Allwinner sun7i (A20) Family
> > > > Workqueue: events switchdev_deferred_process_work
> > > > PC is at 0x0
> > > > LR is at dsa_switch_event+0x570/0x620
> > > > pc : [<00000000>] lr : [<c08533ec>] psr: 80070013
> > > > sp : ee871db8 ip : 00000000 fp : ee98d0a4
> > > > r10: 0000000c r9 : 00000008 r8 : ee89f710
> > > > r7 : ee98d040 r6 : ee98d088 r5 : c0f04c48 r4 : ee98d04c
> > > > r3 : 00000000 r2 : ee89f710 r1 : 00000008 r0 : ee98d040
> > > > Flags: Nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none
> > > > Control: 10c5387d Table: 6deb406a DAC: 00000051
> > > > Process kworker/1:2 (pid: 134, stack limit = 0x(ptrval))
> > > > Stack: (0xee871db8 to 0xee872000)
> > > > 1da0: ee871e14 103ace2d
> > > > 1dc0: 00000000 ffffffff 00000000 ee871e14 00000005 00000000 c08524a0 00000000
> > > > 1de0: ffffe000 c014bdfc c0f04c48 ee871e98 c0f04c48 ee9e5000 c0851120 c014bef0
> > > > 1e00: 00000000 b643aea2 ee9b4068 c08509a8 ee2bf940 ee89f710 ee871ecb 00000000
> > > > 1e20: 00000008 103ace2d 00000000 c087e248 ee29c868 103ace2d 00000001 ffffffff
> > > > 1e40: 00000000 ee871e98 00000006 00000000 c0fb2a50 c087e2d0 ffffffff c08523c4
> > > > 1e60: ffffffff c014bdfc 00000006 c0fad2d0 ee871e98 ee89f710 00000000 c014c500
> > > > 1e80: 00000000 ee89f3c0 c0f04c48 00000000 ee9e5000 c087dfb4 ee9e5000 00000000
> > > > 1ea0: ee89f710 ee871ecb 00000001 103ace2d 00000000 c0f04c48 00000000 c087e0a8
> > > > 1ec0: 00000000 efd9a3e0 0089f3c0 103ace2d ee89f700 ee89f710 ee9e5000 00000122
> > > > 1ee0: 00000100 c087e130 ee89f700 c0fad2c8 c1003ef0 c087de4c 2e928000 c0fad2ec
> > > > 1f00: c0fad2ec ee839580 ef7a62c0 ef7a9400 00000000 c087def8 c0fad2ec c01447dc
> > > > 1f20: ef315640 ef7a62c0 00000008 ee839580 ee839594 ef7a62c0 00000008 c0f03d00
> > > > 1f40: ef7a62d8 ef7a62c0 ffffe000 c0145b84 ffffe000 c0fb2420 c0bfaa8c 00000000
> > > > 1f60: ffffe000 ee84b600 ee84b5c0 00000000 ee870000 ee839580 c0145b40 ef0e5ea4
> > > > 1f80: ee84b61c c014a6f8 00000001 ee84b5c0 c014a5b0 00000000 00000000 00000000
> > > > 1fa0: 00000000 00000000 00000000 c01010e8 00000000 00000000 00000000 00000000
> > > > 1fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
> > > > 1fe0: 00000000 00000000 00000000 00000000 00000013 00000000 00000000 00000000
> > > > [<c08533ec>] (dsa_switch_event) from [<c014bdfc>] (notifier_call_chain+0x48/0x84)
> > > > [<c014bdfc>] (notifier_call_chain) from [<c014bef0>] (raw_notifier_call_chain+0x18/0x20)
> > > > [<c014bef0>] (raw_notifier_call_chain) from [<c08509a8>] (dsa_port_mdb_add+0x48/0x74)
> > > > [<c08509a8>] (dsa_port_mdb_add) from [<c087e248>] (__switchdev_handle_port_obj_add+0x54/0xd4)
> > > > [<c087e248>] (__switchdev_handle_port_obj_add) from [<c087e2d0>] (switchdev_handle_port_obj_add+0x8/0x14)
> > > > [<c087e2d0>] (switchdev_handle_port_obj_add) from [<c08523c4>] (dsa_slave_switchdev_blocking_event+0x94/0xa4)
> > > > [<c08523c4>] (dsa_slave_switchdev_blocking_event) from [<c014bdfc>] (notifier_call_chain+0x48/0x84)
> > > > [<c014bdfc>] (notifier_call_chain) from [<c014c500>] (blocking_notifier_call_chain+0x50/0x68)
> > > > [<c014c500>] (blocking_notifier_call_chain) from [<c087dfb4>] (switchdev_port_obj_notify+0x44/0xa8)
> > > > [<c087dfb4>] (switchdev_port_obj_notify) from [<c087e0a8>] (switchdev_port_obj_add_now+0x90/0x104)
> > > > [<c087e0a8>] (switchdev_port_obj_add_now) from [<c087e130>] (switchdev_port_obj_add_deferred+0x14/0x5c)
> > > > [<c087e130>] (switchdev_port_obj_add_deferred) from [<c087de4c>] (switchdev_deferred_process+0x64/0x104)
> > > > [<c087de4c>] (switchdev_deferred_process) from [<c087def8>] (switchdev_deferred_process_work+0xc/0x14)
> > > > [<c087def8>] (switchdev_deferred_process_work) from [<c01447dc>] (process_one_work+0x218/0x50c)
> > > > [<c01447dc>] (process_one_work) from [<c0145b84>] (worker_thread+0x44/0x5bc)
> > > > [<c0145b84>] (worker_thread) from [<c014a6f8>] (kthread+0x148/0x150)
> > > > [<c014a6f8>] (kthread) from [<c01010e8>] (ret_from_fork+0x14/0x2c)
> > > > Exception stack(0xee871fb0 to 0xee871ff8)
> > > > 1fa0: 00000000 00000000 00000000 00000000
> > > > 1fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
> > > > 1fe0: 00000000 00000000 00000000 00000000 00000013 00000000
> > > > Code: bad PC value
> > > > ---[ end trace 1292c61abd17b130 ]---
> > > >
> > > > [<c08533ec>] (dsa_switch_event) from [<c014bdfc>] (notifier_call_chain+0x48/0x84)
> > > > corresponds to
> > > >
> > > > $ arm-linux-gnueabihf-addr2line -C -i -e vmlinux c08533ec
> > > >
> > > > linux/net/dsa/switch.c:156
> > > > linux/net/dsa/switch.c:178
> > > > linux/net/dsa/switch.c:328
> > > >
> > > > Fixes: e6db98db8a95 ("net: dsa: add switch mdb bitmap functions")
> > > > Signed-off-by: Chen-Yu Tsai <wens@csie.org>
> > > > ---
> > > > Changes since v1:
> > > >
> > > > - Moved the check to the beginning of dsa_switch_mdb_add()
> > > >
> > > > Looks like we could also move the ops check out of
> > > > dsa_switch_mdb_prepare_bitmap(), though I suppose keeping the code the
> > > > way it is now is clearer.
> > > >
> > > > ---
> > > > net/dsa/switch.c | 3 +++
> > > > 1 file changed, 3 insertions(+)
> > > >
> > > > diff --git a/net/dsa/switch.c b/net/dsa/switch.c
> > > > index 4ec5b7f85d51..231af5268656 100644
> > > > --- a/net/dsa/switch.c
> > > > +++ b/net/dsa/switch.c
> > > > @@ -164,6 +164,9 @@ static int dsa_switch_mdb_add(struct dsa_switch *ds,
> > > > struct switchdev_trans *trans = info->trans;
> > > > int port;
> > > >
> > > > + if (!ds->ops->port_mdb_add)
> > > > + return -EOPNOTSUPP;
> > > > +
> > > > /* Build a mask of Multicast group members */
> > > > bitmap_zero(ds->bitmap, ds->num_ports);
> > > > if (ds->index == info->sw_index)
> > > > --
> > > > 2.20.1
> > > >
> > >
> > > I don't understand the crash here, nor the fix. dsa_switch_mdb_add()
> > > is supposed to be called through switchdev with a prepare phase,
> > > which checks for ds->ops->port_mdb_add. Do you mean that a switchdev
> > > MDB object is added somewhere without a prepare phase? If that's
> > > the case, this is what the commit message must say. Then the
> >
> > I had pretty much zero understanding of how switchdev and dsa work.
> > The symptom is a NULL pointer reference, resulting from an unsupported
> > callback that was not checked before being called, as described above.
> > And that is what I mean. A NULL pointer reference happened when it
> > should not have.
> >
> > Based on what you just mentioned, yes it does look like an object was
> > added without a prepare phase. Randomly looking through the net/dsa
> > code, it seems only dsa_port_vid_add() does a prepare phase, judging
> > by .ph_prepare being set. dsa_port_{vlan,mbd,fdb}_add directly call
> > the add phase, without the prepare phase. So I'm guessing "supposed
> > to be called with a prepare phase" is not quite accurate. This also
> > exceeds the scope of the simple fix I had in mind.
> >
> > > ds->ops->port_mdb_add check must go where it is used, that is to say
> > > at the beginning of dsa_switch_mdb_add_bitmap() (similarly to what
> > > dsa_switch_mdb_prepare_bitmap() does), not in dsa_switch_mdb_add.
> >
> > Andrew asked me to move it to where it is now. Please take a look at
> > v1 [2] if it's what you would like.
> >
> > I'm ok either way.
>
> I still cannot find in the code where a SWITCHDEV_OBJ_ID_PORT_MDB object
> gets added without a prepare phase or a trans object, but it wouldn't hurt to
> double check the presence of ds->ops->port_mdb_add before calling it anyway,
> since a patch may actually bypass this prepare phase.
I dug a bit more and I couldn't find it either. AFAICS the only place it
gets called is through the notification chain invoked in
switchdev_port_obj_add_now(),
which specifically has prepare and commit phases.
> Your v1 patch was a bit confusing and changed the signature of the function
> at the same time. Please check the callback where it is used, like this:
>
> diff --git a/net/dsa/switch.c b/net/dsa/switch.c
> index 4ec5b7f85d51..09d9286b27cc 100644
> --- a/net/dsa/switch.c
> +++ b/net/dsa/switch.c
> @@ -153,6 +153,9 @@ static void dsa_switch_mdb_add_bitmap(struct dsa_switch *ds,
> {
> int port;
>
> + if (!ds->ops->port_mdb_add)
> + return;
> +
> for_each_set_bit(port, bitmap, ds->num_ports)
> ds->ops->port_mdb_add(ds, port, mdb);
> }
>
>
> This will be easier to maintain. Please provide a simpler commit message,
OK. How about:
The dsa framework has optional .port_mdb_{prepare,add,delete} callback fields
for drivers to handle multicast database entries. When adding an entry, the
framework goes through a prepare phase, then a commit phase. Drivers not
providing these callbacks should be detected in the prepare phase.
For some unknown reason, the guard in the prepare phase is insufficient,
and the framework ends up calling an undefined .port_mdb_add callback.
This results in a NULL pointer dereference, as shown in the log below.
Add a check for .port_mdb_add before calling it in dsa_switch_mdb_add_bitmap().
<attach log>
> this one is not relevant. Are you actually able to reproduce this stack
> trace? If not, that is not necessary to add it to the commit message...
Yes I can reproduce it. Just did on a vanilla v5.3-rc3 kernel.
[ 71.827837] br0: port 1(eth0.1) entered blocking state
[ 71.844072] br0: port 1(eth0.1) entered disabled state
[ 71.849669] device eth0.1 entered promiscuous mode
[ 71.865848] device eth0 entered promiscuous mode
[ 71.876535] br0: port 2(wan) entered blocking state
[ 71.881475] br0: port 2(wan) entered disabled state
[ 71.888631] device wan entered promiscuous mode
[ 71.914341] bcm53xx stmmac-0:1e wan: configuring for phy/gmii link mode
[ 71.921505] 8021q: adding VLAN 0 to HW filter on device wan
[ 71.930146] bcm53xx stmmac-0:1e wan: Link is Up - 1Gbps/Full - flow
control rx/tx
[ 71.932360] br0: port 3(lan1) entered blocking state
[ 71.942751] br0: port 3(lan1) entered disabled state
[ 71.950128] device lan1 entered promiscuous mode
[ 71.970024] bcm53xx stmmac-0:1e lan1: configuring for phy/gmii link mode
[ 71.977157] 8021q: adding VLAN 0 to HW filter on device lan1
[ 71.988159] br0: port 4(lan2) entered blocking state
[ 71.993158] br0: port 4(lan2) entered disabled state
[ 72.001235] device lan2 entered promiscuous mode
[ 72.020994] bcm53xx stmmac-0:1e lan2: configuring for phy/gmii link mode
[ 72.028081] 8021q: adding VLAN 0 to HW filter on device lan2
[ 72.035797] bcm53xx stmmac-0:1e lan2: Link is Up - 1Gbps/Full -
flow control rx/tx
[ 72.038767] br0: port 5(lan3) entered blocking state
[ 72.048450] br0: port 5(lan3) entered disabled state
[ 72.057129] device lan3 entered promiscuous mode
[ 72.076956] bcm53xx stmmac-0:1e lan3: configuring for phy/gmii link mode
[ 72.084004] 8021q: adding VLAN 0 to HW filter on device lan3
[ 72.091352] bcm53xx stmmac-0:1e lan3: Link is Up - 1Gbps/Full -
flow control rx/tx
[ 72.095168] br0: port 6(lan4) entered blocking state
[ 72.103926] br0: port 6(lan4) entered disabled state
[ 72.113313] device lan4 entered promiscuous mode
[ 72.133293] bcm53xx stmmac-0:1e lan4: configuring for phy/gmii link mode
[ 72.140410] 8021q: adding VLAN 0 to HW filter on device lan4
[ 72.147682] bcm53xx stmmac-0:1e lan4: Link is Up - 100Mbps/Full -
flow control rx/tx
[ 72.155915] br0: port 6(lan4) entered blocking state
[ 72.160905] br0: port 6(lan4) entered forwarding state
[ 72.166135] br0: port 5(lan3) entered blocking state
[ 72.171111] br0: port 5(lan3) entered forwarding state
[ 72.176300] br0: port 4(lan2) entered blocking state
[ 72.181277] br0: port 4(lan2) entered forwarding state
[ 72.186472] br0: port 2(wan) entered blocking state
[ 72.191361] br0: port 2(wan) entered forwarding state
[ 72.196453] br0: port 1(eth0.1) entered blocking state
[ 72.201601] br0: port 1(eth0.1) entered forwarding state
Waiting for br0 to get ready (MAXWAIT is 32 seconds).
[ 72.244956] 8<--- cut here ---
[ 72.248072] Unable to handle kernel NULL pointer dereference at
virtual address 00000000
[ 72.256331] pgd = (ptrval)
[ 72.259048] [00000000] *pgd=00000000
[ 72.262642] Internal error: Oops: 80000005 [#1] SMP ARM
[ 72.267869] Modules linked in:
[ 72.270934] CPU: 1 PID: 134 Comm: kworker/1:2 Not tainted 5.3.0-rc3 #1
[ 72.277458] Hardware name: Allwinner sun7i (A20) Family
[ 72.282699] Workqueue: events switchdev_deferred_process_work
[ 72.288447] PC is at 0x0
[ 72.290988] LR is at dsa_switch_event+0x570/0x620
[ 72.295691] pc : [<00000000>] lr : [<c0853890>] psr: 80070113
[ 72.301955] sp : ef3cfdb8 ip : 00000000 fp : ef0ad0a4
[ 72.307180] r10: 0000000c r9 : 00000008 r8 : eeb3b450
[ 72.312405] r7 : ef0ad040 r6 : ef0ad088 r5 : c0f04c48 r4 : ef0ad04c
[ 72.318931] r3 : 00000000 r2 : eeb3b450 r1 : 00000008 r0 : ef0ad040
[ 72.325459] Flags: Nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none
[ 72.332593] Control: 10c5387d Table: 6de6406a DAC: 00000051
[ 72.338340] Process kworker/1:2 (pid: 134, stack limit = 0x(ptrval))
[ 72.344692] Stack: (0xef3cfdb8 to 0xef3d0000)
[ 72.349085] fda0:
ef3cfe14 46893f30
[ 72.357329] fdc0: 00000000 ffffffff 00000000 ef3cfe14 00000005
00000000 c0852944 00000000
[ 72.365550] fde0: ffffe000 c014be24 c0f04c48 ef3cfe98 c0f04c48
ee9ea800 c08515d8 c014bf18
[ 72.373739] fe00: 00000000 63a7c876 ee9b0068 c0850e60 ef26e800
eeb3b450 ef3cfecb 00000000
[ 72.381921] fe20: 00000008 46893f30 00000000 c087e718 eebe2068
46893f30 c0e72400 ffffffff
[ 72.390102] fe40: 00000000 ef3cfe98 00000006 00000000 c0fb2a50
c087e7a0 ffffffff c0852868
[ 72.398283] fe60: ffffffff c014be24 00000006 c0fad2d0 ef3cfe98
eeb3b450 00000000 c014c528
[ 72.406463] fe80: 00000000 c015dcf8 c0f04c48 00000000 ee9ea800
c087e484 ee9ea800 00000000
[ 72.414644] fea0: eeb3b450 ef3cfecb 00000001 46893f30 00000000
c0f04c48 00000000 c087e578
[ 72.422825] fec0: 00000000 ef26e780 00024400 46893f30 eeb3b440
eeb3b450 ee9ea800 00000122
[ 72.430998] fee0: 00000100 c087e600 eeb3b440 c0fad2c8 c1003ef0
c087e31c 2e928000 c0fad2ec
[ 72.439179] ff00: c0fad2ec ee858080 ef7a62c0 ef7a9400 00000000
c087e3c8 c0fad2ec c0144804
[ 72.447360] ff20: ef312280 ef7a62c0 00000008 ee858080 ee858094
ef7a62c0 00000008 c0f03d00
[ 72.455541] ff40: ef7a62d8 ef7a62c0 ffffe000 c0145bac ffffe000
c0fb2420 c0bfa1ec 00000000
[ 72.463721] ff60: ffffe000 ee855c40 ee855c00 00000000 ef3ce000
ee858080 c0145b68 ef0e5ea4
[ 72.471902] ff80: ee855c5c c014a720 0000000b ee855c00 c014a5d8
00000000 00000000 00000000
[ 72.480082] ffa0: 00000000 00000000 00000000 c01010e8 00000000
00000000 00000000 00000000
[ 72.488262] ffc0: 00000000 00000000 00000000 00000000 00000000
00000000 00000000 00000000
[ 72.496442] ffe0: 00000000 00000000 00000000 00000000 00000013
00000000 00000000 00000000
[ 72.504652] [<c0853890>] (dsa_switch_event) from [<c014be24>]
(notifier_call_chain+0x48/0x84)
[ 72.513184] [<c014be24>] (notifier_call_chain) from [<c014bf18>]
(raw_notifier_call_chain+0x18/0x20)
[ 72.522321] [<c014bf18>] (raw_notifier_call_chain) from
[<c0850e60>] (dsa_port_mdb_add+0x48/0x74)
[ 72.531200] [<c0850e60>] (dsa_port_mdb_add) from [<c087e718>]
(__switchdev_handle_port_obj_add+0x54/0xd4)
[ 72.540773] [<c087e718>] (__switchdev_handle_port_obj_add) from
[<c087e7a0>] (switchdev_handle_port_obj_add+0x8/0x14)
[ 72.551385] [<c087e7a0>] (switchdev_handle_port_obj_add) from
[<c0852868>] (dsa_slave_switchdev_blocking_event+0x94/0xa4)
[ 72.562343] [<c0852868>] (dsa_slave_switchdev_blocking_event) from
[<c014be24>] (notifier_call_chain+0x48/0x84)
[ 72.572434] [<c014be24>] (notifier_call_chain) from [<c014c528>]
(blocking_notifier_call_chain+0x50/0x68)
[ 72.582004] [<c014c528>] (blocking_notifier_call_chain) from
[<c087e484>] (switchdev_port_obj_notify+0x44/0xa8)
[ 72.592094] [<c087e484>] (switchdev_port_obj_notify) from
[<c087e578>] (switchdev_port_obj_add_now+0x90/0x104)
[ 72.602098] [<c087e578>] (switchdev_port_obj_add_now) from
[<c087e600>] (switchdev_port_obj_add_deferred+0x14/0x5c)
[ 72.612537] [<c087e600>] (switchdev_port_obj_add_deferred) from
[<c087e31c>] (switchdev_deferred_process+0x64/0x104)
[ 72.623060] [<c087e31c>] (switchdev_deferred_process) from
[<c087e3c8>] (switchdev_deferred_process_work+0xc/0x14)
[ 72.633412] [<c087e3c8>] (switchdev_deferred_process_work) from
[<c0144804>] (process_one_work+0x218/0x50c)
[ 72.643156] [<c0144804>] (process_one_work) from [<c0145bac>]
(worker_thread+0x44/0x5bc)
[ 72.651253] [<c0145bac>] (worker_thread) from [<c014a720>]
(kthread+0x148/0x150)
[ 72.658657] [<c014a720>] (kthread) from [<c01010e8>]
(ret_from_fork+0x14/0x2c)
[ 72.665879] Exception stack(0xef3cffb0 to 0xef3cfff8)
[ 72.670932] ffa0: 00000000
00000000 00000000 00000000
[ 72.679112] ffc0: 00000000 00000000 00000000 00000000 00000000
00000000 00000000 00000000
[ 72.687291] ffe0: 00000000 00000000 00000000 00000000 00000013 00000000
[ 72.693911] Code: bad PC value
[ 72.697022] ---[ end trace c7626868564873c8 ]---
ChenYu
^ permalink raw reply
* WARNING in cgroup_rstat_updated
From: syzbot @ 2019-08-07 3:18 UTC (permalink / raw)
To: linux-kernel, linux-mm, netdev, syzkaller-bugs
Hello,
syzbot found the following crash on:
HEAD commit: 31cc088a Merge tag 'drm-next-2019-07-19' of git://anongit...
git tree: net-next
console output: https://syzkaller.appspot.com/x/log.txt?x=102db48c600000
kernel config: https://syzkaller.appspot.com/x/.config?x=4dba67bf8b8c9ad7
dashboard link: https://syzkaller.appspot.com/bug?extid=370e4739fa489334a4ef
compiler: gcc (GCC) 9.0.0 20181231 (experimental)
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=16dd57dc600000
IMPORTANT: if you fix the bug, please add the following tag to the commit:
Reported-by: syzbot+370e4739fa489334a4ef@syzkaller.appspotmail.com
8021q: adding VLAN 0 to HW filter on device batadv0
WARNING: CPU: 1 PID: 9095 at mm/page_counter.c:62 page_counter_cancel
mm/page_counter.c:62 [inline]
WARNING: CPU: 1 PID: 9095 at mm/page_counter.c:62
page_counter_cancel+0x5a/0x70 mm/page_counter.c:55
Kernel panic - not syncing: panic_on_warn set ...
Shutting down cpus with NMI
Kernel Offset: disabled
======================================================
WARNING: possible circular locking dependency detected
5.2.0+ #67 Not tainted
------------------------------------------------------
syz-executor.2/9306 is trying to acquire lock:
00000000e4252251 ((console_sem).lock){-.-.}, at: down_trylock+0x13/0x70
kernel/locking/semaphore.c:135
but task is already holding lock:
000000000fdb8781 (per_cpu_ptr(&cgroup_rstat_cpu_lock, cpu)){-...}, at:
cgroup_rstat_updated+0x115/0x2f0 kernel/cgroup/rstat.c:49
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #3 (per_cpu_ptr(&cgroup_rstat_cpu_lock, cpu)){-...}:
__raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
_raw_spin_lock_irqsave+0x95/0xcd kernel/locking/spinlock.c:159
cgroup_rstat_updated+0x115/0x2f0 kernel/cgroup/rstat.c:49
cgroup_base_stat_cputime_account_end.isra.0+0x1d/0x60
kernel/cgroup/rstat.c:361
__cgroup_account_cputime+0x9e/0xd0 kernel/cgroup/rstat.c:371
cgroup_account_cputime include/linux/cgroup.h:782 [inline]
update_curr+0x3c8/0x8d0 kernel/sched/fair.c:862
dequeue_entity+0x1e/0x1100 kernel/sched/fair.c:4014
dequeue_task_fair+0x65/0x870 kernel/sched/fair.c:5306
dequeue_task+0x77/0x2e0 kernel/sched/core.c:1195
sched_move_task+0x1fb/0x350 kernel/sched/core.c:6847
cpu_cgroup_attach+0x6d/0xb0 kernel/sched/core.c:6970
cgroup_migrate_execute+0xc56/0x1350 kernel/cgroup/cgroup.c:2524
cgroup_migrate+0x14f/0x1f0 kernel/cgroup/cgroup.c:2780
cgroup_attach_task+0x57f/0x860 kernel/cgroup/cgroup.c:2817
cgroup_procs_write+0x340/0x400 kernel/cgroup/cgroup.c:4777
cgroup_file_write+0x241/0x790 kernel/cgroup/cgroup.c:3754
kernfs_fop_write+0x2b8/0x480 fs/kernfs/file.c:315
__vfs_write+0x8a/0x110 fs/read_write.c:494
vfs_write+0x268/0x5d0 fs/read_write.c:558
ksys_write+0x14f/0x290 fs/read_write.c:611
__do_sys_write fs/read_write.c:623 [inline]
__se_sys_write fs/read_write.c:620 [inline]
__x64_sys_write+0x73/0xb0 fs/read_write.c:620
do_syscall_64+0xfd/0x6a0 arch/x86/entry/common.c:296
entry_SYSCALL_64_after_hwframe+0x49/0xbe
-> #2 (&rq->lock){-.-.}:
__raw_spin_lock include/linux/spinlock_api_smp.h:142 [inline]
_raw_spin_lock+0x2f/0x40 kernel/locking/spinlock.c:151
rq_lock kernel/sched/sched.h:1207 [inline]
task_fork_fair+0x6a/0x520 kernel/sched/fair.c:9940
sched_fork+0x3af/0x900 kernel/sched/core.c:2783
copy_process+0x1b04/0x6b00 kernel/fork.c:1987
_do_fork+0x146/0xfa0 kernel/fork.c:2369
kernel_thread+0xbb/0xf0 kernel/fork.c:2456
rest_init+0x28/0x37b init/main.c:417
arch_call_rest_init+0xe/0x1b
start_kernel+0x912/0x951 init/main.c:785
x86_64_start_reservations+0x29/0x2b arch/x86/kernel/head64.c:472
x86_64_start_kernel+0x77/0x7b arch/x86/kernel/head64.c:453
secondary_startup_64+0xa4/0xb0 arch/x86/kernel/head_64.S:243
-> #1 (&p->pi_lock){-.-.}:
__raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
_raw_spin_lock_irqsave+0x95/0xcd kernel/locking/spinlock.c:159
try_to_wake_up+0xb0/0x1aa0 kernel/sched/core.c:2432
wake_up_process+0x10/0x20 kernel/sched/core.c:2548
__up.isra.0+0x136/0x1a0 kernel/locking/semaphore.c:261
up+0x9c/0xe0 kernel/locking/semaphore.c:186
__up_console_sem+0xb7/0x1c0 kernel/printk/printk.c:244
console_unlock+0x695/0xf10 kernel/printk/printk.c:2481
vprintk_emit+0x2a0/0x700 kernel/printk/printk.c:1986
vprintk_default+0x28/0x30 kernel/printk/printk.c:2013
vprintk_func+0x7e/0x189 kernel/printk/printk_safe.c:386
printk+0xba/0xed kernel/printk/printk.c:2046
check_stack_usage kernel/exit.c:765 [inline]
do_exit.cold+0x18b/0x314 kernel/exit.c:927
do_group_exit+0x135/0x360 kernel/exit.c:981
__do_sys_exit_group kernel/exit.c:992 [inline]
__se_sys_exit_group kernel/exit.c:990 [inline]
__x64_sys_exit_group+0x44/0x50 kernel/exit.c:990
do_syscall_64+0xfd/0x6a0 arch/x86/entry/common.c:296
entry_SYSCALL_64_after_hwframe+0x49/0xbe
-> #0 ((console_sem).lock){-.-.}:
check_prev_add kernel/locking/lockdep.c:2405 [inline]
check_prevs_add kernel/locking/lockdep.c:2507 [inline]
validate_chain kernel/locking/lockdep.c:2897 [inline]
__lock_acquire+0x25a9/0x4c30 kernel/locking/lockdep.c:3880
lock_acquire+0x190/0x410 kernel/locking/lockdep.c:4413
__raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
_raw_spin_lock_irqsave+0x95/0xcd kernel/locking/spinlock.c:159
down_trylock+0x13/0x70 kernel/locking/semaphore.c:135
__down_trylock_console_sem+0xa8/0x210 kernel/printk/printk.c:227
console_trylock+0x15/0xa0 kernel/printk/printk.c:2297
console_trylock_spinning kernel/printk/printk.c:1706 [inline]
vprintk_emit+0x283/0x700 kernel/printk/printk.c:1985
vprintk_default+0x28/0x30 kernel/printk/printk.c:2013
vprintk_func+0x7e/0x189 kernel/printk/printk_safe.c:386
printk+0xba/0xed kernel/printk/printk.c:2046
kasan_die_handler arch/x86/mm/kasan_init_64.c:254 [inline]
kasan_die_handler.cold+0x11/0x23 arch/x86/mm/kasan_init_64.c:249
notifier_call_chain+0xc2/0x230 kernel/notifier.c:95
__atomic_notifier_call_chain+0xa6/0x1a0 kernel/notifier.c:185
atomic_notifier_call_chain kernel/notifier.c:195 [inline]
notify_die+0xfb/0x180 kernel/notifier.c:551
do_general_protection+0x13d/0x300 arch/x86/kernel/traps.c:558
general_protection+0x1e/0x30 arch/x86/entry/entry_64.S:1181
cgroup_rstat_updated+0x174/0x2f0 kernel/cgroup/rstat.c:64
cgroup_base_stat_cputime_account_end.isra.0+0x1d/0x60
kernel/cgroup/rstat.c:361
__cgroup_account_cputime_field+0xd3/0x130 kernel/cgroup/rstat.c:395
cgroup_account_cputime_field include/linux/cgroup.h:797 [inline]
task_group_account_field kernel/sched/cputime.c:109 [inline]
account_system_index_time+0x1f7/0x390 kernel/sched/cputime.c:172
irqtime_account_process_tick.isra.0+0x386/0x490
kernel/sched/cputime.c:389
account_process_tick+0x27f/0x350 kernel/sched/cputime.c:484
update_process_times+0x25/0x80 kernel/time/timer.c:1637
tick_sched_handle+0xa2/0x190 kernel/time/tick-sched.c:167
tick_sched_timer+0x53/0x140 kernel/time/tick-sched.c:1296
__run_hrtimer kernel/time/hrtimer.c:1389 [inline]
__hrtimer_run_queues+0x364/0xe40 kernel/time/hrtimer.c:1451
hrtimer_interrupt+0x314/0x770 kernel/time/hrtimer.c:1509
local_apic_timer_interrupt arch/x86/kernel/apic/apic.c:1068 [inline]
smp_apic_timer_interrupt+0x160/0x610 arch/x86/kernel/apic/apic.c:1093
apic_timer_interrupt+0xf/0x20 arch/x86/entry/entry_64.S:828
other info that might help us debug this:
Chain exists of:
(console_sem).lock --> &rq->lock --> per_cpu_ptr(&cgroup_rstat_cpu_lock,
cpu)
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(per_cpu_ptr(&cgroup_rstat_cpu_lock, cpu));
lock(&rq->lock);
lock(per_cpu_ptr(&cgroup_rstat_cpu_lock,
cpu));
lock((console_sem).lock);
*** DEADLOCK ***
6 locks held by syz-executor.2/9306:
#0: 0000000032d2cedf (&sb->s_type->i_mutex_key#12){+.+.}, at: inode_lock
include/linux/fs.h:778 [inline]
#0: 0000000032d2cedf (&sb->s_type->i_mutex_key#12){+.+.}, at:
__sock_release+0x89/0x280 net/socket.c:589
#1: 000000002033d24d (sk_lock-AF_INET6){+.+.}, at: lock_sock
include/net/sock.h:1522 [inline]
#1: 000000002033d24d (sk_lock-AF_INET6){+.+.}, at: tcp_close+0x27/0x10e0
net/ipv4/tcp.c:2329
#2: 0000000067f2fc6a (rcu_read_lock){....}, at: tcp_bpf_unhash+0x0/0x390
net/ipv4/tcp_bpf.c:480
#3: 0000000067f2fc6a (rcu_read_lock){....}, at: arch_atomic64_add
arch/x86/include/asm/atomic64_64.h:46 [inline]
#3: 0000000067f2fc6a (rcu_read_lock){....}, at: atomic64_add
include/asm-generic/atomic-instrumented.h:873 [inline]
#3: 0000000067f2fc6a (rcu_read_lock){....}, at: account_group_system_time
include/linux/sched/cputime.h:154 [inline]
#3: 0000000067f2fc6a (rcu_read_lock){....}, at:
account_system_index_time+0xf7/0x390 kernel/sched/cputime.c:169
#4: 000000000fdb8781 (per_cpu_ptr(&cgroup_rstat_cpu_lock, cpu)){-...}, at:
cgroup_rstat_updated+0x115/0x2f0 kernel/cgroup/rstat.c:49
#5: 0000000067f2fc6a (rcu_read_lock){....}, at:
__atomic_notifier_call_chain+0x0/0x1a0 kernel/notifier.c:404
stack backtrace:
CPU: 0 PID: 9306 Comm: syz-executor.2 Not tainted 5.2.0+ #67
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
Google 01/01/2011
Call Trace:
<IRQ>
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0x172/0x1f0 lib/dump_stack.c:113
print_circular_bug.cold+0x163/0x172 kernel/locking/lockdep.c:1617
check_noncircular+0x345/0x3e0 kernel/locking/lockdep.c:1741
check_prev_add kernel/locking/lockdep.c:2405 [inline]
check_prevs_add kernel/locking/lockdep.c:2507 [inline]
validate_chain kernel/locking/lockdep.c:2897 [inline]
__lock_acquire+0x25a9/0x4c30 kernel/locking/lockdep.c:3880
lock_acquire+0x190/0x410 kernel/locking/lockdep.c:4413
__raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
_raw_spin_lock_irqsave+0x95/0xcd kernel/locking/spinlock.c:159
down_trylock+0x13/0x70 kernel/locking/semaphore.c:135
__down_trylock_console_sem+0xa8/0x210 kernel/printk/printk.c:227
console_trylock+0x15/0xa0 kernel/printk/printk.c:2297
console_trylock_spinning kernel/printk/printk.c:1706 [inline]
vprintk_emit+0x283/0x700 kernel/printk/printk.c:1985
vprintk_default+0x28/0x30 kernel/printk/printk.c:2013
vprintk_func+0x7e/0x189 kernel/printk/printk_safe.c:386
printk+0xba/0xed kernel/printk/printk.c:2046
kasan_die_handler arch/x86/mm/kasan_init_64.c:254 [inline]
kasan_die_handler.cold+0x11/0x23 arch/x86/mm/kasan_init_64.c:249
notifier_call_chain+0xc2/0x230 kernel/notifier.c:95
__atomic_notifier_call_chain+0xa6/0x1a0 kernel/notifier.c:185
atomic_notifier_call_chain kernel/notifier.c:195 [inline]
notify_die+0xfb/0x180 kernel/notifier.c:551
do_general_protection+0x13d/0x300 arch/x86/kernel/traps.c:558
general_protection+0x1e/0x30 arch/x86/entry/entry_64.S:1181
RIP: 0010:cgroup_rstat_updated+0x174/0x2f0 kernel/cgroup/rstat.c:64
Code: 00 fc ff df 48 8b 45 c0 48 c1 e8 03 4c 01 f8 48 89 45 c8 eb 60 e8 6c
e1 05 00 49 8d 7c 24 30 48 8b 55 d0 49 89 f9 49 c1 e9 03 <43> 80 3c 39 00
0f 85 00 01 00 00 49 8b 7c 24 30 48 89 7a 38 49 8d
RSP: 0018:ffff8880ae809c08 EFLAGS: 00010006
RAX: ffff88809378a480 RBX: 0000000000000000 RCX: ffffffff8159c5ca
RDX: ffff8880ae800000 RSI: ffffffff816ca374 RDI: 47ff8883313e8861
RBP: ffff8880ae809c58 R08: 0000000000000004 R09: 08fff1106627d10c
R10: ffffed1015d0136d R11: 0000000000000003 R12: 47ff8883313e8831
R13: ffff88807b60a280 R14: ffffffff8626cbf5 R15: dffffc0000000000
cgroup_base_stat_cputime_account_end.isra.0+0x1d/0x60
kernel/cgroup/rstat.c:361
__cgroup_account_cputime_field+0xd3/0x130 kernel/cgroup/rstat.c:395
cgroup_account_cputime_field include/linux/cgroup.h:797 [inline]
task_group_account_field kernel/sched/cputime.c:109 [inline]
account_system_index_time+0x1f7/0x390 kernel/sched/cputime.c:172
irqtime_account_process_tick.isra.0+0x386/0x490 kernel/sched/cputime.c:389
account_process_tick+0x27f/0x350 kernel/sched/cputime.c:484
update_process_times+0x25/0x80 kernel/time/timer.c:1637
tick_sched_handle+0xa2/0x190 kernel/time/tick-sched.c:167
tick_sched_timer+0x53/0x140 kernel/time/tick-sched.c:1296
__run_hrtimer kernel/time/hrtimer.c:1389 [inline]
__hrtimer_run_queues+0x364/0xe40 kernel/time/hrtimer.c:1451
hrtimer_interrupt+0x314/0x770 kernel/time/hrtimer.c:1509
local_apic_timer_interrupt arch/x86/kernel/apic/apic.c:1068 [inline]
smp_apic_timer_interrupt+0x160/0x610 arch/x86/kernel/apic/apic.c:1093
apic_timer_interrupt+0xf/0x20 arch/x86/entry/entry_64.S:828
</IRQ>
Rebooting in 86400 seconds..
---
This bug 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 bug report. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
syzbot can test patches for this bug, for details see:
https://goo.gl/tpsmEJ#testing-patches
^ 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