* [PATCH bpf-next 5/5] tools: bpftool: implement "bpftool btf show|list"
From: Quentin Monnet @ 2019-08-15 15:00 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
In-Reply-To: <20190815150019.8523-1-quentin.monnet@netronome.com>
Add a "btf list" (alias: "btf show") subcommand to bpftool in order to
dump all BTF objects loaded on a system.
When running the command, hash tables are built in bpftool to retrieve
all the associations between BTF objects and BPF maps and programs. This
allows for printing all such associations when listing the BTF objects.
The command is added at the top of the subcommands for "bpftool btf", so
that typing only "bpftool btf" also comes down to listing the programs.
We could not have this with the previous command ("dump"), which
required a BTF object id, so it should not break any previous behaviour.
This also makes the "btf" command behaviour consistent with "prog" or
"map".
Bash completion is updated to use "bpftool btf" instead of "bpftool
prog" to list the BTF ids, as it looks more consistent.
Example output (plain):
# bpftool btf show
9: size 2989B prog_ids 21 map_ids 15
17: size 2847B prog_ids 36 map_ids 30,29,28
26: size 2847B
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
.../bpf/bpftool/Documentation/bpftool-btf.rst | 7 +
tools/bpf/bpftool/bash-completion/bpftool | 20 +-
tools/bpf/bpftool/btf.c | 342 +++++++++++++++++-
3 files changed, 363 insertions(+), 6 deletions(-)
diff --git a/tools/bpf/bpftool/Documentation/bpftool-btf.rst b/tools/bpf/bpftool/Documentation/bpftool-btf.rst
index 6694a0fc8f99..39615f8e145b 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-btf.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-btf.rst
@@ -19,6 +19,7 @@ SYNOPSIS
BTF COMMANDS
=============
+| **bpftool** **btf** { **show** | **list** } [**id** *BTF_ID*]
| **bpftool** **btf dump** *BTF_SRC* [**format** *FORMAT*]
| **bpftool** **btf help**
|
@@ -29,6 +30,12 @@ BTF COMMANDS
DESCRIPTION
===========
+ **bpftool btf { show | list }** [**id** *BTF_ID*]
+ Show information about loaded BTF objects. If a BTF ID is
+ specified, show information only about given BTF object,
+ otherwise list all BTF objects currently loaded on the
+ system.
+
**bpftool btf dump** *BTF_SRC*
Dump BTF entries from a given *BTF_SRC*.
diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
index df16c5415444..2c0081121b2b 100644
--- a/tools/bpf/bpftool/bash-completion/bpftool
+++ b/tools/bpf/bpftool/bash-completion/bpftool
@@ -73,8 +73,8 @@ _bpftool_get_prog_tags()
_bpftool_get_btf_ids()
{
- COMPREPLY+=( $( compgen -W "$( bpftool -jp prog 2>&1 | \
- command sed -n 's/.*"btf_id": \(.*\),\?$/\1/p' )" -- "$cur" ) )
+ COMPREPLY+=( $( compgen -W "$( bpftool -jp btf 2>&1 | \
+ command sed -n 's/.*"id": \(.*\),$/\1/p' )" -- "$cur" ) )
}
_bpftool_get_obj_map_names()
@@ -674,7 +674,7 @@ _bpftool()
map)
_bpftool_get_map_ids
;;
- dump)
+ $command)
_bpftool_get_btf_ids
;;
esac
@@ -702,9 +702,21 @@ _bpftool()
;;
esac
;;
+ show|list)
+ case $prev in
+ $command)
+ COMPREPLY+=( $( compgen -W "id" -- "$cur" ) )
+ ;;
+ id)
+ _bpftool_get_btf_ids
+ ;;
+ esac
+ return 0
+ ;;
*)
[[ $prev == $object ]] && \
- COMPREPLY=( $( compgen -W 'dump help' -- "$cur" ) )
+ COMPREPLY=( $( compgen -W 'dump help show list' \
+ -- "$cur" ) )
;;
esac
;;
diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c
index 1b8ec91899e6..839b76af689c 100644
--- a/tools/bpf/bpftool/btf.c
+++ b/tools/bpf/bpftool/btf.c
@@ -11,6 +11,7 @@
#include <bpf.h>
#include <libbpf.h>
#include <linux/btf.h>
+#include <linux/hashtable.h>
#include "btf.h"
#include "json_writer.h"
@@ -35,6 +36,16 @@ static const char * const btf_kind_str[NR_BTF_KINDS] = {
[BTF_KIND_DATASEC] = "DATASEC",
};
+struct btf_attach_table {
+ DECLARE_HASHTABLE(table, 16);
+};
+
+struct btf_attach_point {
+ __u32 obj_id;
+ __u32 btf_id;
+ struct hlist_node hash;
+};
+
static const char *btf_int_enc_str(__u8 encoding)
{
switch (encoding) {
@@ -522,6 +533,330 @@ static int do_dump(int argc, char **argv)
return err;
}
+static int btf_parse_fd(int *argc, char ***argv)
+{
+ unsigned int id;
+ char *endptr;
+ int fd;
+
+ if (!is_prefix(*argv[0], "id")) {
+ p_err("expected 'id', got: '%s'?", **argv);
+ return -1;
+ }
+ NEXT_ARGP();
+
+ id = strtoul(**argv, &endptr, 0);
+ if (*endptr) {
+ p_err("can't parse %s as ID", **argv);
+ return -1;
+ }
+ NEXT_ARGP();
+
+ fd = bpf_btf_get_fd_by_id(id);
+ if (fd < 0)
+ p_err("can't get BTF object by id (%u): %s",
+ id, strerror(errno));
+
+ return fd;
+}
+
+static void delete_btf_table(struct btf_attach_table *tab)
+{
+ struct btf_attach_point *obj;
+ struct hlist_node *tmp;
+
+ unsigned int bkt;
+
+ hash_for_each_safe(tab->table, bkt, tmp, obj, hash) {
+ hash_del(&obj->hash);
+ free(obj);
+ }
+}
+
+static int
+build_btf_type_table(struct btf_attach_table *tab, enum bpf_obj_type type,
+ void *info, __u32 *len)
+{
+ static const char * const names[] = {
+ [BPF_OBJ_UNKNOWN] = "unknown",
+ [BPF_OBJ_PROG] = "prog",
+ [BPF_OBJ_MAP] = "map",
+ };
+ struct btf_attach_point *obj_node;
+ __u32 btf_id, id = 0;
+ int err;
+ int fd;
+
+ while (true) {
+ switch (type) {
+ case BPF_OBJ_PROG:
+ err = bpf_prog_get_next_id(id, &id);
+ break;
+ case BPF_OBJ_MAP:
+ err = bpf_map_get_next_id(id, &id);
+ break;
+ default:
+ err = -1;
+ p_err("unexpected object type: %d", type);
+ goto err_free;
+ }
+ if (err) {
+ if (errno == ENOENT) {
+ err = 0;
+ break;
+ }
+ p_err("can't get next %s: %s%s", names[type],
+ strerror(errno),
+ errno == EINVAL ? " -- kernel too old?" : "");
+ goto err_free;
+ }
+
+ switch (type) {
+ case BPF_OBJ_PROG:
+ fd = bpf_prog_get_fd_by_id(id);
+ break;
+ case BPF_OBJ_MAP:
+ fd = bpf_map_get_fd_by_id(id);
+ break;
+ default:
+ err = -1;
+ p_err("unexpected object type: %d", type);
+ goto err_free;
+ }
+ if (fd < 0) {
+ if (errno == ENOENT)
+ continue;
+ p_err("can't get %s by id (%u): %s", names[type], id,
+ strerror(errno));
+ err = -1;
+ goto err_free;
+ }
+
+ memset(info, 0, *len);
+ err = bpf_obj_get_info_by_fd(fd, info, len);
+ close(fd);
+ if (err) {
+ p_err("can't get %s info: %s", names[type],
+ strerror(errno));
+ goto err_free;
+ }
+
+ switch (type) {
+ case BPF_OBJ_PROG:
+ btf_id = ((struct bpf_prog_info *)info)->btf_id;
+ break;
+ case BPF_OBJ_MAP:
+ btf_id = ((struct bpf_map_info *)info)->btf_id;
+ break;
+ default:
+ err = -1;
+ p_err("unexpected object type: %d", type);
+ goto err_free;
+ }
+ if (!btf_id)
+ continue;
+
+ obj_node = calloc(1, sizeof(*obj_node));
+ if (!obj_node) {
+ p_err("failed to allocate memory: %s", strerror(errno));
+ goto err_free;
+ }
+
+ obj_node->obj_id = id;
+ obj_node->btf_id = btf_id;
+ hash_add(tab->table, &obj_node->hash, obj_node->btf_id);
+ }
+
+ return 0;
+
+err_free:
+ delete_btf_table(tab);
+ return err;
+}
+
+static int
+build_btf_tables(struct btf_attach_table *btf_prog_table,
+ struct btf_attach_table *btf_map_table)
+{
+ struct bpf_prog_info prog_info;
+ __u32 prog_len = sizeof(prog_info);
+ struct bpf_map_info map_info;
+ __u32 map_len = sizeof(map_info);
+ int err = 0;
+
+ err = build_btf_type_table(btf_prog_table, BPF_OBJ_PROG, &prog_info,
+ &prog_len);
+ if (err)
+ return err;
+
+ err = build_btf_type_table(btf_map_table, BPF_OBJ_MAP, &map_info,
+ &map_len);
+ if (err) {
+ delete_btf_table(btf_prog_table);
+ return err;
+ }
+
+ return 0;
+}
+
+static void
+show_btf_plain(struct bpf_btf_info *info, int fd,
+ struct btf_attach_table *btf_prog_table,
+ struct btf_attach_table *btf_map_table)
+{
+ struct btf_attach_point *obj;
+ int n;
+
+ printf("%u: ", info->id);
+ printf("size %uB", info->btf_size);
+
+ n = 0;
+ hash_for_each_possible(btf_prog_table->table, obj, hash, info->id) {
+ if (obj->btf_id == info->id)
+ printf("%s%u", n++ == 0 ? " prog_ids " : ",",
+ obj->obj_id);
+ }
+
+ n = 0;
+ hash_for_each_possible(btf_map_table->table, obj, hash, info->id) {
+ if (obj->btf_id == info->id)
+ printf("%s%u", n++ == 0 ? " map_ids " : ",",
+ obj->obj_id);
+ }
+
+ printf("\n");
+}
+
+static void
+show_btf_json(struct bpf_btf_info *info, int fd,
+ struct btf_attach_table *btf_prog_table,
+ struct btf_attach_table *btf_map_table)
+{
+ struct btf_attach_point *obj;
+
+ jsonw_start_object(json_wtr); /* btf object */
+ jsonw_uint_field(json_wtr, "id", info->id);
+ jsonw_uint_field(json_wtr, "size", info->btf_size);
+
+ jsonw_name(json_wtr, "prog_ids");
+ jsonw_start_array(json_wtr); /* prog_ids */
+ hash_for_each_possible(btf_prog_table->table, obj, hash,
+ info->id) {
+ if (obj->btf_id == info->id)
+ jsonw_uint(json_wtr, obj->obj_id);
+ }
+ jsonw_end_array(json_wtr); /* prog_ids */
+
+ jsonw_name(json_wtr, "map_ids");
+ jsonw_start_array(json_wtr); /* map_ids */
+ hash_for_each_possible(btf_map_table->table, obj, hash,
+ info->id) {
+ if (obj->btf_id == info->id)
+ jsonw_uint(json_wtr, obj->obj_id);
+ }
+ jsonw_end_array(json_wtr); /* map_ids */
+ jsonw_end_object(json_wtr); /* btf object */
+}
+
+static int
+show_btf(int fd, struct btf_attach_table *btf_prog_table,
+ struct btf_attach_table *btf_map_table)
+{
+ struct bpf_btf_info info = {};
+ __u32 len = sizeof(info);
+ int err;
+
+ err = bpf_obj_get_info_by_fd(fd, &info, &len);
+ if (err) {
+ p_err("can't get BTF object info: %s", strerror(errno));
+ return -1;
+ }
+
+ if (json_output)
+ show_btf_json(&info, fd, btf_prog_table, btf_map_table);
+ else
+ show_btf_plain(&info, fd, btf_prog_table, btf_map_table);
+
+ return 0;
+}
+
+static int do_show(int argc, char **argv)
+{
+ struct btf_attach_table btf_prog_table;
+ struct btf_attach_table btf_map_table;
+ int err, fd = -1;
+ __u32 id = 0;
+
+ if (argc == 2) {
+ fd = btf_parse_fd(&argc, &argv);
+ if (fd < 0)
+ return -1;
+ }
+
+ if (argc) {
+ if (fd >= 0)
+ close(fd);
+ return BAD_ARG();
+ }
+
+ hash_init(btf_prog_table.table);
+ hash_init(btf_map_table.table);
+ err = build_btf_tables(&btf_prog_table, &btf_map_table);
+ if (err) {
+ if (fd >= 0)
+ close(fd);
+ return err;
+ }
+
+ if (fd >= 0) {
+ err = show_btf(fd, &btf_prog_table, &btf_map_table);
+ close(fd);
+ goto exit_free;
+ }
+
+ if (json_output)
+ jsonw_start_array(json_wtr); /* root array */
+
+ while (true) {
+ err = bpf_btf_get_next_id(id, &id);
+ if (err) {
+ if (errno == ENOENT) {
+ err = 0;
+ break;
+ }
+ p_err("can't get next BTF object: %s%s",
+ strerror(errno),
+ errno == EINVAL ? " -- kernel too old?" : "");
+ err = -1;
+ break;
+ }
+
+ fd = bpf_btf_get_fd_by_id(id);
+ if (fd < 0) {
+ if (errno == ENOENT)
+ continue;
+ p_err("can't get BTF object by id (%u): %s",
+ id, strerror(errno));
+ err = -1;
+ break;
+ }
+
+ err = show_btf(fd, &btf_prog_table, &btf_map_table);
+ close(fd);
+ if (err)
+ break;
+ }
+
+ if (json_output)
+ jsonw_end_array(json_wtr); /* root array */
+
+exit_free:
+ delete_btf_table(&btf_prog_table);
+ delete_btf_table(&btf_map_table);
+
+ return err;
+}
+
static int do_help(int argc, char **argv)
{
if (json_output) {
@@ -530,7 +865,8 @@ static int do_help(int argc, char **argv)
}
fprintf(stderr,
- "Usage: %s btf dump BTF_SRC [format FORMAT]\n"
+ "Usage: %s btf { show | list } [id BTF_ID]\n"
+ " %s btf dump BTF_SRC [format FORMAT]\n"
" %s btf help\n"
"\n"
" BTF_SRC := { id BTF_ID | prog PROG | map MAP [{key | value | kv | all}] | file FILE }\n"
@@ -539,12 +875,14 @@ static int do_help(int argc, char **argv)
" " HELP_SPEC_PROGRAM "\n"
" " HELP_SPEC_OPTIONS "\n"
"",
- bin_name, bin_name);
+ bin_name, bin_name, bin_name);
return 0;
}
static const struct cmd cmds[] = {
+ { "show", do_show },
+ { "list", do_show },
{ "help", do_help },
{ "dump", do_dump },
{ 0 }
--
2.17.1
^ permalink raw reply related
* [PATCH bpf-next 4/5] libbpf: add bpf_btf_get_next_id() to cycle through BTF objects
From: Quentin Monnet @ 2019-08-15 15:00 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
In-Reply-To: <20190815150019.8523-1-quentin.monnet@netronome.com>
Add an API function taking a BTF object id and providing the id of the
next BTF object in the kernel. This can be used to list all BTF objects
loaded on the system.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
tools/lib/bpf/Makefile | 2 +-
tools/lib/bpf/bpf.c | 5 +++++
tools/lib/bpf/bpf.h | 1 +
tools/lib/bpf/libbpf.map | 5 +++++
4 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile
index 9312066a1ae3..f6c0295c6a5e 100644
--- a/tools/lib/bpf/Makefile
+++ b/tools/lib/bpf/Makefile
@@ -3,7 +3,7 @@
BPF_VERSION = 0
BPF_PATCHLEVEL = 0
-BPF_EXTRAVERSION = 4
+BPF_EXTRAVERSION = 5
MAKEFLAGS += --no-print-directory
diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index 1439e99c9be5..cbb933532981 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -593,6 +593,11 @@ int bpf_map_get_next_id(__u32 start_id, __u32 *next_id)
return bpf_obj_get_next_id(start_id, next_id, BPF_MAP_GET_NEXT_ID);
}
+int bpf_btf_get_next_id(__u32 start_id, __u32 *next_id)
+{
+ return bpf_obj_get_next_id(start_id, next_id, BPF_BTF_GET_NEXT_ID);
+}
+
int bpf_prog_get_fd_by_id(__u32 id)
{
union bpf_attr attr;
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index ff42ca043dc8..0db01334740f 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -156,6 +156,7 @@ LIBBPF_API int bpf_prog_test_run(int prog_fd, int repeat, void *data,
__u32 *retval, __u32 *duration);
LIBBPF_API int bpf_prog_get_next_id(__u32 start_id, __u32 *next_id);
LIBBPF_API int bpf_map_get_next_id(__u32 start_id, __u32 *next_id);
+LIBBPF_API int bpf_btf_get_next_id(__u32 start_id, __u32 *next_id);
LIBBPF_API int bpf_prog_get_fd_by_id(__u32 id);
LIBBPF_API int bpf_map_get_fd_by_id(__u32 id);
LIBBPF_API int bpf_btf_get_fd_by_id(__u32 id);
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index f9d316e873d8..664ce8e7a60e 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -184,3 +184,8 @@ LIBBPF_0.0.4 {
perf_buffer__new_raw;
perf_buffer__poll;
} LIBBPF_0.0.3;
+
+LIBBPF_0.0.5 {
+ global:
+ bpf_btf_get_next_id;
+} LIBBPF_0.0.4;
--
2.17.1
^ permalink raw reply related
* [PATCH bpf-next 3/5] libbpf: refactor bpf_*_get_next_id() functions
From: Quentin Monnet @ 2019-08-15 15:00 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
In-Reply-To: <20190815150019.8523-1-quentin.monnet@netronome.com>
In preparation for the introduction of a similar function for retrieving
the id of the next BTF object, consolidate the code from
bpf_prog_get_next_id() and bpf_map_get_next_id() in libbpf.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
tools/lib/bpf/bpf.c | 21 ++++++++-------------
1 file changed, 8 insertions(+), 13 deletions(-)
diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index c7d7993c44bb..1439e99c9be5 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -568,7 +568,7 @@ int bpf_prog_test_run_xattr(struct bpf_prog_test_run_attr *test_attr)
return ret;
}
-int bpf_prog_get_next_id(__u32 start_id, __u32 *next_id)
+static int bpf_obj_get_next_id(__u32 start_id, __u32 *next_id, int cmd)
{
union bpf_attr attr;
int err;
@@ -576,26 +576,21 @@ int bpf_prog_get_next_id(__u32 start_id, __u32 *next_id)
memset(&attr, 0, sizeof(attr));
attr.start_id = start_id;
- err = sys_bpf(BPF_PROG_GET_NEXT_ID, &attr, sizeof(attr));
+ err = sys_bpf(cmd, &attr, sizeof(attr));
if (!err)
*next_id = attr.next_id;
return err;
}
-int bpf_map_get_next_id(__u32 start_id, __u32 *next_id)
+int bpf_prog_get_next_id(__u32 start_id, __u32 *next_id)
{
- union bpf_attr attr;
- int err;
-
- memset(&attr, 0, sizeof(attr));
- attr.start_id = start_id;
-
- err = sys_bpf(BPF_MAP_GET_NEXT_ID, &attr, sizeof(attr));
- if (!err)
- *next_id = attr.next_id;
+ return bpf_obj_get_next_id(start_id, next_id, BPF_PROG_GET_NEXT_ID);
+}
- return err;
+int bpf_map_get_next_id(__u32 start_id, __u32 *next_id)
+{
+ return bpf_obj_get_next_id(start_id, next_id, BPF_MAP_GET_NEXT_ID);
}
int bpf_prog_get_fd_by_id(__u32 id)
--
2.17.1
^ permalink raw reply related
* [PATCH bpf-next 1/5] bpf: add new BPF_BTF_GET_NEXT_ID syscall command
From: Quentin Monnet @ 2019-08-15 15:00 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
In-Reply-To: <20190815150019.8523-1-quentin.monnet@netronome.com>
Add a new command for the bpf() system call: BPF_BTF_GET_NEXT_ID is used
to cycle through all BTF objects loaded on the system.
The motivation is to be able to inspect (list) all BTF objects presents
on the system.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
include/linux/bpf.h | 3 +++
include/uapi/linux/bpf.h | 1 +
kernel/bpf/btf.c | 4 ++--
kernel/bpf/syscall.c | 4 ++++
4 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index f9a506147c8a..279ea762c34e 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -24,6 +24,9 @@ struct seq_file;
struct btf;
struct btf_type;
+extern struct idr btf_idr;
+extern spinlock_t btf_idr_lock;
+
/* map is generic key/value storage optionally accesible by eBPF programs */
struct bpf_map_ops {
/* funcs callable from userspace (via syscall) */
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 4393bd4b2419..874bc5eefee1 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -106,6 +106,7 @@ enum bpf_cmd {
BPF_TASK_FD_QUERY,
BPF_MAP_LOOKUP_AND_DELETE_ELEM,
BPF_MAP_FREEZE,
+ BPF_BTF_GET_NEXT_ID,
};
enum bpf_map_type {
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 5fcc7a17eb5a..e716a64b2f7f 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -195,8 +195,8 @@
i < btf_type_vlen(struct_type); \
i++, member++)
-static DEFINE_IDR(btf_idr);
-static DEFINE_SPINLOCK(btf_idr_lock);
+DEFINE_IDR(btf_idr);
+DEFINE_SPINLOCK(btf_idr_lock);
struct btf {
void *data;
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 5d141f16f6fa..407b7f840874 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -2874,6 +2874,10 @@ SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, siz
err = bpf_obj_get_next_id(&attr, uattr,
&map_idr, &map_idr_lock);
break;
+ case BPF_BTF_GET_NEXT_ID:
+ err = bpf_obj_get_next_id(&attr, uattr,
+ &btf_idr, &btf_idr_lock);
+ break;
case BPF_PROG_GET_FD_BY_ID:
err = bpf_prog_get_fd_by_id(&attr);
break;
--
2.17.1
^ permalink raw reply related
* [PATCH bpf-next 0/5] bpf: list BTF objects loaded on system
From: Quentin Monnet @ 2019-08-15 15:00 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
Hi,
This set adds a new command BPF_BTF_GET_NEXT_ID to the bpf() system call,
adds the relevant API function in libbpf, and uses it in bpftool to list
all BTF objects loaded on the system (and to dump the ids of maps and
programs associated with them, if any).
The main motivation of listing BTF objects is introspection and debugging
purposes. By getting BPF program and map information, it should already be
possible to list all BTF objects associated to at least one map or one
program. But there may be unattached BTF objects, held by a file descriptor
from a user space process only, and we may want to list them too.
As a side note, it also turned useful for examining the BTF objects
attached to offloaded programs, which would not show in program information
because the BTF id is not copied when retrieving such info. A fix is in
progress on that side.
Quentin Monnet (5):
bpf: add new BPF_BTF_GET_NEXT_ID syscall command
tools: bpf: synchronise BPF UAPI header with tools
libbpf: refactor bpf_*_get_next_id() functions
libbpf: add bpf_btf_get_next_id() to cycle through BTF objects
tools: bpftool: implement "bpftool btf show|list"
include/linux/bpf.h | 3 +
include/uapi/linux/bpf.h | 1 +
kernel/bpf/btf.c | 4 +-
kernel/bpf/syscall.c | 4 +
.../bpf/bpftool/Documentation/bpftool-btf.rst | 7 +
tools/bpf/bpftool/bash-completion/bpftool | 20 +-
tools/bpf/bpftool/btf.c | 342 +++++++++++++++++-
tools/include/uapi/linux/bpf.h | 1 +
tools/lib/bpf/Makefile | 2 +-
tools/lib/bpf/bpf.c | 24 +-
tools/lib/bpf/bpf.h | 1 +
tools/lib/bpf/libbpf.map | 5 +
12 files changed, 393 insertions(+), 21 deletions(-)
--
2.17.1
^ permalink raw reply
* Re: [PATCH net-next] page_pool: remove unnecessary variable init
From: Jesper Dangaard Brouer @ 2019-08-15 14:56 UTC (permalink / raw)
To: Yunsheng Lin; +Cc: brouer, hawk, ilias.apalodimas, davem, netdev, linux-kernel
In-Reply-To: <1565872860-63524-1-git-send-email-linyunsheng@huawei.com>
On Thu, 15 Aug 2019 20:41:00 +0800
Yunsheng Lin <linyunsheng@huawei.com> wrote:
> Remove variable initializations in functions that
> are followed by assignments before use
>
> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
^ permalink raw reply
* [PATCH net-next v2 4/4] rds: check for excessive looping in rds_send_xmit
From: Gerd Rausch @ 2019-08-15 14:43 UTC (permalink / raw)
To: Santosh Shilimkar, netdev, linux-rdma, rds-devel; +Cc: David Miller
In-Reply-To: <cover.1565879451.git.gerd.rausch@oracle.com>
From: Andy Grover <andy.grover@oracle.com>
Date: Thu, 13 Jan 2011 11:40:31 -0800
Original commit from 2011 updated to include a change by
Yuval Shaia <yuval.shaia@oracle.com>
that adds a new statistic counter "send_stuck_rm"
to capture the messages looping exessively
in the send path.
Signed-off-by: Gerd Rausch <gerd.rausch@oracle.com>
---
net/rds/rds.h | 2 +-
net/rds/send.c | 12 ++++++++++++
net/rds/stats.c | 1 +
3 files changed, 14 insertions(+), 1 deletion(-)
diff --git a/net/rds/rds.h b/net/rds/rds.h
index f0066d168499..ad605fd61655 100644
--- a/net/rds/rds.h
+++ b/net/rds/rds.h
@@ -717,7 +717,7 @@ struct rds_statistics {
uint64_t s_cong_send_blocked;
uint64_t s_recv_bytes_added_to_socket;
uint64_t s_recv_bytes_removed_from_socket;
-
+ uint64_t s_send_stuck_rm;
};
/* af_rds.c */
diff --git a/net/rds/send.c b/net/rds/send.c
index 031b1e97a466..9ce552abf9e9 100644
--- a/net/rds/send.c
+++ b/net/rds/send.c
@@ -145,6 +145,7 @@ int rds_send_xmit(struct rds_conn_path *cp)
LIST_HEAD(to_be_dropped);
int batch_count;
unsigned long send_gen = 0;
+ int same_rm = 0;
restart:
batch_count = 0;
@@ -200,6 +201,17 @@ int rds_send_xmit(struct rds_conn_path *cp)
rm = cp->cp_xmit_rm;
+ if (!rm) {
+ same_rm = 0;
+ } else {
+ same_rm++;
+ if (same_rm >= 4096) {
+ rds_stats_inc(s_send_stuck_rm);
+ ret = -EAGAIN;
+ break;
+ }
+ }
+
/*
* If between sending messages, we can send a pending congestion
* map update.
diff --git a/net/rds/stats.c b/net/rds/stats.c
index 6bbab4d74c4f..9e87da43c004 100644
--- a/net/rds/stats.c
+++ b/net/rds/stats.c
@@ -78,6 +78,7 @@ static const char *const rds_stat_names[] = {
"cong_send_blocked",
"recv_bytes_added_to_sock",
"recv_bytes_freed_fromsock",
+ "send_stuck_rm",
};
void rds_stats_info_copy(struct rds_info_iterator *iter,
--
2.22.1
^ permalink raw reply related
* [PATCH net-next v2 3/4] net/rds: Add a few missing rds_stat_names entries
From: Gerd Rausch @ 2019-08-15 14:42 UTC (permalink / raw)
To: Santosh Shilimkar, netdev, linux-rdma, rds-devel; +Cc: David Miller
In-Reply-To: <cover.1565879451.git.gerd.rausch@oracle.com>
From: Gerd Rausch <gerd.rausch@oracle.com>
Date: Thu, 11 Jul 2019 12:15:50 -0700
In a previous commit, fields were added to "struct rds_statistics"
but array "rds_stat_names" was not updated accordingly.
Please note the inconsistent naming of the string representations
that is done in the name of compatibility
with the Oracle internal code-base.
s_recv_bytes_added_to_socket -> "recv_bytes_added_to_sock"
s_recv_bytes_removed_from_socket -> "recv_bytes_freed_fromsock"
Fixes: 192a798f5299 ("RDS: add stat for socket recv memory usage")
Signed-off-by: Gerd Rausch <gerd.rausch@oracle.com>
---
net/rds/stats.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/net/rds/stats.c b/net/rds/stats.c
index 73be187d389e..6bbab4d74c4f 100644
--- a/net/rds/stats.c
+++ b/net/rds/stats.c
@@ -76,6 +76,8 @@ static const char *const rds_stat_names[] = {
"cong_update_received",
"cong_send_error",
"cong_send_blocked",
+ "recv_bytes_added_to_sock",
+ "recv_bytes_freed_fromsock",
};
void rds_stats_info_copy(struct rds_info_iterator *iter,
--
2.22.1
^ permalink raw reply related
* [net-next v2 1/1] tipc: clean up skb list lock handling on send path
From: Jon Maloy @ 2019-08-15 14:42 UTC (permalink / raw)
To: davem, netdev
Cc: tung.q.nguyen, hoang.h.le, jon.maloy, lxin, shuali, ying.xue,
edumazet, tipc-discussion
The policy for handling the skb list locks on the send and receive paths
is simple.
- On the send path we never need to grab the lock on the 'xmitq' list
when the destination is an exernal node.
- On the receive path we always need to grab the lock on the 'inputq'
list, irrespective of source node.
However, when transmitting node local messages those will eventually
end up on the receive path of a local socket, meaning that the argument
'xmitq' in tipc_node_xmit() will become the 'ínputq' argument in the
function tipc_sk_rcv(). This has been handled by always initializing
the spinlock of the 'xmitq' list at message creation, just in case it
may end up on the receive path later, and despite knowing that the lock
in most cases never will be used.
This approach is inaccurate and confusing, and has also concealed the
fact that the stated 'no lock grabbing' policy for the send path is
violated in some cases.
We now clean up this by never initializing the lock at message creation,
instead doing this at the moment we find that the message actually will
enter the receive path. At the same time we fix the four locations
where we incorrectly access the spinlock on the send/error path.
This patch also reverts commit d12cffe9329f ("tipc: ensure head->lock
is initialised") which has now become redundant.
CC: Eric Dumazet <edumazet@google.com>
Reported-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
---
v2: removed more unnecessary lock initializations after feedback
from Xin Long.
---
net/tipc/bcast.c | 10 +++++-----
net/tipc/group.c | 4 ++--
net/tipc/link.c | 14 +++++++-------
net/tipc/name_distr.c | 2 +-
net/tipc/node.c | 7 ++++---
net/tipc/socket.c | 14 +++++++-------
6 files changed, 26 insertions(+), 25 deletions(-)
diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c
index 34f3e56..6ef1abd 100644
--- a/net/tipc/bcast.c
+++ b/net/tipc/bcast.c
@@ -185,7 +185,7 @@ static void tipc_bcbase_xmit(struct net *net, struct sk_buff_head *xmitq)
}
/* We have to transmit across all bearers */
- skb_queue_head_init(&_xmitq);
+ __skb_queue_head_init(&_xmitq);
for (bearer_id = 0; bearer_id < MAX_BEARERS; bearer_id++) {
if (!bb->dests[bearer_id])
continue;
@@ -256,7 +256,7 @@ static int tipc_bcast_xmit(struct net *net, struct sk_buff_head *pkts,
struct sk_buff_head xmitq;
int rc = 0;
- skb_queue_head_init(&xmitq);
+ __skb_queue_head_init(&xmitq);
tipc_bcast_lock(net);
if (tipc_link_bc_peers(l))
rc = tipc_link_xmit(l, pkts, &xmitq);
@@ -286,7 +286,7 @@ static int tipc_rcast_xmit(struct net *net, struct sk_buff_head *pkts,
u32 dnode, selector;
selector = msg_link_selector(buf_msg(skb_peek(pkts)));
- skb_queue_head_init(&_pkts);
+ __skb_queue_head_init(&_pkts);
list_for_each_entry_safe(dst, tmp, &dests->list, list) {
dnode = dst->node;
@@ -344,7 +344,7 @@ static int tipc_mcast_send_sync(struct net *net, struct sk_buff *skb,
msg_set_size(_hdr, MCAST_H_SIZE);
msg_set_is_rcast(_hdr, !msg_is_rcast(hdr));
- skb_queue_head_init(&tmpq);
+ __skb_queue_head_init(&tmpq);
__skb_queue_tail(&tmpq, _skb);
if (method->rcast)
tipc_bcast_xmit(net, &tmpq, cong_link_cnt);
@@ -378,7 +378,7 @@ int tipc_mcast_xmit(struct net *net, struct sk_buff_head *pkts,
int rc = 0;
skb_queue_head_init(&inputq);
- skb_queue_head_init(&localq);
+ __skb_queue_head_init(&localq);
/* Clone packets before they are consumed by next call */
if (dests->local && !tipc_msg_reassemble(pkts, &localq)) {
diff --git a/net/tipc/group.c b/net/tipc/group.c
index 5f98d38..89257e2 100644
--- a/net/tipc/group.c
+++ b/net/tipc/group.c
@@ -199,7 +199,7 @@ void tipc_group_join(struct net *net, struct tipc_group *grp, int *sk_rcvbuf)
struct tipc_member *m, *tmp;
struct sk_buff_head xmitq;
- skb_queue_head_init(&xmitq);
+ __skb_queue_head_init(&xmitq);
rbtree_postorder_for_each_entry_safe(m, tmp, tree, tree_node) {
tipc_group_proto_xmit(grp, m, GRP_JOIN_MSG, &xmitq);
tipc_group_update_member(m, 0);
@@ -435,7 +435,7 @@ bool tipc_group_cong(struct tipc_group *grp, u32 dnode, u32 dport,
return true;
if (state == MBR_PENDING && adv == ADV_IDLE)
return true;
- skb_queue_head_init(&xmitq);
+ __skb_queue_head_init(&xmitq);
tipc_group_proto_xmit(grp, m, GRP_ADV_MSG, &xmitq);
tipc_node_distr_xmit(grp->net, &xmitq);
return true;
diff --git a/net/tipc/link.c b/net/tipc/link.c
index dd3155b..289e848 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -959,7 +959,7 @@ int tipc_link_xmit(struct tipc_link *l, struct sk_buff_head *list,
pr_warn("Too large msg, purging xmit list %d %d %d %d %d!\n",
skb_queue_len(list), msg_user(hdr),
msg_type(hdr), msg_size(hdr), mtu);
- skb_queue_purge(list);
+ __skb_queue_purge(list);
return -EMSGSIZE;
}
@@ -988,7 +988,7 @@ int tipc_link_xmit(struct tipc_link *l, struct sk_buff_head *list,
if (likely(skb_queue_len(transmq) < maxwin)) {
_skb = skb_clone(skb, GFP_ATOMIC);
if (!_skb) {
- skb_queue_purge(list);
+ __skb_queue_purge(list);
return -ENOBUFS;
}
__skb_dequeue(list);
@@ -1668,7 +1668,7 @@ void tipc_link_create_dummy_tnl_msg(struct tipc_link *l,
struct sk_buff *skb;
u32 dnode = l->addr;
- skb_queue_head_init(&tnlq);
+ __skb_queue_head_init(&tnlq);
skb = tipc_msg_create(TUNNEL_PROTOCOL, FAILOVER_MSG,
INT_H_SIZE, BASIC_H_SIZE,
dnode, onode, 0, 0, 0);
@@ -1708,9 +1708,9 @@ void tipc_link_tnl_prepare(struct tipc_link *l, struct tipc_link *tnl,
if (!tnl)
return;
- skb_queue_head_init(&tnlq);
- skb_queue_head_init(&tmpxq);
- skb_queue_head_init(&frags);
+ __skb_queue_head_init(&tnlq);
+ __skb_queue_head_init(&tmpxq);
+ __skb_queue_head_init(&frags);
/* At least one packet required for safe algorithm => add dummy */
skb = tipc_msg_create(TIPC_LOW_IMPORTANCE, TIPC_DIRECT_MSG,
@@ -1720,7 +1720,7 @@ void tipc_link_tnl_prepare(struct tipc_link *l, struct tipc_link *tnl,
pr_warn("%sunable to create tunnel packet\n", link_co_err);
return;
}
- skb_queue_tail(&tnlq, skb);
+ __skb_queue_tail(&tnlq, skb);
tipc_link_xmit(l, &tnlq, &tmpxq);
__skb_queue_purge(&tmpxq);
diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c
index 44abc8e..61219f0 100644
--- a/net/tipc/name_distr.c
+++ b/net/tipc/name_distr.c
@@ -190,7 +190,7 @@ void tipc_named_node_up(struct net *net, u32 dnode)
struct name_table *nt = tipc_name_table(net);
struct sk_buff_head head;
- skb_queue_head_init(&head);
+ __skb_queue_head_init(&head);
read_lock_bh(&nt->cluster_scope_lock);
named_distribute(net, &head, dnode, &nt->cluster_scope);
diff --git a/net/tipc/node.c b/net/tipc/node.c
index 1bdcf0f..c8f6177 100644
--- a/net/tipc/node.c
+++ b/net/tipc/node.c
@@ -1444,13 +1444,14 @@ int tipc_node_xmit(struct net *net, struct sk_buff_head *list,
if (in_own_node(net, dnode)) {
tipc_loopback_trace(net, list);
+ spin_lock_init(&list->lock);
tipc_sk_rcv(net, list);
return 0;
}
n = tipc_node_find(net, dnode);
if (unlikely(!n)) {
- skb_queue_purge(list);
+ __skb_queue_purge(list);
return -EHOSTUNREACH;
}
@@ -1459,7 +1460,7 @@ int tipc_node_xmit(struct net *net, struct sk_buff_head *list,
if (unlikely(bearer_id == INVALID_BEARER_ID)) {
tipc_node_read_unlock(n);
tipc_node_put(n);
- skb_queue_purge(list);
+ __skb_queue_purge(list);
return -EHOSTUNREACH;
}
@@ -1491,7 +1492,7 @@ int tipc_node_xmit_skb(struct net *net, struct sk_buff *skb, u32 dnode,
{
struct sk_buff_head head;
- skb_queue_head_init(&head);
+ __skb_queue_head_init(&head);
__skb_queue_tail(&head, skb);
tipc_node_xmit(net, &head, dnode, selector);
return 0;
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index 83ae41d..3b9f8cc 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -809,7 +809,7 @@ static int tipc_sendmcast(struct socket *sock, struct tipc_name_seq *seq,
msg_set_nameupper(hdr, seq->upper);
/* Build message as chain of buffers */
- skb_queue_head_init(&pkts);
+ __skb_queue_head_init(&pkts);
rc = tipc_msg_build(hdr, msg, 0, dlen, mtu, &pkts);
/* Send message if build was successful */
@@ -853,7 +853,7 @@ static int tipc_send_group_msg(struct net *net, struct tipc_sock *tsk,
msg_set_grp_bc_seqno(hdr, bc_snd_nxt);
/* Build message as chain of buffers */
- skb_queue_head_init(&pkts);
+ __skb_queue_head_init(&pkts);
mtu = tipc_node_get_mtu(net, dnode, tsk->portid);
rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts);
if (unlikely(rc != dlen))
@@ -1058,7 +1058,7 @@ static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m,
msg_set_grp_bc_ack_req(hdr, ack);
/* Build message as chain of buffers */
- skb_queue_head_init(&pkts);
+ __skb_queue_head_init(&pkts);
rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts);
if (unlikely(rc != dlen))
return rc;
@@ -1387,7 +1387,7 @@ static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dlen)
if (unlikely(rc))
return rc;
- skb_queue_head_init(&pkts);
+ __skb_queue_head_init(&pkts);
mtu = tipc_node_get_mtu(net, dnode, tsk->portid);
rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts);
if (unlikely(rc != dlen))
@@ -1445,7 +1445,7 @@ static int __tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dlen)
int send, sent = 0;
int rc = 0;
- skb_queue_head_init(&pkts);
+ __skb_queue_head_init(&pkts);
if (unlikely(dlen > INT_MAX))
return -EMSGSIZE;
@@ -1805,7 +1805,7 @@ static int tipc_recvmsg(struct socket *sock, struct msghdr *m,
/* Send group flow control advertisement when applicable */
if (tsk->group && msg_in_group(hdr) && !grp_evt) {
- skb_queue_head_init(&xmitq);
+ __skb_queue_head_init(&xmitq);
tipc_group_update_rcv_win(tsk->group, tsk_blocks(hlen + dlen),
msg_orignode(hdr), msg_origport(hdr),
&xmitq);
@@ -2674,7 +2674,7 @@ static void tipc_sk_timeout(struct timer_list *t)
struct sk_buff_head list;
int rc = 0;
- skb_queue_head_init(&list);
+ __skb_queue_head_init(&list);
bh_lock_sock(sk);
/* Try again later if socket is busy */
--
2.1.4
^ permalink raw reply related
* [PATCH net-next v2 2/4] RDS: don't use GFP_ATOMIC for sk_alloc in rds_create
From: Gerd Rausch @ 2019-08-15 14:42 UTC (permalink / raw)
To: Santosh Shilimkar, netdev, linux-rdma, rds-devel; +Cc: David Miller
In-Reply-To: <cover.1565879451.git.gerd.rausch@oracle.com>
From: Chris Mason <chris.mason@oracle.com>
Date: Fri, 3 Feb 2012 11:08:51 -0500
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Bang Nguyen <bang.nguyen@oracle.com>
Signed-off-by: Gerd Rausch <gerd.rausch@oracle.com>
Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
---
net/rds/af_rds.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/rds/af_rds.c b/net/rds/af_rds.c
index 2b969f99ef13..7228892046cf 100644
--- a/net/rds/af_rds.c
+++ b/net/rds/af_rds.c
@@ -705,7 +705,7 @@ static int rds_create(struct net *net, struct socket *sock, int protocol,
if (sock->type != SOCK_SEQPACKET || protocol)
return -ESOCKTNOSUPPORT;
- sk = sk_alloc(net, AF_RDS, GFP_ATOMIC, &rds_proto, kern);
+ sk = sk_alloc(net, AF_RDS, GFP_KERNEL, &rds_proto, kern);
if (!sk)
return -ENOMEM;
--
2.22.1
^ permalink raw reply related
* [PATCH net-next v2 1/4] RDS: limit the number of times we loop in rds_send_xmit
From: Gerd Rausch @ 2019-08-15 14:42 UTC (permalink / raw)
To: Santosh Shilimkar, netdev, linux-rdma, rds-devel; +Cc: David Miller
In-Reply-To: <cover.1565879451.git.gerd.rausch@oracle.com>
From: Chris Mason <chris.mason@oracle.com>
Date: Fri, 3 Feb 2012 11:07:54 -0500
This will kick the RDS worker thread if we have been looping
too long.
Original commit from 2012 updated to include a change by
Venkat Venkatsubra <venkat.x.venkatsubra@oracle.com>
that triggers "must_wake" if "rds_ib_recv_refill_one" fails.
Signed-off-by: Gerd Rausch <gerd.rausch@oracle.com>
---
net/rds/ib_recv.c | 12 +++++++++++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/net/rds/ib_recv.c b/net/rds/ib_recv.c
index 3cae88cbdaa0..1a8a4a760b84 100644
--- a/net/rds/ib_recv.c
+++ b/net/rds/ib_recv.c
@@ -385,6 +385,7 @@ void rds_ib_recv_refill(struct rds_connection *conn, int prefill, gfp_t gfp)
unsigned int posted = 0;
int ret = 0;
bool can_wait = !!(gfp & __GFP_DIRECT_RECLAIM);
+ bool must_wake = false;
u32 pos;
/* the goal here is to just make sure that someone, somewhere
@@ -405,6 +406,7 @@ void rds_ib_recv_refill(struct rds_connection *conn, int prefill, gfp_t gfp)
recv = &ic->i_recvs[pos];
ret = rds_ib_recv_refill_one(conn, recv, gfp);
if (ret) {
+ must_wake = true;
break;
}
@@ -423,6 +425,11 @@ void rds_ib_recv_refill(struct rds_connection *conn, int prefill, gfp_t gfp)
}
posted++;
+
+ if ((posted > 128 && need_resched()) || posted > 8192) {
+ must_wake = true;
+ break;
+ }
}
/* We're doing flow control - update the window. */
@@ -445,10 +452,13 @@ void rds_ib_recv_refill(struct rds_connection *conn, int prefill, gfp_t gfp)
* if we should requeue.
*/
if (rds_conn_up(conn) &&
- ((can_wait && rds_ib_ring_low(&ic->i_recv_ring)) ||
+ (must_wake ||
+ (can_wait && rds_ib_ring_low(&ic->i_recv_ring)) ||
rds_ib_ring_empty(&ic->i_recv_ring))) {
queue_delayed_work(rds_wq, &conn->c_recv_w, 1);
}
+ if (can_wait)
+ cond_resched();
}
/*
--
2.22.1
^ permalink raw reply related
* [PATCH net-next v2 0/4] net/rds: Fixes from internal Oracle repo
From: Gerd Rausch @ 2019-08-15 14:40 UTC (permalink / raw)
To: Santosh Shilimkar, netdev, linux-rdma, rds-devel; +Cc: David Miller
In-Reply-To: <20190814.212525.326606319186601317.davem@davemloft.net>
This is the first set of (mostly old) patches from our internal repository
in an effort to synchronize what Oracle had been using internally
with what is shipped with the Linux kernel.
Andy Grover (1):
rds: check for excessive looping in rds_send_xmit
Chris Mason (2):
RDS: limit the number of times we loop in rds_send_xmit
RDS: don't use GFP_ATOMIC for sk_alloc in rds_create
Gerd Rausch (1):
net/rds: Add a few missing rds_stat_names entries
net/rds/af_rds.c | 2 +-
net/rds/ib_recv.c | 12 +++++++++++-
net/rds/rds.h | 2 +-
net/rds/send.c | 12 ++++++++++++
net/rds/stats.c | 3 +++
5 files changed, 28 insertions(+), 3 deletions(-)
--
Changes in submitted patch v2:
* Dropped the controversial "sysctl" patch:
https://lore.kernel.org/netdev/20190814.142112.1080694155114782651.davem@davemloft.net/
^ permalink raw reply
* [PATCH bpf 6/6] tools: bpftool: move "__printf()" attributes to header file
From: Quentin Monnet @ 2019-08-15 14:32 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
In-Reply-To: <20190815143220.4199-1-quentin.monnet@netronome.com>
Some functions in bpftool have a "__printf()" format attributes to tell
the compiler they should expect printf()-like arguments. But because
these attributes are not used for the function prototypes in the header
files, the compiler does not run the checks everywhere the functions are
used, and some mistakes on format string and corresponding arguments
slipped in over time.
Let's move the __printf() attributes to the correct places.
Note: We add guards around the definition of GCC_VERSION in
tools/include/linux/compiler-gcc.h to prevent a conflict in jit_disasm.c
on GCC_VERSION from headers pulled via libbfd.
Fixes: c101189bc968 ("tools: bpftool: fix -Wmissing declaration warnings")
Reported-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
tools/bpf/bpftool/common.c | 4 ++--
tools/bpf/bpftool/json_writer.c | 6 ++----
tools/bpf/bpftool/json_writer.h | 6 ++++--
tools/bpf/bpftool/main.h | 4 ++--
tools/include/linux/compiler-gcc.h | 2 ++
5 files changed, 12 insertions(+), 10 deletions(-)
diff --git a/tools/bpf/bpftool/common.c b/tools/bpf/bpftool/common.c
index 6a71324be628..88264abaa738 100644
--- a/tools/bpf/bpftool/common.c
+++ b/tools/bpf/bpftool/common.c
@@ -29,7 +29,7 @@
#define BPF_FS_MAGIC 0xcafe4a11
#endif
-void __printf(1, 2) p_err(const char *fmt, ...)
+void p_err(const char *fmt, ...)
{
va_list ap;
@@ -47,7 +47,7 @@ void __printf(1, 2) p_err(const char *fmt, ...)
va_end(ap);
}
-void __printf(1, 2) p_info(const char *fmt, ...)
+void p_info(const char *fmt, ...)
{
va_list ap;
diff --git a/tools/bpf/bpftool/json_writer.c b/tools/bpf/bpftool/json_writer.c
index 6046dcab51cc..86501cd3c763 100644
--- a/tools/bpf/bpftool/json_writer.c
+++ b/tools/bpf/bpftool/json_writer.c
@@ -15,7 +15,6 @@
#include <malloc.h>
#include <inttypes.h>
#include <stdint.h>
-#include <linux/compiler.h>
#include "json_writer.h"
@@ -153,8 +152,7 @@ void jsonw_name(json_writer_t *self, const char *name)
putc(' ', self->out);
}
-void __printf(2, 0)
-jsonw_vprintf_enquote(json_writer_t *self, const char *fmt, va_list ap)
+void jsonw_vprintf_enquote(json_writer_t *self, const char *fmt, va_list ap)
{
jsonw_eor(self);
putc('"', self->out);
@@ -162,7 +160,7 @@ jsonw_vprintf_enquote(json_writer_t *self, const char *fmt, va_list ap)
putc('"', self->out);
}
-void __printf(2, 3) jsonw_printf(json_writer_t *self, const char *fmt, ...)
+void jsonw_printf(json_writer_t *self, const char *fmt, ...)
{
va_list ap;
diff --git a/tools/bpf/bpftool/json_writer.h b/tools/bpf/bpftool/json_writer.h
index cb9a1993681c..35cf1f00f96c 100644
--- a/tools/bpf/bpftool/json_writer.h
+++ b/tools/bpf/bpftool/json_writer.h
@@ -14,6 +14,7 @@
#include <stdbool.h>
#include <stdint.h>
#include <stdarg.h>
+#include <linux/compiler.h>
/* Opaque class structure */
typedef struct json_writer json_writer_t;
@@ -30,8 +31,9 @@ void jsonw_pretty(json_writer_t *self, bool on);
void jsonw_name(json_writer_t *self, const char *name);
/* Add value */
-void jsonw_vprintf_enquote(json_writer_t *self, const char *fmt, va_list ap);
-void jsonw_printf(json_writer_t *self, const char *fmt, ...);
+void __printf(2, 0) jsonw_vprintf_enquote(json_writer_t *self, const char *fmt,
+ va_list ap);
+void __printf(2, 3) jsonw_printf(json_writer_t *self, const char *fmt, ...);
void jsonw_string(json_writer_t *self, const char *value);
void jsonw_bool(json_writer_t *self, bool value);
void jsonw_float(json_writer_t *self, double number);
diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h
index 7031a4bf87a0..af9ad56c303a 100644
--- a/tools/bpf/bpftool/main.h
+++ b/tools/bpf/bpftool/main.h
@@ -98,8 +98,8 @@ extern int bpf_flags;
extern struct pinned_obj_table prog_table;
extern struct pinned_obj_table map_table;
-void p_err(const char *fmt, ...);
-void p_info(const char *fmt, ...);
+void __printf(1, 2) p_err(const char *fmt, ...);
+void __printf(1, 2) p_info(const char *fmt, ...);
bool is_prefix(const char *pfx, const char *str);
int detect_common_prefix(const char *arg, ...);
diff --git a/tools/include/linux/compiler-gcc.h b/tools/include/linux/compiler-gcc.h
index 0d35f18006a1..95c072b70d0e 100644
--- a/tools/include/linux/compiler-gcc.h
+++ b/tools/include/linux/compiler-gcc.h
@@ -6,9 +6,11 @@
/*
* Common definitions for all gcc versions go here.
*/
+#ifndef GCC_VERSION
#define GCC_VERSION (__GNUC__ * 10000 \
+ __GNUC_MINOR__ * 100 \
+ __GNUC_PATCHLEVEL__)
+#endif
#if GCC_VERSION >= 70000 && !defined(__CHECKER__)
# define __fallthrough __attribute__ ((fallthrough))
--
2.17.1
^ permalink raw reply related
* [PATCH bpf 3/6] tools: bpftool: fix argument for p_err() in BTF do_dump()
From: Quentin Monnet @ 2019-08-15 14:32 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
In-Reply-To: <20190815143220.4199-1-quentin.monnet@netronome.com>
The last argument passed to one call to the p_err() function is not
correct, it should be "*argv" instead of "**argv". This may lead to a
segmentation fault error if BTF id cannot be parsed correctly. Let's fix
this.
Fixes: c93cc69004dt ("bpftool: add ability to dump BTF types")
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
tools/bpf/bpftool/btf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c
index 1b8ec91899e6..8805637f1a7e 100644
--- a/tools/bpf/bpftool/btf.c
+++ b/tools/bpf/bpftool/btf.c
@@ -449,7 +449,7 @@ static int do_dump(int argc, char **argv)
btf_id = strtoul(*argv, &endptr, 0);
if (*endptr) {
- p_err("can't parse %s as ID", **argv);
+ p_err("can't parse %s as ID", *argv);
return -1;
}
NEXT_ARG();
--
2.17.1
^ permalink raw reply related
* [PATCH bpf 5/6] tools: bpftool: fix format string for p_err() in detect_common_prefix()
From: Quentin Monnet @ 2019-08-15 14:32 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
In-Reply-To: <20190815143220.4199-1-quentin.monnet@netronome.com>
There is one call to the p_err() function in detect_common_prefix()
where the message to print is passed directly as the first argument,
without using a format string. This is harmless, but may trigger
warnings if the "__printf()" attribute is used correctly for the p_err()
function. Let's fix it by using a "%s" format string.
Fixes: ba95c7452439 ("tools: bpftool: add "prog run" subcommand to test-run programs")
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
tools/bpf/bpftool/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/bpf/bpftool/main.c b/tools/bpf/bpftool/main.c
index e916ff25697f..93d008687020 100644
--- a/tools/bpf/bpftool/main.c
+++ b/tools/bpf/bpftool/main.c
@@ -139,7 +139,7 @@ int detect_common_prefix(const char *arg, ...)
strncat(msg, "'", sizeof(msg) - strlen(msg) - 1);
if (count >= 2) {
- p_err(msg);
+ p_err("%s", msg);
return -1;
}
--
2.17.1
^ permalink raw reply related
* [PATCH bpf 4/6] tools: bpftool: fix format string for p_err() in query_flow_dissector()
From: Quentin Monnet @ 2019-08-15 14:32 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
In-Reply-To: <20190815143220.4199-1-quentin.monnet@netronome.com>
The format string passed to one call to the p_err() function in
query_flow_dissector() does not match the value that should be printed,
resulting in some garbage integer being printed instead of
strerror(errno) if /proc/self/ns/net cannot be open. Let's fix the
format string.
Fixes: 7f0c57fec80f ("bpftool: show flow_dissector attachment status")
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
tools/bpf/bpftool/net.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/bpf/bpftool/net.c b/tools/bpf/bpftool/net.c
index 67e99c56bc88..e3b770082a39 100644
--- a/tools/bpf/bpftool/net.c
+++ b/tools/bpf/bpftool/net.c
@@ -197,7 +197,7 @@ static int query_flow_dissector(struct bpf_attach_info *attach_info)
fd = open("/proc/self/ns/net", O_RDONLY);
if (fd < 0) {
- p_err("can't open /proc/self/ns/net: %d",
+ p_err("can't open /proc/self/ns/net: %s",
strerror(errno));
return -1;
}
--
2.17.1
^ permalink raw reply related
* [PATCH bpf 2/6] tools: bpftool: fix format strings and arguments for jsonw_printf()
From: Quentin Monnet @ 2019-08-15 14:32 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
In-Reply-To: <20190815143220.4199-1-quentin.monnet@netronome.com>
There are some mismatches between format strings and arguments passed to
jsonw_printf() in the BTF dumper for bpftool, which seems harmless but
may result in warnings if the "__printf()" attribute is used correctly
for jsonw_printf(). Let's fix relevant format strings and type cast.
Fixes: b12d6ec09730 ("bpf: btf: add btf print functionality")
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
tools/bpf/bpftool/btf_dumper.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/tools/bpf/bpftool/btf_dumper.c b/tools/bpf/bpftool/btf_dumper.c
index 8cafb9b31467..d66131f69689 100644
--- a/tools/bpf/bpftool/btf_dumper.c
+++ b/tools/bpf/bpftool/btf_dumper.c
@@ -26,9 +26,9 @@ static void btf_dumper_ptr(const void *data, json_writer_t *jw,
bool is_plain_text)
{
if (is_plain_text)
- jsonw_printf(jw, "%p", *(unsigned long *)data);
+ jsonw_printf(jw, "%p", data);
else
- jsonw_printf(jw, "%u", *(unsigned long *)data);
+ jsonw_printf(jw, "%lu", *(unsigned long *)data);
}
static int btf_dumper_modifier(const struct btf_dumper *d, __u32 type_id,
@@ -216,7 +216,7 @@ static int btf_dumper_int(const struct btf_type *t, __u8 bit_offset,
switch (BTF_INT_ENCODING(*int_type)) {
case 0:
if (BTF_INT_BITS(*int_type) == 64)
- jsonw_printf(jw, "%lu", *(__u64 *)data);
+ jsonw_printf(jw, "%llu", *(__u64 *)data);
else if (BTF_INT_BITS(*int_type) == 32)
jsonw_printf(jw, "%u", *(__u32 *)data);
else if (BTF_INT_BITS(*int_type) == 16)
@@ -229,7 +229,7 @@ static int btf_dumper_int(const struct btf_type *t, __u8 bit_offset,
break;
case BTF_INT_SIGNED:
if (BTF_INT_BITS(*int_type) == 64)
- jsonw_printf(jw, "%ld", *(long long *)data);
+ jsonw_printf(jw, "%lld", *(long long *)data);
else if (BTF_INT_BITS(*int_type) == 32)
jsonw_printf(jw, "%d", *(int *)data);
else if (BTF_INT_BITS(*int_type) == 16)
--
2.17.1
^ permalink raw reply related
* [PATCH bpf 1/6] tools: bpftool: fix arguments for p_err() in do_event_pipe()
From: Quentin Monnet @ 2019-08-15 14:32 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
In-Reply-To: <20190815143220.4199-1-quentin.monnet@netronome.com>
The last argument passed to some calls to the p_err() functions is not
correct, it should be "*argv" instead of "**argv". This may lead to a
segmentation fault error if CPU IDs or indices from the command line
cannot be parsed correctly. Let's fix this.
Fixes: f412eed9dfde ("tools: bpftool: add simple perf event output reader")
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
tools/bpf/bpftool/map_perf_ring.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/bpf/bpftool/map_perf_ring.c b/tools/bpf/bpftool/map_perf_ring.c
index 3f108ab17797..4c5531d1a450 100644
--- a/tools/bpf/bpftool/map_perf_ring.c
+++ b/tools/bpf/bpftool/map_perf_ring.c
@@ -157,7 +157,7 @@ int do_event_pipe(int argc, char **argv)
NEXT_ARG();
ctx.cpu = strtoul(*argv, &endptr, 0);
if (*endptr) {
- p_err("can't parse %s as CPU ID", **argv);
+ p_err("can't parse %s as CPU ID", *argv);
goto err_close_map;
}
@@ -168,7 +168,7 @@ int do_event_pipe(int argc, char **argv)
NEXT_ARG();
ctx.idx = strtoul(*argv, &endptr, 0);
if (*endptr) {
- p_err("can't parse %s as index", **argv);
+ p_err("can't parse %s as index", *argv);
goto err_close_map;
}
--
2.17.1
^ permalink raw reply related
* [PATCH bpf 0/6] tools: bpftool: fix printf()-like functions
From: Quentin Monnet @ 2019-08-15 14:32 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
Hi,
Because the "__printf()" attributes were used only where the functions are
implemented, and not in header files, the checks have not been enforced on
all the calls to printf()-like functions, and a number of errors slipped in
bpftool over time.
This set cleans up such errors, and then moves the "__printf()" attributes
to header files, so that the checks are performed at all locations.
Quentin Monnet (6):
tools: bpftool: fix arguments for p_err() in do_event_pipe()
tools: bpftool: fix format strings and arguments for jsonw_printf()
tools: bpftool: fix argument for p_err() in BTF do_dump()
tools: bpftool: fix format string for p_err() in
query_flow_dissector()
tools: bpftool: fix format string for p_err() in
detect_common_prefix()
tools: bpftool: move "__printf()" attributes to header file
tools/bpf/bpftool/btf.c | 2 +-
tools/bpf/bpftool/btf_dumper.c | 8 ++++----
tools/bpf/bpftool/common.c | 4 ++--
tools/bpf/bpftool/json_writer.c | 6 ++----
tools/bpf/bpftool/json_writer.h | 6 ++++--
tools/bpf/bpftool/main.c | 2 +-
tools/bpf/bpftool/main.h | 4 ++--
tools/bpf/bpftool/map_perf_ring.c | 4 ++--
tools/bpf/bpftool/net.c | 2 +-
tools/include/linux/compiler-gcc.h | 2 ++
10 files changed, 21 insertions(+), 19 deletions(-)
--
2.17.1
^ permalink raw reply
* [PATCH bpf] tools: bpftool: close prog FD before exit on showing a single program
From: Quentin Monnet @ 2019-08-15 14:22 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: bpf, netdev, oss-drivers, Quentin Monnet
When showing metadata about a single program by invoking
"bpftool prog show PROG", the file descriptor referring to the program
is not closed before returning from the function. Let's close it.
Fixes: 71bb428fe2c1 ("tools: bpf: add bpftool")
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
tools/bpf/bpftool/prog.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index 66f04a4846a5..43fdbbfe41bb 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -363,7 +363,9 @@ static int do_show(int argc, char **argv)
if (fd < 0)
return -1;
- return show_prog(fd);
+ err = show_prog(fd);
+ close(fd);
+ return err;
}
if (argc)
--
2.17.1
^ permalink raw reply related
* [PATCH -next] btf: fix return value check in btf_vmlinux_init()
From: Wei Yongjun @ 2019-08-15 14:24 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann, Martin KaFai Lau, Song Liu,
Yonghong Song, Andrii Nakryiko
Cc: Wei Yongjun, netdev, bpf, kernel-janitors
In case of error, the function kobject_create_and_add() returns NULL
pointer not ERR_PTR(). The IS_ERR() test in the return value check
should be replaced with NULL test.
Fixes: 341dfcf8d78e ("btf: expose BTF info through sysfs")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
---
kernel/bpf/sysfs_btf.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/kernel/bpf/sysfs_btf.c b/kernel/bpf/sysfs_btf.c
index 4659349fc795..be5557deb958 100644
--- a/kernel/bpf/sysfs_btf.c
+++ b/kernel/bpf/sysfs_btf.c
@@ -30,16 +30,13 @@ static struct kobject *btf_kobj;
static int __init btf_vmlinux_init(void)
{
- int err;
-
if (!_binary__btf_vmlinux_bin_start)
return 0;
btf_kobj = kobject_create_and_add("btf", kernel_kobj);
- if (IS_ERR(btf_kobj)) {
- err = PTR_ERR(btf_kobj);
+ if (!btf_kobj) {
btf_kobj = NULL;
- return err;
+ return -ENOMEM;
}
bin_attr_btf_vmlinux.size = _binary__btf_vmlinux_bin_end -
^ permalink raw reply related
* Re: [RFC PATCH 1/5] x86: tsc: add tsc to art helpers
From: Thomas Gleixner @ 2019-08-15 14:16 UTC (permalink / raw)
To: Felipe Balbi
Cc: Richard Cochran, netdev, Ingo Molnar, Borislav Petkov,
H . Peter Anvin, x86, linux-kernel, Christopher S . Hall
In-Reply-To: <87y2zvt1hk.fsf@gmail.com>
Felipe,
On Thu, 15 Aug 2019, Felipe Balbi wrote:
> Thomas Gleixner <tglx@linutronix.de> writes:
> > On Tue, 16 Jul 2019, Felipe Balbi wrote:
> >
> > So some information what those interfaces are used for and why they are
> > needed would be really helpful.
>
> Okay, I have some more details about this. The TGPIO device itself uses
> ART since TSC is not directly available to anything other than the
> CPU. The 'problem' here is that reading ART incurs extra latency which
> we would like to avoid. Therefore, we use TSC and scale it to
> nanoseconds which, would be the same as ART to ns.
Fine. But that's not really correct:
TSC = art_to_tsc_offset + ART * scale;
> >> +void get_tsc_ns(struct system_counterval_t *tsc_counterval, u64 *tsc_ns)
Why is this not returning the result instead of having that pointer
indirection?
> >> +{
> >> + u64 tmp, res, rem;
> >> + u64 cycles;
> >> +
> >> + tsc_counterval->cycles = clocksource_tsc.read(NULL);
> >> + cycles = tsc_counterval->cycles;
> >> + tsc_counterval->cs = art_related_clocksource;
So this does more than returning the TSC time converted to nanoseconds. The
function name should reflect this. Plus both functions want kernel-doc
explaining what they do.
> >> + rem = do_div(cycles, tsc_khz);
> >> +
> >> + res = cycles * USEC_PER_SEC;
> >> + tmp = rem * USEC_PER_SEC;
> >> +
> >> + do_div(tmp, tsc_khz);
> >> + res += tmp;
> >> +
> >> + *tsc_ns = res;
> >> +}
> >> +EXPORT_SYMBOL(get_tsc_ns);
> >> +
> >> +u64 get_art_ns_now(void)
> >> +{
> >> + struct system_counterval_t tsc_cycles;
> >> + u64 tsc_ns;
> >> +
> >> + get_tsc_ns(&tsc_cycles, &tsc_ns);
> >> +
> >> + return tsc_ns;
> >> +}
> >> +EXPORT_SYMBOL(get_art_ns_now);
> >
> > While the changes look innocuous I'm missing the big picture why this needs
> > to emulate ART instead of simply using TSC directly.
>
> i don't think we're emulating ART here (other than the name in the
> function). We're just reading TSC and converting to nanoseconds, right?
Well, the function name says clearly: get_art_ns_now(). But you are not
using ART, you use the TSC and derive the ART value from it (incorrectly).
Thanks,
tglx
^ permalink raw reply
* Re: [RFC bpf-next 0/3] tools: bpftool: add subcommand to count map entries
From: Quentin Monnet @ 2019-08-15 14:15 UTC (permalink / raw)
To: Edward Cree, Alexei Starovoitov
Cc: Alexei Starovoitov, Daniel Borkmann, bpf, netdev, oss-drivers
In-Reply-To: <84aa97e3-5fde-e041-12c6-85863e27d2d9@solarflare.com>
2019-08-14 18:14 UTC+0100 ~ Edward Cree <ecree@solarflare.com>
> On 14/08/2019 17:58, Quentin Monnet wrote:
>> 2019-08-14 17:45 UTC+0100 ~ Edward Cree <ecree@solarflare.com>
>>> This might be a really dumb suggestion, but: you're wanting to collect a
>>> summary statistic over an in-kernel data structure in a single syscall,
>>> because making a series of syscalls to examine every entry is slow and
>>> racy. Isn't that exactly a job for an in-kernel virtual machine, and
>>> could you not supply an eBPF program which the kernel runs on each entry
>>> in the map, thus supporting people who want to calculate something else
>>> (mean, min and max, whatever) instead of count?
>>>
>> Hi Edward, I like the approach, thanks for the suggestion.
>>
>> But I did not mention that we were using offloaded maps: Tracing the
>> kernel would probably work for programs running on the host, but this is
>> not a solution we could extend to hardware offload.
> I don't see where "tracing" comes into it; this is a new program type and
> a new map op under the bpf() syscall.
> Could the user-supplied BPF program not then be passed down to the device
> for it to run against its offloaded maps?
>
Sorry, I misunderstood your suggestion :s (I thought you meant tracing
the insert and delete operations).
So if I understand correctly, we would use the bpf() syscall to trigger
a run of such program on all map entries (for map implementing the new
operation), and the context would include pointers to the key and the
value for the entry being processed so we can count/sum/compute an
average of the values or any other kind of processing?
Thanks,
Quentin
^ permalink raw reply
* Re: [PATCH v5 10/18] compat_ioctl: move SIOCOUTQ out of compat_ioctl.c
From: Greg Kroah-Hartman @ 2019-08-15 14:09 UTC (permalink / raw)
To: Arnd Bergmann
Cc: linux-kernel, viro, linux-fsdevel, Jiri Slaby, David S. Miller,
netdev
In-Reply-To: <20190814205521.122180-1-arnd@arndb.de>
On Wed, Aug 14, 2019 at 10:54:45PM +0200, Arnd Bergmann wrote:
> All users of this call are in socket or tty code, so handling
> it there means we can avoid the table entry in fs/compat_ioctl.c.
>
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
> drivers/tty/tty_io.c | 1 +
> fs/compat_ioctl.c | 2 --
> net/socket.c | 2 ++
> 3 files changed, 3 insertions(+), 2 deletions(-)a
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
^ permalink raw reply
* Re: [RFC bpf-next 0/3] tools: bpftool: add subcommand to count map entries
From: Quentin Monnet @ 2019-08-15 14:02 UTC (permalink / raw)
To: Andrii Nakryiko
Cc: Alexei Starovoitov, Edward Cree, Alexei Starovoitov,
Daniel Borkmann, bpf, Network Development, oss-drivers
In-Reply-To: <CAEf4BzYqsT4OmWQ9WK9dmnKT9cMcjbhgHZmboUBgkEvtbaeUeA@mail.gmail.com>
2019-08-14 13:18 UTC-0700 ~ Andrii Nakryiko <andrii.nakryiko@gmail.com>
> On Wed, Aug 14, 2019 at 10:12 AM Quentin Monnet
> <quentin.monnet@netronome.com> wrote:
>>
>> 2019-08-14 09:58 UTC-0700 ~ Alexei Starovoitov
>> <alexei.starovoitov@gmail.com>
>>> On Wed, Aug 14, 2019 at 9:45 AM Edward Cree <ecree@solarflare.com> wrote:
>>>>
>>>> On 14/08/2019 10:42, Quentin Monnet wrote:
>>>>> 2019-08-13 18:51 UTC-0700 ~ Alexei Starovoitov
>>>>> <alexei.starovoitov@gmail.com>
>>>>>> The same can be achieved by 'bpftool map dump|grep key|wc -l', no?
>>>>> To some extent (with subtleties for some other map types); and we use a
>>>>> similar command line as a workaround for now. But because of the rate of
>>>>> inserts/deletes in the map, the process often reports a number higher
>>>>> than the max number of entries (we observed up to ~750k when max_entries
>>>>> is 500k), even is the map is only half-full on average during the count.
>>>>> On the worst case (though not frequent), an entry is deleted just before
>>>>> we get the next key from it, and iteration starts all over again. This
>>>>> is not reliable to determine how much space is left in the map.
>>>>>
>>>>> I cannot see a solution that would provide a more accurate count from
>>>>> user space, when the map is under pressure?
>>>> This might be a really dumb suggestion, but: you're wanting to collect a
>>>> summary statistic over an in-kernel data structure in a single syscall,
>>>> because making a series of syscalls to examine every entry is slow and
>>>> racy. Isn't that exactly a job for an in-kernel virtual machine, and
>>>> could you not supply an eBPF program which the kernel runs on each entry
>>>> in the map, thus supporting people who want to calculate something else
>>>> (mean, min and max, whatever) instead of count?
>>>
>>> Pretty much my suggestion as well :)
>
> I also support the suggestion to count it from BPF side. It's flexible
> and powerful approach and doesn't require adding more and more nuanced
> sub-APIs to kernel to support subset of bulk operations on map
> (subset, because we'll expose count, but what about, e.g., p50, etc,
> there will always be something more that someone will want and it just
> doesn't scale).
Hi Andrii,
Yes, that makes sense.
>
>>>
>>> It seems the better fix for your nat threshold is to keep count of
>>> elements in the map in a separate global variable that
>>> bpf program manually increments and decrements.
>>> bpftool will dump it just as regular map of single element.
>>> (I believe it doesn't recognize global variables properly yet)
>>> and BTF will be there to pick exactly that 'count' variable.
>>>
>>
>> It would be with an offloaded map, but yes, I suppose we could keep
>> track of the numbers in a separate map. We'll have a look into this.
>
> See if you can use a global variable, that way you completely
> eliminate any overhead from BPF side of things, except for atomic
> increment.
Offloaded maps do not implement the map_direct_value_addr() operation,
so global variables are not supported at the moment. I need to dive
deeper into this and see what is required to add that support.
Thanks for your advice!
Quentin
^ 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