Re: [PATCH v3 bpf-next 4/4] bpftool: support loading flow dissector

On 11/08, Quentin Monnet wrote:

2018-11-08 10:01 UTC-0800 ~ Stanislav Fomichev sdf@fomichev.me

...

On 11/08, Quentin Monnet wrote:

...

Hi Stanislav, thanks for the changes! More comments below.

Thank you for another round of review!

...

2018-11-07 21:39 UTC-0800 ~ Stanislav Fomichev sdf@google.com

...

This commit adds support for loading/attaching/detaching flow dissector program. The structure of the flow dissector program is assumed to be the same as in the selftests:

• flow_dissector section with the main entry point
• a bunch of tail call progs
• a jmp_table map that is populated with the tail call progs

When `bpftool load` is called with a flow_dissector prog (i.e. when the first section is flow_dissector of 'type flow_dissector' argument is passed), we load and pin all the programs/maps. User is responsible to construct the jump table for the tail calls.

The last argument of `bpftool attach` is made optional for this use case.

Example: bpftool prog load tools/testing/selftests/bpf/bpf_flow.o \ /sys/fs/bpf/flow type flow_dissector

bpftool map update pinned /sys/fs/bpf/flow/jmp_table \ key 0 0 0 0 \ value pinned /sys/fs/bpf/flow/IP

bpftool map update pinned /sys/fs/bpf/flow/jmp_table \ key 1 0 0 0 \ value pinned /sys/fs/bpf/flow/IPV6

bpftool map update pinned /sys/fs/bpf/flow/jmp_table \ key 2 0 0 0 \ value pinned /sys/fs/bpf/flow/IPV6OP

bpftool map update pinned /sys/fs/bpf/flow/jmp_table \ key 3 0 0 0 \ value pinned /sys/fs/bpf/flow/IPV6FR

bpftool map update pinned /sys/fs/bpf/flow/jmp_table \ key 4 0 0 0 \ value pinned /sys/fs/bpf/flow/MPLS

bpftool map update pinned /sys/fs/bpf/flow/jmp_table \ key 5 0 0 0 \ value pinned /sys/fs/bpf/flow/VLAN

bpftool prog attach pinned /sys/fs/bpf/flow/flow_dissector flow_dissector

Tested by using the above lines to load the prog in the test_flow_dissector.sh selftest.

Signed-off-by: Stanislav Fomichev sdf@google.com

.../bpftool/Documentation/bpftool-prog.rst | 36 ++++-- tools/bpf/bpftool/bash-completion/bpftool | 6 +- tools/bpf/bpftool/common.c | 30 ++--- tools/bpf/bpftool/main.h | 1 + tools/bpf/bpftool/prog.c | 112 +++++++++++++----- 5 files changed, 126 insertions(+), 59 deletions(-)

diff --git a/tools/bpf/bpftool/Documentation/bpftool-prog.rst b/tools/bpf/bpftool/Documentation/bpftool-prog.rst index ac4e904b10fb..0374634c3087 100644 --- a/tools/bpf/bpftool/Documentation/bpftool-prog.rst +++ b/tools/bpf/bpftool/Documentation/bpftool-prog.rst @@ -15,7 +15,8 @@ SYNOPSIS *OPTIONS* := { { **-j** | **--json** } [{ **-p** | **--pretty** }] | { **-f** | **--bpffs** } } *COMMANDS* :=

• { **show** | **list** | **dump xlated** | **dump jited** | **pin** | **load** | **help** }

• { **show** | **list** | **dump xlated** | **dump jited** | **pin** | **load**
• | **loadall** | **help** } MAP COMMANDS =============

@@ -24,9 +25,9 @@ MAP COMMANDS | **bpftool** **prog dump xlated** *PROG* [{**file** *FILE* | **opcodes** | **visual**}] | **bpftool** **prog dump jited** *PROG* [{**file** *FILE* | **opcodes**}] | **bpftool** **prog pin** *PROG* *FILE* -| **bpftool** **prog load** *OBJ* *FILE* [**type** *TYPE*] [**map** {**idx** *IDX* | **name** *NAME*} *MAP*] [**dev** *NAME*] -| **bpftool** **prog attach** *PROG* *ATTACH_TYPE* *MAP* -| **bpftool** **prog detach** *PROG* *ATTACH_TYPE* *MAP* +| **bpftool** **prog { load | loadall }** *OBJ* *FILE* [**type** *TYPE*] [**map** {**idx** *IDX* | **name** *NAME*} *MAP*] [**dev** *NAME*] +| **bpftool** **prog attach** *PROG* *ATTACH_TYPE* [*MAP*] +| **bpftool** **prog detach** *PROG* *ATTACH_TYPE* [*MAP*] | **bpftool** **prog help** | | *MAP* := { **id** *MAP_ID* | **pinned** *FILE* } @@ -39,7 +40,9 @@ MAP COMMANDS | **cgroup/bind4** | **cgroup/bind6** | **cgroup/post_bind4** | **cgroup/post_bind6** | | **cgroup/connect4** | **cgroup/connect6** | **cgroup/sendmsg4** | **cgroup/sendmsg6** | } -| *ATTACH_TYPE* := { **msg_verdict** | **skb_verdict** | **skb_parse** } +| *ATTACH_TYPE* := { +| **msg_verdict** | **skb_verdict** | **skb_parse** | **flow_dissector** +| } DESCRIPTION @@ -79,8 +82,11 @@ DESCRIPTION contain a dot character ('.'), which is reserved for future extensions of *bpffs*.

• **bpftool prog load** *OBJ* *FILE* [**type** *TYPE*] [**map** {**idx** *IDX* | **name** *NAME*} *MAP*] [**dev** *NAME*]

• **bpftool prog { load | loadall }** *OBJ* *FILE* [**type** *TYPE*] [**map** {**idx** *IDX* | **name** *NAME*} *MAP*] [**dev** *NAME*] Load bpf program from binary *OBJ* and pin as *FILE*.

•   **bpftool prog load** will pin only the first bpf program
  

•   from the *OBJ*, **bpftool prog loadall** will pin all maps
  

•   and programs from the *OBJ*.
  

This could be improved regarding maps: with "bpftool prog load" I think we also load and pin all maps, but your description implies this is only the case with "loadall"

I don't think we pin any maps with `bpftool prog load`, we certainly load them, but we don't pin any afaict. Can you point me to the code where we pin the maps?

My bad. I read "pin" but thought "load". It does not pin them indeed, sorry about that.

...

...

...

  **type** is optional, if not specified program type will be
  inferred from section names.
  By default bpftool will create new maps as declared in the ELF

@@ -97,13 +103,17 @@ DESCRIPTION contain a dot character ('.'), which is reserved for future extensions of *bpffs*.

•    **bpftool prog attach** *PROG* *ATTACH_TYPE* *MAP*
  

•              Attach bpf program *PROG* (with type specified by *ATTACH_TYPE*)
  

•              to the map *MAP*.
  

•    **bpftool prog detach** *PROG* *ATTACH_TYPE* *MAP*
  

•              Detach bpf program *PROG* (with type specified by *ATTACH_TYPE*)
  

•              from the map *MAP*.
  

•    **bpftool prog attach** *PROG* *ATTACH_TYPE* [*MAP*]
  

•              Attach bpf program *PROG* (with type specified by
  

•              *ATTACH_TYPE*). Most *ATTACH_TYPEs* require a *MAP*
  

•              parameter, with the exception of *flow_dissector* which is
  

•              attached to current networking name space.
  

•    **bpftool prog detach** *PROG* *ATTACH_TYPE* [*MAP*]
  

•              Detach bpf program *PROG* (with type specified by
  

•              *ATTACH_TYPE*). Most *ATTACH_TYPEs* require a *MAP*
  

•              parameter, with the exception of *flow_dissector* which is
  

•              detached from the current networking name space.
  

While at it could you please fix those two paragraphs to use tabs for indentation, as the rest of the doc? Thanks!

Time to teach my vim to use tabs in .rst files. Sorry about that.

Those paragraphs were using spaces already, so you didn't introduce that :). But all others use tabs so its a good occasion to fix it.

...

...

...

**bpftool prog help** Print short help message. diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool index 3f78e6404589..ad0fc919f7ec 100644 --- a/tools/bpf/bpftool/bash-completion/bpftool +++ b/tools/bpf/bpftool/bash-completion/bpftool @@ -243,7 +243,7 @@ _bpftool() # Completion depends on object and command in use case $object in prog)

•        if [[ $command != "load" ]]; then
  

•        if [[ $command != "load" && $command != "loadall" ]]; then
              case $prev in
                  id)
                      _bpftool_get_prog_ids
  

@@ -299,7 +299,7 @@ _bpftool() fi if [[ ${#words[@]} == 6 ]]; then

•                    COMPREPLY=( $( compgen -W "msg_verdict skb_verdict skb_parse" -- "$cur" ) )
  

•                    COMPREPLY=( $( compgen -W "msg_verdict skb_verdict skb_parse flow_dissector" -- "$cur" ) )
                      return 0
                  fi
  

@@ -309,7 +309,7 @@ _bpftool() fi return 0 ;;

•            load)
  

•            load|loadall)
                  local obj
                  if [[ ${#words[@]} -lt 6 ]]; then
  

You also want to update completion for the program types, at line 341 or so. Feel free to split that list on several lines, by the way :).

Will do, thanks!

...

...

diff --git a/tools/bpf/bpftool/common.c b/tools/bpf/bpftool/common.c index 25af85304ebe..f671a921dec5 100644 --- a/tools/bpf/bpftool/common.c +++ b/tools/bpf/bpftool/common.c @@ -169,34 +169,24 @@ int open_obj_pinned_any(char *path, enum bpf_obj_type exp_type) return fd; } -int do_pin_fd(int fd, const char *name) +int mount_bpffs_for_pin(const char *name) { char err_str[ERR_MAX_LEN]; char *file; char *dir; int err = 0;

• err = bpf_obj_pin(fd, name);
• if (!err)

• goto out;
  

• file = malloc(strlen(name) + 1); strcpy(file, name); dir = dirname(file);
• if (errno != EPERM || is_bpffs(dir)) {

• p_err("can't pin the object (%s): %s", name, strerror(errno));
  

• if (is_bpffs(dir)) {

• /* nothing to do if already mounted */
  

  goto out_free; }

Nitpick: unnecessary brackets.

Ack.

...

...

• /* Attempt to mount bpffs, then retry pinning. */ err = mnt_bpffs(dir, err_str, ERR_MAX_LEN);
• if (!err) {

• err = bpf_obj_pin(fd, name);
  

• if (err)
  

• 	p_err("can't pin the object (%s): %s", name,
  

• 	      strerror(errno));
  

• } else {

• if (err) { err_str[ERR_MAX_LEN - 1] = '\0'; p_err("can't mount BPF file system to pin the object (%s): %s", name, err_str);

@@ -204,10 +194,20 @@ int do_pin_fd(int fd, const char *name) out_free: free(file); -out: return err; } +int do_pin_fd(int fd, const char *name) +{

• int err;
• err = mount_bpffs_for_pin(name);
• if (err)

• return err;
  

• return bpf_obj_pin(fd, name);

+}

• int do_pin_any(int argc, char **argv, int (*get_fd_by_id)(__u32)) { unsigned int id;

diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h index 28322ace2856..1383824c9baf 100644 --- a/tools/bpf/bpftool/main.h +++ b/tools/bpf/bpftool/main.h @@ -129,6 +129,7 @@ const char *get_fd_type_name(enum bpf_obj_type type); char *get_fdinfo(int fd, const char *key); int open_obj_pinned(char *path); int open_obj_pinned_any(char *path, enum bpf_obj_type exp_type); +int mount_bpffs_for_pin(const char *name); int do_pin_any(int argc, char **argv, int (*get_fd_by_id)(__u32)); int do_pin_fd(int fd, const char *name); diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c index 5302ee282409..a4346dd673b1 100644 --- a/tools/bpf/bpftool/prog.c +++ b/tools/bpf/bpftool/prog.c @@ -81,6 +81,7 @@ static const char * const attach_type_strings[] = { [BPF_SK_SKB_STREAM_PARSER] = "stream_parser", [BPF_SK_SKB_STREAM_VERDICT] = "stream_verdict", [BPF_SK_MSG_VERDICT] = "msg_verdict",

• [BPF_FLOW_DISSECTOR] = "flow_dissector", [__MAX_BPF_ATTACH_TYPE] = NULL, };

@@ -724,10 +725,11 @@ int map_replace_compar(const void *p1, const void *p2) static int do_attach(int argc, char **argv) { enum bpf_attach_type attach_type;

• int err, mapfd, progfd;

• int err, progfd;
• int mapfd = 0;

• if (!REQ_ARGS(5)) {

• p_err("too few parameters for map attach");
  

• if (!REQ_ARGS(3)) {

• p_err("too few parameters for attach");
  

  return -EINVAL; }

@@ -740,11 +742,17 @@ static int do_attach(int argc, char **argv) p_err("invalid attach type"); return -EINVAL; }

• NEXT_ARG();

• if (attach_type != BPF_FLOW_DISSECTOR) {

• NEXT_ARG();
  

• if (!REQ_ARGS(2)) {
  

• 	p_err("too few parameters for map attach");
  

• 	return -EINVAL;
  

• }
  

• mapfd = map_parse_fd(&argc, &argv);
• if (mapfd < 0)

• return mapfd;
  

• mapfd = map_parse_fd(&argc, &argv);
  

• if (mapfd < 0)
  

• 	return mapfd;
  

• } err = bpf_prog_attach(progfd, mapfd, attach_type, 0); if (err) {

@@ -760,10 +768,11 @@ static int do_attach(int argc, char **argv) static int do_detach(int argc, char **argv) { enum bpf_attach_type attach_type;

• int err, mapfd, progfd;

• int err, progfd;
• int mapfd = 0;

• if (!REQ_ARGS(5)) {

• p_err("too few parameters for map detach");
  

• if (!REQ_ARGS(3)) {

• p_err("too few parameters for detach");
  

  return -EINVAL; }

@@ -776,11 +785,17 @@ static int do_detach(int argc, char **argv) p_err("invalid attach type"); return -EINVAL; }

• NEXT_ARG();

• if (attach_type != BPF_FLOW_DISSECTOR) {

• NEXT_ARG();
  

• if (!REQ_ARGS(2)) {
  

• 	p_err("too few parameters for map detach");
  

• 	return -EINVAL;
  

• }
  

Would that make sense to factor argument checks or parsing for do_attach() and do_detach() to some extent? In order to reduce the number of attach-type-based exceptions to add in the code if we have other attach types that do not take maps in the future.

I can move all argument parsing into a new function and use it from both do_attach and do_detach.

Sounds good to me, thanks!

...

...

...

• mapfd = map_parse_fd(&argc, &argv);
• if (mapfd < 0)

• return mapfd;
  

• mapfd = map_parse_fd(&argc, &argv);
  

• if (mapfd < 0)
  

• 	return mapfd;
  

• } err = bpf_prog_detach2(progfd, mapfd, attach_type); if (err) {

@@ -792,15 +807,16 @@ static int do_detach(int argc, char **argv) jsonw_null(json_wtr); return 0; } -static int do_load(int argc, char **argv)

+static int load_with_options(int argc, char **argv, bool first_prog_only) { enum bpf_attach_type expected_attach_type; struct bpf_object_open_attr attr = { .prog_type = BPF_PROG_TYPE_UNSPEC, }; struct map_replace *map_replace = NULL;

• struct bpf_program *prog = NULL, *pos; unsigned int old_map_fds = 0;

• struct bpf_program *prog; struct bpf_object *obj; struct bpf_map *map; const char *pinfile;

@@ -918,14 +934,20 @@ static int do_load(int argc, char **argv) goto err_free_reuse_maps; }

• prog = bpf_program__next(NULL, obj);
• if (!prog) {

• p_err("object file doesn't contain any bpf program");
  

• goto err_close_obj;
  

• if (first_prog_only) {

• prog = bpf_program__next(NULL, obj);
  

• if (!prog) {
  

• 	p_err("object file doesn't contain any bpf program");
  

• 	goto err_close_obj;
  

• }
  

  }

• bpf_program__set_ifindex(prog, ifindex); if (attr.prog_type == BPF_PROG_TYPE_UNSPEC) {

• if (!prog) {
  

• 	p_err("can not guess program type when loading all programs\n");
  

• 	goto err_close_obj;
  

• }
  

• const char *sec_name = bpf_program__title(prog, false); err = libbpf_prog_type_by_name(sec_name, &attr.prog_type,

@@ -936,8 +958,13 @@ static int do_load(int argc, char **argv) goto err_close_obj; } }

• bpf_program__set_type(prog, attr.prog_type);
• bpf_program__set_expected_attach_type(prog, expected_attach_type);

• bpf_object__for_each_program(pos, obj) {

• bpf_program__set_ifindex(pos, ifindex);
  

• bpf_program__set_type(pos, attr.prog_type);
  

• bpf_program__set_expected_attach_type(pos,
  

• 				      expected_attach_type);
  

• }

I still believe you can have programs of different types here, and be able to load them. I tried it and managed to have it working fine. If no type is provided from command line we can retrieve types for each program from its section name. If a type is provided on the command line, we can do the same, but I am not sure we should do it, or impose that type for all programs instead.

I can move auto-detection into this new bpf_object__for_each_program loop. So if no type is specified, try to infer the type from each prog section name, otherwise, use the provided one for all progs. Do we want something like that?

This is what I have in mind. But others may disagree.

Hm, I did another look and here is the problem. Current **load** implementation actually loads all progs (and maps) from the file: do_load -> bpf_object__load -> bpf_object__load_progs. It pins only the first prog, but loads them all. That's why I don't understand how multiprog loading can work currently (because we don't set correct prog_type/expected_attach_type for all progs, only the first one).

Should we have the following functionality instead: Both **load** and **loadall** load all maps and programs from the *OBJ* and differ only in pinning. **load** pins only the first program from the *OBJ* as *FILE*. **loadall** pins all programs and maps from the *OBJ* under *FILE* directory.

If we want **load** to load only the first prog, we should probably amend bpf_object__load to be able to filter the progs we (don't) want.

Thoughts?

...

Btw, do you have some existing real life example of where it's needed so I can test this new implementation? (maybe something under samples/ ?)

I thought about an ELF file containing both an XDP and a TC classifier program for example. XDP can mark programs for TC, then TC process them with all the facilities we have for skbs. It does not _have_ to be in the same ELF file, but could be.

I haven't searched samples/bpf/ in depth, but a grep on SEC shows a couple of files with several types (kprobe/kretprobe, classifier/xdp). samples/bpf/xdp2skb_meta_kern.c looks like a good candidate. Or actually for testing purposes, I simply used the following:

#define SEC(NAME) __attribute__((section(NAME), used))

int _version SEC("version") = 1;

SEC("classifier") int func() { return 1; }

SEC("xdp") int funcbar() { return 0; }

Thanks, I just found socket_cookie_prog.c in selftests which has cgroup/connect6 and sockops sections, I can probably use that for testing.

...

...

...

qsort(map_replace, old_map_fds, sizeof(*map_replace), map_replace_compar); @@ -1001,9 +1028,25 @@ static int do_load(int argc, char **argv) goto err_close_obj; }

• if (do_pin_fd(bpf_program__fd(prog), pinfile))

• err = mount_bpffs_for_pin(pinfile);
• if (err) goto err_close_obj;
• if (prog) {

Nit: Maybe "if (first_prog_only) {" instead? If I understand correctly, at this stage it should be equivalent, but in my opinion it would make it easier to understand why we have two cases here.

Sure, I can do that if you think that's more readable, I don't have a preference.

Thanks! Quentin