[bpf-next] bpf: Load vmlinux btf for any struct_ops map

Commit Message

David Vernet Dec. 8, 2023, 6:17 a.m. UTC

In libbpf, when determining whether we need to load vmlinux btf, we're
currently (among other things) checking whether there is any struct_ops
program present in the object. This works for most realistic struct_ops
maps, as a struct_ops map is of course typically composed of one or more
struct_ops programs. However, that technically need not be the case. A
struct_ops interface could be defined which allows a map to be specified
which one or more non-prog fields, and which provides default behavior
if no struct_ops progs is actually provided otherwise. For sched_ext,
for example, you technically only need to specify the name of the
scheduler in the struct_ops map, with the core scheduler logic providing
default behavior if no prog is actually specified.

If we were to define and try to load such a struct_ops map, we would
crash in libbpf when initializing it as obj->btf_vmlinux will be NULL:

Reading symbols from minimal...
(gdb) r
Starting program: minimal_example
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/usr/lib/libthread_db.so.1".

Program received signal SIGSEGV, Segmentation fault.
0x000055555558308c in btf__type_cnt (btf=0x0) at btf.c:612
612             return btf->start_id + btf->nr_types;
(gdb) bt
    type_name=0x5555555d99e3 "sched_ext_ops", kind=4) at btf.c:914
    kind=4) at btf.c:942
    type=0x7fffffffe558, type_id=0x7fffffffe548, ...
    data_member=0x7fffffffe568) at libbpf.c:948
    kern_btf=0x0) at libbpf.c:1017
    at libbpf.c:8059

So as to account for such bare-bones struct_ops maps, let's update
obj_needs_vmlinux_btf() to also iterate over an obj's maps and check
whether any of them are struct_ops maps.

Signed-off-by: David Vernet <[e-mail]>
---
 tools/lib/bpf/libbpf.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

Comments

Alan Maguire Dec. 8, 2023, 10:05 a.m. UTC | #1

On 08/12/2023 06:17, David Vernet wrote:
> In libbpf, when determining whether we need to load vmlinux btf, we're
> currently (among other things) checking whether there is any struct_ops
> program present in the object. This works for most realistic struct_ops
> maps, as a struct_ops map is of course typically composed of one or more
> struct_ops programs. However, that technically need not be the case. A
> struct_ops interface could be defined which allows a map to be specified
> which one or more non-prog fields, and which provides default behavior
> if no struct_ops progs is actually provided otherwise. For sched_ext,
> for example, you technically only need to specify the name of the
> scheduler in the struct_ops map, with the core scheduler logic providing
> default behavior if no prog is actually specified.
> 
> If we were to define and try to load such a struct_ops map, we would
> crash in libbpf when initializing it as obj->btf_vmlinux will be NULL:
> 
> Reading symbols from minimal...
> (gdb) r
> Starting program: minimal_example
> [Thread debugging using libthread_db enabled]
> Using host libthread_db library "/usr/lib/libthread_db.so.1".
> 
> Program received signal SIGSEGV, Segmentation fault.
> 0x000055555558308c in btf__type_cnt (btf=0x0) at btf.c:612
> 612             return btf->start_id + btf->nr_types;
> (gdb) bt
>     type_name=0x5555555d99e3 "sched_ext_ops", kind=4) at btf.c:914
>     kind=4) at btf.c:942
>     type=0x7fffffffe558, type_id=0x7fffffffe548, ...
>     data_member=0x7fffffffe568) at libbpf.c:948
>     kern_btf=0x0) at libbpf.c:1017
>     at libbpf.c:8059
> 
> So as to account for such bare-bones struct_ops maps, let's update
> obj_needs_vmlinux_btf() to also iterate over an obj's maps and check
> whether any of them are struct_ops maps.
> 
> Signed-off-by: David Vernet <[e-mail]>

Makes sense to me.

Reviewed-by: Alan Maguire <[e-mail]>

> ---
>  tools/lib/bpf/libbpf.c | 11 +++++++++++
>  1 file changed, 11 insertions(+)
> 
> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
> index ea9b8158c20d..ac54ebc0629f 100644
> --- a/tools/lib/bpf/libbpf.c
> +++ b/tools/lib/bpf/libbpf.c
> @@ -3054,9 +3054,15 @@ static bool prog_needs_vmlinux_btf(struct bpf_program *prog)
>  	return false;
>  }
>  
> +static bool map_needs_vmlinux_btf(struct bpf_map *map)
> +{
> +	return bpf_map__is_struct_ops(map);
> +}
> +
>  static bool obj_needs_vmlinux_btf(const struct bpf_object *obj)
>  {
>  	struct bpf_program *prog;
> +	struct bpf_map *map;
>  	int i;
>  
>  	/* CO-RE relocations need kernel BTF, only when btf_custom_path
> @@ -3081,6 +3087,11 @@ static bool obj_needs_vmlinux_btf(const struct bpf_object *obj)
>  			return true;
>  	}
>  
> +	bpf_object__for_each_map(map, obj) {
> +		if (map_needs_vmlinux_btf(map))
> +			return true;
> +	}
> +
>  	return false;
>  }
>

[e-mail] Dec. 8, 2023, 5:40 p.m. UTC | #2

Hello:

This patch was applied to bpf/bpf-next.git (master)
by Andrii Nakryiko <[e-mail]>:

On Fri,  8 Dec 2023 00:17:03 -0600 you wrote:
> In libbpf, when determining whether we need to load vmlinux btf, we're
> currently (among other things) checking whether there is any struct_ops
> program present in the object. This works for most realistic struct_ops
> maps, as a struct_ops map is of course typically composed of one or more
> struct_ops programs. However, that technically need not be the case. A
> struct_ops interface could be defined which allows a map to be specified
> which one or more non-prog fields, and which provides default behavior
> if no struct_ops progs is actually provided otherwise. For sched_ext,
> for example, you technically only need to specify the name of the
> scheduler in the struct_ops map, with the core scheduler logic providing
> default behavior if no prog is actually specified.
> 
> [...]

Here is the summary with links:
  - [bpf-next] bpf: Load vmlinux btf for any struct_ops map
    https://git.kernel.org/bpf/bpf-next/c/8b7b0e5fe47d

You are awesome, thank you!

Patch

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index ea9b8158c20d..ac54ebc0629f 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -3054,9 +3054,15 @@  static bool prog_needs_vmlinux_btf(struct bpf_program *prog)
 	return false;
 }
 
+static bool map_needs_vmlinux_btf(struct bpf_map *map)
+{
+	return bpf_map__is_struct_ops(map);
+}
+
 static bool obj_needs_vmlinux_btf(const struct bpf_object *obj)
 {
 	struct bpf_program *prog;
+	struct bpf_map *map;
 	int i;
 
 	/* CO-RE relocations need kernel BTF, only when btf_custom_path
@@ -3081,6 +3087,11 @@  static bool obj_needs_vmlinux_btf(const struct bpf_object *obj)
 			return true;
 	}
 
+	bpf_object__for_each_map(map, obj) {
+		if (map_needs_vmlinux_btf(map))
+			return true;
+	}
+
 	return false;
 }