Received-SPF: pass (google.com: domain of linux-kernel+bounces-99754-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240311093526.1010158-1-dongmenglong.8@bytedance.com>
 <20240311093526.1010158-8-dongmenglong.8@bytedance.com> <CAADnVQK4tdefa3s=sim69Sc+ztd-hHohPEDXaUNVTU-mLNYUiw@mail.gmail.com>
In-Reply-To: <CAADnVQK4tdefa3s=sim69Sc+ztd-hHohPEDXaUNVTU-mLNYUiw@mail.gmail.com>
From: =?UTF-8?B?5qKm6b6Z6JGj?= <dongmenglong.8@bytedance.com>
Date: Tue, 12 Mar 2024 10:05:35 +0800
Message-ID: <CALz3k9iabeOwHSrPb9mkfCuOebanh3+bAfi7xh3kBBN0DzHC3A@mail.gmail.com>
Subject: Re: [External] Re: [PATCH bpf-next v2 7/9] libbpf: don't free btf if
 program of multi-link tracing existing
To: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Andrii Nakryiko <andrii@kernel.org>, Alexei Starovoitov <ast@kernel.org>, 
	Daniel Borkmann <daniel@iogearbox.net>, Martin KaFai Lau <martin.lau@linux.dev>, Eddy Z <eddyz87@gmail.com>, 
	Song Liu <song@kernel.org>, Yonghong Song <yonghong.song@linux.dev>, 
	John Fastabend <john.fastabend@gmail.com>, KP Singh <kpsingh@kernel.org>, 
	Stanislav Fomichev <sdf@google.com>, Hao Luo <haoluo@google.com>, Jiri Olsa <jolsa@kernel.org>, 
	Alexander Gordeev <agordeev@linux.ibm.com>, Christian Borntraeger <borntraeger@linux.ibm.com>, 
	Sven Schnelle <svens@linux.ibm.com>, "David S. Miller" <davem@davemloft.net>, 
	David Ahern <dsahern@kernel.org>, Dave Hansen <dave.hansen@linux.intel.com>, 
	X86 ML <x86@kernel.org>, Steven Rostedt <rostedt@goodmis.org>, 
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>, Quentin Monnet <quentin@isovalent.com>, 
	bpf <bpf@vger.kernel.org>, 
	linux-arm-kernel <linux-arm-kernel@lists.infradead.org>, LKML <linux-kernel@vger.kernel.org>, 
	linux-riscv <linux-riscv@lists.infradead.org>, linux-s390 <linux-s390@vger.kernel.org>, 
	Network Development <netdev@vger.kernel.org>, linux-trace-kernel@vger.kernel.org, 
	"open list:KERNEL SELFTEST FRAMEWORK" <linux-kselftest@vger.kernel.org>, linux-stm32@st-md-mailman.stormreply.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Tue, Mar 12, 2024 at 9:55=E2=80=AFAM Alexei Starovoitov
<alexei.starovoitov@gmail.com> wrote:
>
> On Mon, Mar 11, 2024 at 2:35=E2=80=AFAM Menglong Dong
> <dongmenglong.8@bytedance.com> wrote:
> >
> > By default, the kernel btf that we load during loading program will be
> > freed after the programs are loaded in bpf_object_load(). However, we
> > still need to use these btf for tracing of multi-link during attaching.
> > Therefore, we don't free the btfs until the bpf object is closed if any
> > bpf programs of the type multi-link tracing exist.
> >
> > Meanwhile, introduce the new api bpf_object__free_btf() to manually fre=
e
> > the btfs after attaching.
> >
> > Signed-off-by: Menglong Dong <dongmenglong.8@bytedance.com>
> > ---
> >  tools/lib/bpf/libbpf.c   | 47 ++++++++++++++++++++++++++++++----------
> >  tools/lib/bpf/libbpf.h   |  2 ++
> >  tools/lib/bpf/libbpf.map |  1 +
> >  3 files changed, 38 insertions(+), 12 deletions(-)
> >
> > diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
> > index 567ad367e7aa..fd5428494a7e 100644
> > --- a/tools/lib/bpf/libbpf.c
> > +++ b/tools/lib/bpf/libbpf.c
> > @@ -8267,6 +8267,39 @@ static int bpf_object_prepare_struct_ops(struct =
bpf_object *obj)
> >         return 0;
> >  }
> >
> > +void bpf_object__free_btfs(struct bpf_object *obj)
> > +{
> > +       int i;
> > +
> > +       /* clean up module BTFs */
> > +       for (i =3D 0; i < obj->btf_module_cnt; i++) {
> > +               close(obj->btf_modules[i].fd);
> > +               btf__free(obj->btf_modules[i].btf);
> > +               free(obj->btf_modules[i].name);
> > +       }
> > +       free(obj->btf_modules);
> > +       obj->btf_modules =3D NULL;
> > +       obj->btf_module_cnt =3D 0;
> > +
> > +       /* clean up vmlinux BTF */
> > +       btf__free(obj->btf_vmlinux);
> > +       obj->btf_vmlinux =3D NULL;
> > +}
> > +
> > +static void bpf_object_early_free_btf(struct bpf_object *obj)
> > +{
> > +       struct bpf_program *prog;
> > +
> > +       bpf_object__for_each_program(prog, obj) {
> > +               if (prog->expected_attach_type =3D=3D BPF_TRACE_FENTRY_=
MULTI ||
> > +                   prog->expected_attach_type =3D=3D BPF_TRACE_FEXIT_M=
ULTI ||
> > +                   prog->expected_attach_type =3D=3D BPF_MODIFY_RETURN=
_MULTI)
> > +                       return;
> > +       }
> > +
> > +       bpf_object__free_btfs(obj);
> > +}
> > +
> >  static int bpf_object_load(struct bpf_object *obj, int extra_log_level=
, const char *target_btf_path)
> >  {
> >         int err, i;
> > @@ -8307,18 +8340,7 @@ static int bpf_object_load(struct bpf_object *ob=
j, int extra_log_level, const ch
> >         /* clean up fd_array */
> >         zfree(&obj->fd_array);
> >
> > -       /* clean up module BTFs */
> > -       for (i =3D 0; i < obj->btf_module_cnt; i++) {
> > -               close(obj->btf_modules[i].fd);
> > -               btf__free(obj->btf_modules[i].btf);
> > -               free(obj->btf_modules[i].name);
> > -       }
> > -       free(obj->btf_modules);
> > -
> > -       /* clean up vmlinux BTF */
> > -       btf__free(obj->btf_vmlinux);
> > -       obj->btf_vmlinux =3D NULL;
> > -
> > +       bpf_object_early_free_btf(obj);
> >         obj->loaded =3D true; /* doesn't matter if successfully or not =
*/
> >
> >         if (err)
> > @@ -8791,6 +8813,7 @@ void bpf_object__close(struct bpf_object *obj)
> >         usdt_manager_free(obj->usdt_man);
> >         obj->usdt_man =3D NULL;
> >
> > +       bpf_object__free_btfs(obj);
> >         bpf_gen__free(obj->gen_loader);
> >         bpf_object__elf_finish(obj);
> >         bpf_object_unload(obj);
> > diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
> > index 5723cbbfcc41..c41a909ea4c1 100644
> > --- a/tools/lib/bpf/libbpf.h
> > +++ b/tools/lib/bpf/libbpf.h
> > @@ -299,6 +299,8 @@ LIBBPF_API struct bpf_program *
> >  bpf_object__find_program_by_name(const struct bpf_object *obj,
> >                                  const char *name);
> >
> > +LIBBPF_API void bpf_object__free_btfs(struct bpf_object *obj);
> > +
>
> It shouldn't be exported.
> libbpf should clean it up when bpf_object is freed.

Yes, libbpf will clean up the btfs when bpf_object is freed in
this commit. And I'm trying to offer a way to early free the btfs
by the users manual to reduce the memory usage. Or, the
btfs that we opened will keep existing until we close the
bpf_object.

This is optional, I can remove it if you prefer.

Thanks!
Menglong Dong