netdev - Re: [PATCH 2/3] bpf_encoder: Translate SHN_XINDEX in symbol's st

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAEf4BzZquSn0Th7bpVuM0M4XbTPU5-9jDPPd5RJBS5AH2zqaMA@mail.gmail.com>
Date:   Thu, 21 Jan 2021 15:32:40 -0800
From:   Andrii Nakryiko <andrii.nakryiko@...il.com>
To:     Jiri Olsa <jolsa@...nel.org>
Cc:     Arnaldo Carvalho de Melo <acme@...nel.org>,
        Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Andrii Nakryiko <andriin@...com>, dwarves@...r.kernel.org,
        Networking <netdev@...r.kernel.org>, bpf <bpf@...r.kernel.org>,
        Yonghong Song <yhs@...com>, Hao Luo <haoluo@...gle.com>,
        Martin KaFai Lau <kafai@...com>,
        Song Liu <songliubraving@...com>,
        John Fastabend <john.fastabend@...il.com>,
        KP Singh <kpsingh@...omium.org>,
        Joe Lawrence <joe.lawrence@...hat.com>,
        Mark Wielaard <mjw@...hat.com>
Subject: Re: [PATCH 2/3] bpf_encoder: Translate SHN_XINDEX in symbol's
 st_shndx values

On Thu, Jan 21, 2021 at 12:25 PM Jiri Olsa <jolsa@...nel.org> wrote:
>
> For very large ELF objects (with many sections), we could
> get special value SHN_XINDEX (65535) for symbol's st_shndx.
>
> This patch is adding code to detect the optional extended
> section index table and use it to resolve symbol's section
> index.
>
> Adding elf_symtab__for_each_symbol_index macro that returns
> symbol's section index and usign it in collect_symbols function.
>
> Signed-off-by: Jiri Olsa <jolsa@...nel.org>
> ---

You missed fixing up collect_function() as well, which is using
elf_sym__section(), which doesn't know about extended numbering.

>  btf_encoder.c | 36 ++++++++++++++++++++++++++++++++----
>  elf_symtab.c  | 39 ++++++++++++++++++++++++++++++++++++++-
>  elf_symtab.h  |  2 ++
>  3 files changed, 72 insertions(+), 5 deletions(-)
>
> diff --git a/btf_encoder.c b/btf_encoder.c
> index 5557c9efd365..6e6f22c438ce 100644
> --- a/btf_encoder.c
> +++ b/btf_encoder.c
> @@ -585,12 +585,13 @@ static int collect_percpu_var(struct btf_elf *btfe, GElf_Sym *sym)
>         return 0;
>  }
>
> -static void collect_symbol(GElf_Sym *sym, struct funcs_layout *fl)
> +static void collect_symbol(GElf_Sym *sym, struct funcs_layout *fl,
> +                          Elf32_Word sym_sec_idx)
>  {
>         if (!fl->mcount_start &&
>             !strcmp("__start_mcount_loc", elf_sym__name(sym, btfe->symtab))) {
>                 fl->mcount_start = sym->st_value;
> -               fl->mcount_sec_idx = sym->st_shndx;
> +               fl->mcount_sec_idx = sym_sec_idx;
>         }
>
>         if (!fl->mcount_stop &&
> @@ -598,9 +599,36 @@ static void collect_symbol(GElf_Sym *sym, struct funcs_layout *fl)
>                 fl->mcount_stop = sym->st_value;
>  }
>
> +static bool elf_sym__get(Elf_Data *syms, Elf_Data *syms_sec_idx_table,
> +                        int id, GElf_Sym *sym, Elf32_Word *sym_sec_idx)
> +{
> +       if (!gelf_getsym(syms, id, sym))
> +               return false;
> +
> +       *sym_sec_idx = sym->st_shndx;
> +
> +       if (sym->st_shndx == SHN_XINDEX) {
> +               if (!syms_sec_idx_table)
> +                       return false;
> +               if (!gelf_getsymshndx(syms, syms_sec_idx_table,
> +                                     id, sym, sym_sec_idx))


gelf_getsymshndx() is supposed to work even for cases that don't use
extended numbering, so this should work, right?

if (!gelf_getsymshndx(syms, syms_sec_idx_table, id, sym, sym_sec_idx))
    return false;

if (sym->st_shndx == SHN_XINDEX)
  *sym_sec_idx = sym->st_shndx;

return true;

?

> +                       return false;
> +       }
> +
> +       return true;
> +}
> +
> +#define elf_symtab__for_each_symbol_index(symtab, id, sym, sym_sec_idx)                \
> +       for (id = 0, elf_sym__get(symtab->syms, symtab->syms_sec_idx_table,     \
> +                                 id, &sym, &sym_sec_idx);                      \
> +            id < symtab->nr_syms;                                              \
> +            id++, elf_sym__get(symtab->syms, symtab->syms_sec_idx_table,       \
> +                               id, &sym, &sym_sec_idx))

what do we want to do if elf_sym__get() returns error (false)? We can
either stop or ignore that symbol, right? But currently you are
returning invalid symbol data.

so either

for (id = 0; id < symtab->nr_syms && elf_sym__get(symtab->syms,
symtab->syms_sec_idx_table, d, &sym, &sym_sec_idx); id++)

or

for (id = 0; id < symtab->nr_syms; id++)
  if (elf_sym__get(symtab->syms, symtab->syms_sec_idx_table, d, &sym,
&sym_sec_idx))


But the current variant looks broken. Oh, and
elf_symtab__for_each_symbol() is similarly broken, can you please fix
that as well?

And this new macro should probably be in elf_symtab.h, along the
elf_symtab__for_each_symbol.


> +
>  static int collect_symbols(struct btf_elf *btfe, bool collect_percpu_vars)
>  {
>         struct funcs_layout fl = { };
> +       Elf32_Word sym_sec_idx;
>         uint32_t core_id;
>         GElf_Sym sym;
>

[...]