On Tue, Feb 15, 2022 at 2:59 PM Mauricio Vásquez <mauricio@xxxxxxxxxx> wrote:
>
> The last part of the BTFGen algorithm is to create a new BTF object with
> all the types that were recorded in the previous steps.
>
> This function performs two different steps:
> 1. Add the types to the new BTF object by using btf__add_type(). Some
> special logic around struct and unions is implemented to only add the
> members that are really used in the field-based relocations. The type
> ID on the new and old BTF objects is stored on a map.
> 2. Fix all the type IDs on the new BTF object by using the IDs saved in
> the previous step.
>
> Signed-off-by: Mauricio Vásquez <mauricio@xxxxxxxxxx>
> Signed-off-by: Rafael David Tinoco <rafael.tinoco@xxxxxxxxxxx>
> Signed-off-by: Lorenzo Fontana <lorenzo.fontana@xxxxxxxxxx>
> Signed-off-by: Leonardo Di Donato <leonardo.didonato@xxxxxxxxxx>
> ---
> tools/bpf/bpftool/gen.c | 100 +++++++++++++++++++++++++++++++++++++++-
> 1 file changed, 99 insertions(+), 1 deletion(-)
>
[...]
> + cloned_m = btf_members(cloned_type);
> + m = btf_members(type);
> + vlen = btf_vlen(cloned_type);
> + for (idx_src = 0; idx_src < vlen; idx_src++, cloned_m++, m++) {
> + /* add only members that are marked as used */
> + if (cloned_m->name_off != MARKED)
> + continue;
> +
> + name = btf__str_by_offset(info->src_btf, m->name_off);
> + err = btf__add_field(btf_new, name, m->type,
> + BTF_MEMBER_BIT_OFFSET(m->offset),
> + BTF_MEMBER_BITFIELD_SIZE(m->offset));
BTF_MEMBER_BIT_OFFSET() and BTF_MEMBER_BIT_OFFSET() shouldn't be used
unconditionally, only if kflag is set. It's better to use
btf_member_bit_offset() and btf_member_bitfield_size() helpers here,
they handle this transparently.
> + if (err < 0)
> + goto err_out;
> + }
> + } else {
> + err = btf__add_type(btf_new, info->src_btf, type);
> + if (err < 0)
> + goto err_out;
> + new_id = err;
> + }
> +
> + /* add ID mapping */
> + ids[i] = new_id;
> + }
> +
[...]
