On Wed, Nov 06, 2019 at 06:41:15PM -0800, Andrii Nakryiko wrote: > On Wed, Nov 6, 2019 at 5:49 PM Martin KaFai Lau <kafai@xxxxxx> wrote: > > > > This patch adds array support to btf_struct_access(). > > It supports array of int, array of struct and multidimensional > > array. > > > > It also allows using u8[] as a scratch space. For example, > > it allows access the "char cb[48]" with size larger than > > the array's element "char". Another potential use case is > > "u64 icsk_ca_priv[]" in the tcp congestion control. > > > > btf_resolve_size() is added to resolve the size of any type. > > It will follow the modifier if there is any. Please > > see the function comment for details. > > > > This patch also adds the "off < moff" check at the beginning > > of the for loop. It is to reject cases when "off" is pointing > > to a "hole" in a struct. > > > > Signed-off-by: Martin KaFai Lau <kafai@xxxxxx> > > --- > > Looks good, just two small nits. > > Acked-by: Andrii Nakryiko <andriin@xxxxxx> > > > kernel/bpf/btf.c | 187 +++++++++++++++++++++++++++++++++++++++-------- > > 1 file changed, 157 insertions(+), 30 deletions(-) > > > > diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c > > index 128d89601d73..5c4b6aa7b9f0 100644 > > --- a/kernel/bpf/btf.c > > +++ b/kernel/bpf/btf.c > > @@ -1036,6 +1036,82 @@ static const struct resolve_vertex *env_stack_peak(struct btf_verifier_env *env) > > return env->top_stack ? &env->stack[env->top_stack - 1] : NULL; > > } > > > > [...] > > > - if (off + size <= moff / 8) > > - /* won't find anything, field is already too far */ > > + /* offset of the field in bytes */ > > + moff = btf_member_bit_offset(t, member) / 8; > > + if (off + size <= moff) > > you dropped useful comment :( good catch. will undo. > > > break; > > + /* In case of "off" is pointing to holes of a struct */ > > + if (off < moff) > > + continue; > > > > [...] > > > + > > + mtrue_end = moff + msize; > > nit: there is no other _end, so might be just mend (in line with moff) I prefer to keep it. For array, this _end is not the end of mtype. The intention is to distinguish it from the mtype/msize convention such that it is the true_end of the current struct's member. I will add some comments to clarify. > > > + if (off >= mtrue_end) > > /* no overlap with member, keep iterating */ > > continue; > > + > > [...]
