Re: [PATCH bpf-next] bpf: check for insn arrays in check_ptr_alignment

[Anton Protopopov <a.s.protopopov@gmail.com>]

On 25/11/27 09:25PM, bot+bpf-ci@kernel.org wrote:
> > diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
> > index 766695491..43991449f 100644
> > --- a/kernel/bpf/verifier.c
> > +++ b/kernel/bpf/verifier.c
> > @@ -6482,6 +6482,8 @@ static int check_ptr_alignment(struct bpf_verifier_env *env,
> >  		break;
> >  	case PTR_TO_MAP_VALUE:
> >  		pointer_desc = "value ";
> > +		if (reg->map_ptr->map_type == BPF_MAP_TYPE_INSN_ARRAY)
> > +			strict = true;
> >  		break;
> >  	case PTR_TO_CTX:
> >  		pointer_desc = "context ";
> > @@ -7529,8 +7531,6 @@ static int check_mem_access(struct bpf_verifier_env *env, int insn_idx, u32 regn
> >  {
> >  	struct bpf_reg_state *regs = cur_regs(env);
> >  	struct bpf_reg_state *reg = regs + regno;
> > -	bool insn_array = reg->type == PTR_TO_MAP_VALUE &&
> > -			  reg->map_ptr->map_type == BPF_MAP_TYPE_INSN_ARRAY;
> >  	int size, err = 0;
> >
> >  	size = bpf_size_to_bytes(bpf_size);
> > @@ -7538,7 +7538,7 @@ static int check_mem_access(struct bpf_verifier_env *env, int insn_idx, u32 regn
> >  		return size;
> >
> >  	/* alignment checks will add in reg->off themselves */
> > -	err = check_ptr_alignment(env, reg, off, size, strict_alignment_once || insn_array);
> > +	err = check_ptr_alignment(env, reg, off, size, strict_alignment_once);
> >  	if (err)
> >  		return err;
> 
> The insn_array variable declaration was removed, but this code still
> uses it at check_mem_access():7568. Does this compile?

Robot is right, I should have sent it as a series with
https://lore.kernel.org/bpf/20251127210656.3239541-1-a.s.protopopov@gmail.com/
Will resend it now.

> 	if (t == BPF_WRITE && insn_array) {
> 		verbose(env, "writes into insn_array not allowed\n");
> 		return -EACCES;
> 	}
> 
> The fix would be to check the map type directly:
> 
> 	if (t == BPF_WRITE && reg->type == PTR_TO_MAP_VALUE &&
> 	    reg->map_ptr->map_type == BPF_MAP_TYPE_INSN_ARRAY) {
> 
> 
> ---
> AI reviewed your patch. Please fix the bug or email reply why it's not a bug.
> See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md
> 
> CI run summary: https://github.com/kernel-patches/bpf/actions/runs/19747897022