Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Mon, 07 Dec 2020 22:42:02 -0800
From:   John Fastabend <john.fastabend@gmail.com>
To:     Brendan Jackman <jackmanb@google.com>, bpf@vger.kernel.org
Cc:     Alexei Starovoitov <ast@kernel.org>, Yonghong Song <yhs@fb.com>,
        Daniel Borkmann <daniel@iogearbox.net>,
        KP Singh <kpsingh@chromium.org>,
        Florent Revest <revest@chromium.org>,
        linux-kernel@vger.kernel.org, Jann Horn <jannh@google.com>,
        Brendan Jackman <jackmanb@google.com>
Message-ID: <5fcf203ab087a_d22720855@john-XPS-13-9370.notmuch>
In-Reply-To: <20201207160734.2345502-8-jackmanb@google.com>
References: <20201207160734.2345502-1-jackmanb@google.com>
 <20201207160734.2345502-8-jackmanb@google.com>
Subject: RE: [PATCH bpf-next v4 07/11] bpf: Add instructions for
 atomic_[cmp]xchg
Mime-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

Brendan Jackman wrote:
> This adds two atomic opcodes, both of which include the BPF_FETCH
> flag. XCHG without the BPF_FETCH flag would naturally encode
> atomic_set. This is not supported because it would be of limited
> value to userspace (it doesn't imply any barriers). CMPXCHG without
> BPF_FETCH woulud be an atomic compare-and-write. We don't have such
> an operation in the kernel so it isn't provided to BPF either.
> 
> There are two significant design decisions made for the CMPXCHG
> instruction:
> 
>  - To solve the issue that this operation fundamentally has 3
>    operands, but we only have two register fields. Therefore the
>    operand we compare against (the kernel's API calls it 'old') is
>    hard-coded to be R0. x86 has similar design (and A64 doesn't
>    have this problem).
> 
>    A potential alternative might be to encode the other operand's
>    register number in the immediate field.
> 
>  - The kernel's atomic_cmpxchg returns the old value, while the C11
>    userspace APIs return a boolean indicating the comparison
>    result. Which should BPF do? A64 returns the old value. x86 returns
>    the old value in the hard-coded register (and also sets a
>    flag). That means return-old-value is easier to JIT.
> 
> Signed-off-by: Brendan Jackman <jackmanb@google.com>
> ---

Sorry if this is a dup, client crashed while I sent the previous version
and don't see it on the list.

> --- a/kernel/bpf/verifier.c
> +++ b/kernel/bpf/verifier.c
> @@ -3608,11 +3608,14 @@ static int check_mem_access(struct bpf_verifier_env *env, int insn_idx, u32 regn
>  
>  static int check_atomic(struct bpf_verifier_env *env, int insn_idx, struct bpf_insn *insn)
>  {
> +	int load_reg;
>  	int err;
>  
>  	switch (insn->imm) {
>  	case BPF_ADD:
>  	case BPF_ADD | BPF_FETCH:
> +	case BPF_XCHG:
> +	case BPF_CMPXCHG:
>  		break;
>  	default:
>  		verbose(env, "BPF_ATOMIC uses invalid atomic opcode %02x\n", insn->imm);
> @@ -3634,6 +3637,13 @@ static int check_atomic(struct bpf_verifier_env *env, int insn_idx, struct bpf_i
>  	if (err)
>  		return err;
>  
> +	if (insn->imm == BPF_CMPXCHG) {
> +		/* Check comparison of R0 with memory location */
> +		err = check_reg_arg(env, BPF_REG_0, SRC_OP);
> +		if (err)
> +			return err;
> +	}
> +

Need to think a bit more on this, but do we need to update is_reg64() here
as well?

>  	if (is_pointer_value(env, insn->src_reg)) {
>  		verbose(env, "R%d leaks addr into mem\n", insn->src_reg);
>  		return -EACCES;
> @@ -3664,8 +3674,13 @@ static int check_atomic(struct bpf_verifier_env *env, int insn_idx, struct bpf_i
>  	if (!(insn->imm & BPF_FETCH))
>  		return 0;
>  
> -	/* check and record load of old value into src reg  */
> -	err = check_reg_arg(env, insn->src_reg, DST_OP);
> +	if (insn->imm == BPF_CMPXCHG)
> +		load_reg = BPF_REG_0;
> +	else
> +		load_reg = insn->src_reg;
> +
> +	/* check and record load of old value */
> +	err = check_reg_arg(env, load_reg, DST_OP);
>  	if (err)
>  		return err;
>  

Thanks,
John