On Thu, Sep 11, 2014 at 1:33 PM, Alexei Starovoitov <a...@plumgrid.com> wrote:
> On Thu, Sep 11, 2014 at 12:47 PM, Daniel Borkmann <dbork...@redhat.com> wrote:
>> On 09/10/2014 07:32 PM, Alexei Starovoitov wrote:
>>>
>>> On Wed, Sep 10, 2014 at 2:03 AM, Daniel Borkmann <dbork...@redhat.com>
>>> wrote:
>>>>>
>>>>> struct { /* anonymous struct used by BPF_PROG_LOAD command
>>>>> */
>>>>> enum bpf_prog_type prog_type;
>>>>> __u32 insn_cnt;
>>>>> const struct bpf_insn *insns;
>>>>> const char *license;
>>>>> __u32 log_level; /* verbosity level of
>>>>> eBPF verifier */
>>>>> __u32 log_size; /* size of user buffer
>>>>> */
>>>>> void *log_buf; /* user supplied
>>>>> buffer
>>>>> */
>>>>
>>>>
>>>>
>>>> What is log buffer? Would that mean the verifier will return an error
>>>> string if the program will not pass it, or if not, what other data?
>>>> I think the man page is missing how to examine the returned verifier
>>>> log buffer data.
>>>
>>>
>>> yes. it's an error log (as text string for humans) from verifier.
>>
>> I was confused due to the void pointer. But that also means that the text
>
> ahh. ok. will change it to 'char *' then.
>
>> string becomes part of the ABI; aren't eBPF specific error codes (perhaps
>> a tuple of [line + error code]), though ugly as well, but perhaps the better
>> solution to this [which user space can then map to an actual string]?
>
> the verifier log contains full trace. Last unsafe instruction + error
> in many cases is useless. What we found empirically from using
> it over last 2 years is that developers have different learning curve
> to adjust to 'safe' style of C. Pretty much everyone couldn't
> figure out why program is rejected based on last error. Therefore
> verifier emits full log. From the 1st insn all the way till the last
> 'unsafe' instruction. So the log is multiline output.
> 'Understanding eBPF verifier messages' section of
> Documentation/networking/filter.txt provides few trivial
> examples of these multiline messages.
> Like for the program:
> BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
> BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
> BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
> BPF_LD_MAP_FD(BPF_REG_1, 0),
> BPF_CALL_FUNC(BPF_FUNC_map_lookup_elem),
> BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
> BPF_ST_MEM(BPF_DW, BPF_REG_0, 4, 0),
> BPF_EXIT_INSN(),
> the verifier log_buf is:
> 0: (7a) *(u64 *)(r10 -8) = 0
> 1: (bf) r2 = r10
> 2: (07) r2 += -8
> 3: (b7) r1 = 0
> 4: (85) call 1
> 5: (15) if r0 == 0x0 goto pc+1
> R0=map_ptr R10=fp
> 6: (7a) *(u64 *)(r0 +4) = 0
> misaligned access off 4 size 8
>
> It will surely change over time as verifier becomes smarter,
> supports new types, optimizations and so on.
> So this log is not an ABI. It's for humans to read.
> The log explains _how_ verifier came to conclusion
> that the program is unsafe.
Given that you've already arranged (I think) for the verifier to be
compilable in the kernel and in userspace, would it make more sense to
have the kernel version just say yes or no and to make it easy for
user code to retry verification in userspace if they want a full
explanation?
--Andy
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
