On 2/5/19 6:36 PM, Alexei Starovoitov wrote:
> On Tue, Feb 05, 2019 at 02:28:44PM -0800, Yonghong Song wrote:
>> Commit 9c651127445c ("selftests/btf: add initial BTF dedup tests")
>> added dedup tests in test_btf.c.
>> It broke the raw test:
>> BTF raw test[71] (func proto (Bad arg name_off)):
>> btf_raw_create:2905:FAIL Error getting string #65535, strs_cnt:1
>>
>> The test itself encodes invalid func_proto parameter name
>> offset 0xffffFFFF as a negative test for the kernel.
>> The above commit changed the meaning of that offset and
>> resulted in a user space error.
>> #define NAME_NTH(N) (0xffff0000 | N)
>> #define IS_NAME_NTH(X) ((X & 0xffff0000) == 0xffff0000)
>> #define GET_NAME_NTH_IDX(X) (X & 0x0000ffff)
>>
>> Currently, the kernel permits maximum name offset 0xffff.
>> Set the test name off as 0x0fffFFFF to trigger the kernel
>> verification failure.
>>
>> Cc: Andrii Nakryiko <andr...@fb.com>
>> Fixes: 9c651127445c ("selftests/btf: add initial BTF dedup tests")
>> Signed-off-by: Yonghong Song <y...@fb.com>
>
> Applied, Thanks
>
> Also I see the following new error in test_btf:
> BTF libbpf test[2] (test_btf_nokv.o): libbpf: map:btf_map
> container_name:____btf_map_btf_map cannot be found in BTF. Missing
> BPF_ANNOTATE_KV_PAIR?
> OK
>
> A bunch of similar errors are in test_progs as well.
> I suspect it's related to the last few btf changes.
> Andrii, Yonghong, Please investigate.
It is due to one of previous patches to refector btf__get_map_kv_tids(),
accidentally using pr_warning instead of original pr_debug.
Will submit a patch soon.
>
