Re: [PATCH v4 perf,bpf 07/15] perf, bpf: save bpf_prog_info information as headers to perf.data

[Song Liu]

> On Feb 27, 2019, at 5:21 AM, Jiri Olsa <jo...@redhat.com> wrote:
> 
> On Mon, Feb 25, 2019 at 04:20:11PM -0800, Song Liu wrote:
> 
> SNIP
> 
>> +            if (info_len > sizeof(struct bpf_prog_info)) {
>> +                    pr_warning("detected invalid bpf_prog_info\n");
>> +                    goto out;
>> +            }
>> +
>> +            info_linear = malloc(sizeof(struct bpf_prog_info_linear) +
>> +                                 data_len);
>> +            if (!info_linear)
>> +                    goto out;
>> +            info_linear->info_len = sizeof(struct bpf_prog_info);
>> +            info_linear->data_len = data_len;
>> +            if (do_read_u64(ff, (u64 *)(&info_linear->arrays)))
>> +                    goto out;
>> +            if (__do_read(ff, &info_linear->info, info_len))
>> +                    goto out;
>> +            if (info_len < sizeof(struct bpf_prog_info))
>> +                    memset(((void *)(&info_linear->info)) + info_len, 0,
>> +                           sizeof(struct bpf_prog_info) - info_len);
>> +
>> +            if (__do_read(ff, info_linear->data, data_len))
>> +                    goto out;
>> +
>> +            /* endian mismatch, drop the info, continue */
> 
> so there's no way we can swap the data? why?
> 
> jirka

Yes, we can swap the data. OTOH, the set is big enough. How about we 
implement the swap later? Running perf-report on a system of with 
different endianness is a relatively rare case. 

Thanks,
Song



>> +            if (ff->ph->needs_swap) {
>> +                    free(info_linear);
>> +                    continue;
>> +            }
> 
> SNIP