RFR(M): 8020775: PPC64 (part 12): posix signal printing

Vladimir Kozlov vladimir.kozlov at oracle.com
Tue Aug 20 14:39:29 PDT 2013


Since we have consensus I'm pushing these changes into ppc64/stage repo 
using JPRT.

Thanks,
Vladimir

On 8/19/13 8:17 AM, Vladimir Kozlov wrote:
> On 8/19/13 12:23 AM, Lindenmaier, Goetz wrote:
>> Hi Vladimir,
>>
>> I updated the webrev:
>> http://cr.openjdk.java.net/~goetz/webrevs/8020775-print_sig-2/
>>
>>> const char* os::Posix::describe_signal_set_short(const sigset_t* set,
>>> char* buffer, size_t buf_size) {
>> I used your code.
>
> Thanks.
>
>>
>>> The code in the loop in describe_sa_flags() is incorrect because
>>> strlen(p) gives size of all stored sa names. It should be
>>> strlen(flaginfo[idx].s). I also don't see why it is while() loop and not
>>> simple for().
>> No, the code is correct.  p points to the beginning of the recently
>
> Yes, you are right.
>
>> printed string.  Taking the size of p includes the '|' printed.
>> I changed it to a for loop.
>>
>>  From your other mail:
>>> It would be nice if you can produce the same alignment in output:
>>>
>>> SIGSEGV: [libjvm.so+0xc03f0d], sa_mask[0]=1,
>>> sa_flags=SA_RESTART|SA_SIGINFO
>>
>> The alignment of the signals would have to be done in
>> print_signal_handler(), which I did not change yet.
>> But this also looks like a candidate for moving to
>> the posix file.
>
> It is not big deal, so we can leave it for now. I looked on
> print_signal_handler() and it would require several #ifdef in common code.
>
> Thanks,
> Vladimir
>
>>
>> Best regards,
>>    Goetz.
>>
>>


More information about the ppc-aix-port-dev mailing list