RFR (S): 8012260: ciReplay: Include PID into the name of replay data file

[Christian Thalinger]

On Apr 25, 2013, at 7:50 AM, Vladimir Ivanov <vladimir.x.ivanov at oracle.com> wrote:

> Chris,
> 
> You were right. Here's updated version:
> http://cr.openjdk.java.net/~vlivanov/8012260/webrev.03/

Much better.  Thank you.  Perhaps you also want to change your method comments to Doxygen-style ones?

-- Chris

> 
> Best regards,
> Vladimir Ivanov
> 
> On 4/20/13 8:03 AM, Christian Thalinger wrote:
>> 
>> On Apr 19, 2013, at 3:17 PM, Vladimir Ivanov <vladimir.x.ivanov at oracle.com> wrote:
>> 
>>> Thank you for review, Chris!
>>> 
>>>> src/os/windows/vm/os_windows.cpp:
>>>> 
>>>> +   if (buflen > INT_MAX)  n = INT_MAX;
>>>> 
>>>> Why do we have this check on Windows?
>>> Since _getcwd requires int on Windows, this is a consequence of size_t -> int conversion.
>> 
>> Got it.
>> 
>>> 
>>>> src/share/vm/utilities/ostream.hpp:
>>>> 
>>>> !   fileStream(FILE* file, bool need_close=false) { _file = file; _need_close = need_close; }
>>>> 
>>>> Please put spaces before and after "=".
>>> Sure, will do.
>>> 
>>>> +        if (pos > 0 &&
>>>> +            Arguments::copy_expand_pid(default_pattern, strlen(default_pattern),
>>>> +                                       &buf[pos], buflen-pos)) {
>>>> +          fd = open(buf, O_RDWR | O_CREAT | O_TRUNC, 0666);
>>>> +        }
>>>> 
>>>> Could you factor this code into a method?  It's duplicated three times.
>>> Yeah, I don't like it either :-) But I don't think that a function call will help here. Maybe factoring parameters may help, but I'm not sure. What do you prefer?
>> 
>> Putting Arguments::copy_expand_pid and open calls into a method sounds like a win to me.
>> 
>> -- Chris
>> 
>>> 
>>> Best regards,
>>> Vladimir Ivanov
>>> 
>>>> Otherwise this looks good.
>>>> 
>>>> -- Chris
>>>> 
>>>> On Apr 19, 2013, at 10:03 AM, Vladimir Ivanov <vladimir.x.ivanov at oracle.com> wrote:
>>>> 
>>>>> I need one more "official" review. Any volunteers? :-)
>>>>> 
>>>>> Thanks!
>>>>> 
>>>>> Best regards,
>>>>> Vladimir Ivanov
>>>>> 
>>>>> On 4/19/13 7:22 PM, Vladimir Kozlov wrote:
>>>>>> Looks good.
>>>>>> 
>>>>>> Thanks,
>>>>>> Vladimir
>>>>>> 
>>>>>> On 4/19/13 7:17 AM, Vladimir Ivanov wrote:
>>>>>>> Updated fix:
>>>>>>> http://cr.openjdk.java.net/~vlivanov/8012260/webrev.02/
>>>>>>> 
>>>>>>> Did some additional cleanup: changed get_current_directory to accept
>>>>>>> size_t instead of int as a buffer size.
>>>>>>> 
>>>>>>> Best regards,
>>>>>>> Vladimir Ivanov
>>>>>>> 
>>>>>>> On 4/17/13 8:30 PM, Vladimir Kozlov wrote:
>>>>>>>> On 4/17/13 9:03 AM, Vladimir Ivanov wrote:
>>>>>>>>> I considered it redundant since fdopen returns NULL when a call fails
>>>>>>>>> [1] and I already check replay_data_file. Was my reasoning wrong?
>>>>>>>> 
>>>>>>>> My understanding was that you can't pass fd == -1 to fdopen() and I
>>>>>>>> can't find any docs about such case so you may be right. And I am more
>>>>>>>> concern about Windows/Cygwin implementation of fdopen.
>>>>>>>> 
>>>>>>>> Vladimir
>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> I checked on MacOS and when (fd == -1) it produces the following error
>>>>>>>>> message:
>>>>>>>>> # Can't open file to dump replay data. Error: Bad file descriptor
>>>>>>>>> 
>>>>>>>>> Best regards,
>>>>>>>>> Vladimir Ivanov
>>>>>>>>> 
>>>>>>>>> [1] Excerpt from fdopen man page:
>>>>>>>>> RETURN VALUES
>>>>>>>>>      Upon successful completion fopen(), fdopen(), and freopen()
>>>>>>>>> return
>>>>>>>>> a FILE pointer.  Otherwise, NULL is returned and the global variable
>>>>>>>>> errno is set to indicate the error.
>>>>>>>>> 
>>>>>>>>> On 4/17/13 7:54 PM, Vladimir Kozlov wrote:
>>>>>>>>>> Missing check (fd != -1) before fdopen:
>>>>>>>>>> 
>>>>>>>>>> 1036       FILE* replay_data_file = fdopen(fd, "w");
>>>>>>>>>> 
>>>>>>>>>> Vladimir
>>>>>>>>>> 
>>>>>>>>>> On 4/17/13 8:19 AM, Vladimir Ivanov wrote:
>>>>>>>>>>> Thanks for feedback, Vladimir!
>>>>>>>>>>> 
>>>>>>>>>>> Here's updated version:
>>>>>>>>>>> http://cr.openjdk.java.net/~vlivanov/8012260/webrev.01
>>>>>>>>>>> 
>>>>>>>>>>> Best regards,
>>>>>>>>>>> Vladimir Ivanov
>>>>>>>>>>> 
>>>>>>>>>>> On 4/17/13 9:22 AM, Vladimir Kozlov wrote:
>>>>>>>>>>>> Make it product flag since it is ON by default (and it is not
>>>>>>>>>>>> diagnostic
>>>>>>>>>>>> but error dump).
>>>>>>>>>>>> 
>>>>>>>>>>>> Also why not reuse code which opens ErrorFile (hs_err*)?
>>>>>>>>>>>> 
>>>>>>>>>>>> Compilation::compile_only_this_method() allocates fileStream on
>>>>>>>>>>>> stack:
>>>>>>>>>>>> 
>>>>>>>>>>>>   fileStream stream(fopen("c1_compile_only", "wt"));
>>>>>>>>>>>> 
>>>>>>>>>>>> thanks,
>>>>>>>>>>>> Vladimir
>>>>>>>>>>>> 
>>>>>>>>>>>> On 4/16/13 5:01 PM, Vladimir Ivanov wrote:
>>>>>>>>>>>>> http://cr.openjdk.java.net/~vlivanov/8012260/webrev.00
>>>>>>>>>>>>> 43 lines changed: 16 ins; 21 del; 6 mod;
>>>>>>>>>>>>> 
>>>>>>>>>>>>> Changed naming scheme for compilation replay data files from
>>>>>>>>>>>>> 'replay.txt' to 'replay_pid%p.log', where %p is substituted w/
>>>>>>>>>>>>> pid. It
>>>>>>>>>>>>> allows to easily correspond replay file with hs_err file and makes
>>>>>>>>>>>>> overwriting of existing files very unlikely.
>>>>>>>>>>>>> 
>>>>>>>>>>>>> Also, converted ReplayDataFile & DumpReplayDataOnError flags from
>>>>>>>>>>>>> develop to diagnostic, since they are useful in product binaries.
>>>>>>>>>>>>> Did
>>>>>>>>>>>>> some small cleanups.
>>>>>>>>>>>>> 
>>>>>>>>>>>>> One question still bothers me. There's a memory allocation in
>>>>>>>>>>>>> VMError::report_and_die() I don't know how to avoid:
>>>>>>>>>>>>> +         fileStream* replay_data_stream = new
>>>>>>>>>>>>> (ResourceObj::C_HEAP,
>>>>>>>>>>>>> mtCompiler) fileStream(replay_data_file);
>>>>>>>>>>>>> 
>>>>>>>>>>>>> It's not new, I just moved it up from auxiliary method. Any ideas
>>>>>>>>>>>>> how to
>>>>>>>>>>>>> make VMError::report_and_die() allocation-free again? :-)
>>>>>>>>>>>>> 
>>>>>>>>>>>>> Best regards,
>>>>>>>>>>>>> Vladimir Ivanov
>>>>>>>>>>>>> 
>>>> 
>>