quickfix-developers

[Quickfix-developers] process_sleep has unitialized pointer

[Clark S. <cla...@ya...>]

The pointer timespeck hasn't been initialized. Is this perhaps the cause of random wierdness?

 

void process_sleep( double s )
{ QF_STACK_PUSH(process_sleep)

#ifdef _MSC_VER
    Sleep( ( long ) s * 1000 );
#else
/* this causes random wierdness under Solaris
  timespec time;
  double intpart;
  time.tv_nsec = (int)modf(s, &intpart);
  time.tv_nsec *= 1000000000;
  time.tv_sec = (int)intpart;
  timespec* actual;
  nanosleep(&time, actual);*/
 sleep( ( long ) s );
#endif

  QF_STACK_POP
}


 

I think this would work better:

void process_sleep( double s )
{ QF_STACK_PUSH(process_sleep)

#ifdef _MSC_VER
    Sleep( ( long ) s * 1000 );
#else
  timespec time, actual;
  double intpart;
  time.tv_nsec = (int)modf(s, &intpart);
  time.tv_nsec *= 1000000000;
  time.tv_sec = (int)intpart;
   nanosleep(&time, &actual);
 sleep( ( long ) s );
#endif

  QF_STACK_POP
}





		
---------------------------------
Do you Yahoo!?
Yahoo! Mail is new and improved - Check it out!

Re: [Quickfix-developers] process_sleep has unitialized pointer

[Oren M. <or...@qu...>]

Yeah, this makes sense.

--oren

On Jun 21, 2004, at 7:25 PM, Clark Sims wrote:

> The pointer timespeck hasn't been initialized. Is this perhaps the=20
> cause of random wierdness?
>
> =A0
>
> void process_sleep( double s )
> { QF_STACK_PUSH(process_sleep)
>
> #ifdef _MSC_VER
> =A0=A0=A0 Sleep( ( long ) s * 1000 );
> #else
> /* this causes random wierdness under Solaris
> =A0 timespec time;
> =A0 double intpart;
> =A0 time.tv_nsec =3D (int)modf(s, &intpart);
> =A0 time.tv_nsec *=3D 1000000000;
> =A0 time.tv_sec =3D (int)intpart;
> =A0 timespec* actual;
> =A0 nanosleep(&time, actual);*/
> =A0sleep( ( long ) s );
> #endif
>
> =A0 QF_STACK_POP
> }
>
> =A0
>
> I think this would work better:
>
>
> void process_sleep( double s )
> { QF_STACK_PUSH(process_sleep)
>
> #ifdef _MSC_VER
> =A0=A0=A0 Sleep( ( long ) s * 1000 );
> #else
> =A0 timespec time, actual;
> =A0 double intpart;
> =A0 time.tv_nsec =3D (int)modf(s, &intpart);
> =A0 time.tv_nsec *=3D 1000000000;
> =A0 time.tv_sec =3D (int)intpart;
> =A0=A0 nanosleep(&time, &actual);
> =A0sleep( ( long ) s );
> #endif
>
> =A0 QF_STACK_POP
> }
>
>
> Do you Yahoo!?
>  Yahoo! Mail is new and improved - Check it out!=

Re: [Quickfix-developers] process_sleep has unitialized pointer

[Oren M. <or...@qu...>]

Actually, if we aren't making use of that parameter, we can just use=20
NULL.  The only time that the structure is written out if a signal has=20=

been delivered to the process, thus interrupting the call and allowing=20=

us to continue with the remainder.  I don't think this is a situation=20
we need to really concern ourselves with, and it certainly isn't=20
handling this case right now.  If the structure is set to NULL, the=20
system call will not try to write to the structure.  So perhaps we=20
should just change the method to the following.

void process_sleep( double s )
{ QF_STACK_PUSH(process_sleep)

#ifdef _MSC_VER
   Sleep( ( long ) s * 1000 );
#else
   timespec time;
   double intpart;
   time.tv_nsec =3D (int)modf(s, &intpart);
   time.tv_nsec *=3D 1000000000;
   time.tv_sec =3D (int)intpart;
   nanosleep(&time, NULL);
#endif

   QF_STACK_POP
}

On Jun 21, 2004, at 7:25 PM, Clark Sims wrote:

> The pointer timespeck hasn't been initialized. Is this perhaps the=20
> cause of random wierdness?
>
> =A0
>
> void process_sleep( double s )
> { QF_STACK_PUSH(process_sleep)
>
> #ifdef _MSC_VER
> =A0=A0=A0 Sleep( ( long ) s * 1000 );
> #else
> /* this causes random wierdness under Solaris
> =A0 timespec time;
> =A0 double intpart;
> =A0 time.tv_nsec =3D (int)modf(s, &intpart);
> =A0 time.tv_nsec *=3D 1000000000;
> =A0 time.tv_sec =3D (int)intpart;
> =A0 timespec* actual;
> =A0 nanosleep(&time, actual);*/
> =A0sleep( ( long ) s );
> #endif
>
> =A0 QF_STACK_POP
> }
>
> =A0
>
> I think this would work better:
>
>
> void process_sleep( double s )
> { QF_STACK_PUSH(process_sleep)
>
> #ifdef _MSC_VER
> =A0=A0=A0 Sleep( ( long ) s * 1000 );
> #else
> =A0 timespec time, actual;
> =A0 double intpart;
> =A0 time.tv_nsec =3D (int)modf(s, &intpart);
> =A0 time.tv_nsec *=3D 1000000000;
> =A0 time.tv_sec =3D (int)intpart;
> =A0=A0 nanosleep(&time, &actual);
> =A0sleep( ( long ) s );
> #endif
>
> =A0 QF_STACK_POP
> }
>
>
> Do you Yahoo!?
>  Yahoo! Mail is new and improved - Check it out!=

Re: [Quickfix-developers] process_sleep has unitialized pointer

[Joerg T. <Joe...@ma...>]

Oren Miller wrote:
> Actually, if we aren't making use of that parameter, we can just use 
> NULL. The only time that the structure is written out if a signal has 
> been delivered to the process, thus interrupting the call and allowing 
> us to continue with the remainder. I don't think this is a situation we 
> need to really concern ourselves with, and it certainly isn't handling 
> this case right now.

I would suggest to do it right now. nanosleep() is the POSIX compliant 
way to wait and it is designed to make restarts after a signal easy:

	while( -1 == nanosleep( &time_to_sleep, &remainder ) ) {
		time_to_sleep = remainder;
         }

> If the structure is set to NULL, the system call 
> will not try to write to the structure. So perhaps we should just change 
> the method to the following.
> 
> void process_sleep( double s )
> { QF_STACK_PUSH(process_sleep)
> 
> #ifdef _MSC_VER
> Sleep( ( long ) s * 1000 );
> #else
> timespec time;

	timespec time, remainder;

> double intpart;
> time.tv_nsec = (int)modf(s, &intpart);
> time.tv_nsec *= 1000000000;

Am I stupid? If I cast the fractional part returned from modf() to int, 
wouldn't it get 0? I would expect something like:

	time.tv_nsec = (long) ((modf(s, &intpart) * 1e6);

> time.tv_sec = (int)intpart;

	time.tv_sec = (time_t) intpart;

> nanosleep(&time, NULL);

	while ( -1 == nanosleep( &time, &remainder ) ) {
	    time = remainder;
	}

> #endif
> 
> QF_STACK_POP
> }

-- 
Joerg Thoennes
                                http://macd.com
Tel.: +49 (0)241 44597-24      Macdonald Associates GmbH
Fax : +49 (0)241 44597-10      Lothringer Str. 52, D-52070 Aachen

Re: [Quickfix-developers] process_sleep has unitialized pointer

[Oren M. <or...@qu...>]

Yeah, but I think it should be, time.tv_nsec = (long) ((modf(s, 
&intpart) * 1e9);, as a nanosecond is a billionth of a second.

--oren

On Jun 22, 2004, at 3:33 AM, Joerg Thoennes wrote:

> Oren Miller wrote:
>> Actually, if we aren't making use of that parameter, we can just use 
>> NULL. The only time that the structure is written out if a signal has 
>> been delivered to the process, thus interrupting the call and 
>> allowing us to continue with the remainder. I don't think this is a 
>> situation we need to really concern ourselves with, and it certainly 
>> isn't handling this case right now.
>
> I would suggest to do it right now. nanosleep() is the POSIX compliant 
> way to wait and it is designed to make restarts after a signal easy:
>
> 	while( -1 == nanosleep( &time_to_sleep, &remainder ) ) {
> 		time_to_sleep = remainder;
>         }
>
>> If the structure is set to NULL, the system call will not try to 
>> write to the structure. So perhaps we should just change the method 
>> to the following.
>> void process_sleep( double s )
>> { QF_STACK_PUSH(process_sleep)
>> #ifdef _MSC_VER
>> Sleep( ( long ) s * 1000 );
>> #else
>> timespec time;
>
> 	timespec time, remainder;
>
>> double intpart;
>> time.tv_nsec = (int)modf(s, &intpart);
>> time.tv_nsec *= 1000000000;
>
> Am I stupid? If I cast the fractional part returned from modf() to 
> int, wouldn't it get 0? I would expect something like:
>
> 	time.tv_nsec = (long) ((modf(s, &intpart) * 1e6);
>
>> time.tv_sec = (int)intpart;
>
> 	time.tv_sec = (time_t) intpart;
>
>> nanosleep(&time, NULL);
>
> 	while ( -1 == nanosleep( &time, &remainder ) ) {
> 	    time = remainder;
> 	}
>
>> #endif
>> QF_STACK_POP
>> }
>
> -- 
> Joerg Thoennes
>                                http://macd.com
> Tel.: +49 (0)241 44597-24      Macdonald Associates GmbH
> Fax : +49 (0)241 44597-10      Lothringer Str. 52, D-52070 Aachen
>

Re: [Quickfix-developers] process_sleep has unitialized pointer

[Joerg T. <Joe...@ma...>]

Oren Miller wrote:
> Yeah, but I think it should be, time.tv_nsec = (long) ((modf(s, 
> &intpart) * 1e9);, as a nanosecond is a billionth of a second.

OK, stupid mistake :-(

Will you do the checkin?

Thanks, Jörg

> --oren
> 
> On Jun 22, 2004, at 3:33 AM, Joerg Thoennes wrote:
> 
>> Oren Miller wrote:
>>
>>> Actually, if we aren't making use of that parameter, we can just use 
>>> NULL. The only time that the structure is written out if a signal has 
>>> been delivered to the process, thus interrupting the call and 
>>> allowing us to continue with the remainder. I don't think this is a 
>>> situation we need to really concern ourselves with, and it certainly 
>>> isn't handling this case right now.
>>
>>
>> I would suggest to do it right now. nanosleep() is the POSIX compliant 
>> way to wait and it is designed to make restarts after a signal easy:
>>
>>     while( -1 == nanosleep( &time_to_sleep, &remainder ) ) {
>>         time_to_sleep = remainder;
>>         }
>>
>>> If the structure is set to NULL, the system call will not try to 
>>> write to the structure. So perhaps we should just change the method 
>>> to the following.
>>> void process_sleep( double s )
>>> { QF_STACK_PUSH(process_sleep)
>>> #ifdef _MSC_VER
>>> Sleep( ( long ) s * 1000 );
>>> #else
>>> timespec time;
>>
>>
>>     timespec time, remainder;
>>
>>> double intpart;
>>> time.tv_nsec = (int)modf(s, &intpart);
>>> time.tv_nsec *= 1000000000;
>>
>>
>> Am I stupid? If I cast the fractional part returned from modf() to 
>> int, wouldn't it get 0? I would expect something like:
>>
>>     time.tv_nsec = (long) ((modf(s, &intpart) * 1e6);
>>
>>> time.tv_sec = (int)intpart;
>>
>>
>>     time.tv_sec = (time_t) intpart;
>>
>>> nanosleep(&time, NULL);
>>
>>
>>     while ( -1 == nanosleep( &time, &remainder ) ) {
>>         time = remainder;
>>     }
>>
>>> #endif
>>> QF_STACK_POP
>>> }
>>
>>
>> -- 
>> Joerg Thoennes
>>                                http://macd.com
>> Tel.: +49 (0)241 44597-24      Macdonald Associates GmbH
>> Fax : +49 (0)241 44597-10      Lothringer Str. 52, D-52070 Aachen
>>
> 
> 


-- 
Joerg Thoennes
                                http://macd.com
Tel.: +49 (0)241 44597-24      Macdonald Associates GmbH
Fax : +49 (0)241 44597-10      Lothringer Str. 52, D-52070 Aachen