Request for review (#4) 7174978: NPG: Fix bactrace builder for class redefinition

[Coleen Phillimore]

On 1/18/2013 9:26 AM, Vitaly Davidovich wrote:
>
> Isn't signed overflow undefined in C and C++? Not sure if this code 
> should explicitly check if it would overflow (version == max int) 
> after increment to be pedantic.
>

Unfortunately, I already checked it in.   We do this sort of thing in a 
couple of places, so I'll file a bug to fix these also.
Thanks,
Coleen

> Sent from my phone
>
> On Jan 18, 2013 12:49 AM, "serguei.spitsyn at oracle.com 
> <mailto:serguei.spitsyn at oracle.com>" <serguei.spitsyn at oracle.com 
> <mailto:serguei.spitsyn at oracle.com>> wrote:
>
>
>     The fix is good.
>
>     Another way would be to use prefix increment instead of postfix
>     increment:
>       _saved._version = (version >=0) ? ++version : version; // keep
>
>     Nice catch, David.
>
>     Thanks,
>     Serguei
>
>
>     On 1/17/13 6:51 PM, Coleen Phillimore wrote:
>
>         On 1/17/2013 9:15 PM, David Holmes wrote:
>
>                 void increment_and_save_version(int version) {
>                 _saved._version = (version >=0) ? version++ : version;
>                 // keep
>                 overflowed value
>                 }
>
>
>             That saves the pre-incremented value. But otherwise it is
>             a simpler formulation.
>
>             David
>
>
>         I fixed it.  Please review.
>
>         open webrev at http://cr.openjdk.java.net/~coleenp/8006548/
>         <http://cr.openjdk.java.net/%7Ecoleenp/8006548/>
>         bug link at http://bugs.sun.com/view_bug.do?bug_id=8006548
>
>         thanks,
>         Coleen
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/hotspot-runtime-dev/attachments/20130118/6c2410d0/attachment.html