cvs commit: src/lib/libc/stdlib malloc.c

[Christian S.J. Peron]

Jason Evans wrote:

> On Jan 26, 2006, at 6:51 PM, Andrey Chernov wrote:
>
>> I worry about my system performance. Could you please make all  
>> statistics
>> and debug code conditional, debending on some flags which could be  
>> set as
>> malloc options? Even in case this is a test period now, some people
>> perhaps don't want some tests. It is much easy to turn some tests  
>> off via
>> malloc options than to comment out manually corresponding malloc.c  
>> defines
>> after each cvsup.
>
>
> Unfortunately, run-time checks for the debugging features are  
> expensive, since the debugging code is scattered throughout  
> malloc.c.  This pretty much mandates compile-time configuration, for  
> performance reasons.
>
> I can disable statistics and debugging, if there's a general  
> consensus to do so.  How much performance difference is the debug/ 
> stats code making for you?  I've only been disabling debug/stats for  
> benchmarking purposes, so I don't have a good feel for how much it  
> impacts overall system performance.


These malloc debugging/stats features have been invaluable. I vote to 
accept the performance hit in exchange for useful debugging/stats 
features in -CURRENT, and disable them in -STABLE/-RELEASE when they are 
ready.

-- 
Christian S.J. Peron
csjp at FreeBSD.ORG
FreeBSD Committer
FreeBSD Security Team