On Mon, 2007-03-26 at 12:17 -0400, Alan D. Brunelle wrote: > Ming Zhang wrote: > > Hi > > > > I am now using btt to analyze a 700MB trace data and it always run oom > > in my 2GB ram laptop. it also eat 2GB swap as well. Any idea why btt > > need so many memory? > > > > Well, BTT builds trees based upon outstanding IO traces. That could > indicate that BTT is having a hard time dealing with your data - I find > that sometimes the kernel does weird things, causing strange IO trees > being built that BTT can't put back together. I've got some multi-GB > sized binary files that I've handled, let me see what the memory > footprint is for those... > > > > ps, why we always need double in the code, can float fit the bill as > > well? > > > > Probably could, but I'm not too sure about some of the conversions done > with LBAs. I can look into that if need be.
i did a quick check, changing to float will not have much help on that footprint. double is not used in some key data structures. > > Alan > > > Ming > > > > > > - > To unsubscribe from this list: send the line "unsubscribe linux-btrace" in > the body of a message to [EMAIL PROTECTED] > More majordomo info at http://vger.kernel.org/majordomo-info.html - To unsubscribe from this list: send the line "unsubscribe linux-btrace" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
