Hi,,, On Mon, May 19, 2008 at 1:42 PM, Manoj Kumar Santubhai Jain <ManojKumarS_Jain@xxxxxxxxxxx> wrote: > > Hi, > > When the file /proc/slabinfo does get updated? it just reads current condition of slab... so whenever something new is added or something is released from slab, at the earliest moment you read /proc/slabinfo...you shall see the difference there. > I started new process and I was thinking mm_struct active objects should be increased but it stays constant. Big chance they are sharing mm_struct...or at least since under shared page(s). Or....somehow the newly created process/thread used just released mm_struct > > Output Before staring process: > #cat /proc/slabinfo | grep mm_struct > mm_struct 52 64 512 8 1 : tunables 0 0 0 : slabdata 8 8 0 > # ps -fu root | wc > 70 615 4645 > > Started process in another shell (shell was already open) /home/manoj/hello > > Output after Starting process > # cat /proc/slabinfo | grep mm_struct > mm_struct 52 64 512 8 1 : tunables 0 0 0 : slabdata 8 8 0 > # ps -fu root | wc > 71 623 4711 > > Is it possible that parent(shell) and child (hello) using same mm_struct ? Yeah, there is a chance that happens.... > I also tried running /home/manoj/hello&, but the result is same. > > Is there something that I need to enable in my config file? Or there is a possible bug with update on /proc/slabinfo? Bug? I don't think so...however you must rememeber that slab condition is highly volatile...thus IMO even if you read it per 1 milisecond, there is a slight chance you miss the whole slab addition/reduction history. CMIIW people... regards, Mulyadi. -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
