Messages in this thread | | | Date | Tue, 24 Jan 2012 17:45:28 -0800 | From | Arun Sharma <> | Subject | Re: [PATCH] mm: Enable MAP_UNINITIALIZED for archs with mmu |
| |
On 1/23/12 7:07 PM, KAMEZAWA Hiroyuki wrote:
> You can see reduction of clear_page() cost by removing GFP_ZERO but > what's your application's total performance ? Is it good enough considering > many risks ?
I see 90k calls/sec to clear_page_c when running our application. I don't have data on the impact of GFP_ZERO alone, but an earlier experiment when we tuned malloc to not call madvise(MADV_DONTNEED) aggressively saved us 3% CPU. So I'm expecting this to be a 1-2% win.
But not calling madvise() increases our RSS and increases the risk of OOM.
Agree with your analysis that removing the cache misses at clear_page() is not always a win, since it moves the misses to the code where the app first touches the data.
-Arun
| |