On Wed, Apr 19, 2017 at 03:33:39PM +0200, Michal Hocko wrote: > On Wed 19-04-17 15:22:16, Stanislaw Gruszka wrote: > > On Wed, Apr 19, 2017 at 01:13:43PM +0200, Michal Hocko wrote: > > > On Tue 18-04-17 14:47:32, David Rientjes wrote: > > > [...] > > > > I think the debug_guardpage_minorder() check makes sense for failed > > > > allocations because we are essentially removing memory from the system for > > > > debug, failed allocations as a result of low on memory or fragmentation > > > > aren't concerning if we are removing memory from the system. > > > > > > I really fail to see how this is any different from booting with > > > mem=$SIZE to reduce the amount of available memory. > > > > mem= shrink upper memory limit, debug_guardpage_minorder= fragments > > available physical memory (deliberately to catch unintended access). > > Yeah but both make allocation failures (especially higher order ones) > more likely. So I really fail to see the point inhibit allocation > failure warning for one and not for the other. There is difference for buddy allocator. If you limit to 1/2 of memory such only upper half is not available, buddy allocator can easily find pages for higher order allocations in lower half of memory. When you limit to 1/2 of memory such every second page is not available, buddy allocator can not make successful any order 1 or higher allocations. > This whole special casing > of debug_guardpage_minorder is just too strange to me. We do have a rate > limit to not flood the log. I added this check to skip warning if buddy allocator fail, what I considered likely scenario taking the conditions. The check remove warning completely, rate limit only limit the speed warnings shows in logs. Stanislaw -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
