On Fri 06-01-17 13:57:59, Paul Cassella wrote: > On Fri, 6 Jan 2017, Michal Hocko wrote: > > On Thu 05-01-17 16:48:03, Mike Kravetz wrote: > > > On 01/05/2017 07:15 AM, Michal Hocko wrote: > > > > > we have a customer report on an older kernel (3.12) but I believe the > > > > same issue is present in the current vanilla kernel. There is a race > > > > between mmap trying to do a reservation which fails when racing with > > > > truncate_hugepages. See the reproduced attached. > > > > > > > > It should go like this (analysis come from the customer and I hope I > > > > haven't screwed their write up). > > Hi Michal, > > There may have been a step missing from what was sent to you, right at the > point Mike asked about. I've added it below. No, I guess the info I got from you was complete. I was trying to reduce the info to the minimum and wasn't explicit about this fact enough assuming it would be clear from the test case which I forgot to add... Sorry about that! -- Michal Hocko SUSE Labs -- 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>