Messages in this thread | | | Date | Tue, 8 Oct 2013 09:15:30 +0200 | Subject | Re: [RFC] perf: mmap2 not covering VM_CLONE regions | From | Stephane Eranian <> |
| |
On Tue, Oct 8, 2013 at 8:54 AM, Peter Zijlstra <peterz@infradead.org> wrote: > On Mon, Oct 07, 2013 at 11:04:38PM +0200, Stephane Eranian wrote: >> Peter, >> >> Ok, so I tried this. It does not cover shmat() cases unfortunately. >> We need that same ++global_mm_id somewhere on that code path. >> Haven't looked at it in details just yet. > > shmat() is a regular shared memory thingy right? So that should be > covered by the regular inode,dev bits, right?
Yes, it is but I am trying to see whether or not we could unify that and use a single u64 number to uniquely identify each mapping.
| |