Re: [patch] fs, proc: unconditional cond_resched when reading smaps

2017-08-21 Thread Kirill A. Shutemov
On Mon, Aug 21, 2017 at 02:06:45PM -0700, David Rientjes wrote: > If there are large numbers of hugepages to iterate while reading > /proc/pid/smaps, the page walk never does cond_resched(). On archs > without split pmd locks, there can be significant and observable > contention on

Re: [patch] fs, proc: unconditional cond_resched when reading smaps

2017-08-21 Thread Kirill A. Shutemov
On Mon, Aug 21, 2017 at 02:06:45PM -0700, David Rientjes wrote: > If there are large numbers of hugepages to iterate while reading > /proc/pid/smaps, the page walk never does cond_resched(). On archs > without split pmd locks, there can be significant and observable > contention on

[patch] fs, proc: unconditional cond_resched when reading smaps

2017-08-21 Thread David Rientjes
If there are large numbers of hugepages to iterate while reading /proc/pid/smaps, the page walk never does cond_resched(). On archs without split pmd locks, there can be significant and observable contention on mm->page_table_lock which cause lengthy delays without rescheduling. Always

[patch] fs, proc: unconditional cond_resched when reading smaps

2017-08-21 Thread David Rientjes
If there are large numbers of hugepages to iterate while reading /proc/pid/smaps, the page walk never does cond_resched(). On archs without split pmd locks, there can be significant and observable contention on mm->page_table_lock which cause lengthy delays without rescheduling. Always