Re: [PATCH] mm: set khugepaged_max_ptes_none by 1/8 of HPAGE_PMD_NR

2015-03-02 Thread Vlastimil Babka
On 02/27/2015 10:12 PM, David Rientjes wrote: On Fri, 27 Feb 2015, Rik van Riel wrote: I think we do need to change the default. Why? See this bug: The problem was reported here: https://bugzilla.kernel.org/show_bug.cgi?id=93111 Now, there may be a better value than HPAGE_PMD_NR/8, but I a

Re: [PATCH] mm: set khugepaged_max_ptes_none by 1/8 of HPAGE_PMD_NR

2015-02-27 Thread David Rientjes
On Fri, 27 Feb 2015, Rik van Riel wrote: > >> Using THP, programs can access memory faster, by having the > >> kernel collapse small pages into large pages. The parameter > >> max_ptes_none specifies how many extra small pages (that are > >> not already mapped) can be allocated when collapsing a g

Re: [PATCH] mm: set khugepaged_max_ptes_none by 1/8 of HPAGE_PMD_NR

2015-02-27 Thread Rik van Riel
On 02/27/2015 03:53 PM, David Rientjes wrote: > On Fri, 27 Feb 2015, Ebru Akagunduz wrote: > >> Using THP, programs can access memory faster, by having the >> kernel collapse small pages into large pages. The parameter >> max_ptes_none specifies how many extra small pages (that are >> not already

Re: [PATCH] mm: set khugepaged_max_ptes_none by 1/8 of HPAGE_PMD_NR

2015-02-27 Thread David Rientjes
On Fri, 27 Feb 2015, Ebru Akagunduz wrote: > Using THP, programs can access memory faster, by having the > kernel collapse small pages into large pages. The parameter > max_ptes_none specifies how many extra small pages (that are > not already mapped) can be allocated when collapsing a group > of

[PATCH] mm: set khugepaged_max_ptes_none by 1/8 of HPAGE_PMD_NR

2015-02-27 Thread Ebru Akagunduz
Using THP, programs can access memory faster, by having the kernel collapse small pages into large pages. The parameter max_ptes_none specifies how many extra small pages (that are not already mapped) can be allocated when collapsing a group of small pages into one large page. A larger value of ma