Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-04-15 Thread Minchan Kim
On Sun, Apr 12, 2015 at 11:48:23PM +0900, Minchan Kim wrote: > Hello Hugh, > > On Sat, Apr 11, 2015 at 02:40:46PM -0700, Hugh Dickins wrote: > > On Wed, 11 Mar 2015, Minchan Kim wrote: > > > > > Bascially, MADV_FREE relys on the pte dirty to decide whether > > > it allows VM to discard the page.

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-04-15 Thread Minchan Kim
On Sun, Apr 12, 2015 at 11:48:23PM +0900, Minchan Kim wrote: Hello Hugh, On Sat, Apr 11, 2015 at 02:40:46PM -0700, Hugh Dickins wrote: On Wed, 11 Mar 2015, Minchan Kim wrote: Bascially, MADV_FREE relys on the pte dirty to decide whether it allows VM to discard the page. However, if

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-04-12 Thread Minchan Kim
Hello Hugh, On Sat, Apr 11, 2015 at 02:40:46PM -0700, Hugh Dickins wrote: > On Wed, 11 Mar 2015, Minchan Kim wrote: > > > Bascially, MADV_FREE relys on the pte dirty to decide whether > > it allows VM to discard the page. However, if there is swap-in, > > pte pointed out the page has no

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-04-12 Thread Minchan Kim
Hello Hugh, On Sat, Apr 11, 2015 at 02:40:46PM -0700, Hugh Dickins wrote: On Wed, 11 Mar 2015, Minchan Kim wrote: Bascially, MADV_FREE relys on the pte dirty to decide whether it allows VM to discard the page. However, if there is swap-in, pte pointed out the page has no pte_dirty. So,

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-04-11 Thread Hugh Dickins
On Wed, 11 Mar 2015, Minchan Kim wrote: > Bascially, MADV_FREE relys on the pte dirty to decide whether > it allows VM to discard the page. However, if there is swap-in, > pte pointed out the page has no pte_dirty. So, MADV_FREE checks > PageDirty and PageSwapCache for those pages to not discard

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-04-11 Thread Hugh Dickins
On Wed, 11 Mar 2015, Minchan Kim wrote: Bascially, MADV_FREE relys on the pte dirty to decide whether it allows VM to discard the page. However, if there is swap-in, pte pointed out the page has no pte_dirty. So, MADV_FREE checks PageDirty and PageSwapCache for those pages to not discard it

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-04-09 Thread Andrew Morton
On Thu, 9 Apr 2015 08:50:25 +0900 Minchan Kim wrote: > Bump. I'm getting the feeling that MADV_FREE is out of control. Below is the overall rollup of mm-support-madvisemadv_free.patch mm-support-madvisemadv_free-fix.patch mm-support-madvisemadv_free-fix-2.patch

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-04-09 Thread Andrew Morton
On Thu, 9 Apr 2015 08:50:25 +0900 Minchan Kim minc...@kernel.org wrote: Bump. I'm getting the feeling that MADV_FREE is out of control. Below is the overall rollup of mm-support-madvisemadv_free.patch mm-support-madvisemadv_free-fix.patch mm-support-madvisemadv_free-fix-2.patch

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-04-08 Thread Minchan Kim
Bump. On Wed, Mar 11, 2015 at 10:20:38AM +0900, Minchan Kim wrote: > Bascially, MADV_FREE relys on the pte dirty to decide whether > it allows VM to discard the page. However, if there is swap-in, > pte pointed out the page has no pte_dirty. So, MADV_FREE checks > PageDirty and PageSwapCache for

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-04-08 Thread Minchan Kim
Bump. On Wed, Mar 11, 2015 at 10:20:38AM +0900, Minchan Kim wrote: Bascially, MADV_FREE relys on the pte dirty to decide whether it allows VM to discard the page. However, if there is swap-in, pte pointed out the page has no pte_dirty. So, MADV_FREE checks PageDirty and PageSwapCache for

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-30 Thread Minchan Kim
On Tue, Mar 31, 2015 at 12:14:46AM +0300, Cyrill Gorcunov wrote: > On Mon, Mar 30, 2015 at 05:59:15PM +0900, Minchan Kim wrote: > > Hi Cyrill, > > > > On Mon, Mar 30, 2015 at 11:51:12AM +0300, Cyrill Gorcunov wrote: > > > On Mon, Mar 30, 2015 at 02:22:50PM +0900, Minchan Kim wrote: > > > > 2nd

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-30 Thread Cyrill Gorcunov
On Mon, Mar 30, 2015 at 05:59:15PM +0900, Minchan Kim wrote: > Hi Cyrill, > > On Mon, Mar 30, 2015 at 11:51:12AM +0300, Cyrill Gorcunov wrote: > > On Mon, Mar 30, 2015 at 02:22:50PM +0900, Minchan Kim wrote: > > > 2nd description trial. > > ... > > Hi Minchan, could you please point for which

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-30 Thread Minchan Kim
Hi Cyrill, On Mon, Mar 30, 2015 at 11:51:12AM +0300, Cyrill Gorcunov wrote: > On Mon, Mar 30, 2015 at 02:22:50PM +0900, Minchan Kim wrote: > > 2nd description trial. > ... > Hi Minchan, could you please point for which repo this patch, > linux-next? It was based on

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-30 Thread Cyrill Gorcunov
On Mon, Mar 30, 2015 at 02:22:50PM +0900, Minchan Kim wrote: > 2nd description trial. ... Hi Minchan, could you please point for which repo this patch, linux-next? -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-30 Thread Minchan Kim
Hi Cyrill, On Mon, Mar 30, 2015 at 11:51:12AM +0300, Cyrill Gorcunov wrote: On Mon, Mar 30, 2015 at 02:22:50PM +0900, Minchan Kim wrote: 2nd description trial. ... Hi Minchan, could you please point for which repo this patch, linux-next? It was based on v4.0-rc5-mmotm-2015-03-24-17-02. As

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-30 Thread Cyrill Gorcunov
On Mon, Mar 30, 2015 at 02:22:50PM +0900, Minchan Kim wrote: 2nd description trial. ... Hi Minchan, could you please point for which repo this patch, linux-next? -- To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to majord...@vger.kernel.org More

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-30 Thread Cyrill Gorcunov
On Mon, Mar 30, 2015 at 05:59:15PM +0900, Minchan Kim wrote: Hi Cyrill, On Mon, Mar 30, 2015 at 11:51:12AM +0300, Cyrill Gorcunov wrote: On Mon, Mar 30, 2015 at 02:22:50PM +0900, Minchan Kim wrote: 2nd description trial. ... Hi Minchan, could you please point for which repo this

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-30 Thread Minchan Kim
On Tue, Mar 31, 2015 at 12:14:46AM +0300, Cyrill Gorcunov wrote: On Mon, Mar 30, 2015 at 05:59:15PM +0900, Minchan Kim wrote: Hi Cyrill, On Mon, Mar 30, 2015 at 11:51:12AM +0300, Cyrill Gorcunov wrote: On Mon, Mar 30, 2015 at 02:22:50PM +0900, Minchan Kim wrote: 2nd description

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-29 Thread Minchan Kim
2nd description trial. >From ccfc6c79634f6cec69d8fb23b0e863ebfa5b893c Mon Sep 17 00:00:00 2001 From: Minchan Kim Date: Mon, 30 Mar 2015 13:43:08 +0900 Subject: [PATCH v2] mm: make every pte dirty on do_swap_page Bascially, MADV_FREE relys on the dirty bit in page table entry to decide whether

Re: [PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-29 Thread Minchan Kim
2nd description trial. From ccfc6c79634f6cec69d8fb23b0e863ebfa5b893c Mon Sep 17 00:00:00 2001 From: Minchan Kim minc...@kernel.org Date: Mon, 30 Mar 2015 13:43:08 +0900 Subject: [PATCH v2] mm: make every pte dirty on do_swap_page Bascially, MADV_FREE relys on the dirty bit in page table entry to

[PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-10 Thread Minchan Kim
Bascially, MADV_FREE relys on the pte dirty to decide whether it allows VM to discard the page. However, if there is swap-in, pte pointed out the page has no pte_dirty. So, MADV_FREE checks PageDirty and PageSwapCache for those pages to not discard it because swapped-in page could live on swap

[PATCH 4/4] mm: make every pte dirty on do_swap_page

2015-03-10 Thread Minchan Kim
Bascially, MADV_FREE relys on the pte dirty to decide whether it allows VM to discard the page. However, if there is swap-in, pte pointed out the page has no pte_dirty. So, MADV_FREE checks PageDirty and PageSwapCache for those pages to not discard it because swapped-in page could live on swap