* [PATCH 0/3] Readahead fixes
@ 2011-07-09 19:41 ` Raghavendra D Prabhu
0 siblings, 0 replies; 21+ messages in thread
From: Raghavendra D Prabhu @ 2011-07-09 19:41 UTC (permalink / raw)
To: fengguang.wu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
Minor readahead fixes, details in respective patch mails
Raghavendra D Prabhu (3):
Change the check for PageReadahead into an else-if
Remove file_ra_state from arguments of count_history_pages.
Move the check for ra_pages after VM_SequentialReadHint()
mm/filemap.c | 8 ++++----
mm/readahead.c | 3 +--
2 files changed, 5 insertions(+), 6 deletions(-)
--
1.7.6
^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH 0/3] Readahead fixes
@ 2011-07-09 19:41 ` Raghavendra D Prabhu
0 siblings, 0 replies; 21+ messages in thread
From: Raghavendra D Prabhu @ 2011-07-09 19:41 UTC (permalink / raw)
To: fengguang.wu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
Minor readahead fixes, details in respective patch mails
Raghavendra D Prabhu (3):
Change the check for PageReadahead into an else-if
Remove file_ra_state from arguments of count_history_pages.
Move the check for ra_pages after VM_SequentialReadHint()
mm/filemap.c | 8 ++++----
mm/readahead.c | 3 +--
2 files changed, 5 insertions(+), 6 deletions(-)
--
1.7.6
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH 1/3] mm/readahead: Change the check for PageReadahead into an else-if
2011-07-09 19:41 ` Raghavendra D Prabhu
@ 2011-07-09 19:41 ` Raghavendra D Prabhu
-1 siblings, 0 replies; 21+ messages in thread
From: Raghavendra D Prabhu @ 2011-07-09 19:41 UTC (permalink / raw)
To: fengguang.wu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
>From 51daa88ebd8e0d437289f589af29d4b39379ea76, page_sync_readahead coalesces
async readahead into its readahead window, so another checking for that again is
not required.
Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
---
mm/filemap.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/mm/filemap.c b/mm/filemap.c
index a8251a8..074c23d 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1115,8 +1115,7 @@ find_page:
page = find_get_page(mapping, index);
if (unlikely(page == NULL))
goto no_cached_page;
- }
- if (PageReadahead(page)) {
+ } else if (PageReadahead(page)) {
page_cache_async_readahead(mapping,
ra, filp, page,
index, last_index - index);
--
1.7.6
^ permalink raw reply related [flat|nested] 21+ messages in thread
* [PATCH 1/3] mm/readahead: Change the check for PageReadahead into an else-if
@ 2011-07-09 19:41 ` Raghavendra D Prabhu
0 siblings, 0 replies; 21+ messages in thread
From: Raghavendra D Prabhu @ 2011-07-09 19:41 UTC (permalink / raw)
To: fengguang.wu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
>From 51daa88ebd8e0d437289f589af29d4b39379ea76, page_sync_readahead coalesces
async readahead into its readahead window, so another checking for that again is
not required.
Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
---
mm/filemap.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/mm/filemap.c b/mm/filemap.c
index a8251a8..074c23d 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1115,8 +1115,7 @@ find_page:
page = find_get_page(mapping, index);
if (unlikely(page == NULL))
goto no_cached_page;
- }
- if (PageReadahead(page)) {
+ } else if (PageReadahead(page)) {
page_cache_async_readahead(mapping,
ra, filp, page,
index, last_index - index);
--
1.7.6
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related [flat|nested] 21+ messages in thread
* [PATCH 2/3] mm/readahead: Remove file_ra_state from arguments of count_history_pages.
2011-07-09 19:41 ` Raghavendra D Prabhu
@ 2011-07-09 19:41 ` Raghavendra D Prabhu
-1 siblings, 0 replies; 21+ messages in thread
From: Raghavendra D Prabhu @ 2011-07-09 19:41 UTC (permalink / raw)
To: fengguang.wu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
count_history_pages doesn't require readahead state to calculate the offset from history.
Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
---
mm/readahead.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/mm/readahead.c b/mm/readahead.c
index 867f9dd..925d3b3 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -343,7 +343,6 @@ static unsigned long get_next_ra_size(struct file_ra_state *ra,
* - thrashing threshold in memory tight systems
*/
static pgoff_t count_history_pages(struct address_space *mapping,
- struct file_ra_state *ra,
pgoff_t offset, unsigned long max)
{
pgoff_t head;
@@ -366,7 +365,7 @@ static int try_context_readahead(struct address_space *mapping,
{
pgoff_t size;
- size = count_history_pages(mapping, ra, offset, max);
+ size = count_history_pages(mapping, offset, max);
/*
* no history pages:
--
1.7.6
^ permalink raw reply related [flat|nested] 21+ messages in thread
* [PATCH 2/3] mm/readahead: Remove file_ra_state from arguments of count_history_pages.
@ 2011-07-09 19:41 ` Raghavendra D Prabhu
0 siblings, 0 replies; 21+ messages in thread
From: Raghavendra D Prabhu @ 2011-07-09 19:41 UTC (permalink / raw)
To: fengguang.wu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
count_history_pages doesn't require readahead state to calculate the offset from history.
Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
---
mm/readahead.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/mm/readahead.c b/mm/readahead.c
index 867f9dd..925d3b3 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -343,7 +343,6 @@ static unsigned long get_next_ra_size(struct file_ra_state *ra,
* - thrashing threshold in memory tight systems
*/
static pgoff_t count_history_pages(struct address_space *mapping,
- struct file_ra_state *ra,
pgoff_t offset, unsigned long max)
{
pgoff_t head;
@@ -366,7 +365,7 @@ static int try_context_readahead(struct address_space *mapping,
{
pgoff_t size;
- size = count_history_pages(mapping, ra, offset, max);
+ size = count_history_pages(mapping, offset, max);
/*
* no history pages:
--
1.7.6
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related [flat|nested] 21+ messages in thread
* [PATCH 3/3] mm/readahead: Move the check for ra_pages after VM_SequentialReadHint()
2011-07-09 19:41 ` Raghavendra D Prabhu
@ 2011-07-09 19:41 ` Raghavendra D Prabhu
-1 siblings, 0 replies; 21+ messages in thread
From: Raghavendra D Prabhu @ 2011-07-09 19:41 UTC (permalink / raw)
To: fengguang.wu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
page_cache_sync_readahead checks for ra->ra_pages again, so moving the check after VM_SequentialReadHint.
Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
---
mm/filemap.c | 5 +++--
1 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/mm/filemap.c b/mm/filemap.c
index 074c23d..748f720 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1566,8 +1566,6 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
/* If we don't want any read-ahead, don't bother */
if (VM_RandomReadHint(vma))
return;
- if (!ra->ra_pages)
- return;
if (VM_SequentialReadHint(vma)) {
page_cache_sync_readahead(mapping, ra, file, offset,
@@ -1575,6 +1573,9 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
return;
}
+ if (!ra->ra_pages)
+ return;
+
/* Avoid banging the cache line if not needed */
if (ra->mmap_miss < MMAP_LOTSAMISS * 10)
ra->mmap_miss++;
--
1.7.6
^ permalink raw reply related [flat|nested] 21+ messages in thread
* [PATCH 3/3] mm/readahead: Move the check for ra_pages after VM_SequentialReadHint()
@ 2011-07-09 19:41 ` Raghavendra D Prabhu
0 siblings, 0 replies; 21+ messages in thread
From: Raghavendra D Prabhu @ 2011-07-09 19:41 UTC (permalink / raw)
To: fengguang.wu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
page_cache_sync_readahead checks for ra->ra_pages again, so moving the check after VM_SequentialReadHint.
Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
---
mm/filemap.c | 5 +++--
1 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/mm/filemap.c b/mm/filemap.c
index 074c23d..748f720 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1566,8 +1566,6 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
/* If we don't want any read-ahead, don't bother */
if (VM_RandomReadHint(vma))
return;
- if (!ra->ra_pages)
- return;
if (VM_SequentialReadHint(vma)) {
page_cache_sync_readahead(mapping, ra, file, offset,
@@ -1575,6 +1573,9 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
return;
}
+ if (!ra->ra_pages)
+ return;
+
/* Avoid banging the cache line if not needed */
if (ra->mmap_miss < MMAP_LOTSAMISS * 10)
ra->mmap_miss++;
--
1.7.6
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related [flat|nested] 21+ messages in thread
* Re: [PATCH 1/3] mm/readahead: Change the check for PageReadahead into an else-if
2011-07-09 19:41 ` Raghavendra D Prabhu
@ 2011-07-09 20:47 ` Wu Fengguang
-1 siblings, 0 replies; 21+ messages in thread
From: Wu Fengguang @ 2011-07-09 20:47 UTC (permalink / raw)
To: Raghavendra D Prabhu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
On Sun, Jul 10, 2011 at 03:41:18AM +0800, Raghavendra D Prabhu wrote:
> >From 51daa88ebd8e0d437289f589af29d4b39379ea76, page_sync_readahead coalesces
> async readahead into its readahead window, so another checking for that again is
> not required.
>
> Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
Acked-by: Wu Fengguang <fengguang.wu@intel.com>
Thank you!
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH 1/3] mm/readahead: Change the check for PageReadahead into an else-if
@ 2011-07-09 20:47 ` Wu Fengguang
0 siblings, 0 replies; 21+ messages in thread
From: Wu Fengguang @ 2011-07-09 20:47 UTC (permalink / raw)
To: Raghavendra D Prabhu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
On Sun, Jul 10, 2011 at 03:41:18AM +0800, Raghavendra D Prabhu wrote:
> >From 51daa88ebd8e0d437289f589af29d4b39379ea76, page_sync_readahead coalesces
> async readahead into its readahead window, so another checking for that again is
> not required.
>
> Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
Acked-by: Wu Fengguang <fengguang.wu@intel.com>
Thank you!
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH 2/3] mm/readahead: Remove file_ra_state from arguments of count_history_pages.
2011-07-09 19:41 ` Raghavendra D Prabhu
@ 2011-07-09 20:47 ` Wu Fengguang
-1 siblings, 0 replies; 21+ messages in thread
From: Wu Fengguang @ 2011-07-09 20:47 UTC (permalink / raw)
To: Raghavendra D Prabhu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
On Sun, Jul 10, 2011 at 03:41:19AM +0800, Raghavendra D Prabhu wrote:
> count_history_pages doesn't require readahead state to calculate the offset from history.
>
> Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
Acked-by: Wu Fengguang <fengguang.wu@intel.com>
Thank you!
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH 2/3] mm/readahead: Remove file_ra_state from arguments of count_history_pages.
@ 2011-07-09 20:47 ` Wu Fengguang
0 siblings, 0 replies; 21+ messages in thread
From: Wu Fengguang @ 2011-07-09 20:47 UTC (permalink / raw)
To: Raghavendra D Prabhu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
On Sun, Jul 10, 2011 at 03:41:19AM +0800, Raghavendra D Prabhu wrote:
> count_history_pages doesn't require readahead state to calculate the offset from history.
>
> Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
Acked-by: Wu Fengguang <fengguang.wu@intel.com>
Thank you!
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH 3/3] mm/readahead: Move the check for ra_pages after VM_SequentialReadHint()
2011-07-09 19:41 ` Raghavendra D Prabhu
@ 2011-07-09 20:53 ` Wu Fengguang
-1 siblings, 0 replies; 21+ messages in thread
From: Wu Fengguang @ 2011-07-09 20:53 UTC (permalink / raw)
To: Raghavendra D Prabhu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
On Sun, Jul 10, 2011 at 03:41:20AM +0800, Raghavendra D Prabhu wrote:
> page_cache_sync_readahead checks for ra->ra_pages again, so moving the check after VM_SequentialReadHint.
NAK. This patch adds nothing but overheads.
> --- a/mm/filemap.c
> +++ b/mm/filemap.c
> @@ -1566,8 +1566,6 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
> /* If we don't want any read-ahead, don't bother */
> if (VM_RandomReadHint(vma))
> return;
> - if (!ra->ra_pages)
> - return;
>
> if (VM_SequentialReadHint(vma)) {
> page_cache_sync_readahead(mapping, ra, file, offset,
> @@ -1575,6 +1573,9 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
> return;
> }
>
> + if (!ra->ra_pages)
> + return;
> +
page_cache_sync_readahead() has the same
if (!ra->ra_pages)
return;
So the patch adds the call into page_cache_sync_readahead() just to return..
Thanks,
Fengguang
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH 3/3] mm/readahead: Move the check for ra_pages after VM_SequentialReadHint()
@ 2011-07-09 20:53 ` Wu Fengguang
0 siblings, 0 replies; 21+ messages in thread
From: Wu Fengguang @ 2011-07-09 20:53 UTC (permalink / raw)
To: Raghavendra D Prabhu; +Cc: linux-mm, linux-kernel, akpm, Raghavendra D Prabhu
On Sun, Jul 10, 2011 at 03:41:20AM +0800, Raghavendra D Prabhu wrote:
> page_cache_sync_readahead checks for ra->ra_pages again, so moving the check after VM_SequentialReadHint.
NAK. This patch adds nothing but overheads.
> --- a/mm/filemap.c
> +++ b/mm/filemap.c
> @@ -1566,8 +1566,6 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
> /* If we don't want any read-ahead, don't bother */
> if (VM_RandomReadHint(vma))
> return;
> - if (!ra->ra_pages)
> - return;
>
> if (VM_SequentialReadHint(vma)) {
> page_cache_sync_readahead(mapping, ra, file, offset,
> @@ -1575,6 +1573,9 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
> return;
> }
>
> + if (!ra->ra_pages)
> + return;
> +
page_cache_sync_readahead() has the same
if (!ra->ra_pages)
return;
So the patch adds the call into page_cache_sync_readahead() just to return..
Thanks,
Fengguang
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [TOME] Re: [PATCH 3/3] mm/readahead: Move the check for ra_pages after VM_SequentialReadHint()
2011-07-09 20:53 ` Wu Fengguang
(?)
@ 2011-07-10 12:59 ` Raghavendra D Prabhu
2011-07-10 15:59 ` Wu Fengguang
-1 siblings, 1 reply; 21+ messages in thread
From: Raghavendra D Prabhu @ 2011-07-10 12:59 UTC (permalink / raw)
To: Wu Fengguang; +Cc: linux-mm, linux-kernel, akpm
[-- Attachment #1: Type: text/plain, Size: 1650 bytes --]
* On Sat, Jul 09, 2011 at 01:53:08PM -0700, Wu Fengguang <fengguang.wu@intel.com> wrote:
>On Sun, Jul 10, 2011 at 03:41:20AM +0800, Raghavendra D Prabhu wrote:
>> page_cache_sync_readahead checks for ra->ra_pages again, so moving the check after VM_SequentialReadHint.
>
>NAK. This patch adds nothing but overheads.
>
>> --- a/mm/filemap.c
>> +++ b/mm/filemap.c
>> @@ -1566,8 +1566,6 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
>> /* If we don't want any read-ahead, don't bother */
>> if (VM_RandomReadHint(vma))
>> return;
>> - if (!ra->ra_pages)
>> - return;
>> if (VM_SequentialReadHint(vma)) {
>> page_cache_sync_readahead(mapping, ra, file, offset,
>> @@ -1575,6 +1573,9 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
>> return;
>> }
>> + if (!ra->ra_pages)
>> + return;
>> +
>
>page_cache_sync_readahead() has the same
>
> if (!ra->ra_pages)
> return;
1. Yes, I saw that and that is why I moved it after the condition, so that duplicate checks are
not needed -- ie., if VM_SequentialReadHint is true, then
(!ra->ra_pages) is checked twice otherwise.
2. Also, another thought, is the check needed at its original place (if
not it can be removed), reasons being -- filesystems like tmpfs which
have ra_pages set to 0 don't use filemap_fault in their VMA ops and also
do_sync_mmap_readahead is called in a major page fault context.
>
>So the patch adds the call into page_cache_sync_readahead() just to return..
>
>Thanks,
>Fengguang
>
--------------------------
Raghavendra Prabhu
GPG Id : 0xD72BE977
Fingerprint: B93F EBCB 8E05 7039 CD3C A4B8 A616 DCA1 D72B E977
www: wnohang.net
[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [TOME] Re: [PATCH 3/3] mm/readahead: Move the check for ra_pages after VM_SequentialReadHint()
2011-07-10 12:59 ` [TOME] " Raghavendra D Prabhu
@ 2011-07-10 15:59 ` Wu Fengguang
0 siblings, 0 replies; 21+ messages in thread
From: Wu Fengguang @ 2011-07-10 15:59 UTC (permalink / raw)
To: Raghavendra D Prabhu; +Cc: linux-mm, linux-kernel, akpm
On Sun, Jul 10, 2011 at 08:59:09PM +0800, Raghavendra D Prabhu wrote:
> * On Sat, Jul 09, 2011 at 01:53:08PM -0700, Wu Fengguang <fengguang.wu@intel.com> wrote:
> >On Sun, Jul 10, 2011 at 03:41:20AM +0800, Raghavendra D Prabhu wrote:
> >>page_cache_sync_readahead checks for ra->ra_pages again, so moving the check after VM_SequentialReadHint.
> >
> >NAK. This patch adds nothing but overheads.
> >
> >>--- a/mm/filemap.c
> >>+++ b/mm/filemap.c
> >>@@ -1566,8 +1566,6 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
> >> /* If we don't want any read-ahead, don't bother */
> >> if (VM_RandomReadHint(vma))
> >> return;
> >>- if (!ra->ra_pages)
> >>- return;
>
> >> if (VM_SequentialReadHint(vma)) {
> >> page_cache_sync_readahead(mapping, ra, file, offset,
> >>@@ -1575,6 +1573,9 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
> >> return;
> >> }
>
> >>+ if (!ra->ra_pages)
> >>+ return;
> >>+
> >
> >page_cache_sync_readahead() has the same
> >
> > if (!ra->ra_pages)
> > return;
> 1. Yes, I saw that and that is why I moved it after the condition, so that duplicate checks are
> not needed -- ie., if VM_SequentialReadHint is true, then
> (!ra->ra_pages) is checked twice otherwise.
Ok, I see.
> 2. Also, another thought, is the check needed at its original place (if
> not it can be removed), reasons being -- filesystems like tmpfs which
> have ra_pages set to 0 don't use filemap_fault in their VMA ops and also
Good point. tmpfs is using shmem_fault().. Can you remove the test?
> do_sync_mmap_readahead is called in a major page fault context.
Right. This is irrelevant however, because if pa_pages==0, the
page faults will normally be major ones.
Thanks,
Fengguang
> >So the patch adds the call into page_cache_sync_readahead() just to return..
> >
> >Thanks,
> >Fengguang
> >
> --------------------------
> Raghavendra Prabhu
> GPG Id : 0xD72BE977
> Fingerprint: B93F EBCB 8E05 7039 CD3C A4B8 A616 DCA1 D72B E977
> www: wnohang.net
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [TOME] Re: [PATCH 3/3] mm/readahead: Move the check for ra_pages after VM_SequentialReadHint()
@ 2011-07-10 15:59 ` Wu Fengguang
0 siblings, 0 replies; 21+ messages in thread
From: Wu Fengguang @ 2011-07-10 15:59 UTC (permalink / raw)
To: Raghavendra D Prabhu; +Cc: linux-mm, linux-kernel, akpm
On Sun, Jul 10, 2011 at 08:59:09PM +0800, Raghavendra D Prabhu wrote:
> * On Sat, Jul 09, 2011 at 01:53:08PM -0700, Wu Fengguang <fengguang.wu@intel.com> wrote:
> >On Sun, Jul 10, 2011 at 03:41:20AM +0800, Raghavendra D Prabhu wrote:
> >>page_cache_sync_readahead checks for ra->ra_pages again, so moving the check after VM_SequentialReadHint.
> >
> >NAK. This patch adds nothing but overheads.
> >
> >>--- a/mm/filemap.c
> >>+++ b/mm/filemap.c
> >>@@ -1566,8 +1566,6 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
> >> /* If we don't want any read-ahead, don't bother */
> >> if (VM_RandomReadHint(vma))
> >> return;
> >>- if (!ra->ra_pages)
> >>- return;
>
> >> if (VM_SequentialReadHint(vma)) {
> >> page_cache_sync_readahead(mapping, ra, file, offset,
> >>@@ -1575,6 +1573,9 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
> >> return;
> >> }
>
> >>+ if (!ra->ra_pages)
> >>+ return;
> >>+
> >
> >page_cache_sync_readahead() has the same
> >
> > if (!ra->ra_pages)
> > return;
> 1. Yes, I saw that and that is why I moved it after the condition, so that duplicate checks are
> not needed -- ie., if VM_SequentialReadHint is true, then
> (!ra->ra_pages) is checked twice otherwise.
Ok, I see.
> 2. Also, another thought, is the check needed at its original place (if
> not it can be removed), reasons being -- filesystems like tmpfs which
> have ra_pages set to 0 don't use filemap_fault in their VMA ops and also
Good point. tmpfs is using shmem_fault().. Can you remove the test?
> do_sync_mmap_readahead is called in a major page fault context.
Right. This is irrelevant however, because if pa_pages==0, the
page faults will normally be major ones.
Thanks,
Fengguang
> >So the patch adds the call into page_cache_sync_readahead() just to return..
> >
> >Thanks,
> >Fengguang
> >
> --------------------------
> Raghavendra Prabhu
> GPG Id : 0xD72BE977
> Fingerprint: B93F EBCB 8E05 7039 CD3C A4B8 A616 DCA1 D72B E977
> www: wnohang.net
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re. Revised [PATCH 3/3] mm/readahead: Remove the check for ra->ra_pages
2011-07-10 15:59 ` Wu Fengguang
@ 2011-07-11 23:02 ` Raghavendra D Prabhu
-1 siblings, 0 replies; 21+ messages in thread
From: Raghavendra D Prabhu @ 2011-07-11 23:02 UTC (permalink / raw)
To: Wu Fengguang; +Cc: linux-mm, linux-kernel, akpm
* On Sun, Jul 10, 2011 at 08:59:06AM -0700, Wu Fengguang <fengguang.wu@intel.com> wrote:
>On Sun, Jul 10, 2011 at 08:59:09PM +0800, Raghavendra D Prabhu wrote:
>> * On Sat, Jul 09, 2011 at 01:53:08PM -0700, Wu Fengguang <fengguang.wu@intel.com> wrote:
>> >On Sun, Jul 10, 2011 at 03:41:20AM +0800, Raghavendra D Prabhu wrote:
>> >>page_cache_sync_readahead checks for ra->ra_pages again, so moving the check after VM_SequentialReadHint.
>> >NAK. This patch adds nothing but overheads.
>> >>--- a/mm/filemap.c
>> >>+++ b/mm/filemap.c
>> >>@@ -1566,8 +1566,6 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
>> >> /* If we don't want any read-ahead, don't bother */
>> >> if (VM_RandomReadHint(vma))
>> >> return;
>> >>- if (!ra->ra_pages)
>> >>- return;
>> >> if (VM_SequentialReadHint(vma)) {
>> >> page_cache_sync_readahead(mapping, ra, file, offset,
>> >>@@ -1575,6 +1573,9 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
>> >> return;
>> >> }
>> >>+ if (!ra->ra_pages)
>> >>+ return;
>> >>+
>> >page_cache_sync_readahead() has the same
>> > if (!ra->ra_pages)
>> > return;
>> 1. Yes, I saw that and that is why I moved it after the condition, so that duplicate checks are
>> not needed -- ie., if VM_SequentialReadHint is true, then
>> (!ra->ra_pages) is checked twice otherwise.
>
>Ok, I see.
>
>> 2. Also, another thought, is the check needed at its original place (if
>> not it can be removed), reasons being -- filesystems like tmpfs which
>> have ra_pages set to 0 don't use filemap_fault in their VMA ops and also
>
>Good point. tmpfs is using shmem_fault().. Can you remove the test?
I have removed that test. Patch attached.
>
>> do_sync_mmap_readahead is called in a major page fault context.
>
>Right. This is irrelevant however, because if pa_pages==0, the
>page faults will normally be major ones.
>
>Thanks,
>Fengguang
>
>> >So the patch adds the call into page_cache_sync_readahead() just to return..
>> >Thanks,
>> >Fengguang
>> --------------------------
>> Raghavendra Prabhu
>> GPG Id : 0xD72BE977
>> Fingerprint: B93F EBCB 8E05 7039 CD3C A4B8 A616 DCA1 D72B E977
>> www: wnohang.net
>
>
======================================================================
The check for ra->ra_pages is not required since fs like tmpfs which have
ra_pages set to 0 don't use filemap_fault as part of their VMA ops (it uses
shmem_fault). Also, page_cache_sync_readahead does its own check for ra_pages.
Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
---
mm/filemap.c | 2 --
1 files changed, 0 insertions(+), 2 deletions(-)
diff --git a/mm/filemap.c b/mm/filemap.c
index 074c23d..0bcd276 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1566,8 +1566,6 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
/* If we don't want any read-ahead, don't bother */
if (VM_RandomReadHint(vma))
return;
- if (!ra->ra_pages)
- return;
if (VM_SequentialReadHint(vma)) {
page_cache_sync_readahead(mapping, ra, file, offset,
--
1.7.6
--------------------------
Raghavendra Prabhu
GPG Id : 0xD72BE977
Fingerprint: B93F EBCB 8E05 7039 CD3C A4B8 A616 DCA1 D72B E977
www: wnohang.net
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re. Revised [PATCH 3/3] mm/readahead: Remove the check for ra->ra_pages
@ 2011-07-11 23:02 ` Raghavendra D Prabhu
0 siblings, 0 replies; 21+ messages in thread
From: Raghavendra D Prabhu @ 2011-07-11 23:02 UTC (permalink / raw)
To: Wu Fengguang; +Cc: linux-mm, linux-kernel, akpm
* On Sun, Jul 10, 2011 at 08:59:06AM -0700, Wu Fengguang <fengguang.wu@intel.com> wrote:
>On Sun, Jul 10, 2011 at 08:59:09PM +0800, Raghavendra D Prabhu wrote:
>> * On Sat, Jul 09, 2011 at 01:53:08PM -0700, Wu Fengguang <fengguang.wu@intel.com> wrote:
>> >On Sun, Jul 10, 2011 at 03:41:20AM +0800, Raghavendra D Prabhu wrote:
>> >>page_cache_sync_readahead checks for ra->ra_pages again, so moving the check after VM_SequentialReadHint.
>> >NAK. This patch adds nothing but overheads.
>> >>--- a/mm/filemap.c
>> >>+++ b/mm/filemap.c
>> >>@@ -1566,8 +1566,6 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
>> >> /* If we don't want any read-ahead, don't bother */
>> >> if (VM_RandomReadHint(vma))
>> >> return;
>> >>- if (!ra->ra_pages)
>> >>- return;
>> >> if (VM_SequentialReadHint(vma)) {
>> >> page_cache_sync_readahead(mapping, ra, file, offset,
>> >>@@ -1575,6 +1573,9 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
>> >> return;
>> >> }
>> >>+ if (!ra->ra_pages)
>> >>+ return;
>> >>+
>> >page_cache_sync_readahead() has the same
>> > if (!ra->ra_pages)
>> > return;
>> 1. Yes, I saw that and that is why I moved it after the condition, so that duplicate checks are
>> not needed -- ie., if VM_SequentialReadHint is true, then
>> (!ra->ra_pages) is checked twice otherwise.
>
>Ok, I see.
>
>> 2. Also, another thought, is the check needed at its original place (if
>> not it can be removed), reasons being -- filesystems like tmpfs which
>> have ra_pages set to 0 don't use filemap_fault in their VMA ops and also
>
>Good point. tmpfs is using shmem_fault().. Can you remove the test?
I have removed that test. Patch attached.
>
>> do_sync_mmap_readahead is called in a major page fault context.
>
>Right. This is irrelevant however, because if pa_pages==0, the
>page faults will normally be major ones.
>
>Thanks,
>Fengguang
>
>> >So the patch adds the call into page_cache_sync_readahead() just to return..
>> >Thanks,
>> >Fengguang
>> --------------------------
>> Raghavendra Prabhu
>> GPG Id : 0xD72BE977
>> Fingerprint: B93F EBCB 8E05 7039 CD3C A4B8 A616 DCA1 D72B E977
>> www: wnohang.net
>
>
======================================================================
The check for ra->ra_pages is not required since fs like tmpfs which have
ra_pages set to 0 don't use filemap_fault as part of their VMA ops (it uses
shmem_fault). Also, page_cache_sync_readahead does its own check for ra_pages.
Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
---
mm/filemap.c | 2 --
1 files changed, 0 insertions(+), 2 deletions(-)
diff --git a/mm/filemap.c b/mm/filemap.c
index 074c23d..0bcd276 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1566,8 +1566,6 @@ static void do_sync_mmap_readahead(struct vm_area_struct *vma,
/* If we don't want any read-ahead, don't bother */
if (VM_RandomReadHint(vma))
return;
- if (!ra->ra_pages)
- return;
if (VM_SequentialReadHint(vma)) {
page_cache_sync_readahead(mapping, ra, file, offset,
--
1.7.6
--------------------------
Raghavendra Prabhu
GPG Id : 0xD72BE977
Fingerprint: B93F EBCB 8E05 7039 CD3C A4B8 A616 DCA1 D72B E977
www: wnohang.net
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: Re. Revised [PATCH 3/3] mm/readahead: Remove the check for ra->ra_pages
2011-07-11 23:02 ` Raghavendra D Prabhu
@ 2011-07-12 18:09 ` Wu Fengguang
-1 siblings, 0 replies; 21+ messages in thread
From: Wu Fengguang @ 2011-07-12 18:09 UTC (permalink / raw)
To: Raghavendra D Prabhu; +Cc: linux-mm, linux-kernel, akpm
> The check for ra->ra_pages is not required since fs like tmpfs which have
> ra_pages set to 0 don't use filemap_fault as part of their VMA ops (it uses
> shmem_fault). Also, page_cache_sync_readahead does its own check for ra_pages.
>
> Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
Acked-by: Wu Fengguang <fengguang.wu@intel.com>
Thank you!
^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: Re. Revised [PATCH 3/3] mm/readahead: Remove the check for ra->ra_pages
@ 2011-07-12 18:09 ` Wu Fengguang
0 siblings, 0 replies; 21+ messages in thread
From: Wu Fengguang @ 2011-07-12 18:09 UTC (permalink / raw)
To: Raghavendra D Prabhu; +Cc: linux-mm, linux-kernel, akpm
> The check for ra->ra_pages is not required since fs like tmpfs which have
> ra_pages set to 0 don't use filemap_fault as part of their VMA ops (it uses
> shmem_fault). Also, page_cache_sync_readahead does its own check for ra_pages.
>
> Signed-off-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
Acked-by: Wu Fengguang <fengguang.wu@intel.com>
Thank you!
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 21+ messages in thread
end of thread, other threads:[~2011-07-12 18:10 UTC | newest]
Thread overview: 21+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2011-07-09 19:41 [PATCH 0/3] Readahead fixes Raghavendra D Prabhu
2011-07-09 19:41 ` Raghavendra D Prabhu
2011-07-09 19:41 ` [PATCH 1/3] mm/readahead: Change the check for PageReadahead into an else-if Raghavendra D Prabhu
2011-07-09 19:41 ` Raghavendra D Prabhu
2011-07-09 20:47 ` Wu Fengguang
2011-07-09 20:47 ` Wu Fengguang
2011-07-09 19:41 ` [PATCH 2/3] mm/readahead: Remove file_ra_state from arguments of count_history_pages Raghavendra D Prabhu
2011-07-09 19:41 ` Raghavendra D Prabhu
2011-07-09 20:47 ` Wu Fengguang
2011-07-09 20:47 ` Wu Fengguang
2011-07-09 19:41 ` [PATCH 3/3] mm/readahead: Move the check for ra_pages after VM_SequentialReadHint() Raghavendra D Prabhu
2011-07-09 19:41 ` Raghavendra D Prabhu
2011-07-09 20:53 ` Wu Fengguang
2011-07-09 20:53 ` Wu Fengguang
2011-07-10 12:59 ` [TOME] " Raghavendra D Prabhu
2011-07-10 15:59 ` Wu Fengguang
2011-07-10 15:59 ` Wu Fengguang
2011-07-11 23:02 ` Re. Revised [PATCH 3/3] mm/readahead: Remove the check for ra->ra_pages Raghavendra D Prabhu
2011-07-11 23:02 ` Raghavendra D Prabhu
2011-07-12 18:09 ` Wu Fengguang
2011-07-12 18:09 ` Wu Fengguang
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.