Patchwork [v2] ext4: avoid redundant cache finding if flags not cantian PUNCH_OUT

login
register
mail settings
Submitter Robin Dong
Date June 27, 2011, 10:45 a.m.
Message ID <1309171513-5246-1-git-send-email-sanbai@taobao.com>
Download mbox | patch
Permalink /patch/102139/
State Superseded
Headers show

Comments

Robin Dong - June 27, 2011, 10:45 a.m.
In old code, ext4 will find extent in cache first even flags does
not contian PUNCH_OUT, that may waste some time.

Signed-off-by: Robin Dong <sanbai@taobao.com>
Reviewed-by: Lukas Czerner <lczerner@redhat.com>
---
 fs/ext4/extents.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)
Steven Liu - June 27, 2011, 1:45 p.m.
Hi Robin Dong

     Why don't modify this line?
   if ((flags & EXT4_GET_BLOCKS_CREATE) == 0) {
    if(!(flags & EXT4_GET_BLOCKS_CREATE))


2011/6/27 Robin Dong <hao.bigrat@gmail.com>:
> In old code, ext4 will find extent in cache first even flags does
> not contian PUNCH_OUT, that may waste some time.
>
> Signed-off-by: Robin Dong <sanbai@taobao.com>
> Reviewed-by: Lukas Czerner <lczerner@redhat.com>
> ---
>  fs/ext4/extents.c |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
> index eb63c7b..fed2633 100644
> --- a/fs/ext4/extents.c
> +++ b/fs/ext4/extents.c
> @@ -3358,8 +3358,8 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
>        trace_ext4_ext_map_blocks_enter(inode, map->m_lblk, map->m_len, flags);
>
>        /* check in cache */
> -       if (ext4_ext_in_cache(inode, map->m_lblk, &newex) &&
> -               ((flags & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) == 0)) {
> +       if (!(flags & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) &&
> +               ext4_ext_in_cache(inode, map->m_lblk, &newex)) {
>                if (!newex.ee_start_lo && !newex.ee_start_hi) {
>                        if ((flags & EXT4_GET_BLOCKS_CREATE) == 0) {
>                                /*
> --
> 1.7.1
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

Best Regards
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Darrick J. Wong - June 27, 2011, 7:59 p.m.
On Mon, Jun 27, 2011 at 06:45:13PM +0800, Robin Dong wrote:
> Subject: [PATCH v2] ext4: avoid redundant cache finding if flags not cantian

I suspect you mean "contain" as opposed to "cantian".

> In old code, ext4 will find extent in cache first even flags does
> not contian PUNCH_OUT, that may waste some time.

Same here ("contian").

--D
> 
> Signed-off-by: Robin Dong <sanbai@taobao.com>
> Reviewed-by: Lukas Czerner <lczerner@redhat.com>
> ---
>  fs/ext4/extents.c |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
> index eb63c7b..fed2633 100644
> --- a/fs/ext4/extents.c
> +++ b/fs/ext4/extents.c
> @@ -3358,8 +3358,8 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
>  	trace_ext4_ext_map_blocks_enter(inode, map->m_lblk, map->m_len, flags);
> 
>  	/* check in cache */
> -	if (ext4_ext_in_cache(inode, map->m_lblk, &newex) &&
> -		((flags & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) == 0)) {
> +	if (!(flags & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) &&
> +		ext4_ext_in_cache(inode, map->m_lblk, &newex)) {
>  		if (!newex.ee_start_lo && !newex.ee_start_hi) {
>  			if ((flags & EXT4_GET_BLOCKS_CREATE) == 0) {
>  				/*
> -- 
> 1.7.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Eric Gouriou - June 28, 2011, 12:44 a.m.
On Mon, Jun 27, 2011 at 12:59, Darrick J. Wong <djwong@us.ibm.com> wrote:
> On Mon, Jun 27, 2011 at 06:45:13PM +0800, Robin Dong wrote:
>> Subject: [PATCH v2] ext4: avoid redundant cache finding if flags not cantian
>
> I suspect you mean "contain" as opposed to "cantian".

The patch summary line could also use some shortening.

>> In old code, ext4 will find extent in cache first even flags does
>> not contian PUNCH_OUT, that may waste some time.
>
> Same here ("contian").

How about something along the lines of:
---------
ext4: avoid wasted extent cache lookup if !PUNCH_OUT_EXT

This patch avoids an extraneous lookup of the extent cache
in ext4_ext_map_blocks() when the flag
EXT4_GET_BLOCKS_PUNCH_OUT_EXT is absent.

The existing logic was performing the lookup but not making
use of the result. The patch simply reverses the order of evaluation
in the condition.

Since ext4_ext_in_cache() does not initialize newex on misses, bypassing
its invocation does not introduce any new issue in this regard.
---------

 Regards - Eric

>
> --D
>>
>> Signed-off-by: Robin Dong <sanbai@taobao.com>
>> Reviewed-by: Lukas Czerner <lczerner@redhat.com>
>> ---
>>  fs/ext4/extents.c |    4 ++--
>>  1 files changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
>> index eb63c7b..fed2633 100644
>> --- a/fs/ext4/extents.c
>> +++ b/fs/ext4/extents.c
>> @@ -3358,8 +3358,8 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
>>       trace_ext4_ext_map_blocks_enter(inode, map->m_lblk, map->m_len, flags);
>>
>>       /* check in cache */
>> -     if (ext4_ext_in_cache(inode, map->m_lblk, &newex) &&
>> -             ((flags & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) == 0)) {
>> +     if (!(flags & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) &&
>> +             ext4_ext_in_cache(inode, map->m_lblk, &newex)) {
>>               if (!newex.ee_start_lo && !newex.ee_start_hi) {
>>                       if ((flags & EXT4_GET_BLOCKS_CREATE) == 0) {
>>                               /*
>> --
>> 1.7.1
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index eb63c7b..fed2633 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -3358,8 +3358,8 @@  int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 	trace_ext4_ext_map_blocks_enter(inode, map->m_lblk, map->m_len, flags);
 
 	/* check in cache */
-	if (ext4_ext_in_cache(inode, map->m_lblk, &newex) &&
-		((flags & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) == 0)) {
+	if (!(flags & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) &&
+		ext4_ext_in_cache(inode, map->m_lblk, &newex)) {
 		if (!newex.ee_start_lo && !newex.ee_start_hi) {
 			if ((flags & EXT4_GET_BLOCKS_CREATE) == 0) {
 				/*