* Re: [PATCH 1/3] writeback: change nr_pages_dirtied argument into local variable [not found] <1345283309-7768-1-git-send-email-linkinjeon@gmail.com> @ 2012-08-19 2:15 ` Fengguang Wu 2012-08-23 8:06 ` Namjae Jeon 0 siblings, 1 reply; 3+ messages in thread From: Fengguang Wu @ 2012-08-19 2:15 UTC (permalink / raw) To: Namjae Jeon Cc: chris.mason, mfasheh, jlbec, akpm, linux-kernel, Vivek Trivedi, linux-fsdevel On Sat, Aug 18, 2012 at 05:48:29AM -0400, Namjae Jeon wrote: > There is no reason nr_pages_dirtied is argument anymore. > because nr_pages_dirtied value from caller is not used in > balance_dirty_pages_ratelimited_nr function. Looks good to me. If the fs developers don't have problems with this, I'll apply it to the writeback tree. Thanks, Fengguang > Signed-off-by: Namjae Jeon <linkinjeon@gmail.com> > Signed-off-by: Vivek Trivedi <vtrivedi018@gmail.com> > --- > fs/btrfs/disk-io.c | 8 ++++---- > fs/btrfs/file.c | 3 +-- > fs/btrfs/ioctl.c | 2 +- > fs/ocfs2/file.c | 5 +---- > fs/splice.c | 5 +---- > include/linux/writeback.h | 9 +-------- > mm/page-writeback.c | 11 +++++------ > 7 files changed, 14 insertions(+), 29 deletions(-) > > diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c > index 62e0caf..10b6128 100644 > --- a/fs/btrfs/disk-io.c > +++ b/fs/btrfs/disk-io.c > @@ -3344,8 +3344,8 @@ void btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr) > num_dirty = root->fs_info->dirty_metadata_bytes; > > if (num_dirty > thresh) { > - balance_dirty_pages_ratelimited_nr( > - root->fs_info->btree_inode->i_mapping, 1); > + balance_dirty_pages_ratelimited( > + root->fs_info->btree_inode->i_mapping); > } > return; > } > @@ -3365,8 +3365,8 @@ void __btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr) > num_dirty = root->fs_info->dirty_metadata_bytes; > > if (num_dirty > thresh) { > - balance_dirty_pages_ratelimited_nr( > - root->fs_info->btree_inode->i_mapping, 1); > + balance_dirty_pages_ratelimited( > + root->fs_info->btree_inode->i_mapping); > } > return; > } > diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c > index f6b40e8..0dae684 100644 > --- a/fs/btrfs/file.c > +++ b/fs/btrfs/file.c > @@ -1314,8 +1314,7 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, > > cond_resched(); > > - balance_dirty_pages_ratelimited_nr(inode->i_mapping, > - dirty_pages); > + balance_dirty_pages_ratelimited(inode->i_mapping); > if (dirty_pages < (root->leafsize >> PAGE_CACHE_SHIFT) + 1) > btrfs_btree_balance_dirty(root, 1); > > diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c > index 7bb7556..df2e9fb 100644 > --- a/fs/btrfs/ioctl.c > +++ b/fs/btrfs/ioctl.c > @@ -1199,7 +1199,7 @@ int btrfs_defrag_file(struct inode *inode, struct file *file, > } > > defrag_count += ret; > - balance_dirty_pages_ratelimited_nr(inode->i_mapping, ret); > + balance_dirty_pages_ratelimited(inode->i_mapping); > mutex_unlock(&inode->i_mutex); > > if (newer_than) { > diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c > index 46a1f6d..a83c964 100644 > --- a/fs/ocfs2/file.c > +++ b/fs/ocfs2/file.c > @@ -2515,18 +2515,15 @@ static ssize_t ocfs2_file_splice_write(struct pipe_inode_info *pipe, > ret = sd.num_spliced; > > if (ret > 0) { > - unsigned long nr_pages; > int err; > > - nr_pages = (ret + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; > - > err = generic_write_sync(out, *ppos, ret); > if (err) > ret = err; > else > *ppos += ret; > > - balance_dirty_pages_ratelimited_nr(mapping, nr_pages); > + balance_dirty_pages_ratelimited(mapping); > } > > return ret; > diff --git a/fs/splice.c b/fs/splice.c > index 41514dd..36b6c17 100644 > --- a/fs/splice.c > +++ b/fs/splice.c > @@ -1024,17 +1024,14 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, > ret = sd.num_spliced; > > if (ret > 0) { > - unsigned long nr_pages; > int err; > > - nr_pages = (ret + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; > - > err = generic_write_sync(out, *ppos, ret); > if (err) > ret = err; > else > *ppos += ret; > - balance_dirty_pages_ratelimited_nr(mapping, nr_pages); > + balance_dirty_pages_ratelimited(mapping); > } > sb_end_write(inode->i_sb); > > diff --git a/include/linux/writeback.h b/include/linux/writeback.h > index 50c3e8f..b82a83a 100644 > --- a/include/linux/writeback.h > +++ b/include/linux/writeback.h > @@ -161,14 +161,7 @@ void __bdi_update_bandwidth(struct backing_dev_info *bdi, > unsigned long start_time); > > void page_writeback_init(void); > -void balance_dirty_pages_ratelimited_nr(struct address_space *mapping, > - unsigned long nr_pages_dirtied); > - > -static inline void > -balance_dirty_pages_ratelimited(struct address_space *mapping) > -{ > - balance_dirty_pages_ratelimited_nr(mapping, 1); > -} > +void balance_dirty_pages_ratelimited(struct address_space *mapping); > > typedef int (*writepage_t)(struct page *page, struct writeback_control *wbc, > void *data); > diff --git a/mm/page-writeback.c b/mm/page-writeback.c > index 5ad5ce2..49133b6 100644 > --- a/mm/page-writeback.c > +++ b/mm/page-writeback.c > @@ -1069,7 +1069,7 @@ static void bdi_update_bandwidth(struct backing_dev_info *bdi, > } > > /* > - * After a task dirtied this many pages, balance_dirty_pages_ratelimited_nr() > + * After a task dirtied this many pages, balance_dirty_pages_ratelimited() > * will look to see if it needs to start dirty throttling. > * > * If dirty_poll_interval is too low, big NUMA machines will call the expensive > @@ -1436,9 +1436,8 @@ static DEFINE_PER_CPU(int, bdp_ratelimits); > DEFINE_PER_CPU(int, dirty_throttle_leaks) = 0; > > /** > - * balance_dirty_pages_ratelimited_nr - balance dirty memory state > + * balance_dirty_pages_ratelimited - balance dirty memory state > * @mapping: address_space which was dirtied > - * @nr_pages_dirtied: number of pages which the caller has just dirtied > * > * Processes which are dirtying memory should call in here once for each page > * which was newly dirtied. The function will periodically check the system's > @@ -1449,8 +1448,7 @@ DEFINE_PER_CPU(int, dirty_throttle_leaks) = 0; > * limit we decrease the ratelimiting by a lot, to prevent individual processes > * from overshooting the limit by (ratelimit_pages) each. > */ > -void balance_dirty_pages_ratelimited_nr(struct address_space *mapping, > - unsigned long nr_pages_dirtied) > +void balance_dirty_pages_ratelimited(struct address_space *mapping) > { > struct backing_dev_info *bdi = mapping->backing_dev_info; > int ratelimit; > @@ -1484,6 +1482,7 @@ void balance_dirty_pages_ratelimited_nr(struct address_space *mapping, > */ > p = &__get_cpu_var(dirty_throttle_leaks); > if (*p > 0 && current->nr_dirtied < ratelimit) { > + unsigned long nr_pages_dirtied; > nr_pages_dirtied = min(*p, ratelimit - current->nr_dirtied); > *p -= nr_pages_dirtied; > current->nr_dirtied += nr_pages_dirtied; > @@ -1493,7 +1492,7 @@ void balance_dirty_pages_ratelimited_nr(struct address_space *mapping, > if (unlikely(current->nr_dirtied >= ratelimit)) > balance_dirty_pages(mapping, current->nr_dirtied); > } > -EXPORT_SYMBOL(balance_dirty_pages_ratelimited_nr); > +EXPORT_SYMBOL(balance_dirty_pages_ratelimited); > > void throttle_vm_writeout(gfp_t gfp_mask) > { > -- > 1.7.9.5 ^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH 1/3] writeback: change nr_pages_dirtied argument into local variable 2012-08-19 2:15 ` [PATCH 1/3] writeback: change nr_pages_dirtied argument into local variable Fengguang Wu @ 2012-08-23 8:06 ` Namjae Jeon 2012-08-24 16:12 ` David Sterba 0 siblings, 1 reply; 3+ messages in thread From: Namjae Jeon @ 2012-08-23 8:06 UTC (permalink / raw) To: Fengguang Wu, chris.mason, mfasheh, jlbec Cc: akpm, linux-kernel, Vivek Trivedi, linux-fsdevel, linux-btrfs, ocfs2-devel 2012/8/19, Fengguang Wu <fengguang.wu@intel.com>: > On Sat, Aug 18, 2012 at 05:48:29AM -0400, Namjae Jeon wrote: >> There is no reason nr_pages_dirtied is argument anymore. >> because nr_pages_dirtied value from caller is not used in >> balance_dirty_pages_ratelimited_nr function. > > Looks good to me. If the fs developers don't have problems with this, > I'll apply it to the writeback tree. Thanks Wu! Hi Chris, Mark, Joel. Would you check this patch ? Thanks. > > Thanks, > Fengguang > >> Signed-off-by: Namjae Jeon <linkinjeon@gmail.com> >> Signed-off-by: Vivek Trivedi <vtrivedi018@gmail.com> >> --- >> fs/btrfs/disk-io.c | 8 ++++---- >> fs/btrfs/file.c | 3 +-- >> fs/btrfs/ioctl.c | 2 +- >> fs/ocfs2/file.c | 5 +---- >> fs/splice.c | 5 +---- >> include/linux/writeback.h | 9 +-------- >> mm/page-writeback.c | 11 +++++------ >> 7 files changed, 14 insertions(+), 29 deletions(-) >> >> diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c >> index 62e0caf..10b6128 100644 >> --- a/fs/btrfs/disk-io.c >> +++ b/fs/btrfs/disk-io.c >> @@ -3344,8 +3344,8 @@ void btrfs_btree_balance_dirty(struct btrfs_root >> *root, unsigned long nr) >> num_dirty = root->fs_info->dirty_metadata_bytes; >> >> if (num_dirty > thresh) { >> - balance_dirty_pages_ratelimited_nr( >> - root->fs_info->btree_inode->i_mapping, 1); >> + balance_dirty_pages_ratelimited( >> + root->fs_info->btree_inode->i_mapping); >> } >> return; >> } >> @@ -3365,8 +3365,8 @@ void __btrfs_btree_balance_dirty(struct btrfs_root >> *root, unsigned long nr) >> num_dirty = root->fs_info->dirty_metadata_bytes; >> >> if (num_dirty > thresh) { >> - balance_dirty_pages_ratelimited_nr( >> - root->fs_info->btree_inode->i_mapping, 1); >> + balance_dirty_pages_ratelimited( >> + root->fs_info->btree_inode->i_mapping); >> } >> return; >> } >> diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c >> index f6b40e8..0dae684 100644 >> --- a/fs/btrfs/file.c >> +++ b/fs/btrfs/file.c >> @@ -1314,8 +1314,7 @@ static noinline ssize_t >> __btrfs_buffered_write(struct file *file, >> >> cond_resched(); >> >> - balance_dirty_pages_ratelimited_nr(inode->i_mapping, >> - dirty_pages); >> + balance_dirty_pages_ratelimited(inode->i_mapping); >> if (dirty_pages < (root->leafsize >> PAGE_CACHE_SHIFT) + 1) >> btrfs_btree_balance_dirty(root, 1); >> >> diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c >> index 7bb7556..df2e9fb 100644 >> --- a/fs/btrfs/ioctl.c >> +++ b/fs/btrfs/ioctl.c >> @@ -1199,7 +1199,7 @@ int btrfs_defrag_file(struct inode *inode, struct >> file *file, >> } >> >> defrag_count += ret; >> - balance_dirty_pages_ratelimited_nr(inode->i_mapping, ret); >> + balance_dirty_pages_ratelimited(inode->i_mapping); >> mutex_unlock(&inode->i_mutex); >> >> if (newer_than) { >> diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c >> index 46a1f6d..a83c964 100644 >> --- a/fs/ocfs2/file.c >> +++ b/fs/ocfs2/file.c >> @@ -2515,18 +2515,15 @@ static ssize_t ocfs2_file_splice_write(struct >> pipe_inode_info *pipe, >> ret = sd.num_spliced; >> >> if (ret > 0) { >> - unsigned long nr_pages; >> int err; >> >> - nr_pages = (ret + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; >> - >> err = generic_write_sync(out, *ppos, ret); >> if (err) >> ret = err; >> else >> *ppos += ret; >> >> - balance_dirty_pages_ratelimited_nr(mapping, nr_pages); >> + balance_dirty_pages_ratelimited(mapping); >> } >> >> return ret; >> diff --git a/fs/splice.c b/fs/splice.c >> index 41514dd..36b6c17 100644 >> --- a/fs/splice.c >> +++ b/fs/splice.c >> @@ -1024,17 +1024,14 @@ generic_file_splice_write(struct pipe_inode_info >> *pipe, struct file *out, >> ret = sd.num_spliced; >> >> if (ret > 0) { >> - unsigned long nr_pages; >> int err; >> >> - nr_pages = (ret + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; >> - >> err = generic_write_sync(out, *ppos, ret); >> if (err) >> ret = err; >> else >> *ppos += ret; >> - balance_dirty_pages_ratelimited_nr(mapping, nr_pages); >> + balance_dirty_pages_ratelimited(mapping); >> } >> sb_end_write(inode->i_sb); >> >> diff --git a/include/linux/writeback.h b/include/linux/writeback.h >> index 50c3e8f..b82a83a 100644 >> --- a/include/linux/writeback.h >> +++ b/include/linux/writeback.h >> @@ -161,14 +161,7 @@ void __bdi_update_bandwidth(struct backing_dev_info >> *bdi, >> unsigned long start_time); >> >> void page_writeback_init(void); >> -void balance_dirty_pages_ratelimited_nr(struct address_space *mapping, >> - unsigned long nr_pages_dirtied); >> - >> -static inline void >> -balance_dirty_pages_ratelimited(struct address_space *mapping) >> -{ >> - balance_dirty_pages_ratelimited_nr(mapping, 1); >> -} >> +void balance_dirty_pages_ratelimited(struct address_space *mapping); >> >> typedef int (*writepage_t)(struct page *page, struct writeback_control >> *wbc, >> void *data); >> diff --git a/mm/page-writeback.c b/mm/page-writeback.c >> index 5ad5ce2..49133b6 100644 >> --- a/mm/page-writeback.c >> +++ b/mm/page-writeback.c >> @@ -1069,7 +1069,7 @@ static void bdi_update_bandwidth(struct >> backing_dev_info *bdi, >> } >> >> /* >> - * After a task dirtied this many pages, >> balance_dirty_pages_ratelimited_nr() >> + * After a task dirtied this many pages, >> balance_dirty_pages_ratelimited() >> * will look to see if it needs to start dirty throttling. >> * >> * If dirty_poll_interval is too low, big NUMA machines will call the >> expensive >> @@ -1436,9 +1436,8 @@ static DEFINE_PER_CPU(int, bdp_ratelimits); >> DEFINE_PER_CPU(int, dirty_throttle_leaks) = 0; >> >> /** >> - * balance_dirty_pages_ratelimited_nr - balance dirty memory state >> + * balance_dirty_pages_ratelimited - balance dirty memory state >> * @mapping: address_space which was dirtied >> - * @nr_pages_dirtied: number of pages which the caller has just dirtied >> * >> * Processes which are dirtying memory should call in here once for each >> page >> * which was newly dirtied. The function will periodically check the >> system's >> @@ -1449,8 +1448,7 @@ DEFINE_PER_CPU(int, dirty_throttle_leaks) = 0; >> * limit we decrease the ratelimiting by a lot, to prevent individual >> processes >> * from overshooting the limit by (ratelimit_pages) each. >> */ >> -void balance_dirty_pages_ratelimited_nr(struct address_space *mapping, >> - unsigned long nr_pages_dirtied) >> +void balance_dirty_pages_ratelimited(struct address_space *mapping) >> { >> struct backing_dev_info *bdi = mapping->backing_dev_info; >> int ratelimit; >> @@ -1484,6 +1482,7 @@ void balance_dirty_pages_ratelimited_nr(struct >> address_space *mapping, >> */ >> p = &__get_cpu_var(dirty_throttle_leaks); >> if (*p > 0 && current->nr_dirtied < ratelimit) { >> + unsigned long nr_pages_dirtied; >> nr_pages_dirtied = min(*p, ratelimit - current->nr_dirtied); >> *p -= nr_pages_dirtied; >> current->nr_dirtied += nr_pages_dirtied; >> @@ -1493,7 +1492,7 @@ void balance_dirty_pages_ratelimited_nr(struct >> address_space *mapping, >> if (unlikely(current->nr_dirtied >= ratelimit)) >> balance_dirty_pages(mapping, current->nr_dirtied); >> } >> -EXPORT_SYMBOL(balance_dirty_pages_ratelimited_nr); >> +EXPORT_SYMBOL(balance_dirty_pages_ratelimited); >> >> void throttle_vm_writeout(gfp_t gfp_mask) >> { >> -- >> 1.7.9.5 > ^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH 1/3] writeback: change nr_pages_dirtied argument into local variable 2012-08-23 8:06 ` Namjae Jeon @ 2012-08-24 16:12 ` David Sterba 0 siblings, 0 replies; 3+ messages in thread From: David Sterba @ 2012-08-24 16:12 UTC (permalink / raw) To: Namjae Jeon Cc: Fengguang Wu, chris.mason, mfasheh, jlbec, akpm, linux-kernel, Vivek Trivedi, linux-fsdevel, linux-btrfs, ocfs2-devel On Thu, Aug 23, 2012 at 05:06:19PM +0900, Namjae Jeon wrote: > 2012/8/19, Fengguang Wu <fengguang.wu@intel.com>: > > On Sat, Aug 18, 2012 at 05:48:29AM -0400, Namjae Jeon wrote: > >> There is no reason nr_pages_dirtied is argument anymore. > >> because nr_pages_dirtied value from caller is not used in > >> balance_dirty_pages_ratelimited_nr function. > > > > Looks good to me. If the fs developers don't have problems with this, > > I'll apply it to the writeback tree. > Thanks Wu! > > Hi Chris, Mark, Joel. > > Would you check this patch ? Patch is ok from the btrfs side. david ^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2012-08-24 16:12 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <1345283309-7768-1-git-send-email-linkinjeon@gmail.com>
2012-08-19 2:15 ` [PATCH 1/3] writeback: change nr_pages_dirtied argument into local variable Fengguang Wu
2012-08-23 8:06 ` Namjae Jeon
2012-08-24 16:12 ` David Sterba
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).