akpm at linux-foundation.org
2017-Nov-30 22:24 UTC
[Ocfs2-devel] [patch 10/11] ocfs2: add ocfs2_overwrite_io()
From: Gang He <ghe at suse.com>
Subject: ocfs2: add ocfs2_overwrite_io()
Add ocfs2_overwrite_io(), which is used to judge if overwrite allocated
blocks, otherwise, the write will bring extra block allocation overhead.
[ghe at suse.com: v2]
Link:
https://urldefense.proofpoint.com/v2/url?u=http-3A__lkml.kernel.org_r_1511944612-2D9629-2D3-2Dgit-2Dsend-2Demail-2Dghe-40suse.com&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=C7gAd4uDxlAvTdc0vmU6X8CMk6L2iDY8-HD0qT6Fo7Y&m=C4AZ7lGI2-gfFtbla8puNvYIA3a9Cr_XOH6oyx94oGo&s=QhxdPyP9gdEJZTa2UhuSk0x7ENXyGXUOIBV_LyP8oZw&eLink:
https://urldefense.proofpoint.com/v2/url?u=http-3A__lkml.kernel.org_r_1511775987-2D841-2D3-2Dgit-2Dsend-2Demail-2Dghe-40suse.com&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=C7gAd4uDxlAvTdc0vmU6X8CMk6L2iDY8-HD0qT6Fo7Y&m=C4AZ7lGI2-gfFtbla8puNvYIA3a9Cr_XOH6oyx94oGo&s=T9Ljku-nipBqCaeXWukY2hAwcHQL95gtlU7pS3l1fgs&eSigned-off-by:
Gang He <ghe at suse.com>
Cc: Mark Fasheh <mfasheh at versity.com>
Cc: Joel Becker <jlbec at evilplan.org>
Cc: Junxiao Bi <junxiao.bi at oracle.com>
Cc: Joseph Qi <jiangqi903 at gmail.com>
Cc: Changwei Ge <ge.changwei at h3c.com>
Signed-off-by: Andrew Morton <akpm at linux-foundation.org>
---
fs/ocfs2/extent_map.c | 41 ++++++++++++++++++++++++++++++++++++++++
fs/ocfs2/extent_map.h | 3 ++
2 files changed, 44 insertions(+)
diff -puN fs/ocfs2/extent_map.c~ocfs2-add-ocfs2_overwrite_io-function
fs/ocfs2/extent_map.c
--- a/fs/ocfs2/extent_map.c~ocfs2-add-ocfs2_overwrite_io-function
+++ a/fs/ocfs2/extent_map.c
@@ -832,6 +832,47 @@ out:
return ret;
}
+/* Is IO overwriting allocated blocks? */
+int ocfs2_overwrite_io(struct inode *inode, struct buffer_head *di_bh,
+ u64 map_start, u64 map_len)
+{
+ int ret = 0, is_last;
+ u32 mapping_end, cpos;
+ struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+ struct ocfs2_extent_rec rec;
+
+ if ((OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) &&
+ ((map_start + map_len) <= i_size_read(inode)))
+ goto out;
+
+ cpos = map_start >> osb->s_clustersize_bits;
+ mapping_end = ocfs2_clusters_for_bytes(inode->i_sb,
+ map_start + map_len);
+ is_last = 0;
+ while (cpos < mapping_end && !is_last) {
+ ret = ocfs2_get_clusters_nocache(inode, di_bh, cpos,
+ NULL, &rec, &is_last);
+ if (ret) {
+ mlog_errno(ret);
+ goto out;
+ }
+
+ if (rec.e_blkno == 0ULL)
+ break;
+
+ if (rec.e_flags & OCFS2_EXT_REFCOUNTED)
+ break;
+
+ cpos = le32_to_cpu(rec.e_cpos) +
+ le16_to_cpu(rec.e_leaf_clusters);
+ }
+
+ if (cpos < mapping_end)
+ ret = -EAGAIN;
+out:
+ return ret;
+}
+
int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int whence)
{
struct inode *inode = file->f_mapping->host;
diff -puN fs/ocfs2/extent_map.h~ocfs2-add-ocfs2_overwrite_io-function
fs/ocfs2/extent_map.h
--- a/fs/ocfs2/extent_map.h~ocfs2-add-ocfs2_overwrite_io-function
+++ a/fs/ocfs2/extent_map.h
@@ -53,6 +53,9 @@ int ocfs2_extent_map_get_blocks(struct i
int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
u64 map_start, u64 map_len);
+int ocfs2_overwrite_io(struct inode *inode, struct buffer_head *di_bh,
+ u64 map_start, u64 map_len);
+
int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int origin);
int ocfs2_xattr_get_clusters(struct inode *inode, u32 v_cluster,
_
alex chen
2017-Dec-19 12:53 UTC
[Ocfs2-devel] [patch 10/11] ocfs2: add ocfs2_overwrite_io()
Hi Gang, On 2017/12/1 6:24, akpm at linux-foundation.org wrote:> From: Gang He <ghe at suse.com> > Subject: ocfs2: add ocfs2_overwrite_io() > > Add ocfs2_overwrite_io(), which is used to judge if overwrite allocated > blocks, otherwise, the write will bring extra block allocation overhead. > > [ghe at suse.com: v2] > Link: https://urldefense.proofpoint.com/v2/url?u=http-3A__lkml.kernel.org_r_1511944612-2D9629-2D3-2Dgit-2Dsend-2Demail-2Dghe-40suse.com&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=C7gAd4uDxlAvTdc0vmU6X8CMk6L2iDY8-HD0qT6Fo7Y&m=C4AZ7lGI2-gfFtbla8puNvYIA3a9Cr_XOH6oyx94oGo&s=QhxdPyP9gdEJZTa2UhuSk0x7ENXyGXUOIBV_LyP8oZw&e> Link: https://urldefense.proofpoint.com/v2/url?u=http-3A__lkml.kernel.org_r_1511775987-2D841-2D3-2Dgit-2Dsend-2Demail-2Dghe-40suse.com&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=C7gAd4uDxlAvTdc0vmU6X8CMk6L2iDY8-HD0qT6Fo7Y&m=C4AZ7lGI2-gfFtbla8puNvYIA3a9Cr_XOH6oyx94oGo&s=T9Ljku-nipBqCaeXWukY2hAwcHQL95gtlU7pS3l1fgs&e> Signed-off-by: Gang He <ghe at suse.com> > Cc: Mark Fasheh <mfasheh at versity.com> > Cc: Joel Becker <jlbec at evilplan.org> > Cc: Junxiao Bi <junxiao.bi at oracle.com> > Cc: Joseph Qi <jiangqi903 at gmail.com> > Cc: Changwei Ge <ge.changwei at h3c.com> > Signed-off-by: Andrew Morton <akpm at linux-foundation.org> > --- > > fs/ocfs2/extent_map.c | 41 ++++++++++++++++++++++++++++++++++++++++ > fs/ocfs2/extent_map.h | 3 ++ > 2 files changed, 44 insertions(+) > > diff -puN fs/ocfs2/extent_map.c~ocfs2-add-ocfs2_overwrite_io-function fs/ocfs2/extent_map.c > --- a/fs/ocfs2/extent_map.c~ocfs2-add-ocfs2_overwrite_io-function > +++ a/fs/ocfs2/extent_map.c > @@ -832,6 +832,47 @@ out: > return ret; > } > > +/* Is IO overwriting allocated blocks? */ > +int ocfs2_overwrite_io(struct inode *inode, struct buffer_head *di_bh, > + u64 map_start, u64 map_len) > +{ > + int ret = 0, is_last; > + u32 mapping_end, cpos; > + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); > + struct ocfs2_extent_rec rec; > + > + if ((OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) && > + ((map_start + map_len) <= i_size_read(inode))) > + goto out; > +Should we return -EAGAIN directly when the condition ((OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) && ((map_start + map_len) > i_size_read(inode))) is true? Thanks, Alex> + cpos = map_start >> osb->s_clustersize_bits; > + mapping_end = ocfs2_clusters_for_bytes(inode->i_sb, > + map_start + map_len); > + is_last = 0; > + while (cpos < mapping_end && !is_last) { > + ret = ocfs2_get_clusters_nocache(inode, di_bh, cpos, > + NULL, &rec, &is_last); > + if (ret) { > + mlog_errno(ret); > + goto out; > + } > + > + if (rec.e_blkno == 0ULL) > + break; > + > + if (rec.e_flags & OCFS2_EXT_REFCOUNTED) > + break; > + > + cpos = le32_to_cpu(rec.e_cpos) + > + le16_to_cpu(rec.e_leaf_clusters); > + } > + > + if (cpos < mapping_end) > + ret = -EAGAIN; > +out: > + return ret; > +} > + > int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int whence) > { > struct inode *inode = file->f_mapping->host; > diff -puN fs/ocfs2/extent_map.h~ocfs2-add-ocfs2_overwrite_io-function fs/ocfs2/extent_map.h > --- a/fs/ocfs2/extent_map.h~ocfs2-add-ocfs2_overwrite_io-function > +++ a/fs/ocfs2/extent_map.h > @@ -53,6 +53,9 @@ int ocfs2_extent_map_get_blocks(struct i > int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, > u64 map_start, u64 map_len); > > +int ocfs2_overwrite_io(struct inode *inode, struct buffer_head *di_bh, > + u64 map_start, u64 map_len); > + > int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int origin); > > int ocfs2_xattr_get_clusters(struct inode *inode, u32 v_cluster, > _ > > _______________________________________________ > Ocfs2-devel mailing list > Ocfs2-devel at oss.oracle.com > https://oss.oracle.com/mailman/listinfo/ocfs2-devel > > . >