Message ID | 99b317af0f20a170fba2e70695d7cca1597fb19a.1570100361.git.mbobrowski@mbobrowski.org (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | ext4: port direct I/O to iomap infrastructure | expand |
On Thu 03-10-19 21:33:29, Matthew Bobrowski wrote: > In preparation for porting across the direct I/O path to iomap, split > out the IOMAP_WRITE logic into a separate helper. This way, we don't > need to clutter the ext4_iomap_begin() callback. > > Signed-off-by: Matthew Bobrowski <mbobrowski@mbobrowski.org> The patch looks good to me. You can add: Reviewed-by: Jan Kara <jack@suse.cz> Just please reformat the comments to use full 80 column lines. Your Emacs still doesn't seem to get it :) Honza > --- > fs/ext4/inode.c | 110 ++++++++++++++++++++++++++---------------------- > 1 file changed, 60 insertions(+), 50 deletions(-) > > diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c > index 1ccdc14c4d69..caeb3dec0dec 100644 > --- a/fs/ext4/inode.c > +++ b/fs/ext4/inode.c > @@ -3439,6 +3439,62 @@ static int ext4_set_iomap(struct inode *inode, struct iomap *iomap, u16 type, > return 0; > } > > +static int ext4_iomap_alloc(struct inode *inode, > + unsigned flags, > + unsigned long first_block, > + struct ext4_map_blocks *map) > +{ > + handle_t *handle; > + u8 blkbits = inode->i_blkbits; > + int ret, dio_credits, retries = 0; > + > + /* > + * Trim mapping request to the maximum value that we can map > + * at once for direct I/O. > + */ > + if (map->m_len > DIO_MAX_BLOCKS) > + map->m_len = DIO_MAX_BLOCKS; > + dio_credits = ext4_chunk_trans_blocks(inode, map->m_len); > +retry: > + /* > + * Either we allocate blocks and then we don't get unwritten > + * extent so we have reserved enough credits, or the blocks > + * are already allocated and unwritten. In that case, the > + * extent conversion fits in the credits as well. > + */ > + handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, dio_credits); > + if (IS_ERR(handle)) > + return PTR_ERR(handle); > + > + ret = ext4_map_blocks(handle, inode, map, EXT4_GET_BLOCKS_CREATE_ZERO); > + if (ret < 0) > + goto journal_stop; > + > + /* > + * If we have allocated blocks beyond the EOF, we need to make > + * sure that they get truncate if we crash before updating the > + * inode size metadata in ext4_iomap_end(). For faults, we > + * don't need to do that (and cannot due to the orphan list > + * operations needing an inode_lock()). If we happen to > + * instantiate blocks beyond EOF, it is because we race with a > + * truncate operation, which already has added the inode onto > + * the orphan list. > + */ > + if (!(flags & IOMAP_FAULT) && first_block + map->m_len > > + (i_size_read(inode) + (1 << blkbits) - 1) >> blkbits) { > + int err; > + > + err = ext4_orphan_add(handle, inode); > + if (err < 0) > + ret = err; > + } > +journal_stop: > + ext4_journal_stop(handle); > + if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) > + goto retry; > + return ret; > +} > + > static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length, > unsigned flags, struct iomap *iomap) > { > @@ -3500,62 +3556,16 @@ static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length, > } > } > } else if (flags & IOMAP_WRITE) { > - int dio_credits; > - handle_t *handle; > - int retries = 0; > - > - /* Trim mapping request to maximum we can map at once for DIO */ > - if (map.m_len > DIO_MAX_BLOCKS) > - map.m_len = DIO_MAX_BLOCKS; > - dio_credits = ext4_chunk_trans_blocks(inode, map.m_len); > -retry: > - /* > - * Either we allocate blocks and then we don't get unwritten > - * extent so we have reserved enough credits, or the blocks > - * are already allocated and unwritten and in that case > - * extent conversion fits in the credits as well. > - */ > - handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, > - dio_credits); > - if (IS_ERR(handle)) > - return PTR_ERR(handle); > - > - ret = ext4_map_blocks(handle, inode, &map, > - EXT4_GET_BLOCKS_CREATE_ZERO); > - if (ret < 0) { > - ext4_journal_stop(handle); > - if (ret == -ENOSPC && > - ext4_should_retry_alloc(inode->i_sb, &retries)) > - goto retry; > - return ret; > - } > - > - /* > - * If we added blocks beyond i_size, we need to make sure they > - * will get truncated if we crash before updating i_size in > - * ext4_iomap_end(). For faults we don't need to do that (and > - * even cannot because for orphan list operations inode_lock is > - * required) - if we happen to instantiate block beyond i_size, > - * it is because we race with truncate which has already added > - * the inode to the orphan list. > - */ > - if (!(flags & IOMAP_FAULT) && first_block + map.m_len > > - (i_size_read(inode) + (1 << blkbits) - 1) >> blkbits) { > - int err; > - > - err = ext4_orphan_add(handle, inode); > - if (err < 0) { > - ext4_journal_stop(handle); > - return err; > - } > - } > - ext4_journal_stop(handle); > + ret = ext4_iomap_alloc(inode, flags, first_block, &map); > } else { > ret = ext4_map_blocks(NULL, inode, &map, 0); > if (ret < 0) > return ret; > } > > + if (ret < 0) > + return ret; > + > if (!ret) > type = delalloc ? IOMAP_DELALLOC : IOMAP_HOLE; > return ext4_set_iomap(inode, iomap, type, first_block, &map); > -- > 2.20.1 >
On 10/3/19 5:03 PM, Matthew Bobrowski wrote: > In preparation for porting across the direct I/O path to iomap, split > out the IOMAP_WRITE logic into a separate helper. This way, we don't > need to clutter the ext4_iomap_begin() callback. > > Signed-off-by: Matthew Bobrowski <mbobrowski@mbobrowski.org> Minor comment, but otherwise. Patch looks good to me. You may add: Reviewed-by: Ritesh Harjani <riteshh@linux.ibm.com> > --- > fs/ext4/inode.c | 110 ++++++++++++++++++++++++++---------------------- > 1 file changed, 60 insertions(+), 50 deletions(-) > > diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c > index 1ccdc14c4d69..caeb3dec0dec 100644 > --- a/fs/ext4/inode.c > +++ b/fs/ext4/inode.c > @@ -3439,6 +3439,62 @@ static int ext4_set_iomap(struct inode *inode, struct iomap *iomap, u16 type, > return 0; > } > > +static int ext4_iomap_alloc(struct inode *inode, > + unsigned flags, > + unsigned long first_block, > + struct ext4_map_blocks *map) > +{ > + handle_t *handle; > + u8 blkbits = inode->i_blkbits; > + int ret, dio_credits, retries = 0; > + > + /* > + * Trim mapping request to the maximum value that we can map > + * at once for direct I/O. > + */ > + if (map->m_len > DIO_MAX_BLOCKS) > + map->m_len = DIO_MAX_BLOCKS; > + dio_credits = ext4_chunk_trans_blocks(inode, map->m_len); > +retry: > + /* > + * Either we allocate blocks and then we don't get unwritten > + * extent so we have reserved enough credits, or the blocks > + * are already allocated and unwritten. In that case, the > + * extent conversion fits in the credits as well. > + */ > + handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, dio_credits); > + if (IS_ERR(handle)) > + return PTR_ERR(handle); > + > + ret = ext4_map_blocks(handle, inode, map, EXT4_GET_BLOCKS_CREATE_ZERO); > + if (ret < 0) > + goto journal_stop; > + > + /* > + * If we have allocated blocks beyond the EOF, we need to make > + * sure that they get truncate if we crash before updating the > + * inode size metadata in ext4_iomap_end(). For faults, we > + * don't need to do that (and cannot due to the orphan list > + * operations needing an inode_lock()). If we happen to > + * instantiate blocks beyond EOF, it is because we race with a > + * truncate operation, which already has added the inode onto > + * the orphan list. > + */ > + if (!(flags & IOMAP_FAULT) && first_block + map->m_len > > + (i_size_read(inode) + (1 << blkbits) - 1) >> blkbits) { > + int err; > + > + err = ext4_orphan_add(handle, inode); > + if (err < 0) > + ret = err; > + } > +journal_stop: > + ext4_journal_stop(handle); > + if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) > + goto retry; > + return ret; > +} > + > static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length, > unsigned flags, struct iomap *iomap) > { > @@ -3500,62 +3556,16 @@ static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length, > } > } > } else if (flags & IOMAP_WRITE) { > - int dio_credits; > - handle_t *handle; > - int retries = 0; > - > - /* Trim mapping request to maximum we can map at once for DIO */ > - if (map.m_len > DIO_MAX_BLOCKS) > - map.m_len = DIO_MAX_BLOCKS; > - dio_credits = ext4_chunk_trans_blocks(inode, map.m_len); > -retry: > - /* > - * Either we allocate blocks and then we don't get unwritten > - * extent so we have reserved enough credits, or the blocks > - * are already allocated and unwritten and in that case > - * extent conversion fits in the credits as well. > - */ > - handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, > - dio_credits); > - if (IS_ERR(handle)) > - return PTR_ERR(handle); > - > - ret = ext4_map_blocks(handle, inode, &map, > - EXT4_GET_BLOCKS_CREATE_ZERO); > - if (ret < 0) { > - ext4_journal_stop(handle); > - if (ret == -ENOSPC && > - ext4_should_retry_alloc(inode->i_sb, &retries)) > - goto retry; > - return ret; > - } > - > - /* > - * If we added blocks beyond i_size, we need to make sure they > - * will get truncated if we crash before updating i_size in > - * ext4_iomap_end(). For faults we don't need to do that (and > - * even cannot because for orphan list operations inode_lock is > - * required) - if we happen to instantiate block beyond i_size, > - * it is because we race with truncate which has already added > - * the inode to the orphan list. > - */ > - if (!(flags & IOMAP_FAULT) && first_block + map.m_len > > - (i_size_read(inode) + (1 << blkbits) - 1) >> blkbits) { > - int err; > - > - err = ext4_orphan_add(handle, inode); > - if (err < 0) { > - ext4_journal_stop(handle); > - return err; > - } > - } > - ext4_journal_stop(handle); > + ret = ext4_iomap_alloc(inode, flags, first_block, &map); We don't need "first_block" argument here. Since map->m_lblk saves first_block directly above in the same function. No strong objection against ext4_iomap_alloc, but maybe ext4_iomap_map_write sounds better? Either way is fine though. > } else { > ret = ext4_map_blocks(NULL, inode, &map, 0); > if (ret < 0) > return ret; > } > > + if (ret < 0) > + return ret; > + > if (!ret) > type = delalloc ? IOMAP_DELALLOC : IOMAP_HOLE; > return ext4_set_iomap(inode, iomap, type, first_block, &map); >
On Tue, Oct 08, 2019 at 12:31:37PM +0200, Jan Kara wrote: > On Thu 03-10-19 21:33:29, Matthew Bobrowski wrote: > > In preparation for porting across the direct I/O path to iomap, split > > out the IOMAP_WRITE logic into a separate helper. This way, we don't > > need to clutter the ext4_iomap_begin() callback. > > > > Signed-off-by: Matthew Bobrowski <mbobrowski@mbobrowski.org> > > The patch looks good to me. You can add: > > Reviewed-by: Jan Kara <jack@suse.cz> > > Just please reformat the comments to use full 80 column lines. Your Emacs > still doesn't seem to get it :) *nod* :) --<M>--
On Wed, Oct 09, 2019 at 11:52:41AM +0530, Ritesh Harjani wrote: > On 10/3/19 5:03 PM, Matthew Bobrowski wrote: > Minor comment, but otherwise. > Patch looks good to me. You may add: > > Reviewed-by: Ritesh Harjani <riteshh@linux.ibm.com> *nod* - Thank you! > > static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length, > > unsigned flags, struct iomap *iomap) > > { > > @@ -3500,62 +3556,16 @@ static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length, > > } > > } > > } else if (flags & IOMAP_WRITE) { > > - int dio_credits; > > - handle_t *handle; > > - int retries = 0; > > - > > - /* Trim mapping request to maximum we can map at once for DIO */ > > - if (map.m_len > DIO_MAX_BLOCKS) > > - map.m_len = DIO_MAX_BLOCKS; > > - dio_credits = ext4_chunk_trans_blocks(inode, map.m_len); > > -retry: > > - /* > > - * Either we allocate blocks and then we don't get unwritten > > - * extent so we have reserved enough credits, or the blocks > > - * are already allocated and unwritten and in that case > > - * extent conversion fits in the credits as well. > > - */ > > - handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, > > - dio_credits); > > - if (IS_ERR(handle)) > > - return PTR_ERR(handle); > > - > > - ret = ext4_map_blocks(handle, inode, &map, > > - EXT4_GET_BLOCKS_CREATE_ZERO); > > - if (ret < 0) { > > - ext4_journal_stop(handle); > > - if (ret == -ENOSPC && > > - ext4_should_retry_alloc(inode->i_sb, &retries)) > > - goto retry; > > - return ret; > > - } > > - > > - /* > > - * If we added blocks beyond i_size, we need to make sure they > > - * will get truncated if we crash before updating i_size in > > - * ext4_iomap_end(). For faults we don't need to do that (and > > - * even cannot because for orphan list operations inode_lock is > > - * required) - if we happen to instantiate block beyond i_size, > > - * it is because we race with truncate which has already added > > - * the inode to the orphan list. > > - */ > > - if (!(flags & IOMAP_FAULT) && first_block + map.m_len > > > - (i_size_read(inode) + (1 << blkbits) - 1) >> blkbits) { > > - int err; > > - > > - err = ext4_orphan_add(handle, inode); > > - if (err < 0) { > > - ext4_journal_stop(handle); > > - return err; > > - } > > - } > > - ext4_journal_stop(handle); > > + ret = ext4_iomap_alloc(inode, flags, first_block, &map); > > We don't need "first_block" argument here. Since > map->m_lblk saves first_block directly above in the same function. You're right. I will change that. > No strong objection against ext4_iomap_alloc, but > maybe ext4_iomap_map_write sounds better? > Either way is fine though. I like 'ext4_iomap_alloc', because it's performing allocation in preparation for a write being performed on behalf of iomap. :) --<M>--
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 1ccdc14c4d69..caeb3dec0dec 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3439,6 +3439,62 @@ static int ext4_set_iomap(struct inode *inode, struct iomap *iomap, u16 type, return 0; } +static int ext4_iomap_alloc(struct inode *inode, + unsigned flags, + unsigned long first_block, + struct ext4_map_blocks *map) +{ + handle_t *handle; + u8 blkbits = inode->i_blkbits; + int ret, dio_credits, retries = 0; + + /* + * Trim mapping request to the maximum value that we can map + * at once for direct I/O. + */ + if (map->m_len > DIO_MAX_BLOCKS) + map->m_len = DIO_MAX_BLOCKS; + dio_credits = ext4_chunk_trans_blocks(inode, map->m_len); +retry: + /* + * Either we allocate blocks and then we don't get unwritten + * extent so we have reserved enough credits, or the blocks + * are already allocated and unwritten. In that case, the + * extent conversion fits in the credits as well. + */ + handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, dio_credits); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + ret = ext4_map_blocks(handle, inode, map, EXT4_GET_BLOCKS_CREATE_ZERO); + if (ret < 0) + goto journal_stop; + + /* + * If we have allocated blocks beyond the EOF, we need to make + * sure that they get truncate if we crash before updating the + * inode size metadata in ext4_iomap_end(). For faults, we + * don't need to do that (and cannot due to the orphan list + * operations needing an inode_lock()). If we happen to + * instantiate blocks beyond EOF, it is because we race with a + * truncate operation, which already has added the inode onto + * the orphan list. + */ + if (!(flags & IOMAP_FAULT) && first_block + map->m_len > + (i_size_read(inode) + (1 << blkbits) - 1) >> blkbits) { + int err; + + err = ext4_orphan_add(handle, inode); + if (err < 0) + ret = err; + } +journal_stop: + ext4_journal_stop(handle); + if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) + goto retry; + return ret; +} + static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length, unsigned flags, struct iomap *iomap) { @@ -3500,62 +3556,16 @@ static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length, } } } else if (flags & IOMAP_WRITE) { - int dio_credits; - handle_t *handle; - int retries = 0; - - /* Trim mapping request to maximum we can map at once for DIO */ - if (map.m_len > DIO_MAX_BLOCKS) - map.m_len = DIO_MAX_BLOCKS; - dio_credits = ext4_chunk_trans_blocks(inode, map.m_len); -retry: - /* - * Either we allocate blocks and then we don't get unwritten - * extent so we have reserved enough credits, or the blocks - * are already allocated and unwritten and in that case - * extent conversion fits in the credits as well. - */ - handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, - dio_credits); - if (IS_ERR(handle)) - return PTR_ERR(handle); - - ret = ext4_map_blocks(handle, inode, &map, - EXT4_GET_BLOCKS_CREATE_ZERO); - if (ret < 0) { - ext4_journal_stop(handle); - if (ret == -ENOSPC && - ext4_should_retry_alloc(inode->i_sb, &retries)) - goto retry; - return ret; - } - - /* - * If we added blocks beyond i_size, we need to make sure they - * will get truncated if we crash before updating i_size in - * ext4_iomap_end(). For faults we don't need to do that (and - * even cannot because for orphan list operations inode_lock is - * required) - if we happen to instantiate block beyond i_size, - * it is because we race with truncate which has already added - * the inode to the orphan list. - */ - if (!(flags & IOMAP_FAULT) && first_block + map.m_len > - (i_size_read(inode) + (1 << blkbits) - 1) >> blkbits) { - int err; - - err = ext4_orphan_add(handle, inode); - if (err < 0) { - ext4_journal_stop(handle); - return err; - } - } - ext4_journal_stop(handle); + ret = ext4_iomap_alloc(inode, flags, first_block, &map); } else { ret = ext4_map_blocks(NULL, inode, &map, 0); if (ret < 0) return ret; } + if (ret < 0) + return ret; + if (!ret) type = delalloc ? IOMAP_DELALLOC : IOMAP_HOLE; return ext4_set_iomap(inode, iomap, type, first_block, &map);
In preparation for porting across the direct I/O path to iomap, split out the IOMAP_WRITE logic into a separate helper. This way, we don't need to clutter the ext4_iomap_begin() callback. Signed-off-by: Matthew Bobrowski <mbobrowski@mbobrowski.org> --- fs/ext4/inode.c | 110 ++++++++++++++++++++++++++---------------------- 1 file changed, 60 insertions(+), 50 deletions(-)