[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <Ypec4YV5nBNvo4H4@google.com>
Date: Wed, 1 Jun 2022 10:07:45 -0700
From: Jaegeuk Kim <jaegeuk@...nel.org>
To: Chao Yu <chao@...nel.org>
Cc: linux-f2fs-devel@...ts.sourceforge.net,
linux-kernel@...r.kernel.org, Chao Yu <chao.yu@...o.com>
Subject: Re: [PATCH v4] f2fs: separate NOCoW and pinfile semantics
On 06/01, Chao Yu wrote:
> On 2022/6/1 4:18, Jaegeuk Kim wrote:
> > On 05/31, Chao Yu wrote:
> > > On 2022/5/18 1:21, Jaegeuk Kim wrote:
> > > > On 05/17, Chao Yu wrote:
> > > > > Pinning a file is heavy, because skipping pinned files make GC
> > > > > running with heavy load or no effect.
> > > > >
> > > > > So that this patch proposes to separate nocow and pinfile semantics:
> > > > > - NOCoW flag can only be set on regular file.
> > > > > - NOCoW file will only trigger IPU at common writeback/flush.
> > > > > - NOCow file will do OPU during GC.
> > > >
> > > > How about adding
> > > > - NOCow file will allocate 2MB-aligned space via fallocate.
> > > >
> > > > So, it'd be same as file pinning except allowing GCs. wdyt?
> > >
> > > Well, it seems the segment-aligned allocation feature should not couple with
> > > pin_file or NoCow flag, what about introducing another ioctl for that? something
> > > like: F2FS_IOC_ALIGNMENT_PREALLOCATION w/ arg.alignment_start, arg.alignment_len,
> > > arg.total_len?
> >
> > We should allocate 2MB to pin the file, since android is using it.
>
> I mean we can use nocow + new ioctl instead of pinfile, thoughts?
Can we focus on the original problem that you saw where there were some
pinned files alived in Android logcat?
>
> Thanks,
>
> >
> > >
> > > Thanks,
> > >
> > > >
> > > > >
> > > > > This flag can satisfying the demand of:
> > > > > 1) avoiding fragment of file's physical block
> > > > > 2) userspace doesn't want to pin file's physical address
> > > > >
> > > > > After commit 5d539245cb18 ("f2fs: export FS_NOCOW_FL flag to user"),
> > > > > Pin_file and NOCoW flags have already been twined closely. e.g.
> > > > > once we set pinfile flag in file, nocow flag will be shown; and after
> > > > > clearing pinfile flag, nocow flag will disappear.
> > > > >
> > > > > So, in order to keep backward compatibility, let use below semantics:
> > > > >
> > > > > f2fs_ioc_set_pin_file/f2fs_fileattr_set logic:
> > > > > pinfile nocow
> > > > > set set pinfile | nocow set nocow
> > > > > clear clear pinfile | nocow clear nocow
> > > > >
> > > > > File Behaviors:
> > > > > w/ pinfile, w/ nocow: use pinfile semantics
> > > > > w/ pinfile, w/o nocow: use pinfile semantics
> > > > > w/o pinfile, w/ nocow: use nocow semantics
> > > > > w/o pinfile, w/o nocow: no pinfile or nocow semantics
> > > > >
> > > > > NOCoW can also be set on directory, and it will have no effect on
> > > > > directory, however, new files created in nocow directory will have the
> > > > > flag set.
> > > > >
> > > > > Signed-off-by: Chao Yu <chao.yu@...o.com>
> > > > > ---
> > > > > v4:
> > > > > - allow IPU only for NoCowed regular inode.
> > > > > fs/f2fs/data.c | 3 +++
> > > > > fs/f2fs/f2fs.h | 13 +++++++++++--
> > > > > fs/f2fs/file.c | 18 +++++++++++++++++-
> > > > > 3 files changed, 31 insertions(+), 3 deletions(-)
> > > > >
> > > > > diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
> > > > > index 54a7a8ad994d..42d95ac6b508 100644
> > > > > --- a/fs/f2fs/data.c
> > > > > +++ b/fs/f2fs/data.c
> > > > > @@ -2498,6 +2498,9 @@ bool f2fs_should_update_inplace(struct inode *inode, struct f2fs_io_info *fio)
> > > > > if (f2fs_is_pinned_file(inode))
> > > > > return true;
> > > > > + if (S_ISREG(inode->i_mode) && F2FS_I(inode)->i_flags & F2FS_NOCOW_FL)
> > > > > + return true;
> > > > > +
> > > > > /* if this is cold file, we should overwrite to avoid fragmentation */
> > > > > if (file_is_cold(inode))
> > > > > return true;
> > > > > diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
> > > > > index 492af5b96de1..5c67736000a7 100644
> > > > > --- a/fs/f2fs/f2fs.h
> > > > > +++ b/fs/f2fs/f2fs.h
> > > > > @@ -2916,13 +2916,15 @@ static inline void f2fs_change_bit(unsigned int nr, char *addr)
> > > > > #define F2FS_NOCOMP_FL 0x00000400 /* Don't compress */
> > > > > #define F2FS_INDEX_FL 0x00001000 /* hash-indexed directory */
> > > > > #define F2FS_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */
> > > > > +#define F2FS_NOCOW_FL 0x00800000 /* Do not cow file */
> > > > > #define F2FS_PROJINHERIT_FL 0x20000000 /* Create with parents projid */
> > > > > #define F2FS_CASEFOLD_FL 0x40000000 /* Casefolded file */
> > > > > /* Flags that should be inherited by new inodes from their parent. */
> > > > > #define F2FS_FL_INHERITED (F2FS_SYNC_FL | F2FS_NODUMP_FL | F2FS_NOATIME_FL | \
> > > > > F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL | \
> > > > > - F2FS_CASEFOLD_FL | F2FS_COMPR_FL | F2FS_NOCOMP_FL)
> > > > > + F2FS_CASEFOLD_FL | F2FS_COMPR_FL | F2FS_NOCOMP_FL | \
> > > > > + F2FS_NOCOW_FL)
> > > > > /* Flags that are appropriate for regular files (all but dir-specific ones). */
> > > > > #define F2FS_REG_FLMASK (~(F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL | \
> > > > > @@ -2954,9 +2956,16 @@ static inline void __mark_inode_dirty_flag(struct inode *inode,
> > > > > fallthrough;
> > > > > case FI_DATA_EXIST:
> > > > > case FI_INLINE_DOTS:
> > > > > - case FI_PIN_FILE:
> > > > > case FI_COMPRESS_RELEASED:
> > > > > f2fs_mark_inode_dirty_sync(inode, true);
> > > > > + break;
> > > > > + case FI_PIN_FILE:
> > > > > + if (set)
> > > > > + F2FS_I(inode)->i_flags |= F2FS_NOCOW_FL;
> > > > > + else
> > > > > + F2FS_I(inode)->i_flags &= ~F2FS_NOCOW_FL;
> > > > > + f2fs_mark_inode_dirty_sync(inode, true);
> > > > > + break;
> > > > > }
> > > > > }
> > > > > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
> > > > > index 216081ea8c81..9e71ce8601f9 100644
> > > > > --- a/fs/f2fs/file.c
> > > > > +++ b/fs/f2fs/file.c
> > > > > @@ -1851,6 +1851,20 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
> > > > > if (IS_NOQUOTA(inode))
> > > > > return -EPERM;
> > > > > + if ((iflags ^ masked_flags) & F2FS_NOCOW_FL) {
> > > > > + int ret;
> > > > > +
> > > > > + if (!S_ISREG(inode->i_mode) && !S_ISDIR(inode->i_mode))
> > > > > + return -EINVAL;
> > > > > + if (S_ISREG(inode->i_mode)) {
> > > > > + if (f2fs_should_update_outplace(inode, NULL))
> > > > > + return -EINVAL;
> > > > > + ret = f2fs_convert_inline_inode(inode);
> > > > > + if (ret)
> > > > > + return ret;
> > > > > + }
> > > > > + }
> > > > > +
> > > > > if ((iflags ^ masked_flags) & F2FS_CASEFOLD_FL) {
> > > > > if (!f2fs_sb_has_casefold(F2FS_I_SB(inode)))
> > > > > return -EOPNOTSUPP;
> > > > > @@ -1926,6 +1940,7 @@ static const struct {
> > > > > { F2FS_NOCOMP_FL, FS_NOCOMP_FL },
> > > > > { F2FS_INDEX_FL, FS_INDEX_FL },
> > > > > { F2FS_DIRSYNC_FL, FS_DIRSYNC_FL },
> > > > > + { F2FS_NOCOW_FL, FS_NOCOW_FL },
> > > > > { F2FS_PROJINHERIT_FL, FS_PROJINHERIT_FL },
> > > > > { F2FS_CASEFOLD_FL, FS_CASEFOLD_FL },
> > > > > };
> > > > > @@ -1957,7 +1972,8 @@ static const struct {
> > > > > FS_NOCOMP_FL | \
> > > > > FS_DIRSYNC_FL | \
> > > > > FS_PROJINHERIT_FL | \
> > > > > - FS_CASEFOLD_FL)
> > > > > + FS_CASEFOLD_FL | \
> > > > > + FS_NOCOW_FL)
> > > > > /* Convert f2fs on-disk i_flags to FS_IOC_{GET,SET}FLAGS flags */
> > > > > static inline u32 f2fs_iflags_to_fsflags(u32 iflags)
> > > > > --
> > > > > 2.25.1
Powered by blists - more mailing lists