| Message ID | 20190916235642.167583-2-khazhy@google.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [v3,1/2] fuse: on 64-bit store time in d_fsdata directly | expand |
On Tue, Sep 17, 2019 at 1:56 AM Khazhismel Kumykov <khazhy@google.com> wrote: > > account per-file, dentry, and inode data > > blockdev/superblock and temporary per-request data was left alone, as > this usually isn't accounted > > Signed-off-by: Khazhismel Kumykov <khazhy@google.com> > Reviewed-by: Shakeel Butt <shakeelb@google.com> > --- > fs/fuse/dir.c | 3 ++- > fs/fuse/file.c | 5 +++-- > fs/fuse/inode.c | 3 ++- > 3 files changed, 7 insertions(+), 4 deletions(-) > > diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c > index 58557d4817e9..d572c900bb0f 100644 > --- a/fs/fuse/dir.c > +++ b/fs/fuse/dir.c > @@ -279,7 +279,8 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags) > #if BITS_PER_LONG < 64 > static int fuse_dentry_init(struct dentry *dentry) > { > - dentry->d_fsdata = kzalloc(sizeof(union fuse_dentry), GFP_KERNEL); > + dentry->d_fsdata = kzalloc(sizeof(union fuse_dentry), > + GFP_KERNEL_ACCOUNT | __GFP_RECLAIMABLE); > > return dentry->d_fsdata ? 0 : -ENOMEM; > } > diff --git a/fs/fuse/file.c b/fs/fuse/file.c > index a2ea347c4d2c..862aff3665b5 100644 > --- a/fs/fuse/file.c > +++ b/fs/fuse/file.c > @@ -63,12 +63,13 @@ struct fuse_file *fuse_file_alloc(struct fuse_conn *fc) > { > struct fuse_file *ff; > > - ff = kzalloc(sizeof(struct fuse_file), GFP_KERNEL); > + ff = kzalloc(sizeof(struct fuse_file), GFP_KERNEL_ACCOUNT); > if (unlikely(!ff)) > return NULL; > > ff->fc = fc; > - ff->release_args = kzalloc(sizeof(*ff->release_args), GFP_KERNEL); > + ff->release_args = kzalloc(sizeof(*ff->release_args), > + GFP_KERNEL_ACCOUNT); > if (!ff->release_args) { > kfree(ff); > return NULL; > diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c > index 3d598a5bb5b5..6cb445bed89d 100644 > --- a/fs/fuse/inode.c > +++ b/fs/fuse/inode.c > @@ -66,7 +66,8 @@ static struct file_system_type fuseblk_fs_type; > > struct fuse_forget_link *fuse_alloc_forget(void) > { > - return kzalloc(sizeof(struct fuse_forget_link), GFP_KERNEL); > + return kzalloc(sizeof(struct fuse_forget_link), > + GFP_KERNEL_ACCOUNT | __GFP_RECLAIMABLE); What does __GFP_RECLAIMBALE signify in slab allocs? You understand that the forget_link is not reclaimable in the sense, that it requires action (reading requests from the fuse device) from the userspace filesystem daemon? Thanks, Miklos
On Tue, Sep 17, 2019 at 12:52 AM Miklos Szeredi <miklos@szeredi.hu> wrote: > > On Tue, Sep 17, 2019 at 1:56 AM Khazhismel Kumykov <khazhy@google.com> wrote: > > struct fuse_forget_link *fuse_alloc_forget(void) > > { > > - return kzalloc(sizeof(struct fuse_forget_link), GFP_KERNEL); > > + return kzalloc(sizeof(struct fuse_forget_link), > > + GFP_KERNEL_ACCOUNT | __GFP_RECLAIMABLE); > > What does __GFP_RECLAIMBALE signify in slab allocs? > Marking these allocations reclaimable hints to mm how much we can reclaim overall by shrinking, so if it is reclaimable we should mark it as such. For d_fsdata, the lifetime is linked to the dentry, which is reclaimable, so it makes sense here. > You understand that the forget_link is not reclaimable in the sense, > that it requires action (reading requests from the fuse device) from > the userspace filesystem daemon? > Ah, I see, whenever we evict the fuse_inode, we queue a forget command, which usually waits for userspace. So it's not actually linked to the inode itself, and yeah, if we need userspace to intervene we shouldn't treat forget_link as reclaimable. I had figured since fuse_inode is reclaimable, this should be too, but missed that disconnect, thanks.
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 58557d4817e9..d572c900bb0f 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -279,7 +279,8 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags) #if BITS_PER_LONG < 64 static int fuse_dentry_init(struct dentry *dentry) { - dentry->d_fsdata = kzalloc(sizeof(union fuse_dentry), GFP_KERNEL); + dentry->d_fsdata = kzalloc(sizeof(union fuse_dentry), + GFP_KERNEL_ACCOUNT | __GFP_RECLAIMABLE); return dentry->d_fsdata ? 0 : -ENOMEM; } diff --git a/fs/fuse/file.c b/fs/fuse/file.c index a2ea347c4d2c..862aff3665b5 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -63,12 +63,13 @@ struct fuse_file *fuse_file_alloc(struct fuse_conn *fc) { struct fuse_file *ff; - ff = kzalloc(sizeof(struct fuse_file), GFP_KERNEL); + ff = kzalloc(sizeof(struct fuse_file), GFP_KERNEL_ACCOUNT); if (unlikely(!ff)) return NULL; ff->fc = fc; - ff->release_args = kzalloc(sizeof(*ff->release_args), GFP_KERNEL); + ff->release_args = kzalloc(sizeof(*ff->release_args), + GFP_KERNEL_ACCOUNT); if (!ff->release_args) { kfree(ff); return NULL; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 3d598a5bb5b5..6cb445bed89d 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -66,7 +66,8 @@ static struct file_system_type fuseblk_fs_type; struct fuse_forget_link *fuse_alloc_forget(void) { - return kzalloc(sizeof(struct fuse_forget_link), GFP_KERNEL); + return kzalloc(sizeof(struct fuse_forget_link), + GFP_KERNEL_ACCOUNT | __GFP_RECLAIMABLE); } static struct inode *fuse_alloc_inode(struct super_block *sb)