fs: Add a filesystem flag for THPs
Subsystems (4)
Matthew Wilcox (Oracle)
maintainerPATCH 1/2 fs: Add a filesystem flag for THPs
The page cache needs to know whether the filesystem supports THPs so that it doesn't send THPs to filesystems which can't handle them. Dave Chinner points out that getting from the page mapping to the filesystem type is too many steps (mapping->host->i_sb->s_type->fs_flags) so cache that information in the address space flags.
--- fs/inode.c | 2 ++ include/linux/fs.h | 1 + include/linux/pagemap.h | 6 ++++++ mm/shmem.c | 2 +- 4 files changed, 10 insertions(+), 1 deletion(-)
fs/inode.c +2 -0
@@ -181,6 +181,8 @@ int inode_init_always(struct super_block *sb, struct inode *inode) mapping->a_ops = &empty_aops; mapping->host = inode; mapping->flags =0 ; +if (sb->s_type->fs_flags &FS_THP_SUPPORT ) +__set_bit (AS_THP_SUPPORT , &mapping->flags); mapping->wb_err =0 ;atomic_set (&mapping->i_mmap_writable,0 );#ifdef CONFIG_READ_ONLY_THP_FOR_FS
include/linux/fs.h +1 -0
@@ -2217,6 +2217,7 @@ struct file_system_type {#define FS_HAS_SUBTYPE 4#define FS_USERNS_MOUNT 8/* Can be mounted by userns root */ #define FS_DISALLOW_NOTIFY_PERM 16/* Disable fanotify permission events */ +#define FS_THP_SUPPORT 8192/* Remove once all fs converted */ #define FS_RENAME_DOES_D_MOVE 32768/* FS will handle d_move() during rename() internally. */ int (*init_fs_context)(struct fs_context *);const struct fs_parameter_spec *parameters;
include/linux/pagemap.h +6 -0
@@ -34,6 +34,7 @@ enum mapping_flags {AS_EXITING =4 ,/* final truncate in progress */ /* writeback related tags are not used */ AS_NO_WRITEBACK_TAGS =5 , + AS_THP_SUPPORT = 6, /* THPs supported */ };/**
@@ -124,6 +125,11 @@ static inline void mapping_set_gfp_mask(struct address_space *m, gfp_t mask) m->gfp_mask = mask; } +static inline bool mapping_thp_support (struct address_space *mapping) +{ +return test_bit (AS_THP_SUPPORT , &mapping->flags); +} +void release_pages (struct page **pages,int nr);/*
mm/shmem.c +1 -1
@@ -3844,7 +3844,7 @@ static struct file_system_type shmem_fs_type = { .parameters = shmem_fs_parameters, #endif .kill_sb = kill_litter_super, - .fs_flags = FS_USERNS_MOUNT, + .fs_flags = FS_USERNS_MOUNT | FS_THP_SUPPORT, };int __init shmem_init (void )
Signature · 1 line
2.28.0
Matthew Wilcox (Oracle)
maintainerPATCH 2/2 fs: Do not update nr_thps for mappings which support THPs
The nr_thps counter is to support THPs in the page cache when the filesystem doesn't understand THPs. Eventually it will be removed, but we should still support filesystems which do not understand THPs yet. Move the nr_thp manipulation functions to filemap.h since they're page-cache specific.
--- include/linux/fs.h | 27 --------------------------- include/linux/pagemap.h | 29 +++++++++++++++++++++++++++++ 2 files changed, 29 insertions(+), 27 deletions(-)
include/linux/fs.h +0 -27
@@ -2701,33 +2701,6 @@ static inline errseq_t file_sample_sb_err(struct file *file)return errseq_sample (&file->f_path.dentry->d_sb->s_wb_err); } -static inline int filemap_nr_thps (struct address_space *mapping) -{ -#ifdef CONFIG_READ_ONLY_THP_FOR_FS -return atomic_read (&mapping->nr_thps); -#else -return 0 ; -#endif -} - -static inline void filemap_nr_thps_inc (struct address_space *mapping) -{ -#ifdef CONFIG_READ_ONLY_THP_FOR_FS -atomic_inc (&mapping->nr_thps); -#else -WARN_ON_ONCE (1 ); -#endif -} - -static inline void filemap_nr_thps_dec (struct address_space *mapping) -{ -#ifdef CONFIG_READ_ONLY_THP_FOR_FS -atomic_dec (&mapping->nr_thps); -#else -WARN_ON_ONCE (1 ); -#endif -} -extern int vfs_fsync_range (struct file *file,loff_t start,loff_t end,int datasync);extern int vfs_fsync (struct file *file,int datasync);
include/linux/pagemap.h +29 -0
@@ -130,6 +130,35 @@ static inline bool mapping_thp_support(struct address_space *mapping)return test_bit (AS_THP_SUPPORT , &mapping->flags); } +static inline int filemap_nr_thps (struct address_space *mapping) +{ +#ifdef CONFIG_READ_ONLY_THP_FOR_FS +return atomic_read (&mapping->nr_thps); +#else +return 0 ; +#endif +} + +static inline void filemap_nr_thps_inc (struct address_space *mapping) +{ +#ifdef CONFIG_READ_ONLY_THP_FOR_FS +if (!mapping_thp_support (mapping)) +atomic_inc (&mapping->nr_thps); +#else +WARN_ON_ONCE (1 ); +#endif +} + +static inline void filemap_nr_thps_dec (struct address_space *mapping) +{ +#ifdef CONFIG_READ_ONLY_THP_FOR_FS +if (!mapping_thp_support (mapping)) +atomic_dec (&mapping->nr_thps); +#else +WARN_ON_ONCE (1 ); +#endif +} +void release_pages (struct page **pages,int nr);/*
Signature · 1 line
2.28.0
Re: [PATCH 2/2] fs: Do not update nr_thps for mappings which support THPs
Quoted 5 lines · Matthew Wilcox (Oracle)
The
nr_thpscounter is to support THPs in the page cache when the filesystem doesn't understand THPs. Eventually it will be removed, but we should still support filesystems which do not understand THPs yet. Move thenr_thpmanipulation functions tofilemap.hsince they're page-cache specific.
Honestly I don't think we should support the read-only THP crap. We should in fact never have merged that bandaid to start with given that you did good progress on the real thing.
Matthew Wilcox
maintainerRe: [PATCH 2/2] fs: Do not update nr_thps for mappings which support THPs
Quoted 10 lines · Christoph Hellwig
Quoted 5 lines · Matthew Wilcox (Oracle)
The
nr_thpscounter is to support THPs in the page cache when the filesystem doesn't understand THPs. Eventually it will be removed, but we should still support filesystems which do not understand THPs yet. Move thenr_thpmanipulation functions tofilemap.hsince they're page-cache specific.Honestly I don't think we should support the read-only THP crap. We should in fact never have merged that bandaid to start with given that you did good progress on the real thing.
I'd like to see the feature ripped out again, yes. Once we have a few more filesystems converted, I think that'll be a reasonable thing to do.
It was a good step along the way; Song fixed a number of problems, and worked on other things that I never had to learn anything about (like uprobes and khugepaged). I wouldn't go so far as to say we should never have merged it, but I think we can remove it in about six months.