* [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache @ 2023-04-14 10:43 ` Wu Bo 0 siblings, 0 replies; 16+ messages in thread From: Wu Bo via Linux-f2fs-devel @ 2023-04-14 10:43 UTC (permalink / raw) To: Jaegeuk Kim, Chao Yu; +Cc: wubo.oduw, linux-kernel, Wu Bo, linux-f2fs-devel It would be better to use the dedicated slab to store path. Signed-off-by: Wu Bo <bo.wu@vivo.com> --- fs/f2fs/file.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 15dabeac4690..27137873958f 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) struct inode *inode = file_inode(iocb->ki_filp); char *buf, *path; - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); + buf = __getname(); if (!buf) return; path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) trace_f2fs_dataread_start(inode, iocb->ki_pos, count, current->pid, path, current->comm); free_buf: - kfree(buf); + __putname(buf); } static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) -- 2.35.3 _______________________________________________ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 1/1] f2fs: allocate trace path buffer from names_cache @ 2023-04-14 10:43 ` Wu Bo 0 siblings, 0 replies; 16+ messages in thread From: Wu Bo @ 2023-04-14 10:43 UTC (permalink / raw) To: Jaegeuk Kim, Chao Yu; +Cc: linux-f2fs-devel, linux-kernel, wubo.oduw, Wu Bo It would be better to use the dedicated slab to store path. Signed-off-by: Wu Bo <bo.wu@vivo.com> --- fs/f2fs/file.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 15dabeac4690..27137873958f 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) struct inode *inode = file_inode(iocb->ki_filp); char *buf, *path; - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); + buf = __getname(); if (!buf) return; path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) trace_f2fs_dataread_start(inode, iocb->ki_pos, count, current->pid, path, current->comm); free_buf: - kfree(buf); + __putname(buf); } static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) -- 2.35.3 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* Re: [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache 2023-04-14 10:43 ` Wu Bo @ 2023-04-18 15:51 ` Chao Yu -1 siblings, 0 replies; 16+ messages in thread From: Chao Yu @ 2023-04-18 15:51 UTC (permalink / raw) To: Wu Bo, Jaegeuk Kim; +Cc: wubo.oduw, linux-kernel, linux-f2fs-devel On 2023/4/14 18:43, Wu Bo wrote: > It would be better to use the dedicated slab to store path. > > Signed-off-by: Wu Bo <bo.wu@vivo.com> > --- > fs/f2fs/file.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c > index 15dabeac4690..27137873958f 100644 > --- a/fs/f2fs/file.c > +++ b/fs/f2fs/file.c > @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > struct inode *inode = file_inode(iocb->ki_filp); > char *buf, *path; > > - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); > + buf = __getname(); How about: buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); > if (!buf) > return; > path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); > @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > trace_f2fs_dataread_start(inode, iocb->ki_pos, count, > current->pid, path, current->comm); > free_buf: > - kfree(buf); > + __putname(buf); kmem_cache_free(names_cachep, buf); Thanks, > } > > static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) _______________________________________________ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel ^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH 1/1] f2fs: allocate trace path buffer from names_cache @ 2023-04-18 15:51 ` Chao Yu 0 siblings, 0 replies; 16+ messages in thread From: Chao Yu @ 2023-04-18 15:51 UTC (permalink / raw) To: Wu Bo, Jaegeuk Kim; +Cc: linux-f2fs-devel, linux-kernel, wubo.oduw On 2023/4/14 18:43, Wu Bo wrote: > It would be better to use the dedicated slab to store path. > > Signed-off-by: Wu Bo <bo.wu@vivo.com> > --- > fs/f2fs/file.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c > index 15dabeac4690..27137873958f 100644 > --- a/fs/f2fs/file.c > +++ b/fs/f2fs/file.c > @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > struct inode *inode = file_inode(iocb->ki_filp); > char *buf, *path; > > - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); > + buf = __getname(); How about: buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); > if (!buf) > return; > path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); > @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > trace_f2fs_dataread_start(inode, iocb->ki_pos, count, > current->pid, path, current->comm); > free_buf: > - kfree(buf); > + __putname(buf); kmem_cache_free(names_cachep, buf); Thanks, > } > > static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) ^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache 2023-04-18 15:51 ` Chao Yu @ 2023-04-18 16:07 ` Jaegeuk Kim -1 siblings, 0 replies; 16+ messages in thread From: Jaegeuk Kim @ 2023-04-18 16:07 UTC (permalink / raw) To: Chao Yu; +Cc: wubo.oduw, linux-kernel, Wu Bo, linux-f2fs-devel On 04/18, Chao Yu wrote: > On 2023/4/14 18:43, Wu Bo wrote: > > It would be better to use the dedicated slab to store path. > > > > Signed-off-by: Wu Bo <bo.wu@vivo.com> > > --- > > fs/f2fs/file.c | 4 ++-- > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c > > index 15dabeac4690..27137873958f 100644 > > --- a/fs/f2fs/file.c > > +++ b/fs/f2fs/file.c > > @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > > struct inode *inode = file_inode(iocb->ki_filp); > > char *buf, *path; > > - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); > > + buf = __getname(); > > How about: > > buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); This looks like a hack using names_cachep? > > > if (!buf) > > return; > > path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); > > @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > > trace_f2fs_dataread_start(inode, iocb->ki_pos, count, > > current->pid, path, current->comm); > > free_buf: > > - kfree(buf); > > + __putname(buf); > > kmem_cache_free(names_cachep, buf); > > Thanks, > > > } > > static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) _______________________________________________ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel ^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH 1/1] f2fs: allocate trace path buffer from names_cache @ 2023-04-18 16:07 ` Jaegeuk Kim 0 siblings, 0 replies; 16+ messages in thread From: Jaegeuk Kim @ 2023-04-18 16:07 UTC (permalink / raw) To: Chao Yu; +Cc: Wu Bo, linux-f2fs-devel, linux-kernel, wubo.oduw On 04/18, Chao Yu wrote: > On 2023/4/14 18:43, Wu Bo wrote: > > It would be better to use the dedicated slab to store path. > > > > Signed-off-by: Wu Bo <bo.wu@vivo.com> > > --- > > fs/f2fs/file.c | 4 ++-- > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c > > index 15dabeac4690..27137873958f 100644 > > --- a/fs/f2fs/file.c > > +++ b/fs/f2fs/file.c > > @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > > struct inode *inode = file_inode(iocb->ki_filp); > > char *buf, *path; > > - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); > > + buf = __getname(); > > How about: > > buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); This looks like a hack using names_cachep? > > > if (!buf) > > return; > > path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); > > @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > > trace_f2fs_dataread_start(inode, iocb->ki_pos, count, > > current->pid, path, current->comm); > > free_buf: > > - kfree(buf); > > + __putname(buf); > > kmem_cache_free(names_cachep, buf); > > Thanks, > > > } > > static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) ^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache 2023-04-18 16:07 ` Jaegeuk Kim @ 2023-04-19 6:28 ` Chao Yu -1 siblings, 0 replies; 16+ messages in thread From: Chao Yu @ 2023-04-19 6:28 UTC (permalink / raw) To: Jaegeuk Kim; +Cc: wubo.oduw, linux-kernel, Wu Bo, linux-f2fs-devel On 2023/4/19 0:07, Jaegeuk Kim wrote: > On 04/18, Chao Yu wrote: >> On 2023/4/14 18:43, Wu Bo wrote: >>> It would be better to use the dedicated slab to store path. >>> >>> Signed-off-by: Wu Bo <bo.wu@vivo.com> >>> --- >>> fs/f2fs/file.c | 4 ++-- >>> 1 file changed, 2 insertions(+), 2 deletions(-) >>> >>> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c >>> index 15dabeac4690..27137873958f 100644 >>> --- a/fs/f2fs/file.c >>> +++ b/fs/f2fs/file.c >>> @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) >>> struct inode *inode = file_inode(iocb->ki_filp); >>> char *buf, *path; >>> - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); >>> + buf = __getname(); >> >> How about: >> >> buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); > > This looks like a hack using names_cachep? names_cachep was exported in fs.h. > Using f2fs_kmem_cache_alloc is able to inject malloc error. > But here is a trace event, is it ok to inject error in a trace path? Yes, the fail path handling is very simple, so it's fine to leave it as it is. Reviewed-by: Chao Yu <chao@kernel.org> Thanks, > >> >>> if (!buf) >>> return; >>> path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); >>> @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) >>> trace_f2fs_dataread_start(inode, iocb->ki_pos, count, >>> current->pid, path, current->comm); >>> free_buf: >>> - kfree(buf); >>> + __putname(buf); >> >> kmem_cache_free(names_cachep, buf); >> >> Thanks, >> >>> } >>> static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) _______________________________________________ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel ^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH 1/1] f2fs: allocate trace path buffer from names_cache @ 2023-04-19 6:28 ` Chao Yu 0 siblings, 0 replies; 16+ messages in thread From: Chao Yu @ 2023-04-19 6:28 UTC (permalink / raw) To: Jaegeuk Kim; +Cc: Wu Bo, linux-f2fs-devel, linux-kernel, wubo.oduw On 2023/4/19 0:07, Jaegeuk Kim wrote: > On 04/18, Chao Yu wrote: >> On 2023/4/14 18:43, Wu Bo wrote: >>> It would be better to use the dedicated slab to store path. >>> >>> Signed-off-by: Wu Bo <bo.wu@vivo.com> >>> --- >>> fs/f2fs/file.c | 4 ++-- >>> 1 file changed, 2 insertions(+), 2 deletions(-) >>> >>> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c >>> index 15dabeac4690..27137873958f 100644 >>> --- a/fs/f2fs/file.c >>> +++ b/fs/f2fs/file.c >>> @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) >>> struct inode *inode = file_inode(iocb->ki_filp); >>> char *buf, *path; >>> - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); >>> + buf = __getname(); >> >> How about: >> >> buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); > > This looks like a hack using names_cachep? names_cachep was exported in fs.h. > Using f2fs_kmem_cache_alloc is able to inject malloc error. > But here is a trace event, is it ok to inject error in a trace path? Yes, the fail path handling is very simple, so it's fine to leave it as it is. Reviewed-by: Chao Yu <chao@kernel.org> Thanks, > >> >>> if (!buf) >>> return; >>> path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); >>> @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) >>> trace_f2fs_dataread_start(inode, iocb->ki_pos, count, >>> current->pid, path, current->comm); >>> free_buf: >>> - kfree(buf); >>> + __putname(buf); >> >> kmem_cache_free(names_cachep, buf); >> >> Thanks, >> >>> } >>> static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) ^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache 2023-04-19 6:28 ` Chao Yu @ 2023-04-19 18:18 ` Jaegeuk Kim -1 siblings, 0 replies; 16+ messages in thread From: Jaegeuk Kim @ 2023-04-19 18:18 UTC (permalink / raw) To: Chao Yu; +Cc: wubo.oduw, linux-kernel, Wu Bo, linux-f2fs-devel On 04/19, Chao Yu wrote: > On 2023/4/19 0:07, Jaegeuk Kim wrote: > > On 04/18, Chao Yu wrote: > > > On 2023/4/14 18:43, Wu Bo wrote: > > > > It would be better to use the dedicated slab to store path. > > > > > > > > Signed-off-by: Wu Bo <bo.wu@vivo.com> > > > > --- > > > > fs/f2fs/file.c | 4 ++-- > > > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > > > > > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c > > > > index 15dabeac4690..27137873958f 100644 > > > > --- a/fs/f2fs/file.c > > > > +++ b/fs/f2fs/file.c > > > > @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > > > > struct inode *inode = file_inode(iocb->ki_filp); > > > > char *buf, *path; > > > > - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); > > > > + buf = __getname(); > > > > > > How about: > > > > > > buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); > > > > This looks like a hack using names_cachep? > > names_cachep was exported in fs.h. I think that's for __getname() in general, which doesn't indicate you can hack. No one is using like that. $ grep names_cachep fs/* -R fs/dcache.c:struct kmem_cache *names_cachep __read_mostly; fs/dcache.c:EXPORT_SYMBOL(names_cachep); fs/dcache.c: names_cachep = kmem_cache_create_usercopy("names_cache", PATH_MAX, 0, $ grep __getname fs/* -R fs/ceph/mds_client.c: path = __getname(); fs/cifs/cifsproto.h: return __getname(); fs/dcache.c:/* SLAB cache for __getname() consumers */ fs/d_path.c: char *page = __getname(); fs/exfat/dir.c: nb->lfn = __getname(); fs/f2fs/file.c: buf = __getname(); fs/fat/dir.c: *unicode = __getname(); fs/fat/namei_vfat.c: uname = __getname(); fs/hostfs/hostfs_kern.c: char *name = __getname(); fs/namei.c: result = __getname(); fs/namei.c: result = __getname(); fs/ntfs3/dir.c: name = __getname(); fs/ntfs3/xattr.c: buf = __getname(); fs/ntfs3/inode.c: new_de = __getname(); fs/ntfs3/inode.c: de = __getname(); fs/ntfs3/inode.c: de = __getname(); fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); fs/ntfs3/namei.c: de = __getname(); fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); fs/ntfs3/namei.c: uni = __getname(); fs/ntfs3/namei.c: uni1 = __getname(); fs/vboxsf/utils.c: * Returns a shfl_string allocated through __getname (must be freed using fs/vboxsf/utils.c: buf = __getname(); fs/vboxsf/utils.c: shfl_path = __getname(); > > > Using f2fs_kmem_cache_alloc is able to inject malloc error. > > But here is a trace event, is it ok to inject error in a trace path? > > Yes, the fail path handling is very simple, so it's fine to leave it > as it is. > > Reviewed-by: Chao Yu <chao@kernel.org> What is this for? > > Thanks, > > > > > > > > > > if (!buf) > > > > return; > > > > path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); > > > > @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > > > > trace_f2fs_dataread_start(inode, iocb->ki_pos, count, > > > > current->pid, path, current->comm); > > > > free_buf: > > > > - kfree(buf); > > > > + __putname(buf); > > > > > > kmem_cache_free(names_cachep, buf); > > > > > > Thanks, > > > > > > > } > > > > static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) _______________________________________________ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel ^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH 1/1] f2fs: allocate trace path buffer from names_cache @ 2023-04-19 18:18 ` Jaegeuk Kim 0 siblings, 0 replies; 16+ messages in thread From: Jaegeuk Kim @ 2023-04-19 18:18 UTC (permalink / raw) To: Chao Yu; +Cc: Wu Bo, linux-f2fs-devel, linux-kernel, wubo.oduw On 04/19, Chao Yu wrote: > On 2023/4/19 0:07, Jaegeuk Kim wrote: > > On 04/18, Chao Yu wrote: > > > On 2023/4/14 18:43, Wu Bo wrote: > > > > It would be better to use the dedicated slab to store path. > > > > > > > > Signed-off-by: Wu Bo <bo.wu@vivo.com> > > > > --- > > > > fs/f2fs/file.c | 4 ++-- > > > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > > > > > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c > > > > index 15dabeac4690..27137873958f 100644 > > > > --- a/fs/f2fs/file.c > > > > +++ b/fs/f2fs/file.c > > > > @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > > > > struct inode *inode = file_inode(iocb->ki_filp); > > > > char *buf, *path; > > > > - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); > > > > + buf = __getname(); > > > > > > How about: > > > > > > buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); > > > > This looks like a hack using names_cachep? > > names_cachep was exported in fs.h. I think that's for __getname() in general, which doesn't indicate you can hack. No one is using like that. $ grep names_cachep fs/* -R fs/dcache.c:struct kmem_cache *names_cachep __read_mostly; fs/dcache.c:EXPORT_SYMBOL(names_cachep); fs/dcache.c: names_cachep = kmem_cache_create_usercopy("names_cache", PATH_MAX, 0, $ grep __getname fs/* -R fs/ceph/mds_client.c: path = __getname(); fs/cifs/cifsproto.h: return __getname(); fs/dcache.c:/* SLAB cache for __getname() consumers */ fs/d_path.c: char *page = __getname(); fs/exfat/dir.c: nb->lfn = __getname(); fs/f2fs/file.c: buf = __getname(); fs/fat/dir.c: *unicode = __getname(); fs/fat/namei_vfat.c: uname = __getname(); fs/hostfs/hostfs_kern.c: char *name = __getname(); fs/namei.c: result = __getname(); fs/namei.c: result = __getname(); fs/ntfs3/dir.c: name = __getname(); fs/ntfs3/xattr.c: buf = __getname(); fs/ntfs3/inode.c: new_de = __getname(); fs/ntfs3/inode.c: de = __getname(); fs/ntfs3/inode.c: de = __getname(); fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); fs/ntfs3/namei.c: de = __getname(); fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); fs/ntfs3/namei.c: uni = __getname(); fs/ntfs3/namei.c: uni1 = __getname(); fs/vboxsf/utils.c: * Returns a shfl_string allocated through __getname (must be freed using fs/vboxsf/utils.c: buf = __getname(); fs/vboxsf/utils.c: shfl_path = __getname(); > > > Using f2fs_kmem_cache_alloc is able to inject malloc error. > > But here is a trace event, is it ok to inject error in a trace path? > > Yes, the fail path handling is very simple, so it's fine to leave it > as it is. > > Reviewed-by: Chao Yu <chao@kernel.org> What is this for? > > Thanks, > > > > > > > > > > if (!buf) > > > > return; > > > > path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); > > > > @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > > > > trace_f2fs_dataread_start(inode, iocb->ki_pos, count, > > > > current->pid, path, current->comm); > > > > free_buf: > > > > - kfree(buf); > > > > + __putname(buf); > > > > > > kmem_cache_free(names_cachep, buf); > > > > > > Thanks, > > > > > > > } > > > > static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) ^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache 2023-04-19 18:18 ` Jaegeuk Kim @ 2023-04-19 19:45 ` Jaegeuk Kim -1 siblings, 0 replies; 16+ messages in thread From: Jaegeuk Kim @ 2023-04-19 19:45 UTC (permalink / raw) To: Chao Yu; +Cc: wubo.oduw, linux-kernel, Wu Bo, linux-f2fs-devel On 04/19, Jaegeuk Kim wrote: > On 04/19, Chao Yu wrote: > > On 2023/4/19 0:07, Jaegeuk Kim wrote: > > > On 04/18, Chao Yu wrote: > > > > On 2023/4/14 18:43, Wu Bo wrote: > > > > > It would be better to use the dedicated slab to store path. > > > > > > > > > > Signed-off-by: Wu Bo <bo.wu@vivo.com> > > > > > --- > > > > > fs/f2fs/file.c | 4 ++-- > > > > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > > > > > > > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c > > > > > index 15dabeac4690..27137873958f 100644 > > > > > --- a/fs/f2fs/file.c > > > > > +++ b/fs/f2fs/file.c > > > > > @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > > > > > struct inode *inode = file_inode(iocb->ki_filp); > > > > > char *buf, *path; > > > > > - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); > > > > > + buf = __getname(); > > > > > > > > How about: > > > > > > > > buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); > > > > > > This looks like a hack using names_cachep? > > > > names_cachep was exported in fs.h. > > I think that's for __getname() in general, which doesn't indicate you can hack. > No one is using like that. > > $ grep names_cachep fs/* -R > fs/dcache.c:struct kmem_cache *names_cachep __read_mostly; > fs/dcache.c:EXPORT_SYMBOL(names_cachep); > fs/dcache.c: names_cachep = kmem_cache_create_usercopy("names_cache", PATH_MAX, 0, > > $ grep __getname fs/* -R > fs/ceph/mds_client.c: path = __getname(); > fs/cifs/cifsproto.h: return __getname(); > fs/dcache.c:/* SLAB cache for __getname() consumers */ > fs/d_path.c: char *page = __getname(); > fs/exfat/dir.c: nb->lfn = __getname(); > fs/f2fs/file.c: buf = __getname(); > fs/fat/dir.c: *unicode = __getname(); > fs/fat/namei_vfat.c: uname = __getname(); > fs/hostfs/hostfs_kern.c: char *name = __getname(); > fs/namei.c: result = __getname(); > fs/namei.c: result = __getname(); > fs/ntfs3/dir.c: name = __getname(); > fs/ntfs3/xattr.c: buf = __getname(); > fs/ntfs3/inode.c: new_de = __getname(); > fs/ntfs3/inode.c: de = __getname(); > fs/ntfs3/inode.c: de = __getname(); > fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); > fs/ntfs3/namei.c: de = __getname(); > fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); > fs/ntfs3/namei.c: uni = __getname(); > fs/ntfs3/namei.c: uni1 = __getname(); > fs/vboxsf/utils.c: * Returns a shfl_string allocated through __getname (must be freed using > fs/vboxsf/utils.c: buf = __getname(); > fs/vboxsf/utils.c: shfl_path = __getname(); > > > > > > Using f2fs_kmem_cache_alloc is able to inject malloc error. > > > But here is a trace event, is it ok to inject error in a trace path? > > > > Yes, the fail path handling is very simple, so it's fine to leave it > > as it is. > > > > Reviewed-by: Chao Yu <chao@kernel.org> > > What is this for? > If we want to keep the error injection, how about this? Signed-off-by: Wu Bo <bo.wu@vivo.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org> --- fs/f2fs/f2fs.h | 13 +++++++++++++ fs/f2fs/file.c | 4 ++-- 2 files changed, 15 insertions(+), 2 deletions(-) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 6cae94d51821..d87044516fe9 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -3347,6 +3347,19 @@ static inline void *f2fs_kmalloc(struct f2fs_sb_info *sbi, return kmalloc(size, flags); } +static inline void *f2fs_getname(struct f2fs_sb_info *sbi) +{ + if (time_to_inject(sbi, FAULT_KMALLOC)) + return NULL; + + return __getname(); +} + +static inline void f2fs_putname(char *buf) +{ + __putname(buf); +} + static inline void *f2fs_kzalloc(struct f2fs_sb_info *sbi, size_t size, gfp_t flags) { diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 1b4411271f54..5ac53d2627d2 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -4372,7 +4372,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) struct inode *inode = file_inode(iocb->ki_filp); char *buf, *path; - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); + buf = f2fs_getname(F2FS_I_SB(inode)); if (!buf) return; path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); @@ -4385,7 +4385,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) trace_f2fs_dataread_start(inode, iocb->ki_pos, count, current->pid, path, current->comm); free_buf: - kfree(buf); + f2fs_putname(buf); } static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) -- 2.40.0.634.g4ca3ef3211-goog _______________________________________________ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel ^ permalink raw reply related [flat|nested] 16+ messages in thread
* Re: [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache @ 2023-04-19 19:45 ` Jaegeuk Kim 0 siblings, 0 replies; 16+ messages in thread From: Jaegeuk Kim @ 2023-04-19 19:45 UTC (permalink / raw) To: Chao Yu; +Cc: wubo.oduw, linux-kernel, Wu Bo, linux-f2fs-devel On 04/19, Jaegeuk Kim wrote: > On 04/19, Chao Yu wrote: > > On 2023/4/19 0:07, Jaegeuk Kim wrote: > > > On 04/18, Chao Yu wrote: > > > > On 2023/4/14 18:43, Wu Bo wrote: > > > > > It would be better to use the dedicated slab to store path. > > > > > > > > > > Signed-off-by: Wu Bo <bo.wu@vivo.com> > > > > > --- > > > > > fs/f2fs/file.c | 4 ++-- > > > > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > > > > > > > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c > > > > > index 15dabeac4690..27137873958f 100644 > > > > > --- a/fs/f2fs/file.c > > > > > +++ b/fs/f2fs/file.c > > > > > @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > > > > > struct inode *inode = file_inode(iocb->ki_filp); > > > > > char *buf, *path; > > > > > - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); > > > > > + buf = __getname(); > > > > > > > > How about: > > > > > > > > buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); > > > > > > This looks like a hack using names_cachep? > > > > names_cachep was exported in fs.h. > > I think that's for __getname() in general, which doesn't indicate you can hack. > No one is using like that. > > $ grep names_cachep fs/* -R > fs/dcache.c:struct kmem_cache *names_cachep __read_mostly; > fs/dcache.c:EXPORT_SYMBOL(names_cachep); > fs/dcache.c: names_cachep = kmem_cache_create_usercopy("names_cache", PATH_MAX, 0, > > $ grep __getname fs/* -R > fs/ceph/mds_client.c: path = __getname(); > fs/cifs/cifsproto.h: return __getname(); > fs/dcache.c:/* SLAB cache for __getname() consumers */ > fs/d_path.c: char *page = __getname(); > fs/exfat/dir.c: nb->lfn = __getname(); > fs/f2fs/file.c: buf = __getname(); > fs/fat/dir.c: *unicode = __getname(); > fs/fat/namei_vfat.c: uname = __getname(); > fs/hostfs/hostfs_kern.c: char *name = __getname(); > fs/namei.c: result = __getname(); > fs/namei.c: result = __getname(); > fs/ntfs3/dir.c: name = __getname(); > fs/ntfs3/xattr.c: buf = __getname(); > fs/ntfs3/inode.c: new_de = __getname(); > fs/ntfs3/inode.c: de = __getname(); > fs/ntfs3/inode.c: de = __getname(); > fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); > fs/ntfs3/namei.c: de = __getname(); > fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); > fs/ntfs3/namei.c: uni = __getname(); > fs/ntfs3/namei.c: uni1 = __getname(); > fs/vboxsf/utils.c: * Returns a shfl_string allocated through __getname (must be freed using > fs/vboxsf/utils.c: buf = __getname(); > fs/vboxsf/utils.c: shfl_path = __getname(); > > > > > > Using f2fs_kmem_cache_alloc is able to inject malloc error. > > > But here is a trace event, is it ok to inject error in a trace path? > > > > Yes, the fail path handling is very simple, so it's fine to leave it > > as it is. > > > > Reviewed-by: Chao Yu <chao@kernel.org> > > What is this for? > If we want to keep the error injection, how about this? Signed-off-by: Wu Bo <bo.wu@vivo.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org> --- fs/f2fs/f2fs.h | 13 +++++++++++++ fs/f2fs/file.c | 4 ++-- 2 files changed, 15 insertions(+), 2 deletions(-) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 6cae94d51821..d87044516fe9 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -3347,6 +3347,19 @@ static inline void *f2fs_kmalloc(struct f2fs_sb_info *sbi, return kmalloc(size, flags); } +static inline void *f2fs_getname(struct f2fs_sb_info *sbi) +{ + if (time_to_inject(sbi, FAULT_KMALLOC)) + return NULL; + + return __getname(); +} + +static inline void f2fs_putname(char *buf) +{ + __putname(buf); +} + static inline void *f2fs_kzalloc(struct f2fs_sb_info *sbi, size_t size, gfp_t flags) { diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 1b4411271f54..5ac53d2627d2 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -4372,7 +4372,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) struct inode *inode = file_inode(iocb->ki_filp); char *buf, *path; - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); + buf = f2fs_getname(F2FS_I_SB(inode)); if (!buf) return; path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); @@ -4385,7 +4385,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) trace_f2fs_dataread_start(inode, iocb->ki_pos, count, current->pid, path, current->comm); free_buf: - kfree(buf); + f2fs_putname(buf); } static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) -- 2.40.0.634.g4ca3ef3211-goog ^ permalink raw reply related [flat|nested] 16+ messages in thread
* Re: [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache 2023-04-19 19:45 ` Jaegeuk Kim @ 2023-04-20 1:10 ` Chao Yu -1 siblings, 0 replies; 16+ messages in thread From: Chao Yu @ 2023-04-20 1:10 UTC (permalink / raw) To: Jaegeuk Kim; +Cc: wubo.oduw, linux-kernel, Wu Bo, linux-f2fs-devel On 2023/4/20 3:45, Jaegeuk Kim wrote: > On 04/19, Jaegeuk Kim wrote: >> On 04/19, Chao Yu wrote: >>> On 2023/4/19 0:07, Jaegeuk Kim wrote: >>>> On 04/18, Chao Yu wrote: >>>>> On 2023/4/14 18:43, Wu Bo wrote: >>>>>> It would be better to use the dedicated slab to store path. >>>>>> >>>>>> Signed-off-by: Wu Bo <bo.wu@vivo.com> >>>>>> --- >>>>>> fs/f2fs/file.c | 4 ++-- >>>>>> 1 file changed, 2 insertions(+), 2 deletions(-) >>>>>> >>>>>> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c >>>>>> index 15dabeac4690..27137873958f 100644 >>>>>> --- a/fs/f2fs/file.c >>>>>> +++ b/fs/f2fs/file.c >>>>>> @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) >>>>>> struct inode *inode = file_inode(iocb->ki_filp); >>>>>> char *buf, *path; >>>>>> - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); >>>>>> + buf = __getname(); >>>>> >>>>> How about: >>>>> >>>>> buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); >>>> >>>> This looks like a hack using names_cachep? >>> >>> names_cachep was exported in fs.h. >> >> I think that's for __getname() in general, which doesn't indicate you can hack. >> No one is using like that. >> >> $ grep names_cachep fs/* -R >> fs/dcache.c:struct kmem_cache *names_cachep __read_mostly; >> fs/dcache.c:EXPORT_SYMBOL(names_cachep); >> fs/dcache.c: names_cachep = kmem_cache_create_usercopy("names_cache", PATH_MAX, 0, >> >> $ grep __getname fs/* -R >> fs/ceph/mds_client.c: path = __getname(); >> fs/cifs/cifsproto.h: return __getname(); >> fs/dcache.c:/* SLAB cache for __getname() consumers */ >> fs/d_path.c: char *page = __getname(); >> fs/exfat/dir.c: nb->lfn = __getname(); >> fs/f2fs/file.c: buf = __getname(); >> fs/fat/dir.c: *unicode = __getname(); >> fs/fat/namei_vfat.c: uname = __getname(); >> fs/hostfs/hostfs_kern.c: char *name = __getname(); >> fs/namei.c: result = __getname(); >> fs/namei.c: result = __getname(); >> fs/ntfs3/dir.c: name = __getname(); >> fs/ntfs3/xattr.c: buf = __getname(); >> fs/ntfs3/inode.c: new_de = __getname(); >> fs/ntfs3/inode.c: de = __getname(); >> fs/ntfs3/inode.c: de = __getname(); >> fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); >> fs/ntfs3/namei.c: de = __getname(); >> fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); >> fs/ntfs3/namei.c: uni = __getname(); >> fs/ntfs3/namei.c: uni1 = __getname(); >> fs/vboxsf/utils.c: * Returns a shfl_string allocated through __getname (must be freed using >> fs/vboxsf/utils.c: buf = __getname(); >> fs/vboxsf/utils.c: shfl_path = __getname(); >> >>> >>>> Using f2fs_kmem_cache_alloc is able to inject malloc error. >>>> But here is a trace event, is it ok to inject error in a trace path? >>> >>> Yes, the fail path handling is very simple, so it's fine to leave it >>> as it is. >>> >>> Reviewed-by: Chao Yu <chao@kernel.org> >> >> What is this for? Oh, I mean I'm okay w/ original patch, because f2fs_trace_rw_file_path() doesn't have complicated error handling. >> > > If we want to keep the error injection, how about this? Both original patch or below patch w/ fault injection is fine to me. Free feel to add rvb tag of me. :) Thanks, > > Signed-off-by: Wu Bo <bo.wu@vivo.com> > Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org> > --- > fs/f2fs/f2fs.h | 13 +++++++++++++ > fs/f2fs/file.c | 4 ++-- > 2 files changed, 15 insertions(+), 2 deletions(-) > > diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h > index 6cae94d51821..d87044516fe9 100644 > --- a/fs/f2fs/f2fs.h > +++ b/fs/f2fs/f2fs.h > @@ -3347,6 +3347,19 @@ static inline void *f2fs_kmalloc(struct f2fs_sb_info *sbi, > return kmalloc(size, flags); > } > > +static inline void *f2fs_getname(struct f2fs_sb_info *sbi) > +{ > + if (time_to_inject(sbi, FAULT_KMALLOC)) > + return NULL; > + > + return __getname(); > +} > + > +static inline void f2fs_putname(char *buf) > +{ > + __putname(buf); > +} > + > static inline void *f2fs_kzalloc(struct f2fs_sb_info *sbi, > size_t size, gfp_t flags) > { > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c > index 1b4411271f54..5ac53d2627d2 100644 > --- a/fs/f2fs/file.c > +++ b/fs/f2fs/file.c > @@ -4372,7 +4372,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > struct inode *inode = file_inode(iocb->ki_filp); > char *buf, *path; > > - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); > + buf = f2fs_getname(F2FS_I_SB(inode)); > if (!buf) > return; > path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); > @@ -4385,7 +4385,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > trace_f2fs_dataread_start(inode, iocb->ki_pos, count, > current->pid, path, current->comm); > free_buf: > - kfree(buf); > + f2fs_putname(buf); > } > > static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) _______________________________________________ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel ^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache @ 2023-04-20 1:10 ` Chao Yu 0 siblings, 0 replies; 16+ messages in thread From: Chao Yu @ 2023-04-20 1:10 UTC (permalink / raw) To: Jaegeuk Kim; +Cc: wubo.oduw, linux-kernel, Wu Bo, linux-f2fs-devel On 2023/4/20 3:45, Jaegeuk Kim wrote: > On 04/19, Jaegeuk Kim wrote: >> On 04/19, Chao Yu wrote: >>> On 2023/4/19 0:07, Jaegeuk Kim wrote: >>>> On 04/18, Chao Yu wrote: >>>>> On 2023/4/14 18:43, Wu Bo wrote: >>>>>> It would be better to use the dedicated slab to store path. >>>>>> >>>>>> Signed-off-by: Wu Bo <bo.wu@vivo.com> >>>>>> --- >>>>>> fs/f2fs/file.c | 4 ++-- >>>>>> 1 file changed, 2 insertions(+), 2 deletions(-) >>>>>> >>>>>> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c >>>>>> index 15dabeac4690..27137873958f 100644 >>>>>> --- a/fs/f2fs/file.c >>>>>> +++ b/fs/f2fs/file.c >>>>>> @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) >>>>>> struct inode *inode = file_inode(iocb->ki_filp); >>>>>> char *buf, *path; >>>>>> - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); >>>>>> + buf = __getname(); >>>>> >>>>> How about: >>>>> >>>>> buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, F2FS_I_SB(inode)); >>>> >>>> This looks like a hack using names_cachep? >>> >>> names_cachep was exported in fs.h. >> >> I think that's for __getname() in general, which doesn't indicate you can hack. >> No one is using like that. >> >> $ grep names_cachep fs/* -R >> fs/dcache.c:struct kmem_cache *names_cachep __read_mostly; >> fs/dcache.c:EXPORT_SYMBOL(names_cachep); >> fs/dcache.c: names_cachep = kmem_cache_create_usercopy("names_cache", PATH_MAX, 0, >> >> $ grep __getname fs/* -R >> fs/ceph/mds_client.c: path = __getname(); >> fs/cifs/cifsproto.h: return __getname(); >> fs/dcache.c:/* SLAB cache for __getname() consumers */ >> fs/d_path.c: char *page = __getname(); >> fs/exfat/dir.c: nb->lfn = __getname(); >> fs/f2fs/file.c: buf = __getname(); >> fs/fat/dir.c: *unicode = __getname(); >> fs/fat/namei_vfat.c: uname = __getname(); >> fs/hostfs/hostfs_kern.c: char *name = __getname(); >> fs/namei.c: result = __getname(); >> fs/namei.c: result = __getname(); >> fs/ntfs3/dir.c: name = __getname(); >> fs/ntfs3/xattr.c: buf = __getname(); >> fs/ntfs3/inode.c: new_de = __getname(); >> fs/ntfs3/inode.c: de = __getname(); >> fs/ntfs3/inode.c: de = __getname(); >> fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); >> fs/ntfs3/namei.c: de = __getname(); >> fs/ntfs3/namei.c: struct cpu_str *uni = __getname(); >> fs/ntfs3/namei.c: uni = __getname(); >> fs/ntfs3/namei.c: uni1 = __getname(); >> fs/vboxsf/utils.c: * Returns a shfl_string allocated through __getname (must be freed using >> fs/vboxsf/utils.c: buf = __getname(); >> fs/vboxsf/utils.c: shfl_path = __getname(); >> >>> >>>> Using f2fs_kmem_cache_alloc is able to inject malloc error. >>>> But here is a trace event, is it ok to inject error in a trace path? >>> >>> Yes, the fail path handling is very simple, so it's fine to leave it >>> as it is. >>> >>> Reviewed-by: Chao Yu <chao@kernel.org> >> >> What is this for? Oh, I mean I'm okay w/ original patch, because f2fs_trace_rw_file_path() doesn't have complicated error handling. >> > > If we want to keep the error injection, how about this? Both original patch or below patch w/ fault injection is fine to me. Free feel to add rvb tag of me. :) Thanks, > > Signed-off-by: Wu Bo <bo.wu@vivo.com> > Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org> > --- > fs/f2fs/f2fs.h | 13 +++++++++++++ > fs/f2fs/file.c | 4 ++-- > 2 files changed, 15 insertions(+), 2 deletions(-) > > diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h > index 6cae94d51821..d87044516fe9 100644 > --- a/fs/f2fs/f2fs.h > +++ b/fs/f2fs/f2fs.h > @@ -3347,6 +3347,19 @@ static inline void *f2fs_kmalloc(struct f2fs_sb_info *sbi, > return kmalloc(size, flags); > } > > +static inline void *f2fs_getname(struct f2fs_sb_info *sbi) > +{ > + if (time_to_inject(sbi, FAULT_KMALLOC)) > + return NULL; > + > + return __getname(); > +} > + > +static inline void f2fs_putname(char *buf) > +{ > + __putname(buf); > +} > + > static inline void *f2fs_kzalloc(struct f2fs_sb_info *sbi, > size_t size, gfp_t flags) > { > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c > index 1b4411271f54..5ac53d2627d2 100644 > --- a/fs/f2fs/file.c > +++ b/fs/f2fs/file.c > @@ -4372,7 +4372,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > struct inode *inode = file_inode(iocb->ki_filp); > char *buf, *path; > > - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); > + buf = f2fs_getname(F2FS_I_SB(inode)); > if (!buf) > return; > path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); > @@ -4385,7 +4385,7 @@ static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw) > trace_f2fs_dataread_start(inode, iocb->ki_pos, count, > current->pid, path, current->comm); > free_buf: > - kfree(buf); > + f2fs_putname(buf); > } > > static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) ^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache 2023-04-18 15:51 ` Chao Yu @ 2023-04-19 1:32 ` Wu Bo -1 siblings, 0 replies; 16+ messages in thread From: Wu Bo via Linux-f2fs-devel @ 2023-04-19 1:32 UTC (permalink / raw) To: chao; +Cc: wubo.oduw, jaegeuk, linux-kernel, bo.wu, linux-f2fs-devel On 2023/4/18 23:51, Chao Yu wrote: > On 2023/4/14 18:43, Wu Bo wrote: >> It would be better to use the dedicated slab to store path. >> >> Signed-off-by: Wu Bo <bo.wu@vivo.com> >> --- >> fs/f2fs/file.c | 4 ++-- >> 1 file changed, 2 insertions(+), 2 deletions(-) >> >> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c >> index 15dabeac4690..27137873958f 100644 >> --- a/fs/f2fs/file.c >> +++ b/fs/f2fs/file.c >> @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct >> kiocb *iocb, size_t count, int rw) >> struct inode *inode = file_inode(iocb->ki_filp); >> char *buf, *path; >> - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); >> + buf = __getname(); > > How about: > > buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, > F2FS_I_SB(inode)); Using f2fs_kmem_cache_alloc is able to inject malloc error. But here is a trace event, is it ok to inject error in a trace path? > >> if (!buf) >> return; >> path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); >> @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct >> kiocb *iocb, size_t count, int rw) >> trace_f2fs_dataread_start(inode, iocb->ki_pos, count, >> current->pid, path, current->comm); >> free_buf: >> - kfree(buf); >> + __putname(buf); > > kmem_cache_free(names_cachep, buf); > > Thanks, > >> } >> static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct >> iov_iter *to) > > > _______________________________________________ > Linux-f2fs-devel mailing list > Linux-f2fs-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel > _______________________________________________ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel ^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache @ 2023-04-19 1:32 ` Wu Bo 0 siblings, 0 replies; 16+ messages in thread From: Wu Bo @ 2023-04-19 1:32 UTC (permalink / raw) To: chao; +Cc: bo.wu, jaegeuk, linux-f2fs-devel, linux-kernel, wubo.oduw On 2023/4/18 23:51, Chao Yu wrote: > On 2023/4/14 18:43, Wu Bo wrote: >> It would be better to use the dedicated slab to store path. >> >> Signed-off-by: Wu Bo <bo.wu@vivo.com> >> --- >> fs/f2fs/file.c | 4 ++-- >> 1 file changed, 2 insertions(+), 2 deletions(-) >> >> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c >> index 15dabeac4690..27137873958f 100644 >> --- a/fs/f2fs/file.c >> +++ b/fs/f2fs/file.c >> @@ -4361,7 +4361,7 @@ static void f2fs_trace_rw_file_path(struct >> kiocb *iocb, size_t count, int rw) >> struct inode *inode = file_inode(iocb->ki_filp); >> char *buf, *path; >> - buf = f2fs_kmalloc(F2FS_I_SB(inode), PATH_MAX, GFP_KERNEL); >> + buf = __getname(); > > How about: > > buf = f2fs_kmem_cache_alloc(names_cachep, GFP_KERNEL, NULL, > F2FS_I_SB(inode)); Using f2fs_kmem_cache_alloc is able to inject malloc error. But here is a trace event, is it ok to inject error in a trace path? > >> if (!buf) >> return; >> path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX); >> @@ -4374,7 +4374,7 @@ static void f2fs_trace_rw_file_path(struct >> kiocb *iocb, size_t count, int rw) >> trace_f2fs_dataread_start(inode, iocb->ki_pos, count, >> current->pid, path, current->comm); >> free_buf: >> - kfree(buf); >> + __putname(buf); > > kmem_cache_free(names_cachep, buf); > > Thanks, > >> } >> static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct >> iov_iter *to) > > > _______________________________________________ > Linux-f2fs-devel mailing list > Linux-f2fs-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel > ^ permalink raw reply [flat|nested] 16+ messages in thread
end of thread, other threads:[~2023-04-20 1:10 UTC | newest] Thread overview: 16+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2023-04-14 10:43 [f2fs-dev] [PATCH 1/1] f2fs: allocate trace path buffer from names_cache Wu Bo via Linux-f2fs-devel 2023-04-14 10:43 ` Wu Bo 2023-04-18 15:51 ` [f2fs-dev] " Chao Yu 2023-04-18 15:51 ` Chao Yu 2023-04-18 16:07 ` [f2fs-dev] " Jaegeuk Kim 2023-04-18 16:07 ` Jaegeuk Kim 2023-04-19 6:28 ` [f2fs-dev] " Chao Yu 2023-04-19 6:28 ` Chao Yu 2023-04-19 18:18 ` [f2fs-dev] " Jaegeuk Kim 2023-04-19 18:18 ` Jaegeuk Kim 2023-04-19 19:45 ` [f2fs-dev] " Jaegeuk Kim 2023-04-19 19:45 ` Jaegeuk Kim 2023-04-20 1:10 ` Chao Yu 2023-04-20 1:10 ` Chao Yu 2023-04-19 1:32 ` Wu Bo via Linux-f2fs-devel 2023-04-19 1:32 ` Wu Bo
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.