NFSD: filecache: add STATX_DIOALIGN and STATX_DIO_READ_ALIGN support

Use STATX_DIOALIGN and STATX_DIO_READ_ALIGN to get DIO alignment
attributes from the underlying filesystem and store them in the
associated nfsd_file. This is done when the nfsd_file is first
opened for each regular file.

Signed-off-by: Mike Snitzer <snitzer@kernel.org>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: NeilBrown <neil@brown.name>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>

Author: Mike Snitzer

fs/nfsd/filecache.c

@@ -231,6 +231,9 @@ nfsd_file_alloc(struct net *net, struct inode *inode, unsigned char need,
 	refcount_set(&nf->nf_ref, 1);
 	nf->nf_may = need;
 	nf->nf_mark = NULL;
+	nf->nf_dio_mem_align = 0;
+	nf->nf_dio_offset_align = 0;
+	nf->nf_dio_read_offset_align = 0;
 	return nf;
 }
 
@@ -1069,6 +1072,35 @@ nfsd_file_is_cached(struct inode *inode)
 	return ret;
 }
 
+static __be32
+nfsd_file_get_dio_attrs(const struct svc_fh *fhp, struct nfsd_file *nf)
+{
+	struct inode *inode = file_inode(nf->nf_file);
+	struct kstat stat;
+	__be32 status;
+
+	/* Currently only need to get DIO alignment info for regular files */
+	if (!S_ISREG(inode->i_mode))
+		return nfs_ok;
+
+	status = fh_getattr(fhp, &stat);
+	if (status != nfs_ok)
+		return status;
+
+	trace_nfsd_file_get_dio_attrs(inode, &stat);
+
+	if (stat.result_mask & STATX_DIOALIGN) {
+		nf->nf_dio_mem_align = stat.dio_mem_align;
+		nf->nf_dio_offset_align = stat.dio_offset_align;
+	}
+	if (stat.result_mask & STATX_DIO_READ_ALIGN)
+		nf->nf_dio_read_offset_align = stat.dio_read_offset_align;
+	else
+		nf->nf_dio_read_offset_align = nf->nf_dio_offset_align;
+
+	return nfs_ok;
+}
+
 static __be32
 nfsd_file_do_acquire(struct svc_rqst *rqstp, struct net *net,
 		     struct svc_cred *cred,
@@ -1187,6 +1219,8 @@ nfsd_file_do_acquire(struct svc_rqst *rqstp, struct net *net,
 			}
 			status = nfserrno(ret);
 			trace_nfsd_file_open(nf, status);
+			if (status == nfs_ok)
+				status = nfsd_file_get_dio_attrs(fhp, nf);
 		}
 	} else
 		status = nfserr_jukebox;

fs/nfsd/filecache.h

@@ -54,6 +54,10 @@ struct nfsd_file {
 	struct list_head	nf_gc;
 	struct rcu_head		nf_rcu;
 	ktime_t			nf_birthtime;
+
+	u32			nf_dio_mem_align;
+	u32			nf_dio_offset_align;
+	u32			nf_dio_read_offset_align;
 };
 
 int nfsd_file_cache_init(void);

fs/nfsd/trace.h

@@ -1133,6 +1133,33 @@ TRACE_EVENT(nfsd_file_alloc,
 	)
 );
 
+TRACE_EVENT(nfsd_file_get_dio_attrs,
+	TP_PROTO(
+		const struct inode *inode,
+		const struct kstat *stat
+	),
+	TP_ARGS(inode, stat),
+	TP_STRUCT__entry(
+		__field(const void *, inode)
+		__field(unsigned long, mask)
+		__field(u32, mem_align)
+		__field(u32, offset_align)
+		__field(u32, read_offset_align)
+	),
+	TP_fast_assign(
+		__entry->inode = inode;
+		__entry->mask = stat->result_mask;
+		__entry->mem_align = stat->dio_mem_align;
+		__entry->offset_align = stat->dio_offset_align;
+		__entry->read_offset_align = stat->dio_read_offset_align;
+	),
+	TP_printk("inode=%p flags=%s mem_align=%u offset_align=%u read_offset_align=%u",
+		__entry->inode, show_statx_mask(__entry->mask),
+		__entry->mem_align, __entry->offset_align,
+		__entry->read_offset_align
+	)
+);
+
 TRACE_EVENT(nfsd_file_acquire,
 	TP_PROTO(
 		const struct svc_rqst *rqstp,

fs/nfsd/vfs.h

@@ -185,6 +185,10 @@ static inline __be32 fh_getattr(const struct svc_fh *fh, struct kstat *stat)
 	u32 request_mask = STATX_BASIC_STATS;
 	struct path p = {.mnt = fh->fh_export->ex_path.mnt,
 			 .dentry = fh->fh_dentry};
+	struct inode *inode = d_inode(p.dentry);
+
+	if (S_ISREG(inode->i_mode))
+		request_mask |= (STATX_DIOALIGN | STATX_DIO_READ_ALIGN);
 
 	if (fh->fh_maxsize == NFS4_FHSIZE)
 		request_mask |= (STATX_BTIME | STATX_CHANGE_COOKIE);

include/trace/misc/fs.h

@@ -141,3 +141,25 @@
 		{ ATTR_TIMES_SET,	"TIMES_SET" },	\
 		{ ATTR_TOUCH,		"TOUCH"},	\
 		{ ATTR_DELEG,		"DELEG"})
+
+#define show_statx_mask(flags)					\
+	__print_flags(flags, "|",				\
+		{ STATX_TYPE,		"TYPE" },		\
+		{ STATX_MODE,		"MODE" },		\
+		{ STATX_NLINK,		"NLINK" },		\
+		{ STATX_UID,		"UID" },		\
+		{ STATX_GID,		"GID" },		\
+		{ STATX_ATIME,		"ATIME" },		\
+		{ STATX_MTIME,		"MTIME" },		\
+		{ STATX_CTIME,		"CTIME" },		\
+		{ STATX_INO,		"INO" },		\
+		{ STATX_SIZE,		"SIZE" },		\
+		{ STATX_BLOCKS,		"BLOCKS" },		\
+		{ STATX_BASIC_STATS,	"BASIC_STATS" },	\
+		{ STATX_BTIME,		"BTIME" },		\
+		{ STATX_MNT_ID,		"MNT_ID" },		\
+		{ STATX_DIOALIGN,	"DIOALIGN" },		\
+		{ STATX_MNT_ID_UNIQUE,	"MNT_ID_UNIQUE" },	\
+		{ STATX_SUBVOL,		"SUBVOL" },		\
+		{ STATX_WRITE_ATOMIC,	"WRITE_ATOMIC" },	\
+		{ STATX_DIO_READ_ALIGN,	"DIO_READ_ALIGN" })