Merge tag 'xfs-5.3-merge-12' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

Pull xfs updates from Darrick Wong:
 "In this release there are a significant amounts of consolidations and
  cleanups in the log code; restructuring of the log to issue struct
  bios directly; new bulkstat ioctls to return v5 fs inode information
  (and fix all the padding problems of the old ioctl); the beginnings of
  multithreaded inode walks (e.g. quotacheck); and a reduction in memory
  usage in the online scrub code leading to reduced runtimes.

   - Refactor inode geometry calculation into a single structure instead
     of open-coding pieces everywhere.

   - Add online repair to build options.

   - Remove unnecessary function call flags and functions.

   - Claim maintainership of various loose xfs documentation and header
     files.

   - Use struct bio directly for log buffer IOs instead of struct
     xfs_buf.

   - Reduce log item boilerplate code requirements.

   - Merge log item code spread across too many files.

   - Further distinguish between log item commits and cancellations.

   - Various small cleanups to the ag small allocator.

   - Support cgroup-aware writeback

   - libxfs refactoring for mkfs cleanup

   - Remove unneeded #includes

   - Fix a memory allocation miscalculation in the new log bio code

   - Fix bisection problems

   - Fix a crash in ioend processing caused by tripping over freeing of
     preallocated transactions

   - Split out a generic inode walk mechanism from the bulkstat code,
     hook up all the internal users to use the walking code, then clean
     up bulkstat to serve only the bulkstat ioctls.

   - Add a multithreaded iwalk implementation to speed up quotacheck on
     fast storage with many CPUs.

   - Remove unnecessary return values in logging teardown functions.

   - Supplement the bstat and inogrp structures with new bulkstat and
     inumbers structures that have all the fields we need for v5
     filesystem features and none of the padding problems of their
     predecessors.

   - Wire up new ioctls that use the new structures with a much simpler
     bulk_ireq structure at the head instead of the pointerhappy mess we
     had before.

   - Enable userspace to constrain bulkstat returns to a single AG or a
     single special inode so that we can phase out a lot of geometry
     guesswork in userspace.

   - Reduce memory consumption and zeroing overhead in extended
     attribute scrub code.

   - Fix some behavioral regressions in the new bulkstat backend code.

   - Fix some behavioral regressions in the new log bio code"

* tag 'xfs-5.3-merge-12' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: (100 commits)
  xfs: chain bios the right way around in xfs_rw_bdev
  xfs: bump INUMBERS cursor correctly in xfs_inumbers_walk
  xfs: don't update lastino for FSBULKSTAT_SINGLE
  xfs: online scrub needn't bother zeroing its temporary buffer
  xfs: only allocate memory for scrubbing attributes when we need it
  xfs: refactor attr scrub memory allocation function
  xfs: refactor extended attribute buffer pointer functions
  xfs: attribute scrub should use seen_enough to pass error values
  xfs: allow single bulkstat of special inodes
  xfs: specify AG in bulk req
  xfs: wire up the v5 inumbers ioctl
  xfs: wire up new v5 bulkstat ioctls
  xfs: introduce v5 inode group structure
  xfs: introduce new v5 bulkstat structure
  xfs: rename bulkstat functions
  xfs: remove various bulk request typedef usage
  fs: xfs: xfs_log: Change return type from int to void
  xfs: poll waiting for quotacheck
  xfs: multithreaded iwalk implementation
  xfs: refactor INUMBERS to use iwalk functions
  ...

fs/xfs/xfs_ioctl.c

@@ -11,9 +11,8 @@
 #include "xfs_trans_resv.h"
 #include "xfs_mount.h"
 #include "xfs_inode.h"
-#include "xfs_ioctl.h"
-#include "xfs_alloc.h"
 #include "xfs_rtalloc.h"
+#include "xfs_iwalk.h"
 #include "xfs_itable.h"
 #include "xfs_error.h"
 #include "xfs_attr.h"
@@ -25,7 +24,6 @@
 #include "xfs_export.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
-#include "xfs_symlink.h"
 #include "xfs_trans.h"
 #include "xfs_acl.h"
 #include "xfs_btree.h"
@@ -36,14 +34,8 @@
 #include "xfs_ag.h"
 #include "xfs_health.h"
 
-#include <linux/capability.h>
-#include <linux/cred.h>
-#include <linux/dcache.h>
 #include <linux/mount.h>
 #include <linux/namei.h>
-#include <linux/pagemap.h>
-#include <linux/slab.h>
-#include <linux/exportfs.h>
 
 /*
  * xfs_find_handle maps from userspace xfs_fsop_handlereq structure to
@@ -721,16 +713,45 @@ out_unlock:
 	return error;
 }
 
+/* Return 0 on success or positive error */
+int
+xfs_fsbulkstat_one_fmt(
+	struct xfs_ibulk		*breq,
+	const struct xfs_bulkstat	*bstat)
+{
+	struct xfs_bstat		bs1;
+
+	xfs_bulkstat_to_bstat(breq->mp, &bs1, bstat);
+	if (copy_to_user(breq->ubuffer, &bs1, sizeof(bs1)))
+		return -EFAULT;
+	return xfs_ibulk_advance(breq, sizeof(struct xfs_bstat));
+}
+
+int
+xfs_fsinumbers_fmt(
+	struct xfs_ibulk		*breq,
+	const struct xfs_inumbers	*igrp)
+{
+	struct xfs_inogrp		ig1;
+
+	xfs_inumbers_to_inogrp(&ig1, igrp);
+	if (copy_to_user(breq->ubuffer, &ig1, sizeof(struct xfs_inogrp)))
+		return -EFAULT;
+	return xfs_ibulk_advance(breq, sizeof(struct xfs_inogrp));
+}
+
 STATIC int
-xfs_ioc_bulkstat(
+xfs_ioc_fsbulkstat(
 	xfs_mount_t		*mp,
 	unsigned int		cmd,
 	void			__user *arg)
 {
-	xfs_fsop_bulkreq_t	bulkreq;
-	int			count;	/* # of records returned */
-	xfs_ino_t		inlast;	/* last inode number */
-	int			done;
+	struct xfs_fsop_bulkreq	bulkreq;
+	struct xfs_ibulk	breq = {
+		.mp		= mp,
+		.ocount		= 0,
+	};
+	xfs_ino_t		lastino;
 	int			error;
 
 	/* done = 1 if there are more stats to get and if bulkstat */
@@ -742,41 +763,243 @@ xfs_ioc_bulkstat(
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return -EIO;
 
-	if (copy_from_user(&bulkreq, arg, sizeof(xfs_fsop_bulkreq_t)))
+	if (copy_from_user(&bulkreq, arg, sizeof(struct xfs_fsop_bulkreq)))
 		return -EFAULT;
 
-	if (copy_from_user(&inlast, bulkreq.lastip, sizeof(__s64)))
+	if (copy_from_user(&lastino, bulkreq.lastip, sizeof(__s64)))
 		return -EFAULT;
 
-	if ((count = bulkreq.icount) <= 0)
+	if (bulkreq.icount <= 0)
 		return -EINVAL;
 
 	if (bulkreq.ubuffer == NULL)
 		return -EINVAL;
 
-	if (cmd == XFS_IOC_FSINUMBERS)
-		error = xfs_inumbers(mp, &inlast, &count,
-					bulkreq.ubuffer, xfs_inumbers_fmt);
-	else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE)
-		error = xfs_bulkstat_one(mp, inlast, bulkreq.ubuffer,
-					sizeof(xfs_bstat_t), NULL, &done);
-	else	/* XFS_IOC_FSBULKSTAT */
-		error = xfs_bulkstat(mp, &inlast, &count, xfs_bulkstat_one,
-				     sizeof(xfs_bstat_t), bulkreq.ubuffer,
-				     &done);
+	breq.ubuffer = bulkreq.ubuffer;
+	breq.icount = bulkreq.icount;
+
+	/*
+	 * FSBULKSTAT_SINGLE expects that *lastip contains the inode number
+	 * that we want to stat.  However, FSINUMBERS and FSBULKSTAT expect
+	 * that *lastip contains either zero or the number of the last inode to
+	 * be examined by the previous call and return results starting with
+	 * the next inode after that.  The new bulk request back end functions
+	 * take the inode to start with, so we have to compute the startino
+	 * parameter from lastino to maintain correct function.  lastino == 0
+	 * is a special case because it has traditionally meant "first inode
+	 * in filesystem".
+	 */
+	if (cmd == XFS_IOC_FSINUMBERS) {
+		breq.startino = lastino ? lastino + 1 : 0;
+		error = xfs_inumbers(&breq, xfs_fsinumbers_fmt);
+		lastino = breq.startino - 1;
+	} else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE) {
+		breq.startino = lastino;
+		breq.icount = 1;
+		error = xfs_bulkstat_one(&breq, xfs_fsbulkstat_one_fmt);
+	} else {	/* XFS_IOC_FSBULKSTAT */
+		breq.startino = lastino ? lastino + 1 : 0;
+		error = xfs_bulkstat(&breq, xfs_fsbulkstat_one_fmt);
+		lastino = breq.startino - 1;
+	}
 
 	if (error)
 		return error;
 
-	if (bulkreq.ocount != NULL) {
-		if (copy_to_user(bulkreq.lastip, &inlast,
-						sizeof(xfs_ino_t)))
-			return -EFAULT;
+	if (bulkreq.lastip != NULL &&
+	    copy_to_user(bulkreq.lastip, &lastino, sizeof(xfs_ino_t)))
+		return -EFAULT;
 
-		if (copy_to_user(bulkreq.ocount, &count, sizeof(count)))
-			return -EFAULT;
+	if (bulkreq.ocount != NULL &&
+	    copy_to_user(bulkreq.ocount, &breq.ocount, sizeof(__s32)))
+		return -EFAULT;
+
+	return 0;
+}
+
+/* Return 0 on success or positive error */
+static int
+xfs_bulkstat_fmt(
+	struct xfs_ibulk		*breq,
+	const struct xfs_bulkstat	*bstat)
+{
+	if (copy_to_user(breq->ubuffer, bstat, sizeof(struct xfs_bulkstat)))
+		return -EFAULT;
+	return xfs_ibulk_advance(breq, sizeof(struct xfs_bulkstat));
+}
+
+/*
+ * Check the incoming bulk request @hdr from userspace and initialize the
+ * internal @breq bulk request appropriately.  Returns 0 if the bulk request
+ * should proceed; XFS_ITER_ABORT if there's nothing to do; or the usual
+ * negative error code.
+ */
+static int
+xfs_bulk_ireq_setup(
+	struct xfs_mount	*mp,
+	struct xfs_bulk_ireq	*hdr,
+	struct xfs_ibulk	*breq,
+	void __user		*ubuffer)
+{
+	if (hdr->icount == 0 ||
+	    (hdr->flags & ~XFS_BULK_IREQ_FLAGS_ALL) ||
+	    memchr_inv(hdr->reserved, 0, sizeof(hdr->reserved)))
+		return -EINVAL;
+
+	breq->startino = hdr->ino;
+	breq->ubuffer = ubuffer;
+	breq->icount = hdr->icount;
+	breq->ocount = 0;
+	breq->flags = 0;
+
+	/*
+	 * The @ino parameter is a special value, so we must look it up here.
+	 * We're not allowed to have IREQ_AGNO, and we only return one inode
+	 * worth of data.
+	 */
+	if (hdr->flags & XFS_BULK_IREQ_SPECIAL) {
+		if (hdr->flags & XFS_BULK_IREQ_AGNO)
+			return -EINVAL;
+
+		switch (hdr->ino) {
+		case XFS_BULK_IREQ_SPECIAL_ROOT:
+			hdr->ino = mp->m_sb.sb_rootino;
+			break;
+		default:
+			return -EINVAL;
+		}
+		breq->icount = 1;
 	}
 
+	/*
+	 * The IREQ_AGNO flag means that we only want results from a given AG.
+	 * If @hdr->ino is zero, we start iterating in that AG.  If @hdr->ino is
+	 * beyond the specified AG then we return no results.
+	 */
+	if (hdr->flags & XFS_BULK_IREQ_AGNO) {
+		if (hdr->agno >= mp->m_sb.sb_agcount)
+			return -EINVAL;
+
+		if (breq->startino == 0)
+			breq->startino = XFS_AGINO_TO_INO(mp, hdr->agno, 0);
+		else if (XFS_INO_TO_AGNO(mp, breq->startino) < hdr->agno)
+			return -EINVAL;
+
+		breq->flags |= XFS_IBULK_SAME_AG;
+
+		/* Asking for an inode past the end of the AG?  We're done! */
+		if (XFS_INO_TO_AGNO(mp, breq->startino) > hdr->agno)
+			return XFS_ITER_ABORT;
+	} else if (hdr->agno)
+		return -EINVAL;
+
+	/* Asking for an inode past the end of the FS?  We're done! */
+	if (XFS_INO_TO_AGNO(mp, breq->startino) >= mp->m_sb.sb_agcount)
+		return XFS_ITER_ABORT;
+
+	return 0;
+}
+
+/*
+ * Update the userspace bulk request @hdr to reflect the end state of the
+ * internal bulk request @breq.
+ */
+static void
+xfs_bulk_ireq_teardown(
+	struct xfs_bulk_ireq	*hdr,
+	struct xfs_ibulk	*breq)
+{
+	hdr->ino = breq->startino;
+	hdr->ocount = breq->ocount;
+}
+
+/* Handle the v5 bulkstat ioctl. */
+STATIC int
+xfs_ioc_bulkstat(
+	struct xfs_mount		*mp,
+	unsigned int			cmd,
+	struct xfs_bulkstat_req __user	*arg)
+{
+	struct xfs_bulk_ireq		hdr;
+	struct xfs_ibulk		breq = {
+		.mp			= mp,
+	};
+	int				error;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return -EIO;
+
+	if (copy_from_user(&hdr, &arg->hdr, sizeof(hdr)))
+		return -EFAULT;
+
+	error = xfs_bulk_ireq_setup(mp, &hdr, &breq, arg->bulkstat);
+	if (error == XFS_ITER_ABORT)
+		goto out_teardown;
+	if (error < 0)
+		return error;
+
+	error = xfs_bulkstat(&breq, xfs_bulkstat_fmt);
+	if (error)
+		return error;
+
+out_teardown:
+	xfs_bulk_ireq_teardown(&hdr, &breq);
+	if (copy_to_user(&arg->hdr, &hdr, sizeof(hdr)))
+		return -EFAULT;
+
+	return 0;
+}
+
+STATIC int
+xfs_inumbers_fmt(
+	struct xfs_ibulk		*breq,
+	const struct xfs_inumbers	*igrp)
+{
+	if (copy_to_user(breq->ubuffer, igrp, sizeof(struct xfs_inumbers)))
+		return -EFAULT;
+	return xfs_ibulk_advance(breq, sizeof(struct xfs_inumbers));
+}
+
+/* Handle the v5 inumbers ioctl. */
+STATIC int
+xfs_ioc_inumbers(
+	struct xfs_mount		*mp,
+	unsigned int			cmd,
+	struct xfs_inumbers_req __user	*arg)
+{
+	struct xfs_bulk_ireq		hdr;
+	struct xfs_ibulk		breq = {
+		.mp			= mp,
+	};
+	int				error;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return -EIO;
+
+	if (copy_from_user(&hdr, &arg->hdr, sizeof(hdr)))
+		return -EFAULT;
+
+	error = xfs_bulk_ireq_setup(mp, &hdr, &breq, arg->inumbers);
+	if (error == XFS_ITER_ABORT)
+		goto out_teardown;
+	if (error < 0)
+		return error;
+
+	error = xfs_inumbers(&breq, xfs_inumbers_fmt);
+	if (error)
+		return error;
+
+out_teardown:
+	xfs_bulk_ireq_teardown(&hdr, &breq);
+	if (copy_to_user(&arg->hdr, &hdr, sizeof(hdr)))
+		return -EFAULT;
+
 	return 0;
 }
 
@@ -1926,7 +2149,12 @@ xfs_file_ioctl(
 	case XFS_IOC_FSBULKSTAT_SINGLE:
 	case XFS_IOC_FSBULKSTAT:
 	case XFS_IOC_FSINUMBERS:
+		return xfs_ioc_fsbulkstat(mp, cmd, arg);
+
+	case XFS_IOC_BULKSTAT:
 		return xfs_ioc_bulkstat(mp, cmd, arg);
+	case XFS_IOC_INUMBERS:
+		return xfs_ioc_inumbers(mp, cmd, arg);
 
 	case XFS_IOC_FSGEOMETRY_V1:
 		return xfs_ioc_fsgeometry(mp, arg, 3);