svn commit: r290970 - head/sys/fs/nfsclient

[Rick Macklem]

Author: rmacklem
Date: Tue Nov 17 01:44:26 2015
New Revision: 290970
URL: https://svnweb.freebsd.org/changeset/base/290970

Log:
  mnt_stat.f_iosize (which is used to set bo_bsize) must be set to
  the largest size of buffer cache block or the mapping of the buffer
  is bogus. When a mount with rsize=4096,wsize=4096 was done, f_iosize
  would be set to 4096. This resulted in corrupted directory data, since
  the buffer cache block size for directories is NFS_DIRBLKSIZ (8192).
  This patch fixes the code so that it always sets f_iosize to at least
  NFS_DIRBLKSIZ.
  
  Tested by:	krichy at cflinux.hu
  PR:		177971
  MFC after:	2 weeks

Modified:
  head/sys/fs/nfsclient/nfs_clvfsops.c

Modified: head/sys/fs/nfsclient/nfs_clvfsops.c
==============================================================================
--- head/sys/fs/nfsclient/nfs_clvfsops.c	Tue Nov 17 01:07:21 2015	(r290969)
+++ head/sys/fs/nfsclient/nfs_clvfsops.c	Tue Nov 17 01:44:26 2015	(r290970)
@@ -216,10 +216,12 @@ newnfs_iosize(struct nfsmount *nmp)
 	 * Calculate the size used for io buffers.  Use the larger
 	 * of the two sizes to minimise nfs requests but make sure
 	 * that it is at least one VM page to avoid wasting buffer
-	 * space.
+	 * space.  It must also be at least NFS_DIRBLKSIZ, since
+	 * that is the buffer size used for directories.
 	 */
 	iosize = imax(nmp->nm_rsize, nmp->nm_wsize);
 	iosize = imax(iosize, PAGE_SIZE);
+	iosize = imax(iosize, NFS_DIRBLKSIZ);
 	nmp->nm_mountp->mnt_stat.f_iosize = iosize;
 	return (iosize);
 }