Re: [PATCH 07/11] xfs: Remove kmem_realloc

["Darrick J. Wong" <darrick.wong@oracle.com>]

On Wed, Nov 13, 2019 at 09:40:54AM -0800, Darrick J. Wong wrote:
> On Wed, Nov 13, 2019 at 03:23:31PM +0100, Carlos Maiolino wrote:
> > We can use krealloc() with __GFP_NOFAIL directly
> > 
> > Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
> > ---
> >  fs/xfs/kmem.c                  | 22 ----------------------
> >  fs/xfs/kmem.h                  |  1 -
> >  fs/xfs/libxfs/xfs_iext_tree.c  |  2 +-
> >  fs/xfs/libxfs/xfs_inode_fork.c |  8 ++++----
> >  fs/xfs/xfs_log_recover.c       |  2 +-
> >  fs/xfs/xfs_mount.c             |  4 ++--
> >  6 files changed, 8 insertions(+), 31 deletions(-)
> > 
> > diff --git a/fs/xfs/kmem.c b/fs/xfs/kmem.c
> > index 2644fdaa0549..6e10e565632c 100644
> > --- a/fs/xfs/kmem.c
> > +++ b/fs/xfs/kmem.c
> > @@ -93,25 +93,3 @@ kmem_alloc_large(size_t size, xfs_km_flags_t flags)
> >  		return ptr;
> >  	return __kmem_vmalloc(size, flags);
> >  }
> > -
> > -void *
> > -kmem_realloc(const void *old, size_t newsize, xfs_km_flags_t flags)
> > -{
> > -	int	retries = 0;
> > -	gfp_t	lflags = kmem_flags_convert(flags);
> > -	void	*ptr;
> > -
> > -	trace_kmem_realloc(newsize, flags, _RET_IP_);
> > -
> > -	do {
> > -		ptr = krealloc(old, newsize, lflags);
> > -		if (ptr || (flags & KM_MAYFAIL))
> > -			return ptr;
> > -		if (!(++retries % 100))
> > -			xfs_err(NULL,
> > -	"%s(%u) possible memory allocation deadlock size %zu in %s (mode:0x%x)",
> > -				current->comm, current->pid,
> > -				newsize, __func__, lflags);
> > -		congestion_wait(BLK_RW_ASYNC, HZ/50);

I left a more detailed comment attached to patch 9, but is it ok to
remove this retry logic?

> > -	} while (1);
> > -}
> > diff --git a/fs/xfs/kmem.h b/fs/xfs/kmem.h
> > index 46c8c5546674..18b62eee3177 100644
> > --- a/fs/xfs/kmem.h
> > +++ b/fs/xfs/kmem.h
> > @@ -55,7 +55,6 @@ kmem_flags_convert(xfs_km_flags_t flags)
> >  extern void *kmem_alloc(size_t, xfs_km_flags_t);
> >  extern void *kmem_alloc_io(size_t size, int align_mask, xfs_km_flags_t flags);
> >  extern void *kmem_alloc_large(size_t size, xfs_km_flags_t);
> > -extern void *kmem_realloc(const void *, size_t, xfs_km_flags_t);
> >  static inline void  kmem_free(const void *ptr)
> >  {
> >  	kvfree(ptr);
> > diff --git a/fs/xfs/libxfs/xfs_iext_tree.c b/fs/xfs/libxfs/xfs_iext_tree.c
> > index f2005671e86c..a929ea0b09b7 100644
> > --- a/fs/xfs/libxfs/xfs_iext_tree.c
> > +++ b/fs/xfs/libxfs/xfs_iext_tree.c
> > @@ -607,7 +607,7 @@ xfs_iext_realloc_root(
> >  	if (new_size / sizeof(struct xfs_iext_rec) == RECS_PER_LEAF)
> >  		new_size = NODE_SIZE;
> >  
> > -	new = kmem_realloc(ifp->if_u1.if_root, new_size, KM_NOFS);
> > +	new = krealloc(ifp->if_u1.if_root, new_size, GFP_NOFS | __GFP_NOFAIL);
> >  	memset(new + ifp->if_bytes, 0, new_size - ifp->if_bytes);
> >  	ifp->if_u1.if_root = new;
> >  	cur->leaf = new;
> > diff --git a/fs/xfs/libxfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c
> > index 2bffaa31d62a..34c336f45796 100644
> > --- a/fs/xfs/libxfs/xfs_inode_fork.c
> > +++ b/fs/xfs/libxfs/xfs_inode_fork.c
> > @@ -387,8 +387,8 @@ xfs_iroot_realloc(
> >  		cur_max = xfs_bmbt_maxrecs(mp, ifp->if_broot_bytes, 0);
> >  		new_max = cur_max + rec_diff;
> >  		new_size = XFS_BMAP_BROOT_SPACE_CALC(mp, new_max);
> > -		ifp->if_broot = kmem_realloc(ifp->if_broot, new_size,
> > -				KM_NOFS);
> > +		ifp->if_broot = krealloc(ifp->if_broot, new_size,
> > +				GFP_NOFS | __GFP_NOFAIL);
> >  		op = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1,
> >  						     ifp->if_broot_bytes);
> >  		np = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1,
> > @@ -497,8 +497,8 @@ xfs_idata_realloc(
> >  	 * in size so that it can be logged and stay on word boundaries.
> >  	 * We enforce that here.
> >  	 */
> > -	ifp->if_u1.if_data = kmem_realloc(ifp->if_u1.if_data,
> > -			roundup(new_size, 4), KM_NOFS);
> > +	ifp->if_u1.if_data = krealloc(ifp->if_u1.if_data, roundup(new_size, 4),
> > +				      GFP_NOFS | __GFP_NOFAIL);
> >  	ifp->if_bytes = new_size;
> >  }
> >  
> > diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
> > index bc5c0aef051c..a7f1dcecc640 100644
> > --- a/fs/xfs/xfs_log_recover.c
> > +++ b/fs/xfs/xfs_log_recover.c
> > @@ -4211,7 +4211,7 @@ xlog_recover_add_to_cont_trans(
> >  	old_ptr = item->ri_buf[item->ri_cnt-1].i_addr;
> >  	old_len = item->ri_buf[item->ri_cnt-1].i_len;
> >  
> > -	ptr = kmem_realloc(old_ptr, len + old_len, 0);
> > +	ptr = krealloc(old_ptr, len + old_len, GFP_KERNEL | __GFP_NOFAIL);
> >  	memcpy(&ptr[old_len], dp, len);
> >  	item->ri_buf[item->ri_cnt-1].i_len += len;
> >  	item->ri_buf[item->ri_cnt-1].i_addr = ptr;
> > diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
> > index 91a5354f20fb..a14046314c1f 100644
> > --- a/fs/xfs/xfs_mount.c
> > +++ b/fs/xfs/xfs_mount.c
> > @@ -80,9 +80,9 @@ xfs_uuid_mount(
> >  	}
> >  
> >  	if (hole < 0) {
> > -		xfs_uuid_table = kmem_realloc(xfs_uuid_table,
> > +		xfs_uuid_table = krealloc(xfs_uuid_table,
> >  			(xfs_uuid_table_size + 1) * sizeof(*xfs_uuid_table),
> > -			0);
> > +			GFP_KERNEL | __GFP_NOFAIL);
> >  		hole = xfs_uuid_table_size++;
> >  	}
> >  	xfs_uuid_table[hole] = *uuid;
> 
> This whole uuid table thing is pretty ugly (what happens when someone
> tries to mount a few thousand xfs images?) but I guess fixing it is for
> another day...
> 
> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>

  ^^^^^^^^^^^ Withdrawn for the time being.

--D

> --D
> 
> > -- 
> > 2.23.0
> >