xfsrestore: fix 2GB directory dump limitation for multi-stream

xfsrestore: fix 2GB directory dump limitation for multi-stream

[Rich Johnston]

[-- Attachment #1: fix-2GB-directory-dump-limitation-for-multi-stream.patch --]
[-- Type: text/plain, Size: 1492 bytes --]


The drive_simple restore path has a 2GB directory limit. Instead of
ASSERTing if nreadneeded64 is greater than INTGENMAX (2GB), add a loop
to handle it.

Signed-off-by: Rich Johnston <rjohnston@sgi.com>
---
 common/drive_simple.c |   24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

Index: b/common/drive_simple.c
===================================================================
--- a/common/drive_simple.c
+++ b/common/drive_simple.c
@@ -765,18 +765,20 @@ do_seek_mark( drive_t *drivep, drive_mar
 	/* use read_buf util func to eat up difference
 	 */
 	nreadneeded64 = mark - strmoff;
-	ASSERT( nreadneeded64 <= INTGENMAX );
-	nreadneeded = ( intgen_t )nreadneeded64;
-	nread = read_buf( 0,
-			  ( size_t )nreadneeded,
-			  ( void * )drivep,
-			  ( rfp_t )drivep->d_opsp->do_read,
-			  ( rrbfp_t )drivep->d_opsp->do_return_read_buf,
-			  &rval );
-	if  ( rval ) {
-		return rval;
+	while ( nreadneeded64 > 0 ) {
+		if ( nreadneeded64 > INTGENMAX )
+			nreadneeded = INTGENMAX;
+		else
+			nreadneeded = ( intgen_t )nreadneeded64;
+		nread = read_buf( 0, nreadneeded, drivep,
+				  ( rfp_t )drivep->d_opsp->do_read,
+				( rrbfp_t )drivep->d_opsp->do_return_read_buf,
+				  &rval );
+		if  ( rval ) {
+			return rval;
+		}
+		nreadneeded64 -= nread;
 	}
-	ASSERT( nread == nreadneeded );
 
 	/* verify we are on the mark
 	 */

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: xfsrestore: fix 2GB directory dump limitation for multi-stream

[Rich Johnston]

ping, Does anyone have time to review this please?

Thanks
--Rich

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: xfsrestore: fix 2GB directory dump limitation for multi-stream

[Brian Foster]

On Thu, Aug 27, 2015 at 01:54:45PM -0500, Rich Johnston wrote:
> 
> The drive_simple restore path has a 2GB directory limit. Instead of
> ASSERTing if nreadneeded64 is greater than INTGENMAX (2GB), add a loop
> to handle it.
> 
> Signed-off-by: Rich Johnston <rjohnston@sgi.com>
> ---

I'm not a huge fan of all the spaces used in this code (around the
braces and whatnot). It is consistent with the rest of the code but I'd
probably get away from that as code is modified.

That aside, the change seems fine to me:

Reviewed-by: Brian Foster <bfoster@redhat.com>

>  common/drive_simple.c |   24 +++++++++++++-----------
>  1 file changed, 13 insertions(+), 11 deletions(-)
> 
> Index: b/common/drive_simple.c
> ===================================================================
> --- a/common/drive_simple.c
> +++ b/common/drive_simple.c
> @@ -765,18 +765,20 @@ do_seek_mark( drive_t *drivep, drive_mar
>  	/* use read_buf util func to eat up difference
>  	 */
>  	nreadneeded64 = mark - strmoff;
> -	ASSERT( nreadneeded64 <= INTGENMAX );
> -	nreadneeded = ( intgen_t )nreadneeded64;
> -	nread = read_buf( 0,
> -			  ( size_t )nreadneeded,
> -			  ( void * )drivep,
> -			  ( rfp_t )drivep->d_opsp->do_read,
> -			  ( rrbfp_t )drivep->d_opsp->do_return_read_buf,
> -			  &rval );
> -	if  ( rval ) {
> -		return rval;
> +	while ( nreadneeded64 > 0 ) {
> +		if ( nreadneeded64 > INTGENMAX )
> +			nreadneeded = INTGENMAX;
> +		else
> +			nreadneeded = ( intgen_t )nreadneeded64;
> +		nread = read_buf( 0, nreadneeded, drivep,
> +				  ( rfp_t )drivep->d_opsp->do_read,
> +				( rrbfp_t )drivep->d_opsp->do_return_read_buf,
> +				  &rval );
> +		if  ( rval ) {
> +			return rval;
> +		}
> +		nreadneeded64 -= nread;
>  	}
> -	ASSERT( nread == nreadneeded );
>  
>  	/* verify we are on the mark
>  	 */
> 
> _______________________________________________
> xfs mailing list
> xfs@oss.sgi.com
> http://oss.sgi.com/mailman/listinfo/xfs

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs