[PATCH] fideduperange.2: fix the discussion of maximum sizes

[PATCH] fideduperange.2: fix the discussion of maximum sizes

[Darrick J. Wong]

Fix the discussion of the limitations on the dest_count and src_length
parameters to the fideduperange ioctl to reflect what's actually in the
kernel.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 man2/ioctl_fideduperange.2 |   12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/man2/ioctl_fideduperange.2 b/man2/ioctl_fideduperange.2
index 2112d10..59f0ea9 100644
--- a/man2/ioctl_fideduperange.2
+++ b/man2/ioctl_fideduperange.2
@@ -95,7 +95,7 @@ struct file_dedupe_range_info {
 .in
 
 Each deduplication operation targets
-.IR length
+.IR src_length
 bytes in file descriptor
 .IR dest_fd
 at offset
@@ -108,8 +108,14 @@ During the call,
 must be open for reading and
 .IR dest_fd
 must be open for writing.
-For any call to this ioctl, there may not be more than 65,536
-requests attached; each request may not exceed 16MiB.
+The combined size of the struct
+.IR file_dedupe_range
+and the struct
+.IR file_dedupe_range_info
+array must not exceed the system page size.
+The maximum size of
+.IR src_length
+is filesystem dependent and is typically 16MiB.
 By convention, the storage used by
 .IR src_fd
 is mapped into

Re: [PATCH] fideduperange.2: fix the discussion of maximum sizes

[Omar Sandoval]

On Tue, Nov 22, 2016 at 08:22:22PM -0800, Darrick J. Wong wrote:
> Fix the discussion of the limitations on the dest_count and src_length
> parameters to the fideduperange ioctl to reflect what's actually in the
> kernel.
> 
> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> ---
>  man2/ioctl_fideduperange.2 |   12 +++++++++---
>  1 file changed, 9 insertions(+), 3 deletions(-)
> 
> diff --git a/man2/ioctl_fideduperange.2 b/man2/ioctl_fideduperange.2
> index 2112d10..59f0ea9 100644
> --- a/man2/ioctl_fideduperange.2
> +++ b/man2/ioctl_fideduperange.2
> @@ -95,7 +95,7 @@ struct file_dedupe_range_info {
>  .in
>  
>  Each deduplication operation targets
> -.IR length
> +.IR src_length
>  bytes in file descriptor
>  .IR dest_fd
>  at offset
> @@ -108,8 +108,14 @@ During the call,
>  must be open for reading and
>  .IR dest_fd
>  must be open for writing.
> -For any call to this ioctl, there may not be more than 65,536
> -requests attached; each request may not exceed 16MiB.
> +The combined size of the struct
> +.IR file_dedupe_range
> +and the struct
> +.IR file_dedupe_range_info
> +array must not exceed the system page size.
> +The maximum size of
> +.IR src_length
> +is filesystem dependent and is typically 16MiB.
>  By convention, the storage used by
>  .IR src_fd
>  is mapped into

Might be worth it to clarify that the file_dedupe_range_info limit is a
hard limit that will result in an error if exceeded and that the
src_length limit is silently enforced. These semantics suck...

-- 
Omar

Re: [PATCH] fideduperange.2: fix the discussion of maximum sizes

[Omar Sandoval]

On Tue, Nov 22, 2016 at 08:22:22PM -0800, Darrick J. Wong wrote:
> Fix the discussion of the limitations on the dest_count and src_length
> parameters to the fideduperange ioctl to reflect what's actually in the
> kernel.
> 
> Signed-off-by: Darrick J. Wong <darrick.wong-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>
> ---
>  man2/ioctl_fideduperange.2 |   12 +++++++++---
>  1 file changed, 9 insertions(+), 3 deletions(-)
> 
> diff --git a/man2/ioctl_fideduperange.2 b/man2/ioctl_fideduperange.2
> index 2112d10..59f0ea9 100644
> --- a/man2/ioctl_fideduperange.2
> +++ b/man2/ioctl_fideduperange.2
> @@ -95,7 +95,7 @@ struct file_dedupe_range_info {
>  .in
>  
>  Each deduplication operation targets
> -.IR length
> +.IR src_length
>  bytes in file descriptor
>  .IR dest_fd
>  at offset
> @@ -108,8 +108,14 @@ During the call,
>  must be open for reading and
>  .IR dest_fd
>  must be open for writing.
> -For any call to this ioctl, there may not be more than 65,536
> -requests attached; each request may not exceed 16MiB.
> +The combined size of the struct
> +.IR file_dedupe_range
> +and the struct
> +.IR file_dedupe_range_info
> +array must not exceed the system page size.
> +The maximum size of
> +.IR src_length
> +is filesystem dependent and is typically 16MiB.
>  By convention, the storage used by
>  .IR src_fd
>  is mapped into

Might be worth it to clarify that the file_dedupe_range_info limit is a
hard limit that will result in an error if exceeded and that the
src_length limit is silently enforced. These semantics suck...

-- 
Omar

Re: [PATCH] fideduperange.2: fix the discussion of maximum sizes

[Darrick J. Wong]

On Tue, Nov 22, 2016 at 08:39:18PM -0800, Omar Sandoval wrote:
> On Tue, Nov 22, 2016 at 08:22:22PM -0800, Darrick J. Wong wrote:
> > Fix the discussion of the limitations on the dest_count and src_length
> > parameters to the fideduperange ioctl to reflect what's actually in the
> > kernel.
> > 
> > Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> > ---
> >  man2/ioctl_fideduperange.2 |   12 +++++++++---
> >  1 file changed, 9 insertions(+), 3 deletions(-)
> > 
> > diff --git a/man2/ioctl_fideduperange.2 b/man2/ioctl_fideduperange.2
> > index 2112d10..59f0ea9 100644
> > --- a/man2/ioctl_fideduperange.2
> > +++ b/man2/ioctl_fideduperange.2
> > @@ -95,7 +95,7 @@ struct file_dedupe_range_info {
> >  .in
> >  
> >  Each deduplication operation targets
> > -.IR length
> > +.IR src_length
> >  bytes in file descriptor
> >  .IR dest_fd
> >  at offset
> > @@ -108,8 +108,14 @@ During the call,
> >  must be open for reading and
> >  .IR dest_fd
> >  must be open for writing.
> > -For any call to this ioctl, there may not be more than 65,536
> > -requests attached; each request may not exceed 16MiB.
> > +The combined size of the struct
> > +.IR file_dedupe_range
> > +and the struct
> > +.IR file_dedupe_range_info
> > +array must not exceed the system page size.
> > +The maximum size of
> > +.IR src_length
> > +is filesystem dependent and is typically 16MiB.
> >  By convention, the storage used by
> >  .IR src_fd
> >  is mapped into
> 
> Might be worth it to clarify that the file_dedupe_range_info limit is a
> hard limit that will result in an error if exceeded and that the
> src_length limit is silently enforced. These semantics suck...

Yeah... :(

Anyway I'll post a v2 patch.

--D

> 
> -- 
> Omar
> --
> To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html