Re: [PATCH] man2/vmsplice: splice pages from pipe to memory

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Michael,

What is wrong with this patch?

On Mon, Nov 06, 2017 at 11:34:17PM -0800, Andrei Vagin wrote:
> The man page describes that vmsplice can splice pages from
> memory to a pipe, but it can work to anonther direction as well.
> 
> Cc: Jens Axboe <axboe@xxxxxxxxx>
> Signed-off-by: Andrei Vagin <avagin@xxxxxxxxxx>
> ---
>  man2/vmsplice.2 | 24 +++++++++++++++++++++---
>  1 file changed, 21 insertions(+), 3 deletions(-)
> 
> diff --git a/man2/vmsplice.2 b/man2/vmsplice.2
> index e3e61cf27..8f48b8809 100644
> --- a/man2/vmsplice.2
> +++ b/man2/vmsplice.2
> @@ -25,7 +25,7 @@
>  .\"
>  .TH VMSPLICE 2 2014-10-02 "Linux" "Linux Programmer's Manual"
>  .SH NAME
> -vmsplice \- splice user pages into a pipe
> +vmsplice \- splice user pages to/from a pipe
>  .SH SYNOPSIS
>  .nf
>  .BR "#define _GNU_SOURCE" "         /* See feature_test_macros(7) */"
> @@ -44,13 +44,25 @@ vmsplice \- splice user pages into a pipe
>  .\" the kernel-space one (contrast this to "write()", which copies
>  .\" the actual data, and you can thus reuse the buffer immediately
>  .\" after a successful write), but that is often easy to do.
> -The
> +If
> +.I fd
> +is opened for writting, the
>  .BR vmsplice ()
>  system call maps
>  .I nr_segs
>  ranges of user memory described by
>  .I iov
>  into a pipe.
> +.\" commit 6a14b90bb6bc7cd83e2a444bf457a2ea645cbfe7
> +If
> +.I fd
> +is opened for reading, the
> +.BR vmsplice ()
> +system call fills
> +.I nr_segs
> +ranges of user memory described by
> +.I iov
> +from a pipe.
>  The file descriptor
>  .I fd
>  must refer to a pipe.
> @@ -70,7 +82,6 @@ struct iovec {
>  };
>  .in
>  .fi
> -
>  The
>  .I flags
>  argument is a bit mask that is composed by ORing together
> @@ -163,6 +174,13 @@ as defined in
>  Currently,
>  .\" UIO_MAXIOV in kernel source
>  this limit is 1024.
> +
> +.\" commit 6a14b90bb6bc7cd83e2a444bf457a2ea645cbfe7
> +.BR vmsplice ()
> +only really supports true splicing from user memory to a pipe. In another
> +direction, it actually just copies the data to userspace. But this makes
> +the interface nice and symmetric and enables people to build on splice,
> +with room for future improvement in performance.
>  .SH SEE ALSO
>  .BR splice (2),
>  .BR tee (2)
> -- 
> 2.13.6
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-api" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux