linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Benny Halevy <bhalevy@panasas.com>
To: andros@netapp.com
Cc: linux-nfs@vger.kernel.org
Subject: Re: [PATCH 4/7] SQUASHME pnfs-submit consolidate write_pagelist parameters
Date: Tue, 20 Jul 2010 07:26:37 +0300	[thread overview]
Message-ID: <4C45257D.7000507@panasas.com> (raw)
In-Reply-To: <1279210434-2772-5-git-send-email-andros@netapp.com>

On Jul. 15, 2010, 19:13 +0300, andros@netapp.com wrote:
> From: Andy Adamson <andros@netapp.com>
> 
> Remove the numpages calculation which is not used by the file layout driver.
> Layout drivers that need the number of pages can call nfs_page_array_len in
> their write_pagelist operation.
> 
> NOTE: API change.
> 
> Signed-off-by: Andy Adamson <andros@netapp.com>
> ---
>  fs/nfs/pnfs.c             |   24 +++---------------------
>  include/linux/nfs4_pnfs.h |    5 +----
>  2 files changed, 4 insertions(+), 25 deletions(-)
> 
> diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
> index 576946b..51bd66f 100644
> --- a/fs/nfs/pnfs.c
> +++ b/fs/nfs/pnfs.c
> @@ -1271,39 +1271,21 @@ enum pnfs_try_status
>  pnfs_try_to_write_data(struct nfs_write_data *wdata,
>  			const struct rpc_call_ops *call_ops, int how)
>  {
> -	struct nfs_writeargs *args = &wdata->args;
>  	struct inode *inode = wdata->inode;
>  	enum pnfs_try_status trypnfs;
>  	struct nfs_server *nfss = NFS_SERVER(inode);
> -	struct nfs_inode *nfsi = NFS_I(inode);
>  	struct pnfs_layout_segment *lseg = wdata->req->wb_lseg;
> -	int numpages;
>  
>  	wdata->pdata.call_ops = call_ops;
>  	wdata->pdata.how = how;
>  
> -	dprintk("%s: Writing ino:%lu %u@%llu\n", __func__,
> -		inode->i_ino, args->count, args->offset);
> +	dprintk("%s: Writing ino:%lu %u@%llu (how %d)\n", __func__,
> +		inode->i_ino, wdata->args.count, wdata->args.offset, how);
>  
>  	get_lseg(lseg);
>  
> -	/* Determine number of pages
> -	 */
> -	numpages = nfs_page_array_len(args->pgbase, args->count);

Andy, removing this from the interface is a bit problematic
since nfs_page_array_len is private to the nfs module and I want to leave
the logic there rather than copying it to the layout driver.
Also, we still pass in its equivalent to read_pagelist so for symmetry
reasons it would be cleaner to pass to to the write path as well.

Benny

> -
> -	dprintk("%s: Calling layout driver (how %d) write with %d pages\n",
> -		__func__, how, numpages);
> -
>  	wdata->pdata.lseg = lseg;
> -	trypnfs = nfss->pnfs_curr_ld->ld_io_ops->write_pagelist(
> -							nfsi->layout,
> -							args->pages,
> -							args->pgbase,
> -							numpages,
> -							(loff_t)args->offset,
> -							args->count,
> -							how,
> -							wdata);
> +	trypnfs = nfss->pnfs_curr_ld->ld_io_ops->write_pagelist(wdata, how);
>  
>  	if (trypnfs == PNFS_NOT_ATTEMPTED) {
>  		wdata->pdata.lseg = NULL;
> diff --git a/include/linux/nfs4_pnfs.h b/include/linux/nfs4_pnfs.h
> index b379225..13f71ad 100644
> --- a/include/linux/nfs4_pnfs.h
> +++ b/include/linux/nfs4_pnfs.h
> @@ -130,10 +130,7 @@ struct layoutdriver_io_operations {
>  	enum pnfs_try_status
>  	(*read_pagelist) (struct nfs_read_data *nfs_data, unsigned nr_pages);
>  	enum pnfs_try_status
> -	(*write_pagelist) (struct pnfs_layout_type *layoutid,
> -			   struct page **pages, unsigned int pgbase,
> -			   unsigned nr_pages, loff_t offset, size_t count,
> -			   int sync, struct nfs_write_data *nfs_data);
> +	(*write_pagelist) (struct nfs_write_data *nfs_data, int how);
>  
>  	/* Consistency ops */
>  	/* 2 problems:

  parent reply	other threads:[~2010-07-20  4:26 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-07-15 16:13 [PATCH 0/7] pnfs-submit layoutdriver api parameter consolidation andros
2010-07-15 16:13 ` [PATCH 1/7] SQUASHME pnfs-submit fix CONFIG_NFS_V4_1 not set compile error andros
2010-07-15 16:13   ` [PATCH 2/7] SQUASHME pnfs-submit consolidate read_pagelist parameters andros
2010-07-15 16:13     ` [PATCH 3/7] SQUASHME pnfs-submit filelayout use new read_pagelist API andros
2010-07-15 16:13       ` [PATCH 4/7] SQUASHME pnfs-submit consolidate write_pagelist parameters andros
2010-07-15 16:13         ` [PATCH 5/7] SQUASHME pnfs-submit filelayout use new write_pagelist API andros
2010-07-15 16:13           ` [PATCH 6/7] SQUASHME pnfs-submit consolidate commit parameters andros
2010-07-15 16:13             ` [PATCH 7/7] SQUASHME pnfs-submit filelayout_commit use new API andros
2010-07-20  4:26         ` Benny Halevy [this message]
2010-07-20  4:33           ` [PATCH 4/7] SQUASHME pnfs-submit consolidate write_pagelist parameters Benny Halevy
2010-07-20 13:51             ` William A. (Andy) Adamson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4C45257D.7000507@panasas.com \
    --to=bhalevy@panasas.com \
    --cc=andros@netapp.com \
    --cc=linux-nfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).