All of lore.kernel.org
 help / color / mirror / Atom feed
From: Gao Xiang <xiang@kernel.org>
To: Jeffle Xu <jefflexu@linux.alibaba.com>
Cc: tianzichen@kuaishou.com, linux-erofs@lists.ozlabs.org,
	fannaihao@baidu.com, willy@infradead.org,
	linux-kernel@vger.kernel.org, dhowells@redhat.com,
	joseph.qi@linux.alibaba.com, linux-cachefs@redhat.com,
	gregkh@linuxfoundation.org, linux-fsdevel@vger.kernel.org,
	luodaowen.backend@bytedance.com, gerry@linux.alibaba.com,
	torvalds@linux-foundation.org
Subject: Re: [PATCH v8 18/20] erofs: implement fscache-based data read for inline layout
Date: Thu, 7 Apr 2022 22:31:54 +0800	[thread overview]
Message-ID: <Yk712oM2xrEUmbhZ@debian> (raw)
In-Reply-To: <20220406075612.60298-19-jefflexu@linux.alibaba.com>

On Wed, Apr 06, 2022 at 03:56:10PM +0800, Jeffle Xu wrote:
> Implement the data plane of reading data from data blobs over fscache
> for inline layout.
> 
> For the heading non-inline part, the data plane for non-inline layout is
> reused, while only the tail packing part needs special handling.
> 
> Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com>
> ---
>  fs/erofs/fscache.c | 40 ++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 40 insertions(+)
> 
> diff --git a/fs/erofs/fscache.c b/fs/erofs/fscache.c
> index 65de1c754e80..d32cb5840c6d 100644
> --- a/fs/erofs/fscache.c
> +++ b/fs/erofs/fscache.c
> @@ -60,6 +60,40 @@ static int erofs_fscache_meta_readpage(struct file *data, struct page *page)
>  	return ret;
>  }
>  
> +static int erofs_fscache_readpage_inline(struct folio *folio,
> +					 struct erofs_map_blocks *map)
> +{
> +	struct inode *inode = folio_file_mapping(folio)->host;
> +	struct super_block *sb = inode->i_sb;
> +	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
> +	erofs_blk_t blknr;
> +	size_t offset, len;
> +	void *src, *dst;
> +
> +	/*
> +	 * For inline (tail packing) layout, the offset may be non-zero, which
> +	 * can be calculated from corresponding physical address directly.
> +	 */
> +	offset = erofs_blkoff(map->m_pa);
> +	blknr = erofs_blknr(map->m_pa);
> +	len = map->m_llen;
> +
> +	src = erofs_read_metabuf(&buf, sb, blknr, EROFS_KMAP);
> +	if (IS_ERR(src))
> +		return PTR_ERR(src);
> +
> +	DBG_BUGON(folio_size(folio) != PAGE_SIZE);
> +
> +	dst = kmap(folio_page(folio, 0));

kmap_local_folio?

> +	memcpy(dst, src + offset, len);
> +	memset(dst + len, 0, PAGE_SIZE - len);
> +	kunmap(folio_page(folio, 0));
> +
> +	erofs_put_metabuf(&buf);
> +
> +	return 0;
> +}
> +
>  static int erofs_fscache_readpage(struct file *file, struct page *page)
>  {
>  	struct folio *folio = page_folio(page);
> @@ -85,6 +119,12 @@ static int erofs_fscache_readpage(struct file *file, struct page *page)
>  		goto out_uptodate;
>  	}
>  
> +	/* inline readpage */

I think the code below is self-explained.

> +	if (map.m_flags & EROFS_MAP_META) {
> +		ret = erofs_fscache_readpage_inline(folio, &map);
> +		goto out_uptodate;
> +	}
> +
>  	/* no-inline readpage */

Same here.

Thanks,
Gao Xiang

>  	mdev = (struct erofs_map_dev) {
>  		.m_deviceid = map.m_deviceid,
> -- 
> 2.27.0
> 

WARNING: multiple messages have this Message-ID (diff)
From: Gao Xiang <xiang@kernel.org>
To: Jeffle Xu <jefflexu@linux.alibaba.com>
Cc: dhowells@redhat.com, linux-cachefs@redhat.com, xiang@kernel.org,
	chao@kernel.org, linux-erofs@lists.ozlabs.org,
	torvalds@linux-foundation.org, gregkh@linuxfoundation.org,
	willy@infradead.org, linux-fsdevel@vger.kernel.org,
	joseph.qi@linux.alibaba.com, bo.liu@linux.alibaba.com,
	tao.peng@linux.alibaba.com, gerry@linux.alibaba.com,
	eguan@linux.alibaba.com, linux-kernel@vger.kernel.org,
	luodaowen.backend@bytedance.com, tianzichen@kuaishou.com,
	fannaihao@baidu.com
Subject: Re: [PATCH v8 18/20] erofs: implement fscache-based data read for inline layout
Date: Thu, 7 Apr 2022 22:31:54 +0800	[thread overview]
Message-ID: <Yk712oM2xrEUmbhZ@debian> (raw)
In-Reply-To: <20220406075612.60298-19-jefflexu@linux.alibaba.com>

On Wed, Apr 06, 2022 at 03:56:10PM +0800, Jeffle Xu wrote:
> Implement the data plane of reading data from data blobs over fscache
> for inline layout.
> 
> For the heading non-inline part, the data plane for non-inline layout is
> reused, while only the tail packing part needs special handling.
> 
> Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com>
> ---
>  fs/erofs/fscache.c | 40 ++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 40 insertions(+)
> 
> diff --git a/fs/erofs/fscache.c b/fs/erofs/fscache.c
> index 65de1c754e80..d32cb5840c6d 100644
> --- a/fs/erofs/fscache.c
> +++ b/fs/erofs/fscache.c
> @@ -60,6 +60,40 @@ static int erofs_fscache_meta_readpage(struct file *data, struct page *page)
>  	return ret;
>  }
>  
> +static int erofs_fscache_readpage_inline(struct folio *folio,
> +					 struct erofs_map_blocks *map)
> +{
> +	struct inode *inode = folio_file_mapping(folio)->host;
> +	struct super_block *sb = inode->i_sb;
> +	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
> +	erofs_blk_t blknr;
> +	size_t offset, len;
> +	void *src, *dst;
> +
> +	/*
> +	 * For inline (tail packing) layout, the offset may be non-zero, which
> +	 * can be calculated from corresponding physical address directly.
> +	 */
> +	offset = erofs_blkoff(map->m_pa);
> +	blknr = erofs_blknr(map->m_pa);
> +	len = map->m_llen;
> +
> +	src = erofs_read_metabuf(&buf, sb, blknr, EROFS_KMAP);
> +	if (IS_ERR(src))
> +		return PTR_ERR(src);
> +
> +	DBG_BUGON(folio_size(folio) != PAGE_SIZE);
> +
> +	dst = kmap(folio_page(folio, 0));

kmap_local_folio?

> +	memcpy(dst, src + offset, len);
> +	memset(dst + len, 0, PAGE_SIZE - len);
> +	kunmap(folio_page(folio, 0));
> +
> +	erofs_put_metabuf(&buf);
> +
> +	return 0;
> +}
> +
>  static int erofs_fscache_readpage(struct file *file, struct page *page)
>  {
>  	struct folio *folio = page_folio(page);
> @@ -85,6 +119,12 @@ static int erofs_fscache_readpage(struct file *file, struct page *page)
>  		goto out_uptodate;
>  	}
>  
> +	/* inline readpage */

I think the code below is self-explained.

> +	if (map.m_flags & EROFS_MAP_META) {
> +		ret = erofs_fscache_readpage_inline(folio, &map);
> +		goto out_uptodate;
> +	}
> +
>  	/* no-inline readpage */

Same here.

Thanks,
Gao Xiang

>  	mdev = (struct erofs_map_dev) {
>  		.m_deviceid = map.m_deviceid,
> -- 
> 2.27.0
> 

  reply	other threads:[~2022-04-07 14:32 UTC|newest]

Thread overview: 110+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-06  7:55 [PATCH v8 00/20] fscache, erofs: fscache-based on-demand read semantics Jeffle Xu
2022-04-06  7:55 ` [PATCH v8 00/20] fscache,erofs: " Jeffle Xu
2022-04-06  7:55 ` [PATCH v8 01/20] cachefiles: unmark inode in use in error path Jeffle Xu
2022-04-06  7:55   ` Jeffle Xu
2022-04-06  7:55 ` [PATCH v8 02/20] cachefiles: extract write routine Jeffle Xu
2022-04-06  7:55   ` Jeffle Xu
2022-04-11 12:13   ` David Howells
2022-04-11 12:13     ` David Howells
2022-04-11 12:29     ` JeffleXu
2022-04-11 12:29       ` JeffleXu
2022-04-06  7:55 ` [PATCH v8 03/20] cachefiles: notify user daemon with anon_fd when looking up cookie Jeffle Xu
2022-04-06  7:55   ` Jeffle Xu
2022-04-11 12:28   ` David Howells
2022-04-11 12:28     ` David Howells
2022-04-11 12:36     ` JeffleXu
2022-04-11 12:36       ` JeffleXu
2022-04-11 12:32   ` David Howells
2022-04-11 12:32     ` David Howells
2022-04-11 12:36     ` JeffleXu
2022-04-11 12:36       ` JeffleXu
2022-04-06  7:55 ` [PATCH v8 04/20] cachefiles: notify user daemon when withdrawing cookie Jeffle Xu
2022-04-06  7:55   ` Jeffle Xu
2022-04-11 12:35   ` David Howells
2022-04-11 12:35     ` David Howells
2022-04-11 12:48     ` JeffleXu
2022-04-11 12:48       ` JeffleXu
2022-04-11 13:42       ` David Howells
2022-04-11 13:42         ` David Howells
2022-04-12  3:35         ` JeffleXu
2022-04-12  3:35           ` JeffleXu
2022-04-06  7:55 ` [PATCH v8 05/20] cachefiles: implement on-demand read Jeffle Xu
2022-04-06  7:55   ` Jeffle Xu
2022-04-11 12:44   ` David Howells
2022-04-11 12:44     ` David Howells
2022-04-11 12:50     ` JeffleXu
2022-04-11 12:50       ` JeffleXu
2022-04-06  7:55 ` [PATCH v8 06/20] cachefiles: enable on-demand read mode Jeffle Xu
2022-04-06  7:55   ` Jeffle Xu
2022-04-06  7:55 ` [PATCH v8 07/20] cachefiles: document " Jeffle Xu
2022-04-06  7:55   ` Jeffle Xu
2022-04-11 13:38   ` David Howells
2022-04-11 13:38     ` David Howells
2022-04-12  3:17     ` JeffleXu
2022-04-12  3:17       ` JeffleXu
2022-04-06  7:56 ` [PATCH v8 08/20] erofs: make erofs_map_blocks() generally available Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07  2:44   ` Gao Xiang
2022-04-07  2:44     ` Gao Xiang
2022-04-06  7:56 ` [PATCH v8 09/20] erofs: add mode checking helper Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07  2:46   ` Gao Xiang
2022-04-07  2:46     ` Gao Xiang
2022-04-06  7:56 ` [PATCH v8 10/20] erofs: register fscache volume Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07  2:50   ` Gao Xiang
2022-04-07  2:50     ` Gao Xiang
2022-04-06  7:56 ` [PATCH v8 11/20] erofs: add fscache context helper functions Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07  3:25   ` Gao Xiang
2022-04-07  3:25     ` Gao Xiang
2022-04-06  7:56 ` [PATCH v8 12/20] erofs: add anonymous inode managing page cache for data blob Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07  5:31   ` Gao Xiang
2022-04-07  5:31     ` Gao Xiang
2022-04-08  2:14     ` JeffleXu
2022-04-06  7:56 ` [PATCH v8 13/20] erofs: add erofs_fscache_read_folios() helper Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07 14:05   ` Gao Xiang
2022-04-07 14:05     ` Gao Xiang
2022-04-06  7:56 ` [PATCH v8 14/20] erofs: register fscache context for primary data blob Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07 14:09   ` Gao Xiang
2022-04-07 14:09     ` Gao Xiang
2022-04-06  7:56 ` [PATCH v8 15/20] erofs: register fscache context for extra data blobs Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07 14:15   ` Gao Xiang
2022-04-07 14:15     ` Gao Xiang
2022-04-08  2:11     ` JeffleXu
2022-04-06  7:56 ` [PATCH v8 16/20] erofs: implement fscache-based metadata read Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07 14:19   ` Gao Xiang
2022-04-07 14:19     ` Gao Xiang
2022-04-06  7:56 ` [PATCH v8 17/20] erofs: implement fscache-based data read for non-inline layout Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07 14:24   ` Gao Xiang
2022-04-07 14:24     ` Gao Xiang
2022-04-06  7:56 ` [PATCH v8 18/20] erofs: implement fscache-based data read for inline layout Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07 14:31   ` Gao Xiang [this message]
2022-04-07 14:31     ` Gao Xiang
2022-04-06  7:56 ` [PATCH v8 19/20] erofs: implement fscache-based data readahead Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07 14:36   ` Gao Xiang
2022-04-07 14:36     ` Gao Xiang
2022-04-06  7:56 ` [PATCH v8 20/20] erofs: add 'fsid' mount option Jeffle Xu
2022-04-06  7:56   ` Jeffle Xu
2022-04-07 14:39   ` Gao Xiang
2022-04-07 14:39     ` Gao Xiang
2022-04-10 12:51 ` [PATCH v8 00/20] fscache,erofs: fscache-based on-demand read semantics Gao Xiang
2022-04-10 12:51   ` Gao Xiang
2022-04-13 12:27   ` 田子晨
2022-04-13 12:27     ` 田子晨
2022-04-14  8:10   ` Re: [PATCH v8 00/20] fscache, erofs: " Jiachen Zhang
2022-04-14  8:10     ` Re: [PATCH v8 00/20] fscache,erofs: " Jiachen Zhang
2022-04-14  9:29     ` Re: [PATCH v8 00/20] fscache, erofs: " Gao Xiang
2022-04-14  9:29       ` Re: [PATCH v8 00/20] fscache,erofs: " Gao Xiang
2022-04-11 13:43 ` [PATCH v8 00/20] fscache, erofs: " David Howells
2022-04-11 13:43   ` [PATCH v8 00/20] fscache,erofs: " David Howells
2022-04-12  3:18   ` JeffleXu
2022-04-12  3:18     ` JeffleXu

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=Yk712oM2xrEUmbhZ@debian \
    --to=xiang@kernel.org \
    --cc=dhowells@redhat.com \
    --cc=fannaihao@baidu.com \
    --cc=gerry@linux.alibaba.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=jefflexu@linux.alibaba.com \
    --cc=joseph.qi@linux.alibaba.com \
    --cc=linux-cachefs@redhat.com \
    --cc=linux-erofs@lists.ozlabs.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luodaowen.backend@bytedance.com \
    --cc=tianzichen@kuaishou.com \
    --cc=torvalds@linux-foundation.org \
    --cc=willy@infradead.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.