All of lore.kernel.org
 help / color / mirror / Atom feed
From: Gao Xiang <hsiangkao@linux.alibaba.com>
To: Yangtao Li <frank.li@vivo.com>, tytso@mit.edu, adilger.kernel@dilger.ca
Cc: linux-ext4@vger.kernel.org, linux-erofs@lists.ozlabs.org,
	linux-kernel@vger.kernel.org, viro@zeniv.linux.org.uk
Subject: Re: [PATCH] ext4: convert to DIV_ROUND_UP() in mpage_process_page_bufs()
Date: Fri, 10 Mar 2023 14:37:11 +0800	[thread overview]
Message-ID: <85434b6a-e7bd-d0c0-654f-233e56112cd0@linux.alibaba.com> (raw)
In-Reply-To: <b47f5003-c7ad-da27-dab6-077f57aaae18@linux.alibaba.com>



On 2023/3/10 14:35, Gao Xiang wrote:
> 
> 
> On 2023/3/10 14:27, Yangtao Li wrote:
>>> Please don't do this.  This makes the code compile down to a division, which is
>>> far less efficient.  I've verified this by checking the assembly generated.
>>
>> How much is the performance impact? So should the following be modified as shift operations as well?
>>
>> fs/erofs/namei.c:92:    int head = 0, back = DIV_ROUND_UP(dir->i_size, EROFS_BLKSIZ) - 1;
>> fs/erofs/zmap.c:252:    const unsigned int totalidx = DIV_ROUND_UP(inode->i_size, EROFS_BLKSIZ);
>> fs/erofs/decompressor.c:14:#define LZ4_MAX_DISTANCE_PAGES       (DIV_ROUND_UP(LZ4_DISTANCE_MAX, PAGE_SIZE) + 1)
>> fs/erofs/decompressor.c:56:                                     DIV_ROUND_UP(distance, PAGE_SIZE) + 1 :
>> fs/erofs/decompressor.c:70:     unsigned long bounced[DIV_ROUND_UP(LZ4_MAX_DISTANCE_PAGES,
>> fs/erofs/data.c:84:     nblocks = DIV_ROUND_UP(inode->i_size, EROFS_BLKSIZ);
> 
> Please stop taking EROFS as example on ext4 patches
> and they will all be changed due to subpage support.

Here EROFS_BLKSIZ == PAGE_SIZE is a constant, so no difference
to use shift or division.

> 
>>
>> Thx,
>> Yangtao

WARNING: multiple messages have this Message-ID (diff)
From: Gao Xiang <hsiangkao@linux.alibaba.com>
To: Yangtao Li <frank.li@vivo.com>, tytso@mit.edu, adilger.kernel@dilger.ca
Cc: linux-ext4@vger.kernel.org, linux-kernel@vger.kernel.org,
	viro@zeniv.linux.org.uk, linux-erofs@lists.ozlabs.org
Subject: Re: [PATCH] ext4: convert to DIV_ROUND_UP() in mpage_process_page_bufs()
Date: Fri, 10 Mar 2023 14:37:11 +0800	[thread overview]
Message-ID: <85434b6a-e7bd-d0c0-654f-233e56112cd0@linux.alibaba.com> (raw)
In-Reply-To: <b47f5003-c7ad-da27-dab6-077f57aaae18@linux.alibaba.com>



On 2023/3/10 14:35, Gao Xiang wrote:
> 
> 
> On 2023/3/10 14:27, Yangtao Li wrote:
>>> Please don't do this.  This makes the code compile down to a division, which is
>>> far less efficient.  I've verified this by checking the assembly generated.
>>
>> How much is the performance impact? So should the following be modified as shift operations as well?
>>
>> fs/erofs/namei.c:92:    int head = 0, back = DIV_ROUND_UP(dir->i_size, EROFS_BLKSIZ) - 1;
>> fs/erofs/zmap.c:252:    const unsigned int totalidx = DIV_ROUND_UP(inode->i_size, EROFS_BLKSIZ);
>> fs/erofs/decompressor.c:14:#define LZ4_MAX_DISTANCE_PAGES       (DIV_ROUND_UP(LZ4_DISTANCE_MAX, PAGE_SIZE) + 1)
>> fs/erofs/decompressor.c:56:                                     DIV_ROUND_UP(distance, PAGE_SIZE) + 1 :
>> fs/erofs/decompressor.c:70:     unsigned long bounced[DIV_ROUND_UP(LZ4_MAX_DISTANCE_PAGES,
>> fs/erofs/data.c:84:     nblocks = DIV_ROUND_UP(inode->i_size, EROFS_BLKSIZ);
> 
> Please stop taking EROFS as example on ext4 patches
> and they will all be changed due to subpage support.

Here EROFS_BLKSIZ == PAGE_SIZE is a constant, so no difference
to use shift or division.

> 
>>
>> Thx,
>> Yangtao

  reply	other threads:[~2023-03-10  6:37 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-10  6:07 [PATCH] ext4: convert to DIV_ROUND_UP() in mpage_process_page_bufs() Yangtao Li
2023-03-10  6:17 ` Eric Biggers
2023-03-10  6:27   ` Yangtao Li
2023-03-10  6:27     ` Yangtao Li
2023-03-10  6:35     ` Gao Xiang
2023-03-10  6:35       ` Gao Xiang
2023-03-10  6:37       ` Gao Xiang [this message]
2023-03-10  6:37         ` Gao Xiang
2023-03-10  6:41     ` Eric Biggers
2023-03-10  6:41       ` Eric Biggers
2023-03-10  6:37   ` Al Viro
2023-03-10  6:43     ` Al Viro
2023-03-10  6:43     ` Eric Biggers
2023-03-10  6:46       ` Al Viro
2023-03-10  6:54         ` Eric Biggers
2023-03-10  7:05           ` Al Viro
2023-03-10  7:12             ` Al Viro
2023-03-10 21:47               ` Eric Biggers
2023-03-10 19:07 ` kernel test robot
2023-03-10 19:38 ` kernel test robot
2023-03-10 23:21 ` Theodore Ts'o
2023-03-13  9:25   ` David Laight

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=85434b6a-e7bd-d0c0-654f-233e56112cd0@linux.alibaba.com \
    --to=hsiangkao@linux.alibaba.com \
    --cc=adilger.kernel@dilger.ca \
    --cc=frank.li@vivo.com \
    --cc=linux-erofs@lists.ozlabs.org \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=tytso@mit.edu \
    --cc=viro@zeniv.linux.org.uk \
    /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.