[PATCH v8 6/9] erofs: introduce the page cache share feature

[Hongbo Li]

Hi Xiang,

On 2025/11/17 11:14, Hongbo Li wrote:
> Hi Xiang
> 
> On 2025/11/17 11:06, Gao Xiang wrote:
>>
>>
>> On 2025/11/14 17:55, Hongbo Li wrote:
>>> From: Hongzhen Luo <hongzhen at linux.alibaba.com>
>>>
>>> Currently, reading files with different paths (or names) but the same
>>> content will consume multiple copies of the page cache, even if the
>>> content of these page caches is the same. For example, reading
>>> identical files (e.g., *.so files) from two different minor versions of
>>> container images will cost multiple copies of the same page cache,
>>> since different containers have different mount points. Therefore,
>>> sharing the page cache for files with the same content can save memory.
>>>
>>> This introduces the page cache share feature in erofs. It allocate a
>>> deduplicated inode and use its page cache as shared. Reads for files
>>> with identical content will ultimately be routed to the page cache of
>>> the deduplicated inode. In this way, a single page cache satisfies
>>> multiple read requests for different files with the same contents.
>>>
>>> Signed-off-by: Hongzhen Luo <hongzhen at linux.alibaba.com>
>>> Signed-off-by: Hongbo Li <lihongbo22 at huawei.com>
>>> ---
>>
>> ...
>>
>>
>>> +
>>> +static int erofs_ishare_file_open(struct inode *inode, struct file 
>>> *file)
>>> +{
>>> +    struct file *realfile;
>>> +    struct inode *dedup;
>>> +
>>> +    dedup = EROFS_I(inode)->ishare;
>>> +    if (!dedup)
>>> +        return -EINVAL;
>>> +
>>> +    realfile = alloc_file_pseudo(dedup, erofs_ishare_mnt, 
>>> "erofs_ishare_file",
>>> +                     O_RDONLY, &erofs_file_fops);
>>> +    if (IS_ERR(realfile))
>>> +        return PTR_ERR(realfile);
>>> +
>>> +    file_ra_state_init(&realfile->f_ra, file->f_mapping);
>>> +    realfile->private_data = EROFS_I(inode);
>>> +    file->private_data = realfile;
>>> +    return 0;
>>
> 
> My apologies, I got it wrong. The latest code wasn't synced. The most 
> current version should be this one.
> 
> static int erofs_ishare_file_open(struct inode *inode, struct file *file)
> {
>      struct file *realfile;
>      struct inode *dedup;
>      char *buf, *filepath;
> 
>      dedup = EROFS_I(inode)->ishare;
>      if (!dedup)
>          return -EINVAL;
> 
>      buf = kmalloc(PATH_MAX, GFP_KERNEL);
>      if (!buf)
>          return -ENOMEM;
>      filepath = file_path(file, buf, PATH_MAX);
>      if (IS_ERR(filepath)) {
>          kfree(buf);
>          return -PTR_ERR(filepath);
>      }
>      realfile = alloc_file_pseudo(dedup, erofs_ishare_mnt, filepath + 1,
>                       O_RDONLY, &erofs_file_fops);
>      kfree(buf);
>      if (IS_ERR(realfile))
>          return PTR_ERR(realfile);
> 
>      file_ra_state_init(&realfile->f_ra, file->f_mapping);
>      ihold(dedup);
>      realfile->private_data = EROFS_I(inode);
>      file->private_data = realfile;
>      return 0;
> }
> 
> I changed the "erofs_ishare_file" with filepath + 1 to display the 
> realpath of the original file.

I made this change in patch 7 which caused the misunderstanding here.

Thanks,
Hongbo

> 
> Thanks,
> Hongbo
> 
>> Again, as Amir mentioned before, it should be converted to use (at least)
>> some of backing file interfaces, please see:
>>    file_user_path() and file_user_inode() in include/linux/fs.h
>>
>> Or are you sure /proc/<pid>/maps is shown as expected?
>>
>> Thanks,
>> Gao Xiang