Re: [PATCH 1/2] erofs: clean up erofs_iget()

[Xiang Gao <hsiangkao@linux.alibaba.com>]

On 2023/1/13 15:41, Jingbo Xu wrote:
> 
> 
> On 1/13/23 2:52 PM, Gao Xiang wrote:
>> Move inode hash function into inode.c and simplify erofs_iget().
>>
>> Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
>> ---
>>   fs/erofs/inode.c    | 40 +++++++++++++++++++++-------------------
>>   fs/erofs/internal.h |  9 ---------
>>   2 files changed, 21 insertions(+), 28 deletions(-)
>>
>> diff --git a/fs/erofs/inode.c b/fs/erofs/inode.c
>> index d3b8736fa124..57328691582e 100644
>> --- a/fs/erofs/inode.c
>> +++ b/fs/erofs/inode.c
>> @@ -308,47 +308,49 @@ static int erofs_fill_inode(struct inode *inode)
>>   }
>>   
>>   /*
>> - * erofs nid is 64bits, but i_ino is 'unsigned long', therefore
>> - * we should do more for 32-bit platform to find the right inode.
>> + * ino_t is 32-bits on 32-bit arch. We have to squash the 64-bit value down
>> + * so that it will fit.
>>    */
>> -static int erofs_ilookup_test_actor(struct inode *inode, void *opaque)
>> +static ino_t erofs_squash_ino(erofs_nid_t nid)
>>   {
>> -	const erofs_nid_t nid = *(erofs_nid_t *)opaque;
>> +	ino_t ino = (ino_t)nid;
>> +
>> +	if (sizeof(ino_t) < sizeof(erofs_nid_t))
>> +		ino ^= nid >> (sizeof(erofs_nid_t) - sizeof(ino_t)) * 8;
> 
> Shouldn't we do:
> 
> 	ino ^= nid >> sizeof(ino_t) * 8
Actually I copied it from fuse, for 64-bit erofs_nid_t it has no difference
though. I will also update it as your suggestion in v2.

Thanks,
Gao Xiang


> ?
> 
>