Re: [PATCH 3/6] fuse: Allow parallel direct writes for O_DIRECT

[Bernd Schubert <bernd.schubert@fastmail.fm>]

On 8/31/23 10:30, Hao Xu wrote:
> On 8/30/23 00:11, Bernd Schubert wrote:
>> Take a shared lock in fuse_cache_write_iter. This was already
>> done for FOPEN_DIRECT_IO in
>>
>> commit 153524053bbb ("fuse: allow non-extending parallel direct
>> writes on the same file")
>>
>> but so far missing for plain O_DIRECT. Server side needs
>> to set FOPEN_PARALLEL_DIRECT_WRITES in order to signal that
>> it supports parallel dio writes.
>>
>> Cc: Hao Xu <howeyxu@tencent.com>
>> Cc: Miklos Szeredi <miklos@szeredi.hu>
>> Cc: Dharmendra Singh <dsingh@ddn.com>
>> Cc: linux-fsdevel@vger.kernel.org
>> Signed-off-by: Bernd Schubert <bschubert@ddn.com>
>> ---
>>   fs/fuse/file.c | 18 ++++++++++++++++--
>>   1 file changed, 16 insertions(+), 2 deletions(-)
>>
>> diff --git a/fs/fuse/file.c b/fs/fuse/file.c
>> index 6b8b9512c336..a6b99bc80fe7 100644
>> --- a/fs/fuse/file.c
>> +++ b/fs/fuse/file.c
>> @@ -1314,6 +1314,10 @@ static bool fuse_dio_wr_exclusive_lock(struct 
>> kiocb *iocb, struct iov_iter *from
>>       struct file *file = iocb->ki_filp;
>>       struct fuse_file *ff = file->private_data;
>> +    /* this function is about direct IO only */
>> +    if (!(iocb->ki_flags & IOCB_DIRECT))
>> +        return false;
> 
> This means for buffered write in fuse_cache_write_iter(), we grab shared 
> lock, looks not right.
> 

Yeah, sorry, I made all values the other way around, consistently. 
Miklos had already noticed.


Thanks,
Bernd