Re: [PATCH v3] fuse: invalidate the page cache after direct write

[Bernd Schubert]

On 1/11/26 09:18, Markus Elfring wrote:
> [You don't often get email from markus.elfring@xxxxxx. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ]
> 
> …
>> +++ b/fs/fuse/file.c
>> @@ -667,6 +667,18 @@ static void fuse_aio_complete(struct fuse_io_priv *io, int err, ssize_t pos)
>>                       struct inode *inode = file_inode(io->iocb->ki_filp);
>>                       struct fuse_conn *fc = get_fuse_conn(inode);
>>                       struct fuse_inode *fi = get_fuse_inode(inode);
>> +                     struct address_space *mapping = io->iocb->ki_filp->f_mapping;
>> +
>> +                     /*
>> +                      * As in generic_file_direct_write(), invalidate after the
>> +                      * write, to invalidate read-ahead cache that may have competed
>> +                      * with the write.
>> +                      */
>> +                     if (io->write && res && mapping->nrpages) {
>> +                             invalidate_inode_pages2_range(mapping,
>> +                                             io->offset >> PAGE_SHIFT,
>> +                                             (io->offset + res - 1) >> PAGE_SHIFT);
>> +                     }
>>
>>                       spin_lock(&fi->lock);
> …
>> @@ -1160,10 +1174,26 @@ static ssize_t fuse_send_write(struct fuse_io_args *ia, loff_t pos,
> …
> -       return err ?: ia->write.out.size;
>> +     /*
>> +      * Without FOPEN_DIRECT_IO, generic_file_direct_write() does the
>> +      * invalidation for us.
>> +      */
>> +     if (!err && written && mapping->nrpages &&
>> +         (ff->open_flags & FOPEN_DIRECT_IO)) {
>> +             /*
>> +              * As in generic_file_direct_write(), invalidate after the
>> +              * write, to invalidate read-ahead cache that may have competed
>> +              * with the write.
>> +              */
>> +             invalidate_inode_pages2_range(mapping, pos >> PAGE_SHIFT,
>> +                                     (pos + written - 1) >> PAGE_SHIFT);
>> +     }
>> +
>> +     return err ?: written;
> …
> 
> You may omit curly brackets at selected source code places.
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/coding-style.rst?h=v6.19-rc4#n197

We could, but we could also concentrate on code correctness instead of
nit-picky debatable code style.

Thanks,
Bernd