Re: [PATCH v3 1/4] fs: kiocb: introduce IOCB_DONT_DIRTY_PAGE flag for direct IO

[Ming Lei <ming.lei@canonical.com>]

On Tue, May 12, 2015 at 3:24 PM, Christoph Hellwig <hch@infradead.org> wrote:
> On Mon, May 11, 2015 at 09:02:54PM +0800, Ming Lei wrote:
>> > But those would be user pages.  Anyone who passes a kernel page doesn't
>>
>> Inside kernel it is hard to say one page in page cache is kernel or
>> user page, :-)
>
> Inside the kernel it's easy to say that it's the caller business to
> make sure we don't need to carry through hacks like this.
>
>> > expect it do be dirtied for normal kernel interfaces.
>>
>> Currently the coming direct IO introduced for loop should be the 1st direct
>> read inside kernel, so there isn't the case of passing kernel pages, and
>> obviously ITER_BVEC/ITER_KVEC doesn't mean the page is 'kernel' page.
>>
>> Another direct IO inside kernel is __swap_writepage(), and it is still
>> ITER_BVEC and user page. Maybe in the future swap_readpage()
>> can use direct IO with BVEC too.
>
>
> Swap will have very different rules to deal with page dirtying as it's
> part of the VM subsystem.  Again, the caller will now better.

Yes, the caller knows better, that is why I introduce the flag, so
the caller can set the flag if it need direct-io to bypass the dirtying,
otherwise the caller can just clear the flag.  Or I understand you wrong?


Thanks,
Ming Lei