Re: correction: fs/buffer.c underlocking async pages

Chris Mason <[e-mail]>
06/21/01 08:20 PM
Please respond to Chris Mason

        To:     Andrea Arcangeli <[e-mail]>, Linus Torvalds
<[e-mail]>
        cc:     Stefan Bader/Germany/IBM@IBMDE,
[e-mail]
        Subject:        Re: correction: fs/buffer.c underlocking async
pages

On Thursday, June 21, 2001 07:15:22 PM +0200 Andrea Arcangeli
<[e-mail]> wrote:

>> On Thu, Jun 21, 2001 at 09:56:04AM -0700, Linus Torvalds wrote:
>>  What's the problem with the existing code, and why do people want to
add
>> a
>>> (unnecessary) new bit?
>>
>> there's no problem with the existing code, what I understood is that
>> they cannot overwrite the ->b_end_io callback in the lowlevel blkdev
>> layer or the page will be unlocked too early.

>Just to be more explicit, the big problem is mixing different async
>callbacks on the same page.  The patch would also allow things like this:
>
>fs_specific_end_io() {
>    do something special
>    end_buffer_io_async()
>}
>
>-chris

Yes, that was exactly the thing I tried to do. In my case some sort of
bookkeeping
how many IO's where mapped (to a certain path) and how many came back.
My assumption first was, if I am restoring the old pointers before I call
the original
function it should work.
After running into problems this patch was just my quick hack to try out
whether this
was the only place that I did not think of. I wouldn't insist on the exact
way to do it,
but since it worked for me, I thought it might be worth to discuss (or
even be useful
for other people... ;-)).

- Stefan

Linux for eServer development
[e-mail]
Phone: [phone]
----------------------------------------------------------------------------------

  When all other means of communication fail, try words.