Re: [RFC] Stacking bio support

[Ph. Marek]

Hello Daniel!

On Dienstag, 11. März 2008, Daniel Phillips wrote:
> 3: bio.stack-2.6.23.12
>
> This adds an internal stack to each struct bio and introduces two new
> bio operations:
>
>         data = bio_push(bio, worksize, endio);
>         data = bio_pop(bio);
>
> The first is used before submitting a bio and the second is used in the
> endio handler, which gives the driver a nice way to share context
> between the two events.  If the requested amount stack space is not
> available in the bio, the bio stack is automatically extended.
...
> (I learned from Andreas Dilger last month at FAST that Lustre already
> implements a mechanism along these lines.)
Win32 has IRP stacks, which do mostly the same AFAIU.
	http://msdn2.microsoft.com/en-us/library/ms796144.aspx

How do you handle the reallocation?
- If you don't do it (but rely on the fact that the initial allocation is
  enough), you might end up with NO_MORE_IRP_STACK_LOCATIONS
    http://msdn2.microsoft.com/en-us/library/ms793675.aspx
- If you do reallocate, the allocations have to register themselves in
  the emergency pool (see the current thread about swapping over NFS)

I don't say that it's impossible ... just that some "interesting" things will 
await you. 

> Currently the stack size in the bio is set to zero and is always
> extended on the first bio_push.  An upcoming revision will add a
> mechanism for a block driver to specify the initial amount of stack
> space it knows will always be needed.  Some time further in the future,
> a mechanism for discovering the stack requirements of several block
> devices in a stack may be added, so that a typical bio submission is
> able to traverse the entire stack with only a single bio allocation.
That's different from the Win32 way AFAIK - there it's defined that every 
layer *has* to use its own stack location. (But it's been some time since I 
needed that, so I might be wrong.)


But I sure hope you succeed!


Regards,

Phil
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/