Re: [PATCH 2/2] aio: propogate post-EIOCBQUEUED errors to completion event

From: Zach Brown
Date: Tue Feb 20 2007 - 13:27:28 EST

On Feb 20, 2007, at 9:33 AM, Ananiev, Leonid I wrote:

There is change in the patch which is uncommented in preface.
Now aio_ring_insert_entry() is not called if req->ki_users>=1.
Before was called.
Could you comment it?

It is described in the patch description, though perhaps not explicitly enough:

"To get this working we change the aio_complete() path so that the ring
insertion is performed as the final iocb reference is dropped."

This stops an aio_complete() from generating a completion event before the submission path has had a chance to generate an error. This could happen if O_DIRECT bios completed very quickly before the submission path had a chance to return up and call ivalidate_inode_pages2_range().

- z

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at