All of lore.kernel.org
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: Fam Zheng <famz@redhat.com>, qemu-devel@nongnu.org
Cc: Kevin Wolf <kwolf@redhat.com>, Stefan Hajnoczi <stefanha@redhat.com>
Subject: Re: [Qemu-devel] [RFC PATCH 1/9] block: Add bdrv_aio_cancel_async
Date: Thu, 21 Aug 2014 14:14:01 +0200	[thread overview]
Message-ID: <53F5E289.8050703@redhat.com> (raw)
In-Reply-To: <1408622216-9578-2-git-send-email-famz@redhat.com>

Il 21/08/2014 13:56, Fam Zheng ha scritto:
> +        BlockDriverAIOCB *save = g_new(BlockDriverAIOCB, 1);
> +        save->cb               = acb->cb;
> +        save->opaque           = acb->opaque;
> +        acb->cb                = bdrv_aio_cancel_async_cb;
> +        acb->opaque            = acb;
> +        acb->cancel_acb_save   = save;

No need for this extra field.  You can make a struct with {acb->cb,
acb->opaque, acb} and pass it as the opaque.  You also do not need to
allocate it on the heap, since everything is synchronous.

> +
> +        /* Use the synchronous version and hope our cb is called. */
> +        acb->aiocb_info->cancel(acb);

Unfortunately, acb has been freed at this point, so you'll be accessing
a dangling pointer.  I think you need to modify all cancel
implementations.  For example:

- return whether they have called the callback

- only free the acb if they have called the callback

- otherwise, free the acb in bdrv_aio_cancel

Paolo

> +        if (acb->cancel_acb_save) {
> +            /* cb is not called yet, let's call it */
> +            bdrv_aio_cancel_async_cb(acb->opaque, -ECANCELED);
> +        }
> +    }
> +}
> +
>  /**************************************************************/
>  /* async block device emulation */
>  
> diff --git a/include/block/aio.h b/include/block/aio.h
> index c23de3c..fcc5c87 100644
> --- a/include/block/aio.h
> +++ b/include/block/aio.h
> @@ -27,6 +27,7 @@ typedef void BlockDriverCompletionFunc(void *opaque, int ret);
>  
>  typedef struct AIOCBInfo {
>      void (*cancel)(BlockDriverAIOCB *acb);
> +    void (*cancel_async)(BlockDriverAIOCB *acb);
>      size_t aiocb_size;
>  } AIOCBInfo;
>  
> @@ -35,6 +36,7 @@ struct BlockDriverAIOCB {
>      BlockDriverState *bs;
>      BlockDriverCompletionFunc *cb;
>      void *opaque;
> +    BlockDriverAIOCB *cancel_acb_save;
>  };

  reply	other threads:[~2014-08-21 12:14 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-08-21 11:56 [Qemu-devel] [RFC PATCH 0/9] scsi, block: Asynchronous request cancellation Fam Zheng
2014-08-21 11:56 ` [Qemu-devel] [RFC PATCH 1/9] block: Add bdrv_aio_cancel_async Fam Zheng
2014-08-21 12:14   ` Paolo Bonzini [this message]
2014-08-22  1:23     ` Fam Zheng
2014-08-22  8:14       ` Paolo Bonzini
2014-08-22  9:37         ` Fam Zheng
2014-08-22 10:10           ` Paolo Bonzini
2014-08-22 10:51             ` Fam Zheng
2014-08-21 13:44   ` Stefan Hajnoczi
2014-08-21 11:56 ` [Qemu-devel] [RFC PATCH 2/9] tests: Add testing code for bdrv_aio_cancel_async Fam Zheng
2014-08-21 11:56 ` [Qemu-devel] [RFC PATCH 3/9] iscsi: Implement .cancel_async in acb info Fam Zheng
2014-08-21 11:56 ` [Qemu-devel] [RFC PATCH 4/9] linux-aio: Implement .cancel_async Fam Zheng
2014-08-21 16:31   ` Stefan Hajnoczi
2014-08-22  4:56     ` Fam Zheng
2014-08-21 11:56 ` [Qemu-devel] [RFC PATCH 5/9] thread-pool: " Fam Zheng
2014-08-21 11:56 ` [Qemu-devel] [RFC PATCH 6/9] blkdebug: " Fam Zheng
2014-08-21 16:52   ` Stefan Hajnoczi
2014-08-22  4:29     ` Fam Zheng
2014-08-21 11:56 ` [Qemu-devel] [RFC PATCH 7/9] dma: " Fam Zheng
2014-08-21 11:56 ` [Qemu-devel] [RFC PATCH 8/9] block: Implement stub bdrv_em_co_aiocb_info.cancel_async Fam Zheng
2014-08-21 17:01   ` Stefan Hajnoczi
2014-08-22  4:28     ` Fam Zheng
2014-08-21 11:56 ` [Qemu-devel] [RFC PATCH 9/9] scsi: Cancel request asynchronously Fam Zheng
2014-08-21 12:19   ` Paolo Bonzini
2014-08-22  4:57     ` Fam Zheng

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=53F5E289.8050703@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=famz@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.