Re: [Qemu-devel] [PATCH] block/iscsi: use a bh to schedule co reentrance

[Paolo Bonzini]

Il 14/12/2013 17:31, Peter Lieven ha scritto:
> this fixes a potential segfault and performance regression.
> 
> If the coroutine is reentered directly in the iscsi_co_generic_cb
> iscsi_process_{read,write} are interrupted and reentered any
> time later. One the one hand this could happen after an iscsi_close
> where the iscsi context is already gone (segfault). On the
> other hand this limits the number of processed callbacks
> in each aio_dispatch to one (potential performance regression).
> 
> Cc: address@hidden
> Signed-off-by: Peter Lieven <address@hidden>
> ---
>  block/iscsi.c |   11 ++++++++++-
>  1 file changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/block/iscsi.c b/block/iscsi.c
> index fa69408..b0e6eea 100644
> --- a/block/iscsi.c
> +++ b/block/iscsi.c
> @@ -68,6 +68,7 @@ typedef struct IscsiTask {
>      int do_retry;
>      struct scsi_task *task;
>      Coroutine *co;
> +    QEMUBH *bh;
>  } IscsiTask;
>  
>  typedef struct IscsiAIOCB {
> @@ -123,6 +124,13 @@ iscsi_schedule_bh(IscsiAIOCB *acb)
>      qemu_bh_schedule(acb->bh);
>  }
>  
> +static void iscsi_co_generic_bh_cb(void *opaque)
> +{
> +    struct IscsiTask *iTask = opaque;
> +    qemu_bh_delete(iTask->bh);
> +    qemu_coroutine_enter(iTask->co, NULL);
> +}
> +
>  static void
>  iscsi_co_generic_cb(struct iscsi_context *iscsi, int status,
>                          void *command_data, void *opaque)
> @@ -147,7 +155,8 @@ iscsi_co_generic_cb(struct iscsi_context *iscsi, int 
> status,
>  
>  out:
>      if (iTask->co) {
> -        qemu_coroutine_enter(iTask->co, NULL);
> +        iTask->bh = qemu_bh_new(iscsi_co_generic_bh_cb, iTask);
> +        qemu_bh_schedule(iTask->bh);
>      }
>  }
>  
> 

Applied to scsi-next branch, thanks.

Paolo