From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([208.118.235.92]:51242)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <famz@redhat.com>) id 1UfiFe-0001FP-5j
	for qemu-devel@nongnu.org; Thu, 23 May 2013 23:01:09 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <famz@redhat.com>) id 1UfiFa-0006hF-V3
	for qemu-devel@nongnu.org; Thu, 23 May 2013 23:01:06 -0400
Received: from mx1.redhat.com ([209.132.183.28]:60357)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <famz@redhat.com>) id 1UfiFa-0006h3-Mz
	for qemu-devel@nongnu.org; Thu, 23 May 2013 23:01:02 -0400
Date: Fri, 24 May 2013 11:00:56 +0800
From: Fam Zheng <famz@redhat.com>
Message-ID: <20130524030056.GD1705@localhost.nay.redhat.com>
References: <1369280289-20928-1-git-send-email-famz@redhat.com>
	<1369280289-20928-7-git-send-email-famz@redhat.com>
	<20130523140929.GQ9093@stefanha-thinkpad.redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20130523140929.GQ9093@stefanha-thinkpad.redhat.com>
Subject: Re: [Qemu-devel] [PATCH v5 06/11] curl: introduce CURLDataCache
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Stefan Hajnoczi <stefanha@gmail.com>
Cc: kwolf@redhat.com, jcody@redhat.com, qemu-devel@nongnu.org, stefanha@redhat.com

On Thu, 05/23 16:09, Stefan Hajnoczi wrote:
> On Thu, May 23, 2013 at 11:38:04AM +0800, Fam Zheng wrote:
> > +typedef struct CURLDataCache {
> > +    char *data;
> > +    size_t base_pos;
> 
> Must be int64_t.  QEMU compiled on 32-bit hosts would only allow 4 GB
> images with size_t!

OK.

> 
> > +    size_t data_len;
> > +    size_t write_pos;
> > +    /* Ref count for CURLState */
> > +    int use_count;
> 
> It's better to introduce this field when you add code to use it.  When
> possible, don't add unused code in a patch, it makes it harder to
> review.

Moving to later patch.

> 
> > +static void curl_complete_io(BDRVCURLState *bs, CURLAIOCB *acb,
> > +                             CURLDataCache *cache)
> > +{
> > +    size_t aio_base = acb->sector_num * SECTOR_SIZE;
> 
> int64_t
> 
> > +    size_t aio_bytes = acb->nb_sectors * SECTOR_SIZE;
> > +    size_t off = aio_base - cache->base_pos;
> > +
> > +    qemu_iovec_from_buf(acb->qiov, 0, cache->data + off, aio_bytes);
> > +    acb->common.cb(acb->common.opaque, 0);
> > +    DPRINTF("AIO Request OK: %10zd %10zd\n", aio_base, aio_bytes);
> 
> PRId64 for 64-bit aio_base

OK, thanks.

> 
> > @@ -589,26 +577,24 @@ static const AIOCBInfo curl_aiocb_info = {
> >  static void curl_readv_bh_cb(void *p)
> >  {
> >      CURLState *state;
> > -
> > +    CURLDataCache *cache = NULL;
> >      CURLAIOCB *acb = p;
> >      BDRVCURLState *s = acb->common.bs->opaque;
> > +    size_t aio_base, aio_bytes;
> 
> int64_t aio_base;

Yes, will change.

-- 
Fam