qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Zhang, Chen" <chen.zhang@intel.com>
To: Derek Su <dereksu@qnap.com>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>
Cc: "jasowang@redhat.com" <jasowang@redhat.com>,
	"lizhijian@cn.fujitsu.com" <lizhijian@cn.fujitsu.com>
Subject: RE: [PATCH v1 1/2] colo-compare: Fix incorrect data type conversion
Date: Sun, 13 Sep 2020 20:02:13 +0000	[thread overview]
Message-ID: <3c5d8cfd966a415c896c0d8e814e4ab9@intel.com> (raw)
In-Reply-To: <20200911190508.1316487-2-dereksu@qnap.com>



> -----Original Message-----
> From: Derek Su <dereksu@qnap.com>
> Sent: Saturday, September 12, 2020 3:05 AM
> To: qemu-devel@nongnu.org
> Cc: Zhang, Chen <chen.zhang@intel.com>; lizhijian@cn.fujitsu.com;
> jasowang@redhat.com; Derek Su <dereksu@qnap.com>
> Subject: [PATCH v1 1/2] colo-compare: Fix incorrect data type conversion
> 
> Fix data type conversion of compare_timeout. The incorrect conversion
> results in a random compare_timeout value and unexpected stalls in packet
> comparison.
> 

This bug already found on our internal test too. Just waiting to send.
Good catch! But this patch not fixed the root cause.
Change the compare_timeout from uint32_t to uint64_t is better.
I will send a patch for this and tag reported by you.

Thanks
Zhang Chen


> Signed-off-by: Derek Su <dereksu@qnap.com>
> ---
>  net/colo-compare.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/net/colo-compare.c b/net/colo-compare.c index
> 2c20de1537..c4de86ef34 100644
> --- a/net/colo-compare.c
> +++ b/net/colo-compare.c
> @@ -619,11 +619,12 @@ static int colo_packet_compare_other(Packet *spkt,
> Packet *ppkt)
>                                         ppkt->size - offset);  }
> 
> -static int colo_old_packet_check_one(Packet *pkt, int64_t *check_time)
> +static int colo_old_packet_check_one(Packet *pkt, void *user_data)
>  {
>      int64_t now = qemu_clock_get_ms(QEMU_CLOCK_HOST);
> +    uint32_t check_time = *(uint32_t *)user_data;
> 
> -    if ((now - pkt->creation_ms) > (*check_time)) {
> +    if ((now - pkt->creation_ms) > check_time) {
>          trace_colo_old_packet_check_found(pkt->creation_ms);
>          return 0;
>      } else {
> --
> 2.25.1



  reply	other threads:[~2020-09-13 20:03 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-11 19:05 [PATCH v1 0/2] colo-compare bugfixes Derek Su
2020-09-11 19:05 ` [PATCH v1 1/2] colo-compare: Fix incorrect data type conversion Derek Su
2020-09-13 20:02   ` Zhang, Chen [this message]
2020-09-14  0:31     ` Derek Su
2020-09-11 19:05 ` [PATCH v1 2/2] colo-compare: Record packet creation time by QEMU_CLOCK_REALTIME Derek Su
2020-09-13 20:02   ` Zhang, Chen
2020-09-13 20:06     ` Zhang, Chen
2020-09-14  1:00       ` Derek Su
2020-09-14  7:42         ` Zhang, Chen
2020-09-14 16:23           ` Derek Su
2020-09-15  0:09             ` Zhang, Chen
2020-09-15  1:50               ` Derek Su

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=3c5d8cfd966a415c896c0d8e814e4ab9@intel.com \
    --to=chen.zhang@intel.com \
    --cc=dereksu@qnap.com \
    --cc=jasowang@redhat.com \
    --cc=lizhijian@cn.fujitsu.com \
    --cc=qemu-devel@nongnu.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).