Re: [Qemu-devel] [PATCH] vnc: fix segfault in closed connection handling

2018-02-14 Thread Daniel P . Berrangé
On Wed, Feb 07, 2018 at 12:48:44PM +0300, Klim Kireev wrote:
> On one of our client's node, due to trying to read from closed ioc,
> a segmentation fault occured. Corresponding backtrace:
> 
> 0  object_get_class (obj=obj@entry=0x0)
> 1  qio_channel_readv_full (ioc=0x0, iov=0x7ffe55277180 ...
> 2  qio_channel_read (ioc= ...
> 3  vnc_client_read_buf (vs=vs@entry=0x55625f3c6000, ...
> 4  vnc_client_read_plain (vs=0x55625f3c6000)
> 5  vnc_client_read (vs=0x55625f3c6000)
> 6  vnc_client_io (ioc=, condition=G_IO_IN, ...
> 7  g_main_dispatch (context=0x556251568a50)
> 8  g_main_context_dispatch (context=context@entry=0x556251568a50)
> 9  glib_pollfds_poll ()
> 10 os_host_main_loop_wait (timeout=)
> 11 main_loop_wait (nonblocking=nonblocking@entry=0)
> 12 main_loop () at vl.c:1909
> 13 main (argc=, argv=, ...
> 
> Having analyzed the coredump, I understood that the reason is that
> ioc_tag is reset on vnc_disconnect_start and ioc is cleaned
> in vnc_disconnect_finish. Between these two events due to some
> reasons the ioc_tag was set again and after vnc_disconnect_finish
> the handler is running with freed ioc,
> which led to the segmentation fault.
> 
> The patch checks vs->disconnecting in places where we call
> qio_channel_add_watch and resets handler if disconnecting == TRUE
> to prevent such an occurrence.
> 
> Signed-off-by: Klim Kireev 
> ---
> Changelog:
> v2: Attach the backtrace
> 
> v3: Change checks
> 
>  ui/vnc-jobs.c |  6 --
>  ui/vnc.c  | 15 ++-
>  2 files changed, 18 insertions(+), 3 deletions(-)

Reviewed-by: Daniel P. Berrangé 


> 
> diff --git a/ui/vnc-jobs.c b/ui/vnc-jobs.c
> index e326679dd0..868dddef4b 100644
> --- a/ui/vnc-jobs.c
> +++ b/ui/vnc-jobs.c
> @@ -148,8 +148,10 @@ void vnc_jobs_consume_buffer(VncState *vs)
>  if (vs->ioc_tag) {
>  g_source_remove(vs->ioc_tag);
>  }
> -vs->ioc_tag = qio_channel_add_watch(
> -vs->ioc, G_IO_IN | G_IO_OUT, vnc_client_io, vs, NULL);
> +if (vs->disconnecting == FALSE) {
> +vs->ioc_tag = qio_channel_add_watch(
> +vs->ioc, G_IO_IN | G_IO_OUT, vnc_client_io, vs, NULL);
> +}
>  }
>  buffer_move(&vs->output, &vs->jobs_buffer);
>  
> diff --git a/ui/vnc.c b/ui/vnc.c
> index 93731accb6..67ccc8160f 100644
> --- a/ui/vnc.c
> +++ b/ui/vnc.c
> @@ -1536,12 +1536,19 @@ gboolean vnc_client_io(QIOChannel *ioc G_GNUC_UNUSED,
>  VncState *vs = opaque;
>  if (condition & G_IO_IN) {
>  if (vnc_client_read(vs) < 0) {
> -return TRUE;
> +goto end;
>  }
>  }
>  if (condition & G_IO_OUT) {
>  vnc_client_write(vs);
>  }
> +end:
> +if (vs->disconnecting) {
> +if (vs->ioc_tag != 0) {
> +g_source_remove(vs->ioc_tag);
> +}
> +vs->ioc_tag = 0;
> +}
>  return TRUE;
>  }
>  
> @@ -1630,6 +1637,12 @@ void vnc_flush(VncState *vs)
>  if (vs->ioc != NULL && vs->output.offset) {
>  vnc_client_write_locked(vs);
>  }
> +if (vs->disconnecting) {
> +if (vs->ioc_tag != 0) {
> +g_source_remove(vs->ioc_tag);
> +}
> +vs->ioc_tag = 0;
> +}
>  vnc_unlock_output(vs);
>  }
>  
> -- 
> 2.14.3
> 

Regards,
Daniel
-- 
|: https://berrange.com  -o-https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org -o-https://fstop138.berrange.com :|
|: https://entangle-photo.org-o-https://www.instagram.com/dberrange :|



Re: [Qemu-devel] [PATCH] vnc: fix segfault in closed connection handling

2018-01-31 Thread klim

On 01/31/2018 04:16 PM, Marc-André Lureau wrote:

Hi

On Wed, Jan 31, 2018 at 2:06 PM, Klim Kireev  wrote:

On one of our client's node, due to trying to read from closed ioc,
a segmentation fault occured. Corresponding backtrace:


Oops, you probably forgot an extra space before the # interpreted as comment.

Thanks, fixed


Do you have a reproducer?

Unfortunately, no

Having analyzed the coredump, I understood that the reason is that
ioc_tag is reset on vnc_disconnect_start and ioc is cleaned
in vnc_disconnect_finish. Between these two events due to some
reasons the ioc_tag was set again and after vnc_disconnect_finish
the handler is running with freed ioc,
which led to the segmentation fault.

I suggest to check ioc_tag in vnc_disconnect_finish to prevent such
an occurrence.



Signed-off-by: Klim Kireev 
---
  ui/vnc.c | 4 
  1 file changed, 4 insertions(+)

diff --git a/ui/vnc.c b/ui/vnc.c
index 33b087221f..b8bf0180cb 100644
--- a/ui/vnc.c
+++ b/ui/vnc.c
@@ -1270,6 +1270,10 @@ void vnc_disconnect_finish(VncState *vs)
  }
  g_free(vs->lossy_rect);

+if (vs->ioc_tag) {
+g_source_remove(vs->ioc_tag);
+vs->ioc_tag = 0;
+}
  object_unref(OBJECT(vs->ioc));
  vs->ioc = NULL;
  object_unref(OBJECT(vs->sioc));
--
2.14.3










Re: [Qemu-devel] [PATCH] vnc: fix segfault in closed connection handling

2018-01-31 Thread Marc-André Lureau
Hi

On Wed, Jan 31, 2018 at 2:06 PM, Klim Kireev  wrote:
> On one of our client's node, due to trying to read from closed ioc,
> a segmentation fault occured. Corresponding backtrace:
>

Oops, you probably forgot an extra space before the # interpreted as comment.

Do you have a reproducer?

> Having analyzed the coredump, I understood that the reason is that
> ioc_tag is reset on vnc_disconnect_start and ioc is cleaned
> in vnc_disconnect_finish. Between these two events due to some
> reasons the ioc_tag was set again and after vnc_disconnect_finish
> the handler is running with freed ioc,
> which led to the segmentation fault.
>
> I suggest to check ioc_tag in vnc_disconnect_finish to prevent such
> an occurrence.


>
> Signed-off-by: Klim Kireev 
> ---
>  ui/vnc.c | 4 
>  1 file changed, 4 insertions(+)
>
> diff --git a/ui/vnc.c b/ui/vnc.c
> index 33b087221f..b8bf0180cb 100644
> --- a/ui/vnc.c
> +++ b/ui/vnc.c
> @@ -1270,6 +1270,10 @@ void vnc_disconnect_finish(VncState *vs)
>  }
>  g_free(vs->lossy_rect);
>
> +if (vs->ioc_tag) {
> +g_source_remove(vs->ioc_tag);
> +vs->ioc_tag = 0;
> +}
>  object_unref(OBJECT(vs->ioc));
>  vs->ioc = NULL;
>  object_unref(OBJECT(vs->sioc));
> --
> 2.14.3
>
>



-- 
Marc-André Lureau