diff mbox series

[v2] net/colo-compare.c: Fix a crash in COLO Primary.

Message ID 20190420191425.7d1dab82@luklap
State New
Headers show
Series [v2] net/colo-compare.c: Fix a crash in COLO Primary. | expand

Commit Message

Lukas Straub April 20, 2019, 5:14 p.m. UTC
From: Lukas Straub <lukasstraub2@web.de>
Because event_unhandled_count may be accessed concurrently, it needs
to be protected by taking the lock. However the assert is outside the
lock, probably causing it to read garbage and aborting Qemu erroneously.

The Bug only happens when running Qemu in COLO mode.

This Patch fixes the following bug: https://bugs.launchpad.net/qemu/+bug/1824622

Signed-off-by: Lukas Straub <lukasstraub2@web.de>
---
 net/colo-compare.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

--
2.20.1

Comments

Philippe Mathieu-Daudé April 21, 2019, 10:13 a.m. UTC | #1
On 4/20/19 7:14 PM, Lukas Straub wrote:
> From: Lukas Straub <lukasstraub2@web.de>
> Because event_unhandled_count may be accessed concurrently, it needs
> to be protected by taking the lock. However the assert is outside the
> lock, probably causing it to read garbage and aborting Qemu erroneously.
> 
> The Bug only happens when running Qemu in COLO mode.
> 
> This Patch fixes the following bug: https://bugs.launchpad.net/qemu/+bug/1824622
> 
> Signed-off-by: Lukas Straub <lukasstraub2@web.de>
> ---
>  net/colo-compare.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/net/colo-compare.c b/net/colo-compare.c
> index bf10526f05..fcb491121b 100644
> --- a/net/colo-compare.c
> +++ b/net/colo-compare.c
> @@ -813,9 +813,8 @@ static void colo_compare_handle_event(void *opaque)
>          break;
>      }
> 
> -    assert(event_unhandled_count > 0);
> -
>      qemu_mutex_lock(&event_mtx);
> +    assert(event_unhandled_count > 0);
>      event_unhandled_count--;
>      qemu_cond_broadcast(&event_complete_cond);
>      qemu_mutex_unlock(&event_mtx);
> --
> 2.20.1
> 
> 

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Zhang, Chen April 23, 2019, 6:58 a.m. UTC | #2
> -----Original Message-----
> From: Lukas Straub [mailto:lukasstraub2@web.de]
> Sent: Sunday, April 21, 2019 1:14 AM
> To: qemu-devel@nongnu.org
> Cc: Zhang, Chen <chen.zhang@intel.com>
> Subject: [PATCH v2] net/colo-compare.c: Fix a crash in COLO Primary.
> 
> From: Lukas Straub <lukasstraub2@web.de> Because event_unhandled_count
> may be accessed concurrently, it needs to be protected by taking the lock.
> However the assert is outside the lock, probably causing it to read garbage and
> aborting Qemu erroneously.
> 
> The Bug only happens when running Qemu in COLO mode.
> 
> This Patch fixes the following bug:
> https://bugs.launchpad.net/qemu/+bug/1824622
> 
> Signed-off-by: Lukas Straub <lukasstraub2@web.de>

Looks good for me.

Reviewed-by: Zhang Chen <chen.zhang@intel.com>

Thanks
Zhang Chen

> ---
>  net/colo-compare.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/net/colo-compare.c b/net/colo-compare.c index
> bf10526f05..fcb491121b 100644
> --- a/net/colo-compare.c
> +++ b/net/colo-compare.c
> @@ -813,9 +813,8 @@ static void colo_compare_handle_event(void *opaque)
>          break;
>      }
> 
> -    assert(event_unhandled_count > 0);
> -
>      qemu_mutex_lock(&event_mtx);
> +    assert(event_unhandled_count > 0);
>      event_unhandled_count--;
>      qemu_cond_broadcast(&event_complete_cond);
>      qemu_mutex_unlock(&event_mtx);
> --
> 2.20.1
Lukas Straub May 6, 2019, 10:32 a.m. UTC | #3
On Sat, 20 Apr 2019 19:14:25 +0200
Lukas Straub <lukasstraub2@web.de> wrote:

> From: Lukas Straub <lukasstraub2@web.de>
> Because event_unhandled_count may be accessed concurrently, it needs
> to be protected by taking the lock. However the assert is outside the
> lock, probably causing it to read garbage and aborting Qemu
> erroneously.
>
> The Bug only happens when running Qemu in COLO mode.
>
> This Patch fixes the following bug:
> https://bugs.launchpad.net/qemu/+bug/1824622
>
> Signed-off-by: Lukas Straub <lukasstraub2@web.de>
> ---
>  net/colo-compare.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/net/colo-compare.c b/net/colo-compare.c
> index bf10526f05..fcb491121b 100644
> --- a/net/colo-compare.c
> +++ b/net/colo-compare.c
> @@ -813,9 +813,8 @@ static void colo_compare_handle_event(void
> *opaque) break;
>      }
>
> -    assert(event_unhandled_count > 0);
> -
>      qemu_mutex_lock(&event_mtx);
> +    assert(event_unhandled_count > 0);
>      event_unhandled_count--;
>      qemu_cond_broadcast(&event_complete_cond);
>      qemu_mutex_unlock(&event_mtx);

Ping.

Regards,
Lukas Straub
Philippe Mathieu-Daudé May 6, 2019, 11:13 a.m. UTC | #4
Cc'ing Paolo & Stefan

On 4/20/19 7:14 PM, Lukas Straub wrote:
> From: Lukas Straub <lukasstraub2@web.de>
> Because event_unhandled_count may be accessed concurrently, it needs
> to be protected by taking the lock. However the assert is outside the
> lock, probably causing it to read garbage and aborting Qemu erroneously.
> 
> The Bug only happens when running Qemu in COLO mode.
> 
> This Patch fixes the following bug: https://bugs.launchpad.net/qemu/+bug/1824622
> 
> Signed-off-by: Lukas Straub <lukasstraub2@web.de>
> ---
>  net/colo-compare.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/net/colo-compare.c b/net/colo-compare.c
> index bf10526f05..fcb491121b 100644
> --- a/net/colo-compare.c
> +++ b/net/colo-compare.c
> @@ -813,9 +813,8 @@ static void colo_compare_handle_event(void *opaque)
>          break;
>      }
> 
> -    assert(event_unhandled_count > 0);
> -
>      qemu_mutex_lock(&event_mtx);
> +    assert(event_unhandled_count > 0);
>      event_unhandled_count--;
>      qemu_cond_broadcast(&event_complete_cond);
>      qemu_mutex_unlock(&event_mtx);
> --
> 2.20.1
> 
>
Jason Wang May 7, 2019, 12:07 a.m. UTC | #5
On 2019/5/6 下午6:32, Lukas Straub wrote:
> On Sat, 20 Apr 2019 19:14:25 +0200
> Lukas Straub <lukasstraub2@web.de> wrote:
>
>> From: Lukas Straub <lukasstraub2@web.de>
>> Because event_unhandled_count may be accessed concurrently, it needs
>> to be protected by taking the lock. However the assert is outside the
>> lock, probably causing it to read garbage and aborting Qemu
>> erroneously.
>>
>> The Bug only happens when running Qemu in COLO mode.
>>
>> This Patch fixes the following bug:
>> https://bugs.launchpad.net/qemu/+bug/1824622
>>
>> Signed-off-by: Lukas Straub <lukasstraub2@web.de>
>> ---
>>   net/colo-compare.c | 3 +--
>>   1 file changed, 1 insertion(+), 2 deletions(-)
>>
>> diff --git a/net/colo-compare.c b/net/colo-compare.c
>> index bf10526f05..fcb491121b 100644
>> --- a/net/colo-compare.c
>> +++ b/net/colo-compare.c
>> @@ -813,9 +813,8 @@ static void colo_compare_handle_event(void
>> *opaque) break;
>>       }
>>
>> -    assert(event_unhandled_count > 0);
>> -
>>       qemu_mutex_lock(&event_mtx);
>> +    assert(event_unhandled_count > 0);
>>       event_unhandled_count--;
>>       qemu_cond_broadcast(&event_complete_cond);
>>       qemu_mutex_unlock(&event_mtx);
> Ping.
>
> Regards,
> Lukas Straub


Applied.

Thanks

>
diff mbox series

Patch

diff --git a/net/colo-compare.c b/net/colo-compare.c
index bf10526f05..fcb491121b 100644
--- a/net/colo-compare.c
+++ b/net/colo-compare.c
@@ -813,9 +813,8 @@  static void colo_compare_handle_event(void *opaque)
         break;
     }

-    assert(event_unhandled_count > 0);
-
     qemu_mutex_lock(&event_mtx);
+    assert(event_unhandled_count > 0);
     event_unhandled_count--;
     qemu_cond_broadcast(&event_complete_cond);
     qemu_mutex_unlock(&event_mtx);