[v1] tests: fix tpm-crb tpm-tis tests race

[Qemu-devel] [PATCH] tests: fix tpm-crb tpm-tis tests race

Posted by Marc-André Lureau 7 years, 7 months ago

No need to close the TPM data socket on the emulator end, qemu will
close it after a SHUTDOWN. This avoids a race between close() and
read() in the TPM data thread.

Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 tests/tpm-emu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/tpm-emu.c b/tests/tpm-emu.c
index 4dada76834..8c2bd53cad 100644
--- a/tests/tpm-emu.c
+++ b/tests/tpm-emu.c
@@ -125,7 +125,7 @@ void *tpm_emu_ctrl_thread(void *data)
         case CMD_SHUTDOWN: {
             ptm_res res = 0;
             qio_channel_write(ioc, (char *)&res, sizeof(res), &error_abort);
-            qio_channel_close(s->tpm_ioc, &error_abort);
+            /* the tpm data thread is expected to finish now */
             g_thread_join(s->emu_tpm_thread);
             break;
         }
-- 
2.16.2.521.g9aa15f885a

Re: [Qemu-devel] [PATCH] tests: fix tpm-crb tpm-tis tests race

Posted by Daniel P. Berrangé 7 years, 7 months ago

On Fri, Mar 16, 2018 at 03:22:23PM +0100, Marc-André Lureau wrote:
> No need to close the TPM data socket on the emulator end, qemu will
> close it after a SHUTDOWN. This avoids a race between close() and
> read() in the TPM data thread.
> 
> Reported-by: Peter Maydell <peter.maydell@linaro.org>
> Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
> ---
>  tests/tpm-emu.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tests/tpm-emu.c b/tests/tpm-emu.c
> index 4dada76834..8c2bd53cad 100644
> --- a/tests/tpm-emu.c
> +++ b/tests/tpm-emu.c
> @@ -125,7 +125,7 @@ void *tpm_emu_ctrl_thread(void *data)
>          case CMD_SHUTDOWN: {
>              ptm_res res = 0;
>              qio_channel_write(ioc, (char *)&res, sizeof(res), &error_abort);
> -            qio_channel_close(s->tpm_ioc, &error_abort);
> +            /* the tpm data thread is expected to finish now */
>              g_thread_join(s->emu_tpm_thread);

Won't this leave an orphaed FD open in the test suite ? Is it perhaps
sufficient to just swap the order of the g_thread_join and qio_channel_close
calls, so that we join the thread before we close the channel that the
thread is using ?


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] tests: fix tpm-crb tpm-tis tests race

Posted by Marc-André Lureau 7 years, 7 months ago

Hi

On Fri, Mar 16, 2018 at 3:25 PM, Daniel P. Berrangé <berrange@redhat.com> wrote:
> On Fri, Mar 16, 2018 at 03:22:23PM +0100, Marc-André Lureau wrote:
>> No need to close the TPM data socket on the emulator end, qemu will
>> close it after a SHUTDOWN. This avoids a race between close() and
>> read() in the TPM data thread.
>>
>> Reported-by: Peter Maydell <peter.maydell@linaro.org>
>> Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
>> ---
>>  tests/tpm-emu.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/tests/tpm-emu.c b/tests/tpm-emu.c
>> index 4dada76834..8c2bd53cad 100644
>> --- a/tests/tpm-emu.c
>> +++ b/tests/tpm-emu.c
>> @@ -125,7 +125,7 @@ void *tpm_emu_ctrl_thread(void *data)
>>          case CMD_SHUTDOWN: {
>>              ptm_res res = 0;
>>              qio_channel_write(ioc, (char *)&res, sizeof(res), &error_abort);
>> -            qio_channel_close(s->tpm_ioc, &error_abort);
>> +            /* the tpm data thread is expected to finish now */
>>              g_thread_join(s->emu_tpm_thread);
>
> Won't this leave an orphaed FD open in the test suite ? Is it perhaps
> sufficient to just swap the order of the g_thread_join and qio_channel_close
> calls, so that we join the thread before we close the channel that the
> thread is using ?

Isn't the socket close on the last unref? (at end of tpm_emu_tpm_thread())

>
> 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 :|
>



-- 
Marc-André Lureau

Re: [Qemu-devel] [PATCH] tests: fix tpm-crb tpm-tis tests race

Posted by Daniel P. Berrangé 7 years, 7 months ago

On Fri, Mar 16, 2018 at 03:29:04PM +0100, Marc-André Lureau wrote:
> Hi
> 
> On Fri, Mar 16, 2018 at 3:25 PM, Daniel P. Berrangé <berrange@redhat.com> wrote:
> > On Fri, Mar 16, 2018 at 03:22:23PM +0100, Marc-André Lureau wrote:
> >> No need to close the TPM data socket on the emulator end, qemu will
> >> close it after a SHUTDOWN. This avoids a race between close() and
> >> read() in the TPM data thread.
> >>
> >> Reported-by: Peter Maydell <peter.maydell@linaro.org>
> >> Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
> >> ---
> >>  tests/tpm-emu.c | 2 +-
> >>  1 file changed, 1 insertion(+), 1 deletion(-)
> >>
> >> diff --git a/tests/tpm-emu.c b/tests/tpm-emu.c
> >> index 4dada76834..8c2bd53cad 100644
> >> --- a/tests/tpm-emu.c
> >> +++ b/tests/tpm-emu.c
> >> @@ -125,7 +125,7 @@ void *tpm_emu_ctrl_thread(void *data)
> >>          case CMD_SHUTDOWN: {
> >>              ptm_res res = 0;
> >>              qio_channel_write(ioc, (char *)&res, sizeof(res), &error_abort);
> >> -            qio_channel_close(s->tpm_ioc, &error_abort);
> >> +            /* the tpm data thread is expected to finish now */
> >>              g_thread_join(s->emu_tpm_thread);
> >
> > Won't this leave an orphaed FD open in the test suite ? Is it perhaps
> > sufficient to just swap the order of the g_thread_join and qio_channel_close
> > calls, so that we join the thread before we close the channel that the
> > thread is using ?
> 
> Isn't the socket close on the last unref? (at end of tpm_emu_tpm_thread())

Oh yes, that is true. I should have looked at more than just diff context.
In that case

  Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>

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] tests: fix tpm-crb tpm-tis tests race

Posted by Stefan Berger 7 years, 7 months ago

On 03/16/2018 10:30 AM, Daniel P. Berrangé wrote:
> On Fri, Mar 16, 2018 at 03:29:04PM +0100, Marc-André Lureau wrote:
>> Hi
>>
>> On Fri, Mar 16, 2018 at 3:25 PM, Daniel P. Berrangé <berrange@redhat.com> wrote:
>>> On Fri, Mar 16, 2018 at 03:22:23PM +0100, Marc-André Lureau wrote:
>>>> No need to close the TPM data socket on the emulator end, qemu will
>>>> close it after a SHUTDOWN. This avoids a race between close() and
>>>> read() in the TPM data thread.
>>>>
>>>> Reported-by: Peter Maydell <peter.maydell@linaro.org>
>>>> Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
>>>> ---
>>>>   tests/tpm-emu.c | 2 +-
>>>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>> diff --git a/tests/tpm-emu.c b/tests/tpm-emu.c
>>>> index 4dada76834..8c2bd53cad 100644
>>>> --- a/tests/tpm-emu.c
>>>> +++ b/tests/tpm-emu.c
>>>> @@ -125,7 +125,7 @@ void *tpm_emu_ctrl_thread(void *data)
>>>>           case CMD_SHUTDOWN: {
>>>>               ptm_res res = 0;
>>>>               qio_channel_write(ioc, (char *)&res, sizeof(res), &error_abort);
>>>> -            qio_channel_close(s->tpm_ioc, &error_abort);
>>>> +            /* the tpm data thread is expected to finish now */
>>>>               g_thread_join(s->emu_tpm_thread);
>>> Won't this leave an orphaed FD open in the test suite ? Is it perhaps
>>> sufficient to just swap the order of the g_thread_join and qio_channel_close
>>> calls, so that we join the thread before we close the channel that the
>>> thread is using ?
>> Isn't the socket close on the last unref? (at end of tpm_emu_tpm_thread())
> Oh yes, that is true. I should have looked at more than just diff context.
> In that case
>
>    Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>

Ha. It's already done :-)

Reviewed-by: Stefan Berger <stefanb@linux.vnet.ibm.com>


>
> Regards,
> Daniel

Re: [Qemu-devel] [PATCH] tests: fix tpm-crb tpm-tis tests race

Posted by Alex Bennée 7 years, 5 months ago

Marc-André Lureau <marcandre.lureau@redhat.com> writes:

> No need to close the TPM data socket on the emulator end, qemu will
> close it after a SHUTDOWN. This avoids a race between close() and
> read() in the TPM data thread.
>
> Reported-by: Peter Maydell <peter.maydell@linaro.org>
> Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Tested-by: Alex Bennée <alex.bennee@linaro.org>

> ---
>  tests/tpm-emu.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tests/tpm-emu.c b/tests/tpm-emu.c
> index 4dada76834..8c2bd53cad 100644
> --- a/tests/tpm-emu.c
> +++ b/tests/tpm-emu.c
> @@ -125,7 +125,7 @@ void *tpm_emu_ctrl_thread(void *data)
>          case CMD_SHUTDOWN: {
>              ptm_res res = 0;
>              qio_channel_write(ioc, (char *)&res, sizeof(res), &error_abort);
> -            qio_channel_close(s->tpm_ioc, &error_abort);
> +            /* the tpm data thread is expected to finish now */
>              g_thread_join(s->emu_tpm_thread);
>              break;
>          }


--
Alex Bennée