On 27.11.2018 15:00, Marc-André Lureau wrote:
> Hi
> On Tue, Nov 27, 2018 at 3:56 PM Ilya Maximets <i.maxim...@samsung.com> wrote:
>>
>> On 27.11.2018 14:49, Marc-André Lureau wrote:
>>> Hi
>>> On Tue, Nov 27, 2018 at 3:11 PM Ilya Maximets <i.maxim...@samsung.com>
>>> wrote:
>>>>
>>>> If seals are not supported, memfd_create() will fail.
>>>> Furthermore, there is no way to disable it in this case because
>>>> '.seal' property is not registered.
>>>>
>>>> This issue leads to vhost-user-test failures on RHEL 7.2:
>>>>
>>>> qemu-system-x86_64: -object memory-backend-memfd,id=mem,size=2M,: \
>>>> failed to create memfd: Invalid argument
>>>>
>>>> Signed-off-by: Ilya Maximets <i.maxim...@samsung.com>
>>>
>>>
>>> This will change the default behaviour of memfd backend, and may thus
>>> me considered a break.
>>
>> This will change the default behaviour only on systems without sealing
>> support. But current implementation is broken there anyway and does not
>> work.
>>
>>>
>>> Instead, memfd vhost-user-test should skipped (or tuned with
>>> sealed=false if unsupported)
>>
>> vhost-user-test is just an example here. In fact memfd could not be
>> used at all on system without sealing support. And there is no way
>> to disable seals.
>
> which system supports memfd without sealing?
RHEL 7.2. kernel version 3.10.0-327.el7.x86_64
>
>>
>>>
>>>> ---
>>>> backends/hostmem-memfd.c | 4 ++--
>>>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/backends/hostmem-memfd.c b/backends/hostmem-memfd.c
>>>> index b6836b28e5..ee39bdbde6 100644
>>>> --- a/backends/hostmem-memfd.c
>>>> +++ b/backends/hostmem-memfd.c
>>>> @@ -129,8 +129,8 @@ memfd_backend_instance_init(Object *obj)
>>>> {
>>>> HostMemoryBackendMemfd *m = MEMORY_BACKEND_MEMFD(obj);
>>>>
>>>> - /* default to sealed file */
>>>> - m->seal = true;
>>>> + /* default to sealed file if supported */
>>>> + m->seal = qemu_memfd_check(MFD_ALLOW_SEALING);
>>>> }
>>>>
>>>> static void
>>>> --
>>>> 2.17.1
>>>>
>>>
>>>
>>
>
>
