[CRIU] [PATCH 0/2] virtual terminal

[Ruslan Kuprieiev]

Yes, this patch is a first iteration and adds basic
support to c/r tasks that are using vts in a basic way.
It works fine for all our tests and a bunch of tested
programs. No VT specific data is needed for now.

Though, I am going to add more features along with
c/r-ing vt buffers(which seems quite possible
considering /dev/vcsN and /dev/vcsaN).

17.12.2014 10:05, Cyrill Gorcunov пишет:
> On Wed, Dec 17, 2014 at 10:35:48AM +0300, Cyrill Gorcunov wrote:
>> On Wed, Dec 17, 2014 at 01:21:38AM +0200, Ruslan Kuprieiev wrote:
>>> Ruslan Kuprieiev (2):
>>>    tty: add vt support
>>>    zdtm:tty:vt: add simple test
>>>
>>>   include/tty.h                  | 10 +++++++
>>>   protobuf/tty.proto             |  1 +
>>>   test/zdtm.sh                   |  3 ++
>>>   test/zdtm/.gitignore           |  1 +
>>>   test/zdtm/live/static/Makefile |  1 +
>>>   test/zdtm/live/static/vt.c     | 60 ++++++++++++++++++++++++++++++++++++++++
>>>   tty.c                          | 63 ++++++++++++++++++++++++++++++++++--------
>>>   7 files changed, 128 insertions(+), 11 deletions(-)
>>>   create mode 100644 test/zdtm/live/static/vt.c
>>
>> I didn't look into how VTs are implemented inside the kernel,
>> but if this pass the tests and works (I suppose)
>>
>> Reviewed-by: Cyrill Gorcunov <gorcunov at openvz.org>
>>
>> 'cause userspace part looks good to me.
>
> After lookin into kernel's part it looks like rather early rfc than
> a real patches. Why don't we save and restore any VT specific data
> which might be fetched with vt ioctls, are they unneeded?
>