Patchwork [trivial] vl: always define no_frame

login
register
mail settings
Submitter Michael Tokarev
Date June 15, 2013, 10:42 a.m.
Message ID <1371292923-28105-1-git-send-email-mjt@msgid.tls.msk.ru>
Download mbox | patch
Permalink /patch/251621/
State New
Headers show

Comments

Michael Tokarev - June 15, 2013, 10:42 a.m.
Commit 047d4e151dd46 "Unbreak -no-quit for GTK, validate SDL options" broke
build of qemu without sdl, by referencing `no_frame' variable which is defined
inside #if SDL block.  Fix that by defining that variable unconditionally.

This is a better fix for the build issue introduced by that patch than
a revert.  This change keeps the new functinality introduced by that patch
and just fixes the compilation.  It still is not a complete fix around the
original issue (not working -no-frame et al with -display gtk), because it
makes only the legacy interface working, not the new suboption interface,
so a few more changes are needed.

Cc: Peter Wu <lekensteyn@gmail.com>
Cc: qemu-trivial@nongnu.org
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
---
 vl.c |    2 --
 1 file changed, 2 deletions(-)
Peter Wu - June 15, 2013, 10:51 a.m.
On Saturday 15 June 2013 14:42:03 Michael Tokarev wrote:
> Commit 047d4e151dd46 "Unbreak -no-quit for GTK, validate SDL options" broke
> build of qemu without sdl, by referencing `no_frame' variable which is
> defined inside #if SDL block.  Fix that by defining that variable
> unconditionally.
> 
> This is a better fix for the build issue introduced by that patch than
> a revert.  This change keeps the new functinality introduced by that patch
> and just fixes the compilation.  It still is not a complete fix around the
> original issue (not working -no-frame et al with -display gtk), because it
> makes only the legacy interface working, not the new suboption interface,
> so a few more changes are needed.

Reviewed-by: Peter Wu <lekensteyn@gmail.com>

-no-frame was deemed to be a SDL feature that was unnecessary for GTK, hence dropped. 
At run-time, it is checked whether -no-frame makes sense or not (that is, if `-display sdl` 
was specified or not). Original patch and discussion can be found at 
http://lists.nongnu.org/archive/html/qemu-devel/2013-06/msg01297.html.

As for the discussion about legacy options[1], are there plans to remove `-sdl` (replaced by 
`-display sdl`) and move the `-*-grab` and `-no-frame` to `-display sdl` options? If that is 
the idea, what about printing a message that the mentioned options are deprecated and 
removed in a future version?

Regards,
Peter

 [1]: http://lists.nongnu.org/archive/html/qemu-devel/2013-06/msg02437.html

> Cc: Peter Wu <lekensteyn@gmail.com>
> Cc: qemu-trivial@nongnu.org
> Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
> ---
>  vl.c |    2 --
>  1 file changed, 2 deletions(-)
> 
> diff --git a/vl.c b/vl.c
> index 9f8fd6e..f94ec9c 100644
> --- a/vl.c
> +++ b/vl.c
> @@ -199,9 +199,7 @@ static int rtc_date_offset = -1; /* -1 means no change
> */ QEMUClock *rtc_clock;
>  int vga_interface_type = VGA_NONE;
>  static int full_screen = 0;
> -#ifdef CONFIG_SDL
>  static int no_frame = 0;
> -#endif
>  int no_quit = 0;
>  CharDriverState *serial_hds[MAX_SERIAL_PORTS];
>  CharDriverState *parallel_hds[MAX_PARALLEL_PORTS];
Michael Tokarev - June 15, 2013, 11:05 a.m.
15.06.2013 14:51, Peter Wu wrote:
> On Saturday 15 June 2013 14:42:03 Michael Tokarev wrote:
> 
>> Commit 047d4e151dd46 "Unbreak -no-quit for GTK, validate SDL options" broke
>> build of qemu without sdl, by referencing `no_frame' variable which is
>> defined inside #if SDL block. Fix that by defining that variable
>> unconditionally.
> 
> Reviewed-by: Peter Wu <lekensteyn@gmail.com>

Thanks!

> As for the discussion about legacy options[1], are there plans to remove `-sdl` (replaced by `-display sdl`) and move the `-*-grab` and `-no-frame` to `-display sdl` options? If that is the idea, what about printing a message that the mentioned options are deprecated and removed in a future version?

This is already done for sdl, see select_display() function in vl.c --
it handles -display sdl,frame={on|off},...

It is unlikely the old options will be removed any time soon, and
these will not likely print any warnings either -- this hasn't done
for lots of other old options (-stdvga for example).

Thanks,

/mjt
Andreas Färber - June 16, 2013, 5:26 p.m.
Am 15.06.2013 12:42, schrieb Michael Tokarev:
> Commit 047d4e151dd46 "Unbreak -no-quit for GTK, validate SDL options" broke
> build of qemu without sdl, by referencing `no_frame' variable which is defined
> inside #if SDL block.  Fix that by defining that variable unconditionally.
> 
> This is a better fix for the build issue introduced by that patch than
> a revert.  This change keeps the new functinality introduced by that patch
> and just fixes the compilation.  It still is not a complete fix around the
> original issue (not working -no-frame et al with -display gtk), because it
> makes only the legacy interface working, not the new suboption interface,
> so a few more changes are needed.
> 
> Cc: Peter Wu <lekensteyn@gmail.com>
> Cc: qemu-trivial@nongnu.org
> Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>

Tested-by: Andreas Färber <afaerber@suse.de>

Unbreaks my headless ppc build.

Thanks,
Andreas
Anthony Liguori - June 17, 2013, 9:06 p.m.
Applied.  Thanks.

Regards,

Anthony Liguori

Patch

diff --git a/vl.c b/vl.c
index 9f8fd6e..f94ec9c 100644
--- a/vl.c
+++ b/vl.c
@@ -199,9 +199,7 @@  static int rtc_date_offset = -1; /* -1 means no change */
 QEMUClock *rtc_clock;
 int vga_interface_type = VGA_NONE;
 static int full_screen = 0;
-#ifdef CONFIG_SDL
 static int no_frame = 0;
-#endif
 int no_quit = 0;
 CharDriverState *serial_hds[MAX_SERIAL_PORTS];
 CharDriverState *parallel_hds[MAX_PARALLEL_PORTS];