Re: Fwd: [PATCH] Initialize mutexes for static builds for win32

On Mon, Jun 12, 2017 at 12:49:44PM +0200, Alexander Täschner wrote:
> Here is the revised patch.

This sounds safe enough to me, and presumably Uli is ok with the updated
version.  I fixed the indentation error and landed it:

To ssh://git.freedesktop.org/git/cairo
   339fe9a..5f90f0a  master -> master

I'd encourage win32 users to test this usage case.

> Kind regards,
> Alexander Täschner
> ---
>  src/win32/cairo-win32-device.c | 2 ++
>  1 file changed, 2 insertions(+)
> diff --git a/src/win32/cairo-win32-device.c b/src/win32/cairo-win32-device.c
> index 741e49e..3e959f6 100644
> --- a/src/win32/cairo-win32-device.c
> +++ b/src/win32/cairo-win32-device.c
> @@ -131,6 +131,8 @@ _cairo_win32_device_get (void)
>  {
>      cairo_win32_device_t *device;
> 
> + CAIRO_MUTEX_INITIALIZE ();
> +
>      if (__cairo_win32_device)
>   return cairo_device_reference (__cairo_win32_device);
> 
> -- 
> 2.10.2.windows.1
> 
> 
> 2017-06-10 14:38 GMT+02:00 Uli Schlachter <[email protected]>:
> 
> > On 07.06.2017 16:01, Alexander Täschner wrote:
> > > For static builds for win32 the mutexes are not initialized (for dynamic
> > > builds its done when the DLL is loaded). Therefore the patch add
> > > initialization to the create surface calls.
> > >
> > > ---
> > >  src/win32/cairo-win32-display-surface.c | 8 ++++++--
> > >  1 file changed, 6 insertions(+), 2 deletions(-)
> > > diff --git a/src/win32/cairo-win32-display-surface.c
> > > b/src/win32/cairo-win32-display-surface.c
> > > index 025b2c9..56f0e65 100644
> > > --- a/src/win32/cairo-win32-display-surface.c
> > > +++ b/src/win32/cairo-win32-display-surface.c
> > > @@ -953,7 +953,9 @@ cairo_win32_surface_create_with_format (HDC hdc,
> > > cairo_format_t format)
> > >      case CAIRO_FORMAT_RGB24:
> > >   break;
> > >      }
> > > -
> > > +#if CAIRO_WIN32_STATIC_BUILD
> > > +    CAIRO_MUTEX_INITIALIZE ();
> > > +#endif
> > >      surface = malloc (sizeof (*surface));
> > >      if (surface == NULL)
> > >   return _cairo_surface_create_in_error (_cairo_error
> > > (CAIRO_STATUS_NO_MEMORY));
> > > @@ -1031,7 +1033,9 @@ cairo_win32_surface_create_with_dib
> > (cairo_format_t
> > > format,
> > >  {
> > >      if (! CAIRO_FORMAT_VALID (format))
> > >   return _cairo_surface_create_in_error (_cairo_error
> > > (CAIRO_STATUS_INVALID_FORMAT));
> > > -
> > > +#if CAIRO_WIN32_STATIC_BUILD
> > > +    CAIRO_MUTEX_INITIALIZE ();
> > > +#endif
> > >      return _cairo_win32_display_surface_create_for_dc (NULL, format,
> > > width, height);
> > >  }
> >
> > I would suggest doing the above unconditionally (without the #if).
> >
> > Also, how about moving this to the beginning of
> > _cairo_win32_device_get()? That sounds central enough that every code
> > path in the win32 backend should go through it.
> >
> > Uli
> > --
> > 99 little bugs in the code
> > 99 little bugs in the code
> > Take one down, patch it around
> > 117 little bugs in the code
> >   -- @irqed
> >

> -- 
> cairo mailing list
> [email protected]
> https://lists.cairographics.org/mailman/listinfo/cairo

-- 
cairo mailing list
[email protected]
https://lists.cairographics.org/mailman/listinfo/cairo