qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH v2 16/18] ui: introduce egl_init()


From: Marc-André Lureau
Subject: Re: [PATCH v2 16/18] ui: introduce egl_init()
Date: Mon, 13 Mar 2023 13:59:01 +0400

Hi

On Fri, Mar 10, 2023 at 2:07 PM Daniel P. Berrangé <berrange@redhat.com> wrote:
>
> On Tue, Mar 07, 2023 at 03:56:35PM +0400, marcandre.lureau@redhat.com wrote:
> > From: Marc-André Lureau <marcandre.lureau@redhat.com>
> >
> > Future patches will introduce EGL support on win32 (too late for 8.0
> > though). Having a common place for EGL initialization and error handling
> > will make it simpler.
> >
> > Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
> > ---
> >  include/ui/egl-helpers.h |  2 ++
> >  ui/dbus.c                |  7 +------
> >  ui/egl-headless.c        | 16 ++++++++--------
> >  ui/egl-helpers.c         | 25 +++++++++++++++++++++++++
> >  ui/spice-core.c          |  7 +------
> >  5 files changed, 37 insertions(+), 20 deletions(-)
> >
> > diff --git a/include/ui/egl-helpers.h b/include/ui/egl-helpers.h
> > index c92dd90e33..53d953ddf4 100644
> > --- a/include/ui/egl-helpers.h
> > +++ b/include/ui/egl-helpers.h
> > @@ -65,4 +65,6 @@ int qemu_egl_init_dpy_mesa(EGLNativeDisplayType dpy, 
> > DisplayGLMode mode);
> >  EGLContext qemu_egl_init_ctx(void);
> >  bool qemu_egl_has_dmabuf(void);
> >
> > +bool egl_init(const char *rendernode, DisplayGLMode mode, Error **errp);
> > +
> >  #endif /* EGL_HELPERS_H */
> > diff --git a/ui/dbus.c b/ui/dbus.c
> > index f529928f0b..ebf03bd84d 100644
> > --- a/ui/dbus.c
> > +++ b/ui/dbus.c
> > @@ -451,12 +451,7 @@ early_dbus_init(DisplayOptions *opts)
> >      DisplayGLMode mode = opts->has_gl ? opts->gl : DISPLAYGL_MODE_OFF;
> >
> >      if (mode != DISPLAYGL_MODE_OFF) {
> > -        if (egl_rendernode_init(opts->u.dbus.rendernode, mode) < 0) {
> > -            error_report("dbus: render node init failed");
> > -            exit(1);
> > -        }
> > -
> > -        display_opengl = 1;
> > +        egl_init(opts->u.dbus.rendernode, mode, &error_fatal);
> >      }
> >
> >      type_register(&dbus_vc_type_info);
> > diff --git a/ui/egl-headless.c b/ui/egl-headless.c
> > index ae07e91302..ef70e6a18e 100644
> > --- a/ui/egl-headless.c
> > +++ b/ui/egl-headless.c
> > @@ -1,7 +1,7 @@
> >  #include "qemu/osdep.h"
> >  #include "qemu/error-report.h"
> >  #include "qemu/module.h"
> > -#include "sysemu/sysemu.h"
> > +#include "qapi/error.h"
> >  #include "ui/console.h"
> >  #include "ui/egl-helpers.h"
> >  #include "ui/egl-context.h"
> > @@ -191,21 +191,21 @@ static const DisplayGLCtxOps eglctx_ops = {
> >
> >  static void early_egl_headless_init(DisplayOptions *opts)
> >  {
> > -    display_opengl = 1;
> > +    DisplayGLMode mode = DISPLAYGL_MODE_ON;
> > +
> > +    if (opts->has_gl) {
> > +        mode = opts->gl;
> > +    }
> > +
> > +    egl_init(opts->u.egl_headless.rendernode, mode, &error_fatal);
> >  }
> >
> >  static void egl_headless_init(DisplayState *ds, DisplayOptions *opts)
> >  {
> > -    DisplayGLMode mode = opts->has_gl ? opts->gl : DISPLAYGL_MODE_ON;
> >      QemuConsole *con;
> >      egl_dpy *edpy;
> >      int idx;
> >
> > -    if (egl_rendernode_init(opts->u.egl_headless.rendernode, mode) < 0) {
> > -        error_report("egl: render node init failed");
> > -        exit(1);
> > -    }
> > -
> >      for (idx = 0;; idx++) {
> >          DisplayGLCtx *ctx;
>
> Why isn't the egl_init() call being made from this egl_headless_init
> method, so egl_rendernode_init() is called at the same logical point
> as before this change ?

For consistency with the other egl_init() callers, called during
qemu_create_early_backends(). That way display_opengl is set early too
(as before).
>
>
> > diff --git a/ui/egl-helpers.c b/ui/egl-helpers.c
> > index 10772b6471..36b4fc51d9 100644
> > --- a/ui/egl-helpers.c
> > +++ b/ui/egl-helpers.c
> > @@ -19,6 +19,8 @@
> >  #include "qemu/error-report.h"
> >  #include "ui/console.h"
> >  #include "ui/egl-helpers.h"
> > +#include "sysemu/sysemu.h"
> > +#include "qapi/error.h"
> >
> >  EGLDisplay *qemu_egl_display;
> >  EGLConfig qemu_egl_config;
> > @@ -569,3 +571,26 @@ EGLContext qemu_egl_init_ctx(void)
> >
> >      return ectx;
> >  }
> > +
> > +bool egl_init(const char *rendernode, DisplayGLMode mode, Error **errp)
> > +{
> > +    ERRP_GUARD();
> > +
> > +    if (mode == DISPLAYGL_MODE_OFF) {
> > +        error_setg(errp, "egl: turning off GL doesn't make sense");
> > +        return false;
> > +    }
> > +
> > +#ifdef CONFIG_GBM
> > +    if (egl_rendernode_init(rendernode, mode) < 0) {
> > +        error_setg(errp, "egl: render node init failed");
> > +        return false;
> > +    }
> > +#else
> > +    error_setg(errp, "egl: not available on this platform");
> > +    return false;
> > +#endif
> > +
> > +    display_opengl = 1;
> > +    return true;
> > +}
>
> Surely this is going to result in compile errors when !CONFIG_GBM
> because these two lines are going to be flagged as unreachable
> code, due to the 'return false' in the #else branch.

Interestingly, I don't get a warning (-g -O2).. I'll change the code
nonetheless.

>
> > diff --git a/ui/spice-core.c b/ui/spice-core.c
> > index 76f7c2bc3d..b05c830086 100644
> > --- a/ui/spice-core.c
> > +++ b/ui/spice-core.c
> > @@ -820,12 +820,7 @@ static void qemu_spice_init(void)
> >                           "incompatible with -spice port/tls-port");
> >              exit(1);
> >          }
> > -        if (egl_rendernode_init(qemu_opt_get(opts, "rendernode"),
> > -                                DISPLAYGL_MODE_ON) != 0) {
> > -            error_report("Failed to initialize EGL render node for SPICE 
> > GL");
> > -            exit(1);
> > -        }
> > -        display_opengl = 1;
> > +        egl_init(qemu_opt_get(opts, "rendernode"), DISPLAYGL_MODE_ON, 
> > &error_fatal);
> >          spice_opengl = 1;
> >      }
> >  #endif
> > --
> > 2.39.2
> >
>
> With 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



reply via email to

[Prev in Thread] Current Thread [Next in Thread]