Re: [Mesa-dev] [PATCH 1/2] llvmpipe: move lp_jit_screen_init() call after allocation of screen object

2014-09-30 Thread Brian Paul

On 09/29/2014 07:46 PM, Michel Dänzer wrote:

On 30.09.2014 10:45, Michel Dänzer wrote:

On 30.09.2014 07:16, Brian Paul wrote:

The screen argument isn't actually used by lp_jit_screen_init() at this
time,


I guess that's why gcc didn't warn about it?


Nope, it actually does warn about it. Mea culpa for not noticing that.


Yeah, I patched this after seeing the gcc warning.  New, non-leaking, 
patch posted.  Thanks, Michel.


-Brian


___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev


[Mesa-dev] [PATCH 1/2] llvmpipe: move lp_jit_screen_init() call after allocation of screen object

2014-09-29 Thread Brian Paul
The screen argument isn't actually used by lp_jit_screen_init() at this
time, but let's move the call so that we pass a valid pointer.
---
 src/gallium/drivers/llvmpipe/lp_screen.c |6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/gallium/drivers/llvmpipe/lp_screen.c 
b/src/gallium/drivers/llvmpipe/lp_screen.c
index 3025322..e3c3980 100644
--- a/src/gallium/drivers/llvmpipe/lp_screen.c
+++ b/src/gallium/drivers/llvmpipe/lp_screen.c
@@ -557,9 +557,6 @@ llvmpipe_create_screen(struct sw_winsys *winsys)
return NULL;
 #endif
 
-   if (!lp_jit_screen_init(screen))
-  return NULL;
-
 #ifdef DEBUG
LP_DEBUG = debug_get_flags_option(LP_DEBUG, lp_debug_flags, 0 );
 #endif
@@ -570,6 +567,9 @@ llvmpipe_create_screen(struct sw_winsys *winsys)
if (!screen)
   return NULL;
 
+   if (!lp_jit_screen_init(screen))
+  return NULL;
+
screen-winsys = winsys;
 
screen-base.destroy = llvmpipe_destroy_screen;
-- 
1.7.10.4

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev


Re: [Mesa-dev] [PATCH 1/2] llvmpipe: move lp_jit_screen_init() call after allocation of screen object

2014-09-29 Thread Michel Dänzer

On 30.09.2014 07:16, Brian Paul wrote:

The screen argument isn't actually used by lp_jit_screen_init() at this
time,


I guess that's why gcc didn't warn about it?



but let's move the call so that we pass a valid pointer.
---
  src/gallium/drivers/llvmpipe/lp_screen.c |6 +++---
  1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/gallium/drivers/llvmpipe/lp_screen.c 
b/src/gallium/drivers/llvmpipe/lp_screen.c
index 3025322..e3c3980 100644
--- a/src/gallium/drivers/llvmpipe/lp_screen.c
+++ b/src/gallium/drivers/llvmpipe/lp_screen.c
@@ -557,9 +557,6 @@ llvmpipe_create_screen(struct sw_winsys *winsys)
 return NULL;
  #endif

-   if (!lp_jit_screen_init(screen))
-  return NULL;
-
  #ifdef DEBUG
 LP_DEBUG = debug_get_flags_option(LP_DEBUG, lp_debug_flags, 0 );
  #endif
@@ -570,6 +567,9 @@ llvmpipe_create_screen(struct sw_winsys *winsys)
 if (!screen)
return NULL;

+   if (!lp_jit_screen_init(screen))
+  return NULL;
+


This leaks screen in the failure case.


--
Earthling Michel Dänzer|  http://www.amd.com
Libre software enthusiast  |Mesa and X developer
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev


Re: [Mesa-dev] [PATCH 1/2] llvmpipe: move lp_jit_screen_init() call after allocation of screen object

2014-09-29 Thread Michel Dänzer

On 30.09.2014 10:45, Michel Dänzer wrote:

On 30.09.2014 07:16, Brian Paul wrote:

The screen argument isn't actually used by lp_jit_screen_init() at this
time,


I guess that's why gcc didn't warn about it?


Nope, it actually does warn about it. Mea culpa for not noticing that.


--
Earthling Michel Dänzer|  http://www.amd.com
Libre software enthusiast  |Mesa and X developer
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev