summaryrefslogtreecommitdiff
path: root/src/liba52/xine_a52_decoder.c
diff options
context:
space:
mode:
Diffstat (limited to 'src/liba52/xine_a52_decoder.c')
-rw-r--r--src/liba52/xine_a52_decoder.c21
1 files changed, 19 insertions, 2 deletions
diff --git a/src/liba52/xine_a52_decoder.c b/src/liba52/xine_a52_decoder.c
index 6c4442996..b8b500ecd 100644
--- a/src/liba52/xine_a52_decoder.c
+++ b/src/liba52/xine_a52_decoder.c
@@ -673,8 +673,25 @@ static audio_decoder_t *open_plugin (audio_decoder_class_t *class_gen, xine_stre
this->pts_list[0] = 0;
this->pts_list_position = 0;
- if( !this->a52_state )
- this->a52_state = a52_init (xine_mm_accel());
+ if( !this->a52_state ) {
+ this->a52_state =
+#ifdef HAVE_A52DEC_A52_H /* External liba52 */
+ /* When using external liba52, enable _all_ capabilities, even
+ if that might break stuff if they add some new capability
+ that depends on CPU's caps.
+ At the moment the only capability is DJBFFT, which is tested
+ only if djbfft is being used at compile time.
+
+ The actual question would be: why don't they check for
+ capabilities themselves?
+ */
+#warning "Enabling all external liba52 capabilities."
+ a52_init (0xFFFFFFFF)
+#else
+ a52_init (xine_mm_accel())
+#endif
+ ;
+ }
/*
* find out if this driver supports a52 output