[MPlayer-dev-eng] [PATCH] final (hopefully) vo_tdfxfb.c delay patch

[Burton Samograd]

> do usec_sleep(10*1000) then. it's constant 10ms delay.
> at least it won't burn the hardware :)
>

Thanks for the usec_sleep suggestion.  That works and gets rid of the
memcpy in a loop (the card status check didn't seem to do anything :-/
)

burton

-------------- next part --------------
? h
Index: vo_tdfxfb.c
===================================================================
RCS file: /cvsroot/mplayer/main/libvo/vo_tdfxfb.c,v
retrieving revision 1.16
diff -u -r1.16 vo_tdfxfb.c
--- vo_tdfxfb.c	28 Aug 2002 18:38:41 -0000	1.16
+++ vo_tdfxfb.c	28 Aug 2002 19:05:03 -0000
@@ -158,8 +158,7 @@
      A printf before the memset call also seems to work, but this made more sense
      since it actually checks the status of the card. 
   */
-  /* first wait for the card to be ready, do not try to write every time - alex */
-  do {} while((reg_IO->status & 0x1f) < 1);
+  usec_sleep(10*1000);
   memset(vidpage, 0, screenwidth * screenheight * screendepth);
   
   	if(vo_doublebuffering) {