Changes between Initial Version and Version 1 of Ticket #1041, comment 23


Ignore:
Timestamp:
Mar 20, 2012, 8:28:23 PM (5 years ago)
Author:
qyot27
Comment:

Legend:

Unmodified
Added
Removed
Modified
  • Ticket #1041, comment 23

    initial v1  
    11The issue is (as far as I can tell), that the assembly for function ff_sbr_sum_square_sse is somehow not fully correct SSE, perhaps - as was the case with ff_sbr_hf_g_filt_sse before it got fixed - mixed with SSE2 instructions.  Processors that lack SSE2 support, like the Pentium III and Athlon XP, will generate silent output when attempting to process SBR-laced data because that assembly isn't strictly SSE and can't correctly be used on such processors.  Comment out that specific line which calls it, and decoding works properly without having to disable all assembly optimizations.  This was already known. 
     2 
     3As evidence of this, the below patch is roughly what I've been using for the last 10 days since it was pinned down (sans the git-am metadata): 
     4http://pastebin.com/P7nDj6vH