[XviD-devel] performance analysis results
Venkata Tumati
vt35 at cornell.edu
Sun Mar 9 11:33:55 CET 2003
Hi,
I was doing some performance analysis using SGI Speedshop on xvid.
1st setup was using xvid_20021230 and the 2nd setup was the latest
download from yesterday xvidcore-0.9.1.
The results are pretty contrasting. Any way here is the summary. Also I
made some changes to xvid_encraw.c and xvid_decraw.c to encode 3 visual
objects; I create three instances of the encoder and store the results
sequentially in a file. For decoding I use the final output from the
encoding of 3visual objects and decode it. I hope this is how multiple
visual are encoded.
All the tests were run on SGI R12k 300mhz don't know the exact cache
config but I think it has a 2MB L2 cache.
Xvidcore-0.9.1:(look the total time for each config)
1vo 740
L1 Cache miss rate: .77%
L2 Cache miss rate: 13.91%
Cycles: 54.79
1vo 1024
L1 Cache miss rate: .58%
L2 Cache miss rate: 23.12%
Cycles: 55.79
3vo 740
L1 Cache miss rate: .73%
L2 Cache miss rate: 17.03%
Cycles: 165.97
3vo 1024
L1 Cache miss rate: .57%
L2 Cache miss rate: 23.25%
Cycles: 159.83
Xvid_20021230:
1vo 740
L1 Cache miss rate: .09%
L2 Cache miss rate: 17.93%
Cycles: 275.51
1vo 1024
L1 Cache miss rate: .06%
L2 Cache miss rate: 25.24%
Cycles: 280.15
3vo 740
L1 Cache miss rate: .09%
L2 Cache miss rate: 16.51%
Cycles: 893.906
3vo 1024
L1 Cache miss rate: .07%
L2 Cache miss rate: 22.36%
Cycles: 847.92
Does any body know why we are having more L1 cache misses, what changes
to the code are causing this. Is this due to some optimizations when the
code is compiled?
-Venkata
More information about the XviD-devel
mailing list