<div dir="ltr"><font face="arial, helvetica, sans-serif">Perf stat results for shader-db:</font><div><font face="arial, helvetica, sans-serif"><br></font></div><div><font face="arial, helvetica, sans-serif">This is measured on an AMD Kaveri CPU.</font></div><div><font face="arial, helvetica, sans-serif"><br></font></div><div><font face="arial, helvetica, sans-serif">gcc-6.2.0 </font><span style="font-family:arial,helvetica,sans-serif">-fno-omit-frame-pointer -g -O2</span></div><div><div><div><font face="arial, helvetica, sans-serif"><br></font></div><div><span style="font-family:arial,helvetica,sans-serif">**** Unpatched:</span><font face="arial, helvetica, sans-serif"><br></font><br><div><font face="monospace, monospace">$ cd shader-db</font></div><div><font face="monospace, monospace">$ ../run-upstream perfstat-u --repeat=5 -- ./run -1 shaders >/dev/null</font><br style="font-family:monospace,monospace"><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace"> Performance counter stats for './run -1 shaders' (5 runs):</span><br style="font-family:monospace,monospace"><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace">      13689.962374      task-clock (msec)         #    1.000 CPUs utilized            ( +-  0.29% )</span><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace">               138      context-switches          #    0.010 K/sec                    ( +- 17.82% )</span><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace">                 6      cpu-migrations            #    0.000 K/sec                    ( +- 13.36% )</span><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace">            78,559      page-faults               #    0.006 M/sec                    ( +-  0.24% )</span><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace">    53,578,642,861      cycles:u                  #    3.914 GHz                      ( +-  0.29% )</span><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace">    44,813,859,985      instructions:u            #    0.84  insn per cycle           ( +-  0.01% )</span><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace">     1,069,586,875      cache-references:u        #   78.129 M/sec                    ( +-  0.65% )</span><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace">        51,295,256      cache-misses:u            #    4.796 % of all cache refs      ( +-  0.56% )</span><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace">     9,508,996,305      branches:u                #  694.596 M/sec                    ( +-  0.01% )</span><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace">       453,237,236      branch-misses:u           #    4.77% of all branches          ( +-  0.84% )</span><br style="font-family:monospace,monospace"><br style="font-family:monospace,monospace"><span style="font-family:monospace,monospace">      13.692494394 seconds time elapsed                                          ( +-  0.29% )</span><br></div><div><span style="font-family:monospace,monospace"><br></span></div><div><span style="font-family:arial,helvetica,sans-serif">**** Patched:</span><span style="font-family:monospace,monospace"><br></span></div><div><span style="font-family:monospace,monospace"><br></span></div><div><div><font face="monospace, monospace">$ cd shader-db</font></div><div><font face="monospace, monospace">$ ../run-upstream-patched perfstat-u --repeat=5 -- ./run -1 shaders >/dev/null</font></div><div><br></div><div><font face="monospace, monospace"> Performance counter stats for './run -1 shaders' (5 runs):</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">      13602.106171      task-clock (msec)         #    1.000 CPUs utilized            ( +-  0.14% )</font></div><div><font face="monospace, monospace">                86      context-switches          #    0.006 K/sec                    ( +- 13.95% )</font></div><div><font face="monospace, monospace">                 6      cpu-migrations            #    0.000 K/sec                    ( +- 26.35% )</font></div><div><font face="monospace, monospace">            78,271      page-faults               #    0.006 M/sec                    ( +-  0.82% )</font></div><div><font face="monospace, monospace">    53,299,046,681      cycles:u                  #    3.918 GHz                      ( +-  0.13% )</font></div><div><font face="monospace, monospace">    44,577,707,063      instructions:u            #    0.84  insn per cycle           ( +-  0.01% )</font></div><div><font face="monospace, monospace">     1,078,158,307      cache-references:u        #   79.264 M/sec                    ( +-  0.70% )</font></div><div><font face="monospace, monospace">        51,521,287      cache-misses:u            #    4.779 % of all cache refs      ( +-  1.03% )</font></div><div><font face="monospace, monospace">     9,459,962,609      branches:u                #  695.478 M/sec                    ( +-  0.01% )</font></div><div><font face="monospace, monospace">       456,593,871      branch-misses:u           #    4.83% of all branches          ( +-  0.27% )</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">      13.603795247 seconds time elapsed                                          ( +-  0.14% )</font></div></div></div></div></div></div>