[llvm-bugs] [Bug 24714] New: ChromeOs+Clang is slower than gcc

[via llvm-bugs]

https://llvm.org/bugs/show_bug.cgi?id=24714

            Bug ID: 24714
           Summary: ChromeOs+Clang is slower than gcc
           Product: clang
           Version: unspecified
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
          Assignee: unassignedclangbugs at nondot.org
          Reporter: yunlian at chromium.org
                CC: llvm-bugs at lists.llvm.org
    Classification: Unclassified

There is up to 23% slow down when building ChromeOS with clang compared with
gcc.

                                            gcc                 clang  
                                         Amean  StdDev         Amean  StdDev   
GmeanSpeedup  

Benchmark:  page_cycler.intl_ar_fa_he

cold_times__page_load_time (ms)        1340.15  17.84        1442.03   2.80    
 -7.1%     
warm_times__page_load_time (ms)        1138.02  11.49        1208.40   5.67    
 -5.8%   


dom_traverse__dom_traverse (runs_per_s)247.89   1.29        238.58   3.26      
  3.8%  
dom_modify__dom_modify (runs_per_s)    293.15   6.84        266.14   2.37      
  -9.2%  
dom_query__dom_query (runs_per_s)      10641.70  54.41      8846.72 191.93     
-2.2%  
dom_attr__dom_attr (runs_per_s)        420.65   6.70       320.55   1.34       
-23.8%

Benchmark:  page_cycler.intl_es_fr_pt-BR                  
cold_times__page_load_time (ms)       1124.76   7.73        1206.06   7.26     
 -6.7%   
warm_times__page_load_time (ms)       926.05   3.63        1010.50   8.46      
  -8.4%  

Benchmark:  page_cycler.netsim.top_10                  
cold_times__page_load_time (ms)       1574.24 235.63       2125.32 100.59      
 -26.7%    


Benchmark:  page_cycler.tough_layout_cases; 
cold_times__page_load_time (ms)        3825.84  33.86        4130.14  19.08    
  -7.4%   
warm_times__page_load_time (ms)        3593.45  34.40        3918.36   6.02    
  -8.3%   

Benchmark:  kraken;  Iterations: 3 
Total__Total (ms)                     2509.57  10.33         2528.60   7.09    
 -0.8%    

Benchmark:  page_cycler.intl_ja_zh;  Iterations: 3               
cold_times__page_load_time (ms)       1225.20   5.22        1510.64 191.49     
   -18.2% 
warm_times__page_load_time (ms)        1410.32  11.43       1415.92 155.38     
   +0.2% 

Benchmark:  robohornet_pro;  
Total__Total (ms)                    7568.33 143.88        7633.33  50.87      
   -0.9%

Benchmark:  page_cycler.intl_hi_ru;  
cold_times__page_load_time (ms)      1084.93   3.83        1184.85   7.88      
       -8.4%
warm_times__page_load_time (ms)      910.26   4.15         1006.45  13.55      
      -9.6%

Benchmark:  sunspider;  
Total__Total (ms)               397.63   3.79          394.20   0.54           
   +0.9%    

Benchmark:  page_cycler.intl_ko_th_vi; 
cold_times__page_load_time (ms)  1579.75   6.85         1691.37   2.91         
     -6.6%    
warm_times__page_load_time (ms)  1325.57   0.61         1447.71  10.09         
    -8.4%    


Benchmark:  page_cycler.typical_25; 
cold_times__page_load_time (ms)  2039.44  15.94        2203.62  11.45          
 -7.5%     
warm_times__page_load_time (ms)  1825.37  15.06       2000.92  13.10           
  -8.8%

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20150904/80ec25ed/attachment-0001.html>