SPEC Benchmark CINT95 Summary
                                       

                     Base      Base      Base      Peak      Peak      Peak
   Benchmarks    Ref Time  Run Time     Ratio  Ref Time  Run Time     Ratio
   ------------  --------  --------  --------  --------  --------  --------
   099.go            4600       390     11.8       4600       403     11.4  
   124.m88ksim       1900       184     10.3       1900       144     13.2  
   126.gcc           1700       228      7.45      1700       176      9.66 
   129.compress      1800       205      8.80      1800       164     11.0  
   130.li            1900       220      8.65      1900       174     10.9  
   132.ijpeg         2400       199     12.1       2400       202     11.9  
   134.perl          1900       204      9.30      1900       147     12.9  
   147.vortex        2700       288      9.38      2700       234     11.5  
   SPECint_base95 (Geom. Mean)           9.61                              
   SPECint95 (Geom. Mean)                                             11.5 


                              TESTER INFORMATION
                              ------------------
     SPEC License #: 2
          Tested By: Digital NH
          Test Date: Oct-96
     Hardware Avail: Nov-96
     Software Avail: Dec-96


                                   HARDWARE
                                   --------
         Model Name: AlphaServer 1000A 5/400
                CPU: 400 MHz 21164A
                FPU: Integrated
   Number of CPU(s): 1
      Primary Cache: 8KB(I)+8KB(D) on chip
    Secondary Cache: 96KB
        Other Cache: 2MB
             Memory: 128MB
     Disk Subsystem: 1 x 2GB
     Other Hardware: Ethernet


                                   SOFTWARE
                                   --------
   Operating System: Digital UNIX V4.0A
           Compiler: DEC C V5.4-045
        File System: UFS
       System State: Multi User
  Kernel Extensions: 


                                    NOTES
                                   -------
Compiler: cc.alt -std1    Base optimizations: -O4 -arch ev56 -non_shared -om
Portability flags: m88ksim: -DLEHOST   perl: -DI_TIME   vortex: -D__RISC_64__

Peak flags: all use -ifo -non_shared.  Other flags are: 
go: -g3 -O3 -inline speed EXTRA_LDFLAGS -om    m88ksim: -g3 -O4 -speculate all -inline speed with feedback
gcc: -g1 -O4 -inline speed -arch ev56 -xtaso_short -taso with feedback    compress: -g3 -O4 -tune ev5 -32data 
-inline speed -arch ev56 -assume whole_program -om -lsys5    li: -g1 -O4 -inline speed -xtaso_short -speculate 
all -lsys5 -taso with feedback    ijpeg: -g3 -O4 -fast -unsigned -inline speed -arch ev56 -speculate all -lsys5 
with feedback    perl: -g3 -O4 -inline speed -arch ev56 -lsys5 with feedback pass 2 only    vortex: -g3 -O4 
-fast -inline speed -speculate all with feedback

feedback: PASS1=-gen_feedback 
fdo_pre1=mv %binary% %exename%.orig; pixie -pids %exename%.orig -o %exename%
fdo_post1=prof -pixie -feedback ../../src/%exename%.fb %exename%.orig %exename%.orig.Addrs %exename%.orig.Counts.*
PASS2=-feedback $(EXENAME).fb -r 
fdo_pre2=mv %exename% %exename%.rr; ld -o %exename%fb %exename%.rr -lexc; pixie -pids %exename%fb -o %exename%
fdo_run2=%command%
fdo_post2=ld -om %exename%.rr -o %exename%.rrom -lexc; prof -pixie -merge %exename%fb.Counts %exename%fb %exename%fb.Addrs 
%exename%fb.Counts.*; /usr/lib/cmplrs/cc.alt/om -om_ireorg_feedback %exename%fb -o %binary% %exename%.rrom
gcc and li add -taso to feedback commands om and ld