> On 2012-01-10 10:01:06, Nilay Vaish wrote:
> >
> 
> Andreas Hansson wrote:
>     It seems this patch causes the following regression to fail with a minor 
> statsdiff:
>     
> build/ALPHA_SE_MOESI_hammer/tests/opt/quick/50.memtest/alpha/linux/memtest-ruby-MOESI_hammer
>  FAILED!
>     
>     
>     Maximum error magnitude: +0.245398%
>     
>                                       Reference  New Value   Abs Diff   Pct 
> Chg
>     Key statistics:
>     
>       host_mem_usage                     381740     376848      -4892    
> -1.28%
>       sim_ticks                        19175808   19128751     -47057    
> -0.25%
>     
>     Differences > 0%:
>     
>       sim_ticks                        19175808   19128751     -47057    
> -0.25%
>       sim_seconds                      0.019176   0.019129  -0.000047    
> -0.25%
>       system.cpu4.num_writes              53497      53534         37    
> +0.07%
>       system.cpu3.num_writes              53686      53709         23    
> +0.04%
>       system.cpu0.num_writes              53926      53947         21    
> +0.04%
>       system.cpu5.num_reads               99742      99706        -36    
> -0.04%
>       system.cpu4.num_reads               99524      99559         35    
> +0.04%
>       system.cpu6.num_reads               99887      99859        -28    
> -0.03%
>       system.cpu1.num_writes              53707      53693        -14    
> -0.03%
>       system.cpu1.num_reads               98882      98903         21    
> +0.02%
>       system.cpu7.num_reads               99347      99367         20    
> +0.02%
>       system.cpu3.num_reads               99594      99611         17    
> +0.02%
>       system.cpu6.num_writes              53292      53283         -9    
> -0.02%
>       system.cpu2.num_reads               98977      98993         16    
> +0.02%
>       system.cpu5.num_writes              53984      53981         -3    
> -0.01%
>       system.cpu2.num_writes              53060      53059         -1    
> -0.00%
>       system.cpu7.num_writes              53300      53299         -1    
> -0.00%
>
> 
> Andreas Hansson wrote:
>     Nilay, Brad, 
>     
>     Do the changes to the memtest-ruby-MOESI_hammer regression make sense?
>

The functional reads and writes will probably happen more easily. So
this is expected.


- Nilay


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviews.m5sim.org/r/984/#review1882
-----------------------------------------------------------


On 2012-01-10 09:19:23, Andreas Hansson wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://reviews.m5sim.org/r/984/
> -----------------------------------------------------------
> 
> (Updated 2012-01-10 09:19:23)
> 
> 
> Review request for Default.
> 
> 
> Summary
> -------
> 
> Ruby: Change the access permissions for MOESI hammer
> 
> This patch changes the access permission for the WB_E_W state from
> Busy to Read_Write to avoid having issues in follow-on patches with
> functional accesses going through Ruby. This change was made after
> consultation with all involved parties and is more of a work-around
> than a fix.
> 
> 
> Diffs
> -----
> 
>   src/mem/protocol/MOESI_hammer-dir.sm a1d5a0e2e970 
> 
> Diff: http://reviews.m5sim.org/r/984/diff
> 
> 
> Testing
> -------
> 
> util/regress all passing (disregarding t1000 and eio)
> 
> 
> Thanks,
> 
> Andreas
> 
>

_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev

Reply via email to