[prev in list] [next in list] [prev in thread] [next in thread] 

List:       bochs-dev
Subject:    Re: [Bochs-developers] BX_CPU_LEVEL,
From:       "Sebastian Herbszt" <herbszt () gmx ! de>
Date:       2010-05-13 16:38:34
Message-ID: 4320B0673F564CF5BCC83075DD3259A4 () FSCPC
[Download RAW message or body]

Stanislav Shwartsman wrote:
> Yes, the magic for me was to understand that :
> 
> -#if BX_PHY_ADDRESS_LONG
> -    // can share writeStamps between multiple pages
> -    return (Bit32u) ((pAddr >> 12) & (PHY_MEM_PAGES-1));
> -#else
> -    return (Bit32u) (pAddr) >> 12;
> -#endif
> 
> Can be replaced by:
> 
> +    // can share writeStamps between multiple pages if >32 bit phy address
> +    return ((Bit32u) pAddr) >> 12;
>   }
> 
> So where will be no perf difference on every pageWriteStamp access anymore.
> 
> Not I measure less than 1% slowdown from BX_PHY_ADDRESS_LONG. Worth to be
> enabled by default.
> 
> Stanislav

Nice - this enables the use of SeaBIOS on Bochs with BX_CPU_LEVEL >= 5.

Sebastian


------------------------------------------------------------------------------

_______________________________________________
bochs-developers mailing list
bochs-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bochs-developers
[prev in list] [next in list] [prev in thread] [next in thread]