Re: [Gems-users] About Cache Replacement


Date: Wed, 14 Oct 2009 17:25:09 -0400
From: hb166307@xxxxxxxx
Subject: Re: [Gems-users] About Cache Replacement
When a replacement, an event is triggered with the address gotten from 
L2cacheMemory.cacheprobe() and causes the replacement but that request does 
not get poped from the queue and I assume gets recycled some how and since 
there is room now that request triggers a different event like L1_GETS or 
something...

Quoting dai jianwei <adaihf@xxxxxxxxx>:

> --0015175cd13209e9a00475b4a695
> Content-Type: text/plain; charset=ISO-8859-1
> 
> Hi, all
>   I have a question about cache replacement. From .sm file, when a
> replacement happens, e.g L2_Replacement in
> MESI_CMP_directory-L2cache.sm, I
> only see that the status of the victim cache line in the L2 will be
> set to
> NOT PRESENT. However, I do not see when the victim cache line will
> be
> replaced with the new cache line (for example, requested from a L1
> miss). Is
> there anyone who has this information to share with me? thanks a lot
> for
> your help.
> 
> 
> Jianwei
> 
> --0015175cd13209e9a00475b4a695
> Content-Type: text/html; charset=ISO-8859-1
> Content-Transfer-Encoding: quoted-printable
> 
> Hi, all=A0<div><br></div><div>=A0=A0I have a question about cache
> replaceme=
> nt. From .sm file, when a replacement happens, e.g L2_Replacement in
> MESI_C=
> MP_directory-L2cache.sm, I only see that the status of the victim
> cache lin=
> e in the L2 will be set to NOT PRESENT. However, I do not see when
> the vict=
> im cache line will be replaced with the new cache line (for example,
> reques=
> ted from a L1 miss). Is there anyone who has this information to
> share with=
>  me? thanks a lot for your help.</div>
> <div><br></div><div><br></div><div>Jianwei</div>
> 
> --0015175cd13209e9a00475b4a695--
> 
[← Prev in Thread] Current Thread [Next in Thread→]