Date: | Mon, 6 Jul 2009 01:13:44 -0700 (PDT) |
---|---|
From: | Muhammad abid Mughal <mabidm_pieas@xxxxxxxxx> |
Subject: | [Gems-users] Question regarding CacheMsg class |
hi guys:- Hope you all doing good. I just wanna confirm one thing : Lets say simics SPARC processor is going to execute Load/Store/Atomic instruction then CacheMsg::m_Address; CacheMsg::m_PhysicalAddress; represent physical address while Address m_LogicalAddress; represents Virtual address of the data to be fetched by Load/Store/Atomic . How about this CacheMsg::m_ProgramCounter? i think this is the address of Load/Store/Atomic in the Instruction cache. Is not it? If this is the case then for instruction
fetches all these data fields CacheMsg::m_Address; CacheMsg::m_PhysicalAddress; CacheMsg::m_ProgramCounter are the same? Waiting eagerly for speedy reply Thanks Muhammad abid ------------------------------- // file CacheMsg.h class CacheMsg : public Message { public: /** \ //private: //both have the same value and are physical addresses Address m_Address; /**< Line address for this request*/ Address m_PhysicalAddress; /**< Physical address for this request*/ Address m_LogicalAddress; /**<
Virtual address for this request*/ / this is also same as above see SimicsProcessor.c CacheRequestType m_Type; /**< Type of request (LD, ST, etc)*/ Address m_ProgramCounter; /**< Program counter of the instruction that caused the miss*/ // this is the virtual address AccessModeType m_AccessMode; /**< user/supervisor access type*/ int m_Size; /**< size in bytes of access*/ PrefetchBit m_Prefetch; /**< Is this a prefetch request*/ int m_Version; /**< Version associated with this request*/ int m_ThreadID; /**< The SMT thread that initiated this request*/ uint64 m_Timestamp; /**< The transaction timestamp of this request. Last commit time if request is non-transactional*/ bool m_ExposedAction; /**< Is this request part of
an exposed action*/ }; --------------------------------------------------- Get your new Email address! Grab the Email name you've always wanted before someone else does! |
[← Prev in Thread] | Current Thread | [Next in Thread→] |
---|---|---|
|
Previous by Date: | Re: [Gems-users] Running benchmark occurs L1 OVERFLOW and some benckmarks halt for a long time, huitzu shih |
---|---|
Next by Date: | Re: [Gems-users] Running benchmark occurs L1 OVERFLOW and some benckmarks halt for a long time, BYONG WU CHONG |
Previous by Thread: | [Gems-users] Question about PT_TO_PT network latency., BYONG WU CHONG |
Next by Thread: | [Gems-users] Ruby and Opal Compile Problem, Julia Lu |
Indexes: | [Date] [Thread] |