LOGIN   :::   RECOVER PASS   :::   GET ACCOUNT    
Browse
  • Projects
  • Code (CVS)
  • Forums
  • News
  • Articles
  • Polls
  •  
    OpenCores
  • FAQ
  • CVS HowTo
  • Mission
  • Media
  • Tools
  • Advertise
  • Mirrors
  • Logos
  • Contact us
  • Job Opportunity
  •  
    Tools
  • Search
      
  • Download Cores (CVSGet)
  •  
    More
  • Wishbone
  • Perlilog
  • EDA tools
  • OpenTech CD
  •  
    Navigation: All forums > Pci > Message List > Message Post

    Message

    Reply | Reply all
    Date Prev | Date Next | Thread Prev | Thread Next Date Index | Thread Index

    From: ezhcai@n...
    Date: Thu, 28 Mar 2002 05:25:38 +0100
    Subject: Re: [pci] why read is slower than write ?
    Top

    Hi,
    
    For read operation, the AD bus must have a turn-around cycle to ensure 
    there is no contention. So there is at least one wating cycle for a single 
    read operation, even you are implementing a fast-decode target. As for 
    write transfer, there is no such restriction, but for medium speed 
    address decode, there is also a wait state.
    I've ever developed a simple revision 2.1 PCI target module with medium 
    speed decode, which can be fund at 
    www.ntu.edu.sg/home/ezhcai/IPdesign
    It's a rough design, some signals need to be elaborated.
    
    ----- Original Message ----- 
    From: Madhusudhan Rao <madhu_sudhana_rao@y... > 
    To: pci@o...  
    Date: Wed, 27 Mar 2002 19:54:45 -0800 (PST) 
    Subject: Re: [pci] why read is slower than write ? 
    
    > 
    > 
    > hi, 
    >    Thanks for quick response. I am not talking about 
    > open cores. I developed by reading specification. 
    > I am implementing pci target. Let me expalin my doubt 
    > clearly. 
    > When i am reading data from ram(memory) it is taking 3 
    > cycles( 2 are wait cycles), but when i write any data 
    > on same memory it completing in a cycle witout any 
    > wait cycles. I don't have any FIFO's. 
    > 
    > Thanks in advance. 
    > Madhu 
    > 
    > --- Miha Dolenc <mihad@o... > wrote: 
    > > Hi! 
    > > 
    > > I don't know what you mean exactly!  Are you talking 
    > > about OpenCores pci 
    > > bridge? Master or Target? 
    > > If you are referring to why PCI Target has 1 initial 
    > > wait cycle for reads 
    > > and none for writes, the reason is because outputs 
    > > must be registered. 
    > > Target has to pull data out of FIFO and put it in 
    > > output registers before 
    > > asserting TRDY. 
    > > Did this help? 
    > > 
    > > Regards, 
    > > Miha Dolenc 
    > > 
    > > ----- Original Message ----- 
    > > From: "Madhusudhan Rao" 
    > > <madhu_sudhana_rao@y... > 
    > > To: <pci@o... > 
    > > Sent: Wednesday, March 27, 2002 11:11 AM 
    > > Subject: [pci] why read is slower than write ? 
    > > 
    > > 
    > > > Hi, 
    > > >    When i am implementing pci memory read 
    > > transactions 
    > > > i was in a need of keeping 2 wait cycles for 
    > > proper 
    > > > read operation but for write with 1 clock cycle i 
    > > am 
    > > > able to memory write. what may be the reason can 
    > > any 
    > > > body give suggestions. All kinds of inputs are 
    > > > welcome. 
    > > > 
    > > > Thanks in advance 
    > > > Madhu 
    > > > 
    > > > 
    > > > ===== 
    > > > MadhusudhanaRao.M 
    > > > 
    > > > __________________________________________________ 
    > > > Do You Yahoo!? 
    > > > Yahoo! Movies - coverage of the 74th Academy 
    > > Awards® 
    > > > http://movies.yahoo.com/ 
    > > > 
    > > > 
    > > 
    > > 
    > > 
    > 
    > 
    > ===== 
    > MadhusudhanaRao.M 
    > 
    
    
    
     
    Copyright (c) 1999 OPENCORES.ORG. All rights reserved.