On 07/18/11 18:26, Edward Jaffe wrote:
On 7/18/2011 12:55 PM, Shmuel Metz (Seymour J.) wrote:
Are you saying that the processor starts fetching instructions for
the branch prior to executing it?
Absolutely! There is a multi-stage pipeline that allows the processor
to get ahead of the current instruction's execution to fetch and
decode instructions, resolve addresses, fetch operands, etc. in
advance of the actual instruction execution. Branch prediction (with a
history table or target buffer) is used to help ensure the pipeline
keeps flowing even when branches are encountered. IIRC, some models
even fetch down more than one branch target path to minimize the
performance hit if the first prediction is wrong.
Sounds almost thiotimoline like to me. (Sorry, I'm almost in vacation mode)/
/
--
Mark Jacobs
Time Customer Service
Tampa, FL
----
Some people are electrifying, they light up
a room when they leave.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
