Re: About dispatching process

If RB3 is pointed to by TCBRBP, RB3 is newest and is the top RB.  TopRB get executed first when the TCB is given control.  The way it’s chained is RB3-RBLINK field points to RB2.  RB2-RBLINK points to RB1 and RB1-RBLINK points to the TCB.  

 

 

I think to obtain a spin lock (unconditional request) the SETLOCK routine first disables then goes into a spin loop to execute a sequence of instructions.  The instruction sequences are tests-for-and-if-available-obtain-the-lock, return if successful, enable for interrupt, then disable, retry the lock obtain.  So in that sense the spin loop is the same as the disabled loop.  The enablement is accomplished via the WINDOW macro which enables for external interrupts and immediately disables so that emergency signal maybe received.  The number of time to loop I think is limited (the detection) by a value set in some system area, possibly the CVT.

 

(Of course somewhere within the instruction sequence, it would have to check for hierarchy.)

 

 

Regards,

Raymond Wong

Compuware Corp

 





-----Original Message-----
From: Johnny Luo <johnny.xingkui.luo@xxxxxxxxx>
To: IBM-MAIN@xxxxxxxxxxx
Sent: Thu, 25 Oct 2007 7:19 am
Subject: Re: About dispatching process





On 10/24/07, Vernooy, C.P. - SPLXM <Kees.Vernooy@xxxxxxx> wrote:



I think what the manual says is the same as what you said. In a MP
system, a
disabled loop is observed as a spin loop by other processors.

This is not exactly true:
A disabled loop by itsself will not generate a problem. However, other
processors will regularly want to talk to the disabled processor, via
the SIGP instruction. The calling processor will spin (in a disabled
loop, therefor spin loop) waiting for an answer. The disabled called
processor will of course not answer and the calling processor will
detect this and generate an "excessive spinloop time out" condition. So
the spinloop is a normal, short duration, situation; an excisive
spinloop timeout is a condition detected by an active processor, where
an other processor does not answer. This can be for many reasons, being
disabled, hardware problems etc. etc.

Have a look at Initialization and Tuning Reference, member EXCPATxx, for
more information on spinloops.

his leads me to reconsider my understanding about the process of getting a
pin lock. I used to think what SETLOCK does is just keeping on check the
ockword and running disabled. Then, what is the difference between a spin
oop and a disabled loop?
So my guess is, what SETLOCK does is a 'timed disabled loop'. That is, most
f the time it'll loop disabled but not forever. It will enable the
rocessor regularly thus an excessive spin loop can be detected and handled
y system. Did I miss something here?

have another question about RB queue/chain. (A little off-topic but at
east it is involved in dispatching process -_- )
Here is a RB chain after a new RB is added:
TCB <-- RB1 <-- RB2 <-- RB3
I think RB1 is the oldest and RB3 is the newest but is it true?
Then this TCB is dispatched and the RB pointed by TCBRBP will be executed.
hich one? I think it should be RB3 ( the newest) but I'm not sure.
Thanks,
ohnny
----------------------------------------------------------------------
or IBM-MAIN subscribe / signoff / archive access instructions,
end email to listserv@xxxxxxxxxxx with the message: GET IBM-MAIN INFO
earch the archives at http://bama.ua.edu/archives/ibm-main.html


________________________________________________________________________
Email and AIM finally together. You've gotta check out free AOL Mail! - http://mail.aol.com

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to listserv@xxxxxxxxxxx with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

.

Relevant Pages

  • Re: programming language
    ... you will find the source code to my bf interpreter. ... instruction_pointer is the index of the instruction currently being executed in the instruction array. ... execute() is where the action happens. ... executegets a pointer to a bf_vm, where it executes one instruction, increments the instruction pointer of the bf_vm so that it points to the next instruction (or does a loop), and returns. ...
    (comp.programming)
  • Re: How much does it take to execute MMX instruction?
    ... a unrolled loop with lots of nop's in the ... This way we have accurate enough instruction timings. ... Pentium M, in general, has latency one clock cycle less, than Pentium ...
    (comp.lang.asm.x86)
  • Re: IAR MSP430 compiler problem
    ... Does anybody knows how to force compiler to use call instruction ... to next instruction after Spin function..... ... But it doesn't actually co-operate - an eternal loop is not co-operative, even if it you cheat and break out using interrupts. ... Interrupts are inherently asynchronous - if the thread can be suspended by an interrupt function, ...
    (comp.arch.embedded)
  • Re: Is a RISC chip more expensive?
    ... the instruction grouping rules on the 21164 were complicated. ... and these are only as good as the compiler writer... ... What makes you think there even was a scheduler? ... of the scalar b outside the loop that modified the vector a. ...
    (comp.arch)
Loading