[Disksim-users] Disksim-users Digest, Vol 86, Issue 5
Sinhome
chensinhome at gmail.com
Wed Jan 16 18:09:51 EST 2013
從我的 iPhone 傳送
disksim-users-request at ece.cmu.edu 於 2013/1/17 上午1:00 寫道:
> Send Disksim-users mailing list submissions to
> disksim-users at ece.cmu.edu
>
> To subscribe or unsubscribe via the World Wide Web, visit
> https://sos.ece.cmu.edu/mailman/listinfo/disksim-users
> or, via email, send a message with subject or body 'help' to
> disksim-users-request at ece.cmu.edu
>
> You can reach the person managing the list at
> disksim-users-owner at ece.cmu.edu
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Disksim-users digest..."
>
>
> Today's Topics:
>
> 1. A problem in the bus ownership grant (???)
> 2. Stopping simulation because of saturation (??)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 16 Jan 2013 14:39:33 +0800
> From: ??? <yinxunrui at gmail.com>
> Subject: [Disksim-users] A problem in the bus ownership grant
> To: disksim-users at ece.cmu.edu
> Message-ID:
> <CAC3HEeMDFdvk7bwKvtdW79kN4-v0ygcGOqYfNd75S=gvXJGg+w at mail.gmail.com>
> Content-Type: text/plain; charset=ISO-8859-1
>
> Hi all,
>
> I encounter the following assertion failure using disksim-4.0:
> in disksim_diskctlr.c:
> in function: disk_bus_ownership_grant():
> ddbg_assert(arbdelay == (simtime - currdisk->stat.requestedbus));
>
> First, I comment this assertion, and print the corresponding values in
> my simulation, finding that the failure is actually caused by the
> precision problem of floating point numbers. It appears only 3 times
> and my simulation runs well after that...
>
> So I wonder if it is a bug to compare the float numbers in this way,
> or there is some special purpose for this assertion?
> Referring to a similar part in the source code of the SSD extension, I
> comment this assertion and changed the line
> currdisk->stat.waitingforbus += simtime - currdisk->stat.requestedbus;
> into
> currdisk->stat.waitingforbus += arbdelay;
>
> Is that OK?
>
> Thank you for your time!
>
> Best regards,
> Xunrui
>
>
> ------------------------------
>
> Message: 2
> Date: Wed, 16 Jan 2013 19:54:52 +0800
> From: ?? <hshopeful1 at gmail.com>
> Subject: [Disksim-users] Stopping simulation because of saturation
> To: disksim-users at ece.cmu.edu
> Message-ID:
> <CAGSMsAHhoMv8CmqMYDnYxaiVXTV1GwN659tCwu_nx_q1Vrbnag at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hi all,
>
> I used disksim-3.0 to test Financial1 trace.
>
> And I created a procedure iotrace_financial1_get_ioreq_event() in
> Disksim_iotrace.c
>
> I also made some changes in ascii.parv to fit the financial1 trace(mainly
> about the devno).
>
> but the problem occurs:
> Stopping simulation because of saturation: simtime 1919.778808, totalreqs
> 100003
> disksim_run_simulation(): simulated 118191 events
>
> I looked up the Disksim_users Archives,and tried the following two
> solutions:
> 1. change the value of MAX_QUEUE_LENGTH in
> Disksim-3.0/src/disksim_logorg.c
>
It works for me...
> 2.change the value of I/O Trace Time Scale=1.0(amplify it to 2.0,3.0 even
> to 10.0)
>
The value is a unfinished job
Changing it in parv file can't do anything
Try to change it directly in source code
> But they don't work!
>
> If you have the same problem and solution, please drop me a message. I will
> be very appreciated.
>
> Best regards
>
> thanks
>
> Heshuang
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <http://sos.ece.cmu.edu/pipermail/disksim-users/attachments/20130116/0a8a3704/attachment-0001.html>
>
> ------------------------------
>
> _______________________________________________
> Disksim-users mailing list
> Disksim-users at ece.cmu.edu
> https://sos.ece.cmu.edu/mailman/listinfo/disksim-users
>
>
> End of Disksim-users Digest, Vol 86, Issue 5
> ********************************************
More information about the Disksim-users
mailing list