The previous solution should have two more nops at the end of the code. I have fixed it. Thanks to Jingcheng for pointing the error out.
Cheers,
Hui
Dear All,
I made a mistake in the last lecture about the marks distributions of final exam. The fact is that Part 1 and Part have equal marks, i.e., 50 marks to 50 marks, not 60 marks to 40 marks as I told you before. Sorry for the mistake.
Cheers,
Hui
Dear All,
I have scheduled two consultation sessions:
Where: K17-508 Consultation Room
When: 1:00-5:00pm Wed 15 Nov, 10:00am -2:00pm Thu 16 Nov
K17-508 Consultation Room is a small room beside the lift on level 5 of K17. If I'm not there, call me on ext 56572 using the phone on the wall beside the lift.
Cheers,
Hui
Dear All,
Please check all your marks. Note that all the marks are raw marks and will be converted into the final marks using the assessment formula given in the course outline. If you have any queries about your marks, contact me as soon as possible.
Cheers,
Hui
Dear All,
In order to help you manage your time for multiple courses, I have extended the documentation submission deadline to Tuesday midnight.
Note that each group submits one copy only.
Cheers,
Hui
Dear COMP2121 Students,
Please fill up the MyExperience course survey forms. The deadline is Next Tuesday. Your feedback is most appreciated.
Best,
Hui
Dear students,
All lab marks would be in the system by Tuesday. Please check them all and if there was any mistake* please keep note of that and let me know on project demo date and I will fix it for you. (I'll bring all marking sheets with my self)
* by mistake I mean if a tutor has already marked you but the mark on the system is different from what you have been told.
For any other cases, Including extension, please contact course coordinator, Professor Hui.
Best Regards,
Arash Bayat
Please use the following link to book a time slot for your project demo.
If you book between 9am to 13pm you should attend in Clavier LAB to be marked. For those who booked between 2pm to 4pm you should attend in K-E15-G048 - Quad G048 ( Quadrangle G048 [TUSM] ).
There would be 3 or 4 time slot in the morning to book, I am looking to add some extra time slot in the afternoon (not sure if I could).
The link to the booking page will be sent to you when I finalise time slots. Possibly tomorrow.
Those groups that need a board to do the project should come to the lab this Thursday (11-13) or Friday (13-15) to collect a board.
Other groups, please don't forget to bring the extra board back this week.
Dear student,
Due to the lack of time, those students who are not in Friday Lab but wants to get marked then should show up at the beginning of Lab (before 3:15, and we record their names when they arrive) and we mark them in order. If they show up later, there is no guarantee for them to be marked and they will receive 0 as the Lab mark.
Best Regards,
Arash Bayat
Dear All,
We have arranged extra tutors to mark Lab 5 for all the students who are not in Friday Lab. If you cannot finish Lab 5 by your lab time, come to Friday lab class to get your programs marked without any late penalty.
Cheers,
Hui
Please ignore this message if you already have a group and your group information is submitted in the previous google form that I sent.
Dear Students,
All ungrouped student should submit below form by Tuesday 17-Oct-2017 14:00pm. Otherwise, I will take no responsibility for them being ungrouped and fail in project.
Ungrouped Student Form link : https://goo.gl/forms/I7CXNx0iISipHszJ3
Then you can see submission from other students in the google sheet below. You can contact them to make groups. In case you find a group mate, one of you should submit your group information in previous google form (I copy the link below for you).
Find other ungrouped students in this sheet: https://docs.google.com/spreadsheets/d/1m3WvR3i3S2...
Previous google form for group information : https://goo.gl/forms/wOgEnkd3SaH5vuX92
The above procedure is a must otherwise you remain ungrouped and will receive 0 for your project. If you find a group mate by other means and submit your group information you can ignore the above procedure.
Also, I made a public discussion group in facebook " Ungrouped Student UNSW CSE COMP2121 17s2 ". You can join there and see if you can find a group mate.
When you find a group mate you should edit your submission in the ungrouped student form and answer yes to the last question.
I will randomly assign remaining ungrouped student on Thursday 19-Oct-2017 16:00 pm and notify them by email (zid@unsw.edu.au).
Best Regards,
Arash Bayat
Dear Students,
All groups must return their extra board next week.
Note that next week is the last week of the lab and all of your work has to be marked by tutors.
The last 30 minutes of next lab is reserved for tutors to mark you.
Best Regards,
Arash Bayat
All Group must submit their information using the link below. I need this information as soon as possible.
Only one student per group should submit this form .
Please read my previous notice " IMPORTANT: Submit your group information for project ".
Dear Students,
While most of the group have 2 members (happy to work together for the project), there are some groups where
Since creating new groups (allowing students to join from different classes) is a quite difficult process, I decided to automate this process using the form below.
https://goo.gl/forms/t11w69YMt0NoPQy92
When you agree with your partner to work on the project (ask them if they agree), one student from each group should submit the form.
Please make sure you submit the form by Sunday so that we can manage the remaining ungrouped students during next week.
Every student is responsible to make sure that he/she has a partner to do the project and we will do our best to make sure that every group has a board to work with.
Best Regards,
Arash Bayat
Dear students,
As you may noticed Hui asked all students to work in group of 2 for the project. While we allow group of 3 for doing the lab, for the project we have to make sure that all groups have exactly 2 members (and all groups have a board to play with). So if you are in group of 3 please let your tutor know in the earliest time so that he can rearrange your group for the project. In case necessary we may join you with a student from other class.
Since some students do not show up in the class and we want to make some changes in the groups, it is your responsibility to confirm with your group mate if he/she is going to collaborate with you doing the project.
Best Regards,
Arash Bayat
Dear All,
This is a group project. You work in a group of two students. If you cannot find a project mate, contact Arash asap.
You need to demonstrate your emulator on 27 Oct. We will tell you the venue and how to book a demo time slot in Week 12.
You need to use the results of Labs 3, 4 and 5. Please finish Labs 4 and 5 asap so that you can start working on the project.
Happy coding!
Hui
Dear All
Since the last three labs provide basic building blocks for the project, I have extended the deadline of Lab 4 to next week so that you guys can finish it.
Lab 5 is available.
Cheers,
Hui
Dear All,
To make Task C easier, I have changed the data width requirement of 4 bytes for intermediate results and final result to 2 bytes. If you have done Task C with 4 bytes, you will get 2 bonus marks.
I explain how to get decimal digits of a signed integer below.
Suppose we want to convert a 2-byte binary number x into a 5-digit decimal number. We have x=d5*10^4+d4*10^3+d3*10^2+d2*10+d1. By dividing x by 10, we get the least significant digit, which is the the remainder x/10. The quotient is x/10= d5*10^3+d4*10^2+d3*10+d2. Again, by dividing x/10 by 10, we can get d2. The same process is applied to get d3, d4, and d5..
The problem with this approach is that it requires divide-by-10 operation which is not easy to implement as AVR does not support division. An alternative is to use repeated subtractions. For a positive number, we can define four constants 10000, 1000, 100, 10, and use repeated subtractions to get d4, d3, d2, and d1 in this order.
For example, we want to convert 4238 which is stored in binary, into a decimal number. After subtracting 1000 from 4238 4 times, 4238 becomes 238, and we get the most significant digit 4. Then we keep the remainder 238 and use the similar process to get 2, 3 and 8.
As the result can be negative, we need to use signed numbers to represent the intermediate results and the final result. After we get the final result, we need to check if it is a negative number or positive number. This can be done by checking the most significant bit of the 2 byte number. If the most significant bit is 1, it is a negative number. Otherwise, it is a positive number. If it is a negative number, we need to compute its absolute value. This can be done by computing 2's complement of this number (1's complement +1).
With 2 bytes, the maximum signed number is 0x7FFF which is 32767.
Once we get each digit, convert it into ASCII code, and use the subroutine provided in lcd.asm for displaying a character to display it on the lcd.
Cheers,
Hui
Log into your CSE account and type the following command:
2121 classrun -sturec
The field Assignment1 shows your marks. The full marks is 100, which will be converted to 5 marks in the calculation of the final result. You can also see your lab marks.
If you have any questions about you assignment marks, contact Umair asap. You can find his email by clicking on "staff" in the left pane of the course homepage.
As I explained in the lecture, you need to initialize the interrupt vector table from Lab 3 onward by placing the interrupt vectors of all the interrupts your program uses in their locations in the interrupt vector table. All the code should be placed after the interrupt table.
Refer to the sample interrupt vector tables given in the sample code.
The labels on PORTL are reversed, i.e., PLi is actually PL7-i (i=0, 1, …, 7). The code keypad.asm is correct.
Dear All,
I have extended the deadline of Lab 3 by one week. The mid-semester break is 23 Sep - 2 Oct. Therefore, you have more time for Lab 4.
Cheers,
Hui
Don't forget to submit assignment by the deadline.
In the last line of Important Notice of the previous Lab 3 specification, PDX4 and PDx3 should be RDX4 and RDx3. The latest lab specification has fixed the typos.
If your computer uses 32-bit Windows, you need to change the paths of avrdude.exe and avrdude.conf in the command of download.bat. For example:
"C:\Program Files\Arduino\hardware\tools\avr\bin\avrdude.exe" -C "C:\Program Files\Arduino\hardware\tools\avr\etc\avrdude.conf" -c wiring -p m2560 -P %1 -b 115200 -U flash:w:%2:i -D
Dear Students,
Next week you will be given an AVR board to test your program on it. Due to the limited number of boards and two courses COMP2121 and COMP9032 using the boards, each group will be allocated one board only. You will need to perform the following procedure before working on your lab tasks. Also don’t forget to read “Board Introduction”.
Make sure the driver is installed in your computer
Before wiring up the board, you should connect it to the computer using the USB cable and make sure that the board is detected by the computer. You could check this in "Device Manager" (this means the Driver is installed and your board is ok. Also, check the COM port number that you need to program the board). Note that COM port number may change when you reconnect the board.
If the board is not detected, please try with other USB port in your lab PC.
If you use your own laptop you should manage to download and install the driver. The driver can be installed by downloading and installing “arduino-1.0.6-windows.exe” file for the AVR Resources tab in Comp2121 website. You can also download the latest Arduino IDE including a USB driver from the link given in the lab specification.
Wiring up the Board
Next wire up the board following the connections given in "Test Your Board" under Lab 3. Note that there are two PB0 and PB1 labels on the board. One set represents PORTB bit 0 and bit 1 and the other set represents Push Button 0 and Push Button 1. Also, make sure that all 4 jumpers are connected (if you don't have jumpers you can use wires). You can ask more wires if you need. Double check if your wiring is correct.
Testing the Board
Once the wiring is complete, program the board with the given "test.hex" and follow the test procedure (given in separate documents in the lab) and check what does not work. In order to program the board, follow the instruction in "Board Introduction" under Lab 3.
The "download.bat" script is provided for 64-bit windows if you use 32-bit windows you should change the script and fix the paths. Also, note that, in most cases, you cannot program the board when the motor is working.
Be aware that most of the boards cannot pass the following 2 tests:
1.Speaker.
2. The motor on half speed.
That is OK, and just tick fail column. All group members should write down their names and zIDs on the test sheet.
Dear All,
I just released Lab 3. You have two weeks to do it. This lab may take a significant amount of time if you still do not have a global picture of interrupts and ports. You are advised to read the lab specification and the relevant material as soon as possible, write most of the code at home. Next week, at the beginning of your lab class, everyone will be given a board.
If you find any typos/errors, please contact me asap.
Cheers,
Hui
Dear All,
Arash has created a video for implementing a C function. This video is for those who still have difficulty in implementing a C function.
You can find the link to the video under "Labs".
Cheers,
Hui
UNSW is participating in the Student Experience Survey (SES). All undergraduate and postgraduate students studying in Australia are invited to participate in this survey.
The earlier you complete the survey, the more chances you have to win the major weekly prize of $1,000 prepaid VISA gift card! There are heaps of other prizes to be won each week as well! Terms and conditions are here (
http://www.srcentre.com.au/sestcs
).
The survey is now open and can be accessed at this link:
http://www.ses.edu.au/
Login details should have been sent to your UNSW email address.
About SES
The Student Experience Survey (SES) provides current higher education students with the opportunity to talk about their experience of the institution they are enrolled in. The results of this research are used to help UNSW and the government gain insight into students' experiences, and to monitor and improve teaching and learning in Australia.
The SES is an Australian Government Department of Education and Training initiative which is carried out by The Social Research Centre (
http://www.srcentre.com.au/ses
), forming part of the suite of higher education surveys under Quality Indicators for Learning and Teaching (QILT;
https://www.qilt.edu.au/about-this-site/student-experience
). The SES takes place in August of each year.
Privacy
The Social Research Centre respects your privacy. Read their full privacy terms (
http://www.srcentre.com.au/research-participants/privacy
).
Contacts
If you have any queries, please do not hesitate to contact the Social Research Centre:
• From within Australia phone: 1800 055 818
• From outside Australia phone: +61 3 8327 1951
• Email contact: ses@srcentre.com.au
There is a typo in task B, Lab 2. a[n] should be a[n+1]. Thanks Quoc-An for pointing this typo out.
Cheers,
Hui
Dear All,
To allow most students to finish all the tasks of Lab 2, I have extended the deadline by one week, i.e., you get get your lab 2 marked next week. Note that the deadline of Lab 3 remains.
Cheers,
Hui
Dear All,
Q2 and Q6 should be helpful for Lab 2.
I have received several emails asking me to debug their code. It appears that some students did not follow the requirements given by the specification. For all the tasks except the first one, you need to create a frame for each function call. It's important that you construct a layout for all the contents stored in a frame, and understand how a frame is created and de-allocated.
It's time consuming to debug a program. The most efficient way is to check if the logic is correct. Setting a break point and checking the state at the break point is also very helpful.
I may help you debug your program only you are desperate. Please understand that if everyone asked me to do debugging, It would be a nightmare for me.
Cheers,
Hui
Dear All,
Please download Lab 2 specification again. The previous C program in Task B was not consistent with the description before the C program and there were typos in the power function (return should not be in the for loop). Sorry for the mistakes.
For Task B, You need to avoid overflow. Notice that the maximum value of an unsigned integer using two bytes is 2^{16}-1=65535, and 3^10=59049. Therefore, the return value of the power function needs two bytes only. However, sum needs three bytes to avoid overflow.
Cheers,
Hui
Dear All,
In this assignment, you will work alone and write a report about the ISA of an ARM microprocessor. The due date is 10 Sept. For the interrupt part, you need to wait for me to finish it by Week 6. You may start writing the other parts.
Cheers,
Hui
Dear All,
I have released Lab 2. The deadline is Week 6. If you find any errors/typos, please let me know.
Cheers,
Hui
The origin of the data should be set to at least 0x200. The default value for ATmega2560 is 0x200. The upper part before 0x200 is used by general registers, I/O registers and ports.
Cheers,
Hui
Dear All,
In general, You are now allowed to change you lab class. However, if you have a special reason, you must contact the course admin Arash for approval. Thanks.
Hui
After having had a flu for several days, I was completely out of condition at the beginning of today's lecture. When I talked out data memory and program memory, the correct statements should be:
1. Data memory consists of SRAM, and thus is volatile.
2. Program memory consists of FLASH, and thus can store constants and code permanently.
Sorry for the mistakes.
Hui
The code for Task B has a typo. for(i=0;i<5;i++) should be for(i=0;i<=5;i++). I have fixed the typo.
Hui
Hi Guys,
Debugging the assembly code took much more time than I expected. I have simulated all the assembly programs for the first three questions. The sample assembly programs for the first three questions should be enough for the beginners to complete Lab 1.
For students who are not familiar with Hexadecimal numbers, the following link is useful: http://www.binaryhexconverter.com/hex-to-decimal-converter .
I have to take a rest and will release the solutions to Homework One tomorrow morning.
Have a good weekend,
Hui
The course admin Arash has found two good links for testing your GCD and matrix product programs:
Testing matrix multiplication:
http://www.calcul.com/show/calculator/matrix-multiplication
Testing greatest common divisor GDC:
http://www.alcula.com/calculators/math/gcd/
These two links should be very helpful for verifying your programs.
Cheers,
Hui
Hi Guys,
In order to help you do the first lab, I designed the questions catered for it. I will release the solutions no later than tomorrow morning. Sorry for the delay. The worst flu completely disordered my schedule.
Hui
Hi Guys,
I'm still in bad condition and thus have to delay the consultation to 9:00am - 12:00pm next Tuesday.
Hui
AVR Studio is designed for Windows. However, people found ways of running it on linux. I don't have any experience using AVR Studio on linux.
Here is a good link https://www.scribd.com/document/79457367/AVR-Studio-on-Linux for using AVR Studio on linux.
Hui
Hi Guys,
I was knocked down by a severe flu and thus will not be available for consultation tomorrow. I will come to uni for consultation this Friday if I get better. I will keep you informed.
Hui
Lab 1 is available now. For all the labs, you will form a group with another student. If you have any problems in forming a group, contact a lab tutor. If your problem is still not resolved, contact the course Admin Arash. You can see the names of all the tutors by clicking on Timetable in the left pane of the course homepage.
The lab time is not enough. Please write most of your code before attending each lab.
There will be no deadline extension for each lab. If you have special reasons, contact me for approval. In case your program cannot be marked in time,contact a lab tutor, keep your program intact, and get it marked next week.
Things you need to do:
1. Download the latest AVR Studio via the link given at the beginning of Introduction to AVR Studio which is under "Labs" in the left pane of the course homepage.
2. Read Introduction to AVR Studio to understand how to use it to debug and simulate an AVR assembly problem.
3. Read the lab specification. Notice that there is an optional task. If you complete the optional task, you will get bonus marks. Do the optional task only if you finish or the compulsory tasks.
Key technical points about first lab:
1. How to store a multi-byte number or string in the data memory? You need to use either little endian order or big endian order. How a multi-byte number or a string is stored in the program memory by AVR Studio? You will know the answer if you write a sample program defining a string constant and a word constant, and view how they stored.
2. When performing an addition or a subtraction on two number with different sizes, how to sign-extend the shorter number?
3. Which branch instructions should be used for signed numbers, and which branch instructions should be used for unsigned number?
4. How a pointer register should change when an array is accessed?
Cheers,
Hui
As I explained in the lectures, my lectures are a bit out of order due to the lab schedule. I have added the lecture slides on number systems, instruction formats and addressing modes for self-study, to Week 2. I will teach number systems in Week 10. You don't need to read the part about floating point numbers. For the students who are not familiar with the number systems, you may read the part about number conversions and sign extension.
Cheers,
Hui
Homework One is available now. Please try your best to answer all the questions. I will post solutions next week.
Cheers,
Hui
So far only one third of students have responded. I need more input to make a good decision. Please make a choice. Thanks
How to do it? Click on "Activities" in the left pane, click on "Moodle or Webcms3", and make a choice.
Hui
Click on "Lecture Recordings" in the left pane, and sign on Moodle using your zPass.
Hui
I have contacted CSE Student Office about the problem. I will inform you once the problem is resolved. Apologies to Web Stream students for the delay!
Lecture recording is done automatically, and is processed by the ECHO team. After processing it, they send the lecturer an link to the recording. So, I'm quite passive.
Hui
We are in Colombo Theatre A today. It is inside Colombo House (B16) beside High Street and neighbors Goldstein Hall (D16). When you get there, go to the entrance facing Fig Tree Lane, not the one facing High Street. A detailed map is here:
http://fmtoolbox.unsw.edu.au/comms/KensingtonCampus.pdf
Cheers,
Hui
The first Lab will start in Week 3, and the lab is ElecEng233 inside EE building, right above the main entrance of EE building.
Cheers,
Hui
So far I have not received any email about the lecture recording for today's lecture. Once it is available, I will send you an email about where to watch the lecture recordings.
Cheers,
Hui
A student suggested we use Moodle for our course website. I just set up a poll. Please click on Activities in the left pane, click on "Moodle or Webcms3", and make a choice. I will move our home to Moodle if the majority of students select Moodle. Thanks.
Hui
Dear COMP2121 Students,
Welcome to COMP2121! Please read the course outline. See you next Monday.