WEBVTT

1
00:00:27.270 --> 00:00:37.290
William Cheng: This is the live lecture 11 so if you have questions about lecture 11 or about the programming assignment, please feel free to to ask me.

2
00:00:57.090 --> 00:00:57.810
Hardik Mahipal Surana: My professor

3
00:00:58.170 --> 00:01:00.300
Hardik Mahipal Surana: Bonnie, can I start asking questions.

4
00:01:00.600 --> 00:01:12.690
Hardik Mahipal Surana: No. Okay, so I had my first question is, so respect to the virtual file system and actual file system. I want to understand when this when the, when the business gets booted

5
00:01:13.830 --> 00:01:20.160
Hardik Mahipal Surana: How does the virtual file system know what on actual file systems are supported by the hardware.

6
00:01:21.390 --> 00:01:26.040
William Cheng: Yeah, so when you get to the colonel to assignment, you'll see an example how this is done.

7
00:01:32.580 --> 00:01:33.540
William Cheng: Hold on, let me see.

8
00:01:36.450 --> 00:01:38.430
William Cheng: Okay, this is one of the virtual

9
00:01:41.460 --> 00:01:48.960
William Cheng: Yeah so. So what happened is that this is the same thing with a lot of, you know, Polly morphic pointer. Remember what we see the

10
00:01:49.800 --> 00:02:00.300
William Cheng: The example for a disk driver right we have you know the Western Digital 123 the CTA 76 or something like that. So at some point we have to have this binding process right we take the

11
00:02:00.990 --> 00:02:08.250
William Cheng: We take the device driver and then we need to buy into one of those. Well, one of those objects, right. So, so part of the process over here is

12
00:02:08.850 --> 00:02:24.060
William Cheng: For the device driver is device discovery when you discover devices, say, oh, you know, this one, you know, so, so when we put our system, we just detect Western Digital 123 instead of CK 67. So at that point, we need to buying the pointer to, you know, to, to, to the array of function pointer.

13
00:02:25.260 --> 00:02:32.580
William Cheng: Okay, so as you system starts you need to identify what you have and then you find the pointer to it. And then at that point you will be able to use it.

14
00:02:33.600 --> 00:02:35.550
Hardik Mahipal Surana: So this is done as a part of the boot.

15
00:02:37.020 --> 00:02:48.510
William Cheng: Well, I mean, you know, it's really not part of the defense of how you look at it right i mean you know that the food is to get you know I guess that the way we find the routing process is the booking process. Get to the operating system.

16
00:02:49.920 --> 00:02:59.400
William Cheng: Okay, so in that case, you know, it's not part of the booth right but at the time of the operating system start, you need to decide how do you find all these pointers with the

17
00:03:00.870 --> 00:03:14.790
William Cheng: You know, so, so, so there's going to be some sort of a discovery process and we mentioned, you know, also today with the device driver. We have USB devices they are discovered as you run the operating system right so so clearly that's not done at the boot process.

18
00:03:16.230 --> 00:03:16.590
Hardik Mahipal Surana: Okay.

19
00:03:17.430 --> 00:03:22.290
William Cheng: Okay, having but but also we mentioned in the good old days in the old days of Unix. What it will do is that they will

20
00:03:22.500 --> 00:03:33.270
William Cheng: You know when you have all the device driver installed inside your operating system. When you boot the operating system as part of the boot, you need to discover all the devices. So you go to the bus and you have every device driver, try to discover itself.

21
00:03:33.900 --> 00:03:39.600
William Cheng: If the device driver can find itself. But then in that case you can you can create a pointer, you can create the object and the point of it.

22
00:03:40.530 --> 00:03:51.270
William Cheng: But, you know, since that was done a long time ago and it was turns out to be a really, really slow process. People don't do that anymore. Especially because we have dynamic Lincoln and loading. So, so, you know, so there's no need to do that anymore.

23
00:03:52.650 --> 00:03:53.010
Hardik Mahipal Surana: Okay.

24
00:03:53.610 --> 00:03:54.120
William Cheng: Yeah, so, so

25
00:03:54.750 --> 00:04:02.490
William Cheng: Introductory operating system class. We're not going to learn, you know, we were basically witness to introduce a lot of the cars that and your kernel programming assignment. If you

26
00:04:02.850 --> 00:04:12.570
William Cheng: You know, if you read your co very, very carefully, you will see how some of these things are done. Okay, but we're not going to cover every, you know, every little detail exactly how you do this.

27
00:04:13.260 --> 00:04:24.150
Hardik Mahipal Surana: Okay, Gordon and in part three of lecture 11 in slide number 22 you talk about how we actually load data from a disk, right.

28
00:04:26.850 --> 00:04:29.670
William Cheng: How to load data from a desk.

29
00:04:31.980 --> 00:04:36.510
William Cheng: Okay, so I need to. I mean, right now I only have this set of slides loaded

30
00:04:40.140 --> 00:04:54.600
Hardik Mahipal Surana: OK, OK. I think I can just ask it without the slide also. So, my only question is now, when we want to read data from it is that time does the disk controller use direct memory access to copy that into physical memory.

31
00:04:56.670 --> 00:05:04.830
William Cheng: Well, again, depends on what the device, you know, if this fast device or slow device typically that this devices. They are they are the one with the DMA

32
00:05:05.670 --> 00:05:16.260
William Cheng: Okay, so what you need to do is that you need to issue that this a command to say that go transfer data, you know, so, so I guess those are the DMA register I you need to specify what is the source address. What is the destination address

33
00:05:16.530 --> 00:05:24.330
William Cheng: And you took this controller as they go. And then what it will do is that it will start taking over the bus all by itself transmitting data using, you know, the MA.

34
00:05:26.310 --> 00:05:31.290
William Cheng: Okay, right. So, so once you start that you don't know when it's going to finish. So your kernel fair will fall asleep.

35
00:05:32.400 --> 00:05:35.460
Hardik Mahipal Surana: After the IQ, you know, give up the CPU and go to sleep.

36
00:05:36.240 --> 00:05:40.650
William Cheng: And then when the data transfers complete the this the the the

37
00:05:41.280 --> 00:05:53.130
William Cheng: That this controller is going to generate interrupt interrupt interrupt service routine yeah unblock the colonel threat and when the current author wakes up and says, Oh, the operation must be done. Let me check if everything's okay and then what we'll do is we'll go from there.

38
00:05:54.510 --> 00:06:06.360
Hardik Mahipal Surana: Okay, yeah. So my next question is, with regards to pipes. So when we actually create a pipe. Is there any entry that we create in the system file table.

39
00:06:09.720 --> 00:06:24.570
Hardik Mahipal Surana: That entry. So when we create a pipe we create to file descriptors one for eating one for writing. So in both of these cases, where do they point to in the system file table and does. Is there an actual pointer to, you know, some place in memory later. Is there like a

40
00:06:25.710 --> 00:06:26.130
Hardik Mahipal Surana: Pointer

41
00:06:26.820 --> 00:06:29.370
William Cheng: So we don't really specify a file in this case they

42
00:06:30.180 --> 00:06:40.320
William Cheng: Were again, you know. We are just giving you the concept. We're not exactly the details how to do that. I think your kernel source code. There's actually pipe inside the weeks, Colonel.

43
00:06:40.890 --> 00:06:46.800
William Cheng: I'm not sure I haven't because you know you're not, you don't need to do that part. So therefore I, you know, so they are ignored.

44
00:06:47.190 --> 00:06:55.260
William Cheng: But if you look at this, this picture over here we have an array of file descriptor. Right. And we know that every file descriptor entry. It's a pointer to a file object.

45
00:06:56.730 --> 00:06:57.330
William Cheng: Right.

46
00:06:57.450 --> 00:07:04.740
William Cheng: So therefore, this one must be appointed to a file object. And if you read from that thought you know that the file object you're going to get data from the pipe.

47
00:07:05.940 --> 00:07:06.270
Hardik Mahipal Surana: Okay.

48
00:07:06.540 --> 00:07:18.090
William Cheng: And then the next file descriptor over here. Again, this is the partner to a file object, so therefore they point to the file object over here. And if you write using that file descriptor, you're going to end up so so yeah the the pipe is a producer consumer problem, right.

49
00:07:18.810 --> 00:07:27.210
William Cheng: We saw the code for the producer consumer problem before, but now the interface is going to be a little different because you're going to use the file object to connect to the producer consumer

50
00:07:28.590 --> 00:07:29.220
Hardik Mahipal Surana: OK. OK.

51
00:07:30.270 --> 00:07:36.960
Hardik Mahipal Surana: OK, so basically you're saying that the actual implementation of where it points to where the data is temporarily stored

52
00:07:37.680 --> 00:07:48.000
Hardik Mahipal Surana: When it's redundant before it's red is sort of hidden from the user. But there is some minor point and it does point to some temporary location that the data is actually stored

53
00:07:49.590 --> 00:07:56.160
William Cheng: Inside of our system is very complicated. We only introduction and I sort of keep it a concept of, you know, the file descriptor table over here.

54
00:07:56.340 --> 00:08:04.860
William Cheng: Simply point to the fallout jag and exactly how to fall object implement all these different kinds of abstraction. Is that a colonel. Those are implementation dependent right

55
00:08:05.730 --> 00:08:12.960
William Cheng: Okay. Okay, so we don't we don't want to talk about and say that this is the only way to implement that we just did that give you the idea to say this is, you know, this is what the interface looks like.

56
00:08:13.200 --> 00:08:16.170
William Cheng: And then, you know, different operating system at different way to implement it.

57
00:08:17.070 --> 00:08:19.830
William Cheng: Okay but but but what's important is to understand the abstraction.

58
00:08:20.220 --> 00:08:32.040
William Cheng: So again, this picture over here is completely out of Colonel right the you know the user space on the left. So as far as the users concern is that there is, you know, there's something called a pipe. You can refund the pipe. You could write to the pipe using code like this.

59
00:08:33.180 --> 00:08:36.120
William Cheng: Right, the user space program only use the file abstraction.

60
00:08:36.630 --> 00:08:40.110
William Cheng: Okay, it doesn't really care how the colonel implement all these kind of stuff.

61
00:08:41.460 --> 00:08:48.480
William Cheng: Right, so over here. We said why the Colonel, you know, the way so they so the implement the pipe abstraction is to use a producer consumer problem.

62
00:08:50.100 --> 00:08:53.370
William Cheng: Okay detail is really not that important. Okay.

63
00:08:53.580 --> 00:09:01.590
Hardik Mahipal Surana: So from a high level concept I want to understand is this pipe concept of part of the virtual phone system or actual transition or

64
00:09:03.210 --> 00:09:12.330
William Cheng: Well, so at the virtual file system you present the fall abstraction nice default abstraction is, you know, the thousand array of bytes and how do you make the font bigger. How do you make

65
00:09:12.660 --> 00:09:22.530
William Cheng: It smaller and also there are all these you know system call that we just went over and let your 11. So those are all you know system call for you to manipulate this file abstraction.

66
00:09:23.130 --> 00:09:27.990
William Cheng: Okay, but we haven't really talked about how they're implementing the actual file system that will be in chapter six.

67
00:09:29.430 --> 00:09:29.760
Hardik Mahipal Surana: Okay.

68
00:09:31.920 --> 00:09:36.630
William Cheng: Right. So again, this is this is Unix and Linux other operating system. They will implement different kinds of abstraction.

69
00:09:38.790 --> 00:09:42.660
Hardik Mahipal Surana: Okay, so just one last question. Actually, with respect to the colonel assignment.

70
00:09:43.470 --> 00:09:51.360
Hardik Mahipal Surana: So as a part of our final submission. Do we have to, by default, run the tests through occasion, or do we have to

71
00:09:53.280 --> 00:09:57.120
Hardik Mahipal Surana: Probe like run the test separately and an occasion to run later on.

72
00:09:57.720 --> 00:10:04.800
William Cheng: Yes. So, so when the greater grace the grid has to follow the branding guidelines. Right. So if you don't have any way for us to run those tests.

73
00:10:05.370 --> 00:10:11.790
William Cheng: There's no way for for you to get any partial credit. Right. Okay. So, therefore, in order, so we have such an A, which is

74
00:10:12.240 --> 00:10:18.660
William Cheng: We look at the grading guidelines section A are the required care certs right so you should try to get all the, you know, all the credit there.

75
00:10:18.900 --> 00:10:26.790
William Cheng: And then part a Part B is just Keisha right, you need to run those those commands and then parse Section C is the favorite test.

76
00:10:27.510 --> 00:10:31.800
William Cheng: Right, so therefore you need to create a case shell command. I mean, again, if you want full credit

77
00:10:32.220 --> 00:10:42.750
William Cheng: Okay, if you want a full credit, you need to create a case shell command your submission. So this way, the when the greater tie and also I think in the readme file, you have to tell the greater what the title.

78
00:10:43.650 --> 00:10:44.100
Hardik Mahipal Surana: Is probably

79
00:10:45.270 --> 00:10:52.080
William Cheng: Right. So then, in this case, the greater will run a command see everything is perfect. They then you get all the they get full credit. Okay, if you

80
00:10:52.320 --> 00:10:54.330
William Cheng: cannot provide the shell command.

81
00:10:54.900 --> 00:11:00.030
William Cheng: I have no idea how how to test it. I guess if there's a way to test it. Right. You can you can do see us world to test

82
00:11:00.810 --> 00:11:08.460
William Cheng: So you can ask the greater to change the conflict. I am K. But in that case, you're gonna you're gonna lose 50% of the credit or something like that.

83
00:11:09.060 --> 00:11:15.780
William Cheng: So good. The choices. You know, some people get all the way to the case out the words so euthanasia is not working going to end up losing points.

84
00:11:16.740 --> 00:11:18.360
William Cheng: Okay, so if you all full credit

85
00:11:19.230 --> 00:11:23.460
William Cheng: You know, I guess that's what I've been saying the past few lecture is that you feel for every

86
00:11:23.910 --> 00:11:32.850
William Cheng: So for the test in Section seizure create one page document what they are to testing section D you create to work a shell command. Right. So this way to go to

87
00:11:33.120 --> 00:11:38.610
William Cheng: Remove out. You just, you just how the greater than say run this command run this command run this command and then you're going to get full credit

88
00:11:40.110 --> 00:11:48.300
Hardik Mahipal Surana: Got it. Okay, I just, I get a point, I just want to clarify whether running the tests like the the favorite test somehow test the

89
00:11:50.130 --> 00:11:54.810
Hardik Mahipal Surana: Test. I just want to understand whether the interface to actually run those tests are supposed to be through the occasion or

90
00:11:55.440 --> 00:11:56.640
William Cheng: If you are full credit. Yeah.

91
00:11:56.730 --> 00:11:57.840
Hardik Mahipal Surana: Okay, yes.

92
00:11:58.230 --> 00:12:05.700
William Cheng: So some people either cannot get cash out to work, you know, by by by the last day. Well, in that case, they have to come up with a different way to try to get as many

93
00:12:05.760 --> 00:12:07.470
William Cheng: Americans partial credit as possible.

94
00:12:09.000 --> 00:12:13.950
William Cheng: Okay, because some people when they turn on interrupt, nothing works anymore. So in that case, you cannot run case, rather than what do you do

95
00:12:15.150 --> 00:12:29.700
William Cheng: Right, so, so, so people who can't get cash out to run. They need to somehow you know either. Talk to me, and try to find a way to get as many partial credit points as possible. My recommendation is to, you know, I guess I keep saying that in class that you should add these three Keisha commands.

96
00:12:29.970 --> 00:12:32.190
Hardik Mahipal Surana: Good. Okay. Thank you so much. Yeah.

97
00:12:36.240 --> 00:12:41.340
Andrew Strimaitis: Hi Professor, I have a quick question. So slide 43

98
00:12:42.870 --> 00:12:44.430
Andrew Strimaitis: That's invoking the signal handler.

99
00:12:51.150 --> 00:12:51.690
William Cheng: Yes.

100
00:12:53.130 --> 00:12:59.310
Andrew Strimaitis: It's about the yellow box there where it says if SIG handler makes a system called and we are fine.

101
00:12:59.610 --> 00:13:07.290
Andrew Strimaitis: Yeah. Could you explain that a little bit more like I'm a little confused and what makes then the system call work.

102
00:13:08.340 --> 00:13:09.570
William Cheng: So what happened is that, you know,

103
00:13:10.590 --> 00:13:16.230
William Cheng: You know inside your user space program right when you make a system call you expect the colonel stock to be empty.

104
00:13:17.430 --> 00:13:25.110
William Cheng: Mm hmm. Right. So, so we have to make sure that the colonel stack is empty over here because inside a signal handler. You can also make a lot of system call. Right. So there's a

105
00:13:25.380 --> 00:13:29.790
William Cheng: Chapter two, there's a slide to say these are all the, the, the system call you're allowed to make

106
00:13:30.240 --> 00:13:42.210
William Cheng: Is that a signal handler and there's like 50 of them, something like that, right. So, so instead of signal. How will you make system call you're gonna you know you're going to take over the colonel stack. So if there's stuff at the bottom of the colonel stack, they will all get wiped out.

107
00:13:43.650 --> 00:13:43.920
Andrew Strimaitis: Huh.

108
00:13:44.310 --> 00:13:53.310
William Cheng: Okay. So, therefore, you know, in order for you to get ready to execute code is that a signal handler, you better know that that's what we copy the thing you know out so that we can have an empty Colonel stat.

109
00:13:57.090 --> 00:13:57.540
Andrew Strimaitis: Okay.

110
00:13:58.620 --> 00:14:03.780
William Cheng: Right, because if we leave it like this. Like before you can really like this and then you start, you know, making you know

111
00:14:05.400 --> 00:14:17.220
William Cheng: You know, execute code inside of signal handler when it comes execute you know when when it make a system call you're going to wipe out the bottom or the colonel stag over here. So once you do that. Well, later on, there's no way for you to

112
00:14:18.600 --> 00:14:20.970
William Cheng: return back to your original state.

113
00:14:24.930 --> 00:14:33.570
William Cheng: Right, because this is all about, you know, saving contacts and restore contacts. So this is the contact that you need to save in order for you to later on returned back to, yo. Yo what you were doing.

114
00:14:34.080 --> 00:14:40.740
William Cheng: In inside user space program, but they you need to hide it inside the user space stack and they don't restore it and they can go back

115
00:14:40.950 --> 00:14:43.590
Andrew Strimaitis: So then when is it not okay to make the system call

116
00:14:45.630 --> 00:14:50.010
William Cheng: So Robin is that, you know, so when you press Control. See, you're inside a kernel over here, right.

117
00:14:50.160 --> 00:14:57.690
William Cheng: So now your user space program is suspended if somehow the user space, well can now make system called then it's going to wipe out of stuff inside of kernels that

118
00:14:58.170 --> 00:15:00.510
Andrew Strimaitis: Okay, so it just erases memory.

119
00:15:01.230 --> 00:15:10.530
William Cheng: Right. Yes. Why is going to erase your contacts right the contacts is save at the bottom of the colonel set right if you erase it and they, they, you know, you lose that forever.

120
00:15:11.100 --> 00:15:12.150
Andrew Strimaitis: Okay, great.

121
00:15:12.600 --> 00:15:19.260
William Cheng: So, so the idea here is that we, we should only return back into the user space with the colonel stack is completely empty.

122
00:15:19.890 --> 00:15:21.240
William Cheng: Okay, then it will be safe.

123
00:15:22.680 --> 00:15:22.920
Andrew Strimaitis: God

124
00:15:23.580 --> 00:15:28.830
William Cheng: Okay, yeah. Because you know he's had a kernel, you don't really know what the user space program is going to do right so so

125
00:15:29.490 --> 00:15:40.260
William Cheng: When you return back to the user space robot. You don't know if the user space will guy is going to run into user space for a while before to make another system call or it's going to make a system called right away. The Colonel doesn't really know what the user space program is doing.

126
00:15:41.340 --> 00:15:49.110
William Cheng: Okay. So, therefore, you got to make sure that Colonel isn't good day before you leave the Colonel. Why did you leave the Colonel, you really don't know what these are space programs that I do.

127
00:15:50.580 --> 00:15:52.020
Andrew Strimaitis: Okay. Okay. Thank you.

128
00:16:02.130 --> 00:16:03.030
Chao Xiang: Hi Professor

129
00:16:03.300 --> 00:16:13.980
Chao Xiang: Hey, I have a question at page 50 880-858-5858

130
00:16:16.590 --> 00:16:25.230
Chao Xiang: So here it says here is to memory space point to the same fire object right but well

131
00:16:25.530 --> 00:16:28.560
William Cheng: Because otherwise. So, so this is this is a memory segment.

132
00:16:28.830 --> 00:16:31.380
William Cheng: Right, we don't have one address space.

133
00:16:31.890 --> 00:16:38.220
William Cheng: Guys, oh so inside the space with divided into memory segment. So in this case member segment is also known as as regions.

134
00:16:39.660 --> 00:16:49.350
Chao Xiang: Um, yeah. My question is how can to settlement point to wall object while often these righty, and private

135
00:16:50.940 --> 00:16:51.420
Chao Xiang: Setting

136
00:16:52.590 --> 00:17:01.590
William Cheng: Right, so, so, so we so this one is you read an execute right so if you try to write to that you're going to get a segmentation fault. Right. This one is rewrite and we're doing copy on right

137
00:17:01.980 --> 00:17:11.730
William Cheng: Okay, so they pointed this file object is when they need data from the file, they will use this fall object when they don't need a data on this fall object that they can they can get the data somewhere else.

138
00:17:14.010 --> 00:17:14.820
Chao Xiang: So this

139
00:17:15.900 --> 00:17:23.880
Chao Xiang: Program can write and read at same time. And while do the right thing to do the writing copy

140
00:17:24.420 --> 00:17:32.220
William Cheng: Yeah, so, so, so again this is this is just introductory slide. Will you getting into Colonel three things going to get a little more complicated. Okay.

141
00:17:32.640 --> 00:17:41.970
William Cheng: So what happened is that, you know, over here, there's a file right so the file can be very, very big, but we managed the file one page at a time. Right, so therefore going to divide them into pages, each one of them is four kilobytes. Yeah.

142
00:17:42.360 --> 00:17:51.420
William Cheng: So in this case, some of the, you know, some of the four kilobytes are sitting inside of inside of our system. So if you want to get it. You have to go to the desk and you have a cold transfer the data.

143
00:17:51.600 --> 00:17:57.090
William Cheng: For this into memory. Some of them, you're ready ready into memory and some of them. You already are doing copy on right

144
00:17:57.330 --> 00:18:09.180
William Cheng: Right. So in that case, there are additional data structures over here. Try to remember whether this data is on this call in memory. If it's a memory. And if you're doing copy on write, there's some other pointer over here will point it was sort of point to where the data is

145
00:18:11.370 --> 00:18:14.880
William Cheng: OK, so there's more information as we go into chapter seven.

146
00:18:16.560 --> 00:18:16.920
Be great

147
00:18:18.000 --> 00:18:27.090
William Cheng: Again, this is a very, very high level, you know, a picture to say that, you know, the these two guys pointed a file object because if you need data from the file system while you have to use the fall object.

148
00:18:28.470 --> 00:18:32.580
William Cheng: Way, but once you make a copy. Then, then you don't follow this pointer anymore. You follow some other pointer.

149
00:18:34.860 --> 00:18:35.370
Chao Xiang: You

150
00:18:38.700 --> 00:18:40.440
William Cheng: The detail is going to be in chapter seven.

151
00:18:57.210 --> 00:19:10.380
William Cheng: You know, like the kernel code. It's just that you're, you know, warm up code. One more, one warm up to write me you know you can you can make decision based on the current state of your program and will do exactly the same thing inside the crown.

152
00:19:11.430 --> 00:19:15.510
William Cheng: And these are just data structure manipulation. Right. So there's no difference between that and warm up to

153
00:20:05.310 --> 00:20:09.000
William Cheng: Anyone else has a question about the lecture or Colonel one

154
00:20:16.770 --> 00:20:18.930
Andrew Strimaitis: So for this slide.

155
00:20:20.280 --> 00:20:32.280
Andrew Strimaitis: I noticed how in some earlier slides, when we were talking about memory or stuff like in terms of stack or texts or something. We mentioned that the model in the book was flipped.

156
00:20:32.700 --> 00:20:36.480
Andrew Strimaitis: Yes, I was wondering if this is something similar, where

157
00:20:37.770 --> 00:20:40.590
Andrew Strimaitis: The way we're looking at this model is always going to look like this.

158
00:20:41.910 --> 00:20:49.710
William Cheng: Yeah, so, so, so this is a linked list right so we sort them based on the virtual address the small is on the left and then big is on the right.

159
00:20:51.030 --> 00:21:00.240
William Cheng: So in this case, it's not flipped because usually the small is on the left because of the way the number line is drawn the small number is going to be on the left and the big numbers on the right.

160
00:21:01.200 --> 00:21:04.290
William Cheng: Okay, yeah. So, so this is this is normal.

161
00:21:07.950 --> 00:21:10.350
William Cheng: So later on when you do your kernel three assignment.

162
00:21:10.890 --> 00:21:22.410
William Cheng: You, you, you also need to implement the address space. So you have to implement the virtual memory map. So you're going to end up with a link address and then again you need to sort them and then you know when you get a pace file, you're going to search for it.

163
00:21:22.800 --> 00:21:29.850
William Cheng: Now, bye bye bye bye following with by walking down the link list and should find out you know when you go to baseball which your memory segment. Does it belong to

164
00:21:31.050 --> 00:21:34.530
William Cheng: And then you can look at rewrite look at private all these things that you can decide what to do.

165
00:21:39.150 --> 00:21:42.930
William Cheng: And then if you need data from the file you will talk to the fall object, you know, etc. Yeah.

166
00:22:11.850 --> 00:22:14.100
William Cheng: Anyone else has questions.

167
00:22:41.250 --> 00:22:52.350
William Cheng: Okay, if no one has this question. I'm gonna stop the recording and then you can, you know, talk to each other and okay and then you know if you have questions, feel free to come to office hour in the afternoon.