Transcript Address Translation

Linux Operating System
許 富 皓
1
Functionality of a Stack
2
A Linux Process Layout and Stack Operations
EIP
main()
{
kernel address space
high address
:
G(1);
env, argv, argc
main
}
void G(int a)
G
stack
H
{
:
Libraries
H(3);
}
heap
void H(int c)
BSS
{
data
:
}
code
low address
3
Explanation of BOAs (1)
G’s stack frame
G(int a)
{
H(3);
add_g:
}
H( int b)
{ char c[100];
int i=0;
while((c[i++]=getch())!=EOF)
{
}
Input String: abc
}
b
return address add_g
ebp
0xabc
0xabb
0xaba
esp
address of G’s
frame point
H’s stack
frame
C[99]
c
b
a
C[0]
i
4
Address Translation
5
Address Translation
Paging
Unit
Segmentation
Unit
inside a CPU
6
Intel 80386 Data Flow
7
Segmentation
8
Translation of a Logical Address
Selector
Offset
9
Segment Selector Format
10
CPU Privilege Levels

The cs register includes a 2-bit field that
specifies the Current Privilege Level (CPL)
of the CPU.
 The
value 0 denotes the highest privilege
level, while the value 3 denotes the lowest
one.

Linux uses only levels 0 and 3, which are
respectively called Kernel Mode and User
Mode.
11
Segment Descriptors
12
Contents of GDT for Processor n
per-CPU init_tss
Linux’s GDT
Linux’s GDT
n-1
default_ldt
13
Task State Segment
In Linux, each processor has only one
TSS.
 The virtual address space corresponding
to each TSS is a small subset of the liner
address space corresponding to the
kernel data segment.

14
Task State Segment

All the TSSs are sequentially stored in the per-CPU init_tss
variable
struct tss_struct {
unsigned short back_link,__blh;
unsigned long esp0;
unsigned short ss0,__ss0h;
unsigned long esp1;
unsigned short ss1,__ss1h;
unsigned long esp2;
unsigned short ss2,__ss2h;
unsigned long __cr3, eip,eflags;
unsigned long eax,ecx,edx,ebx;
unsigned long esp, ebp, esi, edi;
unsigned short es, __esh, cs, __csh, ss, __ssh, ds, __dsh;
unsigned short fs, __fsh, gs, __gsh, ldt, __ldth;
unsigned short trace, bitmap;
unsigned long io_bitmap[IO_BITMAP_LONGS + 1];
unsigned long io_bitmap_max;
struct thread_struct *io_bitmap_owner;
unsigned long __cacheline_filler[35];
unsigned long stack[64];
};
A TSS
15
Task State Segment

The TSS descriptor for the nth CPU
 The
Base field: point to the nth component of the
per-CPU init_tss variable.
G
flag: 0
 Limit field: 0xeb (each TSS segment is 236 bytes)
 DPL: 0
16
Paging
17
Paging by 80x86 Processors
18
virtual address space
physical memory
low address
process 1
process 2
:
high address
19
I/O Port
20
I/O Ports [text book]


Each device connected to the I/O bus has its
own set of I/O addresses, which are usually
called I/O ports.
In the IBM PC architecture, the I/O address
space provides up to 65,536 8-bit I/O ports.
 Two
consecutive 8-bit ports may be regarded as a
single 16-bit port, which must start on an even
address.
 Similarly, two consecutive 16-bit ports may be
regarded as a single 32-bit port, which must start on
an address that is a multiple of 4.
21
I/O Related Instructions [text book]
Four special assembly language
instructions called in, ins, out, and
outs allow the CPU to read from and
write into an I/O port.
 While executing one of these instructions,
the CPU selects the required I/O port and
transfers the data between a CPU register
and the port.

22
I/O Shared Memory [text book]



I/O ports may also be mapped into addresses of
the physical address space.
The processor is then able to communicate with
an I/O device by issuing assembly language
instructions that operate directly on memory (for
instance, mov, and, or, and so on).
Modern hardware devices are more suited to
mapped I/O, because it is faster and can be
combined with DMA.
23
Physical Address Layout
24
Physical Addresses Used by Kernel


The Linux kernel is installed in RAM starting from
the physical address 0x00100000 --- i.e., from the
second megabyte.
Why?
Answer:
When a PC computer is turned on, before Linux is loaded
into memory and takes the control of the system,




the hardware test
hardware investigation
OS booting
and
some hardware initialization work
are performed by BIOS at real mode, which has special
memory requirements at fixed memory addresses.
25
The First Megabyte of RAM Is Not
Available for Linux Kernel


To avoid loading the kernel into groups of
noncontiguous page frames, Linux prefers to
skip the first megabyte of RAM.
However, page frames not reserved by the
PC architecture will be used by Linux to store
dynamically assigned pages.
26
The First 768 Page Frames (3 MB)
in Linux 2.6
Virtual Address
0xC0000000
0x000a0000 640 K
0x000fffff 1M
Physical Address
The symbol _text, which corresponds to physical address 0x00100000,
denotes the address of the first byte of kernel code.
The end of the kernel code is similarly identified by the symbol _etext.
Kernel data is divided into two groups: initialized and uninitialized.





The initialized data starts right after _etext and ends at _edata.
The uninitialized data follows and ends up at _end.
P.S.:


The symbols appearing in the figure are not defined in Linux source code; they
are produced while compiling the kernel.
You can find the linear address of these symbols in the file system.map,
which is created right after the kernel is compiled.
27
Address Spaces for Different
Modes



Linear addresses from 0x00000000 to 0xbfffffff
can be addressed when the process is in either User or
kernel Mode.
Linear addresses from 0xc0000000 to 0xffffffff
can be addressed only when the process is in kernel
mode.
Macro  # define PAGE_OFFSET 0xc0000000
28
Signal
29
Signals

Linux uses signals to notify processes system
events.

Each event has its own signal number,
which is usually referred to by a symbolic
constant such as SIGTERM.
30
Signal Notification

Asynchronous notifications
 For
instance, a user can send the interrupt
signal SIGINT to a foreground process by
pressing the interrupt keycode (usually
Ctrl-C) at the terminal.

Synchronous notifications
 For
instance, the kernel sends the signal
SIGSEGV to a process when it accesses a
memory location at an invalid address.
31
Processes’ Responses to Signals
 Ignore.
 Asynchronously
execute a signal
handler.
SIGKILL and SIGSTOP can not
be directly handled by a process or
ignored.
Signal
32
Kernel Default Actions to
Signals
 When
a process doesn’t define its
response to a signal, then kernel will
utilize the default action of the signal
to handle it.
 Each signal has its own kernel default
action.
33
Kernel Default Actions to
Signals
 Terminate
the process.
 Core dump and terminate the process
 Ignore
 Suspend
 Resume, if it was stopped.
34