diff --git a/en/tutorials/ddwg/ddwg.sgml b/en/tutorials/ddwg/ddwg.sgml
index a9c40d5c5b..1738fcd42a 100644
--- a/en/tutorials/ddwg/ddwg.sgml
+++ b/en/tutorials/ddwg/ddwg.sgml
@@ -1,1128 +1,1128 @@
FreeBSD Device Driver Writer's Guide
Eric L. Hernes, Wednesday, May 29, 1996
This document describes how to add a device driver to FreeBSD. It is
Overview
The FreeBSD kernel is very well documented, unfortunately it's all
in `C'.
Types of drivers.
Character
Data Structures
Entry Points
d_open()
d_open() takes several arguments, the formal list looks something like:
int
d_open(dev_t dev, int flag, int mode, struct proc *p)
d_open() is called on
The
The
The d_close()
d_close() takes the same argument list as d_open():
int
d_close(dev_t dev , int flag , int mode , struct proc *p)
d_close() is only called on the last close of your device (per minor
device). For example in the following code fragment, d_open() is called
3 times, but d_close() is called only once.
...
fd1=open("/dev/mydev", O_RDONLY);
fd2=open("/dev/mydev", O_RDONLY);
fd3=open("/dev/mydev", O_RDONLY);
...
...
close(fd1);
close(fd2);
close(fd3);
...
The arguments are similar to those described above for
d_open().
d_read() and d_write()
d_read() and d_write take the following argument lists:
int
d_read(dev_t dev, struct uio *uio, int flat)
int
d_write(dev_t dev, struct uio *uio, int flat)
The d_read() and d_write() entry points are called when read(2) and
write(2) are called on your device from user-space. The transfer
of data can be handled through the kernel support routine uiomove().
d_ioctl()
It's argument list is as follows:
int
d_ioctl(dev_t dev, int cmd, caddr_t arg, int flag, struct proc *p)
d_ioctl() is a catch-all for operations which don't make sense in
a read/write paradigm. Probably the most famous of all ioctl's is on
tty devices, through stty(1). The ioctl entry point is called from
ioctl() in sys/kern/sys_generic.c
There are four different types of ioctl's which can be implemented.
<sys/ioccom.h> contains convenience macros for defining these ioctls.
d_stop()
d_reset()
d_devtotty()
d_select()
d_mmap()
d_strategy()
d_strategy()'s argument list is as follows:
void
d_strategy(struct buf *bp)
d_strategy() is used for devices which use some form of scatter-gather
io. It is most common in a block device. This is significantly different
than the System V model, where only the block driver performs scatter-gather
io. Under BSD, character devices are sometimes requested to perform
scatter-gather io via the readv() and writev() system calls.
Header Files
Block
Data Structures
Entry Points
d_open()
Described in the Character device section.
d_close()
Described in the Character device section.
d_strategy()
Described in the Character device section.
d_ioctl()
Described in the Character device section.
d_dump()
d_psize()
Header Files
Network
Data Structures
Entry Points
if_init()
if_output()
if_start()
if_done()
if_ioctl()
if_watchdog()
Header Files
Line Discipline
Data Structures
Entry Points
l_open()
l_close()
l_read()
l_write()
l_ioctl()
l_rint()
l_start()
l_modem()
Header Files
Supported Busses
ISA -- Industry Standard Architecture
Data Structures
This structure is required, but generally it is created by config(8)
from the kernel configuration file. It is required on a per-device
basis, meaning that if you have a driver which controls two serial
boards, you will have two isa_device structures. If you build a
device as an LKM, you must create your own isa_device structure to
reflect your configuration. (lines 85 - 131 in pcaudio_lkm.c) There is
nearly a direct mapping between the config file and the isa_device
structure. The definition from /usr/src/sys/i386/isa/isa_device.h is:
struct isa_device {
int id_id; /* device id */
struct isa_driver *id_driver;
int id_iobase; /* base i/o address */
u_short id_irq; /* interrupt request */
short id_drq; /* DMA request */
caddr_t id_maddr; /* physical i/o memory address on bus (if any)*/
int id_msize; /* size of i/o memory */
inthand2_t *id_intr; /* interrupt interface routine */
int id_unit; /* unit number */
int id_flags; /* flags */
int id_scsiid; /* scsi id if needed */
int id_alive; /* device is present */
#define RI_FAST 1 /* fast interrupt handler */
u_int id_ri_flags; /* flags for register_intr() */
int id_reconfig; /* hot eject device support (such as PCMCIA) */
int id_enabled; /* is device enabled */
int id_conflicts; /* we're allowed to conflict with things */
struct isa_device *id_next; /* used in isa_devlist in userconfig() */
};
This structure is defined in ``/usr/src/sys/i386/isa/isa_device.h''.
These are required on a per-driver basis. The definition is:
struct isa_driver {
int (*probe) __P((struct isa_device *idp));
/* test whether device is present */
int (*attach) __P((struct isa_device *idp));
/* setup driver for a device */
char *name; /* device name */
int sensitive_hw; /* true if other probes confuse us */
};
This is the structure used by the probe/attach code to detect and
initialize your device. The
struct isa_driver mcddriver = { mcd_probe, mcd_attach, "mcd" };
Entry Points
probe()
probe() takes a attach()
attach() also takes a Header Files
EISA -- Extended Industry Standard Architecture
Data Structures
Entry Points
probe()
Described in the ISA device section.
attach()
Described in the ISA device section.
Header Files
PCI -- Peripheral Computer Interconnect
Data Structures
Entry Points
probe()
attach()
shutdown()
Header Files
SCSI -- Small Computer Systems Interface
Data Structures
Entry Points
attach()
init()
Header Files
PCCARD (PCMCIA)
Data Structures
Entry Points
handler()
unload()
suspend()
init()
Header Files
a. <pccard/slot.h>
Linking Into the Kernel.
In FreeBSD, support for the ISA and EISA busses is i386 specific.
While FreeBSD itself is presently available on the i386 platform,
some effort has been made to make the PCI, PCCARD, and SCSI code
portable. The ISA and EISA specific code resides in
/usr/src/sys/i386/isa and /usr/src/sys/i386/eisa respectively.
The machine independent PCI, PCCARD, and SCSI code reside in
/usr/src/sys/{pci,pccard,scsi}. The i386 specific code for these
reside in /usr/src/sys/i386/{pci,pccard,scsi}.
In FreeBSD, a device driver can be either binary or source. There is
no ``official'' place for binary drivers to reside. BSD/OS uses
something like sys/i386/OBJ. Since most drivers are distributed
in source, the following discussion refers to a source driver.
Binary only drivers are sometimes provided by hardware vendors
who wish to maintain the source as proprietary.
A typical driver has the source code in one c-file, say dev.c. The
driver also can have some include files; devreg.h typically contains
public device register declarations, macros, and other driver
specific declarations. Some drivers call this devvar.h instead.
Some drivers, such as the dgb (for the Digiboard PC/Xe),
require microcode to be loaded onto the board. For the dgb driver
the microcode is compiled and dumped into a header file ala file2c(1).
If the driver has data structures and ioctl's which are specific to
the driver/device, and need to be accessible from user-space, they
should be put in a separate include file which will reside in
/usr/include/machine/ (some of these reside in /usr/include/sys/).
These are typically named something like ioctl_dev.h or devio.h.
If a driver is being written which, from user space is
identical to a device which already exists, care should be taken to
use the same ioctl interface and data structures. For example, from
-user space, a SCSI cdrom drive should be identical to an IDE cdrom
+user space, a SCSI CDROM drive should be identical to an IDE cdrom
drive; or a serial line on an intelligent multiport card (Digiboard,
Cyclades, ...) should be identical to the sio devices. These devices
have a fairly well defined interface which should be used.
There are two methods for linking a driver into the kernel, static and
the LKM model. The first method is fairly standard across the
*BSD family. The other method was originally developed by Sun
(I believe), and has been implemented into BSD using the Sun model.
I don't believe that the current implementation uses any Sun code.
Standard Model
The steps required to add your driver to the standard FreeBSD kernel are
- Add to the driver list
- Add an entry to the [bc]devsw
- Add the driver entry to the kernel config file
- config(8), compile, and install the kernel
- make required nodes.
- reboot.
Adding to the driver list.
The standard model for adding a device driver to the Berkeley kernel
is to add your driver to the list of known devices. This list is
dependant on the cpu architecture. If the device is not i386 specific
(PCCARD, PCI, SCSI), the file is in ``/usr/src/sys/conf/files''.
If the device is i386 specific, use ``/usr/src/sys/i386/conf/files.i386''.
A typical line looks like:
i386/isa/joy.c optional joy device-driver
The first field is the pathname of the driver module relative to
/usr/src/sys. For the case of a binary driver the path would be
something like ``i386/OBJ/joy.o''.
The second field tells config(8) that this is an optional driver. Some
devices are required for the kernel to even be built.
The third field is the name of the device.
The fourth field tells config that it's a device driver (as opposed to
just optional). This causes config to create entries for the device
in some structures in /usr/src/sys/compile/KERNEL/ioconf.c.
It is also possible to create a file
``/usr/src/sys/i386/conf/files.KERNEL'' whose contents will override
the default files.i386, but only for the kernel ``KERNEL''.
Make room in conf.c
Now you must edit ``/usr/src/sys/i386/i386/conf.c'' to make an entry
for your driver. Somewhere near the top, you need to declare your
entry points. The entry for the joystick driver is:
#include "joy.h"
#if NJOY > 0
d_open_t joyopen;
d_close_t joyclose;
d_rdwr_t joyread;
d_ioctl_t joyioctl;
#else
#define joyopen nxopen
#define joyclose nxclose
#define joyread nxread
#define joyioctl nxioctl
#endif
This either defines your entry points, or null entry points which
will return ENXIO when called (the #else clause).
The include file ``joy.h'' is automatically generated by config(8) when
the kernel build tree is created. This usually has only one line like:
#define NJOY 1
or
#define NJOY 0
which defines the number of your devices in your kernel.
You must additionally add a slot to either cdevsw[&rsqb, or to
bdevsw[&rsqb, depending on whether it is a character device or
a block device, or both if it is a block device with a raw interface.
The entry for the joystick driver is:
/* open, close, read, write, ioctl, stop, reset, ttys, select, mmap, strat */
struct cdevsw cdevsw[] =
{
...
{ joyopen, joyclose, joyread, nowrite, /*51*/
joyioctl, nostop, nullreset, nodevtotty,/*joystick */
seltrue, nommap, NULL},
...
}
Order is what determines the major number of your device. Which is why
there will always be an entry for your driver, either null entry
points, or actual entry points. It is probably worth noting that this
is significantly different from SCO and other system V derivatives,
where any device can (in theory) have any major number. This is
largely a convenience on FreeBSD, due to the way device nodes are
created. More on this later.
Adding your device to the config file.
This is simply adding a line describing your device.
The joystick description line is:
device joy0 at isa? port "IO_GAME"
This says we have a device called ``joy0'' on the isa bus using
io-port ``IO_GAME'' (IO_GAME is a macro defined in
/usr/src/sys/i386/isa/isa.h).
A slightly more complicated entry is for the ``ix'' driver:
device ix0 at isa? port 0x300 net irq 10 iomem 0xd0000 iosiz 32768 vector ixintr
This says that we have a device called `ix0' on the ISA bus. It uses
io-port 0x300. It's interrupt will be masked with other devices in
the network class. It uses interrupt 10. It uses
32k of shared memory at physical address 0xd0000. It also defines
it's interrupt handler to be ``ixintr()''
config(8) the kernel.
Now with our config file in hand, we can create a kernel compile directory.
This is done by simply typing:
# config KERNEL
where KERNEL is the name of your config file. Config creates a
compile tree for you kernel in /usr/src/sys/compile/KERNEL. It
creates the Makefile, some .c files, and some .h files with macros
defining the number of each device in your kernel.
Now you can go to the compile directory and build. Each time you run
config, your previous build tree will be removed, unless you config
with a -n. If you have config'ed and compiled a GENERIC kernel, you can
``make links'' to avoid compiling a few files on each iteration. I typically
run
# make depend links all
followed by a ``make install'' when the kernel is done to my liking.
Making device nodes.
On FreeBSD, you are responsible for making your own device nodes. The
major number of your device is determined by the slot number in the
device switch. Minor number is driver dependent, of course. You can
either run the mknod's from the command line, or add a section to
/dev/MAKEDEV.local, or even /dev/MAKEDEV to do the work. I sometimes
create a MAKEDEV.dev script that can be run stand-alone or pasted
into /dev/MAKEDEV.local
Reboot.
This is the easy part. There are a number of ways to do this, reboot,
fastboot, shutdown -r, cycle the power, etc. Upon bootup you should
see your XXprobe() called, and if all is successful, your XXattach()
too.
Loadable Kernel Module (LKM)
There are really no defined procedures for writing an LKM driver. The
following is my own conception after experimenting with the LKM device
interface and looking at the standard device driver model, this is one
way of adding an LKM interface to an existing driver without touching
the original driver source (or binary). It is recommended though,
that if you plan to release source to your driver, the LKM specific
parts should be part of the driver itself, conditionally compiled
on the LKM macro (i.e. #ifdef LKM).
This section will focus on writing the LKM specific part of the driver. We
will assume that we have written a driver which will drop into the standard
device driver model, which we would now like to implement as an LKM. We will
use the pcaudio driver as a sample driver, and develop an LKM front-end. The
source and makefile for the pcaudio LKM, ``pcaudio_lkm.c'' and ``Makefile'',
should be placed in /usr/src/lkm/pcaudio. What follows is a breakdown of
pcaudio_lkm.c.
Lines 17 - 26
-- This includes the file ``pca.h'' and conditionally compiles the rest
of the LKM on whether or not we have a pcaudio device defined. This
mimics the behavior of config. In a standard device driver, config(8)
generates the pca.h file from the number pca devices in the config file.
17 /*
18 * figure out how many devices we have..
19 */
20
21 #include "pca.h"
22
23 /*
24 * if we have at least one ...
25 */
26 #if NPCA > 0
Lines 27 - 37
-- Includes required files from various include directories.
27 #include
28 #include
29 #include
30 #include
31 #include
32 #include
33 #include
34 #include
35 #include
36
37
Lines 38 - 51
-- Declares the device driver entry points as external.
38 /*
39 * declare your entry points as externs
40 */
41
42 extern int pcaprobe(struct isa_device *);
43 extern int pcaattach(struct isa_device *);
44 extern int pcaopen(dev_t, int, int, struct proc *);
45 extern int pcaclose(dev_t, int, int, struct proc *);
46 extern int pcawrite(dev_t, struct uio *, int);
47 extern int pcaioctl(dev_t, int, caddr_t);
48 extern int pcaselect(dev_t, int, struct proc *);
49 extern void pcaintr(struct clockframe *);
50 extern struct isa_driver pcadriver;
51
Lines 52 - 70
-- This is creates the device switch entry table for your driver.
This table gets swapped wholesale into the system device switch at
the location specified by your major number. In the standard model,
these are in /usr/src/sys/i386/i386/conf.c. NOTE: you cannot pick a
device major number higher than what exists in conf.c, for example at
present, conf.c rev 1.85, there are 67 slots for character devices,
you cannot use a (character) major device number 67 or greater,
without first reserving space in conf.c.
52 /*
53 * build your device switch entry table
54 */
55
56 static struct cdevsw pcacdevsw = {
57 (d_open_t *) pcaopen, /* open */
58 (d_close_t *) pcaclose, /* close */
59 (d_rdwr_t *) enodev, /* read */
60 (d_rdwr_t *) pcawrite, /* write */
61 (d_ioctl_t *) pcaioctl, /* ioctl */
62 (d_stop_t *) enodev, /* stop?? */
63 (d_reset_t *) enodev, /* reset */
64 (d_ttycv_t *) enodev, /* ttys */
65 (d_select_t *) pcaselect, /* select */
66 (d_mmap_t *) enodev, /* mmap */
67 (d_strategy_t *) enodev /* strategy */
68 };
69
70
Lines 71 - 131
-- This section is analogous to the config file declaration of your
device. The members of the isa_device structure are filled in by what
is known about your device, I/O port, shared memory segment, etc. We
will probably never have a need for two pcaudio devices in the kernel,
but this example shows how multiple devices can be supported.
71 /*
72 * this lkm arbitrarily supports two
73 * instantiations of the pc-audio device.
74 *
75 * this is for illustration purposes
76 * only, it doesn't make much sense
77 * to have two of these beasts...
78 */
79
80
81 /*
82 * these have a direct correlation to the
83 * config file entries...
84 */
85 struct isa_device pcadev[NPCA] = {
86 {
87 11, /* device id */
88 &pcadriver, /* driver pointer */
89 IO_TIMER1, /* base io address */
90 -1, /* interrupt */
91 -1, /* dma channel */
92 (caddr_t)-1, /* physical io memory */
93 0, /* size of io memory */
94 pcaintr , /* interrupt interface */
95 0, /* unit number */
96 0, /* flags */
97 0, /* scsi id */
98 0, /* is alive */
99 0, /* flags for register_intr */
100 0, /* hot eject device support */
101 1 /* is device enabled */
102 },
103 #if NPCA >1
104 {
105
106 /*
107 * these are all zeros, because it doesn't make
108 * much sense to be here
109 * but it may make sense for your device
110 */
111
112 0, /* device id */
113 &pcadriver, /* driver pointer */
114 0, /* base io address */
115 -1, /* interrupt */
116 -1, /* dma channel */
117 -1, /* physical io memory */
118 0, /* size of io memory */
119 NULL, /* interrupt interface */
120 1, /* unit number */
121 0, /* flags */
122 0, /* scsi id */
123 0, /* is alive */
124 0, /* flags for register_intr */
125 0, /* hot eject device support */
126 1 /* is device enabled */
127 },
128 #endif
129
130 };
131
Lines 132 - 139
-- This calls the C-preprocessor macro MOD_DEV, which sets up an LKM device
driver, as opposed to an LKM filesystem, or an LKM system call.
132 /*
133 * this macro maps to a funtion which
134 * sets the LKM up for a driver
135 * as opposed to a filesystem, systemcall, or misc
136 * LKM.
137 */
138 MOD_DEV("pcaudio_mod", LM_DT_CHAR, 24, &pcacdevsw);
139
Lines 140 - 168
-- This is the function which will be called when the driver is
loaded. This function tries to work like sys/i386/isa/isa.c
which does the probe/attach calls for a driver at boot time. The
biggest trick here is that it maps the physical address of the shared
memory segment, which is specified in the isa_device structure to a
kernel virtual address. Normally the physical address is put in the
config file which builds the isa_device structures in
/usr/src/sys/compile/KERNEL/ioconf.c. The probe/attach sequence of
/usr/src/sys/isa/isa.c translates the physical address to a virtual
one so that in your probe/attach routines you can do things like
(int *)id->id_maddr = something;
and just refer to the shared memory segment via pointers.
140 /*
141 * this function is called when the module is
142 * loaded; it tries to mimic the behavior
143 * of the standard probe/attach stuff from
144 * isa.c
145 */
146 int
147 pcaload(){
148 int i;
149 uprintf("PC Audio Driver Loaded\n");
150 for (i=0; i
Lines 169 - 179
-- This is the function called when your driver is unloaded; it just displays
a message to that effect.
169 /*
170 * this function is called
171 * when the module is unloaded
172 */
173
174 int
175 pcaunload(){
176 uprintf("PC Audio Driver Unloaded\n");
177 return 0;
178 }
179
Lines 180 - 190
-- This is the entry point which is specified on the command line of the
modload. By convention it is named <dev>_mod. This is how it is
defined in bsd.lkm.mk, the makefile which builds the LKM. If you name your
module following this convention, you can do ``make load'' and ``make
unload'' from /usr/src/lkm/pcaudio.
Note: this has gone through
180 /*
181 * this is the entry point specified
182 * on the modload command line
183 */
184
185 int
186 pcaudio_mod(struct lkm_table *lkmtp, int cmd, int ver)
187 {
188 DISPATCH(lkmtp, cmd, ver, pcaload, pcaunload, nosys);
189 }
190
191 #endif /* NICP > 0 */
Device Type Idiosyncrasies
Character
Block
Network
Line Discipline
Bus Type Idiosyncrasies
ISA
EISA
PCI
SCSI
PCCARD
Kernel Support
Data Structures
This structure contains some information about the state of the device
and driver. It is defined in /usr/src/sys/sys/devconf.h as:
struct devconf {
char dc_name[MAXDEVNAME]; /* name */
char dc_descr[MAXDEVDESCR]; /* description */
int dc_unit; /* unit number */
int dc_number; /* unique id */
char dc_pname[MAXDEVNAME]; /* name of the parent device */
int dc_punit; /* unit number of the parent */
int dc_pnumber; /* unique id of the parent */
struct machdep_devconf dc_md; /* machine-dependent stuff */
enum dc_state dc_state; /* state of the device (see above) */
enum dc_class dc_class; /* type of device (see above) */
size_t dc_datalen; /* length of data */
char dc_data[1]; /* variable-length data */
};
This structure contains all the information about a process.
It is defined in /usr/src/sys/sys/proc.h:
/*
* Description of a process.
*
* This structure contains the information needed to manage a thread of
* control, known in UN*X as a process; it has references to substructures
* containing descriptions of things that the process uses, but may share
* with related processes. The process structure and the substructures
* are always addressable except for those marked "(PROC ONLY)" below,
* which might be addressable only on a processor on which the process
* is running.
*/
struct proc {
struct proc *p_forw; /* Doubly-linked run/sleep queue. */
struct proc *p_back;
struct proc *p_next; /* Linked list of active procs */
struct proc **p_prev; /* and zombies. */
/* substructures: */
struct pcred *p_cred; /* Process owner's identity. */
struct filedesc *p_fd; /* Ptr to open files structure. */
struct pstats *p_stats; /* Accounting/statistics (PROC ONLY). */ struct plimit *p_limit; /* Process limits. */
struct vmspace *p_vmspace; /* Address space. */
struct sigacts *p_sigacts; /* Signal actions, state (PROC ONLY). */
#define p_ucred p_cred->pc_ucred
#define p_rlimit p_limit->pl_rlimit
int p_flag; /* P_* flags. */
char p_stat; /* S* process status. */
char p_pad1[3];
pid_t p_pid; /* Process identifier. */
struct proc *p_hash; /* Hashed based on p_pid for kill+exit+... */
struct proc *p_pgrpnxt; /* Pointer to next process in process group. */
struct proc *p_pptr; /* Pointer to process structure of parent. */
struct proc *p_osptr; /* Pointer to older sibling processes. */
/* The following fields are all zeroed upon creation in fork. */
#define p_startzero p_ysptr
struct proc *p_ysptr; /* Pointer to younger siblings. */
struct proc *p_cptr; /* Pointer to youngest living child. */
pid_t p_oppid; /* Save parent pid during ptrace. XXX */
int p_dupfd; /* Sideways return value from fdopen. XXX */
/* scheduling */
u_int p_estcpu; /* Time averaged value of p_cpticks. */
int p_cpticks; /* Ticks of cpu time. */
fixpt_t p_pctcpu; /* %cpu for this process during p_swtime */
void *p_wchan; /* Sleep address. */
char *p_wmesg; /* Reason for sleep. */
u_int p_swtime; /* Time swapped in or out. */
u_int p_slptime; /* Time since last blocked. */
struct itimerval p_realtimer; /* Alarm timer. */
struct timeval p_rtime; /* Real time. */
u_quad_t p_uticks; /* Statclock hits in user mode. */
u_quad_t p_sticks; /* Statclock hits in system mode. */
u_quad_t p_iticks; /* Statclock hits processing intr. */
int p_traceflag; /* Kernel trace points. */
struct vnode *p_tracep; /* Trace to vnode. */
int p_siglist; /* Signals arrived but not delivered. */
struct vnode *p_textvp; /* Vnode of executable. */
char p_lock; /* Process lock (prevent swap) count. */
char p_pad2[3]; /* alignment */
/* End area that is zeroed on creation. */
#define p_endzero p_startcopy
/* The following fields are all copied upon creation in fork. */
#define p_startcopy p_sigmask
sigset_t p_sigmask; /* Current signal mask. */
sigset_t p_sigignore; /* Signals being ignored. */
sigset_t p_sigcatch; /* Signals being caught by user. */
u_char p_priority; /* Process priority. */
u_char p_usrpri; /* User-priority based on p_cpu and p_nice. */
char p_nice; /* Process "nice" value. */
char p_comm[MAXCOMLEN+1];
struct pgrp *p_pgrp; /* Pointer to process group. */
struct sysentvec *p_sysent; /* System call dispatch information. */
struct rtprio p_rtprio; /* Realtime priority. */
/* End area that is copied on creation. */
#define p_endcopy p_addr
struct user *p_addr; /* Kernel virtual addr of u-area (PROC ONLY). */
struct mdproc p_md; /* Any machine-dependent fields. */
u_short p_xstat; /* Exit status for wait; also stop signal. */
u_short p_acflag; /* Accounting flags. */
struct rusage *p_ru; /* Exit information. XXX */
};
The
/*
* The buffer header describes an I/O operation in the kernel.
*/
struct buf {
LIST_ENTRY(buf) b_hash; /* Hash chain. */
LIST_ENTRY(buf) b_vnbufs; /* Buffer's associated vnode. */
TAILQ_ENTRY(buf) b_freelist; /* Free list position if not active. */
struct buf *b_actf, **b_actb; /* Device driver queue when active. */
struct proc *b_proc; /* Associated proc; NULL if kernel. */
volatile long b_flags; /* B_* flags. */
int b_qindex; /* buffer queue index */
int b_error; /* Errno value. */
long b_bufsize; /* Allocated buffer size. */
long b_bcount; /* Valid bytes in buffer. */
long b_resid; /* Remaining I/O. */
dev_t b_dev; /* Device associated with buffer. */
struct {
caddr_t b_addr; /* Memory, superblocks, indirect etc. */
} b_un;
void *b_saveaddr; /* Original b_addr for physio. */
daddr_t b_lblkno; /* Logical block number. */
daddr_t b_blkno; /* Underlying physical block number. */
/* Function to call upon completion. */
void (*b_iodone) __P((struct buf *));
/* For nested b_iodone's. */
struct iodone_chain *b_iodone_chain;
struct vnode *b_vp; /* Device vnode. */
int b_pfcent; /* Center page when swapping cluster. */
int b_dirtyoff; /* Offset in buffer of dirty region. */
int b_dirtyend; /* Offset of end of dirty region. */
struct ucred *b_rcred; /* Read credentials reference. */
struct ucred *b_wcred; /* Write credentials reference. */
int b_validoff; /* Offset in buffer of valid region. */
int b_validend; /* Offset of end of valid region. */
daddr_t b_pblkno; /* physical block number */
caddr_t b_savekva; /* saved kva for transfer while bouncing
*/
void *b_driver1; /* for private use by the driver */
void *b_driver2; /* for private use by the driver */
void *b_spc;
struct vm_page *b_pages[(MAXPHYS + PAGE_SIZE - 1)/PAGE_SIZE];
int b_npages;
};
This structure is used for moving data between the kernel and user spaces
through read() and write() system calls. It is defined in
/usr/src/sys/sys/uio.h:
struct uio {
struct iovec *uio_iov;
int uio_iovcnt;
off_t uio_offset;
int uio_resid;
enum uio_seg uio_segflg;
enum uio_rw uio_rw;
struct proc *uio_procp;
};
Functions
lots of 'em
References.
FreeBSD Kernel Sources http://www.freebsd.org
NetBSD Kernel Sources http://www.netbsd.org
Writing Device Drivers: Tutorial and Reference;
Tim Burke, Mark A. Parenti, Al, Wojtas;
Digital Press, ISBN 1-55558-141-2.
Writing A Unix Device Driver;
Janet I. Egan, Thomas J. Teixeira;
John Wiley & Sons, ISBN 0-471-62859-X.
Writing Device Drivers for SCO Unix;
Peter Kettle;
diff --git a/en/tutorials/devel/devel.sgml b/en/tutorials/devel/devel.sgml
index e9d8fbf560..0fde023afd 100644
--- a/en/tutorials/devel/devel.sgml
+++ b/en/tutorials/devel/devel.sgml
@@ -1,1738 +1,1739 @@
+
A User's Guide to FreeBSD Programming Tools
James Raynard, 30th May 1996
This document is an introduction to using some of the programming
tools supplied with FreeBSD, although much of it will be applicable to
many other versions of Unix. It does
Introduction
FreeBSD offers an excellent development environment. Compilers for C,
C++, and Fortran and an assembler come with the basic system, not to
mention a Perl interpreter and classic Unix tools such as sed and awk.
If that isn't enough, there are many more compilers and interpreters
in the Ports collection. FreeBSD is very compatible with standards
such as POSIX and ANSI C, as well with its own BSD heritage, so it is
possible to write applications that will compile and run with little
or no modification on a wide range of platforms.
However, all this power can be rather overwhelming at first if you've
never written programs on a Unix platform before. This document aims
to help you get up and running, without getting too deeply into more
advanced topics. The intention is that this document should give you
enough of the basics to be able to make some sense of the
documentation.
Most of the document requires little or no knowledge of programming,
although it does assume a basic competence with using Unix and a
willingness to learn!
Introduction to Programming
A program is a set of instructions that tell the computer to do
various things; sometimes the instruction it has to perform depends on
what happened when it performed a previous instruction. This section
gives an overview of the two main ways in which you can give these
instructions, or ``commands'' as they're usually called. One way uses
an interpreter, the other a compiler. As human languages are too
difficult for a computer to understand in an unambiguous way, commands
are usually written in one or other languages specially designed for
the purpose.
Interpreters
With an interpreter, the language comes as an environment, where you
type in commands at a prompt and the environment executes them for
you. For more complicated programs, you can type the commands into a
file and get the interpreter to load the file and execute the commands
in it. If anything goes wrong, many interpreters will drop you into a
debugger to help you track down the problem.
The advantage of this is that you can see the results of your commands
immediately, and mistakes can be corrected readily. The biggest
disadvantage comes when you want to share your programs with
someone. They must have the same interpreter (or you must have some
way of giving it to them) and they need to understand how to use
it. Also users may not appreciate being thrown into a debugger if they
press the wrong key! From a performance point of view, interpreters
can use up a lot of memory, and generally do not generate code as
efficiently as compilers.
In my opinion, interpreted languages are the best way to start if you
haven't done any programming before. This kind of environment is
typically found with languages like Lisp, Smalltalk, Perl and
Basic. It could also be argued that the Unix shell (sh, csh) is itself
an interpreter, and many people do in fact write shell `scripts' to
help with various ``housekeeping'' tasks on their machine. Indeed,
part of the original Unix philosophy was to provide lots of small
utility programs that could be linked together in shell scripts to
perform useful tasks.
Interpreters available with FreeBSD
Here is a list of interpreters that are available as , with a brief discussion of some of the more popular
interpreted languages.
To get one of these packages, all you need to do is to click on the
hotlink for the package, then run
pkg_add
as root. Obviously, you'll need to have a fully-functional FreeBSD
2.1.0 system for the package to work!
Basic
Short for Beginner's All-purpose Symbolic Instruction Code. Developed
in the 1950s for teaching University students to program and provided
with every self-respecting personal computer in the 1980s, BASIC has
been the first programming language for many programmers. It's also
the foundation for Visual Basic.
The and the (formerly Rabbit Basic) are
available as FreeBSD
Lisp
A language that was developed in the late 1950s as an alternative to
the ``number-crunching'' languages that were popular at the time.
Instead of being based on numbers, Lisp is based on `lists'; in fact
the name is short for "List Processing". Very popular in AI
(Artificial Intelligence) circles.
Lisp is an extremely powerful and sophisticated language, but can be
rather large and unwieldy.
FreeBSD has available as a package.
Perl
Very popular with system administrators for writing scripts; also
often used on World Wide Web servers for writing CGI scripts.
Version 4, which is probably still the most widely-used version, comes
with FreeBSD; the newer
is available as a package.
Scheme
A dialect of Lisp that is rather more compact and cleaner than Common
Lisp. Popular in Universities as it's simple enough to teach to
undergraduates as a first language, while it has a high enough level
of abstraction to be used in research work.
FreeBSD has packages of the
, the
and the
.
Icon
.
Logo
.
Python
Compilers
Compilers are rather different. First of all, you write your code in a
file (or files) using an editor. You then run the compiler and see if
it accepts your program. If it didn't compile, grit your teeth and go
back to the editor; if it did compile and gave you a program, you can
run it either at a shell command prompt or in a debugger to see if it
works properly. (If you run it in the shell, you may get a core dump).
Obviously, this is not quite as direct as using an interpreter.
However it allows you to do a lot of things which are very difficult
or even impossible with an interpreter, such as writing code which
interacts with the operating system - or even writing your own
operating system! It's also useful if you need to write very efficient
code, as the compiler can take its time and optimise the code, which
wouldn't be acceptable in an interpreter. And distributing a program
written for a compiler is usually more straightforward than one
written for an interpreter - you can just give them a copy of the
executable (assuming they have the same operating system as you).
Compiled languages include Pascal, C and C++. C and C++ are rather
unforgiving languages, and best suited to more experienced
programmers; Pascal, on the other hand, was designed as an educational
language, and is quite a good language to start with. Unfortunately,
FreeBSD doesn't have any Pascal support, except for a Pascal-to-C
converter in the ports.
As the edit-compile-run-debug cycle is rather tedious when using
separate programs, many commercial compiler makers have produced
Integrated Development Environments (IDEs for short). FreeBSD doesn't
have an IDE as such; however it's possible to use Emacs for this
purpose. This is discussed under Emacs.
Compiling with cc
This section deals only with the GNU compiler for C and C++, since
that comes with the base FreeBSD system. It can be invoked by either
`cc' or `gcc'. The details of producing a program with an interpreter
vary considerably between interpreters, and are usually well covered
in the documentation and on-line help for the interpreter.
Once you've written your masterpiece, the next step is to convert it
into something that will (hopefully!) run on FreeBSD. This usually
involves several steps, each of which is done by a separate program.
- Pre-process your source code to remove comments and do other
tricks like expanding `macros' in C.
- Check the syntax of your code to see if you have obeyed the
rules of the language. If you haven't, it will complain!
- Convert the source code into assembler - this is very close to
machine code, but still understandable by humans. Allegedly. (To be
strictly accurate, cc converts the source code into its own,
machine-independent p-code instead of assembler at this stage).
- Convert the assembler into machine code - yep, we're talking
bits and bytes, ones and zeros here.
- Check that you've used things like functions and global
variables in a consistent way (eg if you've called a non-existent
function, it'll complain).
- If you're trying to produce an executable from several source
code files, work out how to fit them all together.
- Work out how to produce something that the system's run-time
loader will be able to load into memory and run.
- (Finally!) Write the executable on the file system.
The word ``compiling'' is often used to refer to just steps 1 to 4 -
the others are referred to as ``linking''. Sometimes step 1 is
referred to as ``pre-processing'' and steps 3-4 as ``assembling''.
Fortunately, almost all this detail is hidden from you, as cc is a
front end that manages calling all these programs with the right
arguments for you; simply typing
cc foobar.c
will cause foobar.c to be compiled by all the steps above. If you have
more than one file to compile, just do something like
cc foo.c bar.c
Note that the syntax checking is just that - checking the syntax. It
won't check for any logical mistakes you may have made, like putting
the program into an infinite loop, or using a bubble sort when you
meant to use a binary sort. {In case you didn't know, a
binary sort is an efficient way of sorting things into order and a
bubble sort isn't.}
There are lots and lots of options for cc, which are all in the man
page. Here are a few of the most important ones, with examples of how
to use them.
cc foobar.c executable is a.out
cc -o foobar foobar.c executable is foobar
cc -c foobar.c
This will produce an ``object file'' (not an executable) called
`foobar.o'. This can be linked together with other object files
into an executable.
cc -g foobar.c
This will produce a debug version of the program. (Note, we didn't use
the -o flag to specify the executable name, so we'll get an executable
called `a.out'. Producing a debug version called `foobar' is left as an
exercise for the reader!)
cc -O -o foobar foobar.c
This will produce an optimised version of `foobar'.
The following three flags will force cc to check that your code
complies to the relevant international standard (often referred to
as the ``ANSI'' standard, though strictly speaking it's an ISO
standard).
Without these flags, cc will allow you to use some of its
non-standard extensions to the standard. Some of these are very
useful, but will not work with other compilers - in fact, one of
the main aims of the standard is to allow people to write code
that will work with any compiler on any system. (This is known as
``portable code'').
Generally, you should try to make your code as portable as
possible, as otherwise you may have to completely re-write the
program later to get it to work somewhere else - and who knows
what you may be using in a few years time?
Example:-
cc -Wall -ansi -pedantic -o foobar foobar.c
will produce an executable `foobar' after checking foobar.c for standard
compliance.
The most common example of this is when compiling a program that
uses some of the mathematical functions in C. Unlike most other
platforms, these are in a separate library from the standard C one
and you have to tell the compiler to add it.
The rule is that if the library is called `libsomething.a', you
give cc the argument `-lsomething'. For example, the maths library
is `libm.a', so you give cc the argument `-lm'. A common
``gotcha'' with the maths library is that it has to be the last
library on the command line.
Example:-
cc -o foobar foobar.c -lm
will link the maths library functions into `foobar'.
If you're compiling C++ code, you need to add `-lg++' to the
command line argument, to link the C++ library functions.
Alternatively, you can run c++ instead of cc, which does this for
you.
Example:-
cc -o foobar foobar.cc -lg++
c++ -o foobar foobar.cc
will both produce an executable `foobar' from the C++ source file
`foobar.cc'. Note that, on Unix systems, C++ source files
traditionally end in `.C', `.cxx' or `.cc', rather than the
DOS-style `.cpp' (which was already used for something else). gcc
used to rely on this to work out what kind of compiler to use on
the source file; however, this restriction no longer applies, so
you may now call your C++ files `.cpp' with impunity!
{c++ can also be invoked as g++ on FreeBSD.}
Common cc Queries and Problems
Q. I'm trying to write a program which uses the sin() function and I get
an error like this. What does it mean?
/var/tmp/cc0143941.o: Undefined symbol `_sin' referenced from text segment
A. When using mathematical functions like sin(), you have to tell cc to link
in the maths library, like so:-
cc -o foobar foobar.c -lm
Q. All right, I wrote this simple program to practice using -lm. All
it does is raise 2.1 to the power of 6.
#include
int main() {
float f;
f = pow(2.1, 6);
printf("2.1 ^ 6 = %f\n", f);
return 0;
}
and I compiled it as
gcc temp.c -lm
like you said I should, but I get this when I run it:-
$ ./a.out
2.1 ^ 6 = 1023.000000
This is
A. When the compiler sees you call a function, it checks if it's
already seen a prototype for it. If it hasn't, it assumes the function
returns an int, which is definitely not what you want here.
Q. So how do I fix this?
A. The prototypes for the mathematical functions are in math.h. If you
include this file, the compiler will be able to find the prototype and
it'll stop doing strange things to your calculation!
#include
#include
int main() {
...
$ ./a.out
2.1 ^ 6 = 85.766121
Morale: if you're using any of the mathematical functions, always
include math.h and remember to link in the maths library.
Q. I've compiled a file called `foobar.c' and I can't find an executable
called `foobar'. Where's it gone?
A. cc will call the executable `a.out' unless you tell it differently. Use
the -o option, eg
cc -o foobar foobar.c
Q. OK, I've got an executable called `foobar', I can see it when I do
`ls', but when I type in 'foobar' at the command prompt it tells me
there's no such file. Why can't it find it?
A. Unlike DOS, Unix won't look in the current directory when it's
trying to find out which executable you want it to run, unless you
tell it to. Either type `./foobar', which means ``run the file
called `foobar' in the current directory'', or change your PATH
environment variable so that it looks something like
bin:/usr/bin:/usr/local/bin:.
(The dot at the end means ``look in the current directory if it's not in
any of the others'')
Q. I called my executable `test', but nothing happens when I run
it. What's going on?
A. Most Unix systems have a program called `test' in /usr/bin and the
shell's picking that one up before it gets to checking the current
directory. Either type
./test
or choose a better name for your program!
Q. I compiled my program and it seemed to run all right at first, then
there was an error and it said something about ``core
dumped''. What does that mean?
A. The name ``core dump'' dates back to the very early days of Unix,
when the machines used core memory for storing data. Basically, if
the program failed under certain conditions, the system would write
the contents of core memory to disk in a file called ``core'',
which the programmer could then pore over to find out what went
wrong.
Q. Fascinating stuff, but what I am supposed to do now?
A. Use gdb to analyse the core (see Debugging).
Q. When my program dumped core, it said something about a segmentation
fault. What's that?
A. This basically means that your program tried to perform some sort
of illegal operation on memory; Unix is designed to protect the
operating system and other programs from ``rogue'' programs.
Common causes for this are:-
- Trying to write to a NULL pointer, eg
char *foo = NULL;
strcpy(foo, "bang!");
- Using a pointer that hasn't been initialised, eg
char *foo;
strcpy(foo, "bang!");
The pointer will have some random value that, with luck,
will point into an area of memory that isn't available to
your program and the kernel will kill your program before
it can do any damage. If you're unlucky, it'll point
somewhere inside your own program and corrupt one of your
data structures, causing the program to fail mysteriously.
- Trying to access past the end of an array, eg
int bar[20];
bar[27] = 6;
- Trying to store something in read-only memory, eg
char *foo = "My string";
strcpy(foo, "bang!");
(Unix compilers often put string literals like ``My string'' into
read-only areas of memory).
- Doing naughty things with malloc() and free(), eg
char bar[80];
free(bar);
or
char *foo = malloc(27);
free(foo);
free(foo);
(Note making one of these mistakes will not always lead to an
error, but they are always bad practice. Some systems and
compilers are more tolerant than others, which is why programs
that ran well on one system can crash when you try them on an
another)
Q. Sometimes when I get a core dump it says ``bus error''. It says in
my Unix book that this means a hardware problem, but the computer
still seems to be working. Is this true?
A. No, fortunately not (unless of course you really do have a hardware
problem...). This is usually another way of saying that you
accessed memory in a way you shouldn't have.
Q. This dumping core business sounds as though it could be quite
useful, if I can make it happen when I want to. Can I do this, or
do I have to wait until there's an error?
A. Yes, just go to another console or xterm, do
ps
to find out the process ID of your program, and do
kill -ABRT
This is useful if your program has got stuck in an infinite loop,
for instance. (If your program traps SIGABRT, there are several
other signals which have a similar effect).
Make
What is make?
When you're working on a simple program with only one or two source
files, typing in
cc file1.c file2.c
is not too bad, but it quickly becomes very tedious when there are
several files - and it can take a while to compile, too.
One way to get around this is to use object files and only recompile
the source file if the source code has changed. So we could have
something like:-
cc file1.o file2.o ... file37.c ...
if we'd changed file37.c, but not any of the others, since the last
time we compiled.
This may speed up the compilation quite a bit, but doesn't solve the
typing problem.
Or we could write a shell script to solve the typing problem, but it
would have to re-compile everything, making it very inefficient on a
large project.
What happens if we have hundreds of source files lying about? What if
we're working in a team with other people who forget to tell us when
they've changed one of their source files that we use?
Perhaps we could put the two solutions together and write something
like a shell script that would contain some kind of magic rule saying
when a source file needs compiling. Now all we need now is a program
that can understand these rules, as it's a bit too complicated for the
shell.
This program is called
Make files are typically kept in the same directory as the source they
apply to, and can be called `makefile', `Makefile' or `MAKEFILE'. Most
programmers use the name 'Makefile', as this puts it near the top of a
directory listing, where it can easily be seen (they don't use the
`MAKEFILE' form as block capitals are often used for documentation
files like `README').
Example of using make
Here's a very simple make file:-
foo: foo.c
cc -o foo foo.c
It consists of two lines, a dependency line and a creation line.
The dependency line here consists of the name of the program (known as
``the target''), followed by a colon, then a gap, then the name of the
source file. When make reads this line, it looks to see if `foo'
exists; if it exists, it compares the time 'foo' was last modified to
the time `foo.c' was last modified. If 'foo' does not exist, or is
older than `foo.c', it then looks at the creation line to find out
what to do. In other words, this is the rule for working out when
foo.c needs to be re-compiled.
The creation line starts with a tab (press the tab key) and then the
command you would type to create `foo' if you were doing it at a
command prompt. If `foo' is out of date, or does not exist, `make'
then executes this command to create it. In other words, this is the
rule which tells make how to re-compile foo.c.
So, when you type `make', it will make sure that `foo' is up to date
with respect to your latest changes to `foo.c'. This principle can be
extended to Makefiles with hundreds of targets - in fact, on FreeBSD,
it is possible to compile the entire operating system just by typing
`make world' in the appropriate directory!
Another useful property of make files is that the targets don't have
to be programs. For instance, we could have a make file that looks
like this:-
foo: foo.c
cc -o foo foo.c
install:
cp foo /home/me
We can tell make which target we want to make by typing
make
make will then only look at that target and ignore any
others. For example, if we type `make foo' with the make file above,
make will ignore the 'install' target.
If we just type `make' on its own, make will always look at the first
target and then stop without looking at any others. So if we typed
`make' here, it will just go to the `foo' target, re-compile `foo' if
necessary, and then stop without going on to the `install' target.
Notice that the `install' target doesn't actually depend on anything!
This means that the command on the following line is always executed
when we try to make that target by typing `make install'. In this
case, it will copy `foo' into the user's home directory. This is often
used by application make files, so that the application can be
installed in the correct directory when it has been correctly
compiled.
This is a slightly confusing subject to try and explain. If you don't
quite understand how make works, the best thing to do is to write a
simple program like ``hello world'' and a make file like the one above
and experiment. Then progress to using more than one source file, or
having the source file include a header file. (The `touch' command is
very useful here - it changes the date on a file without you having to
edit it).
FreeBSD Make Files
Make files can be rather complicated to write. Fortunately, BSD-based
systems like FreeBSD come with some very powerful ones as part of the
system.
One very good example of this is the FreeBSD ports system. Here's the
essential part of a typical ports Makefile:-
MASTER_SITES= ftp://freefall.cdrom.com/pub/FreeBSD/LOCAL_PORTS/
DISTFILES= scheme-microcode+dist-7.3-freebsd.tgz
.include
Now, if we go to the directory for this port and type make, the
following happens:-
- A check is made to see if the source code for this port is
already on the system.
- If it isn't, an FTP connection to the URL in ``MASTER_SITES''
is set up to download the source.
- The checksum for the source is calculated and compared it with
one for a known, good, copy of the source. This is to make sure that
the source was not corrupted while in transit.
- Any changes required to make the source work on FreeBSD are
applied - this is known as ``patching''.
- Any special configuration needed for the source is done. (Many
Unix program distributions try to work out which version of Unix they
are being compiled on and which optional Unix features are present -
this is where they are given the information in the FreeBSD ports
scenario).
- The source code for the program is compiled. In effect, we
change to the directory where the source was unpacked and do 'make' -
the program's own make file has the necessary information to build the
program.
- We now have a compiled version of the program. If we wish, we
can test it now; when we feel confident about the program, we can type
'make install'. This will cause the program and any supporting files
it needs to be copied into the correct location; an entry is also made
into a ``package database'', so that the port can easily be
uninstalled later if we change our mind about it.
Now I think you'll agree that's rather impressive for a four line
script!
The secret lies in the last line, which tells make to look in the
system make file called `bsd.port.mk'. It's easy to overlook this
line, but this is where all the clever stuff comes from - someone has
written a make file that tells make to do all the things above (plus a
couple of other things I didn't mention, not to mention handling any
errors that may occur) and anyone can get access to that just by
putting a single line in their own make file!
If you want to have a look at these system make files, they're in
/usr/share/mk, but it's probably best to wait until you've had a bit
of practice with make files, as they are very complicated (and if you
do look at them, make sure you have a flask of strong coffee handy!)
More advanced uses of make
Make is a very powerful tool, and can do much more than the simple
example above shows. Unfortunately, there are several different
versions of make, and they all differ considerably. The best way to
learn what they can do is probably to read the documentation -
hopefully this introduction will have given you a base from which you
can do this.
The version of make that comes with FreeBSD is the Berkeley make;
there is a tutorial for it in /usr/share/doc/psd/12.make. To view it,
do
zmore paper.ascii.gz
in that directory.
Many applications in the ports use GNU make, which has a very good set
of `info' pages. If you have installed any of these ports, GNU make
will automatically have been installed as `gmake'. It's also available
as a port and package in it's own right.
To view the info pages for GNU make, you will have to edit the `dir'
file in the /usr/local/info directory to add an entry for it. This
involves adding a line like
* Make: (make). The GNU Make utility.
to the file. Once you have done this, you can type `info' and then
select make from the menu (or in Emacs, do C-h i).
Debugging
The Debugger
The debugger that comes with FreeBSD is called `gdb' (GNU debugger). You
start it up by typing
gdb
although most people prefer to run it inside Emacs. You can do this by
M-x gdb RET RET.
Using a debugger allows you to run the program under more controlled
circumstances. Typically, you can step through the program a line at a
time, inspect the value of variables, change them, tell the debugger to run
up to a certain point and then stop, and so on. You can even attach to a
program that's already running, or load a core file to investigate why the
program crashed.
It's even possible to debug the kernel, though that's a little trickier
than the user applications we'll be discussing in this section.
gdb has quite good on-line help, as well as a set of info pages, so this
section will concentrate on a few of the basic commands.
Finally, if you find its text-based command-prompt style off-putting,
there's a graphical front-end for it in the ports.
This section is intended to be an introduction to using gdb and does
not cover specialised topics such as debugging the kernel.
Running a program in the debugger
You'll need to have compiled the program with the `-g' option to get the
most out of using gdb. It will work without, but you'll only see the name
of the function you're in, instead of the source code. If you see a line
like
...(no debugging symbols found)...
when gdb starts up, you'll know that the program wasn't compiled with
the `-g' option.
At the gdb prompt, type `break main'. This will tell the debugger to
skip over the preliminary set-up code in the program and start at the
beginning of your code. Now type `run' to start the program - it will
start at the beginning of the set-up code and then get stopped by the
debugger when it calls main(). (If you've ever wondered where main()
gets called from, now you know!).
You can now step through the program, a line at a time, by pressing
`n'. If you get to a function call, you can step into it by pressing
`s'. Once you're in a function call, you can return from stepping into
a function call by pressing `f'. You can also use `up' and `down' to
take a quick look at the caller.
Here's a simple example of how to spot a mistake in a program with
gdb. This is our program (with a deliberate mistake):-
#include
int bazz(int anint);
main() {
int i;
printf("This is my program\n");
bazz(i);
return 0;
}
int bazz(int anint) {
printf("You gave me %d\n", anint);
return anint;
}
This program sets i to be 5 and passes it to a function bazz() which prints
out the number we gave it.
When we compile and run the program we get
cc -g -o temp temp.c
./temp
This is my program
anint = 4231
That wasn't what we expected! Time to see what's going on!
Current directory is ~/tmp/
GDB is free software and you are welcome to distribute copies of it
under certain conditions; type "show copying" to see the conditions.
There is absolutely no warranty for GDB; type "show warranty" for details.
GDB 4.13 (i386-unknown-freebsd),
Copyright 1994 Free Software Foundation, Inc...
(gdb) break main # Skip the set-up code
Breakpoint 1 at 0x160f: file temp.c, line 9. # gdb puts breakpoint at main()
(gdb) run # Run as far as main()
Starting program: /home/james/tmp/temp # Program starts running
Breakpoint 1, main () at temp.c:9 # gdb stops at main()
(gdb) n # Go to next line
This is my program # Program prints out "This .."
(gdb) s # step into bazz()
bazz (anint=4231) at temp.c:17 # gdb displays stack frame
Hang on a minute! How did anint get to be 4231? Didn't we set it to be 5
in main()? Let's move up to main() and have a look.
(gdb) up # Move up call stack
#1 0x1625 in main () at temp.c:11 # gdb displays stack frame
(gdb) p i # Show us the value of i
$1 = 4231 # gdb displays 4231
Oh dear! Looking at the code, we forgot to initialise i. We meant to put
...
main() {
int i;
i = 5;
printf("This is my program\n");
...
but we missed the `i=5;' line out. As we didn't initialise i, it had
whatever number happened to be in that area of memory when the program
ran, which in this case happened to be 4231.
Note that gdb displays the stack frame every time we go into or out of
a function, even if we're using `up' and `down' to move around the
call stack. This shows the name of the function and the values of its
arguments, which helps us keep track of where we are and what's going
on. (The stack is a storage area where the program stores information
about the arguments passed to functions and where to go when it
returns from a function call).
Examining a core file
A core file is basically a file which contains the complete state of
the process when it crashed. In ``the good old days'', programmers had
to print out hex listings of core files and sweat over machine code
manuals, but now life is a bit easier. Incidentally, under FreeBSD and
other 4.4BSD systems, a core file is called ``progname.core'' instead
of just core, to make it clearer which program a core file belongs to.
To examine a core file, start up gdb in the usual way. Instead of
typing `break' or `run', type
core progname.core
(if you're not in the same directory as the core file, you'll have to
do `dir /path/to/core/file' first).
You should see something like this:-
Current directory is ~/tmp/
GDB is free software and you are welcome to distribute copies of it
under certain conditions; type "show copying" to see the conditions.
There is absolutely no warranty for GDB; type "show warranty" for details.
GDB 4.13 (i386-unknown-freebsd),
Copyright 1994 Free Software Foundation, Inc...
(gdb) core a.out.core
Core was generated by `a.out'.
Program terminated with signal 11, Segmentation fault.
Cannot access memory at address 0x7020796d.
#0 0x164a in foobar (some_arg=0x5) at temp.c:17
In this case, the program was called `a.out', so the core file is
called `a.out.core'. We can see that the program crashed due to trying
to access an area in memory that was not available to it in a function
called `bazz'.
Sometimes it's useful to be able to see how a function was called, as
the problem could have occurred a long way up the call stack in a
complex program. The `bt' command causes gdb to print out a back-trace
of the call stack:-
(gdb) bt
#0 0x164a in bazz (anint=0x5) at temp.c:17
#1 0xefbfd888 in end ()
#2 0x162c in main () at temp.c:11
The end() function is called when a program crashes; in this case, the
bazz() function was called from main().
Attaching to a running program
One of the neatest features about gdb is that it can attach to a
program that's already running. (Of course, that assumes you have
sufficient permissions to do so). A common problem is when you are
stepping through a program that forks, and you want to trace the
child, but the debugger will only let you trace the parent.
What you do is start up another gdb, use `ps' to find the process ID
for the child, and do
attach
in gdb, and then debug as usual.
``That's all very well,'' you're probably thinking, ``but by the time
I've done that, the child process will be over the hill and far
away''. Fear not, gentle reader, here's how to do it (courtesy of the
gdb info pages):-
...
if ((pid = fork()) < 0) /* _Always_ check this */
error();
else if (pid == 0) { /* child */
int PauseMode = 1;
while (PauseMode)
sleep(10); /* Wait until someone attaches to us */
...
} else { /* parent */
...
Now all you have to do is attach to the child, set PauseMode to 0, and
wait for the sleep() call to return!
Using Emacs as a Development Environment
Emacs
Unfortunately, Unix systems don't come with the kind of
everything-you-ever-wanted-and-lots-more-you-didn't-in-one-gigantic-package
integrated development environments that other systems have (at least,
not unless you pay out very large sums of money). However, it is
possible to set up your own environment. It may not be as pretty, and
it may not be quite as integrated, but you can set it up the way you
want it. And it's free. And you have the source to it.
The key to it all is Emacs. Now there are some people who loathe it,
but many who love it. If you're one of the former, I'm afraid this
section will hold little of interest to you. Also, you'll need a fair
amount of memory to run it - I'd recommend 8MB in text mode and 16MB
in X as the bare minimum to get reasonable performance.
Emacs is basically a highly customisable editor - indeed, it has been
customised to the point where it's more like an operating system than
an editor! (Many developers and sysadmins do in fact spend practically
all their time working inside Emacs, leaving it only to log out).
It's impossible even to summarise everything Emacs can do here, but
here are some of the features of interest to developers:-
- Very powerful editor, allowing search-and-replace on both
strings and regular expressions (patterns), jumping to start/end of
block expression, etc, etc.
- Pull-down menus and online help.
- Language-dependent syntax highlighting and indentation.
- Completely customisable.
- You can compile and debug programs within Emacs.
- On a compilation error, you can jump to the offending line of source
code.
- Friendly-ish front-end to the `info' program used for reading GNU
hypertext documentation (including the documentation on Emacs).
- Friendly front-end to GDB, allowing you to look at the source code
as you step through your program.
- You can read Usenet news and mail while your program is compiling ;-)
And doubtless many more that I've overlooked.
Emacs can be installed on FreeBSD using .
Once it's installed, start it up and do C-h t to read an Emacs
tutorial - that means hold down the control key, press `h', let go of
the control key, and then press t. (Alternatively, you can you use
the mouse to select ``Emacs Tutorial'' from the ``Help'' menu).
Although Emacs does have menus, it's well worth learning the key
bindings, as it's much quicker when you're editing something to press
a couple of keys than to try and find the mouse and then click on the
right place. And, when you're talking to seasoned Emacs users, you'll
find they often casually throw around expressions like
M-x replace-s RET foo RET bar RET
so it's useful to know what they mean. And in any case, Emacs has far
too many useful functions for them to all fit on the menu bars.
Fortunately, it's quite easy to pick up the key-bindings, as they're
displayed next to the menu item. (My advice is to use the menu item
for, say, opening a file until you understand how it works and feel
confident with it, then try doing C-x C-f. When you're happy with
that, move on to another menu command).
If you can't remember what a particular combination of keys does,
select ``Describe Key'' from the ``Help'' menu and type it in - Emacs
will tell you what it does. You can also use the ``Command Apropos''
menu item to find out all the commands which contain a particular word
in them, with the key binding next to it.
By the way, the expression above means hold down the Meta key, press
`x', release the Meta key, type replace-s (short for ``replace-string''
- another feature of Emacs is that you can abbreviate commands), press
the return key, type foo (the string you want replaced), press the
return key, type bar (the string you want to replace ``foo'' with) and
press return again. Emacs will then do the search-and-replace
operation you've just requested.
If you're wondering what on earth the Meta key is, it's a special key
that many Unix workstations have. Unfortunately, PC's don't have one,
so it's usually the ``alt'' key (or if you're unlucky, the ``escape''
key).
Oh, and to get out of Emacs, do C-c C-x (that means hold down the
control key, press `c', press `x' and release the control key). If you
have any unsaved files open, Emacs will ask you if you want to save
them. (Ignore the bit in the documentation where it says C-z is the
usual way to leave Emacs - that leaves Emacs hanging around in the
background, and is only really useful if you're on a system which
doesn't have virtual terminals).
Configuring Emacs
Emacs does many wonderful things, some of them are built in, some of
them need to be configured.
Instead of using a proprietary macro language for configuration, Emacs
uses a version of Lisp specially adapted for editors, known as Emacs
Lisp. This can be quite useful if you want to go on and learn
something like Common Lisp, as it's considerably smaller than Common
Lisp (although still quite big!).
The best way to learn Emacs Lisp is to download the
However, there's no need to actually know any Lisp to get started with
configuring Emacs, as I've included a sample ``.emacs'' file, which
should be enough to get you started. Just copy it into your home
directory and restart Emacs if it's already running; it will read the
commands from the file and (hopefully) give you a useful basic setup.
A sample .emacs file
Unfortunately, there's far too much here to explain it in detail;
however there are one or two points worth mentioning.
- Everything beginning with a `;' is a comment and is ignored by Emacs.
- In the first line, the -*- Emacs-Lisp -*- is so that we can edit
the .emacs file itself within Emacs and get all the fancy features for
editing Emacs Lisp (Emacs usually tries to guess this based on the
filename, and may not get it right for .emacs).
- The `tab' key is bound to an indentation function in some modes, so
when you press the tab key, it will indent the current line of
code. If you want to put a tab character in whatever you're writing,
hold the control key down while you're pressing the tab key.
- This file supports syntax highlighting for C, C++, Perl, Lisp and
Scheme (by guessing the language from the filename).
- Emacs already has a pre-defined function called ``next-error''.
In a compilation output window, this allows you to move from one
compilation error to the next by doing M-n; we define a complementary
function, ``previous-error'', that allows you to go to a previous
error by doing M-p. The nicest feature of all is that C-c C-c will
open up the source file in which the error occurred and jump to the
appropriate line.
- We enable Emacs's ability to act as a server, so that if you're doing
something outside Emacs and you want to edit a file, you can just
type in
emacsclient
and then you can edit the file in your Emacs! (Many Emacs users set
their EDITOR environment to `emacsclient' so this happens every time
they need to edit a file).
;; -*-Emacs-Lisp-*-
;; This file is designed to be re-evaled; use the variable first-time
;; to avoid any problems with this.
(defvar first-time t
"Flag signifying this is the first time that .emacs has been evaled")
;; Meta
(global-set-key "\M- " 'set-mark-command)
(global-set-key "\M-\C-h" 'backward-kill-word)
(global-set-key "\M-\C-r" 'query-replace)
(global-set-key "\M-r" 'replace-string)
(global-set-key "\M-g" 'goto-line)
(global-set-key "\M-h" 'help-command)
;; Function keys
(global-set-key [f1] 'manual-entry)
(global-set-key [f2] 'info)
(global-set-key [f3] 'repeat-complex-command)
(global-set-key [f4] 'advertised-undo)
(global-set-key [f5] 'eval-current-buffer)
(global-set-key [f6] 'buffer-menu)
(global-set-key [f7] 'other-window)
(global-set-key [f8] 'find-file)
(global-set-key [f9] 'save-buffer)
(global-set-key [f10] 'next-error)
(global-set-key [f11] 'compile)
(global-set-key [f12] 'grep)
(global-set-key [C-f1] 'compile)
(global-set-key [C-f2] 'grep)
(global-set-key [C-f3] 'next-error)
(global-set-key [C-f4] 'previous-error)
(global-set-key [C-f5] 'display-faces)
(global-set-key [C-f8] 'dired)
(global-set-key [C-f10] 'kill-compilation)
;; Keypad bindings
(global-set-key [up] "\C-p")
(global-set-key [down] "\C-n")
(global-set-key [left] "\C-b")
(global-set-key [right] "\C-f")
(global-set-key [home] "\C-a")
(global-set-key [end] "\C-e")
(global-set-key [prior] "\M-v")
(global-set-key [next] "\C-v")
(global-set-key [C-up] "\M-\C-b")
(global-set-key [C-down] "\M-\C-f")
(global-set-key [C-left] "\M-b")
(global-set-key [C-right] "\M-f")
(global-set-key [C-home] "\M-<")
(global-set-key [C-end] "\M->")
(global-set-key [C-prior] "\M-<")
(global-set-key [C-next] "\M->")
;; Mouse
(global-set-key [mouse-3] 'imenu)
;; Misc
(global-set-key [C-tab] "\C-q\t") ; Control tab quotes a tab.
(setq backup-by-copying-when-mismatch t)
;; Treat 'y' or as yes, 'n' as no.
(fset 'yes-or-no-p 'y-or-n-p)
(define-key query-replace-map [return] 'act)
(define-key query-replace-map [?\C-m] 'act)
;; Load packages
(require 'desktop)
(require 'tar-mode)
;; Pretty diff mode
(autoload 'ediff-buffers "ediff" "Intelligent Emacs interface to diff" t)
(autoload 'ediff-files "ediff" "Intelligent Emacs interface to diff" t)
(autoload 'ediff-files-remote "ediff"
"Intelligent Emacs interface to diff")
(if first-time
(setq auto-mode-alist
(append '(("\\.cpp$" . c++-mode)
("\\.hpp$" . c++-mode)
("\\.lsp$" . lisp-mode)
("\\.scm$" . scheme-mode)
("\\.pl$" . perl-mode)
) auto-mode-alist)))
;; Auto font lock mode
(defvar font-lock-auto-mode-list
(list 'c-mode 'c++-mode 'c++-c-mode 'emacs-lisp-mode 'lisp-mode 'perl-mode 'scheme-mode)
"List of modes to always start in font-lock-mode")
(defvar font-lock-mode-keyword-alist
'((c++-c-mode . c-font-lock-keywords)
(perl-mode . perl-font-lock-keywords))
"Associations between modes and keywords")
(defun font-lock-auto-mode-select ()
"Automatically select font-lock-mode if the current major mode is
in font-lock-auto-mode-list"
(if (memq major-mode font-lock-auto-mode-list)
(progn
(font-lock-mode t))
)
)
(global-set-key [M-f1] 'font-lock-fontify-buffer)
;; New dabbrev stuff
;(require 'new-dabbrev)
(setq dabbrev-always-check-other-buffers t)
(setq dabbrev-abbrev-char-regexp "\\sw\\|\\s_")
(add-hook 'emacs-lisp-mode-hook
'(lambda ()
(set (make-local-variable 'dabbrev-case-fold-search) nil)
(set (make-local-variable 'dabbrev-case-replace) nil)))
(add-hook 'c-mode-hook
'(lambda ()
(set (make-local-variable 'dabbrev-case-fold-search) nil)
(set (make-local-variable 'dabbrev-case-replace) nil)))
(add-hook 'text-mode-hook
'(lambda ()
(set (make-local-variable 'dabbrev-case-fold-search) t)
(set (make-local-variable 'dabbrev-case-replace) t)))
;; C++ and C mode...
(defun my-c++-mode-hook ()
(setq tab-width 4)
(define-key c++-mode-map "\C-m" 'reindent-then-newline-and-indent)
(define-key c++-mode-map "\C-ce" 'c-comment-edit)
(setq c++-auto-hungry-initial-state 'none)
(setq c++-delete-function 'backward-delete-char)
(setq c++-tab-always-indent t)
(setq c-indent-level 4)
(setq c-continued-statement-offset 4)
(setq c++-empty-arglist-indent 4))
(defun my-c-mode-hook ()
(setq tab-width 4)
(define-key c-mode-map "\C-m" 'reindent-then-newline-and-indent)
(define-key c-mode-map "\C-ce" 'c-comment-edit)
(setq c-auto-hungry-initial-state 'none)
(setq c-delete-function 'backward-delete-char)
(setq c-tab-always-indent t)
;; BSD-ish indentation style
(setq c-indent-level 4)
(setq c-continued-statement-offset 4)
(setq c-brace-offset -4)
(setq c-argdecl-indent 0)
(setq c-label-offset -4))
;; Perl mode
(defun my-perl-mode-hook ()
(setq tab-width 4)
(define-key c++-mode-map "\C-m" 'reindent-then-newline-and-indent)
(setq perl-indent-level 4)
(setq perl-continued-statement-offset 4))
;; Scheme mode...
(defun my-scheme-mode-hook ()
(define-key scheme-mode-map "\C-m" 'reindent-then-newline-and-indent))
;; Emacs-Lisp mode...
(defun my-lisp-mode-hook ()
(define-key lisp-mode-map "\C-m" 'reindent-then-newline-and-indent)
(define-key lisp-mode-map "\C-i" 'lisp-indent-line)
(define-key lisp-mode-map "\C-j" 'eval-print-last-sexp))
;; Add all of the hooks...
(add-hook 'c++-mode-hook 'my-c++-mode-hook)
(add-hook 'c-mode-hook 'my-c-mode-hook)
(add-hook 'scheme-mode-hook 'my-scheme-mode-hook)
(add-hook 'emacs-lisp-mode-hook 'my-lisp-mode-hook)
(add-hook 'lisp-mode-hook 'my-lisp-mode-hook)
(add-hook 'perl-mode-hook 'my-perl-mode-hook)
;; Complement to next-error
(defun previous-error (n)
"Visit previous compilation error message and corresponding source code."
(interactive "p")
(next-error (- n)))
;; Misc...
(transient-mark-mode 1)
(setq mark-even-if-inactive t)
(setq visible-bell nil)
(setq next-line-add-newlines nil)
(setq compile-command "make")
(setq suggest-key-bindings nil)
(put 'eval-expression 'disabled nil)
(put 'narrow-to-region 'disabled nil)
(put 'set-goal-column 'disabled nil)
;; Elisp archive searching
(autoload 'format-lisp-code-directory "lispdir" nil t)
(autoload 'lisp-dir-apropos "lispdir" nil t)
(autoload 'lisp-dir-retrieve "lispdir" nil t)
(autoload 'lisp-dir-verify "lispdir" nil t)
;; Font lock mode
(defun my-make-face (face colour &optional bold)
"Create a face from a colour and optionally make it bold"
(make-face face)
(copy-face 'default face)
(set-face-foreground face colour)
(if bold (make-face-bold face))
)
(if (eq window-system 'x)
(progn
(my-make-face 'blue "blue")
(my-make-face 'red "red")
(my-make-face 'green "dark green")
(setq font-lock-comment-face 'blue)
(setq font-lock-string-face 'bold)
(setq font-lock-type-face 'bold)
(setq font-lock-keyword-face 'bold)
(setq font-lock-function-name-face 'red)
(setq font-lock-doc-string-face 'green)
(add-hook 'find-file-hooks 'font-lock-auto-mode-select)
(setq baud-rate 1000000)
(global-set-key "\C-cmm" 'menu-bar-mode)
(global-set-key "\C-cms" 'scroll-bar-mode)
(global-set-key [backspace] 'backward-delete-char)
; (global-set-key [delete] 'delete-char)
(standard-display-european t)
(load-library "iso-transl")))
;; X11 or PC using direct screen writes
(if window-system
(progn
;; (global-set-key [M-f1] 'hilit-repaint-command)
;; (global-set-key [M-f2] [?\C-u M-f1])
(setq hilit-mode-enable-list
'(not text-mode c-mode c++-mode emacs-lisp-mode lisp-mode
scheme-mode)
hilit-auto-highlight nil
hilit-auto-rehighlight 'visible
hilit-inhibit-hooks nil
hilit-inhibit-rebinding t)
(require 'hilit19)
(require 'paren))
(setq baud-rate 2400) ; For slow serial connections
)
;; TTY type terminal
(if (and (not window-system)
(not (equal system-type 'ms-dos)))
(progn
(if first-time
(progn
(keyboard-translate ?\C-h ?\C-?)
(keyboard-translate ?\C-? ?\C-h)))))
;; Under UNIX
(if (not (equal system-type 'ms-dos))
(progn
(if first-time
(server-start))))
;; Add any face changes here
(add-hook 'term-setup-hook 'my-term-setup-hook)
(defun my-term-setup-hook ()
(if (eq window-system 'pc)
(progn
;; (set-face-background 'default "red")
)))
;; Restore the "desktop" - do this as late as possible
(if first-time
(progn
(desktop-load-default)
(desktop-read)))
;; Indicate that this file has been read at least once
(setq first-time nil)
;; No need to debug anything now
(setq debug-on-error nil)
;; All done
(message "All done, %s%s" (user-login-name) ".")
Extending the Range of Languages Emacs Understands
Now, this is all very well if you only want to program in the
languages already catered for in the .emacs file (C, C++, Perl, Lisp
and Scheme), but what happens if a new language called "whizbang"
comes out, full of exciting features?
The first thing to do is find out if "whizbang" comes with any files
that tell Emacs about the language. These usually end in ".el", short
for "Emacs Lisp". For example, if "whizbang" is a FreeBSD port, we can
locate these files by doing
find /usr/ports/lang/whizbang -name *.el -print
-and install them by copying them into Emac's site Lisp directory. On
+and install them by copying them into the Emacs site Lisp directory. On
FreeBSD 2.1.0-RELEASE, this is /usr/local/share/emacs/site-lisp.
So for example, if the output from the find command was
/usr/ports/lang/whizbang/work/misc/whizbang.el
we would do
cp /usr/ports/lang/whizbang/work/misc/whizbang.el /usr/local/share/emacs/site-lisp
Next, we need to decide what extension whizbang source files
have. Let's say for the sake of argument that they all end in
`.wiz'. We need to add an entry to our .emacs file to make sure Emacs
will be able to use the information in whizbang.el.
Find the auto-mode-alist entry in .emacs and add a line for whizbang,
such as:-
...
("\\.lsp$" . lisp-mode)
("\\.wiz$" . whizbang-mode)
("\\.scm$" . scheme-mode)
...
This means that Emacs will automatically go into whizbang-mode when
you edit a file ending in .wiz.
Just below this, you'll find the font-lock-auto-mode-list entry. Add
whizbang-mode to it like so:-
;; Auto font lock mode
(defvar font-lock-auto-mode-list
(list 'c-mode 'c++-mode 'c++-c-mode 'emacs-lisp-mode 'whizbang-mode 'lisp-mode 'perl-mode 'scheme-mode)
"List of modes to always start in font-lock-mode")
This means that Emacs will always enable font-lock-mode (ie syntax
highlighting) when editing a .wiz file.
And that's all that's needed. If there's anything else you want done
automatically when you open up a .wiz file, you can add a
whizbang-mode hook (see my-scheme-mode-hook for a simple example that
adds auto-indent).
Further Reading
Bibliography
-
Brian Harvey and Matthew Wright
Simply Scheme
MIT 1994.
ISBN 0-262-08226-8
-
Randall Schwartz
Learning Perl
O'Reilly 1993
ISBN 1-56592-042-2
-
Patrick Henry Winston and Berthold Klaus Paul Horn
Lisp (3rd Edition)
Addison-Wesley 1989
ISBN 0-201-08319-1
-
Brian W. Kernighan and Rob Pike
The Unix Programming Environment
Prentice-Hall 1984
ISBN 0-13-937681-X
-
Brian W. Kernighan and Dennis M. Ritchie
The C Programming Language (2nd Edition)
Prentice-Hall 1988
ISBN 0-13-110362-8
-
Bjarne Stroustrup
The C++ Programming Language
Addison-Wesley 1991
ISBN 0-201-53992-6
-
W. Richard Stevens
Advanced Programming in the Unix Environment
Addison-Wesley 1992
ISBN 0-201-56317-7
-
W. Richard Stevens
Unix Network Programming
Prentice-Hall 1990
ISBN 0-13-949876-1
diff --git a/en/tutorials/disklessx/disklessx.sgml b/en/tutorials/disklessx/disklessx.sgml
index 94a8b62c87..914236c41e 100644
--- a/en/tutorials/disklessx/disklessx.sgml
+++ b/en/tutorials/disklessx/disklessx.sgml
@@ -1,264 +1,266 @@
-
+
%includes;
]>
+
+
&header;
By Jerry Kendall
With the help of some 'friends' on the FreeBSD-hackers list, I have
been able to create a diskless X terminal... The creation of the X terminal
required first creating a diskless system with minimal utilities mounted
-via NFS. These same steps were used to create 2 seperate diskless systems.
+via NFS. These same steps were used to create 2 separate diskless systems.
The first is 'altair.kcis.com'. A diskless X terminal that I run on my
old 386DX-40. It has a 340Meg hard disk but, I did not want to change it.
So, it boots from 'antares.kcis.com' across a ethernet. The second system
is a 486DX2-66. I setup a diskless FreeBSD (complete) that uses no local
disk. The server in that case is a Sun 670MP running
SunOS 4.1.3. The same setup configuration was needed for both.
NOTE: I am sure that there is stuff that needs to be added to this. Please send me any comments....
Creating the boot floppy (On the diskless system)
Since the network boot loaders will not work with some of
the TSR's and such that MS-DOS uses, it is best to create
a dedicated boot floppy OR, if you can, create an MS-DOS menu
that will (via the config.sys/autoexec.bat files) ask what
configuration to load when the system starts. The later is the
method that I use and it works great. My MS-DOS (6.x) menu is below.
---- config.sys ----
[menu]
menuitem=normal, normal
menuitem=unix, unix
[normal]
....
normal config.sys stuff
...
[unix]
----
---- autoexec.bat ----
@ECHO OFF
goto %config%
:normal
...
normal autoexec.bat stuff
...
goto end
:unix
cd \netboot
nb8390.com
:end
----
Getting the network boot programs (On the server)
Compile the 'net-boot' programs that are located in
/usr/src/sys/i386/boot/netboot. You should read the comments
at the top of the makefile. Adjust as required. !!!! make a
backup of the original in case it gets fobar'd !!! When the build
is done, there should be 2 MS-DOS executables, 'nb8390.com' and
'nb3c509.com'. One of these two programs will be what you need
to run on the diskless server. It will load the kernel from the
boot server. At this point, put both programs on the MS-DOS
boot floppy created earlier.
Determine which program to run (On the diskless system)
If you know the chipset that your ethernet adapter uses, this is
easy. If you have the NS8390 chipset, or a NS8390 based chipset,
use NB8390.COM. If you have a 3Com 509 based chipset, use the
NB3C509.COM boot program. If you are not sure which you have,
try using one, if it says 'No adapter found', try the other.
Beyond that, you are pretty much on your own.
Booting across the network
Boot the diskless system with out any config.sys/autoexec.bat
files. try running the boot program for your ethernet adapter.
My ethernet adapter is running in WD8013 16bit mode so
I run NB8390.COM
C:> cd \netboot
C:> nb8390
Boot from Network (Y/N) ? Y
BOOTP/TFTP/NFS bootstrap loader ESC for menu
Searching for adapter..
WD8013EBT base 0x0300, memory 0x000D8000, addr 00:40:01:43:26:66
Searching for server..
At this point, my diskless system is trying to find a machine to act
as a boot server. Make note of the addr line above, you will need this
number later. Reset the diskless system and modify your config.sys and
autoexec.bat files to do these steps automatically for you. Perhaps in
a menu. If you had to run 'nb3c509.com' instead of 'nb8390.com' the
output is the same as above. If you got 'No adapter found' at the
'Searching for adapter..' message, verify that you did indeed set the
compile time defines in the makefile correctly.
Allowing systems to boot across the network (On the
server)
Make sure the /etc/inetd.conf file has entries for tftp and bootps.
Mine are listed below:
---- /etc/inetd.conf ----
tftp dgram udp wait nobody /usr/libexec/tftpd tftpd
#
# Additions by who ever you are
bootps dgram udp wait root /usr/libexec/bootpd bootpd /etc/bootptab
----
If you have to change the /etc/inetd.conf file, send a HUP signal to
inetd. To do this, get the process ID of inetd with 'ps -ax | grep
inetd | grep -v grep'. Once you have it, send it a HUP signal. Do this
by 'kill -HUP <pid>'. This will force inetd to re-read its config file.
Did you remember to note the 'addr' line from the output of the boot
loader on the diskless system???? Guess what, here is where you need it.
Add an entry to /etc/bootptab (maybe creating the file). It should be
laid out identical to this:
altair:\
:ht=ether:\
:ha=004001432666:\
:sm=255.255.255.0:\
:hn:\
:ds=199.246.76.1:\
:ip=199.246.76.2:\
:gw=199.246.76.1:\
:vm=rfc1048:
The lines are as follows:
'altair' the diskless systems name without the domain name.
'ht=ether' the hardware type of 'ethernet'.
'ha=004001432666' the hardware address (the number noted above).
'sm=255.255.255.0' the subnet mask.
'hn' tells server to send client's hostname to the client.
'ds=199.246.76.1' tells the client who the domain server is.
'ip=199.246.76.2' tells the client what it's IP address is.
'gw=199.246.76.1' tells the client what the default gateway is.
'vm=...' just leave it there...
NOTE:
****** Be sure to setup the IP addresses correctly, the addresses
above are my own......
Create the directory '/tftpboot' on the server it will contain the
configuration files for the diskless systems that the server will
serve. These files will be named 'cfg.<ip>' where <ip> is the IP
address of the diskless system. The config file for 'altair' is
/tftpboot/cfg.199.246.76.2. The contents is:
---- /tftpboot/cfg.199.246.76.2 ----
rootfs 199.246.76.1:/DiskLess/rootfs/altair
hostname altair.kcis.com
----
The line 'hostname altair.kcis.com' simply tells the diskless
system what its fully qualified domain name is.
The line 'rootfs 199.246.76.1:/DiskLess/rootfs/altair' tells the
diskless system where its NFS mountable root filesystem is located.
NOTE:!!!!! The NFS mounted root filesystem will be mounted READ ONLY.
The hierarchy for the diskless system can be re-mounted allowing
read-write operations if required.
I use my spare 386DX-40 as a dedicated X terminal...
The hierarchy for 'altair' is:
/
/bin
/etc
/tmp
/sbin
/dev
/dev/fd
/usr
/var
/var/run
The actual list of files is:
-r-xr-xr-x 1 root wheel 779984 Dec 11 23:44 ./kernel
-r-xr-xr-x 1 root bin 299008 Dec 12 00:22 ./bin/sh
-rw-r--r-- 1 root wheel 499 Dec 15 15:54 ./etc/rc
-rw-r--r-- 1 root wheel 1411 Dec 11 23:19 ./etc/ttys
-rw-r--r-- 1 root wheel 157 Dec 15 15:42 ./etc/hosts
-rw-r--r-- 1 root bin 1569 Dec 15 15:26 ./etc/XF86Config.altair
-r-x------ 1 bin bin 151552 Jun 10 1995 ./sbin/init
-r-xr-xr-x 1 bin bin 176128 Jun 10 1995 ./sbin/ifconfig
-r-xr-xr-x 1 bin bin 110592 Jun 10 1995 ./sbin/mount_nfs
-r-xr-xr-x 1 bin bin 135168 Jun 10 1995 ./sbin/reboot
-r-xr-xr-x 1 root bin 73728 Dec 13 22:38 ./sbin/mount
-r-xr-xr-x 1 root wheel 1992 Jun 10 1995 ./dev/MAKEDEV.local
-r-xr-xr-x 1 root wheel 24419 Jun 10 1995 ./dev/MAKEDEV
Don't forget to 'MAKEDEV all' in the 'dev' directory.
My /etc/rc for 'altair' is:
#!/bin/sh
#
PATH=/bin:/sbin
export PATH
#
# configure the localhost
/sbin/ifconfig lo0 127.0.0.1
#
# configure the ethernet card
/sbin/ifconfig ed0 199.246.76.2 netmask 0xffffff00
#
# mount the root filesystem via NFS
/sbin/mount antares:/DiskLess/rootfs/altair /
#
# mount the /usr filesystem via NFS
/sbin/mount antares:/DiskLess/usr /usr
#
/usr/X11R6/bin/XF86_SVGA -query antares -xf86config /etc/XF86Config.altair > /dev/null 2>&1
#
# Reboot after X exits
/sbin/reboot
#
# We blew up....
exit 1
Any comments and ALL questions welcome....
Jerry Kendall
jerry@kcis.com
&footer;