Boot
that BIOS
by
Lee Alexander
Note:
for Compaq laptop, hit F10 repeatedly upon power ON to enter BIOS.
The
term boot comes from the early days of large mainframe computers.
To start these heavy iron machines, programs known as bootstrap
(as in pulling on a pair of boots by the attached straps) loaders
were employed. Eventually, the refined programs were stored in ROM
(Read Only Memory). As we will see later, this is not entirely factual
as to the Read Only part. This special memory is an IC (Integrated
Circuit) mounted on the motherboard, the “mainframe” of your PC.
An IC is popularly referred to as a chip and a collection
of them as a chipset . The IC involved in booting the computer
is called the BIOS (Basic Input/Output System).
The
transistors employed in the BIOS are not the same as those that
comprise your RAM (Random Access Memory). RAM transistors are FETs
(Field Effect Transistors), volatile memory elements. By that we
mean they must be constantly refreshed, electronically. When power
is turned off, their state (being a 1 or 0) is lost. The
transistors in the BIOS are CMOS (Complementary Metal Oxide Semiconductors)
elements and are nonvolatile; they retain their state without power.
The term Complementary refers to the carriers of information within
the elements. In the n-type component, the electric conductivity
is via electrons; for the p-type it is holes (you can think of a
“hole” as the absence of an electron). The n- and p- channels are
created by doping the silicon substrate with different specific
elements.
Hey,
if that is all it takes, why not make the RAM out of CMOSes? In
a word, cost. CMOS devices are more complex and require more steps
in their fabrication. They are also not as fast at switching as
FETs. Like your vehicle, it takes more to get it started than to
keep it running.
In
later PCs, the BIOS ROM is actually an EPROM (Erasable Programmable
ROM) and can be “flashed” for updating. This is a tricky procedure
with plenty of opportunity for things to go wrong. We will not delve
into this subject except to say – make a full backup of your system
if you are ever tempted to Flash the BIOS.
Here
are two more spoonfuls from our alphabet soup – POST (Power On Self
Test) and RTC (Real Time Clock). If you poke around inside your
computer you will eventually find a “coin battery” resembling an
oversized watch battery – which is exactly what it is. With a mean
time life of three years or more, it is there to keep the RTC running.
That is how your computer knows the date and time you saved a file.
Not the most accurate timepiece, it occasionally needs tweaking
– but it is smart enough to adjust to the Daylight Saving scheme.
The
history of the BIOS begins with the IBM PC. The early versions did
not even recognize the floppy, the hard drive, or the video adapter.
In 1983, with the next generation of PC from IBM, the XT, the BIOS
received a real boost in its firmware (nonvolatile software). It
could now recognize the drives and video cards, but it could still
not control them. Compaq and other OEMs (Original Equipment Manufacturers)
managed to “reverse engineer” IBM's little “black box” and created
proprietary devices. This, in turn, led to compatibility problems
with things like memory and peripherals. You had to go back to the
OEM to upgrade.
Just
a year later, in 1984, Phoenix Technologies and AMI (American Megatrends
Incorporated) came on the scene with commercially available BIOS.
Cooperation between BIOS and CPU (Central Processing Unit) developers
(such as Intel and AMD) is a big factor in the compatibility and
lower cost of today's devices. BIOS development is specific to a
system. There is no need to recognize hard drives in PDAs (Personal
Digital Assistants) or HPCs (Handheld PCs) – they don't have any!
An
OEM will approach Phoenix or AMI with the specifications for a future
PC. The latest versions of BIOS are programmed modularly, thus avoiding
“reinventing the wheel” for a new device. A stockpile of program
modules leads to a versatile system of manufacture. Including or
excluding small building blocks of code facilitates rapid development
and production.
Okay,
let's POWER UP. That first surge of electricity wakes up the CPU
and causes it to reset itself to a blank state. It then reads the
first 16 bytes of code in the BIOS. The BIOS queries the CPU for
its parameters, such as speed, cache size, etc. and contacts the
CMOS and RTC for time, system information, and tests the RAM. This
is the POST part of the boot process. Other tasks for the BIOS are
to identify the keyboard, mouse, video setup, and types of disk
drives. It may also communicate with a “secondary BIOS” which actually
controls a device. In the case of disk drives, the IDE (Integrated
Drive Electronics) interface is part of that function. Microsoft's
innovative PnP (Plug and Play) is enabled or disabled by the BIOS.
The
final step in the boot process is to load the OS (Operating System)
software. Typical sequence to search the floppy, the hard drive(s),
and any CD (Compact Disk) drives for a bootable disk or file. The
boot record is stored in the first sector of the drive and contains
two parts: the location of the root directory (a catalog of all
directories on the drive) and the OS loader. Upon completion of
the OS loading, the OS then passes control to the EU (End User,
that's you!) to get something meaningful done – or play a game.
Startup
complete, the BIOS is not allowed to loaf. All during your session
on the PC, the BIOS manages the IRQs (Interrupt Requests). An IRQ
is the link between the system software and the physical components
of your computer. Standards have been established that assign certain
IRQs to generic devices. For example IRQ 0 links to the system timer;
IRQ 1, to the keyboard; IRQ 6, to the floppy drive controller; IRQ
12, to the PS/2 mouse port; IRQ 14, to the Primary IDE channel;
and IRQ 15 to the Secondary IDE channel. Some IRQs can be shared,
such as IRQ 3 to COM ports (serial connections) 2 and 4, while IRQ
4 services COM 1 and COM 3. For such a tiny and inconspicuous chip,
it has a lot riding on it.
|