plan9fox/sys/man/3/arch

.TH ARCH 3
.SH NAME
arch \- architecture-specific information and control
.SH SYNOPSIS
.nf
.B bind -a #P /dev
.sp 0.3v
.B /dev/archctl
.B /dev/cputype
.B /dev/ioalloc
.B /dev/iob
.B /dev/iol
.B /dev/iow
.B /dev/irqalloc
.B /dev/msr
.SH DESCRIPTION
This device presents textual information about PC hardware and allows
user-level control of the I/O ports on x86-class and DEC Alpha machines.
.PP
Reads from
.I cputype
recover the processor type and clock rate in MHz.
Reads from
.I archctl
yield at least data of this form:
.IP
.EX
cpu AMD64 2201 pge
pge on
coherence mfence
cmpswap cmpswap486
i8253set on
cache default uc
cache 0x0 1073741824 wb
cache 0x3ff00000 1048576 uc
.EE
.LP
Where
.L AMD64
is the processor type,
.L 2201
is the processor speed in MHz,
and
.L pge
is present only if the `page global extension' capability is present;
the next line reflects its setting.
.L coherence
is followed by one of
.LR mb386 ,
.LR mb586 ,
.L mfence
or
.LR nop ,
showing the form of memory barrier used by the kernel.
.L cmpswap
is followed by
.L cmpswap386
or
.LR cmpswap486 ,
reflecting the form of `compare and swap' used by the kernel.
.L i8253set
is a flag, indicating the need to explicitly set
the Intel 8253 or equivalent timer.
There may be lines starting with
.L cache
that reflect the state of memory caching via MTRRs
(memory-type region registers).
The second word on the line is
.L default
or a C-style number which is the base physical address of the region;
the third is a C-style length of the region;
and the fourth is one of
.LR uc
(for uncachable),
.LR wb
(write-back),
.LR wc
(write-combining),
.LR wp
(write-protected),
or
.LR wt
(write-through).
A region may be a subset of another region, and the smaller region
takes precedence.
This may be used to make I/O registers uncachable
in the midst of a write-combining region mostly used
for a video framebuffer, for example.
Control messages may be written to
.I archctl
and use the same syntax as the data read from
.IR archctl .
Known commands include
.LR cache ,
.LR coherence ,
.LR i8253set ,
and
.LR pge .
.
.PP
Reads from
.I ioalloc
return I/O ranges used by each device, one line
per range.
Each line contains three fields separated by white space: first address
in hexadecimal,
last address, name of device.
.PP
Reads from
.I irqalloc
return the enabled interrupts, one line per
interrupt.  Each line contains three fields separated by white space:
the trap number, the IRQ it is assigned to, and the name of
the device using it.
.PP
Reads and writes to
.IR iob ,
.IR iow ,
and
.IR iol
cause 8-bit wide, 16-bit wide, and 32-bit wide requests to
I/O ports.
The port accessed is determined by the byte offset of the
file descriptor.
.PP
Reads and writes to
.I msr
go to the P4/P6/Core/Core2/AMD64 MSRs.
.SH EXAMPLE
The following code reads from an x86 byte I/O port.
.IP
.EX
uchar
inportb(unsigned port)
{
    uchar data;

    if(iobfd == -1)
        iobfd = open("#P/iob", ORDWR);

    seek(iobfd, port, 0);
    if(read(iobfd, &data, sizeof(data)) != sizeof(data))
        sysfatal("inportb(0x%4.4ux): %r", port);
    return data;
}
.EE
.SH SOURCE
.B /sys/src/9/pc/devarch.c