tar xf disc-name/linux/ver/<file>.tar.gz
In this command,
ver
is the architecture of the operating system (for example, i386
or amd64), and
file.tar
is the name of a
.tar.gz
file for that component.
See
Software packages on the Security World installation media
for more about the
component bundles and the additional software supplied on your installation media.
5. To use an nShield module with your Linux system, you must build a kernel driver.
Entrust supplies the source to the (nfp) and a makefile for building the driver as a
loadable module.
The kernel level driver is installed as part of the
hwsp
bundle. To build the driver with
the supplied makefile, you must have the correct headers installed for the kernel that
you are running. They must be headers for the same version of the kernel and must
contain the kernel configuration options with which your kernel was built. You must
also have appropriate versions of
gcc
,
make
, and your C library’s development
package.
The configuration script looks for the kernel headers in the default directory
/lib/modules/'<uname -r>'/build/include#
. If your kernel headers are located in a
different directory, set the
KERNEL_HEADERS
environment variable so that they are in
$KERNEL_HEADERS/include/
. Historically, the headers have resided in
/usr/src/linux/include/
. If the headers for your kernel are not already installed, install
them from your Linux distribution disc, or contact your kernel supplier.
Build the driver as a loadable kernel module. When you have ensured the correct
headers are in place, perform the following steps to use the makefile:
a. Change directory to the nShield PCI driver directory by running the command:
# cd /opt/nfast/driver/
b. Configure the source by running the command:
# ./configure
c. Make the driver by running the command
# make
This produces a driver file that is automatically loaded as part of the normal
installation process.
nShield® Solo and nShield® Solo XC Installation Guide
22 of 49
