From b9945320c2a286dda4ad4c6529fec0f3f85f68a3 Mon Sep 17 00:00:00 2001 From: sobomax Date: Fri, 10 Sep 2004 22:26:31 +0000 Subject: Clarify/extend in several places and make sure that everything matches reality. --- usr.bin/mkuzip/mkuzip.8 | 48 ++++++++++++++++++++++++++++++++---------------- 1 file changed, 32 insertions(+), 16 deletions(-) diff --git a/usr.bin/mkuzip/mkuzip.8 b/usr.bin/mkuzip/mkuzip.8 index 277faf2..bc2d9c50 100644 --- a/usr.bin/mkuzip/mkuzip.8 +++ b/usr.bin/mkuzip/mkuzip.8 @@ -12,9 +12,9 @@ .Os .Sh NAME .Nm mkuzip -.Nd compress a +.Nd compress disk image for use with .Xr geom_uzip 4 -image +class .Sh SYNOPSIS .Nm .Op Fl v @@ -26,11 +26,11 @@ The .Nm utility compresses a disk image file in such a way that the .Xr geom_uzip 4 -class will be able to decompress resulting image in run-time -when it loaded into memory. This allows for significant reduction -of memory footprint of memory-based filesystem at the expense of -some CPU time required to decompress the data each time it is -read. Internally, operation is done in two phases as follows: +class will be able to decompress resulting image in run-time. +This allows for significant reduction of size of disk image at +the expense of some CPU time required to decompress the data each +time it is read. Internally, operation is done in two phases as +follows: .Bl -enum .It An @@ -38,7 +38,7 @@ An image is split into clusters and each cluster compressed using .Xr zlib 3 . .It -Resulting set of clusters along with headers allowing to +Resulting set of compressed clusters along with headers allowing to independently locate each individual cluster is written into output file. .El @@ -54,30 +54,46 @@ The default is to use the input name with the suffix Use .Ar cluster_size as the size of chunks the file being split up into. Default value -is 16K. The +is 16384 bytes. The .Ar cluster_size -should be multiple of block size of the -.Xr geom_uzip 4 -device (usually 512 bytes). +should be multiple of 512 bytes. .It Fl v Display verbose messages. .El .Sh NOTES Compression ratio largely depends on the cluster size used. For -large cluster sizes of (16K and higher) typical compression ratios +large cluster sizes (16K and higher) typical compression ratios are only 1-2% less than those achieved with the .Xr gzip 1 utlity. However, it should be kept in mind that larger cluster sizes lead to higher overhead in the .Xr geom_uzip 4 class, as the class has to decompress the whole cluster even if -literally only several bytes from that cluster have to be read. +only several bytes from that cluster have to be read. +.Pp +The +.Nm +inserts tiny shell script at the beginning of the generated image, +which makes it possible to +.Qq run +the image just like any other shell script. This script tries +to load +.Xr geom_uzip 4 +class if it is not loaded, configure image as a +.Xr md 4 +disk device using +.Xr mdconfig 8 +utility and automatically mount it using +.Xr mount_cd9660 8 +utility to the mount point provided as a first argument. .Sh SEE ALSO .Xr gzip 1 , .Xr zlib 3 , .Xr geom_uzip 4 , -.Xr boot 8 , -.Xr loader 8 +.Xr geom 4 , +.Xr md 4 , +.Xr mdconfig 8 , +.Xr mount_cd9660 8 . .Sh DIAGNOSTICS Exit status is 0 on success and >0 on error. .Sh AUTHORS -- cgit v1.1