On Sun, Jul 01, 2012 at 11:03:02AM -0400, Milan Kupcevic wrote: > On 07/01/2012 07:05 AM, Touko Korpela wrote: > >> ... > >> +if test "$post_2_6_38"; then > >> + XZ="xz --check=crc32 -8" > >> +else > >> + XZ=false > >> +fi > > > > From xz(1) manual page (you can ignore DecMem): > > Preset DictSize CompCPU CompMem DecMem > > -0 256 KiB 0 3 MiB 1 MiB > > -1 1 MiB 1 9 MiB 2 MiB > > -2 2 MiB 2 17 MiB 3 MiB > > -3 4 MiB 3 32 MiB 5 MiB > > -4 4 MiB 4 48 MiB 5 MiB > > -5 8 MiB 5 94 MiB 9 MiB > > -6 8 MiB 6 94 MiB 9 MiB > > -7 16 MiB 6 186 MiB 17 MiB > > -8 32 MiB 6 370 MiB 33 MiB > > -9 64 MiB 6 674 MiB 65 MiB > > > > Better to use default setting "-6" that has lower memory requirement. > > > > Here is timing for re-compression from gzip to xz on Pegasos II with > Freescale 7447 1GhZ processor and 1Gb RAM:
Use of -8 option isn't a good thing for systems having only 512 Mib or less memory. This operation shouldn't force to use slow swap space or OOM. > > set time compressed size > -9 1m43.962s 9.5MB > -8 1m44.824s 9.5MB > -6 1m35.097s 9.6MB > -3 1m10.080s 10.6MB > -2 0m45.609s 10.7MB > -0 0m19.286s 11.1MB > > I've originally decided to go with -8, mainly because the same preset is > used in mkinitramfs form initramfs-tools package. It may be worth > hearing arguments about the timing given that all presets currently > produce vmlinuz smaller than the 12MB limit, but the compression time is > very different. Your size numbers show that -8 produces only marginally smaller images. If mkinitramfs (initramfs-tools) uses -8 for xz compression, it should be changed too. -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
