Hi,
From: Karel Zak <kzak@xxxxxxxxxx>
Subject: Re: [RFC] lscpu - CPU architecture information helper
Date: Sun, 6 Jul 2008 23:44:09 +0200
>
> Hi Cai,
>
> I have a few (minor) notes:
>
> On Fri, Jul 04, 2008 at 11:23:10AM +0800, Cai Qian wrote:
> > I propose a new tool lscpu to util-linux-ng, This program gathers CPU
>
> don't forget that a man page is required for all new tools.
>
> > _______________________________________________________________
> > | | | | | | | | |
> > | cpu0 | | | | | | | |
> > |______| core0 | | L1d | L1i | L2d | | |
> > | | | | 16K | 16K | 256K | | |
> > | cpu1 | | | | | | | |
> > |______|_______| processor0 |_____|_____|______| L2i | L3 |
> > | | | | | | | 1024K | 12288K |
> > | cpu2 | | | | | | | |
> > |______| core1 | | L1d | L1i | L2d | | |
> > | | | | 16K | 16K | 256K | | |
> > | cpu3 | | | | | | | |
> > |______|_______|____________|_____|_____|______|_______|________|
> > | | | | | | | | |
> > | cpu4 | | | | | | | |
> > |______| core2 | | L1d | L1i | L2d | | |
> > | | | | 16K | 16K | 256K | | |
> > | cpu5 | | | | | | | |
> > |______|_______| processor1 |_____|_____|______| L2i | L3 |
> > | | | | | | | 1024K | 12288K |
> > | cpu6 | | | | | | | |
> > |______| core3 | | L1d | L1i | L2d | | |
> > | | | | 16K | 16K | 256K | | |
> > | cpu7 | | | | | | | |
> > |______|_______|____________|_____|_____|______|_______|________|
>
>
> That's really nice ASCII art. People love pictures :-)
>
> > usage()
> > {
> > cat <<EOF
> > Usage: $0 [OPTION]
> > CPU architecture information helper
> >
> > -p, --parse print out in parsable instead of printable format.
> > -h, --help usage information
> > EOF
> > }
>
> Please, read:
> http://www.gnu.org/software/autoconf/manual/gettext/Preparing-Shell-Scripts.html
>
> > arch=$(uname -m)
> > # Only i686, X86_64 and IA64 are supported at the moment.
>
> Do you plan to add any other architectures?
>
On 2.6.18 Kernel, ppc64 does not have a really helpful cpuinfo and
sysfs. I don't know about it on the latest Kernel version though. Also,
It does not make too much sense to support s390(x), as each guest just
have some virtual CPUs. I don't have access to machines of other
architectures though. If it is possible, I'd be happy to test it on
those architectures too.
CaiQian
> > # Dom0 Kernel gives wrong information.
> > if grep "control_d" "/proc/xen/capabilities" >/dev/null 2>&1; then
> > echo "Error: Dom0 Kernel is not supported." >&2
> > exit 1
> > fi
>
> BTW, virsh(8) (command "virtsh nodeinfo") from libvirt provides basic
> information about virtual HW.
>
> > # Read through sysfs.
> > syspath="/sys/devices/system"
>
> maybe you also want to define:
>
> cpu0path="${syspath}/cpu/cpu0"
>
> > # Read through cpuinfo.
> > vendor=$(grep -m 1 "vendor" /proc/cpuinfo | sed 's/.*: \(.*\)/\1/')
> > family=$(grep -m 1 "family" /proc/cpuinfo | sed 's/.*: \(.*\)/\1/')
> > model=$(grep -m 1 "model.[^n]" /proc/cpuinfo | sed 's/.*: \(.*\)/\1/')
> > stepping=$(grep -m 1 "stepping" /proc/cpuinfo | sed 's/.*: \(.*\)/\1/')
> > mhz=$(grep -m 1 "cpu MHz" /proc/cpuinfo | sed 's/.*: \(.*\)/\1/')
>
> Awk lover's note:
>
> export $(gawk '
> BEGIN {FS=": "}
> /^model.[^n]/ {v=$2}
> /^cpu family/ {f=$2}
> /^model.[^n]/ {m=$2}
> /^stepping/ {s=$2}
> /^cpu MHz/ {h=$2}
> { if (v && f && m && s && h) nextfile }
> END {
> printf "vendor=" v;
> printf "family=" f;
> printf "model=" m;
> printf "stepping=" s;
> printf "mhz=" h;
> }' /proc/cpuinfo)
>
> > level=$(cat "${syspath}/cpu/cpu0/cache/${i}/level")
> > type=$(cat "${syspath}/cpu/cpu0/cache/${i}/type")
>
> type=$(<"${syspath}/cpu/cpu0/cache/${i}/type")
>
>
> Karel
>
> --
> Karel Zak <kzak@xxxxxxxxxx>
--
To unsubscribe from this list: send the line "unsubscribe util-linux-ng" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
