Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 7194 - pmac kernel does not have same vendor_id and proc fields as x86
Summary: pmac kernel does not have same vendor_id and proc fields as x86
Status: RESOLVED FIXED
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: Current packages (show other bugs)
Hardware: PPC Linux
: Lowest enhancement (vote)
Assignee: Maik Schreiber
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2002-08-28 20:19 UTC by Kevyn Shortell (RETIRED)
Modified: 2006-02-04 06:03 UTC (History)
3 users (show)

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Kevyn Shortell (RETIRED) gentoo-dev 2002-08-28 20:19:30 UTC
gentoo-stats appears to have problems with /proc/cpuinfo on ppc
Here comes the culprit:

nick $ cat /proc/cpuinfo
cpu             : 750FX
temperature     : 39 C (uncalibrated)
clock           : 700MHz
revision        : 1.18 (pvr 7000 0112)
bogomips        : 1389.36
machine         : PowerBook4,3
motherboard     : PowerBook4,3 MacRISC2 MacRISC Power Macintosh
detected as     : 257 (iBook 2 with 14" LCD)
pmac flags      : 00000003
L2 cache        : 512K unified
memory          : 128MB
pmac-generation : NewWorld


------- Additional Comment #1 From Nicolas Kaiser 2002-08-02 06:10 -------

I'd like to propose these modifications for PPC. So the output would look like

Vendor: PowerBook4,3
Model:  750FX
MHz:    700
Flags:  PowerBook4,3 MacRISC2 MacRISC Power Macintosh


diff -u /usr/sbin/gentoo-stats /usr/sbin/gentoo-stats.modified
--- /usr/sbin/gentoo-stats      Fri Aug  2 12:57:49 2002
+++ /usr/sbin/gentoo-stats.modified     Fri Aug  2 12:56:59 2002
@@ -63,7 +63,7 @@
        {
                s/[\r\n]//g;

-               if (/^vendor_id/)
+               if (/^machine/)
                {
                        if (length($vendor) > 0)
                        {
@@ -77,7 +77,7 @@
                        $vendor =~ s/^[ \t]*//;
                }

-               if (/^model name/)
+               if (/^cpu/)
                {
                        if (length($modelname) > 0)
                        {
@@ -91,7 +91,7 @@
                        $modelname =~ s/^[ \t]*//;
                }

-               if (/^cpu MHz/)
+               if (/^clock/)
                {
                        if (length($mhz) > 0)
                        {
@@ -103,9 +103,10 @@

                        ($x, $mhz) = split(/:/);
                        $mhz =~ s/^[ \t]*//;
+                       $mhz =~ s/MHz//;
                }

-               if (/^flags/)
+               if (/^motherboard/)
                {
                        if (length($flags) > 0)
                        {


------- Additional Comment #2 From Kevyn Shortell 2002-08-02 06:36 -------

Hi Maik,

I think this is meant for you, since it's your baby =)

Some of the fields would be labeled wrong, but we can hammer those out if your 
willing to make the changes. 

We'd need to look at the table you have, vs the info I have in the proc table 
and see if I can get the appropriate info. But the Vendor of the CPU is not 
PowerBook =)


------- Additional Comment #3 From Maik Schreiber 2002-08-02 07:33 -------

Another config I've been sent by Pieter Van den Abeele:
(I'm adding it here so I won't forget)

cpu               : 7450, altivec supported
clock             : 550Mhz
revisions         : 2.1 (pvr 8000 0201)
bogomips          : 548.86
machine           : Powerbook3,3
motherboard       : Powerbook3,3 MacRISC2 MacRISC Power Macintosh
detected as       : 72 (Powerbook Titanium II)
pmac flags        : 00000003
L2 cache          : 256K unified
memory            : 256MB
pmac-generation   : NewWorld


------- Additional Comment #4 From Maik Schreiber 2002-08-02 07:38 -------

Another one by Robert Cole:

cpu             : 7455, altivec supported
clock           : 667MHz
revision        : 2.1 (pvr 8001 0201)
bogomips        : 665.19
machine         : PowerBook3,4
motherboard     : PowerBook3,4 MacRISC2 MacRISC Power Macintosh
detected as     : 73 (PowerBook Titanium III)
pmac flags      : 00000003
L2 cache        : 256K unified
memory          : 1024MB
pmac-generation : NewWorld


------- Additional Comment #5 From Rajiv Manglani 2002-08-06 12:38 -------

from my powermac 6400:

cpu             : 603ev
clock           : 200MHz
revision        : 2.1 (pvr 0007 0201)
bogomips        : 133.12
machine         : Power Macintosh
motherboard     : AAPL,e407 MacRISC
detected as     : 35 (Alchemy)
pmac flags      : 00000000
memory          : 72MB
pmac-generation : OldWorld



------- Additional Comment #6 From Rajiv Manglani 2002-08-14 18:21 -------

until this is fixed, shouldn't gentoo-stats not have 'ppc' in KEYWORDS?


------- Additional Comment #7 From Mark Guertin 2002-08-25 17:35 -------

From a G4 733 Quicksilver 
 
cpu             : 7450, altivec supported 
clock           : 733MHz 
revision        : 2.1 (pvr 8000 0201) 
bogomips        : 730.72 
machine         : PowerMac3,5 
motherboard     : PowerMac3,5 MacRISC2 MacRISC Power Macintosh 
detected as     : 69 (PowerMac G4 Silver) 
pmac flags      : 00000000 
L2 cache        : 256K unified 
memory          : 512MB 
pmac-generation : NewWorld 
 
from iMac DV 350 
 
cpu             : 740/750 
temperature     : 43-45 C (uncalibrated) 
clock           : 350MHz 
revision        : 2.2 (pvr 0008 0202) 
bogomips        : 697.95 
machine         : PowerMac2,1 
motherboard     : PowerMac2,1 MacRISC Power Macintosh 
detected as     : 66 (iMac FireWire) 
pmac flags      : 00000001 
L2 cache        : 512K unified 
memory          : 256MB 
pmac-generation : NewWorld 
 


------- Additional Comment #8 From Kevyn Shortell 2002-08-25 17:39 -------

Don't need to keep sending us new versions, the data will stay the same 
actually, it's just a matter of changing uname and proc cpuinfo, which isn't 
part of this package.

Nor will the vendor str ever say powerbook =)




------- Additional Comment #9 From Mark Guertin 2002-08-25 17:42 -------

I've set this as -ppc temporarily, until this is resolved  


------- Additional Comment #10 From Kevyn Shortell 2002-08-27 01:34 -------

Bug is requesting that valid information be pulled from /proc/cpuinfo to match 
information gathered by x86 info for the result of generating a corect form. 
Will create new bug for uname.c and proc table modifications.
Comment 1 Kevyn Shortell (RETIRED) gentoo-dev 2002-08-28 20:21:47 UTC
Will work with pmac kernel devs to evaluate possibility of parity 
for /proc/cpuinfo so tools such a gentoo stats can get the same data
Comment 2 Mark Guertin 2002-10-25 11:33:00 UTC
Any news on this trance?

There have been suggestions of using gentoo stats for information gathering and
this would be useful
Comment 3 Rajiv Aaron Manglani (RETIRED) gentoo-dev 2002-10-25 13:58:04 UTC
i thought all archs have different info in /proc/cpuinfo. should this really be
fixed just to get gentoo-stats working? instead, gentoo-stats should be updated
to work with different archs.

i think this bug should be closed, marked as wontfix.

(see also bug 5822)
Comment 4 Mark Guertin 2002-10-27 09:59:43 UTC
ok, ppc support has been added to gentoo-stats, closing this bug