Re: svn commit: r205013 - in head/sys: amd64/amd64 i386/i386

2010-03-14 Thread Alexander Best
John Baldwin schrieb am 2010-03-11:
> On Thursday 11 March 2010 12:01:28 pm Alexander Best wrote:
> > thanks for the commit. :)

> > a few thoughts:

> > 1) why does stepping remain to be printed in dec while family and
> >model are in
> > hex? is this the way amd/intel cpu docs refer to
> > stepping/model/family?

> I just left it the way it was.

> > 2) the hex value for "Id" and "(AMD) Features(2)" gets printed with
> >an "0x"
> > prepended to indicate it's in hex. maybe this should also be the
> > case here.

> Because on a typical Intel system you overflow 80 cols if you do
> that. :(
> Otherwise I would have.  We could maybe add an 'h' suffix as that is
> what
> the manuals and it would only add 2 chars rather than 4.

ah i see. not an easy decision. i wouldn't use the 'h' suffix, because there
should be only one way of relating to a hex value and i believe the overall
convention in freebsd is '0x'.so better keep it the way it is. maybe using %X
instead of %x in the printfs makes the output look a bit nicer, but that's
just eyecandy. ;)

cheers.
alex
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"


Re: svn commit: r205013 - in head/sys: amd64/amd64 i386/i386

2010-03-11 Thread John Baldwin
On Thursday 11 March 2010 12:01:28 pm Alexander Best wrote:
> thanks for the commit. :)
> 
> a few thoughts:
> 
> 1) why does stepping remain to be printed in dec while family and model are in
> hex? is this the way amd/intel cpu docs refer to stepping/model/family?

I just left it the way it was.

> 2) the hex value for "Id" and "(AMD) Features(2)" gets printed with an "0x"
> prepended to indicate it's in hex. maybe this should also be the case here.

Because on a typical Intel system you overflow 80 cols if you do that. :(
Otherwise I would have.  We could maybe add an 'h' suffix as that is what
the manuals and it would only add 2 chars rather than 4.

-- 
John Baldwin
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"


Re: svn commit: r205013 - in head/sys: amd64/amd64 i386/i386

2010-03-11 Thread John Baldwin
On Thursday 11 March 2010 12:01:28 pm Alexander Best wrote:
> thanks for the commit. :)
> 
> a few thoughts:
> 
> 1) why does stepping remain to be printed in dec while family and model are 
in
> hex? is this the way amd/intel cpu docs refer to stepping/model/family?
> 2) the hex value for "Id" and "(AMD) Features(2)" gets printed with an "0x"
> prepended to indicate it's in hex. maybe this should also be the case here.
> 
> cheers.
> alex
> 

-- 
John Baldwin
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"


Re: svn commit: r205013 - in head/sys: amd64/amd64 i386/i386

2010-03-11 Thread Alexander Best
thanks for the commit. :)

a few thoughts:

1) why does stepping remain to be printed in dec while family and model are in
hex? is this the way amd/intel cpu docs refer to stepping/model/family?
2) the hex value for "Id" and "(AMD) Features(2)" gets printed with an "0x"
prepended to indicate it's in hex. maybe this should also be the case here.

cheers.
alex
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"


svn commit: r205013 - in head/sys: amd64/amd64 i386/i386

2010-03-11 Thread John Baldwin
Author: jhb
Date: Thu Mar 11 14:17:37 2010
New Revision: 205013
URL: http://svn.freebsd.org/changeset/base/205013

Log:
  Print out the family and model from the cpu_id.  This is especially useful
  given the advent of the extended family and extended model fields.  The
  values are printed in hex to match their common usage in documentation.
  
  Submitted by: Alexander Best
  MFC after:1 week

Modified:
  head/sys/amd64/amd64/identcpu.c
  head/sys/i386/i386/identcpu.c

Modified: head/sys/amd64/amd64/identcpu.c
==
--- head/sys/amd64/amd64/identcpu.c Thu Mar 11 13:16:06 2010
(r205012)
+++ head/sys/amd64/amd64/identcpu.c Thu Mar 11 14:17:37 2010
(r205013)
@@ -187,7 +187,9 @@ printcpuinfo(void)
if (cpu_vendor_id == CPU_VENDOR_INTEL ||
cpu_vendor_id == CPU_VENDOR_AMD ||
cpu_vendor_id == CPU_VENDOR_CENTAUR) {
-   printf("  Stepping = %u", cpu_id & 0xf);
+   printf("  Family = %x", CPUID_TO_FAMILY(cpu_id));
+   printf("  Model = %x", CPUID_TO_MODEL(cpu_id));
+   printf("  Stepping = %u", cpu_id & CPUID_STEPPING);
if (cpu_high > 0) {
 
/*

Modified: head/sys/i386/i386/identcpu.c
==
--- head/sys/i386/i386/identcpu.c   Thu Mar 11 13:16:06 2010
(r205012)
+++ head/sys/i386/i386/identcpu.c   Thu Mar 11 14:17:37 2010
(r205013)
@@ -672,9 +672,11 @@ printcpuinfo(void)
cpu_vendor_id == CPU_VENDOR_NSC ||
(cpu_vendor_id == CPU_VENDOR_CYRIX &&
 ((cpu_id & 0xf00) > 0x500))) {
-   printf("  Stepping = %u", cpu_id & 0xf);
+   printf("  Family = %x", CPUID_TO_FAMILY(cpu_id));
+   printf("  Model = %x", CPUID_TO_MODEL(cpu_id));
+   printf("  Stepping = %u", cpu_id & CPUID_STEPPING);
if (cpu_vendor_id == CPU_VENDOR_CYRIX)
-   printf("  DIR=0x%04x", cyrix_did);
+   printf("\n  DIR=0x%04x", cyrix_did);
if (cpu_high > 0) {
 
/*
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"