Re: libgtop has been branched

2005-12-10 Thread Bastien Nocera
On Sat, 2005-12-10 at 11:36 +0800, Davyd Madeley wrote:
 On Fri, 2005-12-09 at 14:49 -0600, Federico Mena Quintero wrote:
 
   If you have any suggestion ...
  
  Make that an opaque structure and add accessors:
  
guint64 glibtop_get_cpu_user (glibtop_cpu *cpu, int cpu_num);
 
 I am referring people to the bug in which this is discussed, because
 increasing magic numbers just ahead of Moore's law is really dumb.
 
 http://bugzilla.gnome.org/show_bug.cgi?id=323354

It's also easy. Don't need to break any source compat, and adds the
requested feature. The main problem is fixing the support for non-Linux
platforms, and older Linux versions.

---
Bastien Nocera [EMAIL PROTECTED] 


___
desktop-devel-list mailing list
desktop-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/desktop-devel-list


Re: libgtop has been branched

2005-12-10 Thread Bastien Nocera
On Fri, 2005-12-09 at 14:49 -0600, Federico Mena Quintero wrote:
 On Fri, 2005-12-09 at 17:23 +0100, BenoƮt Dejean wrote:
  * include/glibtop/cpu.h:
  * include/glibtop/procmap.h:
  
  Increased GLIBTOP_NCPU to 32.
  Added real_ncpu to struct glibtop.
  Added smaps members to glibtop_map_entry.
 
  If you have any suggestion ...
 
 Make that an opaque structure and add accessors:
 
   guint64 glibtop_get_cpu_user (glibtop_cpu *cpu, int cpu_num);
 
 That's pretty well documented in the GNOME Programming Guidelines, by
 the way.
 
 (And what does a guint64 mean for that value?  The API is not obvious at
 all.)
 
 And having these in the public API is pretty busted:
snip

All those crappy APIs are the ones that were there in libgtop 1.x. You
would have to ask Martin why he chose those APIs...

---
Bastien Nocera [EMAIL PROTECTED] 


___
desktop-devel-list mailing list
desktop-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/desktop-devel-list