Re: removing ci_divisor_recip in mips struct cpu_info

[Masao Uebayashi <uebayasi%gmail.com@localhost>]

> Is the layout that important?  "struct cpu_info" always seems to have
> been a bit volatile - there were a couple of new additions to the MIPS
> one only a week or two ago.

What lines are you referring to?  cvs annotate shows me this, and
nothing has been changed in this year?

1.53         (simonb   04-Sep-01): struct cpu_info {
1.73         (yamt     22-Sep-04):      struct cpu_data ci_data;        /* MI 
per-cpu data */
1.85         (ad       19-Oct-07):      struct cpu_info *ci_next;       /* Next 
CPU in list */
1.85         (ad       19-Oct-07):      cpuid_t ci_cpuid;               /* 
Machine-level identifier */
1.58         (simonb   05-Mar-02):      u_long ci_cpu_freq;             /* CPU 
frequency */
1.58         (simonb   05-Mar-02):      u_long ci_cycles_per_hz;        /* CPU 
freq / hz */
1.58         (simonb   05-Mar-02):      u_long ci_divisor_delay;        /* for 
delay/DELAY */
1.64         (simonb   04-Jun-02):      u_long ci_divisor_recip;        /* 
scaled reciprocal of previous;
1.64         (simonb   04-Jun-02):                                         see 
below */
1.82         (yamt     17-May-07):      struct lwp *ci_curlwp;          /* 
currently running lwp */
1.82         (yamt     17-May-07):      struct lwp *ci_fpcurlwp;        /* the 
current FPU owner */
1.82         (yamt     17-May-07):      int ci_want_resched;            /* user 
preemption pending */
1.78         (ad       09-Feb-07):      int ci_mtx_count;               /* 
negative count of held mutexes */
1.78         (ad       09-Feb-07):      int ci_mtx_oldspl;              /* 
saved SPL value */
1.86         (ad       03-Dec-07):      int ci_idepth;                  /* 
hardware interrupt depth */
1.53         (simonb   04-Sep-01): };

Masao