On Sun, Nov 11, 2018 at 08:51:33AM +0900, Akira Yokosawa wrote:
> On 2018/11/10 20:49:21 +0800, Junchang Wang wrote:
> > Signed-off-by: Junchang Wang <junchangwang@xxxxxxxxx>
> > ---
> > Hi Paul,
> >
> > This is the only patch for Chapter CPU. Please take a look.
> >
> >
> > Thanks,
> > --Junchang
> >
> > --
> > cpu/overheads.tex | 4 ++--
> > cpu/overview.tex | 2 +-
> > 2 files changed, 3 insertions(+), 3 deletions(-)
> >
> > diff --git a/cpu/overheads.tex b/cpu/overheads.tex
> > index 2474cfe..17b120b 100644
> > --- a/cpu/overheads.tex
> > +++ b/cpu/overheads.tex
> > @@ -155,7 +155,7 @@ displayed in
> > Table~\ref{tab:cpu:Performance of Synchronization Mechanisms on 4-CPU 1.8GHz AMD Opteron 844 System}.
> > This system's clock period rounds to 0.6\,ns.
> > Although it is not unusual for modern microprocessors to be able to
> > -retire multiple instructions per clock period, the operations's costs are
> > +retire multiple instructions per clock period, the operations' costs are
> > nevertheless normalized to a clock period in the third column, labeled
> > ``Ratio''.
> > The first thing to note about this table is the large values of many of
> > @@ -360,7 +360,7 @@ It is clear that the combination of speculative execution and cloud
> > computing needs more than a bit of rework!
> >
> > A fifth hardware optimization is large caches, allowing individual
> > -CPUs to operate on larger datasets without incuring expensive cache
> > +CPUs to operate on larger datasets without incurring expensive cache
> > misses.
>
> Nice catches!
Indeed!!! ;-)
> > Although large caches can degrade energy efficiency and cache-miss
> > latency, the ever-growing cache sizes on production microprocessors
> > diff --git a/cpu/overview.tex b/cpu/overview.tex
> > index 071cf7c..9e91a7b 100644
> > --- a/cpu/overview.tex
> > +++ b/cpu/overview.tex
> > @@ -15,7 +15,7 @@ where the race always goes to the swiftest.
> > \ContributedBy{Figure}{fig:cpu:CPU Performance at its Best}{Melissa Broussard}
> > \end{figure}
> >
> > -Although there are a few CPU-bound benchmarks that approach the ideal
> > +Although there are a few CPU-bound benchmarks that approach the ideal case
> > shown in Figure~\ref{fig:cpu:CPU Performance at its Best},
>
> "ideal" can be used as a noun. I don't think this hunk is necessary.
This is quite true, but making it a bit less confusing to non-native
English speakers is worth something. Not that I am volunteering to
confine myself to the 850-word vocabulary of Basic English, mind you! ;-)
So Akira has a good point, but I will take Junchang's patch.
Thanx, Paul
> Thanks, Akira
>
> > the typical program more closely resembles an obstacle course than
> > a race track.
> >
>
