Re: [PATCH v4 3/3] Doc/memory-barriers: Add Korean translation

[Byungchul Park]

On Fri, Jul 08, 2016 at 07:50:39AM +0900, SeongJae Park wrote:
> > I will add my opinion in korean.
> 
> Thank you for kind and faithful review.  I agree with most of your opinions and
> suggestions.  Most of your suggestions looks much better than mine.
> 
> However, I also have some different opinion.  I want to emphasize the fact that
> (1) CPUs 'issue' memory operations to memory system as they want, (2) memory
> system 'executes' those operations as they want, and (3) CPUs 'perceive' the
> 'effects' of the operation executions as they want.  I want to emphasize the
> fact in document because I think most confusion about memory ordering comes
> from vague understanding about the relation.  To my perspective, few of your

Right. I agree with you.

> suggestions could enhance readability but could dim the point, too.  I have
> appended the opinion in Korean line by line, too.  So, if you do not opposed
> to, I will enhance the text again while keeping the point.

...

> >> +
> >> +     STORE *A = 5, x = LOAD *D
> >> +     x = LOAD *D, STORE *A = 5
> >> +
> >> +ëëì ìíì ëìíë ììì _íì_ ììë ììíëë, ìëë ëìì ììí
> >> +êìëë.
> >> +
> >> +
> >> +ëììí
> >> +--------
> >> +
> >> +CPU ìê êëí ì ìë ììíì ëììí ëêìê ììëë:
> >> +
> >> + (*) ìë CPU ë, ìììì ììíë ëëë ìììëì íë CPU ìììê
> >> +     ìììë íë ìì êëë ììëëë. ì, ëìì ëíìë:
> >
> > ììëëë. -> ìíëëë.
> 
> CPU (ëë ëëìì) ë ëëë ìíëììì ëëë ììíì 'ìì'íê ëëë
> ììíì ìí 'ìí'ë ìíëììì 'êê'ë 'ëêë' ëë CPU ê ììíì
> ëëë ììíìì ìíë ìíëììì ìíì ëê ëë êìë, (1) CPU ê
> ìëì ììë ëêì ììí ì ìê, (2) (ììë ííí) ëëë ììíì ììë
> ìíëììëì ììì ììë ìíí ì ìê, (3) ëëë ììíì ìí êêë ê
> CPU ê ììì ììë ë ì ìê ëë ë ìì ëëë ìì êìì êëìê,
> ìëì êë ìë êë ëíìë ìííê ììëë.
> 
> ì êëë ëìíê 'ìííë'ê íê ëë ì ëëì íêëê í ìë ìì ê
> êìì, êëí CPU ë 'ìì' íê ëëë ììíìëëí êêë 'ìëëëë'ë
> ëëì ëì ììììë ììíëê ìëê ììëë.

Yes. Agree.

> ì ëìì êì êë 'ìì'íëê íëê ì ììí ê êìë, ëëë ììíìê
> ììíëë ìì ì ë êìíë ììë ììíëë íêìëë.

Agree. And it seems to be not easy :(

> >> +
> >> +     Q = READ_ONCE(P); smp_read_barrier_depends(); D = READ_ONCE(*Q);
> >> +
> >> +     CPU ë ëìê êì ëëë ìíëììëì ììíëë:
> >
> > (...) ìë ëëë ìíëììì ìííëë:
> 
> ìì 'ìì'ì ëëì ìëëë íêìëë.

Yes.

> >> + (*) íì CPU ëìì êìì ííìë ëëì ìíì ëì ê CPU ìììë ììê
> >
> > ëëë ììì êìë ëëì ìíìê í CPU ëìì ìíë êì, íë CPU
> > ëìì ëìì ëì ê ììë ëëì ììëë. (...)
> 
> ìì 'ìì' ì ëëì ëêëë íêìëë.

Yes.

> > -> íë CPUë ìì ëì ììëë ëëë ìíëììì ìíí êìëë.
> 
> ìì 'ìì' ì ëëì ëêëë íêìëë.

Yes.

> >> +     a = LOAD *X, STORE *X = b
> >> +
> >> +     êëê ëìì ëíìë:
> >> +
> >> +     WRITE_ONCE(*X, c); d = READ_ONCE(*X);
> >> +
> >> +     CPU ë ìëì ëìì ëëëì ëëë:
> >
> > (...) ìëì ìë ììë ìííëë.
> 
> ìì 'ìì' ì ëëì ëêëë íêìëë.

Yes.

> > I think there are still many things literally translated. It makes it much
> > harder to read.
> >
> > I think it's not the version with which I can add opinions line by line.
> > Please review all sentences line by line carefully by yourself and make it
> > more readable.
> >
> > I hope you don't hurry and spend more time to make it done. I believe you
> > agree with it.
> 
> Ok, I agree with your opinion.  Actually, I have reviewed all sentences line by
> line for this version, too.  However, it looks like it wasn't sufficient.  I
> will enhance it again.

Thank you,
Byungchul

> 
> >
> > Thank you,
> > Byungchul