Dear Vasily Khoruzhick,

In message <CA+E=qvexe0wutn4nb6k-zah7uqvf4j0umd33+gxkqb3ypvt...@mail.gmail.com> 
you wrote:
> 
> Doing SDRAM init very early (even in assembly code) is obvious and simple.
> Just few simple writes into RAM controller registers.

This is a misconception.

Experience on this very mailing list has shown that an extremely high
number of issues is cause by incomplete and/or incorrect memory
initialization.  This is why U-Boot (be design!) makes sure we have a
full C environment when doing this taks, where we can use printf() and
friends for useful output etc.

> It's not as simple as previous solution, isn't it? And looks

No, it's not so simple, but far more convenient.  And over time, much
more time-saving.

> overengineered to me. I prefer to keep
> such low-level things simple.

We respect your personal preferences.  But in this case we ask you to
stick with the U-Boot design andimplement as requested. Thanks.

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: w...@denx.de
1 1 was a race-horse, 2 2 was 1 2. When 1 1 1 1 race, 2 2 1 1 2.
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot

Reply via email to