Dear Matthias,
Matthias Ludwig wrote:
Wolfgang, Dirk,
cause for this patch was originally the plan to bring my company's new omap3-based platform into U-Boot. As we do use other CS for a network chip this patch was intended to be the starting point (otherwise our patchset will not apply).
I can prepare a big #define -> c_struct patch for omap next week. The only problem i see, my following patches will be rebased onto that and will only apply after the transition #define -> c_struct is done.
If this is ok for you i will submit this mid/end of next week.
If you do it for all OMAP3 boards currently in mainline and don't break anything, I'm fine with this, too.
Many thanks
Dirk
On Thu, May 07, 2009 at 08:58:17PM +0200, Wolfgang Denk wrote:
Dear Dirk,
In message 4A02FB34.2090907@googlemail.com you wrote:
Please provide proper C structs!
Would you like to have a look to the code snippet visible in Matthias' patch
http://lists.denx.de/pipermail/u-boot/2009-May/052157.html
?
Done.
It's my understanding that what's in code
writel(NET_GPMC_CONFIG1, &gpmc_cs5_base->config1);
is what you want? I.e. as I understand it, the code is correct (we use
It is only part of what I want to see. There are still deficiencies.
Yes, we have a struct gpmc_csx, that's good.
Note though that using such typedef's not in line with the Coding Style, which says: "It's a _mistake_ to use typedef for structures and pointers." Also, checkpatch.pl complains about this.
The next problem is that the entries in these structs are declared as "unsigned int", which just happens to work by chance. "u32" would be more reliable.
Finally, and this is what I really compalin about, is that there is no big structure which includes all the blocks that make up the CPU into one big structure (as it's done for example for PowerPC systems in the include/asm-ppc/*immap* files) - you still use code like
gpmc_csx_t *gpmc_cs5_base = (gpmc_csx_t *)GPMC_CONFIG_CS5_BASE;
to locate each of the individual C structs in the memory map; instead, onle one single pointer to the internal memory should be needed.
Sorry if I missed something, just correct then ;)
Done. 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: wd@denx.de Winners never talk about glorious victories. That's because they're the ones who see what the battlefield looks like afterwards. It's only the losers who have glorious victories. - Terry Pratchett, _Small Gods_