On Wed 24 Jun 2009 19:55, Mike Frysinger pondered:
> On Wed, Jun 24, 2009 at 16:11, Robin Getz wrote:
> > That should make most of the nastiness go away there - and, and the same
> > time - clean up bug 3417
> >
> > http://blackfin.uclinux.org/gf/tracker/3417
> 
> btw, is the last comment in that bug still valid ?  i.e. does
> isram_memcpy() only operate on 64bit chunks ?  i thought you fixed it
> to be like a real memcpy() so it operates down to the individual byte
> level ...

I did.

the internal isram_read and isram_write functions operate on complete cache 
lines.

the isram_memcpy() functions at the byte level, and there is no need for 
alignment.

-Robin

BTW - arch/blackfin/mm/isram-driver.c:isram_check_addr() looks like another 
place for bfin_mem_access_type() (if it not BFIN_MEM_ACCESS_ITEST, return 
false)...

-Robin
_______________________________________________
Linux-kernel-commits mailing list
[email protected]
https://blackfin.uclinux.org/mailman/listinfo/linux-kernel-commits

Reply via email to