This is the mail archive of the ecos-patches@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: edb7211 flash without CYGHWR_IO_FLASH_DEVICE_NOT_IN_RAM


> I'll be glad to review the changes (and even more glad to purge all
> old drivers!).  The only problem is testing, which I will have to
> leave to you (I assume that since you are sending such patches, you
> have access to this hardware).
To this hardware yes but not to the others which still don't use generic
drivers.

> > Only with the cache macros in is the code in 28F equivalent to edb
> 
> I think this may be because I had some trouble with the cache (it
> didn't seem to ever turn off).  Maybe you can isolate this and get it
> fixed while you make these changes.

I never tested it heavier than causal FIS usage and it seemed to work
with and without the
cache macros is 28F.

> I think the StrataFlash driver is the one to use - it's very flexible 
> and can handle most any of that family of devices.  I'm also pretty
> sure that it does 100% supercede the "bootblock" driver.

Then maybe deleting bootblock from the repo would stop somebody accidentaly use it :) 

> As for your changes below, I don't think they are required.  The upper
> level of the driver (io/flash/...) handles the caches fully so that
> the physical drivers never need bother with it.

That's great so does that mean that no low level drivers need to touch
the cache?
A grep DCACHE_ENABLE -r devs/flash/ shows that there are still some
users including the intel drivers.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]