[linux-cirrus] Re: bootmem_init_node failure with 64M TS-7250



Charles Moschel wrote:

I think the costs are approaching parity, but 1 64M chip has been
significantly more costly that 2 32M chips for some time.  We do put on
64M chips from time to time for customers that need it as our 2.4 kernel
modifications allow the chips to be pretty much dropped in.  Both our
RedBoot bootloader and our Linux-based bootloader (Linux booting Linux)
auto-detect the SDRAM chip sizes and pass appropriate ATAGs.


Well, RedBoot was a bit of a red herring for me, initially.  I found it
did pass ATAGs correctly, but the redboot banner still says 32M RAM was
available (0x02000000).  And when I tried Lennert's suggestion of
changing PHYS_OFFSET and loading the kernel at 0xe0008000, RedBoot
helpfully asked if I was sure, because the address appeared to be
outside of RAM (see boot log below).

While talking about reboot, would it be possible for TS to build RedBoot
to pass the correct machine ID?  Or would that break the 2.4 TS kernels?
I'm sure we could do this change, but we're actually trying to get away from RedBoot and uBoot altogether. Our new products boot a minimal and fast-booting Linux 2.4 kernel straight from the boot firmware. There is then an executable "bootload" that can boot other OS and Linux kernels from this bootloader kernel. I definitely would like to be as correct as possible with the "bootload" program since it is a new development for TS. If somebody could tell me exactly what magic model numbers to pass in what registers, I would gladly change both the "bootload" program and TS's default shipped Redboot on older products. 


Is it relatively simple for you to properly fix the problem once you
have access to an afflicted board?

I think so.  [ Maybe I can trade in my 32M ts7250 for a 64M model. :) ]

No problem.  Where would you like it sent?  We could probably give you a
different board too if you wanted.  Since the TS-7250, we've released a
7260, 7300, and 7400 based on the same processor.

//Jesse Off


Great!  It would be great if TS could tell customers 'our boards work
out of the box with Linux 2.6', even if 2.6 is not officially supported
(yet!) by TS.  Lennert has done an outstanding job in getting ep93xx
support upstream, and most of the bits are in place already.  More 2.6
users on TS boards == more development on TS boards, as almost all
new/interesting development is done on 2.6.

Regards, ......... Charlie
I agree Lennert has done an outstanding job, I wish I had some time to play with 2.6 and contribute! If 2.6 was in this state 3 years ago when the TS-7xxx line was started, TS would certainly have chosen this kernel as default and have been more able to participate in its further development/testing. I have been telling people that ask me directly that 2.6 boots on many of our products for many months now. 

//Jesse Off

Other related posts: