Right now I'm working through a https://community.freescale.com/message/89566#89566 issue, and in the course of it, I recalled a passage from the M54455EVB BSP User's Manual, Rev 1.3, page 3-1:
This BSP is known to work on the following board(s) revision(s):
Not knowing exactly how to identify the silicon revision (unsure if "Freescale MCF54455 (Mask:48 Version:1)" in U-Boot means Rev 1 silicon), I Googled and found the chip errata document, which shows that Rev 1 is marked "1M22H" and Rev 2 is "2M22H".
So I checked my (newly acquired) M54455EVB, and while the sticker on the board says "Rev. D", the chip itself is marked "1M22H", meaning that I've got Rev 1 silicon.
Do I have any hope of using the new(ish) Sep-2010 LTIB with Kernel 2.6.29 with this Rev 1 silicon, or do I need to ship my M54455EVB mainboard off for BGA rework to get Rev 2 silicon put on there? Has anyone had any issues (or heck, any success) with running the 20100919 LTIB on an M54455EVB with Rev 1 silicon?
已解决! 转到解答。
I would file an on-line service request on this. It would accomplish several things:
1) Make the engineers aware of the work-around
2) They might have information as to an acutal solution and can pass that back to you,
3) Provide an opportunity to update the documentation.
While we do try to solve problems in these forums (and many thanks to everyone who contributes), sometimes it is best to file a service request to get the issue logged into our database and looked at.
---Tom
The Errata Items current for Mask 1 that are fixed in Mask 2 are:
It is unlikely that any of the above could explain the "slow boot".
If you don't have a debug pod it might be worth getting one. If you do have a debug pod then using it is the easiest option.
Tom
Well, since I've managed to solve the slow-boot issue (PATA driver going nuts), it seems that Linux 2.6.29 runs fine on Rev 1 silicon. It would be nice if someone from Freescale could officially verify this and update the BSP user's manual to indicate that Rev 2 silicon is not strictly required.
As I recall (it has been roughly 3 years since I worked on this) the Rev 1 silicon was not produced in volume. Aside from a few development boards it wasn't made available - and wasn't in any products. Maybe JWW can correct me?
I would file an on-line service request on this. It would accomplish several things:
1) Make the engineers aware of the work-around
2) They might have information as to an acutal solution and can pass that back to you,
3) Provide an opportunity to update the documentation.
While we do try to solve problems in these forums (and many thanks to everyone who contributes), sometimes it is best to file a service request to get the issue logged into our database and looked at.
---Tom