-----Original Message----- From: Dirk Behme [mailto:dirk.behme@googlemail.com] Sent: Wednesday, April 22, 2009 1:04 AM To: Premi, Sanjeev Cc: u-boot@lists.denx.de Subject: Re: [U-Boot] [PATCH] OMAP3: Print correct silicon revision
Dear Premi,
Premi, Sanjeev wrote:
-----Original Message----- From: Premi, Sanjeev Sent: Tuesday, April 21, 2009 11:37 PM To: 'Dirk Behme' Cc: u-boot@lists.denx.de Subject: RE: [U-Boot] [PATCH] OMAP3: Print correct silicon revision
-----Original Message----- From: Dirk Behme [mailto:dirk.behme@googlemail.com] Sent: Tuesday, April 21, 2009 10:26 PM To: Premi, Sanjeev Cc: u-boot@lists.denx.de Subject: Re: [U-Boot] [PATCH] OMAP3: Print correct
silicon revision
Dear Premi,
Sanjeev Premi wrote:
The function display_board_info() displays the silicon revision as 2 - based on the return value from get_cpu_rev().
This is incorrect as the current Si version is 3.1
Thanks for the patch and fixing this!
This patch displays the correct version; but does not change get_cpu_rev() to minimize the code impact.
I wonder if it wouldn't be better (and cleaner) to fix
get_cpu_rev()?
Yes. This is what I started with; but then this is where I
felt that
fix may run 'deeper"
u32 get_board_type(void) { if (get_cpu_rev() == CPU_3430_ES2) return sysinfo.board_type_v2; else return sysinfo.board_type_v1; }
...sorry, mail 'went' before I wanted to!
I couldn't figure out how this impacts boards other than the EVM.
Though I admit not having much time looking for the impact. Beyond this, I believe the fix could be straight forward.
What's about something like in the attachment? Compile tested only. Do you like to test it?
Yes, this works on the EVM.
I did spend some more time & fouund that value from get_board_type is ignored in the display_board_info().
I will submit a patch to remove this function if it is really not needed.
Best regards, Sanjeev
Best regards
Dirk
[...snip...]