On Tue, Oct 20, 2009 at 11:37:02AM -0600, Myles Watson wrote:
I like the idea of having Options that we use more available.
DEBUG_SMI was the default. Now it isn't. DEBUG_RAM_SETUP was the default for i440bx & vx800
True, but that's intentional. Just like the "use lzma" or "execute vga blob" options these are now global, user-configurable (as opposed to per-board in Options.lb in newconfig). The old per-board or per-chipset defaults don't matter much, the user chooses if he/she wants the option.
I think these options are good candidates for filtering in the compare script.
As long as the functionality isn't changed, I agree. I don't have any of the affected boards to test.
in the Kontron code, #if CONFIG_DEFAULT_CONSOLE_LOGLEVEL > 8 This never does anything, since 8 is SPEW.
Due to the recent LOGLEVEL changes, yes.
Before the changes you pointed out to me that SPEW was 8.
We should do a repo-wide fixing of any such occurences (there may be more of them). This is unrelated to my patch though, the check is no longer correct with and without the patch.
I was pointing out that it isn't compiled in, so if it breaks abuild, it will break abuild when the CONSOLE_LOGLEVEL line changes.
I'll post a followup-patch for this issue if nobody beats me to it.
Right now I'd like to match newconfig and kill it as quickly as possible.
Sure, same here. I don't think this patch interferes with this effort though.
As long as all of the code that gets enabled/disabled is only print statements, that's probably true. I was worried that some of these boards may depend on the correct setting of those debug flags to be functional.
Thanks, Myles