3 messages, 2 authors, 2005-08-31 · open the first message on its own page
From: Mark A. Greer <hidden>
Date: 2005-08-30 23:38:51
This dates back to a discussion in May. This patch addresses point 1) in the email below by ripping out the progress msgs from MMU_init(). This would allow us to get rid of all the xxx_set_bat() routines in the various platform files. Does anyone object to this patch? Mark http://ozlabs.org/pipermail/linuxppc-dev/2005-May/019120.html
From: Dan Malek <hidden>
Date: 2005-08-31 01:09:55
On Aug 30, 2005, at 7:38 PM, Mark A. Greer wrote:
Does anyone object to this patch?
Don't you still need this to run stuff like xmon and kgdb? Thanks. -- Dan
From: Mark A. Greer <hidden>
Date: 2005-08-31 01:29:40
On Tue, Aug 30, 2005 at 09:09:55PM -0400, Dan Malek wrote:
On Aug 30, 2005, at 7:38 PM, Mark A. Greer wrote:quoted
Does anyone object to this patch?Don't you still need this to run stuff like xmon and kgdb?
Those should work with just an io_block_mapping() which will be active before setup_arch() calls xmon() and breakpoint(). The problem with progress msgs in MMU_init() is that they try to bang on the uart before ioremaps/io_block_mappings are active. That's why I (and others) had the xxx_set_bat() hack. Mark