Re: eboot, accessing hardware
- From: Valter Minute <v_a_l_t_e_r.m_i_n_u_t_e@xxxxxxxxxxxxx>
- Date: Wed, 03 Oct 2007 03:20:31 -0700
=?Utf-8?B?TWFhcnRlbg==?= <Maarten@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote
in news:7D688E3A-F589-4AB9-AB7F-A943489BB569@xxxxxxxxxxxxx:
During eboot I need to access the I2C bus. Is it allowed to use
the code out of the Chip Support Package to do this? The I2C
component that is used in NK is also using this code.
If the code inside the CSP does not use any OS API, don't rely on
interrups (usually not enabled inside the bootloader) and allows you
to change the addresses used to access the device (if the loader uses
a different virtual memory map or does not enable MMU) you will be
able to use it as is.
If the code uses some OS API you may try to re-implement them inside
the bootloader using the same prototype. Obviously some API may be
easy to "fake" while other may require a lot of work (ex: to
"simulate" a IST or something like that) and will make a rewrite of
those routines quicker than a porting.
If you provide more detail about the CPU you are using, somebody who
know that CPU and it's BSP may provide you more "targeted"
information.
--
Valter Minute
(the reply address of this message is invalid)
(l'indirizzo di reply di questo messaggio non è valido)
.
- Prev by Date: Re: Multiple XIP regions cause kernel crash under CE6.0
- Next by Date: Re: ATL activeX dll under CE6.0 porting problem
- Previous by thread: Exception "Paging in from uncompressed R/O page from XIP module --
- Next by thread: Re: ATL activeX dll under CE6.0 porting problem
- Index(es):
Relevant Pages
|
