This is the mail archive of the
ecos-patches@sources.redhat.com
mailing list for the eCos project.
[Bug 1000096] new AT91 platform: JTST
- From: bugzilla at ecoscentric dot com
- To: ecos-patches at ecos dot sourceware dot org
- Date: Sun, 19 Sep 2004 14:11:12 +0100 (BST)
- Subject: [Bug 1000096] new AT91 platform: JTST
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1000096
------- Additional Comments From andrew.lunn@ascom.ch 2004-19-09 14:11 -------
+#if defined(CYGHWR_HAL_ARM_AT91_JTST)
+ // TC2 is reserved for AD/DA
+ CYG_ADDRESS timer = AT91_TC+AT91_TC_TC1;
+#else
+ CYG_ADDRESS timer = AT91_TC+AT91_TC_TC2;
+#endif
If the JTST the only device with AD/DA or are there others as well? Would it be
better to change all varients to use TC1?
+// Reset.
+#define HAL_PLATFORM_RESET()
All the other at91 HALs do
// Reset.
__externC void hal_at91_reset_cpu(void);
#define HAL_PLATFORM_RESET() hal_at91_reset_cpu()
This uses the watchdog to reset the platform. Does this not work for the jtst?
+hal_at91_led(int val){
+ unsigned led=BIT21|BIT22|BIT23|BIT24|BIT25|BIT26|BIT27;
+ // all leds off
+ HAL_WRITE_UINT32(AT91_PIO+AT91_PIO_SODR,led)
Doesn't this actually turn all LEDs on, not off. Its the same pattern as used for 8.
You redboot import ecm files have big problems with flash packages. I've fixed
this, but now i have a question. Why include the packages needed for flash when
you then disable it all with:
cdl_option CYGOPT_REDBOOT_FIS {
user_value 0
};
cdl_component CYGSEM_REDBOOT_FLASH_CONFIG {
user_value 0
};
You mlt_arm_at91_jtst_romram.ldi file is broken. It puts the rom_vectors where
there is no memory, which the linker does not like.
I will attach a new version of the patch. Please use this version for all your
further changes.
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.