This is the mail archive of the ecos-patches@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

AT91 hal_clock_reset fix


Performing an interrupt acknowledge in both hal_clock_reset and
Cyg_RealTimeClock::isr looks wrong to me.


Best wishes,
  --Daniel


diff -u -5 -p -r1.1.1.5 -r1.11
--- hal/arm/at91/var/current/ChangeLog	2003/08/04 16:08:44	1.1.1.5
+++ hal/arm/at91/var/current/ChangeLog	2003/08/06 12:43:35	1.11
@@ -1,5 +1,11 @@
+2003-08-06  Daniel Néri  <daniel.neri@sigicom.se>
+
+	* src/at91_misc.c (hal_clock_reset):
+	Don't acknowledge interrupt here (this is handled
+	in Cyg_RealTimeClock::isr).
+
 2003-07-18  Nick Garnett  <nickg@balti.calivar.com>
 
 	* cdl/hal_arm_at91.cdl:
         Changed values for CYGNUM_HAL_RTC_NUMERATOR,
         CYGNUM_HAL_RTC_DENOMINATOR and CYGNUM_HAL_RTC_PERIOD to
diff -u -5 -p -r1.1.1.5 -r1.8
--- hal/arm/at91/var/current/src/at91_misc.c	2003/08/04 16:08:44	1.1.1.5
+++ hal/arm/at91/var/current/src/at91_misc.c	2003/08/06 12:41:40	1.8
@@ -101,11 +101,10 @@ void hal_clock_reset(cyg_uint32 vector, 
     cyg_uint32 sr;
 
     CYG_ASSERT(period < 0x10000, "Invalid clock period");
 
     HAL_READ_UINT32(timer+AT91_TC_SR, sr);  // Clear interrupt
-    HAL_INTERRUPT_ACKNOWLEDGE(CYGNUM_HAL_INTERRUPT_RTC);
 
     if (period != _period) {
         hal_clock_initialize(period);
     }
     _period = period;

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]