bug fix: LPM_AWAKE() on MSP430 modifies the stack and may only be called in the interrupt
This commit is contained in:
parent
155b37491c
commit
d16b9ad15d
1 changed files with 4 additions and 2 deletions
|
@ -28,7 +28,7 @@
|
|||
*
|
||||
* This file is part of the Contiki operating system.
|
||||
*
|
||||
* @(#)$Id: tr1001.c,v 1.1 2006/06/18 07:49:33 adamdunkels Exp $
|
||||
* @(#)$Id: tr1001.c,v 1.2 2006/08/17 08:27:54 nifi Exp $
|
||||
*/
|
||||
/**
|
||||
* \addtogroup esb
|
||||
|
@ -343,6 +343,9 @@ interrupt (UART0RX_VECTOR)
|
|||
tr1001_rxhandler(void)
|
||||
{
|
||||
tr1001_default_rxhandler_pt(RXBUF0);
|
||||
if(tr1001_rxstate == RXSTATE_FULL) {
|
||||
LPM_AWAKE();
|
||||
}
|
||||
}
|
||||
/*---------------------------------------------------------------------------*/
|
||||
static void
|
||||
|
@ -501,7 +504,6 @@ PT_THREAD(tr1001_default_rxhandler_pt(unsigned char incoming_byte))
|
|||
|
||||
PACKET_ACCEPTED();
|
||||
tr1001_drv_request_poll();
|
||||
LPM_AWAKE();
|
||||
|
||||
/* We'll set the receive state flag to signal that a full frame
|
||||
is present in the buffer, and we'll wait until the buffer has
|
||||
|
|
Loading…
Reference in a new issue