sergiotarxz/nuttx
sergiotarxz/nuttx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Mark an untested driver as EXPERIMENTAL in its Kconfig; Update some comments.

Browse Source
This commit is contained in:
Gregory Nutt 2017-09-11 09:35:39 -06:00
parent 2ffc2ab875
commit 7fbceacb46
3 changed files with 28 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
drivers/wireless/ieee802154/Kconfig
View File

@ -16,6 +16,7 @@ source drivers/wireless/ieee802154/mrf24j40/Kconfig
config IEEE802154_AT86RF233
bool "ATMEL RF233 IEEE 802.15.4 transceiver"
default n
depends on EXPERIMENTAL
---help---
This selection enables support for the Atmel RF233 device.

4
include/nuttx/net/sixlowpan.h
View File

@ -262,7 +262,9 @@
#define SIXLOWPAN_MAC_STDFRAME 127
/* Space for a two byte FCS must be reserved at the end of the frame */
/* Space for a two byte FCS must be reserved at the end of the frame.
* REVISIT: True for IEEE 802.15.4, but not for some other packet radios.
*/
#define SIXLOWPAN_MAC_FCSSIZE 2

27
sched/sched/sched_verifytcb.c
View File

@ -1,7 +1,7 @@
/****************************************************************************
* sched/ched_verifytcb.c
*
* Copyright (C) 2009 Gregory Nutt. All rights reserved.
* Copyright (C) 2009, 2017 Gregory Nutt. All rights reserved.
* Author: Gregory Nutt <gnutt@nuttx.org>
*
* Redistribution and use in source and binary forms, with or without
@ -59,8 +59,29 @@
bool sched_verifytcb(FAR struct tcb_s *tcb)
{
/* Return true if the PID hashes to this TCB. */
/* Return true if the PID hashes to this TCB. This will catch the case
* where the task associated with the TCB has terminated (note that
* sched_releasedtcb() will nullify the TCB field in that case). The
* following logic will also detect the case where the task assocated
* with the TCB has terminated and another task has been started with a
* different TCB but with a PID hashing to the same entry.
*
* NOTE: In the event that the TCB has terminated, the 'tcb' parameter
* will point at either a stale or a re-allocated memory allocation. The
* PID fetched by the use of the bad pointer(tcb->pid) should not cause
* any memory faults (at least in the FLAT or PROTECTED builds) because
* we do at least know that the pointer refers to valid memory and the
* hash macro, PIDHASH(), will return a valid, in-range index into the
* g_pidhash[] table.
*
* REVISIT: This logic will not, however, catch the case where the task
* originally associated with the TCB has terminated, but a new task was
* started reusing the same memory allocation for its TDB that was freed
* by the terminated task. In this case, a false positive value will be
* returned: The TCB is valid but does not refer to the same task as
* before. This case is not detectable with the limited amount of
* information available.
*/
return tcb == g_pidhash[PIDHASH(tcb->pid)].tcb;
}