2bcb9edd91
Add gettimeofday declaration in sys/time.h
2013-09-08 15:01:29 +08:00
dacd2cb9a9
travis: add scan-build in PATH
2013-09-07 18:31:05 +08:00
4240ebd73f
travis: don't export RTT_CC in before_script if it's been set
...
This gives ways for bsp to set it's own RTT_CC.
2013-09-07 17:17:58 +08:00
eae89b9576
travis: install the gcc-multilib to get the 32bit build OK
2013-09-07 16:15:43 +08:00
5ab9943eb5
travis: try to fix the SDL building
2013-09-07 16:01:03 +08:00
0168048f27
travis: auto build the simulator
...
Building is only the first step. Correctness is what we need. There are
already many GCC builds for other bsp so GCC building for simulator is
not important. So I use clang-analyze to check all the source codes in
simulator project. Hope it will help us.
2013-09-07 15:46:11 +08:00
35cac170fa
simulator: add clang-analyze tool support
...
When CROSS_TOOL or 'RTT_CC' env is setted to 'clang-analyze', it will
use Clang to check the codes.
2013-09-07 15:15:48 +08:00
c289aa9fc1
Scons: add a custom tool to do clang static analyze
...
Bsps can use the clang analyzer as a tool:
env = Environment(toolpath=[os.path.join(RTT_ROOT, 'tools',
'tools')], tools = ['clang-analyze'])
When building the project, the static analyzer will be called to check
all the C code. The warnings are print to stderr.
2013-09-07 15:15:48 +08:00
3b4f3f5931
Scons: run `clang -Wall -fsyntax-only` in clang-analyze
...
The `clang -fsyntax-only` will give us additional warning messages in
the console.
2013-09-07 15:15:47 +08:00
18692c29b8
simulator: get CROSS_TOOL and EXEC_PATH from env
...
Simulator get it's way to Travis CI ready.
2013-09-06 21:54:27 +08:00
c6e81b58d9
simulator: move SDL setting staff into drivers/SConstruct
...
This could not only make a cleaner SConstruct, but could also benefit
RTGUI building.
2013-09-06 21:49:53 +08:00
15b9db80d6
Merge pull request #154 from grissiom/simulator
...
sim/posix: move rt_hw_context_switch_interrupt after it's definition
2013-09-03 22:55:17 -07:00
3ebc766521
sim/posix: move rt_hw_context_switch_interrupt after it's definition
...
This fix a compile error in Clang.
2013-09-04 00:05:03 +08:00
b4166e0b25
Merge pull request #153 from grissiom/stm32f10x
...
stm32f10x: fix the rtgui initialization
2013-09-03 05:46:20 -07:00
d70761222c
stm32f10x: fix the rtgui initialization
...
The rtgui_system_server_init prototype is wrong. And if we use
RT_USING_COMPONENTS_INIT, there is no need to call
rtgui_system_server_init again.
2013-09-03 20:09:45 +08:00
49e28c2629
Merge pull request #152 from grissiom/log-trace
...
log_trace: cleanup
2013-09-03 05:00:40 -07:00
ea66b820e1
log_trace: cleanup
2013-09-03 16:09:16 +08:00
51ce91f30b
Merge pull request #151 from grissiom/finsh
...
finsh: use alias in the variable definition
2013-09-02 00:12:03 -07:00
e2a0b1930b
Merge pull request #150 from grissiom/simulator-linux
...
Simulator linux
2013-08-31 17:14:21 -07:00
c05b205390
finsh: use alias in the variable definition
...
With this patch, one could define two alias for the same function like:
FINSH_FUNCTION_EXPORT_ALIAS(rt_hw_cpu_reset, reboot, reboot the cpu);
FINSH_FUNCTION_EXPORT_ALIAS(rt_hw_cpu_reset, __cmd_reboot, reboot the cpu);
Suitable for being used in MSH/Finsh.
2013-08-31 17:58:34 +08:00
6563ae6cab
simulator: init SDL after rt_components_init
...
SDL may use DFS to open files. It should be run after
rt_components_init.
2013-08-31 16:46:36 +08:00
3d0c0af371
simulator: fix SDL initialization under Linux
...
rt_hw_sdl_start should wait untill the rtgui_graphic_set_device is
called int the sdl_loop. Use pthread_cond_wait to achieve this in Linux.
2013-08-31 16:30:32 +08:00
f7f14cf566
Merge pull request #146 from grissiom/rtt-next
...
Some fix for the portal device
2013-08-26 21:12:55 -07:00
246bdde8de
dev/portal: fix bug in portal callback forwarding
...
When forwarding rx/tx callback from the underlaying device(pipe), the
"dev" argument should be the portal. So the portal callback could think
itself as called.
2013-08-22 16:43:06 +08:00
824261d3bc
dev/portal: add checks on oflag
...
It does not make sense to open portal without a flag. One should open a
portal with RT_DEVICE_OFLAG_RDWR in most cases.
2013-08-22 15:12:00 +08:00
b4ba2657bb
Merge pull request #145 from grissiom/hot-fix
...
finsh/cmd: add description for new pipe and portal device
2013-08-20 20:02:31 -07:00
b4ab82284f
finsh/cmd: add description for new pipe and portal device
2013-08-20 14:27:28 +08:00
59f0f66996
Merge pull request #144 from grissiom/portal-device
...
Portal device
2013-08-19 22:41:31 -07:00
6e676e7754
dev/portal: implement portal device
...
Portal is a device that connect devices. Currently, you can only connect
pipes in portal. Pipes are unidirectional. But with portal, you can
construct a bidirectional device with two pipes.
2013-08-20 12:54:43 +08:00
7e68096a88
dev/pipe: make pipe a type of device
...
Pipe have many differences with char device. The main difference is
altough pipe have both read and write, it is unidirectional in nature.
2013-08-20 10:16:55 +08:00
a8f9f85f66
Merge pull request #143 from grissiom/ringbuffer-pipe
...
Ringbuffer and Pipe
2013-08-19 02:39:56 -07:00
0340300f89
dev/pipe: cleanup
2013-08-19 16:25:17 +08:00
e8cca2315a
dev/pipe: call rx_indicate the time we resume reader
...
Some reader(FinSh again) would like to block on a semaphore which is
released in rx_indicate. So we should invoke rx_indicate in
_rt_pipe_resume_reader.
2013-08-19 15:57:59 +08:00
d683d32bd5
device/pipe: add nonblocking read/write and force write mode
...
The previous implementation will always blocks the reader/writer.
However, at least FinSh would expect the device to be nonblocking ---
read should return 0 when there is no data in it.
2013-08-19 15:35:56 +08:00
4919d29d69
ringbuffer: add put_force and putchar_force API
...
Add the APIs that will discard the old data when rb is full.
2013-08-19 15:32:42 +08:00
981d929b56
ringbuffer: fix the ambiguous name
...
RT_RINGBUFFER_SIZE could mean "the size of the whole buffer", "the size
of the empty space" or "the size of the data". Moreover, it's never a
micro anymore. Change it to rt_ringbuffer_data_len before it's too late.
Also, RT_RINGBUFFER_EMPTY is changed to rt_ringbuffer_space_len.
2013-08-19 14:12:35 +08:00
6f75988cee
Merge branch 'master' of https://github.com/RT-Thread/rt-thread
2013-08-19 13:00:48 +08:00
90c9c8a27a
Merge pull request #142 from grissiom/pipe-device
...
device/pipe: add rt_pipe_{init,detach}
2013-08-18 21:47:46 -07:00
d8695085dc
Merge branch 'master' of https://github.com/RT-Thread/rt-thread
2013-08-19 12:47:40 +08:00
fa934fa7bd
add shell device automatically setting for rx_ind
2013-08-19 12:46:04 +08:00
38ce3879d8
device/pipe: add rt_pipe_{init,detach}
...
This provide the possibility that allocate the buffer of the ringbuffer
on a specific region, instead of always mallocing it. It also bring us
the benefit of using pipe device on the systems without heap.
2013-08-19 12:28:59 +08:00
cb880fd763
Merge pull request #141 from grissiom/fix-schedule-debug
...
fix debug output of scheduler
2013-08-18 19:28:20 -07:00
ad17bc2908
fix debug output of scheduler
...
When print the thread names, we should use "%.*s" instead of bare "%s".
2013-08-19 10:08:11 +08:00
0826efaab3
Merge branch 'master' of https://github.com/RT-Thread/rt-thread
2013-08-12 07:22:04 -07:00
05b9eb0b82
remove compiling warning
2013-08-12 07:21:29 -07:00
dce51ddf05
add buildlib for Keil MDK
2013-08-09 22:26:22 +08:00
60a0529553
Merge branch 'master' of https://github.com/RT-Thread/rt-thread
2013-07-30 23:30:27 +08:00
e301d14979
Merge pull request #132 from MrVan/sep6200
...
SEP6200 Support
2013-07-29 22:35:31 -07:00
ac10730492
fix O_APPEND write issue in RAMFS
2013-07-29 14:50:12 +08:00
ac850ef3f0
add device_open/close in mkfs
2013-07-29 14:43:52 +08:00
Bernard Xiong
Grissiom